From d6a5ea702b5e5bd2e669313df9fa8e485d6409c2 Mon Sep 17 00:00:00 2001 From: ohmayr Date: Thu, 25 Sep 2025 16:51:55 +0000 Subject: [PATCH 1/4] chore: release google.cloud.gkerecommender.v1 using librarian --- .librarian/state.yaml | 20 + .../google-cloud-gkerecommender/.coveragerc | 13 + packages/google-cloud-gkerecommender/.flake8 | 34 + .../.repo-metadata.json | 14 + packages/google-cloud-gkerecommender/LICENSE | 202 + .../google-cloud-gkerecommender/MANIFEST.in | 20 + .../google-cloud-gkerecommender/README.rst | 197 + .../docs/_static/custom.css | 20 + .../docs/_templates/layout.html | 50 + .../google-cloud-gkerecommender/docs/conf.py | 385 ++ .../gke_inference_quickstart.rst | 10 + .../docs/gkerecommender_v1/services_.rst | 6 + .../docs/gkerecommender_v1/types_.rst | 6 + .../docs/index.rst | 10 + .../docs/multiprocessing.rst | 7 + .../docs/summary_overview.md | 22 + .../google/cloud/gkerecommender/__init__.py | 81 + .../cloud/gkerecommender/gapic_version.py | 16 + .../google/cloud/gkerecommender/py.typed | 2 + .../cloud/gkerecommender_v1/__init__.py | 79 + .../gkerecommender_v1/gapic_metadata.json | 118 + .../cloud/gkerecommender_v1/gapic_version.py | 16 + .../google/cloud/gkerecommender_v1/py.typed | 2 + .../gkerecommender_v1/services/__init__.py | 15 + .../gke_inference_quickstart/__init__.py | 22 + .../gke_inference_quickstart/async_client.py | 894 +++ .../gke_inference_quickstart/client.py | 1296 ++++ .../gke_inference_quickstart/pagers.py | 669 ++ .../transports/README.rst | 9 + .../transports/__init__.py | 41 + .../transports/base.py | 253 + .../transports/grpc.py | 536 ++ .../transports/grpc_asyncio.py | 586 ++ .../transports/rest.py | 1532 +++++ .../transports/rest_base.py | 378 + .../cloud/gkerecommender_v1/types/__init__.py | 68 + .../gkerecommender_v1/types/gkerecommender.py | 983 +++ packages/google-cloud-gkerecommender/mypy.ini | 3 + .../google-cloud-gkerecommender/noxfile.py | 592 ++ ...uickstart_fetch_benchmarking_data_async.py | 56 + ...quickstart_fetch_benchmarking_data_sync.py | 56 + ...start_fetch_model_server_versions_async.py | 54 + ...kstart_fetch_model_server_versions_sync.py | 54 + ...ce_quickstart_fetch_model_servers_async.py | 53 + ...nce_quickstart_fetch_model_servers_sync.py | 53 + ...inference_quickstart_fetch_models_async.py | 52 + ..._inference_quickstart_fetch_models_sync.py | 52 + ...ference_quickstart_fetch_profiles_async.py | 52 + ...nference_quickstart_fetch_profiles_sync.py | 52 + ...start_generate_optimized_manifest_async.py | 57 + ...kstart_generate_optimized_manifest_sync.py | 57 + ...tadata_google.cloud.gkerecommender.v1.json | 933 +++ .../fixup_gkerecommender_v1_keywords.py | 181 + packages/google-cloud-gkerecommender/setup.py | 99 + .../testing/constraints-3.10.txt | 6 + .../testing/constraints-3.11.txt | 6 + .../testing/constraints-3.12.txt | 6 + .../testing/constraints-3.13.txt | 11 + .../testing/constraints-3.7.txt | 10 + .../testing/constraints-3.8.txt | 6 + .../testing/constraints-3.9.txt | 6 + .../tests/__init__.py | 15 + .../tests/unit/__init__.py | 15 + .../tests/unit/gapic/__init__.py | 15 + .../unit/gapic/gkerecommender_v1/__init__.py | 15 + .../test_gke_inference_quickstart.py | 6051 +++++++++++++++++ 66 files changed, 17230 insertions(+) create mode 100644 packages/google-cloud-gkerecommender/.coveragerc create mode 100644 packages/google-cloud-gkerecommender/.flake8 create mode 100644 packages/google-cloud-gkerecommender/.repo-metadata.json create mode 100644 packages/google-cloud-gkerecommender/LICENSE create mode 100644 packages/google-cloud-gkerecommender/MANIFEST.in create mode 100644 packages/google-cloud-gkerecommender/README.rst create mode 100644 packages/google-cloud-gkerecommender/docs/_static/custom.css create mode 100644 packages/google-cloud-gkerecommender/docs/_templates/layout.html create mode 100644 packages/google-cloud-gkerecommender/docs/conf.py create mode 100644 packages/google-cloud-gkerecommender/docs/gkerecommender_v1/gke_inference_quickstart.rst create mode 100644 packages/google-cloud-gkerecommender/docs/gkerecommender_v1/services_.rst create mode 100644 packages/google-cloud-gkerecommender/docs/gkerecommender_v1/types_.rst create mode 100644 packages/google-cloud-gkerecommender/docs/index.rst create mode 100644 packages/google-cloud-gkerecommender/docs/multiprocessing.rst create mode 100644 packages/google-cloud-gkerecommender/docs/summary_overview.md create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender/__init__.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender/gapic_version.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender/py.typed create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/__init__.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_metadata.json create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_version.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/py.typed create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/__init__.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/__init__.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/async_client.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/client.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/pagers.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/README.rst create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/__init__.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/base.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc_asyncio.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest_base.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/__init__.py create mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/gkerecommender.py create mode 100644 packages/google-cloud-gkerecommender/mypy.ini create mode 100644 packages/google-cloud-gkerecommender/noxfile.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py create mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/snippet_metadata_google.cloud.gkerecommender.v1.json create mode 100644 packages/google-cloud-gkerecommender/scripts/fixup_gkerecommender_v1_keywords.py create mode 100644 packages/google-cloud-gkerecommender/setup.py create mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.10.txt create mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.11.txt create mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.12.txt create mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.13.txt create mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.7.txt create mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.8.txt create mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.9.txt create mode 100644 packages/google-cloud-gkerecommender/tests/__init__.py create mode 100644 packages/google-cloud-gkerecommender/tests/unit/__init__.py create mode 100644 packages/google-cloud-gkerecommender/tests/unit/gapic/__init__.py create mode 100644 packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/__init__.py create mode 100644 packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/test_gke_inference_quickstart.py diff --git a/.librarian/state.yaml b/.librarian/state.yaml index cc78c5c72c99..936766195236 100644 --- a/.librarian/state.yaml +++ b/.librarian/state.yaml @@ -213,3 +213,23 @@ libraries: remove_regex: - packages/google-apps-chat tag_format: '{id}-v{version}' + - id: google-cloud-gkerecommender + version: 0.0.0 + last_generated_commit: 329ace5e3712a2e37d6159d4dcd998d8c73f261e + apis: + - path: google/cloud/gkerecommender/v1 + service_config: gkerecommender_v1.yaml + source_roots: + - packages/google-cloud-gkerecommender + preserve_regex: + - packages/google-cloud-gkerecommender/CHANGELOG.md + - docs/CHANGELOG.md + - docs/README.rst + - samples/README.txt + - tar.gz + - scripts/client-post-processing + - samples/snippets/README.rst + - tests/system + remove_regex: + - packages/google-cloud-gkerecommender + tag_format: '{{id}}-v{{version}}' diff --git a/packages/google-cloud-gkerecommender/.coveragerc b/packages/google-cloud-gkerecommender/.coveragerc new file mode 100644 index 000000000000..61f8e57597d7 --- /dev/null +++ b/packages/google-cloud-gkerecommender/.coveragerc @@ -0,0 +1,13 @@ +[run] +branch = True + +[report] +show_missing = True +omit = + google/cloud/gkerecommender/__init__.py + google/cloud/gkerecommender/gapic_version.py +exclude_lines = + # Re-enable the standard pragma + pragma: NO COVER + # Ignore debug-only repr + def __repr__ diff --git a/packages/google-cloud-gkerecommender/.flake8 b/packages/google-cloud-gkerecommender/.flake8 new file mode 100644 index 000000000000..90316de21489 --- /dev/null +++ b/packages/google-cloud-gkerecommender/.flake8 @@ -0,0 +1,34 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +[flake8] +# TODO(https://github.com/googleapis/gapic-generator-python/issues/2333): +# Resolve flake8 lint issues +ignore = E203, E231, E266, E501, W503 +exclude = + # TODO(https://github.com/googleapis/gapic-generator-python/issues/2333): + # Ensure that generated code passes flake8 lint + **/gapic/** + **/services/** + **/types/** + # Exclude Protobuf gencode + *_pb2.py + + # Standard linting exemptions. + **/.nox/** + __pycache__, + .git, + *.pyc, + conf.py diff --git a/packages/google-cloud-gkerecommender/.repo-metadata.json b/packages/google-cloud-gkerecommender/.repo-metadata.json new file mode 100644 index 000000000000..4834c4122181 --- /dev/null +++ b/packages/google-cloud-gkerecommender/.repo-metadata.json @@ -0,0 +1,14 @@ +{ + "api_shortname": "gkerecommender", + "name_pretty": "GKE Recommender API", + "product_documentation": "https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart", + "api_description": "GKE Recommender API", + "client_documentation": "https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest", + "issue_tracker": "https://github.com/googleapis/google-cloud-python/issues", + "release_level": "preview", + "language": "python", + "library_type": "GAPIC_AUTO", + "repo": "googleapis/google-cloud-python", + "distribution_name": "google-cloud-gkerecommender", + "api_id": "gkerecommender.googleapis.com" +} diff --git a/packages/google-cloud-gkerecommender/LICENSE b/packages/google-cloud-gkerecommender/LICENSE new file mode 100644 index 000000000000..d64569567334 --- /dev/null +++ b/packages/google-cloud-gkerecommender/LICENSE @@ -0,0 +1,202 @@ + + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/packages/google-cloud-gkerecommender/MANIFEST.in b/packages/google-cloud-gkerecommender/MANIFEST.in new file mode 100644 index 000000000000..dae249ec8976 --- /dev/null +++ b/packages/google-cloud-gkerecommender/MANIFEST.in @@ -0,0 +1,20 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +include README.rst LICENSE +recursive-include google *.py *.pyi *.json *.proto py.typed +recursive-include tests * +global-exclude *.py[co] +global-exclude __pycache__ diff --git a/packages/google-cloud-gkerecommender/README.rst b/packages/google-cloud-gkerecommender/README.rst new file mode 100644 index 000000000000..dd6d4b294bb3 --- /dev/null +++ b/packages/google-cloud-gkerecommender/README.rst @@ -0,0 +1,197 @@ +Python Client for GKE Recommender API +===================================== + +|preview| |pypi| |versions| + +`GKE Recommender API`_: GKE Recommender API + +- `Client Library Documentation`_ +- `Product Documentation`_ + +.. |preview| image:: https://img.shields.io/badge/support-preview-orange.svg + :target: https://github.com/googleapis/google-cloud-python/blob/main/README.rst#stability-levels +.. |pypi| image:: https://img.shields.io/pypi/v/google-cloud-gkerecommender.svg + :target: https://pypi.org/project/google-cloud-gkerecommender/ +.. |versions| image:: https://img.shields.io/pypi/pyversions/google-cloud-gkerecommender.svg + :target: https://pypi.org/project/google-cloud-gkerecommender/ +.. _GKE Recommender API: https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart +.. _Client Library Documentation: https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest/summary_overview +.. _Product Documentation: https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart + +Quick Start +----------- + +In order to use this library, you first need to go through the following steps: + +1. `Select or create a Cloud Platform project.`_ +2. `Enable billing for your project.`_ +3. `Enable the GKE Recommender API.`_ +4. `Set up Authentication.`_ + +.. _Select or create a Cloud Platform project.: https://console.cloud.google.com/project +.. _Enable billing for your project.: https://cloud.google.com/billing/docs/how-to/modify-project#enable_billing_for_a_project +.. _Enable the GKE Recommender API.: https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart +.. _Set up Authentication.: https://googleapis.dev/python/google-api-core/latest/auth.html + +Installation +~~~~~~~~~~~~ + +Install this library in a virtual environment using `venv`_. `venv`_ is a tool that +creates isolated Python environments. These isolated environments can have separate +versions of Python packages, which allows you to isolate one project's dependencies +from the dependencies of other projects. + +With `venv`_, it's possible to install this library without needing system +install permissions, and without clashing with the installed system +dependencies. + +.. _`venv`: https://docs.python.org/3/library/venv.html + + +Code samples and snippets +~~~~~~~~~~~~~~~~~~~~~~~~~ + +Code samples and snippets live in the `samples/`_ folder. + +.. _samples/: https://github.com/googleapis/google-cloud-python/tree/main/packages/google-cloud-gkerecommender/samples + + +Supported Python Versions +^^^^^^^^^^^^^^^^^^^^^^^^^ +Our client libraries are compatible with all current `active`_ and `maintenance`_ versions of +Python. + +Python >= 3.7 + +.. _active: https://devguide.python.org/devcycle/#in-development-main-branch +.. _maintenance: https://devguide.python.org/devcycle/#maintenance-branches + +Unsupported Python Versions +^^^^^^^^^^^^^^^^^^^^^^^^^^^ +Python <= 3.6 + +If you are using an `end-of-life`_ +version of Python, we recommend that you update as soon as possible to an actively supported version. + +.. _end-of-life: https://devguide.python.org/devcycle/#end-of-life-branches + +Mac/Linux +^^^^^^^^^ + +.. code-block:: console + + python3 -m venv + source /bin/activate + pip install google-cloud-gkerecommender + + +Windows +^^^^^^^ + +.. code-block:: console + + py -m venv + .\\Scripts\activate + pip install google-cloud-gkerecommender + +Next Steps +~~~~~~~~~~ + +- Read the `Client Library Documentation`_ for GKE Recommender API + to see other available methods on the client. +- Read the `GKE Recommender API Product documentation`_ to learn + more about the product and see How-to Guides. +- View this `README`_ to see the full list of Cloud + APIs that we cover. + +.. _GKE Recommender API Product documentation: https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart +.. _README: https://github.com/googleapis/google-cloud-python/blob/main/README.rst + +Logging +------- + +This library uses the standard Python :code:`logging` functionality to log some RPC events that could be of interest for debugging and monitoring purposes. +Note the following: + +#. Logs may contain sensitive information. Take care to **restrict access to the logs** if they are saved, whether it be on local storage or on Google Cloud Logging. +#. Google may refine the occurrence, level, and content of various log messages in this library without flagging such changes as breaking. **Do not depend on immutability of the logging events**. +#. By default, the logging events from this library are not handled. You must **explicitly configure log handling** using one of the mechanisms below. + +Simple, environment-based configuration +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +To enable logging for this library without any changes in your code, set the :code:`GOOGLE_SDK_PYTHON_LOGGING_SCOPE` environment variable to a valid Google +logging scope. This configures handling of logging events (at level :code:`logging.DEBUG` or higher) from this library in a default manner, emitting the logged +messages in a structured format. It does not currently allow customizing the logging levels captured nor the handlers, formatters, etc. used for any logging +event. + +A logging scope is a period-separated namespace that begins with :code:`google`, identifying the Python module or package to log. + +- Valid logging scopes: :code:`google`, :code:`google.cloud.asset.v1`, :code:`google.api`, :code:`google.auth`, etc. +- Invalid logging scopes: :code:`foo`, :code:`123`, etc. + +**NOTE**: If the logging scope is invalid, the library does not set up any logging handlers. + +Environment-Based Examples +^^^^^^^^^^^^^^^^^^^^^^^^^^ + +- Enabling the default handler for all Google-based loggers + +.. code-block:: console + + export GOOGLE_SDK_PYTHON_LOGGING_SCOPE=google + +- Enabling the default handler for a specific Google module (for a client library called :code:`library_v1`): + +.. code-block:: console + + export GOOGLE_SDK_PYTHON_LOGGING_SCOPE=google.cloud.library_v1 + + +Advanced, code-based configuration +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +You can also configure a valid logging scope using Python's standard `logging` mechanism. + +Code-Based Examples +^^^^^^^^^^^^^^^^^^^ + +- Configuring a handler for all Google-based loggers + +.. code-block:: python + + import logging + + from google.cloud import library_v1 + + base_logger = logging.getLogger("google") + base_logger.addHandler(logging.StreamHandler()) + base_logger.setLevel(logging.DEBUG) + +- Configuring a handler for a specific Google module (for a client library called :code:`library_v1`): + +.. code-block:: python + + import logging + + from google.cloud import library_v1 + + base_logger = logging.getLogger("google.cloud.library_v1") + base_logger.addHandler(logging.StreamHandler()) + base_logger.setLevel(logging.DEBUG) + +Logging details +~~~~~~~~~~~~~~~ + +#. Regardless of which of the mechanisms above you use to configure logging for this library, by default logging events are not propagated up to the root + logger from the `google`-level logger. If you need the events to be propagated to the root logger, you must explicitly set + :code:`logging.getLogger("google").propagate = True` in your code. +#. You can mix the different logging configurations above for different Google modules. For example, you may want use a code-based logging configuration for + one library, but decide you need to also set up environment-based logging configuration for another library. + + #. If you attempt to use both code-based and environment-based configuration for the same module, the environment-based configuration will be ineffectual + if the code -based configuration gets applied first. + +#. The Google-specific logging configurations (default handlers for environment-based configuration; not propagating logging events to the root logger) get + executed the first time *any* client library is instantiated in your application, and only if the affected loggers have not been previously configured. + (This is the reason for 2.i. above.) diff --git a/packages/google-cloud-gkerecommender/docs/_static/custom.css b/packages/google-cloud-gkerecommender/docs/_static/custom.css new file mode 100644 index 000000000000..b0a295464b23 --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/_static/custom.css @@ -0,0 +1,20 @@ +div#python2-eol { + border-color: red; + border-width: medium; +} + +/* Ensure minimum width for 'Parameters' / 'Returns' column */ +dl.field-list > dt { + min-width: 100px +} + +/* Insert space between methods for readability */ +dl.method { + padding-top: 10px; + padding-bottom: 10px +} + +/* Insert empty space between classes */ +dl.class { + padding-bottom: 50px +} diff --git a/packages/google-cloud-gkerecommender/docs/_templates/layout.html b/packages/google-cloud-gkerecommender/docs/_templates/layout.html new file mode 100644 index 000000000000..95e9c77fcfe1 --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/_templates/layout.html @@ -0,0 +1,50 @@ + +{% extends "!layout.html" %} +{%- block content %} +{%- if theme_fixed_sidebar|lower == 'true' %} +
+ {{ sidebar() }} + {%- block document %} +
+ {%- if render_sidebar %} +
+ {%- endif %} + + {%- block relbar_top %} + {%- if theme_show_relbar_top|tobool %} +
+   + {{- rellink_markup () }} +
+ {%- endif %} + {% endblock %} + +
+
+ As of January 1, 2020 this library no longer supports Python 2 on the latest released version. + Library versions released prior to that date will continue to be available. For more information please + visit Python 2 support on Google Cloud. +
+ {% block body %} {% endblock %} +
+ + {%- block relbar_bottom %} + {%- if theme_show_relbar_bottom|tobool %} +
+   + {{- rellink_markup () }} +
+ {%- endif %} + {% endblock %} + + {%- if render_sidebar %} +
+ {%- endif %} +
+ {%- endblock %} +
+
+{%- else %} +{{ super() }} +{%- endif %} +{%- endblock %} diff --git a/packages/google-cloud-gkerecommender/docs/conf.py b/packages/google-cloud-gkerecommender/docs/conf.py new file mode 100644 index 000000000000..77dcf0d97d44 --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/conf.py @@ -0,0 +1,385 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# +# google-cloud-gkerecommender documentation build configuration file +# +# This file is execfile()d with the current directory set to its +# containing dir. +# +# Note that not all possible configuration values are present in this +# autogenerated file. +# +# All configuration values have a default; values that are commented out +# serve to show the default. + +import os +import shlex +import sys + +# If extensions (or modules to document with autodoc) are in another directory, +# add these directories to sys.path here. If the directory is relative to the +# documentation root, use os.path.abspath to make it absolute, like shown here. +sys.path.insert(0, os.path.abspath("..")) + +# For plugins that can not read conf.py. +# See also: https://github.com/docascode/sphinx-docfx-yaml/issues/85 +sys.path.insert(0, os.path.abspath(".")) + +__version__ = "" + +# -- General configuration ------------------------------------------------ + +# If your documentation needs a minimal Sphinx version, state it here. +needs_sphinx = "4.5.0" + +# Add any Sphinx extension module names here, as strings. They can be +# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom +# ones. +extensions = [ + "sphinx.ext.autodoc", + "sphinx.ext.autosummary", + "sphinx.ext.intersphinx", + "sphinx.ext.coverage", + "sphinx.ext.doctest", + "sphinx.ext.napoleon", + "sphinx.ext.todo", + "sphinx.ext.viewcode", + "recommonmark", +] + +# autodoc/autosummary flags +autoclass_content = "both" +autodoc_default_options = {"members": True} +autosummary_generate = True + + +# Add any paths that contain templates here, relative to this directory. +templates_path = ["_templates"] + +# The suffix(es) of source filenames. +# You can specify multiple suffix as a list of string: +# source_suffix = ['.rst', '.md'] +source_suffix = [".rst", ".md"] + +# The encoding of source files. +# source_encoding = 'utf-8-sig' + +# The root toctree document. +root_doc = "index" + +# General information about the project. +project = "google-cloud-gkerecommender" +copyright = "2025, Google, LLC" +author = "Google APIs" + +# The version info for the project you're documenting, acts as replacement for +# |version| and |release|, also used in various other places throughout the +# built documents. +# +# The full version, including alpha/beta/rc tags. +release = __version__ +# The short X.Y version. +version = ".".join(release.split(".")[0:2]) + +# The language for content autogenerated by Sphinx. Refer to documentation +# for a list of supported languages. +# +# This is also used if you do content translation via gettext catalogs. +# Usually you set "language" from the command line for these cases. +language = None + +# There are two options for replacing |today|: either, you set today to some +# non-false value, then it is used: +# today = '' +# Else, today_fmt is used as the format for a strftime call. +# today_fmt = '%B %d, %Y' + +# List of patterns, relative to source directory, that match files and +# directories to ignore when looking for source files. +exclude_patterns = [ + "_build", + "**/.nox/**/*", + "samples/AUTHORING_GUIDE.md", + "samples/CONTRIBUTING.md", + "samples/snippets/README.rst", +] + +# The reST default role (used for this markup: `text`) to use for all +# documents. +# default_role = None + +# If true, '()' will be appended to :func: etc. cross-reference text. +# add_function_parentheses = True + +# If true, the current module name will be prepended to all description +# unit titles (such as .. function::). +# add_module_names = True + +# If true, sectionauthor and moduleauthor directives will be shown in the +# output. They are ignored by default. +# show_authors = False + +# The name of the Pygments (syntax highlighting) style to use. +pygments_style = "sphinx" + +# A list of ignored prefixes for module index sorting. +# modindex_common_prefix = [] + +# If true, keep warnings as "system message" paragraphs in the built documents. +# keep_warnings = False + +# If true, `todo` and `todoList` produce output, else they produce nothing. +todo_include_todos = True + + +# -- Options for HTML output ---------------------------------------------- + +# The theme to use for HTML and HTML Help pages. See the documentation for +# a list of builtin themes. +html_theme = "alabaster" + +# Theme options are theme-specific and customize the look and feel of a theme +# further. For a list of options available for each theme, see the +# documentation. +html_theme_options = { + "description": "Google Cloud Client Libraries for google-cloud-gkerecommender", + "github_user": "googleapis", + "github_repo": "google-cloud-python", + "github_banner": True, + "font_family": "'Roboto', Georgia, sans", + "head_font_family": "'Roboto', Georgia, serif", + "code_font_family": "'Roboto Mono', 'Consolas', monospace", +} + +# Add any paths that contain custom themes here, relative to this directory. +# html_theme_path = [] + +# The name for this set of Sphinx documents. If None, it defaults to +# " v documentation". +# html_title = None + +# A shorter title for the navigation bar. Default is the same as html_title. +# html_short_title = None + +# The name of an image file (relative to this directory) to place at the top +# of the sidebar. +# html_logo = None + +# The name of an image file (within the static path) to use as favicon of the +# docs. This file should be a Windows icon file (.ico) being 16x16 or 32x32 +# pixels large. +# html_favicon = None + +# Add any paths that contain custom static files (such as style sheets) here, +# relative to this directory. They are copied after the builtin static files, +# so a file named "default.css" will overwrite the builtin "default.css". +html_static_path = ["_static"] + +# Add any extra paths that contain custom files (such as robots.txt or +# .htaccess) here, relative to this directory. These files are copied +# directly to the root of the documentation. +# html_extra_path = [] + +# If not '', a 'Last updated on:' timestamp is inserted at every page bottom, +# using the given strftime format. +# html_last_updated_fmt = '%b %d, %Y' + +# If true, SmartyPants will be used to convert quotes and dashes to +# typographically correct entities. +# html_use_smartypants = True + +# Custom sidebar templates, maps document names to template names. +# html_sidebars = {} + +# Additional templates that should be rendered to pages, maps page names to +# template names. +# html_additional_pages = {} + +# If false, no module index is generated. +# html_domain_indices = True + +# If false, no index is generated. +# html_use_index = True + +# If true, the index is split into individual pages for each letter. +# html_split_index = False + +# If true, links to the reST sources are added to the pages. +# html_show_sourcelink = True + +# If true, "Created using Sphinx" is shown in the HTML footer. Default is True. +# html_show_sphinx = True + +# If true, "(C) Copyright ..." is shown in the HTML footer. Default is True. +# html_show_copyright = True + +# If true, an OpenSearch description file will be output, and all pages will +# contain a tag referring to it. The value of this option must be the +# base URL from which the finished HTML is served. +# html_use_opensearch = '' + +# This is the file name suffix for HTML files (e.g. ".xhtml"). +# html_file_suffix = None + +# Language to be used for generating the HTML full-text search index. +# Sphinx supports the following languages: +# 'da', 'de', 'en', 'es', 'fi', 'fr', 'hu', 'it', 'ja' +# 'nl', 'no', 'pt', 'ro', 'ru', 'sv', 'tr' +# html_search_language = 'en' + +# A dictionary with options for the search language support, empty by default. +# Now only 'ja' uses this config value +# html_search_options = {'type': 'default'} + +# The name of a javascript file (relative to the configuration directory) that +# implements a search results scorer. If empty, the default will be used. +# html_search_scorer = 'scorer.js' + +# Output file base name for HTML help builder. +htmlhelp_basename = "google-cloud-gkerecommender-doc" + +# -- Options for warnings ------------------------------------------------------ + + +suppress_warnings = [ + # Temporarily suppress this to avoid "more than one target found for + # cross-reference" warning, which are intractable for us to avoid while in + # a mono-repo. + # See https://github.com/sphinx-doc/sphinx/blob + # /2a65ffeef5c107c19084fabdd706cdff3f52d93c/sphinx/domains/python.py#L843 + "ref.python" +] + +# -- Options for LaTeX output --------------------------------------------- + +latex_elements = { + # The paper size ('letterpaper' or 'a4paper'). + # 'papersize': 'letterpaper', + # The font size ('10pt', '11pt' or '12pt'). + # 'pointsize': '10pt', + # Additional stuff for the LaTeX preamble. + # 'preamble': '', + # Latex figure (float) alignment + # 'figure_align': 'htbp', +} + +# Grouping the document tree into LaTeX files. List of tuples +# (source start file, target name, title, +# author, documentclass [howto, manual, or own class]). +latex_documents = [ + ( + root_doc, + "google-cloud-gkerecommender.tex", + "google-cloud-gkerecommender Documentation", + author, + "manual", + ) +] + +# The name of an image file (relative to this directory) to place at the top of +# the title page. +# latex_logo = None + +# For "manual" documents, if this is true, then toplevel headings are parts, +# not chapters. +# latex_use_parts = False + +# If true, show page references after internal links. +# latex_show_pagerefs = False + +# If true, show URL addresses after external links. +# latex_show_urls = False + +# Documents to append as an appendix to all manuals. +# latex_appendices = [] + +# If false, no module index is generated. +# latex_domain_indices = True + + +# -- Options for manual page output --------------------------------------- + +# One entry per manual page. List of tuples +# (source start file, name, description, authors, manual section). +man_pages = [ + ( + root_doc, + "google-cloud-gkerecommender", + "google-cloud-gkerecommender Documentation", + [author], + 1, + ) +] + +# If true, show URL addresses after external links. +# man_show_urls = False + + +# -- Options for Texinfo output ------------------------------------------- + +# Grouping the document tree into Texinfo files. List of tuples +# (source start file, target name, title, author, +# dir menu entry, description, category) +texinfo_documents = [ + ( + root_doc, + "google-cloud-gkerecommender", + "google-cloud-gkerecommender Documentation", + author, + "google-cloud-gkerecommender", + "google-cloud-gkerecommender Library", + "APIs", + ) +] + +# Documents to append as an appendix to all manuals. +# texinfo_appendices = [] + +# If false, no module index is generated. +# texinfo_domain_indices = True + +# How to display URL addresses: 'footnote', 'no', or 'inline'. +# texinfo_show_urls = 'footnote' + +# If true, do not generate a @detailmenu in the "Top" node's menu. +# texinfo_no_detailmenu = False + + +# Example configuration for intersphinx: refer to the Python standard library. +intersphinx_mapping = { + "python": ("https://python.readthedocs.org/en/latest/", None), + "google-auth": ("https://googleapis.dev/python/google-auth/latest/", None), + "google.api_core": ( + "https://googleapis.dev/python/google-api-core/latest/", + None, + ), + "grpc": ("https://grpc.github.io/grpc/python/", None), + "proto-plus": ("https://proto-plus-python.readthedocs.io/en/latest/", None), + "protobuf": ("https://googleapis.dev/python/protobuf/latest/", None), +} + + +# Napoleon settings +napoleon_google_docstring = True +napoleon_numpy_docstring = True +napoleon_include_private_with_doc = False +napoleon_include_special_with_doc = True +napoleon_use_admonition_for_examples = False +napoleon_use_admonition_for_notes = False +napoleon_use_admonition_for_references = False +napoleon_use_ivar = False +napoleon_use_param = True +napoleon_use_rtype = True diff --git a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/gke_inference_quickstart.rst b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/gke_inference_quickstart.rst new file mode 100644 index 000000000000..60b290775d65 --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/gke_inference_quickstart.rst @@ -0,0 +1,10 @@ +GkeInferenceQuickstart +---------------------------------------- + +.. automodule:: google.cloud.gkerecommender_v1.services.gke_inference_quickstart + :members: + :inherited-members: + +.. automodule:: google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers + :members: + :inherited-members: diff --git a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/services_.rst b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/services_.rst new file mode 100644 index 000000000000..6db73968996f --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/services_.rst @@ -0,0 +1,6 @@ +Services for Google Cloud Gkerecommender v1 API +=============================================== +.. toctree:: + :maxdepth: 2 + + gke_inference_quickstart diff --git a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/types_.rst b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/types_.rst new file mode 100644 index 000000000000..eb4cb3310dc5 --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/types_.rst @@ -0,0 +1,6 @@ +Types for Google Cloud Gkerecommender v1 API +============================================ + +.. automodule:: google.cloud.gkerecommender_v1.types + :members: + :show-inheritance: diff --git a/packages/google-cloud-gkerecommender/docs/index.rst b/packages/google-cloud-gkerecommender/docs/index.rst new file mode 100644 index 000000000000..8220e3992d15 --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/index.rst @@ -0,0 +1,10 @@ +.. include:: multiprocessing.rst + + +API Reference +------------- +.. toctree:: + :maxdepth: 2 + + gkerecommender_v1/services_ + gkerecommender_v1/types_ diff --git a/packages/google-cloud-gkerecommender/docs/multiprocessing.rst b/packages/google-cloud-gkerecommender/docs/multiprocessing.rst new file mode 100644 index 000000000000..536d17b2ea65 --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/multiprocessing.rst @@ -0,0 +1,7 @@ +.. note:: + + Because this client uses :mod:`grpc` library, it is safe to + share instances across threads. In multiprocessing scenarios, the best + practice is to create client instances *after* the invocation of + :func:`os.fork` by :class:`multiprocessing.pool.Pool` or + :class:`multiprocessing.Process`. diff --git a/packages/google-cloud-gkerecommender/docs/summary_overview.md b/packages/google-cloud-gkerecommender/docs/summary_overview.md new file mode 100644 index 000000000000..5878774fd11e --- /dev/null +++ b/packages/google-cloud-gkerecommender/docs/summary_overview.md @@ -0,0 +1,22 @@ +[ +This is a templated file. Adding content to this file may result in it being +reverted. Instead, if you want to place additional content, create an +"overview_content.md" file in `docs/` directory. The Sphinx tool will +pick up on the content and merge the content. +]: # + +# GKE Recommender API API + +Overview of the APIs available for GKE Recommender API API. + +## All entries + +Classes, methods and properties & attributes for +GKE Recommender API API. + +[classes](https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest/summary_class.html) + +[methods](https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest/summary_method.html) + +[properties and +attributes](https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest/summary_property.html) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/__init__.py new file mode 100644 index 000000000000..937bc88d6b72 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/__init__.py @@ -0,0 +1,81 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from google.cloud.gkerecommender import gapic_version as package_version + +__version__ = package_version.__version__ + + +from google.cloud.gkerecommender_v1.services.gke_inference_quickstart.async_client import ( + GkeInferenceQuickstartAsyncClient, +) +from google.cloud.gkerecommender_v1.services.gke_inference_quickstart.client import ( + GkeInferenceQuickstartClient, +) +from google.cloud.gkerecommender_v1.types.gkerecommender import ( + Amount, + Cost, + FetchBenchmarkingDataRequest, + FetchBenchmarkingDataResponse, + FetchModelServersRequest, + FetchModelServersResponse, + FetchModelServerVersionsRequest, + FetchModelServerVersionsResponse, + FetchModelsRequest, + FetchModelsResponse, + FetchProfilesRequest, + FetchProfilesResponse, + GenerateOptimizedManifestRequest, + GenerateOptimizedManifestResponse, + KubernetesManifest, + MillisecondRange, + ModelServerInfo, + PerformanceRange, + PerformanceRequirements, + PerformanceStats, + Profile, + ResourcesUsed, + StorageConfig, + TokensPerSecondRange, +) + +__all__ = ( + "GkeInferenceQuickstartClient", + "GkeInferenceQuickstartAsyncClient", + "Amount", + "Cost", + "FetchBenchmarkingDataRequest", + "FetchBenchmarkingDataResponse", + "FetchModelServersRequest", + "FetchModelServersResponse", + "FetchModelServerVersionsRequest", + "FetchModelServerVersionsResponse", + "FetchModelsRequest", + "FetchModelsResponse", + "FetchProfilesRequest", + "FetchProfilesResponse", + "GenerateOptimizedManifestRequest", + "GenerateOptimizedManifestResponse", + "KubernetesManifest", + "MillisecondRange", + "ModelServerInfo", + "PerformanceRange", + "PerformanceRequirements", + "PerformanceStats", + "Profile", + "ResourcesUsed", + "StorageConfig", + "TokensPerSecondRange", +) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/gapic_version.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/gapic_version.py new file mode 100644 index 000000000000..20a9cd975b02 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/gapic_version.py @@ -0,0 +1,16 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +__version__ = "0.0.0" # {x-release-please-version} diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/py.typed b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/py.typed new file mode 100644 index 000000000000..ebf2186dedbf --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/py.typed @@ -0,0 +1,2 @@ +# Marker file for PEP 561. +# The google-cloud-gkerecommender package uses inline types. diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/__init__.py new file mode 100644 index 000000000000..97b37003c197 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/__init__.py @@ -0,0 +1,79 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from google.cloud.gkerecommender_v1 import gapic_version as package_version + +__version__ = package_version.__version__ + + +from .services.gke_inference_quickstart import ( + GkeInferenceQuickstartAsyncClient, + GkeInferenceQuickstartClient, +) +from .types.gkerecommender import ( + Amount, + Cost, + FetchBenchmarkingDataRequest, + FetchBenchmarkingDataResponse, + FetchModelServersRequest, + FetchModelServersResponse, + FetchModelServerVersionsRequest, + FetchModelServerVersionsResponse, + FetchModelsRequest, + FetchModelsResponse, + FetchProfilesRequest, + FetchProfilesResponse, + GenerateOptimizedManifestRequest, + GenerateOptimizedManifestResponse, + KubernetesManifest, + MillisecondRange, + ModelServerInfo, + PerformanceRange, + PerformanceRequirements, + PerformanceStats, + Profile, + ResourcesUsed, + StorageConfig, + TokensPerSecondRange, +) + +__all__ = ( + "GkeInferenceQuickstartAsyncClient", + "Amount", + "Cost", + "FetchBenchmarkingDataRequest", + "FetchBenchmarkingDataResponse", + "FetchModelServerVersionsRequest", + "FetchModelServerVersionsResponse", + "FetchModelServersRequest", + "FetchModelServersResponse", + "FetchModelsRequest", + "FetchModelsResponse", + "FetchProfilesRequest", + "FetchProfilesResponse", + "GenerateOptimizedManifestRequest", + "GenerateOptimizedManifestResponse", + "GkeInferenceQuickstartClient", + "KubernetesManifest", + "MillisecondRange", + "ModelServerInfo", + "PerformanceRange", + "PerformanceRequirements", + "PerformanceStats", + "Profile", + "ResourcesUsed", + "StorageConfig", + "TokensPerSecondRange", +) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_metadata.json b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_metadata.json new file mode 100644 index 000000000000..277d36460b3b --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_metadata.json @@ -0,0 +1,118 @@ + { + "comment": "This file maps proto services/RPCs to the corresponding library clients/methods", + "language": "python", + "libraryPackage": "google.cloud.gkerecommender_v1", + "protoPackage": "google.cloud.gkerecommender.v1", + "schema": "1.0", + "services": { + "GkeInferenceQuickstart": { + "clients": { + "grpc": { + "libraryClient": "GkeInferenceQuickstartClient", + "rpcs": { + "FetchBenchmarkingData": { + "methods": [ + "fetch_benchmarking_data" + ] + }, + "FetchModelServerVersions": { + "methods": [ + "fetch_model_server_versions" + ] + }, + "FetchModelServers": { + "methods": [ + "fetch_model_servers" + ] + }, + "FetchModels": { + "methods": [ + "fetch_models" + ] + }, + "FetchProfiles": { + "methods": [ + "fetch_profiles" + ] + }, + "GenerateOptimizedManifest": { + "methods": [ + "generate_optimized_manifest" + ] + } + } + }, + "grpc-async": { + "libraryClient": "GkeInferenceQuickstartAsyncClient", + "rpcs": { + "FetchBenchmarkingData": { + "methods": [ + "fetch_benchmarking_data" + ] + }, + "FetchModelServerVersions": { + "methods": [ + "fetch_model_server_versions" + ] + }, + "FetchModelServers": { + "methods": [ + "fetch_model_servers" + ] + }, + "FetchModels": { + "methods": [ + "fetch_models" + ] + }, + "FetchProfiles": { + "methods": [ + "fetch_profiles" + ] + }, + "GenerateOptimizedManifest": { + "methods": [ + "generate_optimized_manifest" + ] + } + } + }, + "rest": { + "libraryClient": "GkeInferenceQuickstartClient", + "rpcs": { + "FetchBenchmarkingData": { + "methods": [ + "fetch_benchmarking_data" + ] + }, + "FetchModelServerVersions": { + "methods": [ + "fetch_model_server_versions" + ] + }, + "FetchModelServers": { + "methods": [ + "fetch_model_servers" + ] + }, + "FetchModels": { + "methods": [ + "fetch_models" + ] + }, + "FetchProfiles": { + "methods": [ + "fetch_profiles" + ] + }, + "GenerateOptimizedManifest": { + "methods": [ + "generate_optimized_manifest" + ] + } + } + } + } + } + } +} diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_version.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_version.py new file mode 100644 index 000000000000..20a9cd975b02 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_version.py @@ -0,0 +1,16 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +__version__ = "0.0.0" # {x-release-please-version} diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/py.typed b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/py.typed new file mode 100644 index 000000000000..ebf2186dedbf --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/py.typed @@ -0,0 +1,2 @@ +# Marker file for PEP 561. +# The google-cloud-gkerecommender package uses inline types. diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/__init__.py new file mode 100644 index 000000000000..cbf94b283c70 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/__init__.py @@ -0,0 +1,15 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/__init__.py new file mode 100644 index 000000000000..2f911bff8338 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/__init__.py @@ -0,0 +1,22 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from .async_client import GkeInferenceQuickstartAsyncClient +from .client import GkeInferenceQuickstartClient + +__all__ = ( + "GkeInferenceQuickstartClient", + "GkeInferenceQuickstartAsyncClient", +) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/async_client.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/async_client.py new file mode 100644 index 000000000000..19182edb0c6c --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/async_client.py @@ -0,0 +1,894 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from collections import OrderedDict +import logging as std_logging +import re +from typing import ( + Callable, + Dict, + Mapping, + MutableMapping, + MutableSequence, + Optional, + Sequence, + Tuple, + Type, + Union, +) + +from google.api_core import exceptions as core_exceptions +from google.api_core import gapic_v1 +from google.api_core import retry_async as retries +from google.api_core.client_options import ClientOptions +from google.auth import credentials as ga_credentials # type: ignore +from google.oauth2 import service_account # type: ignore +import google.protobuf + +from google.cloud.gkerecommender_v1 import gapic_version as package_version + +try: + OptionalRetry = Union[retries.AsyncRetry, gapic_v1.method._MethodDefault, None] +except AttributeError: # pragma: NO COVER + OptionalRetry = Union[retries.AsyncRetry, object, None] # type: ignore + +from google.cloud.gkerecommender_v1.services.gke_inference_quickstart import pagers +from google.cloud.gkerecommender_v1.types import gkerecommender + +from .client import GkeInferenceQuickstartClient +from .transports.base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport +from .transports.grpc_asyncio import GkeInferenceQuickstartGrpcAsyncIOTransport + +try: + from google.api_core import client_logging # type: ignore + + CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER +except ImportError: # pragma: NO COVER + CLIENT_LOGGING_SUPPORTED = False + +_LOGGER = std_logging.getLogger(__name__) + + +class GkeInferenceQuickstartAsyncClient: + """GKE Inference Quickstart (GIQ) service provides profiles with + performance metrics for popular models and model servers across + multiple accelerators. These profiles help generate optimized + best practices for running inference on GKE. + """ + + _client: GkeInferenceQuickstartClient + + # Copy defaults from the synchronous client for use here. + # Note: DEFAULT_ENDPOINT is deprecated. Use _DEFAULT_ENDPOINT_TEMPLATE instead. + DEFAULT_ENDPOINT = GkeInferenceQuickstartClient.DEFAULT_ENDPOINT + DEFAULT_MTLS_ENDPOINT = GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT + _DEFAULT_ENDPOINT_TEMPLATE = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE + _DEFAULT_UNIVERSE = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE + + common_billing_account_path = staticmethod( + GkeInferenceQuickstartClient.common_billing_account_path + ) + parse_common_billing_account_path = staticmethod( + GkeInferenceQuickstartClient.parse_common_billing_account_path + ) + common_folder_path = staticmethod(GkeInferenceQuickstartClient.common_folder_path) + parse_common_folder_path = staticmethod( + GkeInferenceQuickstartClient.parse_common_folder_path + ) + common_organization_path = staticmethod( + GkeInferenceQuickstartClient.common_organization_path + ) + parse_common_organization_path = staticmethod( + GkeInferenceQuickstartClient.parse_common_organization_path + ) + common_project_path = staticmethod(GkeInferenceQuickstartClient.common_project_path) + parse_common_project_path = staticmethod( + GkeInferenceQuickstartClient.parse_common_project_path + ) + common_location_path = staticmethod( + GkeInferenceQuickstartClient.common_location_path + ) + parse_common_location_path = staticmethod( + GkeInferenceQuickstartClient.parse_common_location_path + ) + + @classmethod + def from_service_account_info(cls, info: dict, *args, **kwargs): + """Creates an instance of this client using the provided credentials + info. + + Args: + info (dict): The service account private key info. + args: Additional arguments to pass to the constructor. + kwargs: Additional arguments to pass to the constructor. + + Returns: + GkeInferenceQuickstartAsyncClient: The constructed client. + """ + return GkeInferenceQuickstartClient.from_service_account_info.__func__(GkeInferenceQuickstartAsyncClient, info, *args, **kwargs) # type: ignore + + @classmethod + def from_service_account_file(cls, filename: str, *args, **kwargs): + """Creates an instance of this client using the provided credentials + file. + + Args: + filename (str): The path to the service account private key json + file. + args: Additional arguments to pass to the constructor. + kwargs: Additional arguments to pass to the constructor. + + Returns: + GkeInferenceQuickstartAsyncClient: The constructed client. + """ + return GkeInferenceQuickstartClient.from_service_account_file.__func__(GkeInferenceQuickstartAsyncClient, filename, *args, **kwargs) # type: ignore + + from_service_account_json = from_service_account_file + + @classmethod + def get_mtls_endpoint_and_cert_source( + cls, client_options: Optional[ClientOptions] = None + ): + """Return the API endpoint and client cert source for mutual TLS. + + The client cert source is determined in the following order: + (1) if `GOOGLE_API_USE_CLIENT_CERTIFICATE` environment variable is not "true", the + client cert source is None. + (2) if `client_options.client_cert_source` is provided, use the provided one; if the + default client cert source exists, use the default one; otherwise the client cert + source is None. + + The API endpoint is determined in the following order: + (1) if `client_options.api_endpoint` if provided, use the provided one. + (2) if `GOOGLE_API_USE_CLIENT_CERTIFICATE` environment variable is "always", use the + default mTLS endpoint; if the environment variable is "never", use the default API + endpoint; otherwise if client cert source exists, use the default mTLS endpoint, otherwise + use the default API endpoint. + + More details can be found at https://google.aip.dev/auth/4114. + + Args: + client_options (google.api_core.client_options.ClientOptions): Custom options for the + client. Only the `api_endpoint` and `client_cert_source` properties may be used + in this method. + + Returns: + Tuple[str, Callable[[], Tuple[bytes, bytes]]]: returns the API endpoint and the + client cert source to use. + + Raises: + google.auth.exceptions.MutualTLSChannelError: If any errors happen. + """ + return GkeInferenceQuickstartClient.get_mtls_endpoint_and_cert_source(client_options) # type: ignore + + @property + def transport(self) -> GkeInferenceQuickstartTransport: + """Returns the transport used by the client instance. + + Returns: + GkeInferenceQuickstartTransport: The transport used by the client instance. + """ + return self._client.transport + + @property + def api_endpoint(self): + """Return the API endpoint used by the client instance. + + Returns: + str: The API endpoint used by the client instance. + """ + return self._client._api_endpoint + + @property + def universe_domain(self) -> str: + """Return the universe domain used by the client instance. + + Returns: + str: The universe domain used + by the client instance. + """ + return self._client._universe_domain + + get_transport_class = GkeInferenceQuickstartClient.get_transport_class + + def __init__( + self, + *, + credentials: Optional[ga_credentials.Credentials] = None, + transport: Optional[ + Union[ + str, + GkeInferenceQuickstartTransport, + Callable[..., GkeInferenceQuickstartTransport], + ] + ] = "grpc_asyncio", + client_options: Optional[ClientOptions] = None, + client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, + ) -> None: + """Instantiates the gke inference quickstart async client. + + Args: + credentials (Optional[google.auth.credentials.Credentials]): The + authorization credentials to attach to requests. These + credentials identify the application to the service; if none + are specified, the client will attempt to ascertain the + credentials from the environment. + transport (Optional[Union[str,GkeInferenceQuickstartTransport,Callable[..., GkeInferenceQuickstartTransport]]]): + The transport to use, or a Callable that constructs and returns a new transport to use. + If a Callable is given, it will be called with the same set of initialization + arguments as used in the GkeInferenceQuickstartTransport constructor. + If set to None, a transport is chosen automatically. + client_options (Optional[Union[google.api_core.client_options.ClientOptions, dict]]): + Custom options for the client. + + 1. The ``api_endpoint`` property can be used to override the + default endpoint provided by the client when ``transport`` is + not explicitly provided. Only if this property is not set and + ``transport`` was not explicitly provided, the endpoint is + determined by the GOOGLE_API_USE_MTLS_ENDPOINT environment + variable, which have one of the following values: + "always" (always use the default mTLS endpoint), "never" (always + use the default regular endpoint) and "auto" (auto-switch to the + default mTLS endpoint if client certificate is present; this is + the default value). + + 2. If the GOOGLE_API_USE_CLIENT_CERTIFICATE environment variable + is "true", then the ``client_cert_source`` property can be used + to provide a client certificate for mTLS transport. If + not provided, the default SSL client certificate will be used if + present. If GOOGLE_API_USE_CLIENT_CERTIFICATE is "false" or not + set, no client certificate will be used. + + 3. The ``universe_domain`` property can be used to override the + default "googleapis.com" universe. Note that ``api_endpoint`` + property still takes precedence; and ``universe_domain`` is + currently not supported for mTLS. + + client_info (google.api_core.gapic_v1.client_info.ClientInfo): + The client info used to send a user-agent string along with + API requests. If ``None``, then default info will be used. + Generally, you only need to set this if you're developing + your own client library. + + Raises: + google.auth.exceptions.MutualTlsChannelError: If mutual TLS transport + creation failed for any reason. + """ + self._client = GkeInferenceQuickstartClient( + credentials=credentials, + transport=transport, + client_options=client_options, + client_info=client_info, + ) + + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + std_logging.DEBUG + ): # pragma: NO COVER + _LOGGER.debug( + "Created client `google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient`.", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "universeDomain": getattr( + self._client._transport._credentials, "universe_domain", "" + ), + "credentialsType": f"{type(self._client._transport._credentials).__module__}.{type(self._client._transport._credentials).__qualname__}", + "credentialsInfo": getattr( + self.transport._credentials, "get_cred_info", lambda: None + )(), + } + if hasattr(self._client._transport, "_credentials") + else { + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "credentialsType": None, + }, + ) + + async def fetch_models( + self, + request: Optional[Union[gkerecommender.FetchModelsRequest, dict]] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> pagers.FetchModelsAsyncPager: + r"""Fetches available models. Open-source models follow the + Huggingface Hub ``owner/model_name`` format. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + async def sample_fetch_models(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelsRequest( + ) + + # Make the request + page_result = client.fetch_models(request=request) + + # Handle the response + async for response in page_result: + print(response) + + Args: + request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchModelsRequest, dict]]): + The request object. Request message for + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. + retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsAsyncPager: + Response message for + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. + + Iterating over this object will yield results and + resolve additional pages automatically. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchModelsRequest): + request = gkerecommender.FetchModelsRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._client._transport._wrapped_methods[ + self._client._transport.fetch_models + ] + + # Validate the universe domain. + self._client._validate_universe_domain() + + # Send the request. + response = await rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # This method is paged; wrap the response in a pager, which provides + # an `__aiter__` convenience method. + response = pagers.FetchModelsAsyncPager( + method=rpc, + request=request, + response=response, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + async def fetch_model_servers( + self, + request: Optional[Union[gkerecommender.FetchModelServersRequest, dict]] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> pagers.FetchModelServersAsyncPager: + r"""Fetches available model servers. Open-source model servers use + simplified, lowercase names (e.g., ``vllm``). + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + async def sample_fetch_model_servers(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelServersRequest( + model="model_value", + ) + + # Make the request + page_result = client.fetch_model_servers(request=request) + + # Handle the response + async for response in page_result: + print(response) + + Args: + request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchModelServersRequest, dict]]): + The request object. Request message for + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. + retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersAsyncPager: + Response message for + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. + + Iterating over this object will yield results and + resolve additional pages automatically. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchModelServersRequest): + request = gkerecommender.FetchModelServersRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._client._transport._wrapped_methods[ + self._client._transport.fetch_model_servers + ] + + # Validate the universe domain. + self._client._validate_universe_domain() + + # Send the request. + response = await rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # This method is paged; wrap the response in a pager, which provides + # an `__aiter__` convenience method. + response = pagers.FetchModelServersAsyncPager( + method=rpc, + request=request, + response=response, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + async def fetch_model_server_versions( + self, + request: Optional[ + Union[gkerecommender.FetchModelServerVersionsRequest, dict] + ] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> pagers.FetchModelServerVersionsAsyncPager: + r"""Fetches available model server versions. Open-source servers use + their own versioning schemas (e.g., ``vllm`` uses semver like + ``v1.0.0``). + + Some model servers have different versioning schemas depending + on the accelerator. For example, ``vllm`` uses semver on GPUs, + but returns nightly build tags on TPUs. All available versions + will be returned when different schemas are present. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + async def sample_fetch_model_server_versions(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelServerVersionsRequest( + model="model_value", + model_server="model_server_value", + ) + + # Make the request + page_result = client.fetch_model_server_versions(request=request) + + # Handle the response + async for response in page_result: + print(response) + + Args: + request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest, dict]]): + The request object. Request message for + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. + retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsAsyncPager: + Response message for + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. + + Iterating over this object will yield results and + resolve additional pages automatically. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchModelServerVersionsRequest): + request = gkerecommender.FetchModelServerVersionsRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._client._transport._wrapped_methods[ + self._client._transport.fetch_model_server_versions + ] + + # Validate the universe domain. + self._client._validate_universe_domain() + + # Send the request. + response = await rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # This method is paged; wrap the response in a pager, which provides + # an `__aiter__` convenience method. + response = pagers.FetchModelServerVersionsAsyncPager( + method=rpc, + request=request, + response=response, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + async def fetch_profiles( + self, + request: Optional[Union[gkerecommender.FetchProfilesRequest, dict]] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> pagers.FetchProfilesAsyncPager: + r"""Fetches available profiles. A profile contains performance + metrics and cost information for a specific model server setup. + Profiles can be filtered by parameters. If no filters are + provided, all profiles are returned. + + Profiles display a single value per performance metric based on + the provided performance requirements. If no requirements are + given, the metrics represent the inflection point. See `Run best + practice inference with GKE Inference Quickstart + recipes `__ + for details. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + async def sample_fetch_profiles(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchProfilesRequest( + ) + + # Make the request + page_result = client.fetch_profiles(request=request) + + # Handle the response + async for response in page_result: + print(response) + + Args: + request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchProfilesRequest, dict]]): + The request object. Request message for + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesAsyncPager: + Response message for + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + + Iterating over this object will yield results and + resolve additional pages automatically. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchProfilesRequest): + request = gkerecommender.FetchProfilesRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._client._transport._wrapped_methods[ + self._client._transport.fetch_profiles + ] + + # Validate the universe domain. + self._client._validate_universe_domain() + + # Send the request. + response = await rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # This method is paged; wrap the response in a pager, which provides + # an `__aiter__` convenience method. + response = pagers.FetchProfilesAsyncPager( + method=rpc, + request=request, + response=response, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + async def generate_optimized_manifest( + self, + request: Optional[ + Union[gkerecommender.GenerateOptimizedManifestRequest, dict] + ] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.GenerateOptimizedManifestResponse: + r"""Generates an optimized deployment manifest for a given model and + model server, based on the specified accelerator, performance + targets, and configurations. See `Run best practice inference + with GKE Inference Quickstart + recipes `__ + for deployment details. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + async def sample_generate_optimized_manifest(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + model_server_info = gkerecommender_v1.ModelServerInfo() + model_server_info.model = "model_value" + model_server_info.model_server = "model_server_value" + + request = gkerecommender_v1.GenerateOptimizedManifestRequest( + model_server_info=model_server_info, + accelerator_type="accelerator_type_value", + ) + + # Make the request + response = await client.generate_optimized_manifest(request=request) + + # Handle the response + print(response) + + Args: + request (Optional[Union[google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest, dict]]): + The request object. Request message for + [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. + retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse: + Response message for + [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.GenerateOptimizedManifestRequest): + request = gkerecommender.GenerateOptimizedManifestRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._client._transport._wrapped_methods[ + self._client._transport.generate_optimized_manifest + ] + + # Validate the universe domain. + self._client._validate_universe_domain() + + # Send the request. + response = await rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + async def fetch_benchmarking_data( + self, + request: Optional[ + Union[gkerecommender.FetchBenchmarkingDataRequest, dict] + ] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.FetchBenchmarkingDataResponse: + r"""Fetches all of the benchmarking data available for a + profile. Benchmarking data returns all of the + performance metrics available for a given model server + setup on a given instance type. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + async def sample_fetch_benchmarking_data(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + model_server_info = gkerecommender_v1.ModelServerInfo() + model_server_info.model = "model_value" + model_server_info.model_server = "model_server_value" + + request = gkerecommender_v1.FetchBenchmarkingDataRequest( + model_server_info=model_server_info, + ) + + # Make the request + response = await client.fetch_benchmarking_data(request=request) + + # Handle the response + print(response) + + Args: + request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest, dict]]): + The request object. Request message for + [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. + retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse: + Response message for + [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchBenchmarkingDataRequest): + request = gkerecommender.FetchBenchmarkingDataRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._client._transport._wrapped_methods[ + self._client._transport.fetch_benchmarking_data + ] + + # Validate the universe domain. + self._client._validate_universe_domain() + + # Send the request. + response = await rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + async def __aenter__(self) -> "GkeInferenceQuickstartAsyncClient": + return self + + async def __aexit__(self, exc_type, exc, tb): + await self.transport.close() + + +DEFAULT_CLIENT_INFO = gapic_v1.client_info.ClientInfo( + gapic_version=package_version.__version__ +) + +if hasattr(DEFAULT_CLIENT_INFO, "protobuf_runtime_version"): # pragma: NO COVER + DEFAULT_CLIENT_INFO.protobuf_runtime_version = google.protobuf.__version__ + + +__all__ = ("GkeInferenceQuickstartAsyncClient",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/client.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/client.py new file mode 100644 index 000000000000..cb2284d0744c --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/client.py @@ -0,0 +1,1296 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from collections import OrderedDict +from http import HTTPStatus +import json +import logging as std_logging +import os +import re +from typing import ( + Callable, + Dict, + Mapping, + MutableMapping, + MutableSequence, + Optional, + Sequence, + Tuple, + Type, + Union, + cast, +) +import warnings + +from google.api_core import client_options as client_options_lib +from google.api_core import exceptions as core_exceptions +from google.api_core import gapic_v1 +from google.api_core import retry as retries +from google.auth import credentials as ga_credentials # type: ignore +from google.auth.exceptions import MutualTLSChannelError # type: ignore +from google.auth.transport import mtls # type: ignore +from google.auth.transport.grpc import SslCredentials # type: ignore +from google.oauth2 import service_account # type: ignore +import google.protobuf + +from google.cloud.gkerecommender_v1 import gapic_version as package_version + +try: + OptionalRetry = Union[retries.Retry, gapic_v1.method._MethodDefault, None] +except AttributeError: # pragma: NO COVER + OptionalRetry = Union[retries.Retry, object, None] # type: ignore + +try: + from google.api_core import client_logging # type: ignore + + CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER +except ImportError: # pragma: NO COVER + CLIENT_LOGGING_SUPPORTED = False + +_LOGGER = std_logging.getLogger(__name__) + +from google.cloud.gkerecommender_v1.services.gke_inference_quickstart import pagers +from google.cloud.gkerecommender_v1.types import gkerecommender + +from .transports.base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport +from .transports.grpc import GkeInferenceQuickstartGrpcTransport +from .transports.grpc_asyncio import GkeInferenceQuickstartGrpcAsyncIOTransport +from .transports.rest import GkeInferenceQuickstartRestTransport + + +class GkeInferenceQuickstartClientMeta(type): + """Metaclass for the GkeInferenceQuickstart client. + + This provides class-level methods for building and retrieving + support objects (e.g. transport) without polluting the client instance + objects. + """ + + _transport_registry = ( + OrderedDict() + ) # type: Dict[str, Type[GkeInferenceQuickstartTransport]] + _transport_registry["grpc"] = GkeInferenceQuickstartGrpcTransport + _transport_registry["grpc_asyncio"] = GkeInferenceQuickstartGrpcAsyncIOTransport + _transport_registry["rest"] = GkeInferenceQuickstartRestTransport + + def get_transport_class( + cls, + label: Optional[str] = None, + ) -> Type[GkeInferenceQuickstartTransport]: + """Returns an appropriate transport class. + + Args: + label: The name of the desired transport. If none is + provided, then the first transport in the registry is used. + + Returns: + The transport class to use. + """ + # If a specific transport is requested, return that one. + if label: + return cls._transport_registry[label] + + # No transport is requested; return the default (that is, the first one + # in the dictionary). + return next(iter(cls._transport_registry.values())) + + +class GkeInferenceQuickstartClient(metaclass=GkeInferenceQuickstartClientMeta): + """GKE Inference Quickstart (GIQ) service provides profiles with + performance metrics for popular models and model servers across + multiple accelerators. These profiles help generate optimized + best practices for running inference on GKE. + """ + + @staticmethod + def _get_default_mtls_endpoint(api_endpoint): + """Converts api endpoint to mTLS endpoint. + + Convert "*.sandbox.googleapis.com" and "*.googleapis.com" to + "*.mtls.sandbox.googleapis.com" and "*.mtls.googleapis.com" respectively. + Args: + api_endpoint (Optional[str]): the api endpoint to convert. + Returns: + str: converted mTLS api endpoint. + """ + if not api_endpoint: + return api_endpoint + + mtls_endpoint_re = re.compile( + r"(?P[^.]+)(?P\.mtls)?(?P\.sandbox)?(?P\.googleapis\.com)?" + ) + + m = mtls_endpoint_re.match(api_endpoint) + name, mtls, sandbox, googledomain = m.groups() + if mtls or not googledomain: + return api_endpoint + + if sandbox: + return api_endpoint.replace( + "sandbox.googleapis.com", "mtls.sandbox.googleapis.com" + ) + + return api_endpoint.replace(".googleapis.com", ".mtls.googleapis.com") + + # Note: DEFAULT_ENDPOINT is deprecated. Use _DEFAULT_ENDPOINT_TEMPLATE instead. + DEFAULT_ENDPOINT = "gkerecommender.googleapis.com" + DEFAULT_MTLS_ENDPOINT = _get_default_mtls_endpoint.__func__( # type: ignore + DEFAULT_ENDPOINT + ) + + _DEFAULT_ENDPOINT_TEMPLATE = "gkerecommender.{UNIVERSE_DOMAIN}" + _DEFAULT_UNIVERSE = "googleapis.com" + + @classmethod + def from_service_account_info(cls, info: dict, *args, **kwargs): + """Creates an instance of this client using the provided credentials + info. + + Args: + info (dict): The service account private key info. + args: Additional arguments to pass to the constructor. + kwargs: Additional arguments to pass to the constructor. + + Returns: + GkeInferenceQuickstartClient: The constructed client. + """ + credentials = service_account.Credentials.from_service_account_info(info) + kwargs["credentials"] = credentials + return cls(*args, **kwargs) + + @classmethod + def from_service_account_file(cls, filename: str, *args, **kwargs): + """Creates an instance of this client using the provided credentials + file. + + Args: + filename (str): The path to the service account private key json + file. + args: Additional arguments to pass to the constructor. + kwargs: Additional arguments to pass to the constructor. + + Returns: + GkeInferenceQuickstartClient: The constructed client. + """ + credentials = service_account.Credentials.from_service_account_file(filename) + kwargs["credentials"] = credentials + return cls(*args, **kwargs) + + from_service_account_json = from_service_account_file + + @property + def transport(self) -> GkeInferenceQuickstartTransport: + """Returns the transport used by the client instance. + + Returns: + GkeInferenceQuickstartTransport: The transport used by the client + instance. + """ + return self._transport + + @staticmethod + def common_billing_account_path( + billing_account: str, + ) -> str: + """Returns a fully-qualified billing_account string.""" + return "billingAccounts/{billing_account}".format( + billing_account=billing_account, + ) + + @staticmethod + def parse_common_billing_account_path(path: str) -> Dict[str, str]: + """Parse a billing_account path into its component segments.""" + m = re.match(r"^billingAccounts/(?P.+?)$", path) + return m.groupdict() if m else {} + + @staticmethod + def common_folder_path( + folder: str, + ) -> str: + """Returns a fully-qualified folder string.""" + return "folders/{folder}".format( + folder=folder, + ) + + @staticmethod + def parse_common_folder_path(path: str) -> Dict[str, str]: + """Parse a folder path into its component segments.""" + m = re.match(r"^folders/(?P.+?)$", path) + return m.groupdict() if m else {} + + @staticmethod + def common_organization_path( + organization: str, + ) -> str: + """Returns a fully-qualified organization string.""" + return "organizations/{organization}".format( + organization=organization, + ) + + @staticmethod + def parse_common_organization_path(path: str) -> Dict[str, str]: + """Parse a organization path into its component segments.""" + m = re.match(r"^organizations/(?P.+?)$", path) + return m.groupdict() if m else {} + + @staticmethod + def common_project_path( + project: str, + ) -> str: + """Returns a fully-qualified project string.""" + return "projects/{project}".format( + project=project, + ) + + @staticmethod + def parse_common_project_path(path: str) -> Dict[str, str]: + """Parse a project path into its component segments.""" + m = re.match(r"^projects/(?P.+?)$", path) + return m.groupdict() if m else {} + + @staticmethod + def common_location_path( + project: str, + location: str, + ) -> str: + """Returns a fully-qualified location string.""" + return "projects/{project}/locations/{location}".format( + project=project, + location=location, + ) + + @staticmethod + def parse_common_location_path(path: str) -> Dict[str, str]: + """Parse a location path into its component segments.""" + m = re.match(r"^projects/(?P.+?)/locations/(?P.+?)$", path) + return m.groupdict() if m else {} + + @classmethod + def get_mtls_endpoint_and_cert_source( + cls, client_options: Optional[client_options_lib.ClientOptions] = None + ): + """Deprecated. Return the API endpoint and client cert source for mutual TLS. + + The client cert source is determined in the following order: + (1) if `GOOGLE_API_USE_CLIENT_CERTIFICATE` environment variable is not "true", the + client cert source is None. + (2) if `client_options.client_cert_source` is provided, use the provided one; if the + default client cert source exists, use the default one; otherwise the client cert + source is None. + + The API endpoint is determined in the following order: + (1) if `client_options.api_endpoint` if provided, use the provided one. + (2) if `GOOGLE_API_USE_CLIENT_CERTIFICATE` environment variable is "always", use the + default mTLS endpoint; if the environment variable is "never", use the default API + endpoint; otherwise if client cert source exists, use the default mTLS endpoint, otherwise + use the default API endpoint. + + More details can be found at https://google.aip.dev/auth/4114. + + Args: + client_options (google.api_core.client_options.ClientOptions): Custom options for the + client. Only the `api_endpoint` and `client_cert_source` properties may be used + in this method. + + Returns: + Tuple[str, Callable[[], Tuple[bytes, bytes]]]: returns the API endpoint and the + client cert source to use. + + Raises: + google.auth.exceptions.MutualTLSChannelError: If any errors happen. + """ + + warnings.warn( + "get_mtls_endpoint_and_cert_source is deprecated. Use the api_endpoint property instead.", + DeprecationWarning, + ) + if client_options is None: + client_options = client_options_lib.ClientOptions() + use_client_cert = os.getenv("GOOGLE_API_USE_CLIENT_CERTIFICATE", "false") + use_mtls_endpoint = os.getenv("GOOGLE_API_USE_MTLS_ENDPOINT", "auto") + if use_client_cert not in ("true", "false"): + raise ValueError( + "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" + ) + if use_mtls_endpoint not in ("auto", "never", "always"): + raise MutualTLSChannelError( + "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" + ) + + # Figure out the client cert source to use. + client_cert_source = None + if use_client_cert == "true": + if client_options.client_cert_source: + client_cert_source = client_options.client_cert_source + elif mtls.has_default_client_cert_source(): + client_cert_source = mtls.default_client_cert_source() + + # Figure out which api endpoint to use. + if client_options.api_endpoint is not None: + api_endpoint = client_options.api_endpoint + elif use_mtls_endpoint == "always" or ( + use_mtls_endpoint == "auto" and client_cert_source + ): + api_endpoint = cls.DEFAULT_MTLS_ENDPOINT + else: + api_endpoint = cls.DEFAULT_ENDPOINT + + return api_endpoint, client_cert_source + + @staticmethod + def _read_environment_variables(): + """Returns the environment variables used by the client. + + Returns: + Tuple[bool, str, str]: returns the GOOGLE_API_USE_CLIENT_CERTIFICATE, + GOOGLE_API_USE_MTLS_ENDPOINT, and GOOGLE_CLOUD_UNIVERSE_DOMAIN environment variables. + + Raises: + ValueError: If GOOGLE_API_USE_CLIENT_CERTIFICATE is not + any of ["true", "false"]. + google.auth.exceptions.MutualTLSChannelError: If GOOGLE_API_USE_MTLS_ENDPOINT + is not any of ["auto", "never", "always"]. + """ + use_client_cert = os.getenv( + "GOOGLE_API_USE_CLIENT_CERTIFICATE", "false" + ).lower() + use_mtls_endpoint = os.getenv("GOOGLE_API_USE_MTLS_ENDPOINT", "auto").lower() + universe_domain_env = os.getenv("GOOGLE_CLOUD_UNIVERSE_DOMAIN") + if use_client_cert not in ("true", "false"): + raise ValueError( + "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" + ) + if use_mtls_endpoint not in ("auto", "never", "always"): + raise MutualTLSChannelError( + "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" + ) + return use_client_cert == "true", use_mtls_endpoint, universe_domain_env + + @staticmethod + def _get_client_cert_source(provided_cert_source, use_cert_flag): + """Return the client cert source to be used by the client. + + Args: + provided_cert_source (bytes): The client certificate source provided. + use_cert_flag (bool): A flag indicating whether to use the client certificate. + + Returns: + bytes or None: The client cert source to be used by the client. + """ + client_cert_source = None + if use_cert_flag: + if provided_cert_source: + client_cert_source = provided_cert_source + elif mtls.has_default_client_cert_source(): + client_cert_source = mtls.default_client_cert_source() + return client_cert_source + + @staticmethod + def _get_api_endpoint( + api_override, client_cert_source, universe_domain, use_mtls_endpoint + ): + """Return the API endpoint used by the client. + + Args: + api_override (str): The API endpoint override. If specified, this is always + the return value of this function and the other arguments are not used. + client_cert_source (bytes): The client certificate source used by the client. + universe_domain (str): The universe domain used by the client. + use_mtls_endpoint (str): How to use the mTLS endpoint, which depends also on the other parameters. + Possible values are "always", "auto", or "never". + + Returns: + str: The API endpoint to be used by the client. + """ + if api_override is not None: + api_endpoint = api_override + elif use_mtls_endpoint == "always" or ( + use_mtls_endpoint == "auto" and client_cert_source + ): + _default_universe = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE + if universe_domain != _default_universe: + raise MutualTLSChannelError( + f"mTLS is not supported in any universe other than {_default_universe}." + ) + api_endpoint = GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT + else: + api_endpoint = ( + GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=universe_domain + ) + ) + return api_endpoint + + @staticmethod + def _get_universe_domain( + client_universe_domain: Optional[str], universe_domain_env: Optional[str] + ) -> str: + """Return the universe domain used by the client. + + Args: + client_universe_domain (Optional[str]): The universe domain configured via the client options. + universe_domain_env (Optional[str]): The universe domain configured via the "GOOGLE_CLOUD_UNIVERSE_DOMAIN" environment variable. + + Returns: + str: The universe domain to be used by the client. + + Raises: + ValueError: If the universe domain is an empty string. + """ + universe_domain = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE + if client_universe_domain is not None: + universe_domain = client_universe_domain + elif universe_domain_env is not None: + universe_domain = universe_domain_env + if len(universe_domain.strip()) == 0: + raise ValueError("Universe Domain cannot be an empty string.") + return universe_domain + + def _validate_universe_domain(self): + """Validates client's and credentials' universe domains are consistent. + + Returns: + bool: True iff the configured universe domain is valid. + + Raises: + ValueError: If the configured universe domain is not valid. + """ + + # NOTE (b/349488459): universe validation is disabled until further notice. + return True + + def _add_cred_info_for_auth_errors( + self, error: core_exceptions.GoogleAPICallError + ) -> None: + """Adds credential info string to error details for 401/403/404 errors. + + Args: + error (google.api_core.exceptions.GoogleAPICallError): The error to add the cred info. + """ + if error.code not in [ + HTTPStatus.UNAUTHORIZED, + HTTPStatus.FORBIDDEN, + HTTPStatus.NOT_FOUND, + ]: + return + + cred = self._transport._credentials + + # get_cred_info is only available in google-auth>=2.35.0 + if not hasattr(cred, "get_cred_info"): + return + + # ignore the type check since pypy test fails when get_cred_info + # is not available + cred_info = cred.get_cred_info() # type: ignore + if cred_info and hasattr(error._details, "append"): + error._details.append(json.dumps(cred_info)) + + @property + def api_endpoint(self): + """Return the API endpoint used by the client instance. + + Returns: + str: The API endpoint used by the client instance. + """ + return self._api_endpoint + + @property + def universe_domain(self) -> str: + """Return the universe domain used by the client instance. + + Returns: + str: The universe domain used by the client instance. + """ + return self._universe_domain + + def __init__( + self, + *, + credentials: Optional[ga_credentials.Credentials] = None, + transport: Optional[ + Union[ + str, + GkeInferenceQuickstartTransport, + Callable[..., GkeInferenceQuickstartTransport], + ] + ] = None, + client_options: Optional[Union[client_options_lib.ClientOptions, dict]] = None, + client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, + ) -> None: + """Instantiates the gke inference quickstart client. + + Args: + credentials (Optional[google.auth.credentials.Credentials]): The + authorization credentials to attach to requests. These + credentials identify the application to the service; if none + are specified, the client will attempt to ascertain the + credentials from the environment. + transport (Optional[Union[str,GkeInferenceQuickstartTransport,Callable[..., GkeInferenceQuickstartTransport]]]): + The transport to use, or a Callable that constructs and returns a new transport. + If a Callable is given, it will be called with the same set of initialization + arguments as used in the GkeInferenceQuickstartTransport constructor. + If set to None, a transport is chosen automatically. + client_options (Optional[Union[google.api_core.client_options.ClientOptions, dict]]): + Custom options for the client. + + 1. The ``api_endpoint`` property can be used to override the + default endpoint provided by the client when ``transport`` is + not explicitly provided. Only if this property is not set and + ``transport`` was not explicitly provided, the endpoint is + determined by the GOOGLE_API_USE_MTLS_ENDPOINT environment + variable, which have one of the following values: + "always" (always use the default mTLS endpoint), "never" (always + use the default regular endpoint) and "auto" (auto-switch to the + default mTLS endpoint if client certificate is present; this is + the default value). + + 2. If the GOOGLE_API_USE_CLIENT_CERTIFICATE environment variable + is "true", then the ``client_cert_source`` property can be used + to provide a client certificate for mTLS transport. If + not provided, the default SSL client certificate will be used if + present. If GOOGLE_API_USE_CLIENT_CERTIFICATE is "false" or not + set, no client certificate will be used. + + 3. The ``universe_domain`` property can be used to override the + default "googleapis.com" universe. Note that the ``api_endpoint`` + property still takes precedence; and ``universe_domain`` is + currently not supported for mTLS. + + client_info (google.api_core.gapic_v1.client_info.ClientInfo): + The client info used to send a user-agent string along with + API requests. If ``None``, then default info will be used. + Generally, you only need to set this if you're developing + your own client library. + + Raises: + google.auth.exceptions.MutualTLSChannelError: If mutual TLS transport + creation failed for any reason. + """ + self._client_options = client_options + if isinstance(self._client_options, dict): + self._client_options = client_options_lib.from_dict(self._client_options) + if self._client_options is None: + self._client_options = client_options_lib.ClientOptions() + self._client_options = cast( + client_options_lib.ClientOptions, self._client_options + ) + + universe_domain_opt = getattr(self._client_options, "universe_domain", None) + + ( + self._use_client_cert, + self._use_mtls_endpoint, + self._universe_domain_env, + ) = GkeInferenceQuickstartClient._read_environment_variables() + self._client_cert_source = GkeInferenceQuickstartClient._get_client_cert_source( + self._client_options.client_cert_source, self._use_client_cert + ) + self._universe_domain = GkeInferenceQuickstartClient._get_universe_domain( + universe_domain_opt, self._universe_domain_env + ) + self._api_endpoint = None # updated below, depending on `transport` + + # Initialize the universe domain validation. + self._is_universe_domain_valid = False + + if CLIENT_LOGGING_SUPPORTED: # pragma: NO COVER + # Setup logging. + client_logging.initialize_logging() + + api_key_value = getattr(self._client_options, "api_key", None) + if api_key_value and credentials: + raise ValueError( + "client_options.api_key and credentials are mutually exclusive" + ) + + # Save or instantiate the transport. + # Ordinarily, we provide the transport, but allowing a custom transport + # instance provides an extensibility point for unusual situations. + transport_provided = isinstance(transport, GkeInferenceQuickstartTransport) + if transport_provided: + # transport is a GkeInferenceQuickstartTransport instance. + if credentials or self._client_options.credentials_file or api_key_value: + raise ValueError( + "When providing a transport instance, " + "provide its credentials directly." + ) + if self._client_options.scopes: + raise ValueError( + "When providing a transport instance, provide its scopes " + "directly." + ) + self._transport = cast(GkeInferenceQuickstartTransport, transport) + self._api_endpoint = self._transport.host + + self._api_endpoint = ( + self._api_endpoint + or GkeInferenceQuickstartClient._get_api_endpoint( + self._client_options.api_endpoint, + self._client_cert_source, + self._universe_domain, + self._use_mtls_endpoint, + ) + ) + + if not transport_provided: + import google.auth._default # type: ignore + + if api_key_value and hasattr( + google.auth._default, "get_api_key_credentials" + ): + credentials = google.auth._default.get_api_key_credentials( + api_key_value + ) + + transport_init: Union[ + Type[GkeInferenceQuickstartTransport], + Callable[..., GkeInferenceQuickstartTransport], + ] = ( + GkeInferenceQuickstartClient.get_transport_class(transport) + if isinstance(transport, str) or transport is None + else cast(Callable[..., GkeInferenceQuickstartTransport], transport) + ) + # initialize with the provided callable or the passed in class + self._transport = transport_init( + credentials=credentials, + credentials_file=self._client_options.credentials_file, + host=self._api_endpoint, + scopes=self._client_options.scopes, + client_cert_source_for_mtls=self._client_cert_source, + quota_project_id=self._client_options.quota_project_id, + client_info=client_info, + always_use_jwt_access=True, + api_audience=self._client_options.api_audience, + ) + + if "async" not in str(self._transport): + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + std_logging.DEBUG + ): # pragma: NO COVER + _LOGGER.debug( + "Created client `google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient`.", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "universeDomain": getattr( + self._transport._credentials, "universe_domain", "" + ), + "credentialsType": f"{type(self._transport._credentials).__module__}.{type(self._transport._credentials).__qualname__}", + "credentialsInfo": getattr( + self.transport._credentials, "get_cred_info", lambda: None + )(), + } + if hasattr(self._transport, "_credentials") + else { + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "credentialsType": None, + }, + ) + + def fetch_models( + self, + request: Optional[Union[gkerecommender.FetchModelsRequest, dict]] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> pagers.FetchModelsPager: + r"""Fetches available models. Open-source models follow the + Huggingface Hub ``owner/model_name`` format. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + def sample_fetch_models(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelsRequest( + ) + + # Make the request + page_result = client.fetch_models(request=request) + + # Handle the response + for response in page_result: + print(response) + + Args: + request (Union[google.cloud.gkerecommender_v1.types.FetchModelsRequest, dict]): + The request object. Request message for + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsPager: + Response message for + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. + + Iterating over this object will yield results and + resolve additional pages automatically. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchModelsRequest): + request = gkerecommender.FetchModelsRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._transport._wrapped_methods[self._transport.fetch_models] + + # Validate the universe domain. + self._validate_universe_domain() + + # Send the request. + response = rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # This method is paged; wrap the response in a pager, which provides + # an `__iter__` convenience method. + response = pagers.FetchModelsPager( + method=rpc, + request=request, + response=response, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + def fetch_model_servers( + self, + request: Optional[Union[gkerecommender.FetchModelServersRequest, dict]] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> pagers.FetchModelServersPager: + r"""Fetches available model servers. Open-source model servers use + simplified, lowercase names (e.g., ``vllm``). + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + def sample_fetch_model_servers(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelServersRequest( + model="model_value", + ) + + # Make the request + page_result = client.fetch_model_servers(request=request) + + # Handle the response + for response in page_result: + print(response) + + Args: + request (Union[google.cloud.gkerecommender_v1.types.FetchModelServersRequest, dict]): + The request object. Request message for + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersPager: + Response message for + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. + + Iterating over this object will yield results and + resolve additional pages automatically. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchModelServersRequest): + request = gkerecommender.FetchModelServersRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._transport._wrapped_methods[self._transport.fetch_model_servers] + + # Validate the universe domain. + self._validate_universe_domain() + + # Send the request. + response = rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # This method is paged; wrap the response in a pager, which provides + # an `__iter__` convenience method. + response = pagers.FetchModelServersPager( + method=rpc, + request=request, + response=response, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + def fetch_model_server_versions( + self, + request: Optional[ + Union[gkerecommender.FetchModelServerVersionsRequest, dict] + ] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> pagers.FetchModelServerVersionsPager: + r"""Fetches available model server versions. Open-source servers use + their own versioning schemas (e.g., ``vllm`` uses semver like + ``v1.0.0``). + + Some model servers have different versioning schemas depending + on the accelerator. For example, ``vllm`` uses semver on GPUs, + but returns nightly build tags on TPUs. All available versions + will be returned when different schemas are present. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + def sample_fetch_model_server_versions(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelServerVersionsRequest( + model="model_value", + model_server="model_server_value", + ) + + # Make the request + page_result = client.fetch_model_server_versions(request=request) + + # Handle the response + for response in page_result: + print(response) + + Args: + request (Union[google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest, dict]): + The request object. Request message for + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsPager: + Response message for + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. + + Iterating over this object will yield results and + resolve additional pages automatically. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchModelServerVersionsRequest): + request = gkerecommender.FetchModelServerVersionsRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._transport._wrapped_methods[ + self._transport.fetch_model_server_versions + ] + + # Validate the universe domain. + self._validate_universe_domain() + + # Send the request. + response = rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # This method is paged; wrap the response in a pager, which provides + # an `__iter__` convenience method. + response = pagers.FetchModelServerVersionsPager( + method=rpc, + request=request, + response=response, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + def fetch_profiles( + self, + request: Optional[Union[gkerecommender.FetchProfilesRequest, dict]] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> pagers.FetchProfilesPager: + r"""Fetches available profiles. A profile contains performance + metrics and cost information for a specific model server setup. + Profiles can be filtered by parameters. If no filters are + provided, all profiles are returned. + + Profiles display a single value per performance metric based on + the provided performance requirements. If no requirements are + given, the metrics represent the inflection point. See `Run best + practice inference with GKE Inference Quickstart + recipes `__ + for details. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + def sample_fetch_profiles(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchProfilesRequest( + ) + + # Make the request + page_result = client.fetch_profiles(request=request) + + # Handle the response + for response in page_result: + print(response) + + Args: + request (Union[google.cloud.gkerecommender_v1.types.FetchProfilesRequest, dict]): + The request object. Request message for + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesPager: + Response message for + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + + Iterating over this object will yield results and + resolve additional pages automatically. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchProfilesRequest): + request = gkerecommender.FetchProfilesRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._transport._wrapped_methods[self._transport.fetch_profiles] + + # Validate the universe domain. + self._validate_universe_domain() + + # Send the request. + response = rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # This method is paged; wrap the response in a pager, which provides + # an `__iter__` convenience method. + response = pagers.FetchProfilesPager( + method=rpc, + request=request, + response=response, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + def generate_optimized_manifest( + self, + request: Optional[ + Union[gkerecommender.GenerateOptimizedManifestRequest, dict] + ] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.GenerateOptimizedManifestResponse: + r"""Generates an optimized deployment manifest for a given model and + model server, based on the specified accelerator, performance + targets, and configurations. See `Run best practice inference + with GKE Inference Quickstart + recipes `__ + for deployment details. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + def sample_generate_optimized_manifest(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + model_server_info = gkerecommender_v1.ModelServerInfo() + model_server_info.model = "model_value" + model_server_info.model_server = "model_server_value" + + request = gkerecommender_v1.GenerateOptimizedManifestRequest( + model_server_info=model_server_info, + accelerator_type="accelerator_type_value", + ) + + # Make the request + response = client.generate_optimized_manifest(request=request) + + # Handle the response + print(response) + + Args: + request (Union[google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest, dict]): + The request object. Request message for + [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse: + Response message for + [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.GenerateOptimizedManifestRequest): + request = gkerecommender.GenerateOptimizedManifestRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._transport._wrapped_methods[ + self._transport.generate_optimized_manifest + ] + + # Validate the universe domain. + self._validate_universe_domain() + + # Send the request. + response = rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + def fetch_benchmarking_data( + self, + request: Optional[ + Union[gkerecommender.FetchBenchmarkingDataRequest, dict] + ] = None, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.FetchBenchmarkingDataResponse: + r"""Fetches all of the benchmarking data available for a + profile. Benchmarking data returns all of the + performance metrics available for a given model server + setup on a given instance type. + + .. code-block:: python + + # This snippet has been automatically generated and should be regarded as a + # code template only. + # It will require modifications to work: + # - It may require correct/in-range values for request initialization. + # - It may require specifying regional endpoints when creating the service + # client as shown in: + # https://googleapis.dev/python/google-api-core/latest/client_options.html + from google.cloud import gkerecommender_v1 + + def sample_fetch_benchmarking_data(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + model_server_info = gkerecommender_v1.ModelServerInfo() + model_server_info.model = "model_value" + model_server_info.model_server = "model_server_value" + + request = gkerecommender_v1.FetchBenchmarkingDataRequest( + model_server_info=model_server_info, + ) + + # Make the request + response = client.fetch_benchmarking_data(request=request) + + # Handle the response + print(response) + + Args: + request (Union[google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest, dict]): + The request object. Request message for + [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse: + Response message for + [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. + + """ + # Create or coerce a protobuf request object. + # - Use the request object if provided (there's no risk of modifying the input as + # there are no flattened fields), or create one. + if not isinstance(request, gkerecommender.FetchBenchmarkingDataRequest): + request = gkerecommender.FetchBenchmarkingDataRequest(request) + + # Wrap the RPC method; this adds retry and timeout information, + # and friendly error handling. + rpc = self._transport._wrapped_methods[self._transport.fetch_benchmarking_data] + + # Validate the universe domain. + self._validate_universe_domain() + + # Send the request. + response = rpc( + request, + retry=retry, + timeout=timeout, + metadata=metadata, + ) + + # Done; return the response. + return response + + def __enter__(self) -> "GkeInferenceQuickstartClient": + return self + + def __exit__(self, type, value, traceback): + """Releases underlying transport's resources. + + .. warning:: + ONLY use as a context manager if the transport is NOT shared + with other clients! Exiting the with block will CLOSE the transport + and may cause errors in other clients! + """ + self.transport.close() + + +DEFAULT_CLIENT_INFO = gapic_v1.client_info.ClientInfo( + gapic_version=package_version.__version__ +) + +if hasattr(DEFAULT_CLIENT_INFO, "protobuf_runtime_version"): # pragma: NO COVER + DEFAULT_CLIENT_INFO.protobuf_runtime_version = google.protobuf.__version__ + +__all__ = ("GkeInferenceQuickstartClient",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/pagers.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/pagers.py new file mode 100644 index 000000000000..6a59cefc936c --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/pagers.py @@ -0,0 +1,669 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from typing import ( + Any, + AsyncIterator, + Awaitable, + Callable, + Iterator, + Optional, + Sequence, + Tuple, + Union, +) + +from google.api_core import gapic_v1 +from google.api_core import retry as retries +from google.api_core import retry_async as retries_async + +try: + OptionalRetry = Union[retries.Retry, gapic_v1.method._MethodDefault, None] + OptionalAsyncRetry = Union[ + retries_async.AsyncRetry, gapic_v1.method._MethodDefault, None + ] +except AttributeError: # pragma: NO COVER + OptionalRetry = Union[retries.Retry, object, None] # type: ignore + OptionalAsyncRetry = Union[retries_async.AsyncRetry, object, None] # type: ignore + +from google.cloud.gkerecommender_v1.types import gkerecommender + + +class FetchModelsPager: + """A pager for iterating through ``fetch_models`` requests. + + This class thinly wraps an initial + :class:`google.cloud.gkerecommender_v1.types.FetchModelsResponse` object, and + provides an ``__iter__`` method to iterate through its + ``models`` field. + + If there are more pages, the ``__iter__`` method will make additional + ``FetchModels`` requests and continue to iterate + through the ``models`` field on the + corresponding responses. + + All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelsResponse` + attributes are available on the pager. If multiple requests are made, only + the most recent response is retained, and thus used for attribute lookup. + """ + + def __init__( + self, + method: Callable[..., gkerecommender.FetchModelsResponse], + request: gkerecommender.FetchModelsRequest, + response: gkerecommender.FetchModelsResponse, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = () + ): + """Instantiate the pager. + + Args: + method (Callable): The method that was originally called, and + which instantiated this pager. + request (google.cloud.gkerecommender_v1.types.FetchModelsRequest): + The initial request object. + response (google.cloud.gkerecommender_v1.types.FetchModelsResponse): + The initial response object. + retry (google.api_core.retry.Retry): Designation of what errors, + if any, should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + """ + self._method = method + self._request = gkerecommender.FetchModelsRequest(request) + self._response = response + self._retry = retry + self._timeout = timeout + self._metadata = metadata + + def __getattr__(self, name: str) -> Any: + return getattr(self._response, name) + + @property + def pages(self) -> Iterator[gkerecommender.FetchModelsResponse]: + yield self._response + while self._response.next_page_token: + self._request.page_token = self._response.next_page_token + self._response = self._method( + self._request, + retry=self._retry, + timeout=self._timeout, + metadata=self._metadata, + ) + yield self._response + + def __iter__(self) -> Iterator[str]: + for page in self.pages: + yield from page.models + + def __repr__(self) -> str: + return "{0}<{1!r}>".format(self.__class__.__name__, self._response) + + +class FetchModelsAsyncPager: + """A pager for iterating through ``fetch_models`` requests. + + This class thinly wraps an initial + :class:`google.cloud.gkerecommender_v1.types.FetchModelsResponse` object, and + provides an ``__aiter__`` method to iterate through its + ``models`` field. + + If there are more pages, the ``__aiter__`` method will make additional + ``FetchModels`` requests and continue to iterate + through the ``models`` field on the + corresponding responses. + + All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelsResponse` + attributes are available on the pager. If multiple requests are made, only + the most recent response is retained, and thus used for attribute lookup. + """ + + def __init__( + self, + method: Callable[..., Awaitable[gkerecommender.FetchModelsResponse]], + request: gkerecommender.FetchModelsRequest, + response: gkerecommender.FetchModelsResponse, + *, + retry: OptionalAsyncRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = () + ): + """Instantiates the pager. + + Args: + method (Callable): The method that was originally called, and + which instantiated this pager. + request (google.cloud.gkerecommender_v1.types.FetchModelsRequest): + The initial request object. + response (google.cloud.gkerecommender_v1.types.FetchModelsResponse): + The initial response object. + retry (google.api_core.retry.AsyncRetry): Designation of what errors, + if any, should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + """ + self._method = method + self._request = gkerecommender.FetchModelsRequest(request) + self._response = response + self._retry = retry + self._timeout = timeout + self._metadata = metadata + + def __getattr__(self, name: str) -> Any: + return getattr(self._response, name) + + @property + async def pages(self) -> AsyncIterator[gkerecommender.FetchModelsResponse]: + yield self._response + while self._response.next_page_token: + self._request.page_token = self._response.next_page_token + self._response = await self._method( + self._request, + retry=self._retry, + timeout=self._timeout, + metadata=self._metadata, + ) + yield self._response + + def __aiter__(self) -> AsyncIterator[str]: + async def async_generator(): + async for page in self.pages: + for response in page.models: + yield response + + return async_generator() + + def __repr__(self) -> str: + return "{0}<{1!r}>".format(self.__class__.__name__, self._response) + + +class FetchModelServersPager: + """A pager for iterating through ``fetch_model_servers`` requests. + + This class thinly wraps an initial + :class:`google.cloud.gkerecommender_v1.types.FetchModelServersResponse` object, and + provides an ``__iter__`` method to iterate through its + ``model_servers`` field. + + If there are more pages, the ``__iter__`` method will make additional + ``FetchModelServers`` requests and continue to iterate + through the ``model_servers`` field on the + corresponding responses. + + All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelServersResponse` + attributes are available on the pager. If multiple requests are made, only + the most recent response is retained, and thus used for attribute lookup. + """ + + def __init__( + self, + method: Callable[..., gkerecommender.FetchModelServersResponse], + request: gkerecommender.FetchModelServersRequest, + response: gkerecommender.FetchModelServersResponse, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = () + ): + """Instantiate the pager. + + Args: + method (Callable): The method that was originally called, and + which instantiated this pager. + request (google.cloud.gkerecommender_v1.types.FetchModelServersRequest): + The initial request object. + response (google.cloud.gkerecommender_v1.types.FetchModelServersResponse): + The initial response object. + retry (google.api_core.retry.Retry): Designation of what errors, + if any, should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + """ + self._method = method + self._request = gkerecommender.FetchModelServersRequest(request) + self._response = response + self._retry = retry + self._timeout = timeout + self._metadata = metadata + + def __getattr__(self, name: str) -> Any: + return getattr(self._response, name) + + @property + def pages(self) -> Iterator[gkerecommender.FetchModelServersResponse]: + yield self._response + while self._response.next_page_token: + self._request.page_token = self._response.next_page_token + self._response = self._method( + self._request, + retry=self._retry, + timeout=self._timeout, + metadata=self._metadata, + ) + yield self._response + + def __iter__(self) -> Iterator[str]: + for page in self.pages: + yield from page.model_servers + + def __repr__(self) -> str: + return "{0}<{1!r}>".format(self.__class__.__name__, self._response) + + +class FetchModelServersAsyncPager: + """A pager for iterating through ``fetch_model_servers`` requests. + + This class thinly wraps an initial + :class:`google.cloud.gkerecommender_v1.types.FetchModelServersResponse` object, and + provides an ``__aiter__`` method to iterate through its + ``model_servers`` field. + + If there are more pages, the ``__aiter__`` method will make additional + ``FetchModelServers`` requests and continue to iterate + through the ``model_servers`` field on the + corresponding responses. + + All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelServersResponse` + attributes are available on the pager. If multiple requests are made, only + the most recent response is retained, and thus used for attribute lookup. + """ + + def __init__( + self, + method: Callable[..., Awaitable[gkerecommender.FetchModelServersResponse]], + request: gkerecommender.FetchModelServersRequest, + response: gkerecommender.FetchModelServersResponse, + *, + retry: OptionalAsyncRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = () + ): + """Instantiates the pager. + + Args: + method (Callable): The method that was originally called, and + which instantiated this pager. + request (google.cloud.gkerecommender_v1.types.FetchModelServersRequest): + The initial request object. + response (google.cloud.gkerecommender_v1.types.FetchModelServersResponse): + The initial response object. + retry (google.api_core.retry.AsyncRetry): Designation of what errors, + if any, should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + """ + self._method = method + self._request = gkerecommender.FetchModelServersRequest(request) + self._response = response + self._retry = retry + self._timeout = timeout + self._metadata = metadata + + def __getattr__(self, name: str) -> Any: + return getattr(self._response, name) + + @property + async def pages(self) -> AsyncIterator[gkerecommender.FetchModelServersResponse]: + yield self._response + while self._response.next_page_token: + self._request.page_token = self._response.next_page_token + self._response = await self._method( + self._request, + retry=self._retry, + timeout=self._timeout, + metadata=self._metadata, + ) + yield self._response + + def __aiter__(self) -> AsyncIterator[str]: + async def async_generator(): + async for page in self.pages: + for response in page.model_servers: + yield response + + return async_generator() + + def __repr__(self) -> str: + return "{0}<{1!r}>".format(self.__class__.__name__, self._response) + + +class FetchModelServerVersionsPager: + """A pager for iterating through ``fetch_model_server_versions`` requests. + + This class thinly wraps an initial + :class:`google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse` object, and + provides an ``__iter__`` method to iterate through its + ``model_server_versions`` field. + + If there are more pages, the ``__iter__`` method will make additional + ``FetchModelServerVersions`` requests and continue to iterate + through the ``model_server_versions`` field on the + corresponding responses. + + All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse` + attributes are available on the pager. If multiple requests are made, only + the most recent response is retained, and thus used for attribute lookup. + """ + + def __init__( + self, + method: Callable[..., gkerecommender.FetchModelServerVersionsResponse], + request: gkerecommender.FetchModelServerVersionsRequest, + response: gkerecommender.FetchModelServerVersionsResponse, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = () + ): + """Instantiate the pager. + + Args: + method (Callable): The method that was originally called, and + which instantiated this pager. + request (google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest): + The initial request object. + response (google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse): + The initial response object. + retry (google.api_core.retry.Retry): Designation of what errors, + if any, should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + """ + self._method = method + self._request = gkerecommender.FetchModelServerVersionsRequest(request) + self._response = response + self._retry = retry + self._timeout = timeout + self._metadata = metadata + + def __getattr__(self, name: str) -> Any: + return getattr(self._response, name) + + @property + def pages(self) -> Iterator[gkerecommender.FetchModelServerVersionsResponse]: + yield self._response + while self._response.next_page_token: + self._request.page_token = self._response.next_page_token + self._response = self._method( + self._request, + retry=self._retry, + timeout=self._timeout, + metadata=self._metadata, + ) + yield self._response + + def __iter__(self) -> Iterator[str]: + for page in self.pages: + yield from page.model_server_versions + + def __repr__(self) -> str: + return "{0}<{1!r}>".format(self.__class__.__name__, self._response) + + +class FetchModelServerVersionsAsyncPager: + """A pager for iterating through ``fetch_model_server_versions`` requests. + + This class thinly wraps an initial + :class:`google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse` object, and + provides an ``__aiter__`` method to iterate through its + ``model_server_versions`` field. + + If there are more pages, the ``__aiter__`` method will make additional + ``FetchModelServerVersions`` requests and continue to iterate + through the ``model_server_versions`` field on the + corresponding responses. + + All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse` + attributes are available on the pager. If multiple requests are made, only + the most recent response is retained, and thus used for attribute lookup. + """ + + def __init__( + self, + method: Callable[ + ..., Awaitable[gkerecommender.FetchModelServerVersionsResponse] + ], + request: gkerecommender.FetchModelServerVersionsRequest, + response: gkerecommender.FetchModelServerVersionsResponse, + *, + retry: OptionalAsyncRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = () + ): + """Instantiates the pager. + + Args: + method (Callable): The method that was originally called, and + which instantiated this pager. + request (google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest): + The initial request object. + response (google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse): + The initial response object. + retry (google.api_core.retry.AsyncRetry): Designation of what errors, + if any, should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + """ + self._method = method + self._request = gkerecommender.FetchModelServerVersionsRequest(request) + self._response = response + self._retry = retry + self._timeout = timeout + self._metadata = metadata + + def __getattr__(self, name: str) -> Any: + return getattr(self._response, name) + + @property + async def pages( + self, + ) -> AsyncIterator[gkerecommender.FetchModelServerVersionsResponse]: + yield self._response + while self._response.next_page_token: + self._request.page_token = self._response.next_page_token + self._response = await self._method( + self._request, + retry=self._retry, + timeout=self._timeout, + metadata=self._metadata, + ) + yield self._response + + def __aiter__(self) -> AsyncIterator[str]: + async def async_generator(): + async for page in self.pages: + for response in page.model_server_versions: + yield response + + return async_generator() + + def __repr__(self) -> str: + return "{0}<{1!r}>".format(self.__class__.__name__, self._response) + + +class FetchProfilesPager: + """A pager for iterating through ``fetch_profiles`` requests. + + This class thinly wraps an initial + :class:`google.cloud.gkerecommender_v1.types.FetchProfilesResponse` object, and + provides an ``__iter__`` method to iterate through its + ``profile`` field. + + If there are more pages, the ``__iter__`` method will make additional + ``FetchProfiles`` requests and continue to iterate + through the ``profile`` field on the + corresponding responses. + + All the usual :class:`google.cloud.gkerecommender_v1.types.FetchProfilesResponse` + attributes are available on the pager. If multiple requests are made, only + the most recent response is retained, and thus used for attribute lookup. + """ + + def __init__( + self, + method: Callable[..., gkerecommender.FetchProfilesResponse], + request: gkerecommender.FetchProfilesRequest, + response: gkerecommender.FetchProfilesResponse, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = () + ): + """Instantiate the pager. + + Args: + method (Callable): The method that was originally called, and + which instantiated this pager. + request (google.cloud.gkerecommender_v1.types.FetchProfilesRequest): + The initial request object. + response (google.cloud.gkerecommender_v1.types.FetchProfilesResponse): + The initial response object. + retry (google.api_core.retry.Retry): Designation of what errors, + if any, should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + """ + self._method = method + self._request = gkerecommender.FetchProfilesRequest(request) + self._response = response + self._retry = retry + self._timeout = timeout + self._metadata = metadata + + def __getattr__(self, name: str) -> Any: + return getattr(self._response, name) + + @property + def pages(self) -> Iterator[gkerecommender.FetchProfilesResponse]: + yield self._response + while self._response.next_page_token: + self._request.page_token = self._response.next_page_token + self._response = self._method( + self._request, + retry=self._retry, + timeout=self._timeout, + metadata=self._metadata, + ) + yield self._response + + def __iter__(self) -> Iterator[gkerecommender.Profile]: + for page in self.pages: + yield from page.profile + + def __repr__(self) -> str: + return "{0}<{1!r}>".format(self.__class__.__name__, self._response) + + +class FetchProfilesAsyncPager: + """A pager for iterating through ``fetch_profiles`` requests. + + This class thinly wraps an initial + :class:`google.cloud.gkerecommender_v1.types.FetchProfilesResponse` object, and + provides an ``__aiter__`` method to iterate through its + ``profile`` field. + + If there are more pages, the ``__aiter__`` method will make additional + ``FetchProfiles`` requests and continue to iterate + through the ``profile`` field on the + corresponding responses. + + All the usual :class:`google.cloud.gkerecommender_v1.types.FetchProfilesResponse` + attributes are available on the pager. If multiple requests are made, only + the most recent response is retained, and thus used for attribute lookup. + """ + + def __init__( + self, + method: Callable[..., Awaitable[gkerecommender.FetchProfilesResponse]], + request: gkerecommender.FetchProfilesRequest, + response: gkerecommender.FetchProfilesResponse, + *, + retry: OptionalAsyncRetry = gapic_v1.method.DEFAULT, + timeout: Union[float, object] = gapic_v1.method.DEFAULT, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = () + ): + """Instantiates the pager. + + Args: + method (Callable): The method that was originally called, and + which instantiated this pager. + request (google.cloud.gkerecommender_v1.types.FetchProfilesRequest): + The initial request object. + response (google.cloud.gkerecommender_v1.types.FetchProfilesResponse): + The initial response object. + retry (google.api_core.retry.AsyncRetry): Designation of what errors, + if any, should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + """ + self._method = method + self._request = gkerecommender.FetchProfilesRequest(request) + self._response = response + self._retry = retry + self._timeout = timeout + self._metadata = metadata + + def __getattr__(self, name: str) -> Any: + return getattr(self._response, name) + + @property + async def pages(self) -> AsyncIterator[gkerecommender.FetchProfilesResponse]: + yield self._response + while self._response.next_page_token: + self._request.page_token = self._response.next_page_token + self._response = await self._method( + self._request, + retry=self._retry, + timeout=self._timeout, + metadata=self._metadata, + ) + yield self._response + + def __aiter__(self) -> AsyncIterator[gkerecommender.Profile]: + async def async_generator(): + async for page in self.pages: + for response in page.profile: + yield response + + return async_generator() + + def __repr__(self) -> str: + return "{0}<{1!r}>".format(self.__class__.__name__, self._response) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/README.rst b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/README.rst new file mode 100644 index 000000000000..bbcb4a6af96c --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/README.rst @@ -0,0 +1,9 @@ + +transport inheritance structure +_______________________________ + +`GkeInferenceQuickstartTransport` is the ABC for all transports. +- public child `GkeInferenceQuickstartGrpcTransport` for sync gRPC transport (defined in `grpc.py`). +- public child `GkeInferenceQuickstartGrpcAsyncIOTransport` for async gRPC transport (defined in `grpc_asyncio.py`). +- private child `_BaseGkeInferenceQuickstartRestTransport` for base REST transport with inner classes `_BaseMETHOD` (defined in `rest_base.py`). +- public child `GkeInferenceQuickstartRestTransport` for sync REST transport with inner classes `METHOD` derived from the parent's corresponding `_BaseMETHOD` classes (defined in `rest.py`). diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/__init__.py new file mode 100644 index 000000000000..4163ed5a0753 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/__init__.py @@ -0,0 +1,41 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from collections import OrderedDict +from typing import Dict, Type + +from .base import GkeInferenceQuickstartTransport +from .grpc import GkeInferenceQuickstartGrpcTransport +from .grpc_asyncio import GkeInferenceQuickstartGrpcAsyncIOTransport +from .rest import ( + GkeInferenceQuickstartRestInterceptor, + GkeInferenceQuickstartRestTransport, +) + +# Compile a registry of transports. +_transport_registry = ( + OrderedDict() +) # type: Dict[str, Type[GkeInferenceQuickstartTransport]] +_transport_registry["grpc"] = GkeInferenceQuickstartGrpcTransport +_transport_registry["grpc_asyncio"] = GkeInferenceQuickstartGrpcAsyncIOTransport +_transport_registry["rest"] = GkeInferenceQuickstartRestTransport + +__all__ = ( + "GkeInferenceQuickstartTransport", + "GkeInferenceQuickstartGrpcTransport", + "GkeInferenceQuickstartGrpcAsyncIOTransport", + "GkeInferenceQuickstartRestTransport", + "GkeInferenceQuickstartRestInterceptor", +) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/base.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/base.py new file mode 100644 index 000000000000..405081650eee --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/base.py @@ -0,0 +1,253 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import abc +from typing import Awaitable, Callable, Dict, Optional, Sequence, Union + +import google.api_core +from google.api_core import exceptions as core_exceptions +from google.api_core import gapic_v1 +from google.api_core import retry as retries +import google.auth # type: ignore +from google.auth import credentials as ga_credentials # type: ignore +from google.oauth2 import service_account # type: ignore +import google.protobuf + +from google.cloud.gkerecommender_v1 import gapic_version as package_version +from google.cloud.gkerecommender_v1.types import gkerecommender + +DEFAULT_CLIENT_INFO = gapic_v1.client_info.ClientInfo( + gapic_version=package_version.__version__ +) + +if hasattr(DEFAULT_CLIENT_INFO, "protobuf_runtime_version"): # pragma: NO COVER + DEFAULT_CLIENT_INFO.protobuf_runtime_version = google.protobuf.__version__ + + +class GkeInferenceQuickstartTransport(abc.ABC): + """Abstract transport class for GkeInferenceQuickstart.""" + + AUTH_SCOPES = ("https://www.googleapis.com/auth/cloud-platform",) + + DEFAULT_HOST: str = "gkerecommender.googleapis.com" + + def __init__( + self, + *, + host: str = DEFAULT_HOST, + credentials: Optional[ga_credentials.Credentials] = None, + credentials_file: Optional[str] = None, + scopes: Optional[Sequence[str]] = None, + quota_project_id: Optional[str] = None, + client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, + always_use_jwt_access: Optional[bool] = False, + api_audience: Optional[str] = None, + **kwargs, + ) -> None: + """Instantiate the transport. + + Args: + host (Optional[str]): + The hostname to connect to (default: 'gkerecommender.googleapis.com'). + credentials (Optional[google.auth.credentials.Credentials]): The + authorization credentials to attach to requests. These + credentials identify the application to the service; if none + are specified, the client will attempt to ascertain the + credentials from the environment. + credentials_file (Optional[str]): A file with credentials that can + be loaded with :func:`google.auth.load_credentials_from_file`. + This argument is mutually exclusive with credentials. + scopes (Optional[Sequence[str]]): A list of scopes. + quota_project_id (Optional[str]): An optional project to use for billing + and quota. + client_info (google.api_core.gapic_v1.client_info.ClientInfo): + The client info used to send a user-agent string along with + API requests. If ``None``, then default info will be used. + Generally, you only need to set this if you're developing + your own client library. + always_use_jwt_access (Optional[bool]): Whether self signed JWT should + be used for service account credentials. + """ + + scopes_kwargs = {"scopes": scopes, "default_scopes": self.AUTH_SCOPES} + + # Save the scopes. + self._scopes = scopes + if not hasattr(self, "_ignore_credentials"): + self._ignore_credentials: bool = False + + # If no credentials are provided, then determine the appropriate + # defaults. + if credentials and credentials_file: + raise core_exceptions.DuplicateCredentialArgs( + "'credentials_file' and 'credentials' are mutually exclusive" + ) + + if credentials_file is not None: + credentials, _ = google.auth.load_credentials_from_file( + credentials_file, **scopes_kwargs, quota_project_id=quota_project_id + ) + elif credentials is None and not self._ignore_credentials: + credentials, _ = google.auth.default( + **scopes_kwargs, quota_project_id=quota_project_id + ) + # Don't apply audience if the credentials file passed from user. + if hasattr(credentials, "with_gdch_audience"): + credentials = credentials.with_gdch_audience( + api_audience if api_audience else host + ) + + # If the credentials are service account credentials, then always try to use self signed JWT. + if ( + always_use_jwt_access + and isinstance(credentials, service_account.Credentials) + and hasattr(service_account.Credentials, "with_always_use_jwt_access") + ): + credentials = credentials.with_always_use_jwt_access(True) + + # Save the credentials. + self._credentials = credentials + + # Save the hostname. Default to port 443 (HTTPS) if none is specified. + if ":" not in host: + host += ":443" + self._host = host + + @property + def host(self): + return self._host + + def _prep_wrapped_messages(self, client_info): + # Precompute the wrapped methods. + self._wrapped_methods = { + self.fetch_models: gapic_v1.method.wrap_method( + self.fetch_models, + default_timeout=60.0, + client_info=client_info, + ), + self.fetch_model_servers: gapic_v1.method.wrap_method( + self.fetch_model_servers, + default_timeout=60.0, + client_info=client_info, + ), + self.fetch_model_server_versions: gapic_v1.method.wrap_method( + self.fetch_model_server_versions, + default_timeout=60.0, + client_info=client_info, + ), + self.fetch_profiles: gapic_v1.method.wrap_method( + self.fetch_profiles, + default_timeout=60.0, + client_info=client_info, + ), + self.generate_optimized_manifest: gapic_v1.method.wrap_method( + self.generate_optimized_manifest, + default_timeout=60.0, + client_info=client_info, + ), + self.fetch_benchmarking_data: gapic_v1.method.wrap_method( + self.fetch_benchmarking_data, + default_timeout=60.0, + client_info=client_info, + ), + } + + def close(self): + """Closes resources associated with the transport. + + .. warning:: + Only call this method if the transport is NOT shared + with other clients - this may cause errors in other clients! + """ + raise NotImplementedError() + + @property + def fetch_models( + self, + ) -> Callable[ + [gkerecommender.FetchModelsRequest], + Union[ + gkerecommender.FetchModelsResponse, + Awaitable[gkerecommender.FetchModelsResponse], + ], + ]: + raise NotImplementedError() + + @property + def fetch_model_servers( + self, + ) -> Callable[ + [gkerecommender.FetchModelServersRequest], + Union[ + gkerecommender.FetchModelServersResponse, + Awaitable[gkerecommender.FetchModelServersResponse], + ], + ]: + raise NotImplementedError() + + @property + def fetch_model_server_versions( + self, + ) -> Callable[ + [gkerecommender.FetchModelServerVersionsRequest], + Union[ + gkerecommender.FetchModelServerVersionsResponse, + Awaitable[gkerecommender.FetchModelServerVersionsResponse], + ], + ]: + raise NotImplementedError() + + @property + def fetch_profiles( + self, + ) -> Callable[ + [gkerecommender.FetchProfilesRequest], + Union[ + gkerecommender.FetchProfilesResponse, + Awaitable[gkerecommender.FetchProfilesResponse], + ], + ]: + raise NotImplementedError() + + @property + def generate_optimized_manifest( + self, + ) -> Callable[ + [gkerecommender.GenerateOptimizedManifestRequest], + Union[ + gkerecommender.GenerateOptimizedManifestResponse, + Awaitable[gkerecommender.GenerateOptimizedManifestResponse], + ], + ]: + raise NotImplementedError() + + @property + def fetch_benchmarking_data( + self, + ) -> Callable[ + [gkerecommender.FetchBenchmarkingDataRequest], + Union[ + gkerecommender.FetchBenchmarkingDataResponse, + Awaitable[gkerecommender.FetchBenchmarkingDataResponse], + ], + ]: + raise NotImplementedError() + + @property + def kind(self) -> str: + raise NotImplementedError() + + +__all__ = ("GkeInferenceQuickstartTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc.py new file mode 100644 index 000000000000..f36abcfd5789 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc.py @@ -0,0 +1,536 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import json +import logging as std_logging +import pickle +from typing import Callable, Dict, Optional, Sequence, Tuple, Union +import warnings + +from google.api_core import gapic_v1, grpc_helpers +import google.auth # type: ignore +from google.auth import credentials as ga_credentials # type: ignore +from google.auth.transport.grpc import SslCredentials # type: ignore +from google.protobuf.json_format import MessageToJson +import google.protobuf.message +import grpc # type: ignore +import proto # type: ignore + +from google.cloud.gkerecommender_v1.types import gkerecommender + +from .base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport + +try: + from google.api_core import client_logging # type: ignore + + CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER +except ImportError: # pragma: NO COVER + CLIENT_LOGGING_SUPPORTED = False + +_LOGGER = std_logging.getLogger(__name__) + + +class _LoggingClientInterceptor(grpc.UnaryUnaryClientInterceptor): # pragma: NO COVER + def intercept_unary_unary(self, continuation, client_call_details, request): + logging_enabled = CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + std_logging.DEBUG + ) + if logging_enabled: # pragma: NO COVER + request_metadata = client_call_details.metadata + if isinstance(request, proto.Message): + request_payload = type(request).to_json(request) + elif isinstance(request, google.protobuf.message.Message): + request_payload = MessageToJson(request) + else: + request_payload = f"{type(request).__name__}: {pickle.dumps(request)}" + + request_metadata = { + key: value.decode("utf-8") if isinstance(value, bytes) else value + for key, value in request_metadata + } + grpc_request = { + "payload": request_payload, + "requestMethod": "grpc", + "metadata": dict(request_metadata), + } + _LOGGER.debug( + f"Sending request for {client_call_details.method}", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": str(client_call_details.method), + "request": grpc_request, + "metadata": grpc_request["metadata"], + }, + ) + response = continuation(client_call_details, request) + if logging_enabled: # pragma: NO COVER + response_metadata = response.trailing_metadata() + # Convert gRPC metadata `` to list of tuples + metadata = ( + dict([(k, str(v)) for k, v in response_metadata]) + if response_metadata + else None + ) + result = response.result() + if isinstance(result, proto.Message): + response_payload = type(result).to_json(result) + elif isinstance(result, google.protobuf.message.Message): + response_payload = MessageToJson(result) + else: + response_payload = f"{type(result).__name__}: {pickle.dumps(result)}" + grpc_response = { + "payload": response_payload, + "metadata": metadata, + "status": "OK", + } + _LOGGER.debug( + f"Received response for {client_call_details.method}.", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": client_call_details.method, + "response": grpc_response, + "metadata": grpc_response["metadata"], + }, + ) + return response + + +class GkeInferenceQuickstartGrpcTransport(GkeInferenceQuickstartTransport): + """gRPC backend transport for GkeInferenceQuickstart. + + GKE Inference Quickstart (GIQ) service provides profiles with + performance metrics for popular models and model servers across + multiple accelerators. These profiles help generate optimized + best practices for running inference on GKE. + + This class defines the same methods as the primary client, so the + primary client can load the underlying transport implementation + and call it. + + It sends protocol buffers over the wire using gRPC (which is built on + top of HTTP/2); the ``grpcio`` package must be installed. + """ + + _stubs: Dict[str, Callable] + + def __init__( + self, + *, + host: str = "gkerecommender.googleapis.com", + credentials: Optional[ga_credentials.Credentials] = None, + credentials_file: Optional[str] = None, + scopes: Optional[Sequence[str]] = None, + channel: Optional[Union[grpc.Channel, Callable[..., grpc.Channel]]] = None, + api_mtls_endpoint: Optional[str] = None, + client_cert_source: Optional[Callable[[], Tuple[bytes, bytes]]] = None, + ssl_channel_credentials: Optional[grpc.ChannelCredentials] = None, + client_cert_source_for_mtls: Optional[Callable[[], Tuple[bytes, bytes]]] = None, + quota_project_id: Optional[str] = None, + client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, + always_use_jwt_access: Optional[bool] = False, + api_audience: Optional[str] = None, + ) -> None: + """Instantiate the transport. + + Args: + host (Optional[str]): + The hostname to connect to (default: 'gkerecommender.googleapis.com'). + credentials (Optional[google.auth.credentials.Credentials]): The + authorization credentials to attach to requests. These + credentials identify the application to the service; if none + are specified, the client will attempt to ascertain the + credentials from the environment. + This argument is ignored if a ``channel`` instance is provided. + credentials_file (Optional[str]): A file with credentials that can + be loaded with :func:`google.auth.load_credentials_from_file`. + This argument is ignored if a ``channel`` instance is provided. + scopes (Optional(Sequence[str])): A list of scopes. This argument is + ignored if a ``channel`` instance is provided. + channel (Optional[Union[grpc.Channel, Callable[..., grpc.Channel]]]): + A ``Channel`` instance through which to make calls, or a Callable + that constructs and returns one. If set to None, ``self.create_channel`` + is used to create the channel. If a Callable is given, it will be called + with the same arguments as used in ``self.create_channel``. + api_mtls_endpoint (Optional[str]): Deprecated. The mutual TLS endpoint. + If provided, it overrides the ``host`` argument and tries to create + a mutual TLS channel with client SSL credentials from + ``client_cert_source`` or application default SSL credentials. + client_cert_source (Optional[Callable[[], Tuple[bytes, bytes]]]): + Deprecated. A callback to provide client SSL certificate bytes and + private key bytes, both in PEM format. It is ignored if + ``api_mtls_endpoint`` is None. + ssl_channel_credentials (grpc.ChannelCredentials): SSL credentials + for the grpc channel. It is ignored if a ``channel`` instance is provided. + client_cert_source_for_mtls (Optional[Callable[[], Tuple[bytes, bytes]]]): + A callback to provide client certificate bytes and private key bytes, + both in PEM format. It is used to configure a mutual TLS channel. It is + ignored if a ``channel`` instance or ``ssl_channel_credentials`` is provided. + quota_project_id (Optional[str]): An optional project to use for billing + and quota. + client_info (google.api_core.gapic_v1.client_info.ClientInfo): + The client info used to send a user-agent string along with + API requests. If ``None``, then default info will be used. + Generally, you only need to set this if you're developing + your own client library. + always_use_jwt_access (Optional[bool]): Whether self signed JWT should + be used for service account credentials. + + Raises: + google.auth.exceptions.MutualTLSChannelError: If mutual TLS transport + creation failed for any reason. + google.api_core.exceptions.DuplicateCredentialArgs: If both ``credentials`` + and ``credentials_file`` are passed. + """ + self._grpc_channel = None + self._ssl_channel_credentials = ssl_channel_credentials + self._stubs: Dict[str, Callable] = {} + + if api_mtls_endpoint: + warnings.warn("api_mtls_endpoint is deprecated", DeprecationWarning) + if client_cert_source: + warnings.warn("client_cert_source is deprecated", DeprecationWarning) + + if isinstance(channel, grpc.Channel): + # Ignore credentials if a channel was passed. + credentials = None + self._ignore_credentials = True + # If a channel was explicitly provided, set it. + self._grpc_channel = channel + self._ssl_channel_credentials = None + + else: + if api_mtls_endpoint: + host = api_mtls_endpoint + + # Create SSL credentials with client_cert_source or application + # default SSL credentials. + if client_cert_source: + cert, key = client_cert_source() + self._ssl_channel_credentials = grpc.ssl_channel_credentials( + certificate_chain=cert, private_key=key + ) + else: + self._ssl_channel_credentials = SslCredentials().ssl_credentials + + else: + if client_cert_source_for_mtls and not ssl_channel_credentials: + cert, key = client_cert_source_for_mtls() + self._ssl_channel_credentials = grpc.ssl_channel_credentials( + certificate_chain=cert, private_key=key + ) + + # The base transport sets the host, credentials and scopes + super().__init__( + host=host, + credentials=credentials, + credentials_file=credentials_file, + scopes=scopes, + quota_project_id=quota_project_id, + client_info=client_info, + always_use_jwt_access=always_use_jwt_access, + api_audience=api_audience, + ) + + if not self._grpc_channel: + # initialize with the provided callable or the default channel + channel_init = channel or type(self).create_channel + self._grpc_channel = channel_init( + self._host, + # use the credentials which are saved + credentials=self._credentials, + # Set ``credentials_file`` to ``None`` here as + # the credentials that we saved earlier should be used. + credentials_file=None, + scopes=self._scopes, + ssl_credentials=self._ssl_channel_credentials, + quota_project_id=quota_project_id, + options=[ + ("grpc.max_send_message_length", -1), + ("grpc.max_receive_message_length", -1), + ], + ) + + self._interceptor = _LoggingClientInterceptor() + self._logged_channel = grpc.intercept_channel( + self._grpc_channel, self._interceptor + ) + + # Wrap messages. This must be done after self._logged_channel exists + self._prep_wrapped_messages(client_info) + + @classmethod + def create_channel( + cls, + host: str = "gkerecommender.googleapis.com", + credentials: Optional[ga_credentials.Credentials] = None, + credentials_file: Optional[str] = None, + scopes: Optional[Sequence[str]] = None, + quota_project_id: Optional[str] = None, + **kwargs, + ) -> grpc.Channel: + """Create and return a gRPC channel object. + Args: + host (Optional[str]): The host for the channel to use. + credentials (Optional[~.Credentials]): The + authorization credentials to attach to requests. These + credentials identify this application to the service. If + none are specified, the client will attempt to ascertain + the credentials from the environment. + credentials_file (Optional[str]): A file with credentials that can + be loaded with :func:`google.auth.load_credentials_from_file`. + This argument is mutually exclusive with credentials. + scopes (Optional[Sequence[str]]): A optional list of scopes needed for this + service. These are only used when credentials are not specified and + are passed to :func:`google.auth.default`. + quota_project_id (Optional[str]): An optional project to use for billing + and quota. + kwargs (Optional[dict]): Keyword arguments, which are passed to the + channel creation. + Returns: + grpc.Channel: A gRPC channel object. + + Raises: + google.api_core.exceptions.DuplicateCredentialArgs: If both ``credentials`` + and ``credentials_file`` are passed. + """ + + return grpc_helpers.create_channel( + host, + credentials=credentials, + credentials_file=credentials_file, + quota_project_id=quota_project_id, + default_scopes=cls.AUTH_SCOPES, + scopes=scopes, + default_host=cls.DEFAULT_HOST, + **kwargs, + ) + + @property + def grpc_channel(self) -> grpc.Channel: + """Return the channel designed to connect to this service.""" + return self._grpc_channel + + @property + def fetch_models( + self, + ) -> Callable[ + [gkerecommender.FetchModelsRequest], gkerecommender.FetchModelsResponse + ]: + r"""Return a callable for the fetch models method over gRPC. + + Fetches available models. Open-source models follow the + Huggingface Hub ``owner/model_name`` format. + + Returns: + Callable[[~.FetchModelsRequest], + ~.FetchModelsResponse]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_models" not in self._stubs: + self._stubs["fetch_models"] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModels", + request_serializer=gkerecommender.FetchModelsRequest.serialize, + response_deserializer=gkerecommender.FetchModelsResponse.deserialize, + ) + return self._stubs["fetch_models"] + + @property + def fetch_model_servers( + self, + ) -> Callable[ + [gkerecommender.FetchModelServersRequest], + gkerecommender.FetchModelServersResponse, + ]: + r"""Return a callable for the fetch model servers method over gRPC. + + Fetches available model servers. Open-source model servers use + simplified, lowercase names (e.g., ``vllm``). + + Returns: + Callable[[~.FetchModelServersRequest], + ~.FetchModelServersResponse]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_model_servers" not in self._stubs: + self._stubs["fetch_model_servers"] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModelServers", + request_serializer=gkerecommender.FetchModelServersRequest.serialize, + response_deserializer=gkerecommender.FetchModelServersResponse.deserialize, + ) + return self._stubs["fetch_model_servers"] + + @property + def fetch_model_server_versions( + self, + ) -> Callable[ + [gkerecommender.FetchModelServerVersionsRequest], + gkerecommender.FetchModelServerVersionsResponse, + ]: + r"""Return a callable for the fetch model server versions method over gRPC. + + Fetches available model server versions. Open-source servers use + their own versioning schemas (e.g., ``vllm`` uses semver like + ``v1.0.0``). + + Some model servers have different versioning schemas depending + on the accelerator. For example, ``vllm`` uses semver on GPUs, + but returns nightly build tags on TPUs. All available versions + will be returned when different schemas are present. + + Returns: + Callable[[~.FetchModelServerVersionsRequest], + ~.FetchModelServerVersionsResponse]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_model_server_versions" not in self._stubs: + self._stubs[ + "fetch_model_server_versions" + ] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModelServerVersions", + request_serializer=gkerecommender.FetchModelServerVersionsRequest.serialize, + response_deserializer=gkerecommender.FetchModelServerVersionsResponse.deserialize, + ) + return self._stubs["fetch_model_server_versions"] + + @property + def fetch_profiles( + self, + ) -> Callable[ + [gkerecommender.FetchProfilesRequest], gkerecommender.FetchProfilesResponse + ]: + r"""Return a callable for the fetch profiles method over gRPC. + + Fetches available profiles. A profile contains performance + metrics and cost information for a specific model server setup. + Profiles can be filtered by parameters. If no filters are + provided, all profiles are returned. + + Profiles display a single value per performance metric based on + the provided performance requirements. If no requirements are + given, the metrics represent the inflection point. See `Run best + practice inference with GKE Inference Quickstart + recipes `__ + for details. + + Returns: + Callable[[~.FetchProfilesRequest], + ~.FetchProfilesResponse]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_profiles" not in self._stubs: + self._stubs["fetch_profiles"] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchProfiles", + request_serializer=gkerecommender.FetchProfilesRequest.serialize, + response_deserializer=gkerecommender.FetchProfilesResponse.deserialize, + ) + return self._stubs["fetch_profiles"] + + @property + def generate_optimized_manifest( + self, + ) -> Callable[ + [gkerecommender.GenerateOptimizedManifestRequest], + gkerecommender.GenerateOptimizedManifestResponse, + ]: + r"""Return a callable for the generate optimized manifest method over gRPC. + + Generates an optimized deployment manifest for a given model and + model server, based on the specified accelerator, performance + targets, and configurations. See `Run best practice inference + with GKE Inference Quickstart + recipes `__ + for deployment details. + + Returns: + Callable[[~.GenerateOptimizedManifestRequest], + ~.GenerateOptimizedManifestResponse]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "generate_optimized_manifest" not in self._stubs: + self._stubs[ + "generate_optimized_manifest" + ] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/GenerateOptimizedManifest", + request_serializer=gkerecommender.GenerateOptimizedManifestRequest.serialize, + response_deserializer=gkerecommender.GenerateOptimizedManifestResponse.deserialize, + ) + return self._stubs["generate_optimized_manifest"] + + @property + def fetch_benchmarking_data( + self, + ) -> Callable[ + [gkerecommender.FetchBenchmarkingDataRequest], + gkerecommender.FetchBenchmarkingDataResponse, + ]: + r"""Return a callable for the fetch benchmarking data method over gRPC. + + Fetches all of the benchmarking data available for a + profile. Benchmarking data returns all of the + performance metrics available for a given model server + setup on a given instance type. + + Returns: + Callable[[~.FetchBenchmarkingDataRequest], + ~.FetchBenchmarkingDataResponse]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_benchmarking_data" not in self._stubs: + self._stubs["fetch_benchmarking_data"] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchBenchmarkingData", + request_serializer=gkerecommender.FetchBenchmarkingDataRequest.serialize, + response_deserializer=gkerecommender.FetchBenchmarkingDataResponse.deserialize, + ) + return self._stubs["fetch_benchmarking_data"] + + def close(self): + self._logged_channel.close() + + @property + def kind(self) -> str: + return "grpc" + + +__all__ = ("GkeInferenceQuickstartGrpcTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc_asyncio.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc_asyncio.py new file mode 100644 index 000000000000..3a719e36d2fc --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc_asyncio.py @@ -0,0 +1,586 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import inspect +import json +import logging as std_logging +import pickle +from typing import Awaitable, Callable, Dict, Optional, Sequence, Tuple, Union +import warnings + +from google.api_core import exceptions as core_exceptions +from google.api_core import gapic_v1, grpc_helpers_async +from google.api_core import retry_async as retries +from google.auth import credentials as ga_credentials # type: ignore +from google.auth.transport.grpc import SslCredentials # type: ignore +from google.protobuf.json_format import MessageToJson +import google.protobuf.message +import grpc # type: ignore +from grpc.experimental import aio # type: ignore +import proto # type: ignore + +from google.cloud.gkerecommender_v1.types import gkerecommender + +from .base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport +from .grpc import GkeInferenceQuickstartGrpcTransport + +try: + from google.api_core import client_logging # type: ignore + + CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER +except ImportError: # pragma: NO COVER + CLIENT_LOGGING_SUPPORTED = False + +_LOGGER = std_logging.getLogger(__name__) + + +class _LoggingClientAIOInterceptor( + grpc.aio.UnaryUnaryClientInterceptor +): # pragma: NO COVER + async def intercept_unary_unary(self, continuation, client_call_details, request): + logging_enabled = CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + std_logging.DEBUG + ) + if logging_enabled: # pragma: NO COVER + request_metadata = client_call_details.metadata + if isinstance(request, proto.Message): + request_payload = type(request).to_json(request) + elif isinstance(request, google.protobuf.message.Message): + request_payload = MessageToJson(request) + else: + request_payload = f"{type(request).__name__}: {pickle.dumps(request)}" + + request_metadata = { + key: value.decode("utf-8") if isinstance(value, bytes) else value + for key, value in request_metadata + } + grpc_request = { + "payload": request_payload, + "requestMethod": "grpc", + "metadata": dict(request_metadata), + } + _LOGGER.debug( + f"Sending request for {client_call_details.method}", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": str(client_call_details.method), + "request": grpc_request, + "metadata": grpc_request["metadata"], + }, + ) + response = await continuation(client_call_details, request) + if logging_enabled: # pragma: NO COVER + response_metadata = await response.trailing_metadata() + # Convert gRPC metadata `` to list of tuples + metadata = ( + dict([(k, str(v)) for k, v in response_metadata]) + if response_metadata + else None + ) + result = await response + if isinstance(result, proto.Message): + response_payload = type(result).to_json(result) + elif isinstance(result, google.protobuf.message.Message): + response_payload = MessageToJson(result) + else: + response_payload = f"{type(result).__name__}: {pickle.dumps(result)}" + grpc_response = { + "payload": response_payload, + "metadata": metadata, + "status": "OK", + } + _LOGGER.debug( + f"Received response to rpc {client_call_details.method}.", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": str(client_call_details.method), + "response": grpc_response, + "metadata": grpc_response["metadata"], + }, + ) + return response + + +class GkeInferenceQuickstartGrpcAsyncIOTransport(GkeInferenceQuickstartTransport): + """gRPC AsyncIO backend transport for GkeInferenceQuickstart. + + GKE Inference Quickstart (GIQ) service provides profiles with + performance metrics for popular models and model servers across + multiple accelerators. These profiles help generate optimized + best practices for running inference on GKE. + + This class defines the same methods as the primary client, so the + primary client can load the underlying transport implementation + and call it. + + It sends protocol buffers over the wire using gRPC (which is built on + top of HTTP/2); the ``grpcio`` package must be installed. + """ + + _grpc_channel: aio.Channel + _stubs: Dict[str, Callable] = {} + + @classmethod + def create_channel( + cls, + host: str = "gkerecommender.googleapis.com", + credentials: Optional[ga_credentials.Credentials] = None, + credentials_file: Optional[str] = None, + scopes: Optional[Sequence[str]] = None, + quota_project_id: Optional[str] = None, + **kwargs, + ) -> aio.Channel: + """Create and return a gRPC AsyncIO channel object. + Args: + host (Optional[str]): The host for the channel to use. + credentials (Optional[~.Credentials]): The + authorization credentials to attach to requests. These + credentials identify this application to the service. If + none are specified, the client will attempt to ascertain + the credentials from the environment. + credentials_file (Optional[str]): A file with credentials that can + be loaded with :func:`google.auth.load_credentials_from_file`. + scopes (Optional[Sequence[str]]): A optional list of scopes needed for this + service. These are only used when credentials are not specified and + are passed to :func:`google.auth.default`. + quota_project_id (Optional[str]): An optional project to use for billing + and quota. + kwargs (Optional[dict]): Keyword arguments, which are passed to the + channel creation. + Returns: + aio.Channel: A gRPC AsyncIO channel object. + """ + + return grpc_helpers_async.create_channel( + host, + credentials=credentials, + credentials_file=credentials_file, + quota_project_id=quota_project_id, + default_scopes=cls.AUTH_SCOPES, + scopes=scopes, + default_host=cls.DEFAULT_HOST, + **kwargs, + ) + + def __init__( + self, + *, + host: str = "gkerecommender.googleapis.com", + credentials: Optional[ga_credentials.Credentials] = None, + credentials_file: Optional[str] = None, + scopes: Optional[Sequence[str]] = None, + channel: Optional[Union[aio.Channel, Callable[..., aio.Channel]]] = None, + api_mtls_endpoint: Optional[str] = None, + client_cert_source: Optional[Callable[[], Tuple[bytes, bytes]]] = None, + ssl_channel_credentials: Optional[grpc.ChannelCredentials] = None, + client_cert_source_for_mtls: Optional[Callable[[], Tuple[bytes, bytes]]] = None, + quota_project_id: Optional[str] = None, + client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, + always_use_jwt_access: Optional[bool] = False, + api_audience: Optional[str] = None, + ) -> None: + """Instantiate the transport. + + Args: + host (Optional[str]): + The hostname to connect to (default: 'gkerecommender.googleapis.com'). + credentials (Optional[google.auth.credentials.Credentials]): The + authorization credentials to attach to requests. These + credentials identify the application to the service; if none + are specified, the client will attempt to ascertain the + credentials from the environment. + This argument is ignored if a ``channel`` instance is provided. + credentials_file (Optional[str]): A file with credentials that can + be loaded with :func:`google.auth.load_credentials_from_file`. + This argument is ignored if a ``channel`` instance is provided. + scopes (Optional[Sequence[str]]): A optional list of scopes needed for this + service. These are only used when credentials are not specified and + are passed to :func:`google.auth.default`. + channel (Optional[Union[aio.Channel, Callable[..., aio.Channel]]]): + A ``Channel`` instance through which to make calls, or a Callable + that constructs and returns one. If set to None, ``self.create_channel`` + is used to create the channel. If a Callable is given, it will be called + with the same arguments as used in ``self.create_channel``. + api_mtls_endpoint (Optional[str]): Deprecated. The mutual TLS endpoint. + If provided, it overrides the ``host`` argument and tries to create + a mutual TLS channel with client SSL credentials from + ``client_cert_source`` or application default SSL credentials. + client_cert_source (Optional[Callable[[], Tuple[bytes, bytes]]]): + Deprecated. A callback to provide client SSL certificate bytes and + private key bytes, both in PEM format. It is ignored if + ``api_mtls_endpoint`` is None. + ssl_channel_credentials (grpc.ChannelCredentials): SSL credentials + for the grpc channel. It is ignored if a ``channel`` instance is provided. + client_cert_source_for_mtls (Optional[Callable[[], Tuple[bytes, bytes]]]): + A callback to provide client certificate bytes and private key bytes, + both in PEM format. It is used to configure a mutual TLS channel. It is + ignored if a ``channel`` instance or ``ssl_channel_credentials`` is provided. + quota_project_id (Optional[str]): An optional project to use for billing + and quota. + client_info (google.api_core.gapic_v1.client_info.ClientInfo): + The client info used to send a user-agent string along with + API requests. If ``None``, then default info will be used. + Generally, you only need to set this if you're developing + your own client library. + always_use_jwt_access (Optional[bool]): Whether self signed JWT should + be used for service account credentials. + + Raises: + google.auth.exceptions.MutualTlsChannelError: If mutual TLS transport + creation failed for any reason. + google.api_core.exceptions.DuplicateCredentialArgs: If both ``credentials`` + and ``credentials_file`` are passed. + """ + self._grpc_channel = None + self._ssl_channel_credentials = ssl_channel_credentials + self._stubs: Dict[str, Callable] = {} + + if api_mtls_endpoint: + warnings.warn("api_mtls_endpoint is deprecated", DeprecationWarning) + if client_cert_source: + warnings.warn("client_cert_source is deprecated", DeprecationWarning) + + if isinstance(channel, aio.Channel): + # Ignore credentials if a channel was passed. + credentials = None + self._ignore_credentials = True + # If a channel was explicitly provided, set it. + self._grpc_channel = channel + self._ssl_channel_credentials = None + else: + if api_mtls_endpoint: + host = api_mtls_endpoint + + # Create SSL credentials with client_cert_source or application + # default SSL credentials. + if client_cert_source: + cert, key = client_cert_source() + self._ssl_channel_credentials = grpc.ssl_channel_credentials( + certificate_chain=cert, private_key=key + ) + else: + self._ssl_channel_credentials = SslCredentials().ssl_credentials + + else: + if client_cert_source_for_mtls and not ssl_channel_credentials: + cert, key = client_cert_source_for_mtls() + self._ssl_channel_credentials = grpc.ssl_channel_credentials( + certificate_chain=cert, private_key=key + ) + + # The base transport sets the host, credentials and scopes + super().__init__( + host=host, + credentials=credentials, + credentials_file=credentials_file, + scopes=scopes, + quota_project_id=quota_project_id, + client_info=client_info, + always_use_jwt_access=always_use_jwt_access, + api_audience=api_audience, + ) + + if not self._grpc_channel: + # initialize with the provided callable or the default channel + channel_init = channel or type(self).create_channel + self._grpc_channel = channel_init( + self._host, + # use the credentials which are saved + credentials=self._credentials, + # Set ``credentials_file`` to ``None`` here as + # the credentials that we saved earlier should be used. + credentials_file=None, + scopes=self._scopes, + ssl_credentials=self._ssl_channel_credentials, + quota_project_id=quota_project_id, + options=[ + ("grpc.max_send_message_length", -1), + ("grpc.max_receive_message_length", -1), + ], + ) + + self._interceptor = _LoggingClientAIOInterceptor() + self._grpc_channel._unary_unary_interceptors.append(self._interceptor) + self._logged_channel = self._grpc_channel + self._wrap_with_kind = ( + "kind" in inspect.signature(gapic_v1.method_async.wrap_method).parameters + ) + # Wrap messages. This must be done after self._logged_channel exists + self._prep_wrapped_messages(client_info) + + @property + def grpc_channel(self) -> aio.Channel: + """Create the channel designed to connect to this service. + + This property caches on the instance; repeated calls return + the same channel. + """ + # Return the channel from cache. + return self._grpc_channel + + @property + def fetch_models( + self, + ) -> Callable[ + [gkerecommender.FetchModelsRequest], + Awaitable[gkerecommender.FetchModelsResponse], + ]: + r"""Return a callable for the fetch models method over gRPC. + + Fetches available models. Open-source models follow the + Huggingface Hub ``owner/model_name`` format. + + Returns: + Callable[[~.FetchModelsRequest], + Awaitable[~.FetchModelsResponse]]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_models" not in self._stubs: + self._stubs["fetch_models"] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModels", + request_serializer=gkerecommender.FetchModelsRequest.serialize, + response_deserializer=gkerecommender.FetchModelsResponse.deserialize, + ) + return self._stubs["fetch_models"] + + @property + def fetch_model_servers( + self, + ) -> Callable[ + [gkerecommender.FetchModelServersRequest], + Awaitable[gkerecommender.FetchModelServersResponse], + ]: + r"""Return a callable for the fetch model servers method over gRPC. + + Fetches available model servers. Open-source model servers use + simplified, lowercase names (e.g., ``vllm``). + + Returns: + Callable[[~.FetchModelServersRequest], + Awaitable[~.FetchModelServersResponse]]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_model_servers" not in self._stubs: + self._stubs["fetch_model_servers"] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModelServers", + request_serializer=gkerecommender.FetchModelServersRequest.serialize, + response_deserializer=gkerecommender.FetchModelServersResponse.deserialize, + ) + return self._stubs["fetch_model_servers"] + + @property + def fetch_model_server_versions( + self, + ) -> Callable[ + [gkerecommender.FetchModelServerVersionsRequest], + Awaitable[gkerecommender.FetchModelServerVersionsResponse], + ]: + r"""Return a callable for the fetch model server versions method over gRPC. + + Fetches available model server versions. Open-source servers use + their own versioning schemas (e.g., ``vllm`` uses semver like + ``v1.0.0``). + + Some model servers have different versioning schemas depending + on the accelerator. For example, ``vllm`` uses semver on GPUs, + but returns nightly build tags on TPUs. All available versions + will be returned when different schemas are present. + + Returns: + Callable[[~.FetchModelServerVersionsRequest], + Awaitable[~.FetchModelServerVersionsResponse]]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_model_server_versions" not in self._stubs: + self._stubs[ + "fetch_model_server_versions" + ] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModelServerVersions", + request_serializer=gkerecommender.FetchModelServerVersionsRequest.serialize, + response_deserializer=gkerecommender.FetchModelServerVersionsResponse.deserialize, + ) + return self._stubs["fetch_model_server_versions"] + + @property + def fetch_profiles( + self, + ) -> Callable[ + [gkerecommender.FetchProfilesRequest], + Awaitable[gkerecommender.FetchProfilesResponse], + ]: + r"""Return a callable for the fetch profiles method over gRPC. + + Fetches available profiles. A profile contains performance + metrics and cost information for a specific model server setup. + Profiles can be filtered by parameters. If no filters are + provided, all profiles are returned. + + Profiles display a single value per performance metric based on + the provided performance requirements. If no requirements are + given, the metrics represent the inflection point. See `Run best + practice inference with GKE Inference Quickstart + recipes `__ + for details. + + Returns: + Callable[[~.FetchProfilesRequest], + Awaitable[~.FetchProfilesResponse]]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_profiles" not in self._stubs: + self._stubs["fetch_profiles"] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchProfiles", + request_serializer=gkerecommender.FetchProfilesRequest.serialize, + response_deserializer=gkerecommender.FetchProfilesResponse.deserialize, + ) + return self._stubs["fetch_profiles"] + + @property + def generate_optimized_manifest( + self, + ) -> Callable[ + [gkerecommender.GenerateOptimizedManifestRequest], + Awaitable[gkerecommender.GenerateOptimizedManifestResponse], + ]: + r"""Return a callable for the generate optimized manifest method over gRPC. + + Generates an optimized deployment manifest for a given model and + model server, based on the specified accelerator, performance + targets, and configurations. See `Run best practice inference + with GKE Inference Quickstart + recipes `__ + for deployment details. + + Returns: + Callable[[~.GenerateOptimizedManifestRequest], + Awaitable[~.GenerateOptimizedManifestResponse]]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "generate_optimized_manifest" not in self._stubs: + self._stubs[ + "generate_optimized_manifest" + ] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/GenerateOptimizedManifest", + request_serializer=gkerecommender.GenerateOptimizedManifestRequest.serialize, + response_deserializer=gkerecommender.GenerateOptimizedManifestResponse.deserialize, + ) + return self._stubs["generate_optimized_manifest"] + + @property + def fetch_benchmarking_data( + self, + ) -> Callable[ + [gkerecommender.FetchBenchmarkingDataRequest], + Awaitable[gkerecommender.FetchBenchmarkingDataResponse], + ]: + r"""Return a callable for the fetch benchmarking data method over gRPC. + + Fetches all of the benchmarking data available for a + profile. Benchmarking data returns all of the + performance metrics available for a given model server + setup on a given instance type. + + Returns: + Callable[[~.FetchBenchmarkingDataRequest], + Awaitable[~.FetchBenchmarkingDataResponse]]: + A function that, when called, will call the underlying RPC + on the server. + """ + # Generate a "stub function" on-the-fly which will actually make + # the request. + # gRPC handles serialization and deserialization, so we just need + # to pass in the functions for each. + if "fetch_benchmarking_data" not in self._stubs: + self._stubs["fetch_benchmarking_data"] = self._logged_channel.unary_unary( + "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchBenchmarkingData", + request_serializer=gkerecommender.FetchBenchmarkingDataRequest.serialize, + response_deserializer=gkerecommender.FetchBenchmarkingDataResponse.deserialize, + ) + return self._stubs["fetch_benchmarking_data"] + + def _prep_wrapped_messages(self, client_info): + """Precompute the wrapped methods, overriding the base class method to use async wrappers.""" + self._wrapped_methods = { + self.fetch_models: self._wrap_method( + self.fetch_models, + default_timeout=60.0, + client_info=client_info, + ), + self.fetch_model_servers: self._wrap_method( + self.fetch_model_servers, + default_timeout=60.0, + client_info=client_info, + ), + self.fetch_model_server_versions: self._wrap_method( + self.fetch_model_server_versions, + default_timeout=60.0, + client_info=client_info, + ), + self.fetch_profiles: self._wrap_method( + self.fetch_profiles, + default_timeout=60.0, + client_info=client_info, + ), + self.generate_optimized_manifest: self._wrap_method( + self.generate_optimized_manifest, + default_timeout=60.0, + client_info=client_info, + ), + self.fetch_benchmarking_data: self._wrap_method( + self.fetch_benchmarking_data, + default_timeout=60.0, + client_info=client_info, + ), + } + + def _wrap_method(self, func, *args, **kwargs): + if self._wrap_with_kind: # pragma: NO COVER + kwargs["kind"] = self.kind + return gapic_v1.method_async.wrap_method(func, *args, **kwargs) + + def close(self): + return self._logged_channel.close() + + @property + def kind(self) -> str: + return "grpc_asyncio" + + +__all__ = ("GkeInferenceQuickstartGrpcAsyncIOTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest.py new file mode 100644 index 000000000000..e59cde7649f9 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest.py @@ -0,0 +1,1532 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import dataclasses +import json # type: ignore +import logging +from typing import Any, Callable, Dict, List, Optional, Sequence, Tuple, Union +import warnings + +from google.api_core import exceptions as core_exceptions +from google.api_core import gapic_v1, rest_helpers, rest_streaming +from google.api_core import retry as retries +from google.auth import credentials as ga_credentials # type: ignore +from google.auth.transport.requests import AuthorizedSession # type: ignore +import google.protobuf +from google.protobuf import json_format +from requests import __version__ as requests_version + +from google.cloud.gkerecommender_v1.types import gkerecommender + +from .base import DEFAULT_CLIENT_INFO as BASE_DEFAULT_CLIENT_INFO +from .rest_base import _BaseGkeInferenceQuickstartRestTransport + +try: + OptionalRetry = Union[retries.Retry, gapic_v1.method._MethodDefault, None] +except AttributeError: # pragma: NO COVER + OptionalRetry = Union[retries.Retry, object, None] # type: ignore + +try: + from google.api_core import client_logging # type: ignore + + CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER +except ImportError: # pragma: NO COVER + CLIENT_LOGGING_SUPPORTED = False + +_LOGGER = logging.getLogger(__name__) + +DEFAULT_CLIENT_INFO = gapic_v1.client_info.ClientInfo( + gapic_version=BASE_DEFAULT_CLIENT_INFO.gapic_version, + grpc_version=None, + rest_version=f"requests@{requests_version}", +) + +if hasattr(DEFAULT_CLIENT_INFO, "protobuf_runtime_version"): # pragma: NO COVER + DEFAULT_CLIENT_INFO.protobuf_runtime_version = google.protobuf.__version__ + + +class GkeInferenceQuickstartRestInterceptor: + """Interceptor for GkeInferenceQuickstart. + + Interceptors are used to manipulate requests, request metadata, and responses + in arbitrary ways. + Example use cases include: + * Logging + * Verifying requests according to service or custom semantics + * Stripping extraneous information from responses + + These use cases and more can be enabled by injecting an + instance of a custom subclass when constructing the GkeInferenceQuickstartRestTransport. + + .. code-block:: python + class MyCustomGkeInferenceQuickstartInterceptor(GkeInferenceQuickstartRestInterceptor): + def pre_fetch_benchmarking_data(self, request, metadata): + logging.log(f"Received request: {request}") + return request, metadata + + def post_fetch_benchmarking_data(self, response): + logging.log(f"Received response: {response}") + return response + + def pre_fetch_models(self, request, metadata): + logging.log(f"Received request: {request}") + return request, metadata + + def post_fetch_models(self, response): + logging.log(f"Received response: {response}") + return response + + def pre_fetch_model_servers(self, request, metadata): + logging.log(f"Received request: {request}") + return request, metadata + + def post_fetch_model_servers(self, response): + logging.log(f"Received response: {response}") + return response + + def pre_fetch_model_server_versions(self, request, metadata): + logging.log(f"Received request: {request}") + return request, metadata + + def post_fetch_model_server_versions(self, response): + logging.log(f"Received response: {response}") + return response + + def pre_fetch_profiles(self, request, metadata): + logging.log(f"Received request: {request}") + return request, metadata + + def post_fetch_profiles(self, response): + logging.log(f"Received response: {response}") + return response + + def pre_generate_optimized_manifest(self, request, metadata): + logging.log(f"Received request: {request}") + return request, metadata + + def post_generate_optimized_manifest(self, response): + logging.log(f"Received response: {response}") + return response + + transport = GkeInferenceQuickstartRestTransport(interceptor=MyCustomGkeInferenceQuickstartInterceptor()) + client = GkeInferenceQuickstartClient(transport=transport) + + + """ + + def pre_fetch_benchmarking_data( + self, + request: gkerecommender.FetchBenchmarkingDataRequest, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchBenchmarkingDataRequest, + Sequence[Tuple[str, Union[str, bytes]]], + ]: + """Pre-rpc interceptor for fetch_benchmarking_data + + Override in a subclass to manipulate the request or metadata + before they are sent to the GkeInferenceQuickstart server. + """ + return request, metadata + + def post_fetch_benchmarking_data( + self, response: gkerecommender.FetchBenchmarkingDataResponse + ) -> gkerecommender.FetchBenchmarkingDataResponse: + """Post-rpc interceptor for fetch_benchmarking_data + + DEPRECATED. Please use the `post_fetch_benchmarking_data_with_metadata` + interceptor instead. + + Override in a subclass to read or manipulate the response + after it is returned by the GkeInferenceQuickstart server but before + it is returned to user code. This `post_fetch_benchmarking_data` interceptor runs + before the `post_fetch_benchmarking_data_with_metadata` interceptor. + """ + return response + + def post_fetch_benchmarking_data_with_metadata( + self, + response: gkerecommender.FetchBenchmarkingDataResponse, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchBenchmarkingDataResponse, + Sequence[Tuple[str, Union[str, bytes]]], + ]: + """Post-rpc interceptor for fetch_benchmarking_data + + Override in a subclass to read or manipulate the response or metadata after it + is returned by the GkeInferenceQuickstart server but before it is returned to user code. + + We recommend only using this `post_fetch_benchmarking_data_with_metadata` + interceptor in new development instead of the `post_fetch_benchmarking_data` interceptor. + When both interceptors are used, this `post_fetch_benchmarking_data_with_metadata` interceptor runs after the + `post_fetch_benchmarking_data` interceptor. The (possibly modified) response returned by + `post_fetch_benchmarking_data` will be passed to + `post_fetch_benchmarking_data_with_metadata`. + """ + return response, metadata + + def pre_fetch_models( + self, + request: gkerecommender.FetchModelsRequest, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchModelsRequest, Sequence[Tuple[str, Union[str, bytes]]] + ]: + """Pre-rpc interceptor for fetch_models + + Override in a subclass to manipulate the request or metadata + before they are sent to the GkeInferenceQuickstart server. + """ + return request, metadata + + def post_fetch_models( + self, response: gkerecommender.FetchModelsResponse + ) -> gkerecommender.FetchModelsResponse: + """Post-rpc interceptor for fetch_models + + DEPRECATED. Please use the `post_fetch_models_with_metadata` + interceptor instead. + + Override in a subclass to read or manipulate the response + after it is returned by the GkeInferenceQuickstart server but before + it is returned to user code. This `post_fetch_models` interceptor runs + before the `post_fetch_models_with_metadata` interceptor. + """ + return response + + def post_fetch_models_with_metadata( + self, + response: gkerecommender.FetchModelsResponse, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchModelsResponse, Sequence[Tuple[str, Union[str, bytes]]] + ]: + """Post-rpc interceptor for fetch_models + + Override in a subclass to read or manipulate the response or metadata after it + is returned by the GkeInferenceQuickstart server but before it is returned to user code. + + We recommend only using this `post_fetch_models_with_metadata` + interceptor in new development instead of the `post_fetch_models` interceptor. + When both interceptors are used, this `post_fetch_models_with_metadata` interceptor runs after the + `post_fetch_models` interceptor. The (possibly modified) response returned by + `post_fetch_models` will be passed to + `post_fetch_models_with_metadata`. + """ + return response, metadata + + def pre_fetch_model_servers( + self, + request: gkerecommender.FetchModelServersRequest, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchModelServersRequest, Sequence[Tuple[str, Union[str, bytes]]] + ]: + """Pre-rpc interceptor for fetch_model_servers + + Override in a subclass to manipulate the request or metadata + before they are sent to the GkeInferenceQuickstart server. + """ + return request, metadata + + def post_fetch_model_servers( + self, response: gkerecommender.FetchModelServersResponse + ) -> gkerecommender.FetchModelServersResponse: + """Post-rpc interceptor for fetch_model_servers + + DEPRECATED. Please use the `post_fetch_model_servers_with_metadata` + interceptor instead. + + Override in a subclass to read or manipulate the response + after it is returned by the GkeInferenceQuickstart server but before + it is returned to user code. This `post_fetch_model_servers` interceptor runs + before the `post_fetch_model_servers_with_metadata` interceptor. + """ + return response + + def post_fetch_model_servers_with_metadata( + self, + response: gkerecommender.FetchModelServersResponse, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchModelServersResponse, + Sequence[Tuple[str, Union[str, bytes]]], + ]: + """Post-rpc interceptor for fetch_model_servers + + Override in a subclass to read or manipulate the response or metadata after it + is returned by the GkeInferenceQuickstart server but before it is returned to user code. + + We recommend only using this `post_fetch_model_servers_with_metadata` + interceptor in new development instead of the `post_fetch_model_servers` interceptor. + When both interceptors are used, this `post_fetch_model_servers_with_metadata` interceptor runs after the + `post_fetch_model_servers` interceptor. The (possibly modified) response returned by + `post_fetch_model_servers` will be passed to + `post_fetch_model_servers_with_metadata`. + """ + return response, metadata + + def pre_fetch_model_server_versions( + self, + request: gkerecommender.FetchModelServerVersionsRequest, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchModelServerVersionsRequest, + Sequence[Tuple[str, Union[str, bytes]]], + ]: + """Pre-rpc interceptor for fetch_model_server_versions + + Override in a subclass to manipulate the request or metadata + before they are sent to the GkeInferenceQuickstart server. + """ + return request, metadata + + def post_fetch_model_server_versions( + self, response: gkerecommender.FetchModelServerVersionsResponse + ) -> gkerecommender.FetchModelServerVersionsResponse: + """Post-rpc interceptor for fetch_model_server_versions + + DEPRECATED. Please use the `post_fetch_model_server_versions_with_metadata` + interceptor instead. + + Override in a subclass to read or manipulate the response + after it is returned by the GkeInferenceQuickstart server but before + it is returned to user code. This `post_fetch_model_server_versions` interceptor runs + before the `post_fetch_model_server_versions_with_metadata` interceptor. + """ + return response + + def post_fetch_model_server_versions_with_metadata( + self, + response: gkerecommender.FetchModelServerVersionsResponse, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchModelServerVersionsResponse, + Sequence[Tuple[str, Union[str, bytes]]], + ]: + """Post-rpc interceptor for fetch_model_server_versions + + Override in a subclass to read or manipulate the response or metadata after it + is returned by the GkeInferenceQuickstart server but before it is returned to user code. + + We recommend only using this `post_fetch_model_server_versions_with_metadata` + interceptor in new development instead of the `post_fetch_model_server_versions` interceptor. + When both interceptors are used, this `post_fetch_model_server_versions_with_metadata` interceptor runs after the + `post_fetch_model_server_versions` interceptor. The (possibly modified) response returned by + `post_fetch_model_server_versions` will be passed to + `post_fetch_model_server_versions_with_metadata`. + """ + return response, metadata + + def pre_fetch_profiles( + self, + request: gkerecommender.FetchProfilesRequest, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchProfilesRequest, Sequence[Tuple[str, Union[str, bytes]]] + ]: + """Pre-rpc interceptor for fetch_profiles + + Override in a subclass to manipulate the request or metadata + before they are sent to the GkeInferenceQuickstart server. + """ + return request, metadata + + def post_fetch_profiles( + self, response: gkerecommender.FetchProfilesResponse + ) -> gkerecommender.FetchProfilesResponse: + """Post-rpc interceptor for fetch_profiles + + DEPRECATED. Please use the `post_fetch_profiles_with_metadata` + interceptor instead. + + Override in a subclass to read or manipulate the response + after it is returned by the GkeInferenceQuickstart server but before + it is returned to user code. This `post_fetch_profiles` interceptor runs + before the `post_fetch_profiles_with_metadata` interceptor. + """ + return response + + def post_fetch_profiles_with_metadata( + self, + response: gkerecommender.FetchProfilesResponse, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.FetchProfilesResponse, Sequence[Tuple[str, Union[str, bytes]]] + ]: + """Post-rpc interceptor for fetch_profiles + + Override in a subclass to read or manipulate the response or metadata after it + is returned by the GkeInferenceQuickstart server but before it is returned to user code. + + We recommend only using this `post_fetch_profiles_with_metadata` + interceptor in new development instead of the `post_fetch_profiles` interceptor. + When both interceptors are used, this `post_fetch_profiles_with_metadata` interceptor runs after the + `post_fetch_profiles` interceptor. The (possibly modified) response returned by + `post_fetch_profiles` will be passed to + `post_fetch_profiles_with_metadata`. + """ + return response, metadata + + def pre_generate_optimized_manifest( + self, + request: gkerecommender.GenerateOptimizedManifestRequest, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.GenerateOptimizedManifestRequest, + Sequence[Tuple[str, Union[str, bytes]]], + ]: + """Pre-rpc interceptor for generate_optimized_manifest + + Override in a subclass to manipulate the request or metadata + before they are sent to the GkeInferenceQuickstart server. + """ + return request, metadata + + def post_generate_optimized_manifest( + self, response: gkerecommender.GenerateOptimizedManifestResponse + ) -> gkerecommender.GenerateOptimizedManifestResponse: + """Post-rpc interceptor for generate_optimized_manifest + + DEPRECATED. Please use the `post_generate_optimized_manifest_with_metadata` + interceptor instead. + + Override in a subclass to read or manipulate the response + after it is returned by the GkeInferenceQuickstart server but before + it is returned to user code. This `post_generate_optimized_manifest` interceptor runs + before the `post_generate_optimized_manifest_with_metadata` interceptor. + """ + return response + + def post_generate_optimized_manifest_with_metadata( + self, + response: gkerecommender.GenerateOptimizedManifestResponse, + metadata: Sequence[Tuple[str, Union[str, bytes]]], + ) -> Tuple[ + gkerecommender.GenerateOptimizedManifestResponse, + Sequence[Tuple[str, Union[str, bytes]]], + ]: + """Post-rpc interceptor for generate_optimized_manifest + + Override in a subclass to read or manipulate the response or metadata after it + is returned by the GkeInferenceQuickstart server but before it is returned to user code. + + We recommend only using this `post_generate_optimized_manifest_with_metadata` + interceptor in new development instead of the `post_generate_optimized_manifest` interceptor. + When both interceptors are used, this `post_generate_optimized_manifest_with_metadata` interceptor runs after the + `post_generate_optimized_manifest` interceptor. The (possibly modified) response returned by + `post_generate_optimized_manifest` will be passed to + `post_generate_optimized_manifest_with_metadata`. + """ + return response, metadata + + +@dataclasses.dataclass +class GkeInferenceQuickstartRestStub: + _session: AuthorizedSession + _host: str + _interceptor: GkeInferenceQuickstartRestInterceptor + + +class GkeInferenceQuickstartRestTransport(_BaseGkeInferenceQuickstartRestTransport): + """REST backend synchronous transport for GkeInferenceQuickstart. + + GKE Inference Quickstart (GIQ) service provides profiles with + performance metrics for popular models and model servers across + multiple accelerators. These profiles help generate optimized + best practices for running inference on GKE. + + This class defines the same methods as the primary client, so the + primary client can load the underlying transport implementation + and call it. + + It sends JSON representations of protocol buffers over HTTP/1.1 + """ + + def __init__( + self, + *, + host: str = "gkerecommender.googleapis.com", + credentials: Optional[ga_credentials.Credentials] = None, + credentials_file: Optional[str] = None, + scopes: Optional[Sequence[str]] = None, + client_cert_source_for_mtls: Optional[Callable[[], Tuple[bytes, bytes]]] = None, + quota_project_id: Optional[str] = None, + client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, + always_use_jwt_access: Optional[bool] = False, + url_scheme: str = "https", + interceptor: Optional[GkeInferenceQuickstartRestInterceptor] = None, + api_audience: Optional[str] = None, + ) -> None: + """Instantiate the transport. + + Args: + host (Optional[str]): + The hostname to connect to (default: 'gkerecommender.googleapis.com'). + credentials (Optional[google.auth.credentials.Credentials]): The + authorization credentials to attach to requests. These + credentials identify the application to the service; if none + are specified, the client will attempt to ascertain the + credentials from the environment. + + credentials_file (Optional[str]): A file with credentials that can + be loaded with :func:`google.auth.load_credentials_from_file`. + This argument is ignored if ``channel`` is provided. + scopes (Optional(Sequence[str])): A list of scopes. This argument is + ignored if ``channel`` is provided. + client_cert_source_for_mtls (Callable[[], Tuple[bytes, bytes]]): Client + certificate to configure mutual TLS HTTP channel. It is ignored + if ``channel`` is provided. + quota_project_id (Optional[str]): An optional project to use for billing + and quota. + client_info (google.api_core.gapic_v1.client_info.ClientInfo): + The client info used to send a user-agent string along with + API requests. If ``None``, then default info will be used. + Generally, you only need to set this if you are developing + your own client library. + always_use_jwt_access (Optional[bool]): Whether self signed JWT should + be used for service account credentials. + url_scheme: the protocol scheme for the API endpoint. Normally + "https", but for testing or local servers, + "http" can be specified. + """ + # Run the base constructor + # TODO(yon-mg): resolve other ctor params i.e. scopes, quota, etc. + # TODO: When custom host (api_endpoint) is set, `scopes` must *also* be set on the + # credentials object + super().__init__( + host=host, + credentials=credentials, + client_info=client_info, + always_use_jwt_access=always_use_jwt_access, + url_scheme=url_scheme, + api_audience=api_audience, + ) + self._session = AuthorizedSession( + self._credentials, default_host=self.DEFAULT_HOST + ) + if client_cert_source_for_mtls: + self._session.configure_mtls_channel(client_cert_source_for_mtls) + self._interceptor = interceptor or GkeInferenceQuickstartRestInterceptor() + self._prep_wrapped_messages(client_info) + + class _FetchBenchmarkingData( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData, + GkeInferenceQuickstartRestStub, + ): + def __hash__(self): + return hash("GkeInferenceQuickstartRestTransport.FetchBenchmarkingData") + + @staticmethod + def _get_response( + host, + metadata, + query_params, + session, + timeout, + transcoded_request, + body=None, + ): + uri = transcoded_request["uri"] + method = transcoded_request["method"] + headers = dict(metadata) + headers["Content-Type"] = "application/json" + response = getattr(session, method)( + "{host}{uri}".format(host=host, uri=uri), + timeout=timeout, + headers=headers, + params=rest_helpers.flatten_query_params(query_params, strict=True), + data=body, + ) + return response + + def __call__( + self, + request: gkerecommender.FetchBenchmarkingDataRequest, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Optional[float] = None, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.FetchBenchmarkingDataResponse: + r"""Call the fetch benchmarking data method over HTTP. + + Args: + request (~.gkerecommender.FetchBenchmarkingDataRequest): + The request object. Request message for + [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + ~.gkerecommender.FetchBenchmarkingDataResponse: + Response message for + [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. + + """ + + http_options = ( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_http_options() + ) + + request, metadata = self._interceptor.pre_fetch_benchmarking_data( + request, metadata + ) + transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_transcoded_request( + http_options, request + ) + + body = _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_request_body_json( + transcoded_request + ) + + # Jsonify the query params + query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_query_params_json( + transcoded_request + ) + + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + request_url = "{host}{uri}".format( + host=self._host, uri=transcoded_request["uri"] + ) + method = transcoded_request["method"] + try: + request_payload = type(request).to_json(request) + except: + request_payload = None + http_request = { + "payload": request_payload, + "requestMethod": method, + "requestUrl": request_url, + "headers": dict(metadata), + } + _LOGGER.debug( + f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchBenchmarkingData", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchBenchmarkingData", + "httpRequest": http_request, + "metadata": http_request["headers"], + }, + ) + + # Send the request + response = GkeInferenceQuickstartRestTransport._FetchBenchmarkingData._get_response( + self._host, + metadata, + query_params, + self._session, + timeout, + transcoded_request, + body, + ) + + # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception + # subclass. + if response.status_code >= 400: + raise core_exceptions.from_http_response(response) + + # Return the response + resp = gkerecommender.FetchBenchmarkingDataResponse() + pb_resp = gkerecommender.FetchBenchmarkingDataResponse.pb(resp) + + json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) + + resp = self._interceptor.post_fetch_benchmarking_data(resp) + response_metadata = [(k, str(v)) for k, v in response.headers.items()] + resp, _ = self._interceptor.post_fetch_benchmarking_data_with_metadata( + resp, response_metadata + ) + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + try: + response_payload = ( + gkerecommender.FetchBenchmarkingDataResponse.to_json(response) + ) + except: + response_payload = None + http_response = { + "payload": response_payload, + "headers": dict(response.headers), + "status": response.status_code, + } + _LOGGER.debug( + "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_benchmarking_data", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchBenchmarkingData", + "metadata": http_response["headers"], + "httpResponse": http_response, + }, + ) + return resp + + class _FetchModels( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchModels, + GkeInferenceQuickstartRestStub, + ): + def __hash__(self): + return hash("GkeInferenceQuickstartRestTransport.FetchModels") + + @staticmethod + def _get_response( + host, + metadata, + query_params, + session, + timeout, + transcoded_request, + body=None, + ): + uri = transcoded_request["uri"] + method = transcoded_request["method"] + headers = dict(metadata) + headers["Content-Type"] = "application/json" + response = getattr(session, method)( + "{host}{uri}".format(host=host, uri=uri), + timeout=timeout, + headers=headers, + params=rest_helpers.flatten_query_params(query_params, strict=True), + ) + return response + + def __call__( + self, + request: gkerecommender.FetchModelsRequest, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Optional[float] = None, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.FetchModelsResponse: + r"""Call the fetch models method over HTTP. + + Args: + request (~.gkerecommender.FetchModelsRequest): + The request object. Request message for + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + ~.gkerecommender.FetchModelsResponse: + Response message for + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. + + """ + + http_options = ( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchModels._get_http_options() + ) + + request, metadata = self._interceptor.pre_fetch_models(request, metadata) + transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModels._get_transcoded_request( + http_options, request + ) + + # Jsonify the query params + query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModels._get_query_params_json( + transcoded_request + ) + + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + request_url = "{host}{uri}".format( + host=self._host, uri=transcoded_request["uri"] + ) + method = transcoded_request["method"] + try: + request_payload = type(request).to_json(request) + except: + request_payload = None + http_request = { + "payload": request_payload, + "requestMethod": method, + "requestUrl": request_url, + "headers": dict(metadata), + } + _LOGGER.debug( + f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchModels", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchModels", + "httpRequest": http_request, + "metadata": http_request["headers"], + }, + ) + + # Send the request + response = GkeInferenceQuickstartRestTransport._FetchModels._get_response( + self._host, + metadata, + query_params, + self._session, + timeout, + transcoded_request, + ) + + # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception + # subclass. + if response.status_code >= 400: + raise core_exceptions.from_http_response(response) + + # Return the response + resp = gkerecommender.FetchModelsResponse() + pb_resp = gkerecommender.FetchModelsResponse.pb(resp) + + json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) + + resp = self._interceptor.post_fetch_models(resp) + response_metadata = [(k, str(v)) for k, v in response.headers.items()] + resp, _ = self._interceptor.post_fetch_models_with_metadata( + resp, response_metadata + ) + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + try: + response_payload = gkerecommender.FetchModelsResponse.to_json( + response + ) + except: + response_payload = None + http_response = { + "payload": response_payload, + "headers": dict(response.headers), + "status": response.status_code, + } + _LOGGER.debug( + "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_models", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchModels", + "metadata": http_response["headers"], + "httpResponse": http_response, + }, + ) + return resp + + class _FetchModelServers( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers, + GkeInferenceQuickstartRestStub, + ): + def __hash__(self): + return hash("GkeInferenceQuickstartRestTransport.FetchModelServers") + + @staticmethod + def _get_response( + host, + metadata, + query_params, + session, + timeout, + transcoded_request, + body=None, + ): + uri = transcoded_request["uri"] + method = transcoded_request["method"] + headers = dict(metadata) + headers["Content-Type"] = "application/json" + response = getattr(session, method)( + "{host}{uri}".format(host=host, uri=uri), + timeout=timeout, + headers=headers, + params=rest_helpers.flatten_query_params(query_params, strict=True), + ) + return response + + def __call__( + self, + request: gkerecommender.FetchModelServersRequest, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Optional[float] = None, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.FetchModelServersResponse: + r"""Call the fetch model servers method over HTTP. + + Args: + request (~.gkerecommender.FetchModelServersRequest): + The request object. Request message for + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + ~.gkerecommender.FetchModelServersResponse: + Response message for + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. + + """ + + http_options = ( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers._get_http_options() + ) + + request, metadata = self._interceptor.pre_fetch_model_servers( + request, metadata + ) + transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers._get_transcoded_request( + http_options, request + ) + + # Jsonify the query params + query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers._get_query_params_json( + transcoded_request + ) + + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + request_url = "{host}{uri}".format( + host=self._host, uri=transcoded_request["uri"] + ) + method = transcoded_request["method"] + try: + request_payload = type(request).to_json(request) + except: + request_payload = None + http_request = { + "payload": request_payload, + "requestMethod": method, + "requestUrl": request_url, + "headers": dict(metadata), + } + _LOGGER.debug( + f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchModelServers", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchModelServers", + "httpRequest": http_request, + "metadata": http_request["headers"], + }, + ) + + # Send the request + response = ( + GkeInferenceQuickstartRestTransport._FetchModelServers._get_response( + self._host, + metadata, + query_params, + self._session, + timeout, + transcoded_request, + ) + ) + + # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception + # subclass. + if response.status_code >= 400: + raise core_exceptions.from_http_response(response) + + # Return the response + resp = gkerecommender.FetchModelServersResponse() + pb_resp = gkerecommender.FetchModelServersResponse.pb(resp) + + json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) + + resp = self._interceptor.post_fetch_model_servers(resp) + response_metadata = [(k, str(v)) for k, v in response.headers.items()] + resp, _ = self._interceptor.post_fetch_model_servers_with_metadata( + resp, response_metadata + ) + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + try: + response_payload = gkerecommender.FetchModelServersResponse.to_json( + response + ) + except: + response_payload = None + http_response = { + "payload": response_payload, + "headers": dict(response.headers), + "status": response.status_code, + } + _LOGGER.debug( + "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_model_servers", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchModelServers", + "metadata": http_response["headers"], + "httpResponse": http_response, + }, + ) + return resp + + class _FetchModelServerVersions( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions, + GkeInferenceQuickstartRestStub, + ): + def __hash__(self): + return hash("GkeInferenceQuickstartRestTransport.FetchModelServerVersions") + + @staticmethod + def _get_response( + host, + metadata, + query_params, + session, + timeout, + transcoded_request, + body=None, + ): + uri = transcoded_request["uri"] + method = transcoded_request["method"] + headers = dict(metadata) + headers["Content-Type"] = "application/json" + response = getattr(session, method)( + "{host}{uri}".format(host=host, uri=uri), + timeout=timeout, + headers=headers, + params=rest_helpers.flatten_query_params(query_params, strict=True), + ) + return response + + def __call__( + self, + request: gkerecommender.FetchModelServerVersionsRequest, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Optional[float] = None, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.FetchModelServerVersionsResponse: + r"""Call the fetch model server + versions method over HTTP. + + Args: + request (~.gkerecommender.FetchModelServerVersionsRequest): + The request object. Request message for + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + ~.gkerecommender.FetchModelServerVersionsResponse: + Response message for + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. + + """ + + http_options = ( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions._get_http_options() + ) + + request, metadata = self._interceptor.pre_fetch_model_server_versions( + request, metadata + ) + transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions._get_transcoded_request( + http_options, request + ) + + # Jsonify the query params + query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions._get_query_params_json( + transcoded_request + ) + + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + request_url = "{host}{uri}".format( + host=self._host, uri=transcoded_request["uri"] + ) + method = transcoded_request["method"] + try: + request_payload = type(request).to_json(request) + except: + request_payload = None + http_request = { + "payload": request_payload, + "requestMethod": method, + "requestUrl": request_url, + "headers": dict(metadata), + } + _LOGGER.debug( + f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchModelServerVersions", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchModelServerVersions", + "httpRequest": http_request, + "metadata": http_request["headers"], + }, + ) + + # Send the request + response = GkeInferenceQuickstartRestTransport._FetchModelServerVersions._get_response( + self._host, + metadata, + query_params, + self._session, + timeout, + transcoded_request, + ) + + # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception + # subclass. + if response.status_code >= 400: + raise core_exceptions.from_http_response(response) + + # Return the response + resp = gkerecommender.FetchModelServerVersionsResponse() + pb_resp = gkerecommender.FetchModelServerVersionsResponse.pb(resp) + + json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) + + resp = self._interceptor.post_fetch_model_server_versions(resp) + response_metadata = [(k, str(v)) for k, v in response.headers.items()] + resp, _ = self._interceptor.post_fetch_model_server_versions_with_metadata( + resp, response_metadata + ) + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + try: + response_payload = ( + gkerecommender.FetchModelServerVersionsResponse.to_json( + response + ) + ) + except: + response_payload = None + http_response = { + "payload": response_payload, + "headers": dict(response.headers), + "status": response.status_code, + } + _LOGGER.debug( + "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_model_server_versions", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchModelServerVersions", + "metadata": http_response["headers"], + "httpResponse": http_response, + }, + ) + return resp + + class _FetchProfiles( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles, + GkeInferenceQuickstartRestStub, + ): + def __hash__(self): + return hash("GkeInferenceQuickstartRestTransport.FetchProfiles") + + @staticmethod + def _get_response( + host, + metadata, + query_params, + session, + timeout, + transcoded_request, + body=None, + ): + uri = transcoded_request["uri"] + method = transcoded_request["method"] + headers = dict(metadata) + headers["Content-Type"] = "application/json" + response = getattr(session, method)( + "{host}{uri}".format(host=host, uri=uri), + timeout=timeout, + headers=headers, + params=rest_helpers.flatten_query_params(query_params, strict=True), + data=body, + ) + return response + + def __call__( + self, + request: gkerecommender.FetchProfilesRequest, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Optional[float] = None, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.FetchProfilesResponse: + r"""Call the fetch profiles method over HTTP. + + Args: + request (~.gkerecommender.FetchProfilesRequest): + The request object. Request message for + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + ~.gkerecommender.FetchProfilesResponse: + Response message for + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + + """ + + http_options = ( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles._get_http_options() + ) + + request, metadata = self._interceptor.pre_fetch_profiles(request, metadata) + transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles._get_transcoded_request( + http_options, request + ) + + body = _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles._get_request_body_json( + transcoded_request + ) + + # Jsonify the query params + query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles._get_query_params_json( + transcoded_request + ) + + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + request_url = "{host}{uri}".format( + host=self._host, uri=transcoded_request["uri"] + ) + method = transcoded_request["method"] + try: + request_payload = type(request).to_json(request) + except: + request_payload = None + http_request = { + "payload": request_payload, + "requestMethod": method, + "requestUrl": request_url, + "headers": dict(metadata), + } + _LOGGER.debug( + f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchProfiles", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchProfiles", + "httpRequest": http_request, + "metadata": http_request["headers"], + }, + ) + + # Send the request + response = GkeInferenceQuickstartRestTransport._FetchProfiles._get_response( + self._host, + metadata, + query_params, + self._session, + timeout, + transcoded_request, + body, + ) + + # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception + # subclass. + if response.status_code >= 400: + raise core_exceptions.from_http_response(response) + + # Return the response + resp = gkerecommender.FetchProfilesResponse() + pb_resp = gkerecommender.FetchProfilesResponse.pb(resp) + + json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) + + resp = self._interceptor.post_fetch_profiles(resp) + response_metadata = [(k, str(v)) for k, v in response.headers.items()] + resp, _ = self._interceptor.post_fetch_profiles_with_metadata( + resp, response_metadata + ) + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + try: + response_payload = gkerecommender.FetchProfilesResponse.to_json( + response + ) + except: + response_payload = None + http_response = { + "payload": response_payload, + "headers": dict(response.headers), + "status": response.status_code, + } + _LOGGER.debug( + "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_profiles", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "FetchProfiles", + "metadata": http_response["headers"], + "httpResponse": http_response, + }, + ) + return resp + + class _GenerateOptimizedManifest( + _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest, + GkeInferenceQuickstartRestStub, + ): + def __hash__(self): + return hash("GkeInferenceQuickstartRestTransport.GenerateOptimizedManifest") + + @staticmethod + def _get_response( + host, + metadata, + query_params, + session, + timeout, + transcoded_request, + body=None, + ): + uri = transcoded_request["uri"] + method = transcoded_request["method"] + headers = dict(metadata) + headers["Content-Type"] = "application/json" + response = getattr(session, method)( + "{host}{uri}".format(host=host, uri=uri), + timeout=timeout, + headers=headers, + params=rest_helpers.flatten_query_params(query_params, strict=True), + data=body, + ) + return response + + def __call__( + self, + request: gkerecommender.GenerateOptimizedManifestRequest, + *, + retry: OptionalRetry = gapic_v1.method.DEFAULT, + timeout: Optional[float] = None, + metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), + ) -> gkerecommender.GenerateOptimizedManifestResponse: + r"""Call the generate optimized + manifest method over HTTP. + + Args: + request (~.gkerecommender.GenerateOptimizedManifestRequest): + The request object. Request message for + [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. + retry (google.api_core.retry.Retry): Designation of what errors, if any, + should be retried. + timeout (float): The timeout for this request. + metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be + sent along with the request as metadata. Normally, each value must be of type `str`, + but for metadata keys ending with the suffix `-bin`, the corresponding values must + be of type `bytes`. + + Returns: + ~.gkerecommender.GenerateOptimizedManifestResponse: + Response message for + [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. + + """ + + http_options = ( + _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_http_options() + ) + + request, metadata = self._interceptor.pre_generate_optimized_manifest( + request, metadata + ) + transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_transcoded_request( + http_options, request + ) + + body = _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_request_body_json( + transcoded_request + ) + + # Jsonify the query params + query_params = _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_query_params_json( + transcoded_request + ) + + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + request_url = "{host}{uri}".format( + host=self._host, uri=transcoded_request["uri"] + ) + method = transcoded_request["method"] + try: + request_payload = type(request).to_json(request) + except: + request_payload = None + http_request = { + "payload": request_payload, + "requestMethod": method, + "requestUrl": request_url, + "headers": dict(metadata), + } + _LOGGER.debug( + f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.GenerateOptimizedManifest", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "GenerateOptimizedManifest", + "httpRequest": http_request, + "metadata": http_request["headers"], + }, + ) + + # Send the request + response = GkeInferenceQuickstartRestTransport._GenerateOptimizedManifest._get_response( + self._host, + metadata, + query_params, + self._session, + timeout, + transcoded_request, + body, + ) + + # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception + # subclass. + if response.status_code >= 400: + raise core_exceptions.from_http_response(response) + + # Return the response + resp = gkerecommender.GenerateOptimizedManifestResponse() + pb_resp = gkerecommender.GenerateOptimizedManifestResponse.pb(resp) + + json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) + + resp = self._interceptor.post_generate_optimized_manifest(resp) + response_metadata = [(k, str(v)) for k, v in response.headers.items()] + resp, _ = self._interceptor.post_generate_optimized_manifest_with_metadata( + resp, response_metadata + ) + if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( + logging.DEBUG + ): # pragma: NO COVER + try: + response_payload = ( + gkerecommender.GenerateOptimizedManifestResponse.to_json( + response + ) + ) + except: + response_payload = None + http_response = { + "payload": response_payload, + "headers": dict(response.headers), + "status": response.status_code, + } + _LOGGER.debug( + "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.generate_optimized_manifest", + extra={ + "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "rpcName": "GenerateOptimizedManifest", + "metadata": http_response["headers"], + "httpResponse": http_response, + }, + ) + return resp + + @property + def fetch_benchmarking_data( + self, + ) -> Callable[ + [gkerecommender.FetchBenchmarkingDataRequest], + gkerecommender.FetchBenchmarkingDataResponse, + ]: + # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. + # In C++ this would require a dynamic_cast + return self._FetchBenchmarkingData(self._session, self._host, self._interceptor) # type: ignore + + @property + def fetch_models( + self, + ) -> Callable[ + [gkerecommender.FetchModelsRequest], gkerecommender.FetchModelsResponse + ]: + # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. + # In C++ this would require a dynamic_cast + return self._FetchModels(self._session, self._host, self._interceptor) # type: ignore + + @property + def fetch_model_servers( + self, + ) -> Callable[ + [gkerecommender.FetchModelServersRequest], + gkerecommender.FetchModelServersResponse, + ]: + # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. + # In C++ this would require a dynamic_cast + return self._FetchModelServers(self._session, self._host, self._interceptor) # type: ignore + + @property + def fetch_model_server_versions( + self, + ) -> Callable[ + [gkerecommender.FetchModelServerVersionsRequest], + gkerecommender.FetchModelServerVersionsResponse, + ]: + # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. + # In C++ this would require a dynamic_cast + return self._FetchModelServerVersions(self._session, self._host, self._interceptor) # type: ignore + + @property + def fetch_profiles( + self, + ) -> Callable[ + [gkerecommender.FetchProfilesRequest], gkerecommender.FetchProfilesResponse + ]: + # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. + # In C++ this would require a dynamic_cast + return self._FetchProfiles(self._session, self._host, self._interceptor) # type: ignore + + @property + def generate_optimized_manifest( + self, + ) -> Callable[ + [gkerecommender.GenerateOptimizedManifestRequest], + gkerecommender.GenerateOptimizedManifestResponse, + ]: + # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. + # In C++ this would require a dynamic_cast + return self._GenerateOptimizedManifest(self._session, self._host, self._interceptor) # type: ignore + + @property + def kind(self) -> str: + return "rest" + + def close(self): + self._session.close() + + +__all__ = ("GkeInferenceQuickstartRestTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest_base.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest_base.py new file mode 100644 index 000000000000..07ac486ec876 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest_base.py @@ -0,0 +1,378 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import json # type: ignore +import re +from typing import Any, Callable, Dict, List, Optional, Sequence, Tuple, Union + +from google.api_core import gapic_v1, path_template +from google.protobuf import json_format + +from google.cloud.gkerecommender_v1.types import gkerecommender + +from .base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport + + +class _BaseGkeInferenceQuickstartRestTransport(GkeInferenceQuickstartTransport): + """Base REST backend transport for GkeInferenceQuickstart. + + Note: This class is not meant to be used directly. Use its sync and + async sub-classes instead. + + This class defines the same methods as the primary client, so the + primary client can load the underlying transport implementation + and call it. + + It sends JSON representations of protocol buffers over HTTP/1.1 + """ + + def __init__( + self, + *, + host: str = "gkerecommender.googleapis.com", + credentials: Optional[Any] = None, + client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, + always_use_jwt_access: Optional[bool] = False, + url_scheme: str = "https", + api_audience: Optional[str] = None, + ) -> None: + """Instantiate the transport. + Args: + host (Optional[str]): + The hostname to connect to (default: 'gkerecommender.googleapis.com'). + credentials (Optional[Any]): The + authorization credentials to attach to requests. These + credentials identify the application to the service; if none + are specified, the client will attempt to ascertain the + credentials from the environment. + client_info (google.api_core.gapic_v1.client_info.ClientInfo): + The client info used to send a user-agent string along with + API requests. If ``None``, then default info will be used. + Generally, you only need to set this if you are developing + your own client library. + always_use_jwt_access (Optional[bool]): Whether self signed JWT should + be used for service account credentials. + url_scheme: the protocol scheme for the API endpoint. Normally + "https", but for testing or local servers, + "http" can be specified. + """ + # Run the base constructor + maybe_url_match = re.match("^(?Phttp(?:s)?://)?(?P.*)$", host) + if maybe_url_match is None: + raise ValueError( + f"Unexpected hostname structure: {host}" + ) # pragma: NO COVER + + url_match_items = maybe_url_match.groupdict() + + host = f"{url_scheme}://{host}" if not url_match_items["scheme"] else host + + super().__init__( + host=host, + credentials=credentials, + client_info=client_info, + always_use_jwt_access=always_use_jwt_access, + api_audience=api_audience, + ) + + class _BaseFetchBenchmarkingData: + def __hash__(self): # pragma: NO COVER + return NotImplementedError("__hash__ must be implemented.") + + __REQUIRED_FIELDS_DEFAULT_VALUES: Dict[str, Any] = {} + + @classmethod + def _get_unset_required_fields(cls, message_dict): + return { + k: v + for k, v in cls.__REQUIRED_FIELDS_DEFAULT_VALUES.items() + if k not in message_dict + } + + @staticmethod + def _get_http_options(): + http_options: List[Dict[str, str]] = [ + { + "method": "post", + "uri": "/v1/benchmarkingData:fetch", + "body": "*", + }, + ] + return http_options + + @staticmethod + def _get_transcoded_request(http_options, request): + pb_request = gkerecommender.FetchBenchmarkingDataRequest.pb(request) + transcoded_request = path_template.transcode(http_options, pb_request) + return transcoded_request + + @staticmethod + def _get_request_body_json(transcoded_request): + # Jsonify the request body + + body = json_format.MessageToJson( + transcoded_request["body"], use_integers_for_enums=True + ) + return body + + @staticmethod + def _get_query_params_json(transcoded_request): + query_params = json.loads( + json_format.MessageToJson( + transcoded_request["query_params"], + use_integers_for_enums=True, + ) + ) + query_params.update( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_unset_required_fields( + query_params + ) + ) + + query_params["$alt"] = "json;enum-encoding=int" + return query_params + + class _BaseFetchModels: + def __hash__(self): # pragma: NO COVER + return NotImplementedError("__hash__ must be implemented.") + + @staticmethod + def _get_http_options(): + http_options: List[Dict[str, str]] = [ + { + "method": "get", + "uri": "/v1/models:fetch", + }, + ] + return http_options + + @staticmethod + def _get_transcoded_request(http_options, request): + pb_request = gkerecommender.FetchModelsRequest.pb(request) + transcoded_request = path_template.transcode(http_options, pb_request) + return transcoded_request + + @staticmethod + def _get_query_params_json(transcoded_request): + query_params = json.loads( + json_format.MessageToJson( + transcoded_request["query_params"], + use_integers_for_enums=True, + ) + ) + + query_params["$alt"] = "json;enum-encoding=int" + return query_params + + class _BaseFetchModelServers: + def __hash__(self): # pragma: NO COVER + return NotImplementedError("__hash__ must be implemented.") + + __REQUIRED_FIELDS_DEFAULT_VALUES: Dict[str, Any] = { + "model": "", + } + + @classmethod + def _get_unset_required_fields(cls, message_dict): + return { + k: v + for k, v in cls.__REQUIRED_FIELDS_DEFAULT_VALUES.items() + if k not in message_dict + } + + @staticmethod + def _get_http_options(): + http_options: List[Dict[str, str]] = [ + { + "method": "get", + "uri": "/v1/modelServers:fetch", + }, + ] + return http_options + + @staticmethod + def _get_transcoded_request(http_options, request): + pb_request = gkerecommender.FetchModelServersRequest.pb(request) + transcoded_request = path_template.transcode(http_options, pb_request) + return transcoded_request + + @staticmethod + def _get_query_params_json(transcoded_request): + query_params = json.loads( + json_format.MessageToJson( + transcoded_request["query_params"], + use_integers_for_enums=True, + ) + ) + query_params.update( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers._get_unset_required_fields( + query_params + ) + ) + + query_params["$alt"] = "json;enum-encoding=int" + return query_params + + class _BaseFetchModelServerVersions: + def __hash__(self): # pragma: NO COVER + return NotImplementedError("__hash__ must be implemented.") + + __REQUIRED_FIELDS_DEFAULT_VALUES: Dict[str, Any] = { + "model": "", + "modelServer": "", + } + + @classmethod + def _get_unset_required_fields(cls, message_dict): + return { + k: v + for k, v in cls.__REQUIRED_FIELDS_DEFAULT_VALUES.items() + if k not in message_dict + } + + @staticmethod + def _get_http_options(): + http_options: List[Dict[str, str]] = [ + { + "method": "get", + "uri": "/v1/modelServerVersions:fetch", + }, + ] + return http_options + + @staticmethod + def _get_transcoded_request(http_options, request): + pb_request = gkerecommender.FetchModelServerVersionsRequest.pb(request) + transcoded_request = path_template.transcode(http_options, pb_request) + return transcoded_request + + @staticmethod + def _get_query_params_json(transcoded_request): + query_params = json.loads( + json_format.MessageToJson( + transcoded_request["query_params"], + use_integers_for_enums=True, + ) + ) + query_params.update( + _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions._get_unset_required_fields( + query_params + ) + ) + + query_params["$alt"] = "json;enum-encoding=int" + return query_params + + class _BaseFetchProfiles: + def __hash__(self): # pragma: NO COVER + return NotImplementedError("__hash__ must be implemented.") + + @staticmethod + def _get_http_options(): + http_options: List[Dict[str, str]] = [ + { + "method": "post", + "uri": "/v1/profiles:fetch", + "body": "*", + }, + ] + return http_options + + @staticmethod + def _get_transcoded_request(http_options, request): + pb_request = gkerecommender.FetchProfilesRequest.pb(request) + transcoded_request = path_template.transcode(http_options, pb_request) + return transcoded_request + + @staticmethod + def _get_request_body_json(transcoded_request): + # Jsonify the request body + + body = json_format.MessageToJson( + transcoded_request["body"], use_integers_for_enums=True + ) + return body + + @staticmethod + def _get_query_params_json(transcoded_request): + query_params = json.loads( + json_format.MessageToJson( + transcoded_request["query_params"], + use_integers_for_enums=True, + ) + ) + + query_params["$alt"] = "json;enum-encoding=int" + return query_params + + class _BaseGenerateOptimizedManifest: + def __hash__(self): # pragma: NO COVER + return NotImplementedError("__hash__ must be implemented.") + + __REQUIRED_FIELDS_DEFAULT_VALUES: Dict[str, Any] = {} + + @classmethod + def _get_unset_required_fields(cls, message_dict): + return { + k: v + for k, v in cls.__REQUIRED_FIELDS_DEFAULT_VALUES.items() + if k not in message_dict + } + + @staticmethod + def _get_http_options(): + http_options: List[Dict[str, str]] = [ + { + "method": "post", + "uri": "/v1/optimizedManifest:generate", + "body": "*", + }, + ] + return http_options + + @staticmethod + def _get_transcoded_request(http_options, request): + pb_request = gkerecommender.GenerateOptimizedManifestRequest.pb(request) + transcoded_request = path_template.transcode(http_options, pb_request) + return transcoded_request + + @staticmethod + def _get_request_body_json(transcoded_request): + # Jsonify the request body + + body = json_format.MessageToJson( + transcoded_request["body"], use_integers_for_enums=True + ) + return body + + @staticmethod + def _get_query_params_json(transcoded_request): + query_params = json.loads( + json_format.MessageToJson( + transcoded_request["query_params"], + use_integers_for_enums=True, + ) + ) + query_params.update( + _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_unset_required_fields( + query_params + ) + ) + + query_params["$alt"] = "json;enum-encoding=int" + return query_params + + +__all__ = ("_BaseGkeInferenceQuickstartRestTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/__init__.py new file mode 100644 index 000000000000..bf87f8b02a3e --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/__init__.py @@ -0,0 +1,68 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from .gkerecommender import ( + Amount, + Cost, + FetchBenchmarkingDataRequest, + FetchBenchmarkingDataResponse, + FetchModelServersRequest, + FetchModelServersResponse, + FetchModelServerVersionsRequest, + FetchModelServerVersionsResponse, + FetchModelsRequest, + FetchModelsResponse, + FetchProfilesRequest, + FetchProfilesResponse, + GenerateOptimizedManifestRequest, + GenerateOptimizedManifestResponse, + KubernetesManifest, + MillisecondRange, + ModelServerInfo, + PerformanceRange, + PerformanceRequirements, + PerformanceStats, + Profile, + ResourcesUsed, + StorageConfig, + TokensPerSecondRange, +) + +__all__ = ( + "Amount", + "Cost", + "FetchBenchmarkingDataRequest", + "FetchBenchmarkingDataResponse", + "FetchModelServersRequest", + "FetchModelServersResponse", + "FetchModelServerVersionsRequest", + "FetchModelServerVersionsResponse", + "FetchModelsRequest", + "FetchModelsResponse", + "FetchProfilesRequest", + "FetchProfilesResponse", + "GenerateOptimizedManifestRequest", + "GenerateOptimizedManifestResponse", + "KubernetesManifest", + "MillisecondRange", + "ModelServerInfo", + "PerformanceRange", + "PerformanceRequirements", + "PerformanceStats", + "Profile", + "ResourcesUsed", + "StorageConfig", + "TokensPerSecondRange", +) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/gkerecommender.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/gkerecommender.py new file mode 100644 index 000000000000..cff65826fa75 --- /dev/null +++ b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/gkerecommender.py @@ -0,0 +1,983 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +from __future__ import annotations + +from typing import MutableMapping, MutableSequence + +import proto # type: ignore + +__protobuf__ = proto.module( + package="google.cloud.gkerecommender.v1", + manifest={ + "FetchModelsRequest", + "FetchModelsResponse", + "FetchModelServersRequest", + "FetchModelServersResponse", + "FetchModelServerVersionsRequest", + "FetchModelServerVersionsResponse", + "FetchBenchmarkingDataRequest", + "FetchBenchmarkingDataResponse", + "FetchProfilesRequest", + "PerformanceRequirements", + "Amount", + "Cost", + "TokensPerSecondRange", + "MillisecondRange", + "PerformanceRange", + "FetchProfilesResponse", + "ModelServerInfo", + "ResourcesUsed", + "PerformanceStats", + "Profile", + "GenerateOptimizedManifestRequest", + "KubernetesManifest", + "GenerateOptimizedManifestResponse", + "StorageConfig", + }, +) + + +class FetchModelsRequest(proto.Message): + r"""Request message for + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. + + + .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields + + Attributes: + page_size (int): + Optional. The target number of results to return in a single + response. If not specified, a default value will be chosen + by the service. Note that the response may include a partial + list and a caller should only rely on the response's + [next_page_token][google.cloud.gkerecommender.v1.FetchModelsResponse.next_page_token] + to determine if there are more instances left to be queried. + + This field is a member of `oneof`_ ``_page_size``. + page_token (str): + Optional. The value of + [next_page_token][google.cloud.gkerecommender.v1.FetchModelsResponse.next_page_token] + received from a previous ``FetchModelsRequest`` call. + Provide this to retrieve the subsequent page in a multi-page + list of results. When paginating, all other parameters + provided to ``FetchModelsRequest`` must match the call that + provided the page token. + + This field is a member of `oneof`_ ``_page_token``. + """ + + page_size: int = proto.Field( + proto.INT32, + number=1, + optional=True, + ) + page_token: str = proto.Field( + proto.STRING, + number=2, + optional=True, + ) + + +class FetchModelsResponse(proto.Message): + r"""Response message for + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. + + Attributes: + models (MutableSequence[str]): + Output only. List of available models. Open-source models + follow the Huggingface Hub ``owner/model_name`` format. + next_page_token (str): + Output only. A token which may be sent as + [page_token][FetchModelsResponse.page_token] in a subsequent + ``FetchModelsResponse`` call to retrieve the next page of + results. If this field is omitted or empty, then there are + no more results to return. + """ + + @property + def raw_page(self): + return self + + models: MutableSequence[str] = proto.RepeatedField( + proto.STRING, + number=1, + ) + next_page_token: str = proto.Field( + proto.STRING, + number=2, + ) + + +class FetchModelServersRequest(proto.Message): + r"""Request message for + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. + + + .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields + + Attributes: + model (str): + Required. The model for which to list model servers. + Open-source models follow the Huggingface Hub + ``owner/model_name`` format. Use + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels] + to find available models. + page_size (int): + Optional. The target number of results to return in a single + response. If not specified, a default value will be chosen + by the service. Note that the response may include a partial + list and a caller should only rely on the response's + [next_page_token][google.cloud.gkerecommender.v1.FetchModelServersResponse.next_page_token] + to determine if there are more instances left to be queried. + + This field is a member of `oneof`_ ``_page_size``. + page_token (str): + Optional. The value of + [next_page_token][google.cloud.gkerecommender.v1.FetchModelServersResponse.next_page_token] + received from a previous ``FetchModelServersRequest`` call. + Provide this to retrieve the subsequent page in a multi-page + list of results. When paginating, all other parameters + provided to ``FetchModelServersRequest`` must match the call + that provided the page token. + + This field is a member of `oneof`_ ``_page_token``. + """ + + model: str = proto.Field( + proto.STRING, + number=1, + ) + page_size: int = proto.Field( + proto.INT32, + number=2, + optional=True, + ) + page_token: str = proto.Field( + proto.STRING, + number=3, + optional=True, + ) + + +class FetchModelServersResponse(proto.Message): + r"""Response message for + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. + + Attributes: + model_servers (MutableSequence[str]): + Output only. List of available model servers. Open-source + model servers use simplified, lowercase names (e.g., + ``vllm``). + next_page_token (str): + Output only. A token which may be sent as + [page_token][FetchModelServersResponse.page_token] in a + subsequent ``FetchModelServersResponse`` call to retrieve + the next page of results. If this field is omitted or empty, + then there are no more results to return. + """ + + @property + def raw_page(self): + return self + + model_servers: MutableSequence[str] = proto.RepeatedField( + proto.STRING, + number=1, + ) + next_page_token: str = proto.Field( + proto.STRING, + number=2, + ) + + +class FetchModelServerVersionsRequest(proto.Message): + r"""Request message for + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. + + + .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields + + Attributes: + model (str): + Required. The model for which to list model server versions. + Open-source models follow the Huggingface Hub + ``owner/model_name`` format. Use + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels] + to find available models. + model_server (str): + Required. The model server for which to list versions. + Open-source model servers use simplified, lowercase names + (e.g., ``vllm``). Use + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers] + to find available model servers. + page_size (int): + Optional. The target number of results to return in a single + response. If not specified, a default value will be chosen + by the service. Note that the response may include a partial + list and a caller should only rely on the response's + [next_page_token][google.cloud.gkerecommender.v1.FetchModelServerVersionsResponse.next_page_token] + to determine if there are more instances left to be queried. + + This field is a member of `oneof`_ ``_page_size``. + page_token (str): + Optional. The value of + [next_page_token][google.cloud.gkerecommender.v1.FetchModelServerVersionsResponse.next_page_token] + received from a previous ``FetchModelServerVersionsRequest`` + call. Provide this to retrieve the subsequent page in a + multi-page list of results. When paginating, all other + parameters provided to ``FetchModelServerVersionsRequest`` + must match the call that provided the page token. + + This field is a member of `oneof`_ ``_page_token``. + """ + + model: str = proto.Field( + proto.STRING, + number=1, + ) + model_server: str = proto.Field( + proto.STRING, + number=2, + ) + page_size: int = proto.Field( + proto.INT32, + number=3, + optional=True, + ) + page_token: str = proto.Field( + proto.STRING, + number=4, + optional=True, + ) + + +class FetchModelServerVersionsResponse(proto.Message): + r"""Response message for + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. + + Attributes: + model_server_versions (MutableSequence[str]): + Output only. A list of available model server + versions. + next_page_token (str): + Output only. A token which may be sent as + [page_token][FetchModelServerVersionsResponse.page_token] in + a subsequent ``FetchModelServerVersionsResponse`` call to + retrieve the next page of results. If this field is omitted + or empty, then there are no more results to return. + """ + + @property + def raw_page(self): + return self + + model_server_versions: MutableSequence[str] = proto.RepeatedField( + proto.STRING, + number=1, + ) + next_page_token: str = proto.Field( + proto.STRING, + number=2, + ) + + +class FetchBenchmarkingDataRequest(proto.Message): + r"""Request message for + [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. + + Attributes: + model_server_info (google.cloud.gkerecommender_v1.types.ModelServerInfo): + Required. The model server configuration to get benchmarking + data for. Use + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] + to find valid configurations. + instance_type (str): + Optional. The instance type to filter benchmarking data. + Instance types are in the format ``a2-highgpu-1g``. If not + provided, all instance types for the given profile's + ``model_server_info`` will be returned. Use + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] + to find available instance types. + pricing_model (str): + Optional. The pricing model to use for the benchmarking + data. Defaults to ``spot``. + """ + + model_server_info: "ModelServerInfo" = proto.Field( + proto.MESSAGE, + number=1, + message="ModelServerInfo", + ) + instance_type: str = proto.Field( + proto.STRING, + number=3, + ) + pricing_model: str = proto.Field( + proto.STRING, + number=4, + ) + + +class FetchBenchmarkingDataResponse(proto.Message): + r"""Response message for + [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. + + Attributes: + profile (MutableSequence[google.cloud.gkerecommender_v1.types.Profile]): + Output only. List of profiles containing + their respective benchmarking data. + """ + + profile: MutableSequence["Profile"] = proto.RepeatedField( + proto.MESSAGE, + number=1, + message="Profile", + ) + + +class FetchProfilesRequest(proto.Message): + r"""Request message for + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + + + .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields + + Attributes: + model (str): + Optional. The model to filter profiles by. Open-source + models follow the Huggingface Hub ``owner/model_name`` + format. If not provided, all models are returned. Use + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels] + to find available models. + model_server (str): + Optional. The model server to filter profiles by. If not + provided, all model servers are returned. Use + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers] + to find available model servers for a given model. + model_server_version (str): + Optional. The model server version to filter profiles by. If + not provided, all model server versions are returned. Use + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions] + to find available versions for a given model and server. + performance_requirements (google.cloud.gkerecommender_v1.types.PerformanceRequirements): + Optional. The performance requirements to + filter profiles. Profiles that do not meet these + requirements are filtered out. If not provided, + all profiles are returned. + page_size (int): + Optional. The target number of results to return in a single + response. If not specified, a default value will be chosen + by the service. Note that the response may include a partial + list and a caller should only rely on the response's + [next_page_token][google.cloud.gkerecommender.v1.FetchProfilesResponse.next_page_token] + to determine if there are more instances left to be queried. + + This field is a member of `oneof`_ ``_page_size``. + page_token (str): + Optional. The value of + [next_page_token][google.cloud.gkerecommender.v1.FetchProfilesResponse.next_page_token] + received from a previous ``FetchProfilesRequest`` call. + Provide this to retrieve the subsequent page in a multi-page + list of results. When paginating, all other parameters + provided to ``FetchProfilesRequest`` must match the call + that provided the page token. + + This field is a member of `oneof`_ ``_page_token``. + """ + + model: str = proto.Field( + proto.STRING, + number=1, + ) + model_server: str = proto.Field( + proto.STRING, + number=2, + ) + model_server_version: str = proto.Field( + proto.STRING, + number=3, + ) + performance_requirements: "PerformanceRequirements" = proto.Field( + proto.MESSAGE, + number=4, + message="PerformanceRequirements", + ) + page_size: int = proto.Field( + proto.INT32, + number=5, + optional=True, + ) + page_token: str = proto.Field( + proto.STRING, + number=6, + optional=True, + ) + + +class PerformanceRequirements(proto.Message): + r"""Performance requirements for a profile and or model + deployment. + + + .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields + + Attributes: + target_ntpot_milliseconds (int): + Optional. The target Normalized Time Per Output Token + (NTPOT) in milliseconds. NTPOT is calculated as + ``request_latency / total_output_tokens``. If not provided, + this target will not be enforced. + + This field is a member of `oneof`_ ``_target_ntpot_milliseconds``. + target_ttft_milliseconds (int): + Optional. The target Time To First Token + (TTFT) in milliseconds. TTFT is the time it + takes to generate the first token for a request. + If not provided, this target will not be + enforced. + + This field is a member of `oneof`_ ``_target_ttft_milliseconds``. + target_cost (google.cloud.gkerecommender_v1.types.Cost): + Optional. The target cost for running a + profile's model server. If not provided, this + requirement will not be enforced. + """ + + target_ntpot_milliseconds: int = proto.Field( + proto.INT32, + number=1, + optional=True, + ) + target_ttft_milliseconds: int = proto.Field( + proto.INT32, + number=2, + optional=True, + ) + target_cost: "Cost" = proto.Field( + proto.MESSAGE, + number=3, + message="Cost", + ) + + +class Amount(proto.Message): + r"""Represents an amount of money in a specific currency. + + Attributes: + units (int): + Output only. The whole units of the amount. For example if + ``currencyCode`` is ``"USD"``, then 1 unit is one US dollar. + nanos (int): + Output only. Number of nano (10^-9) units of the amount. The + value must be between -999,999,999 and +999,999,999 + inclusive. If ``units`` is positive, ``nanos`` must be + positive or zero. If ``units`` is zero, ``nanos`` can be + positive, zero, or negative. If ``units`` is negative, + ``nanos`` must be negative or zero. For example $-1.75 is + represented as ``units``\ =-1 and ``nanos``\ =-750,000,000. + """ + + units: int = proto.Field( + proto.INT64, + number=1, + ) + nanos: int = proto.Field( + proto.INT32, + number=2, + ) + + +class Cost(proto.Message): + r"""Cost for running a model deployment on a given instance type. + Currently, only USD currency code is supported. + + + .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields + + Attributes: + cost_per_million_output_tokens (google.cloud.gkerecommender_v1.types.Amount): + Optional. The cost per million output tokens, calculated as: + $/output token = GPU $/s / (1/output-to-input-cost-ratio \* + input tokens/s + output tokens/s) + cost_per_million_input_tokens (google.cloud.gkerecommender_v1.types.Amount): + Optional. The cost per million input tokens. + $/input token = ($/output token) / + output-to-input-cost-ratio. + pricing_model (str): + Optional. The pricing model used to calculate the cost. Can + be one of: ``3-years-cud``, ``1-year-cud``, ``on-demand``, + ``spot``. If not provided, ``spot`` will be used. + output_input_cost_ratio (float): + Optional. The output-to-input cost ratio. This determines + how the total GPU cost is split between input and output + tokens. If not provided, ``4.0`` is used, assuming a 4:1 + output:input cost ratio. + + This field is a member of `oneof`_ ``_output_input_cost_ratio``. + """ + + cost_per_million_output_tokens: "Amount" = proto.Field( + proto.MESSAGE, + number=1, + message="Amount", + ) + cost_per_million_input_tokens: "Amount" = proto.Field( + proto.MESSAGE, + number=2, + message="Amount", + ) + pricing_model: str = proto.Field( + proto.STRING, + number=3, + ) + output_input_cost_ratio: float = proto.Field( + proto.FLOAT, + number=4, + optional=True, + ) + + +class TokensPerSecondRange(proto.Message): + r"""Represents a range of throughput values in tokens per second. + + Attributes: + min_ (int): + Output only. The minimum value of the range. + max_ (int): + Output only. The maximum value of the range. + """ + + min_: int = proto.Field( + proto.INT32, + number=1, + ) + max_: int = proto.Field( + proto.INT32, + number=2, + ) + + +class MillisecondRange(proto.Message): + r"""Represents a range of latency values in milliseconds. + + Attributes: + min_ (int): + Output only. The minimum value of the range. + max_ (int): + Output only. The maximum value of the range. + """ + + min_: int = proto.Field( + proto.INT32, + number=1, + ) + max_: int = proto.Field( + proto.INT32, + number=2, + ) + + +class PerformanceRange(proto.Message): + r"""Performance range for a model deployment. + + Attributes: + throughput_output_range (google.cloud.gkerecommender_v1.types.TokensPerSecondRange): + Output only. The range of throughput in output tokens per + second. This is measured as + total_output_tokens_generated_by_server / + elapsed_time_in_seconds. + ttft_range (google.cloud.gkerecommender_v1.types.MillisecondRange): + Output only. The range of TTFT (Time To First + Token) in milliseconds. TTFT is the time it + takes to generate the first token for a request. + ntpot_range (google.cloud.gkerecommender_v1.types.MillisecondRange): + Output only. The range of NTPOT (Normalized Time Per Output + Token) in milliseconds. NTPOT is the request latency + normalized by the number of output tokens, measured as + request_latency / total_output_tokens. + """ + + throughput_output_range: "TokensPerSecondRange" = proto.Field( + proto.MESSAGE, + number=1, + message="TokensPerSecondRange", + ) + ttft_range: "MillisecondRange" = proto.Field( + proto.MESSAGE, + number=2, + message="MillisecondRange", + ) + ntpot_range: "MillisecondRange" = proto.Field( + proto.MESSAGE, + number=3, + message="MillisecondRange", + ) + + +class FetchProfilesResponse(proto.Message): + r"""Response message for + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + + Attributes: + profile (MutableSequence[google.cloud.gkerecommender_v1.types.Profile]): + Output only. List of profiles that match the + given model server info and performance + requirements (if provided). + performance_range (google.cloud.gkerecommender_v1.types.PerformanceRange): + Output only. The combined range of + performance values observed across all profiles + in this response. + comments (str): + Output only. Additional comments related to + the response. + next_page_token (str): + Output only. A token which may be sent as + [page_token][FetchProfilesResponse.page_token] in a + subsequent ``FetchProfilesResponse`` call to retrieve the + next page of results. If this field is omitted or empty, + then there are no more results to return. + """ + + @property + def raw_page(self): + return self + + profile: MutableSequence["Profile"] = proto.RepeatedField( + proto.MESSAGE, + number=1, + message="Profile", + ) + performance_range: "PerformanceRange" = proto.Field( + proto.MESSAGE, + number=2, + message="PerformanceRange", + ) + comments: str = proto.Field( + proto.STRING, + number=3, + ) + next_page_token: str = proto.Field( + proto.STRING, + number=4, + ) + + +class ModelServerInfo(proto.Message): + r"""Model server information gives. Valid model server info combinations + can be found using + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. + + Attributes: + model (str): + Required. The model. Open-source models follow the + Huggingface Hub ``owner/model_name`` format. Use + [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels] + to find available models. + model_server (str): + Required. The model server. Open-source model servers use + simplified, lowercase names (e.g., ``vllm``). Use + [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers] + to find available servers. + model_server_version (str): + Optional. The model server version. Use + [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions] + to find available versions. If not provided, the latest + available version is used. + """ + + model: str = proto.Field( + proto.STRING, + number=1, + ) + model_server: str = proto.Field( + proto.STRING, + number=2, + ) + model_server_version: str = proto.Field( + proto.STRING, + number=3, + ) + + +class ResourcesUsed(proto.Message): + r"""Resources used by a model deployment. + + Attributes: + accelerator_count (int): + Output only. The number of accelerators + (e.g., GPUs or TPUs) used by the model + deployment on the Kubernetes node. + """ + + accelerator_count: int = proto.Field( + proto.INT32, + number=1, + ) + + +class PerformanceStats(proto.Message): + r"""Performance statistics for a model deployment. + + Attributes: + queries_per_second (float): + Output only. The number of queries per + second. Note: This metric can vary widely based + on context length and may not be a reliable + measure of LLM throughput. + output_tokens_per_second (int): + Output only. The number of output tokens per second. This is + the throughput measured as + total_output_tokens_generated_by_server / + elapsed_time_in_seconds. + ntpot_milliseconds (int): + Output only. The Normalized Time Per Output Token (NTPOT) in + milliseconds. This is the request latency normalized by the + number of output tokens, measured as request_latency / + total_output_tokens. + ttft_milliseconds (int): + Output only. The Time To First Token (TTFT) + in milliseconds. This is the time it takes to + generate the first token for a request. + cost (MutableSequence[google.cloud.gkerecommender_v1.types.Cost]): + Output only. The cost of running the model + deployment. + """ + + queries_per_second: float = proto.Field( + proto.FLOAT, + number=1, + ) + output_tokens_per_second: int = proto.Field( + proto.INT32, + number=2, + ) + ntpot_milliseconds: int = proto.Field( + proto.INT32, + number=3, + ) + ttft_milliseconds: int = proto.Field( + proto.INT32, + number=4, + ) + cost: MutableSequence["Cost"] = proto.RepeatedField( + proto.MESSAGE, + number=5, + message="Cost", + ) + + +class Profile(proto.Message): + r"""A profile containing information about a model deployment. + + Attributes: + model_server_info (google.cloud.gkerecommender_v1.types.ModelServerInfo): + Output only. The model server configuration. Use + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] + to find valid configurations. + accelerator_type (str): + Output only. The accelerator type. Expected format: + ``nvidia-h100-80gb``. + tpu_topology (str): + Output only. The TPU topology (if + applicable). + instance_type (str): + Output only. The instance type. Expected format: + ``a2-highgpu-1g``. + resources_used (google.cloud.gkerecommender_v1.types.ResourcesUsed): + Output only. The resources used by the model + deployment. + performance_stats (MutableSequence[google.cloud.gkerecommender_v1.types.PerformanceStats]): + Output only. The performance statistics for + this profile. + """ + + model_server_info: "ModelServerInfo" = proto.Field( + proto.MESSAGE, + number=1, + message="ModelServerInfo", + ) + accelerator_type: str = proto.Field( + proto.STRING, + number=2, + ) + tpu_topology: str = proto.Field( + proto.STRING, + number=3, + ) + instance_type: str = proto.Field( + proto.STRING, + number=4, + ) + resources_used: "ResourcesUsed" = proto.Field( + proto.MESSAGE, + number=5, + message="ResourcesUsed", + ) + performance_stats: MutableSequence["PerformanceStats"] = proto.RepeatedField( + proto.MESSAGE, + number=6, + message="PerformanceStats", + ) + + +class GenerateOptimizedManifestRequest(proto.Message): + r"""Request message for + [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. + + Attributes: + model_server_info (google.cloud.gkerecommender_v1.types.ModelServerInfo): + Required. The model server configuration to generate the + manifest for. Use + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] + to find valid configurations. + accelerator_type (str): + Required. The accelerator type. Use + [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] + to find valid accelerators for a given + ``model_server_info``. + kubernetes_namespace (str): + Optional. The kubernetes namespace to deploy + the manifests in. + performance_requirements (google.cloud.gkerecommender_v1.types.PerformanceRequirements): + Optional. The performance requirements to use + for generating Horizontal Pod Autoscaler (HPA) + resources. If provided, the manifest includes + HPA resources to adjust the model server replica + count to maintain the specified targets (e.g., + NTPOT, TTFT) at a P50 latency. Cost targets are + not currently supported for HPA generation. If + the specified targets are not achievable, the + HPA manifest will not be generated. + storage_config (google.cloud.gkerecommender_v1.types.StorageConfig): + Optional. The storage configuration for the + model. If not provided, the model is loaded from + Huggingface. + """ + + model_server_info: "ModelServerInfo" = proto.Field( + proto.MESSAGE, + number=1, + message="ModelServerInfo", + ) + accelerator_type: str = proto.Field( + proto.STRING, + number=2, + ) + kubernetes_namespace: str = proto.Field( + proto.STRING, + number=3, + ) + performance_requirements: "PerformanceRequirements" = proto.Field( + proto.MESSAGE, + number=4, + message="PerformanceRequirements", + ) + storage_config: "StorageConfig" = proto.Field( + proto.MESSAGE, + number=5, + message="StorageConfig", + ) + + +class KubernetesManifest(proto.Message): + r"""A Kubernetes manifest. + + Attributes: + kind (str): + Output only. Kubernetes resource kind. + api_version (str): + Output only. Kubernetes API version. + content (str): + Output only. YAML content. + """ + + kind: str = proto.Field( + proto.STRING, + number=1, + ) + api_version: str = proto.Field( + proto.STRING, + number=2, + ) + content: str = proto.Field( + proto.STRING, + number=3, + ) + + +class GenerateOptimizedManifestResponse(proto.Message): + r"""Response message for + [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. + + Attributes: + kubernetes_manifests (MutableSequence[google.cloud.gkerecommender_v1.types.KubernetesManifest]): + Output only. A list of generated Kubernetes + manifests. + comments (MutableSequence[str]): + Output only. Comments related to deploying + the generated manifests. + manifest_version (str): + Output only. Additional information about the versioned + dependencies used to generate the manifests. See `Run best + practice inference with GKE Inference Quickstart + recipes `__ + for details. + """ + + kubernetes_manifests: MutableSequence["KubernetesManifest"] = proto.RepeatedField( + proto.MESSAGE, + number=1, + message="KubernetesManifest", + ) + comments: MutableSequence[str] = proto.RepeatedField( + proto.STRING, + number=2, + ) + manifest_version: str = proto.Field( + proto.STRING, + number=3, + ) + + +class StorageConfig(proto.Message): + r"""Storage configuration for a model deployment. + + Attributes: + model_bucket_uri (str): + Optional. The Google Cloud Storage bucket URI to load the + model from. This URI must point to the directory containing + the model's config file (``config.json``) and model weights. + A tuned GCSFuse setup can improve LLM Pod startup time by + more than 7x. Expected format: + ``gs:///``. + xla_cache_bucket_uri (str): + Optional. The URI for the GCS bucket containing the XLA + compilation cache. If using TPUs, the XLA cache will be + written to the same path as ``model_bucket_uri``. This can + speed up vLLM model preparation for repeated deployments. + """ + + model_bucket_uri: str = proto.Field( + proto.STRING, + number=1, + ) + xla_cache_bucket_uri: str = proto.Field( + proto.STRING, + number=2, + ) + + +__all__ = tuple(sorted(__protobuf__.manifest)) diff --git a/packages/google-cloud-gkerecommender/mypy.ini b/packages/google-cloud-gkerecommender/mypy.ini new file mode 100644 index 000000000000..574c5aed394b --- /dev/null +++ b/packages/google-cloud-gkerecommender/mypy.ini @@ -0,0 +1,3 @@ +[mypy] +python_version = 3.7 +namespace_packages = True diff --git a/packages/google-cloud-gkerecommender/noxfile.py b/packages/google-cloud-gkerecommender/noxfile.py new file mode 100644 index 000000000000..abcdd0db1687 --- /dev/null +++ b/packages/google-cloud-gkerecommender/noxfile.py @@ -0,0 +1,592 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import os +import pathlib +import re +import shutil +from typing import Dict, List +import warnings + +import nox + +BLACK_VERSION = "black[jupyter]==23.7.0" +ISORT_VERSION = "isort==5.11.0" + +LINT_PATHS = ["docs", "google", "tests", "noxfile.py", "setup.py"] + +ALL_PYTHON = [ + "3.7", + "3.8", + "3.9", + "3.10", + "3.11", + "3.12", + "3.13", +] + +DEFAULT_PYTHON_VERSION = ALL_PYTHON[-1] + +CURRENT_DIRECTORY = pathlib.Path(__file__).parent.absolute() + +LOWER_BOUND_CONSTRAINTS_FILE = CURRENT_DIRECTORY / "constraints.txt" +PACKAGE_NAME = "google-cloud-gkerecommender" + +UNIT_TEST_STANDARD_DEPENDENCIES = [ + "mock", + "asyncmock", + "pytest", + "pytest-cov", + "pytest-asyncio", +] +UNIT_TEST_EXTERNAL_DEPENDENCIES: List[str] = [] +UNIT_TEST_LOCAL_DEPENDENCIES: List[str] = [] +UNIT_TEST_DEPENDENCIES: List[str] = [] +UNIT_TEST_EXTRAS: List[str] = [] +UNIT_TEST_EXTRAS_BY_PYTHON: Dict[str, List[str]] = {} + +SYSTEM_TEST_PYTHON_VERSIONS: List[str] = ["3.8", "3.9", "3.10", "3.11", "3.12", "3.13"] +SYSTEM_TEST_STANDARD_DEPENDENCIES = [ + "mock", + "pytest", + "google-cloud-testutils", +] +SYSTEM_TEST_EXTERNAL_DEPENDENCIES: List[str] = [] +SYSTEM_TEST_LOCAL_DEPENDENCIES: List[str] = [] +SYSTEM_TEST_DEPENDENCIES: List[str] = [] +SYSTEM_TEST_EXTRAS: List[str] = [] +SYSTEM_TEST_EXTRAS_BY_PYTHON: Dict[str, List[str]] = {} + +nox.options.sessions = [ + "unit", + "system", + "cover", + "lint", + "lint_setup_py", + "blacken", + "docs", +] + +# Error if a python version is missing +nox.options.error_on_missing_interpreters = True + + +@nox.session(python=ALL_PYTHON) +def mypy(session): + """Run the type checker.""" + session.install( + # TODO(https://github.com/googleapis/gapic-generator-python/issues/2410): Use the latest version of mypy + "mypy<1.16.0", + "types-requests", + "types-protobuf", + ) + session.install(".") + session.run( + "mypy", + "-p", + "google", + ) + + +@nox.session +def update_lower_bounds(session): + """Update lower bounds in constraints.txt to match setup.py""" + session.install("google-cloud-testutils") + session.install(".") + + session.run( + "lower-bound-checker", + "update", + "--package-name", + PACKAGE_NAME, + "--constraints-file", + str(LOWER_BOUND_CONSTRAINTS_FILE), + ) + + +@nox.session +def check_lower_bounds(session): + """Check lower bounds in setup.py are reflected in constraints file""" + session.install("google-cloud-testutils") + session.install(".") + + session.run( + "lower-bound-checker", + "check", + "--package-name", + PACKAGE_NAME, + "--constraints-file", + str(LOWER_BOUND_CONSTRAINTS_FILE), + ) + + +@nox.session(python=DEFAULT_PYTHON_VERSION) +def lint(session): + """Run linters. + + Returns a failure if the linters find linting errors or sufficiently + serious code quality issues. + """ + session.install("flake8", BLACK_VERSION) + session.run( + "black", + "--check", + *LINT_PATHS, + ) + + session.run("flake8", "google", "tests") + + +@nox.session(python=DEFAULT_PYTHON_VERSION) +def blacken(session): + """Run black. Format code to uniform standard.""" + session.install(BLACK_VERSION) + session.run( + "black", + *LINT_PATHS, + ) + + +@nox.session(python=DEFAULT_PYTHON_VERSION) +def format(session): + """ + Run isort to sort imports. Then run black + to format code to uniform standard. + """ + session.install(BLACK_VERSION, ISORT_VERSION) + # Use the --fss option to sort imports using strict alphabetical order. + # See https://pycqa.github.io/isort/docs/configuration/options.html#force-sort-within-sections + session.run( + "isort", + "--fss", + *LINT_PATHS, + ) + session.run( + "black", + *LINT_PATHS, + ) + + +@nox.session(python=DEFAULT_PYTHON_VERSION) +def lint_setup_py(session): + """Verify that setup.py is valid (including RST check).""" + session.install("setuptools", "docutils", "pygments") + session.run("python", "setup.py", "check", "--restructuredtext", "--strict") + + +def install_unittest_dependencies(session, *constraints): + standard_deps = UNIT_TEST_STANDARD_DEPENDENCIES + UNIT_TEST_DEPENDENCIES + session.install(*standard_deps, *constraints) + + if UNIT_TEST_EXTERNAL_DEPENDENCIES: + warnings.warn( + "'unit_test_external_dependencies' is deprecated. Instead, please " + "use 'unit_test_dependencies' or 'unit_test_local_dependencies'.", + DeprecationWarning, + ) + session.install(*UNIT_TEST_EXTERNAL_DEPENDENCIES, *constraints) + + if UNIT_TEST_LOCAL_DEPENDENCIES: + session.install(*UNIT_TEST_LOCAL_DEPENDENCIES, *constraints) + + if UNIT_TEST_EXTRAS_BY_PYTHON: + extras = UNIT_TEST_EXTRAS_BY_PYTHON.get(session.python, []) + elif UNIT_TEST_EXTRAS: + extras = UNIT_TEST_EXTRAS + else: + extras = [] + + if extras: + session.install("-e", f".[{','.join(extras)}]", *constraints) + else: + session.install("-e", ".", *constraints) + + +@nox.session(python=ALL_PYTHON) +@nox.parametrize( + "protobuf_implementation", + ["python", "upb", "cpp"], +) +def unit(session, protobuf_implementation): + # Install all test dependencies, then install this package in-place. + + if protobuf_implementation == "cpp" and session.python in ("3.11", "3.12", "3.13"): + session.skip("cpp implementation is not supported in python 3.11+") + + constraints_path = str( + CURRENT_DIRECTORY / "testing" / f"constraints-{session.python}.txt" + ) + install_unittest_dependencies(session, "-c", constraints_path) + + # TODO(https://github.com/googleapis/synthtool/issues/1976): + # Remove the 'cpp' implementation once support for Protobuf 3.x is dropped. + # The 'cpp' implementation requires Protobuf<4. + if protobuf_implementation == "cpp": + session.install("protobuf<4") + + # Run py.test against the unit tests. + session.run( + "py.test", + "--quiet", + f"--junitxml=unit_{session.python}_sponge_log.xml", + "--cov=google", + "--cov=tests/unit", + "--cov-append", + "--cov-config=.coveragerc", + "--cov-report=", + "--cov-fail-under=0", + os.path.join("tests", "unit"), + *session.posargs, + env={ + "PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION": protobuf_implementation, + }, + ) + + +def install_systemtest_dependencies(session, *constraints): + session.install("--pre", "grpcio") + + session.install(*SYSTEM_TEST_STANDARD_DEPENDENCIES, *constraints) + + if SYSTEM_TEST_EXTERNAL_DEPENDENCIES: + session.install(*SYSTEM_TEST_EXTERNAL_DEPENDENCIES, *constraints) + + if SYSTEM_TEST_LOCAL_DEPENDENCIES: + session.install("-e", *SYSTEM_TEST_LOCAL_DEPENDENCIES, *constraints) + + if SYSTEM_TEST_DEPENDENCIES: + session.install("-e", *SYSTEM_TEST_DEPENDENCIES, *constraints) + + if SYSTEM_TEST_EXTRAS_BY_PYTHON: + extras = SYSTEM_TEST_EXTRAS_BY_PYTHON.get(session.python, []) + elif SYSTEM_TEST_EXTRAS: + extras = SYSTEM_TEST_EXTRAS + else: + extras = [] + + if extras: + session.install("-e", f".[{','.join(extras)}]", *constraints) + else: + session.install("-e", ".", *constraints) + + +@nox.session(python=SYSTEM_TEST_PYTHON_VERSIONS) +def system(session): + """Run the system test suite.""" + constraints_path = str( + CURRENT_DIRECTORY / "testing" / f"constraints-{session.python}.txt" + ) + system_test_path = os.path.join("tests", "system.py") + system_test_folder_path = os.path.join("tests", "system") + + # Check the value of `RUN_SYSTEM_TESTS` env var. It defaults to true. + if os.environ.get("RUN_SYSTEM_TESTS", "true") == "false": + session.skip("RUN_SYSTEM_TESTS is set to false, skipping") + # Install pyopenssl for mTLS testing. + if os.environ.get("GOOGLE_API_USE_CLIENT_CERTIFICATE", "false") == "true": + session.install("pyopenssl") + + system_test_exists = os.path.exists(system_test_path) + system_test_folder_exists = os.path.exists(system_test_folder_path) + # Sanity check: only run tests if found. + if not system_test_exists and not system_test_folder_exists: + session.skip("System tests were not found") + + install_systemtest_dependencies(session, "-c", constraints_path) + + # Run py.test against the system tests. + if system_test_exists: + session.run( + "py.test", + "--quiet", + f"--junitxml=system_{session.python}_sponge_log.xml", + system_test_path, + *session.posargs, + ) + if system_test_folder_exists: + session.run( + "py.test", + "--quiet", + f"--junitxml=system_{session.python}_sponge_log.xml", + system_test_folder_path, + *session.posargs, + ) + + +@nox.session(python=DEFAULT_PYTHON_VERSION) +def cover(session): + """Run the final coverage report. + + This outputs the coverage report aggregating coverage from the unit + test runs (not system test runs), and then erases coverage data. + """ + session.install("coverage", "pytest-cov") + session.run("coverage", "report", "--show-missing", "--fail-under=100") + + session.run("coverage", "erase") + + +@nox.session(python="3.10") +def docs(session): + """Build the docs for this library.""" + + session.install("-e", ".") + session.install( + # We need to pin to specific versions of the `sphinxcontrib-*` packages + # which still support sphinx 4.x. + # See https://github.com/googleapis/sphinx-docfx-yaml/issues/344 + # and https://github.com/googleapis/sphinx-docfx-yaml/issues/345. + "sphinxcontrib-applehelp==1.0.4", + "sphinxcontrib-devhelp==1.0.2", + "sphinxcontrib-htmlhelp==2.0.1", + "sphinxcontrib-qthelp==1.0.3", + "sphinxcontrib-serializinghtml==1.1.5", + "sphinx==4.5.0", + "alabaster", + "recommonmark", + ) + + shutil.rmtree(os.path.join("docs", "_build"), ignore_errors=True) + session.run( + "sphinx-build", + "-W", # warnings as errors + "-T", # show full traceback on exception + "-N", # no colors + "-b", + "html", # builder + "-d", + os.path.join("docs", "_build", "doctrees", ""), # cache directory + # paths to build: + os.path.join("docs", ""), + os.path.join("docs", "_build", "html", ""), + ) + + +@nox.session(python="3.10") +def docfx(session): + """Build the docfx yaml files for this library.""" + + session.install("-e", ".") + session.install( + # We need to pin to specific versions of the `sphinxcontrib-*` packages + # which still support sphinx 4.x. + # See https://github.com/googleapis/sphinx-docfx-yaml/issues/344 + # and https://github.com/googleapis/sphinx-docfx-yaml/issues/345. + "sphinxcontrib-applehelp==1.0.4", + "sphinxcontrib-devhelp==1.0.2", + "sphinxcontrib-htmlhelp==2.0.1", + "sphinxcontrib-qthelp==1.0.3", + "sphinxcontrib-serializinghtml==1.1.5", + "gcp-sphinx-docfx-yaml", + "alabaster", + "recommonmark", + ) + + shutil.rmtree(os.path.join("docs", "_build"), ignore_errors=True) + session.run( + "sphinx-build", + "-T", # show full traceback on exception + "-N", # no colors + "-D", + ( + "extensions=sphinx.ext.autodoc," + "sphinx.ext.autosummary," + "docfx_yaml.extension," + "sphinx.ext.intersphinx," + "sphinx.ext.coverage," + "sphinx.ext.napoleon," + "sphinx.ext.todo," + "sphinx.ext.viewcode," + "recommonmark" + ), + "-b", + "html", + "-d", + os.path.join("docs", "_build", "doctrees", ""), + os.path.join("docs", ""), + os.path.join("docs", "_build", "html", ""), + ) + + +@nox.session(python=DEFAULT_PYTHON_VERSION) +@nox.parametrize( + "protobuf_implementation", + ["python", "upb", "cpp"], +) +def prerelease_deps(session, protobuf_implementation): + """ + Run all tests with pre-release versions of dependencies installed + rather than the standard non pre-release versions. + Pre-release versions can be installed using + `pip install --pre `. + """ + + if protobuf_implementation == "cpp" and session.python in ("3.11", "3.12", "3.13"): + session.skip("cpp implementation is not supported in python 3.11+") + + # Install all dependencies + session.install("-e", ".") + + # Install dependencies for the unit test environment + unit_deps_all = UNIT_TEST_STANDARD_DEPENDENCIES + UNIT_TEST_EXTERNAL_DEPENDENCIES + session.install(*unit_deps_all) + + # Install dependencies for the system test environment + system_deps_all = ( + SYSTEM_TEST_STANDARD_DEPENDENCIES + + SYSTEM_TEST_EXTERNAL_DEPENDENCIES + + SYSTEM_TEST_EXTRAS + ) + session.install(*system_deps_all) + + # Because we test minimum dependency versions on the minimum Python + # version, the first version we test with in the unit tests sessions has a + # constraints file containing all dependencies and extras. + with open( + CURRENT_DIRECTORY / "testing" / f"constraints-{ALL_PYTHON[0]}.txt", + encoding="utf-8", + ) as constraints_file: + constraints_text = constraints_file.read() + + # Ignore leading whitespace and comment lines. + constraints_deps = [ + match.group(1) + for match in re.finditer( + r"^\s*(\S+)(?===\S+)", constraints_text, flags=re.MULTILINE + ) + ] + + # Install dependencies specified in `testing/constraints-X.txt`. + session.install(*constraints_deps) + + # Note: If a dependency is added to the `prerel_deps` list, + # the `core_dependencies_from_source` list in the `core_deps_from_source` + # nox session should also be updated. + prerel_deps = [ + "googleapis-common-protos", + "google-api-core", + "google-auth", + "grpc-google-iam-v1", + "grpcio", + "grpcio-status", + "protobuf", + "proto-plus", + ] + + for dep in prerel_deps: + session.install("--pre", "--no-deps", "--ignore-installed", dep) + # TODO(https://github.com/grpc/grpc/issues/38965): Add `grpcio-status`` + # to the dictionary below once this bug is fixed. + # TODO(https://github.com/googleapis/google-cloud-python/issues/13643): Add + # `googleapis-common-protos` and `grpc-google-iam-v1` to the dictionary below + # once this bug is fixed. + package_namespaces = { + "google-api-core": "google.api_core", + "google-auth": "google.auth", + "grpcio": "grpc", + "protobuf": "google.protobuf", + "proto-plus": "proto", + } + + version_namespace = package_namespaces.get(dep) + + print(f"Installed {dep}") + if version_namespace: + session.run( + "python", + "-c", + f"import {version_namespace}; print({version_namespace}.__version__)", + ) + + session.run( + "py.test", + "tests/unit", + env={ + "PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION": protobuf_implementation, + }, + ) + + +@nox.session(python=DEFAULT_PYTHON_VERSION) +@nox.parametrize( + "protobuf_implementation", + ["python", "upb"], +) +def core_deps_from_source(session, protobuf_implementation): + """Run all tests with core dependencies installed from source + rather than pulling the dependencies from PyPI. + """ + + # Install all dependencies + session.install("-e", ".") + + # Install dependencies for the unit test environment + unit_deps_all = UNIT_TEST_STANDARD_DEPENDENCIES + UNIT_TEST_EXTERNAL_DEPENDENCIES + session.install(*unit_deps_all) + + # Install dependencies for the system test environment + system_deps_all = ( + SYSTEM_TEST_STANDARD_DEPENDENCIES + + SYSTEM_TEST_EXTERNAL_DEPENDENCIES + + SYSTEM_TEST_EXTRAS + ) + session.install(*system_deps_all) + + # Because we test minimum dependency versions on the minimum Python + # version, the first version we test with in the unit tests sessions has a + # constraints file containing all dependencies and extras. + with open( + CURRENT_DIRECTORY / "testing" / f"constraints-{ALL_PYTHON[0]}.txt", + encoding="utf-8", + ) as constraints_file: + constraints_text = constraints_file.read() + + # Ignore leading whitespace and comment lines. + constraints_deps = [ + match.group(1) + for match in re.finditer( + r"^\s*(\S+)(?===\S+)", constraints_text, flags=re.MULTILINE + ) + ] + + # Install dependencies specified in `testing/constraints-X.txt`. + session.install(*constraints_deps) + + # TODO(https://github.com/googleapis/gapic-generator-python/issues/2358): `grpcio` and + # `grpcio-status` should be added to the list below so that they are installed from source, + # rather than PyPI. + # TODO(https://github.com/googleapis/gapic-generator-python/issues/2357): `protobuf` should be + # added to the list below so that it is installed from source, rather than PyPI + # Note: If a dependency is added to the `core_dependencies_from_source` list, + # the `prerel_deps` list in the `prerelease_deps` nox session should also be updated. + core_dependencies_from_source = [ + "googleapis-common-protos @ git+https://github.com/googleapis/google-cloud-python#egg=googleapis-common-protos&subdirectory=packages/googleapis-common-protos", + "google-api-core @ git+https://github.com/googleapis/python-api-core.git", + "google-auth @ git+https://github.com/googleapis/google-auth-library-python.git", + "grpc-google-iam-v1 @ git+https://github.com/googleapis/google-cloud-python#egg=grpc-google-iam-v1&subdirectory=packages/grpc-google-iam-v1", + "proto-plus @ git+https://github.com/googleapis/proto-plus-python.git", + ] + + for dep in core_dependencies_from_source: + session.install(dep, "--no-deps", "--ignore-installed") + print(f"Installed {dep}") + + session.run( + "py.test", + "tests/unit", + env={ + "PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION": protobuf_implementation, + }, + ) diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py new file mode 100644 index 000000000000..f10a415865e0 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py @@ -0,0 +1,56 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchBenchmarkingData +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_async] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +async def sample_fetch_benchmarking_data(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + model_server_info = gkerecommender_v1.ModelServerInfo() + model_server_info.model = "model_value" + model_server_info.model_server = "model_server_value" + + request = gkerecommender_v1.FetchBenchmarkingDataRequest( + model_server_info=model_server_info, + ) + + # Make the request + response = await client.fetch_benchmarking_data(request=request) + + # Handle the response + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py new file mode 100644 index 000000000000..dab6bce72582 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py @@ -0,0 +1,56 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchBenchmarkingData +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_sync] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +def sample_fetch_benchmarking_data(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + model_server_info = gkerecommender_v1.ModelServerInfo() + model_server_info.model = "model_value" + model_server_info.model_server = "model_server_value" + + request = gkerecommender_v1.FetchBenchmarkingDataRequest( + model_server_info=model_server_info, + ) + + # Make the request + response = client.fetch_benchmarking_data(request=request) + + # Handle the response + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py new file mode 100644 index 000000000000..81463976c4c9 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py @@ -0,0 +1,54 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchModelServerVersions +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_async] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +async def sample_fetch_model_server_versions(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelServerVersionsRequest( + model="model_value", + model_server="model_server_value", + ) + + # Make the request + page_result = client.fetch_model_server_versions(request=request) + + # Handle the response + async for response in page_result: + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py new file mode 100644 index 000000000000..9bc645591a20 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py @@ -0,0 +1,54 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchModelServerVersions +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_sync] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +def sample_fetch_model_server_versions(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelServerVersionsRequest( + model="model_value", + model_server="model_server_value", + ) + + # Make the request + page_result = client.fetch_model_server_versions(request=request) + + # Handle the response + for response in page_result: + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py new file mode 100644 index 000000000000..648669d45067 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py @@ -0,0 +1,53 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchModelServers +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_async] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +async def sample_fetch_model_servers(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelServersRequest( + model="model_value", + ) + + # Make the request + page_result = client.fetch_model_servers(request=request) + + # Handle the response + async for response in page_result: + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py new file mode 100644 index 000000000000..ddbf9d98c5b7 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py @@ -0,0 +1,53 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchModelServers +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_sync] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +def sample_fetch_model_servers(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelServersRequest( + model="model_value", + ) + + # Make the request + page_result = client.fetch_model_servers(request=request) + + # Handle the response + for response in page_result: + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py new file mode 100644 index 000000000000..651f2483904f --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py @@ -0,0 +1,52 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchModels +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_async] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +async def sample_fetch_models(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelsRequest( + ) + + # Make the request + page_result = client.fetch_models(request=request) + + # Handle the response + async for response in page_result: + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py new file mode 100644 index 000000000000..357104b82387 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py @@ -0,0 +1,52 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchModels +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_sync] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +def sample_fetch_models(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchModelsRequest( + ) + + # Make the request + page_result = client.fetch_models(request=request) + + # Handle the response + for response in page_result: + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py new file mode 100644 index 000000000000..e218622fe416 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py @@ -0,0 +1,52 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchProfiles +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_async] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +async def sample_fetch_profiles(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchProfilesRequest( + ) + + # Make the request + page_result = client.fetch_profiles(request=request) + + # Handle the response + async for response in page_result: + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py new file mode 100644 index 000000000000..e9a6dea821f2 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py @@ -0,0 +1,52 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for FetchProfiles +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_sync] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +def sample_fetch_profiles(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + request = gkerecommender_v1.FetchProfilesRequest( + ) + + # Make the request + page_result = client.fetch_profiles(request=request) + + # Handle the response + for response in page_result: + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py new file mode 100644 index 000000000000..286c4c22a63c --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py @@ -0,0 +1,57 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for GenerateOptimizedManifest +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_async] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +async def sample_generate_optimized_manifest(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() + + # Initialize request argument(s) + model_server_info = gkerecommender_v1.ModelServerInfo() + model_server_info.model = "model_value" + model_server_info.model_server = "model_server_value" + + request = gkerecommender_v1.GenerateOptimizedManifestRequest( + model_server_info=model_server_info, + accelerator_type="accelerator_type_value", + ) + + # Make the request + response = await client.generate_optimized_manifest(request=request) + + # Handle the response + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py new file mode 100644 index 000000000000..4d1362ea3b17 --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py @@ -0,0 +1,57 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +# Generated code. DO NOT EDIT! +# +# Snippet for GenerateOptimizedManifest +# NOTE: This snippet has been automatically generated for illustrative purposes only. +# It may require modifications to work in your environment. + +# To install the latest published package dependency, execute the following: +# python3 -m pip install google-cloud-gkerecommender + + +# [START gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_sync] +# This snippet has been automatically generated and should be regarded as a +# code template only. +# It will require modifications to work: +# - It may require correct/in-range values for request initialization. +# - It may require specifying regional endpoints when creating the service +# client as shown in: +# https://googleapis.dev/python/google-api-core/latest/client_options.html +from google.cloud import gkerecommender_v1 + + +def sample_generate_optimized_manifest(): + # Create a client + client = gkerecommender_v1.GkeInferenceQuickstartClient() + + # Initialize request argument(s) + model_server_info = gkerecommender_v1.ModelServerInfo() + model_server_info.model = "model_value" + model_server_info.model_server = "model_server_value" + + request = gkerecommender_v1.GenerateOptimizedManifestRequest( + model_server_info=model_server_info, + accelerator_type="accelerator_type_value", + ) + + # Make the request + response = client.generate_optimized_manifest(request=request) + + # Handle the response + print(response) + +# [END gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/snippet_metadata_google.cloud.gkerecommender.v1.json b/packages/google-cloud-gkerecommender/samples/generated_samples/snippet_metadata_google.cloud.gkerecommender.v1.json new file mode 100644 index 000000000000..d3eacb5ac8dd --- /dev/null +++ b/packages/google-cloud-gkerecommender/samples/generated_samples/snippet_metadata_google.cloud.gkerecommender.v1.json @@ -0,0 +1,933 @@ +{ + "clientLibrary": { + "apis": [ + { + "id": "google.cloud.gkerecommender.v1", + "version": "v1" + } + ], + "language": "PYTHON", + "name": "google-cloud-gkerecommender", + "version": "0.0.0" + }, + "snippets": [ + { + "canonical": true, + "clientMethod": { + "async": true, + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", + "shortName": "GkeInferenceQuickstartAsyncClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_benchmarking_data", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchBenchmarkingData" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse", + "shortName": "fetch_benchmarking_data" + }, + "description": "Sample for FetchBenchmarkingData", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_async", + "segments": [ + { + "end": 55, + "start": 27, + "type": "FULL" + }, + { + "end": 55, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 49, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 52, + "start": 50, + "type": "REQUEST_EXECUTION" + }, + { + "end": 56, + "start": 53, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py" + }, + { + "canonical": true, + "clientMethod": { + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", + "shortName": "GkeInferenceQuickstartClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_benchmarking_data", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchBenchmarkingData" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse", + "shortName": "fetch_benchmarking_data" + }, + "description": "Sample for FetchBenchmarkingData", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_sync", + "segments": [ + { + "end": 55, + "start": 27, + "type": "FULL" + }, + { + "end": 55, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 49, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 52, + "start": 50, + "type": "REQUEST_EXECUTION" + }, + { + "end": 56, + "start": 53, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py" + }, + { + "canonical": true, + "clientMethod": { + "async": true, + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", + "shortName": "GkeInferenceQuickstartAsyncClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_model_server_versions", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchModelServerVersions" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsAsyncPager", + "shortName": "fetch_model_server_versions" + }, + "description": "Sample for FetchModelServerVersions", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_async", + "segments": [ + { + "end": 53, + "start": 27, + "type": "FULL" + }, + { + "end": 53, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 46, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 49, + "start": 47, + "type": "REQUEST_EXECUTION" + }, + { + "end": 54, + "start": 50, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py" + }, + { + "canonical": true, + "clientMethod": { + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", + "shortName": "GkeInferenceQuickstartClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_model_server_versions", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchModelServerVersions" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsPager", + "shortName": "fetch_model_server_versions" + }, + "description": "Sample for FetchModelServerVersions", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_sync", + "segments": [ + { + "end": 53, + "start": 27, + "type": "FULL" + }, + { + "end": 53, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 46, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 49, + "start": 47, + "type": "REQUEST_EXECUTION" + }, + { + "end": 54, + "start": 50, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py" + }, + { + "canonical": true, + "clientMethod": { + "async": true, + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", + "shortName": "GkeInferenceQuickstartAsyncClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_model_servers", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchModelServers" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchModelServersRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersAsyncPager", + "shortName": "fetch_model_servers" + }, + "description": "Sample for FetchModelServers", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_async", + "segments": [ + { + "end": 52, + "start": 27, + "type": "FULL" + }, + { + "end": 52, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 45, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 48, + "start": 46, + "type": "REQUEST_EXECUTION" + }, + { + "end": 53, + "start": 49, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py" + }, + { + "canonical": true, + "clientMethod": { + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", + "shortName": "GkeInferenceQuickstartClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_model_servers", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchModelServers" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchModelServersRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersPager", + "shortName": "fetch_model_servers" + }, + "description": "Sample for FetchModelServers", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_sync", + "segments": [ + { + "end": 52, + "start": 27, + "type": "FULL" + }, + { + "end": 52, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 45, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 48, + "start": 46, + "type": "REQUEST_EXECUTION" + }, + { + "end": 53, + "start": 49, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py" + }, + { + "canonical": true, + "clientMethod": { + "async": true, + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", + "shortName": "GkeInferenceQuickstartAsyncClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_models", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchModels" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchModelsRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsAsyncPager", + "shortName": "fetch_models" + }, + "description": "Sample for FetchModels", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_async", + "segments": [ + { + "end": 51, + "start": 27, + "type": "FULL" + }, + { + "end": 51, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 44, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 47, + "start": 45, + "type": "REQUEST_EXECUTION" + }, + { + "end": 52, + "start": 48, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py" + }, + { + "canonical": true, + "clientMethod": { + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", + "shortName": "GkeInferenceQuickstartClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_models", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchModels" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchModelsRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsPager", + "shortName": "fetch_models" + }, + "description": "Sample for FetchModels", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_sync", + "segments": [ + { + "end": 51, + "start": 27, + "type": "FULL" + }, + { + "end": 51, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 44, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 47, + "start": 45, + "type": "REQUEST_EXECUTION" + }, + { + "end": 52, + "start": 48, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py" + }, + { + "canonical": true, + "clientMethod": { + "async": true, + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", + "shortName": "GkeInferenceQuickstartAsyncClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_profiles", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchProfiles" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchProfilesRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesAsyncPager", + "shortName": "fetch_profiles" + }, + "description": "Sample for FetchProfiles", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_async", + "segments": [ + { + "end": 51, + "start": 27, + "type": "FULL" + }, + { + "end": 51, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 44, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 47, + "start": 45, + "type": "REQUEST_EXECUTION" + }, + { + "end": 52, + "start": 48, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py" + }, + { + "canonical": true, + "clientMethod": { + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", + "shortName": "GkeInferenceQuickstartClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_profiles", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "FetchProfiles" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.FetchProfilesRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesPager", + "shortName": "fetch_profiles" + }, + "description": "Sample for FetchProfiles", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_sync", + "segments": [ + { + "end": 51, + "start": 27, + "type": "FULL" + }, + { + "end": 51, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 44, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 47, + "start": 45, + "type": "REQUEST_EXECUTION" + }, + { + "end": 52, + "start": 48, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py" + }, + { + "canonical": true, + "clientMethod": { + "async": true, + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", + "shortName": "GkeInferenceQuickstartAsyncClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.generate_optimized_manifest", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "GenerateOptimizedManifest" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse", + "shortName": "generate_optimized_manifest" + }, + "description": "Sample for GenerateOptimizedManifest", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_async", + "segments": [ + { + "end": 56, + "start": 27, + "type": "FULL" + }, + { + "end": 56, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 50, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 53, + "start": 51, + "type": "REQUEST_EXECUTION" + }, + { + "end": 57, + "start": 54, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py" + }, + { + "canonical": true, + "clientMethod": { + "client": { + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", + "shortName": "GkeInferenceQuickstartClient" + }, + "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.generate_optimized_manifest", + "method": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest", + "service": { + "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", + "shortName": "GkeInferenceQuickstart" + }, + "shortName": "GenerateOptimizedManifest" + }, + "parameters": [ + { + "name": "request", + "type": "google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest" + }, + { + "name": "retry", + "type": "google.api_core.retry.Retry" + }, + { + "name": "timeout", + "type": "float" + }, + { + "name": "metadata", + "type": "Sequence[Tuple[str, Union[str, bytes]]]" + } + ], + "resultType": "google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse", + "shortName": "generate_optimized_manifest" + }, + "description": "Sample for GenerateOptimizedManifest", + "file": "gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py", + "language": "PYTHON", + "origin": "API_DEFINITION", + "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_sync", + "segments": [ + { + "end": 56, + "start": 27, + "type": "FULL" + }, + { + "end": 56, + "start": 27, + "type": "SHORT" + }, + { + "end": 40, + "start": 38, + "type": "CLIENT_INITIALIZATION" + }, + { + "end": 50, + "start": 41, + "type": "REQUEST_INITIALIZATION" + }, + { + "end": 53, + "start": 51, + "type": "REQUEST_EXECUTION" + }, + { + "end": 57, + "start": 54, + "type": "RESPONSE_HANDLING" + } + ], + "title": "gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py" + } + ] +} diff --git a/packages/google-cloud-gkerecommender/scripts/fixup_gkerecommender_v1_keywords.py b/packages/google-cloud-gkerecommender/scripts/fixup_gkerecommender_v1_keywords.py new file mode 100644 index 000000000000..4c43993a2015 --- /dev/null +++ b/packages/google-cloud-gkerecommender/scripts/fixup_gkerecommender_v1_keywords.py @@ -0,0 +1,181 @@ +#! /usr/bin/env python3 +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import argparse +import os +import libcst as cst +import pathlib +import sys +from typing import (Any, Callable, Dict, List, Sequence, Tuple) + + +def partition( + predicate: Callable[[Any], bool], + iterator: Sequence[Any] +) -> Tuple[List[Any], List[Any]]: + """A stable, out-of-place partition.""" + results = ([], []) + + for i in iterator: + results[int(predicate(i))].append(i) + + # Returns trueList, falseList + return results[1], results[0] + + +class gkerecommenderCallTransformer(cst.CSTTransformer): + CTRL_PARAMS: Tuple[str] = ('retry', 'timeout', 'metadata') + METHOD_TO_PARAMS: Dict[str, Tuple[str]] = { + 'fetch_benchmarking_data': ('model_server_info', 'instance_type', 'pricing_model', ), + 'fetch_models': ('page_size', 'page_token', ), + 'fetch_model_servers': ('model', 'page_size', 'page_token', ), + 'fetch_model_server_versions': ('model', 'model_server', 'page_size', 'page_token', ), + 'fetch_profiles': ('model', 'model_server', 'model_server_version', 'performance_requirements', 'page_size', 'page_token', ), + 'generate_optimized_manifest': ('model_server_info', 'accelerator_type', 'kubernetes_namespace', 'performance_requirements', 'storage_config', ), + } + + def leave_Call(self, original: cst.Call, updated: cst.Call) -> cst.CSTNode: + try: + key = original.func.attr.value + kword_params = self.METHOD_TO_PARAMS[key] + except (AttributeError, KeyError): + # Either not a method from the API or too convoluted to be sure. + return updated + + # If the existing code is valid, keyword args come after positional args. + # Therefore, all positional args must map to the first parameters. + args, kwargs = partition(lambda a: not bool(a.keyword), updated.args) + if any(k.keyword.value == "request" for k in kwargs): + # We've already fixed this file, don't fix it again. + return updated + + kwargs, ctrl_kwargs = partition( + lambda a: a.keyword.value not in self.CTRL_PARAMS, + kwargs + ) + + args, ctrl_args = args[:len(kword_params)], args[len(kword_params):] + ctrl_kwargs.extend(cst.Arg(value=a.value, keyword=cst.Name(value=ctrl)) + for a, ctrl in zip(ctrl_args, self.CTRL_PARAMS)) + + request_arg = cst.Arg( + value=cst.Dict([ + cst.DictElement( + cst.SimpleString("'{}'".format(name)), +cst.Element(value=arg.value) + ) + # Note: the args + kwargs looks silly, but keep in mind that + # the control parameters had to be stripped out, and that + # those could have been passed positionally or by keyword. + for name, arg in zip(kword_params, args + kwargs)]), + keyword=cst.Name("request") + ) + + return updated.with_changes( + args=[request_arg] + ctrl_kwargs + ) + + +def fix_files( + in_dir: pathlib.Path, + out_dir: pathlib.Path, + *, + transformer=gkerecommenderCallTransformer(), +): + """Duplicate the input dir to the output dir, fixing file method calls. + + Preconditions: + * in_dir is a real directory + * out_dir is a real, empty directory + """ + pyfile_gen = ( + pathlib.Path(os.path.join(root, f)) + for root, _, files in os.walk(in_dir) + for f in files if os.path.splitext(f)[1] == ".py" + ) + + for fpath in pyfile_gen: + with open(fpath, 'r') as f: + src = f.read() + + # Parse the code and insert method call fixes. + tree = cst.parse_module(src) + updated = tree.visit(transformer) + + # Create the path and directory structure for the new file. + updated_path = out_dir.joinpath(fpath.relative_to(in_dir)) + updated_path.parent.mkdir(parents=True, exist_ok=True) + + # Generate the updated source file at the corresponding path. + with open(updated_path, 'w') as f: + f.write(updated.code) + + +if __name__ == '__main__': + parser = argparse.ArgumentParser( + description="""Fix up source that uses the gkerecommender client library. + +The existing sources are NOT overwritten but are copied to output_dir with changes made. + +Note: This tool operates at a best-effort level at converting positional + parameters in client method calls to keyword based parameters. + Cases where it WILL FAIL include + A) * or ** expansion in a method call. + B) Calls via function or method alias (includes free function calls) + C) Indirect or dispatched calls (e.g. the method is looked up dynamically) + + These all constitute false negatives. The tool will also detect false + positives when an API method shares a name with another method. +""") + parser.add_argument( + '-d', + '--input-directory', + required=True, + dest='input_dir', + help='the input directory to walk for python files to fix up', + ) + parser.add_argument( + '-o', + '--output-directory', + required=True, + dest='output_dir', + help='the directory to output files fixed via un-flattening', + ) + args = parser.parse_args() + input_dir = pathlib.Path(args.input_dir) + output_dir = pathlib.Path(args.output_dir) + if not input_dir.is_dir(): + print( + f"input directory '{input_dir}' does not exist or is not a directory", + file=sys.stderr, + ) + sys.exit(-1) + + if not output_dir.is_dir(): + print( + f"output directory '{output_dir}' does not exist or is not a directory", + file=sys.stderr, + ) + sys.exit(-1) + + if os.listdir(output_dir): + print( + f"output directory '{output_dir}' is not empty", + file=sys.stderr, + ) + sys.exit(-1) + + fix_files(input_dir, output_dir) diff --git a/packages/google-cloud-gkerecommender/setup.py b/packages/google-cloud-gkerecommender/setup.py new file mode 100644 index 000000000000..b442b627cd44 --- /dev/null +++ b/packages/google-cloud-gkerecommender/setup.py @@ -0,0 +1,99 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import io +import os +import re + +import setuptools # type: ignore + +package_root = os.path.abspath(os.path.dirname(__file__)) + +name = "google-cloud-gkerecommender" + + +description = "Google Cloud Gkerecommender API client library" + +version = None + +with open( + os.path.join(package_root, "google/cloud/gkerecommender/gapic_version.py") +) as fp: + version_candidates = re.findall(r"(?<=\")\d+.\d+.\d+(?=\")", fp.read()) + assert len(version_candidates) == 1 + version = version_candidates[0] + +if version[0] == "0": + release_status = "Development Status :: 4 - Beta" +else: + release_status = "Development Status :: 5 - Production/Stable" + +dependencies = [ + "google-api-core[grpc] >= 1.34.1, <3.0.0,!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,!=2.10.*", + # Exclude incompatible versions of `google-auth` + # See https://github.com/googleapis/google-cloud-python/issues/12364 + "google-auth >= 2.14.1, <3.0.0,!=2.24.0,!=2.25.0", + "proto-plus >= 1.22.3, <2.0.0", + "proto-plus >= 1.25.0, <2.0.0; python_version >= '3.13'", + "protobuf>=3.20.2,<7.0.0,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5", +] +extras = {} +url = "https://github.com/googleapis/google-cloud-python/tree/main/packages/google-cloud-gkerecommender" + +package_root = os.path.abspath(os.path.dirname(__file__)) + +readme_filename = os.path.join(package_root, "README.rst") +with io.open(readme_filename, encoding="utf-8") as readme_file: + readme = readme_file.read() + +packages = [ + package + for package in setuptools.find_namespace_packages() + if package.startswith("google") +] + +setuptools.setup( + name=name, + version=version, + description=description, + long_description=readme, + author="Google LLC", + author_email="googleapis-packages@google.com", + license="Apache 2.0", + url=url, + classifiers=[ + release_status, + "Intended Audience :: Developers", + "License :: OSI Approved :: Apache Software License", + "Programming Language :: Python", + "Programming Language :: Python :: 3", + "Programming Language :: Python :: 3.7", + "Programming Language :: Python :: 3.8", + "Programming Language :: Python :: 3.9", + "Programming Language :: Python :: 3.10", + "Programming Language :: Python :: 3.11", + "Programming Language :: Python :: 3.12", + "Programming Language :: Python :: 3.13", + "Operating System :: OS Independent", + "Topic :: Internet", + ], + platforms="Posix; MacOS X; Windows", + packages=packages, + python_requires=">=3.7", + install_requires=dependencies, + extras_require=extras, + include_package_data=True, + zip_safe=False, +) diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.10.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.10.txt new file mode 100644 index 000000000000..ed7f9aed2559 --- /dev/null +++ b/packages/google-cloud-gkerecommender/testing/constraints-3.10.txt @@ -0,0 +1,6 @@ +# -*- coding: utf-8 -*- +# This constraints file is required for unit tests. +# List all library dependencies and extras in this file. +google-api-core +proto-plus +protobuf diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.11.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.11.txt new file mode 100644 index 000000000000..ed7f9aed2559 --- /dev/null +++ b/packages/google-cloud-gkerecommender/testing/constraints-3.11.txt @@ -0,0 +1,6 @@ +# -*- coding: utf-8 -*- +# This constraints file is required for unit tests. +# List all library dependencies and extras in this file. +google-api-core +proto-plus +protobuf diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.12.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.12.txt new file mode 100644 index 000000000000..ed7f9aed2559 --- /dev/null +++ b/packages/google-cloud-gkerecommender/testing/constraints-3.12.txt @@ -0,0 +1,6 @@ +# -*- coding: utf-8 -*- +# This constraints file is required for unit tests. +# List all library dependencies and extras in this file. +google-api-core +proto-plus +protobuf diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.13.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.13.txt new file mode 100644 index 000000000000..c20a77817caa --- /dev/null +++ b/packages/google-cloud-gkerecommender/testing/constraints-3.13.txt @@ -0,0 +1,11 @@ +# We use the constraints file for the latest Python version +# (currently this file) to check that the latest +# major versions of dependencies are supported in setup.py. +# List all library dependencies and extras in this file. +# Require the latest major version be installed for each dependency. +# e.g., if setup.py has "google-cloud-foo >= 1.14.0, < 2.0.0", +# Then this file should have google-cloud-foo>=1 +google-api-core>=2 +google-auth>=2 +proto-plus>=1 +protobuf>=6 diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.7.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.7.txt new file mode 100644 index 000000000000..a77f12bc13e4 --- /dev/null +++ b/packages/google-cloud-gkerecommender/testing/constraints-3.7.txt @@ -0,0 +1,10 @@ +# This constraints file is used to check that lower bounds +# are correct in setup.py +# List all library dependencies and extras in this file. +# Pin the version to the lower bound. +# e.g., if setup.py has "google-cloud-foo >= 1.14.0, < 2.0.0", +# Then this file should have google-cloud-foo==1.14.0 +google-api-core==1.34.1 +google-auth==2.14.1 +proto-plus==1.22.3 +protobuf==3.20.2 diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.8.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.8.txt new file mode 100644 index 000000000000..ed7f9aed2559 --- /dev/null +++ b/packages/google-cloud-gkerecommender/testing/constraints-3.8.txt @@ -0,0 +1,6 @@ +# -*- coding: utf-8 -*- +# This constraints file is required for unit tests. +# List all library dependencies and extras in this file. +google-api-core +proto-plus +protobuf diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.9.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.9.txt new file mode 100644 index 000000000000..ed7f9aed2559 --- /dev/null +++ b/packages/google-cloud-gkerecommender/testing/constraints-3.9.txt @@ -0,0 +1,6 @@ +# -*- coding: utf-8 -*- +# This constraints file is required for unit tests. +# List all library dependencies and extras in this file. +google-api-core +proto-plus +protobuf diff --git a/packages/google-cloud-gkerecommender/tests/__init__.py b/packages/google-cloud-gkerecommender/tests/__init__.py new file mode 100644 index 000000000000..cbf94b283c70 --- /dev/null +++ b/packages/google-cloud-gkerecommender/tests/__init__.py @@ -0,0 +1,15 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# diff --git a/packages/google-cloud-gkerecommender/tests/unit/__init__.py b/packages/google-cloud-gkerecommender/tests/unit/__init__.py new file mode 100644 index 000000000000..cbf94b283c70 --- /dev/null +++ b/packages/google-cloud-gkerecommender/tests/unit/__init__.py @@ -0,0 +1,15 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# diff --git a/packages/google-cloud-gkerecommender/tests/unit/gapic/__init__.py b/packages/google-cloud-gkerecommender/tests/unit/gapic/__init__.py new file mode 100644 index 000000000000..cbf94b283c70 --- /dev/null +++ b/packages/google-cloud-gkerecommender/tests/unit/gapic/__init__.py @@ -0,0 +1,15 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# diff --git a/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/__init__.py b/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/__init__.py new file mode 100644 index 000000000000..cbf94b283c70 --- /dev/null +++ b/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/__init__.py @@ -0,0 +1,15 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# diff --git a/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/test_gke_inference_quickstart.py b/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/test_gke_inference_quickstart.py new file mode 100644 index 000000000000..417137074417 --- /dev/null +++ b/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/test_gke_inference_quickstart.py @@ -0,0 +1,6051 @@ +# -*- coding: utf-8 -*- +# Copyright 2025 Google LLC +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# +import os + +# try/except added for compatibility with python < 3.8 +try: + from unittest import mock + from unittest.mock import AsyncMock # pragma: NO COVER +except ImportError: # pragma: NO COVER + import mock + +from collections.abc import AsyncIterable, Iterable +import json +import math + +from google.api_core import api_core_version +from google.protobuf import json_format +import grpc +from grpc.experimental import aio +from proto.marshal.rules import wrappers +from proto.marshal.rules.dates import DurationRule, TimestampRule +import pytest +from requests import PreparedRequest, Request, Response +from requests.sessions import Session + +try: + from google.auth.aio import credentials as ga_credentials_async + + HAS_GOOGLE_AUTH_AIO = True +except ImportError: # pragma: NO COVER + HAS_GOOGLE_AUTH_AIO = False + +from google.api_core import gapic_v1, grpc_helpers, grpc_helpers_async, path_template +from google.api_core import client_options +from google.api_core import exceptions as core_exceptions +from google.api_core import retry as retries +import google.auth +from google.auth import credentials as ga_credentials +from google.auth.exceptions import MutualTLSChannelError +from google.oauth2 import service_account + +from google.cloud.gkerecommender_v1.services.gke_inference_quickstart import ( + GkeInferenceQuickstartAsyncClient, + GkeInferenceQuickstartClient, + pagers, + transports, +) +from google.cloud.gkerecommender_v1.types import gkerecommender + +CRED_INFO_JSON = { + "credential_source": "/path/to/file", + "credential_type": "service account credentials", + "principal": "service-account@example.com", +} +CRED_INFO_STRING = json.dumps(CRED_INFO_JSON) + + +async def mock_async_gen(data, chunk_size=1): + for i in range(0, len(data)): # pragma: NO COVER + chunk = data[i : i + chunk_size] + yield chunk.encode("utf-8") + + +def client_cert_source_callback(): + return b"cert bytes", b"key bytes" + + +# TODO: use async auth anon credentials by default once the minimum version of google-auth is upgraded. +# See related issue: https://github.com/googleapis/gapic-generator-python/issues/2107. +def async_anonymous_credentials(): + if HAS_GOOGLE_AUTH_AIO: + return ga_credentials_async.AnonymousCredentials() + return ga_credentials.AnonymousCredentials() + + +# If default endpoint is localhost, then default mtls endpoint will be the same. +# This method modifies the default endpoint so the client can produce a different +# mtls endpoint for endpoint testing purposes. +def modify_default_endpoint(client): + return ( + "foo.googleapis.com" + if ("localhost" in client.DEFAULT_ENDPOINT) + else client.DEFAULT_ENDPOINT + ) + + +# If default endpoint template is localhost, then default mtls endpoint will be the same. +# This method modifies the default endpoint template so the client can produce a different +# mtls endpoint for endpoint testing purposes. +def modify_default_endpoint_template(client): + return ( + "test.{UNIVERSE_DOMAIN}" + if ("localhost" in client._DEFAULT_ENDPOINT_TEMPLATE) + else client._DEFAULT_ENDPOINT_TEMPLATE + ) + + +def test__get_default_mtls_endpoint(): + api_endpoint = "example.googleapis.com" + api_mtls_endpoint = "example.mtls.googleapis.com" + sandbox_endpoint = "example.sandbox.googleapis.com" + sandbox_mtls_endpoint = "example.mtls.sandbox.googleapis.com" + non_googleapi = "api.example.com" + + assert GkeInferenceQuickstartClient._get_default_mtls_endpoint(None) is None + assert ( + GkeInferenceQuickstartClient._get_default_mtls_endpoint(api_endpoint) + == api_mtls_endpoint + ) + assert ( + GkeInferenceQuickstartClient._get_default_mtls_endpoint(api_mtls_endpoint) + == api_mtls_endpoint + ) + assert ( + GkeInferenceQuickstartClient._get_default_mtls_endpoint(sandbox_endpoint) + == sandbox_mtls_endpoint + ) + assert ( + GkeInferenceQuickstartClient._get_default_mtls_endpoint(sandbox_mtls_endpoint) + == sandbox_mtls_endpoint + ) + assert ( + GkeInferenceQuickstartClient._get_default_mtls_endpoint(non_googleapi) + == non_googleapi + ) + + +def test__read_environment_variables(): + assert GkeInferenceQuickstartClient._read_environment_variables() == ( + False, + "auto", + None, + ) + + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): + assert GkeInferenceQuickstartClient._read_environment_variables() == ( + True, + "auto", + None, + ) + + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "false"}): + assert GkeInferenceQuickstartClient._read_environment_variables() == ( + False, + "auto", + None, + ) + + with mock.patch.dict( + os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "Unsupported"} + ): + with pytest.raises(ValueError) as excinfo: + GkeInferenceQuickstartClient._read_environment_variables() + assert ( + str(excinfo.value) + == "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" + ) + + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): + assert GkeInferenceQuickstartClient._read_environment_variables() == ( + False, + "never", + None, + ) + + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "always"}): + assert GkeInferenceQuickstartClient._read_environment_variables() == ( + False, + "always", + None, + ) + + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "auto"}): + assert GkeInferenceQuickstartClient._read_environment_variables() == ( + False, + "auto", + None, + ) + + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "Unsupported"}): + with pytest.raises(MutualTLSChannelError) as excinfo: + GkeInferenceQuickstartClient._read_environment_variables() + assert ( + str(excinfo.value) + == "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" + ) + + with mock.patch.dict(os.environ, {"GOOGLE_CLOUD_UNIVERSE_DOMAIN": "foo.com"}): + assert GkeInferenceQuickstartClient._read_environment_variables() == ( + False, + "auto", + "foo.com", + ) + + +def test__get_client_cert_source(): + mock_provided_cert_source = mock.Mock() + mock_default_cert_source = mock.Mock() + + assert GkeInferenceQuickstartClient._get_client_cert_source(None, False) is None + assert ( + GkeInferenceQuickstartClient._get_client_cert_source( + mock_provided_cert_source, False + ) + is None + ) + assert ( + GkeInferenceQuickstartClient._get_client_cert_source( + mock_provided_cert_source, True + ) + == mock_provided_cert_source + ) + + with mock.patch( + "google.auth.transport.mtls.has_default_client_cert_source", return_value=True + ): + with mock.patch( + "google.auth.transport.mtls.default_client_cert_source", + return_value=mock_default_cert_source, + ): + assert ( + GkeInferenceQuickstartClient._get_client_cert_source(None, True) + is mock_default_cert_source + ) + assert ( + GkeInferenceQuickstartClient._get_client_cert_source( + mock_provided_cert_source, "true" + ) + is mock_provided_cert_source + ) + + +@mock.patch.object( + GkeInferenceQuickstartClient, + "_DEFAULT_ENDPOINT_TEMPLATE", + modify_default_endpoint_template(GkeInferenceQuickstartClient), +) +@mock.patch.object( + GkeInferenceQuickstartAsyncClient, + "_DEFAULT_ENDPOINT_TEMPLATE", + modify_default_endpoint_template(GkeInferenceQuickstartAsyncClient), +) +def test__get_api_endpoint(): + api_override = "foo.com" + mock_client_cert_source = mock.Mock() + default_universe = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE + default_endpoint = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=default_universe + ) + mock_universe = "bar.com" + mock_endpoint = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=mock_universe + ) + + assert ( + GkeInferenceQuickstartClient._get_api_endpoint( + api_override, mock_client_cert_source, default_universe, "always" + ) + == api_override + ) + assert ( + GkeInferenceQuickstartClient._get_api_endpoint( + None, mock_client_cert_source, default_universe, "auto" + ) + == GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT + ) + assert ( + GkeInferenceQuickstartClient._get_api_endpoint( + None, None, default_universe, "auto" + ) + == default_endpoint + ) + assert ( + GkeInferenceQuickstartClient._get_api_endpoint( + None, None, default_universe, "always" + ) + == GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT + ) + assert ( + GkeInferenceQuickstartClient._get_api_endpoint( + None, mock_client_cert_source, default_universe, "always" + ) + == GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT + ) + assert ( + GkeInferenceQuickstartClient._get_api_endpoint( + None, None, mock_universe, "never" + ) + == mock_endpoint + ) + assert ( + GkeInferenceQuickstartClient._get_api_endpoint( + None, None, default_universe, "never" + ) + == default_endpoint + ) + + with pytest.raises(MutualTLSChannelError) as excinfo: + GkeInferenceQuickstartClient._get_api_endpoint( + None, mock_client_cert_source, mock_universe, "auto" + ) + assert ( + str(excinfo.value) + == "mTLS is not supported in any universe other than googleapis.com." + ) + + +def test__get_universe_domain(): + client_universe_domain = "foo.com" + universe_domain_env = "bar.com" + + assert ( + GkeInferenceQuickstartClient._get_universe_domain( + client_universe_domain, universe_domain_env + ) + == client_universe_domain + ) + assert ( + GkeInferenceQuickstartClient._get_universe_domain(None, universe_domain_env) + == universe_domain_env + ) + assert ( + GkeInferenceQuickstartClient._get_universe_domain(None, None) + == GkeInferenceQuickstartClient._DEFAULT_UNIVERSE + ) + + with pytest.raises(ValueError) as excinfo: + GkeInferenceQuickstartClient._get_universe_domain("", None) + assert str(excinfo.value) == "Universe Domain cannot be an empty string." + + +@pytest.mark.parametrize( + "error_code,cred_info_json,show_cred_info", + [ + (401, CRED_INFO_JSON, True), + (403, CRED_INFO_JSON, True), + (404, CRED_INFO_JSON, True), + (500, CRED_INFO_JSON, False), + (401, None, False), + (403, None, False), + (404, None, False), + (500, None, False), + ], +) +def test__add_cred_info_for_auth_errors(error_code, cred_info_json, show_cred_info): + cred = mock.Mock(["get_cred_info"]) + cred.get_cred_info = mock.Mock(return_value=cred_info_json) + client = GkeInferenceQuickstartClient(credentials=cred) + client._transport._credentials = cred + + error = core_exceptions.GoogleAPICallError("message", details=["foo"]) + error.code = error_code + + client._add_cred_info_for_auth_errors(error) + if show_cred_info: + assert error.details == ["foo", CRED_INFO_STRING] + else: + assert error.details == ["foo"] + + +@pytest.mark.parametrize("error_code", [401, 403, 404, 500]) +def test__add_cred_info_for_auth_errors_no_get_cred_info(error_code): + cred = mock.Mock([]) + assert not hasattr(cred, "get_cred_info") + client = GkeInferenceQuickstartClient(credentials=cred) + client._transport._credentials = cred + + error = core_exceptions.GoogleAPICallError("message", details=[]) + error.code = error_code + + client._add_cred_info_for_auth_errors(error) + assert error.details == [] + + +@pytest.mark.parametrize( + "client_class,transport_name", + [ + (GkeInferenceQuickstartClient, "grpc"), + (GkeInferenceQuickstartAsyncClient, "grpc_asyncio"), + (GkeInferenceQuickstartClient, "rest"), + ], +) +def test_gke_inference_quickstart_client_from_service_account_info( + client_class, transport_name +): + creds = ga_credentials.AnonymousCredentials() + with mock.patch.object( + service_account.Credentials, "from_service_account_info" + ) as factory: + factory.return_value = creds + info = {"valid": True} + client = client_class.from_service_account_info(info, transport=transport_name) + assert client.transport._credentials == creds + assert isinstance(client, client_class) + + assert client.transport._host == ( + "gkerecommender.googleapis.com:443" + if transport_name in ["grpc", "grpc_asyncio"] + else "https://gkerecommender.googleapis.com" + ) + + +@pytest.mark.parametrize( + "transport_class,transport_name", + [ + (transports.GkeInferenceQuickstartGrpcTransport, "grpc"), + (transports.GkeInferenceQuickstartGrpcAsyncIOTransport, "grpc_asyncio"), + (transports.GkeInferenceQuickstartRestTransport, "rest"), + ], +) +def test_gke_inference_quickstart_client_service_account_always_use_jwt( + transport_class, transport_name +): + with mock.patch.object( + service_account.Credentials, "with_always_use_jwt_access", create=True + ) as use_jwt: + creds = service_account.Credentials(None, None, None) + transport = transport_class(credentials=creds, always_use_jwt_access=True) + use_jwt.assert_called_once_with(True) + + with mock.patch.object( + service_account.Credentials, "with_always_use_jwt_access", create=True + ) as use_jwt: + creds = service_account.Credentials(None, None, None) + transport = transport_class(credentials=creds, always_use_jwt_access=False) + use_jwt.assert_not_called() + + +@pytest.mark.parametrize( + "client_class,transport_name", + [ + (GkeInferenceQuickstartClient, "grpc"), + (GkeInferenceQuickstartAsyncClient, "grpc_asyncio"), + (GkeInferenceQuickstartClient, "rest"), + ], +) +def test_gke_inference_quickstart_client_from_service_account_file( + client_class, transport_name +): + creds = ga_credentials.AnonymousCredentials() + with mock.patch.object( + service_account.Credentials, "from_service_account_file" + ) as factory: + factory.return_value = creds + client = client_class.from_service_account_file( + "dummy/file/path.json", transport=transport_name + ) + assert client.transport._credentials == creds + assert isinstance(client, client_class) + + client = client_class.from_service_account_json( + "dummy/file/path.json", transport=transport_name + ) + assert client.transport._credentials == creds + assert isinstance(client, client_class) + + assert client.transport._host == ( + "gkerecommender.googleapis.com:443" + if transport_name in ["grpc", "grpc_asyncio"] + else "https://gkerecommender.googleapis.com" + ) + + +def test_gke_inference_quickstart_client_get_transport_class(): + transport = GkeInferenceQuickstartClient.get_transport_class() + available_transports = [ + transports.GkeInferenceQuickstartGrpcTransport, + transports.GkeInferenceQuickstartRestTransport, + ] + assert transport in available_transports + + transport = GkeInferenceQuickstartClient.get_transport_class("grpc") + assert transport == transports.GkeInferenceQuickstartGrpcTransport + + +@pytest.mark.parametrize( + "client_class,transport_class,transport_name", + [ + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartGrpcTransport, + "grpc", + ), + ( + GkeInferenceQuickstartAsyncClient, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + "grpc_asyncio", + ), + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartRestTransport, + "rest", + ), + ], +) +@mock.patch.object( + GkeInferenceQuickstartClient, + "_DEFAULT_ENDPOINT_TEMPLATE", + modify_default_endpoint_template(GkeInferenceQuickstartClient), +) +@mock.patch.object( + GkeInferenceQuickstartAsyncClient, + "_DEFAULT_ENDPOINT_TEMPLATE", + modify_default_endpoint_template(GkeInferenceQuickstartAsyncClient), +) +def test_gke_inference_quickstart_client_client_options( + client_class, transport_class, transport_name +): + # Check that if channel is provided we won't create a new one. + with mock.patch.object(GkeInferenceQuickstartClient, "get_transport_class") as gtc: + transport = transport_class(credentials=ga_credentials.AnonymousCredentials()) + client = client_class(transport=transport) + gtc.assert_not_called() + + # Check that if channel is provided via str we will create a new one. + with mock.patch.object(GkeInferenceQuickstartClient, "get_transport_class") as gtc: + client = client_class(transport=transport_name) + gtc.assert_called() + + # Check the case api_endpoint is provided. + options = client_options.ClientOptions(api_endpoint="squid.clam.whelk") + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(transport=transport_name, client_options=options) + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host="squid.clam.whelk", + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + # Check the case api_endpoint is not provided and GOOGLE_API_USE_MTLS_ENDPOINT is + # "never". + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host=client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ), + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + # Check the case api_endpoint is not provided and GOOGLE_API_USE_MTLS_ENDPOINT is + # "always". + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "always"}): + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host=client.DEFAULT_MTLS_ENDPOINT, + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + # Check the case api_endpoint is not provided and GOOGLE_API_USE_MTLS_ENDPOINT has + # unsupported value. + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "Unsupported"}): + with pytest.raises(MutualTLSChannelError) as excinfo: + client = client_class(transport=transport_name) + assert ( + str(excinfo.value) + == "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" + ) + + # Check the case GOOGLE_API_USE_CLIENT_CERTIFICATE has unsupported value. + with mock.patch.dict( + os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "Unsupported"} + ): + with pytest.raises(ValueError) as excinfo: + client = client_class(transport=transport_name) + assert ( + str(excinfo.value) + == "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" + ) + + # Check the case quota_project_id is provided + options = client_options.ClientOptions(quota_project_id="octopus") + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(client_options=options, transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host=client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ), + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id="octopus", + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + # Check the case api_endpoint is provided + options = client_options.ClientOptions( + api_audience="https://language.googleapis.com" + ) + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(client_options=options, transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host=client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ), + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience="https://language.googleapis.com", + ) + + +@pytest.mark.parametrize( + "client_class,transport_class,transport_name,use_client_cert_env", + [ + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartGrpcTransport, + "grpc", + "true", + ), + ( + GkeInferenceQuickstartAsyncClient, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + "grpc_asyncio", + "true", + ), + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartGrpcTransport, + "grpc", + "false", + ), + ( + GkeInferenceQuickstartAsyncClient, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + "grpc_asyncio", + "false", + ), + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartRestTransport, + "rest", + "true", + ), + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartRestTransport, + "rest", + "false", + ), + ], +) +@mock.patch.object( + GkeInferenceQuickstartClient, + "_DEFAULT_ENDPOINT_TEMPLATE", + modify_default_endpoint_template(GkeInferenceQuickstartClient), +) +@mock.patch.object( + GkeInferenceQuickstartAsyncClient, + "_DEFAULT_ENDPOINT_TEMPLATE", + modify_default_endpoint_template(GkeInferenceQuickstartAsyncClient), +) +@mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "auto"}) +def test_gke_inference_quickstart_client_mtls_env_auto( + client_class, transport_class, transport_name, use_client_cert_env +): + # This tests the endpoint autoswitch behavior. Endpoint is autoswitched to the default + # mtls endpoint, if GOOGLE_API_USE_CLIENT_CERTIFICATE is "true" and client cert exists. + + # Check the case client_cert_source is provided. Whether client cert is used depends on + # GOOGLE_API_USE_CLIENT_CERTIFICATE value. + with mock.patch.dict( + os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": use_client_cert_env} + ): + options = client_options.ClientOptions( + client_cert_source=client_cert_source_callback + ) + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(client_options=options, transport=transport_name) + + if use_client_cert_env == "false": + expected_client_cert_source = None + expected_host = client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ) + else: + expected_client_cert_source = client_cert_source_callback + expected_host = client.DEFAULT_MTLS_ENDPOINT + + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host=expected_host, + scopes=None, + client_cert_source_for_mtls=expected_client_cert_source, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + # Check the case ADC client cert is provided. Whether client cert is used depends on + # GOOGLE_API_USE_CLIENT_CERTIFICATE value. + with mock.patch.dict( + os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": use_client_cert_env} + ): + with mock.patch.object(transport_class, "__init__") as patched: + with mock.patch( + "google.auth.transport.mtls.has_default_client_cert_source", + return_value=True, + ): + with mock.patch( + "google.auth.transport.mtls.default_client_cert_source", + return_value=client_cert_source_callback, + ): + if use_client_cert_env == "false": + expected_host = client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ) + expected_client_cert_source = None + else: + expected_host = client.DEFAULT_MTLS_ENDPOINT + expected_client_cert_source = client_cert_source_callback + + patched.return_value = None + client = client_class(transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host=expected_host, + scopes=None, + client_cert_source_for_mtls=expected_client_cert_source, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + # Check the case client_cert_source and ADC client cert are not provided. + with mock.patch.dict( + os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": use_client_cert_env} + ): + with mock.patch.object(transport_class, "__init__") as patched: + with mock.patch( + "google.auth.transport.mtls.has_default_client_cert_source", + return_value=False, + ): + patched.return_value = None + client = client_class(transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host=client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ), + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + +@pytest.mark.parametrize( + "client_class", [GkeInferenceQuickstartClient, GkeInferenceQuickstartAsyncClient] +) +@mock.patch.object( + GkeInferenceQuickstartClient, + "DEFAULT_ENDPOINT", + modify_default_endpoint(GkeInferenceQuickstartClient), +) +@mock.patch.object( + GkeInferenceQuickstartAsyncClient, + "DEFAULT_ENDPOINT", + modify_default_endpoint(GkeInferenceQuickstartAsyncClient), +) +def test_gke_inference_quickstart_client_get_mtls_endpoint_and_cert_source( + client_class, +): + mock_client_cert_source = mock.Mock() + + # Test the case GOOGLE_API_USE_CLIENT_CERTIFICATE is "true". + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): + mock_api_endpoint = "foo" + options = client_options.ClientOptions( + client_cert_source=mock_client_cert_source, api_endpoint=mock_api_endpoint + ) + api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source( + options + ) + assert api_endpoint == mock_api_endpoint + assert cert_source == mock_client_cert_source + + # Test the case GOOGLE_API_USE_CLIENT_CERTIFICATE is "false". + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "false"}): + mock_client_cert_source = mock.Mock() + mock_api_endpoint = "foo" + options = client_options.ClientOptions( + client_cert_source=mock_client_cert_source, api_endpoint=mock_api_endpoint + ) + api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source( + options + ) + assert api_endpoint == mock_api_endpoint + assert cert_source is None + + # Test the case GOOGLE_API_USE_MTLS_ENDPOINT is "never". + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): + api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source() + assert api_endpoint == client_class.DEFAULT_ENDPOINT + assert cert_source is None + + # Test the case GOOGLE_API_USE_MTLS_ENDPOINT is "always". + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "always"}): + api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source() + assert api_endpoint == client_class.DEFAULT_MTLS_ENDPOINT + assert cert_source is None + + # Test the case GOOGLE_API_USE_MTLS_ENDPOINT is "auto" and default cert doesn't exist. + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): + with mock.patch( + "google.auth.transport.mtls.has_default_client_cert_source", + return_value=False, + ): + api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source() + assert api_endpoint == client_class.DEFAULT_ENDPOINT + assert cert_source is None + + # Test the case GOOGLE_API_USE_MTLS_ENDPOINT is "auto" and default cert exists. + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): + with mock.patch( + "google.auth.transport.mtls.has_default_client_cert_source", + return_value=True, + ): + with mock.patch( + "google.auth.transport.mtls.default_client_cert_source", + return_value=mock_client_cert_source, + ): + ( + api_endpoint, + cert_source, + ) = client_class.get_mtls_endpoint_and_cert_source() + assert api_endpoint == client_class.DEFAULT_MTLS_ENDPOINT + assert cert_source == mock_client_cert_source + + # Check the case api_endpoint is not provided and GOOGLE_API_USE_MTLS_ENDPOINT has + # unsupported value. + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "Unsupported"}): + with pytest.raises(MutualTLSChannelError) as excinfo: + client_class.get_mtls_endpoint_and_cert_source() + + assert ( + str(excinfo.value) + == "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" + ) + + # Check the case GOOGLE_API_USE_CLIENT_CERTIFICATE has unsupported value. + with mock.patch.dict( + os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "Unsupported"} + ): + with pytest.raises(ValueError) as excinfo: + client_class.get_mtls_endpoint_and_cert_source() + + assert ( + str(excinfo.value) + == "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" + ) + + +@pytest.mark.parametrize( + "client_class", [GkeInferenceQuickstartClient, GkeInferenceQuickstartAsyncClient] +) +@mock.patch.object( + GkeInferenceQuickstartClient, + "_DEFAULT_ENDPOINT_TEMPLATE", + modify_default_endpoint_template(GkeInferenceQuickstartClient), +) +@mock.patch.object( + GkeInferenceQuickstartAsyncClient, + "_DEFAULT_ENDPOINT_TEMPLATE", + modify_default_endpoint_template(GkeInferenceQuickstartAsyncClient), +) +def test_gke_inference_quickstart_client_client_api_endpoint(client_class): + mock_client_cert_source = client_cert_source_callback + api_override = "foo.com" + default_universe = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE + default_endpoint = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=default_universe + ) + mock_universe = "bar.com" + mock_endpoint = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=mock_universe + ) + + # If ClientOptions.api_endpoint is set and GOOGLE_API_USE_CLIENT_CERTIFICATE="true", + # use ClientOptions.api_endpoint as the api endpoint regardless. + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): + with mock.patch( + "google.auth.transport.requests.AuthorizedSession.configure_mtls_channel" + ): + options = client_options.ClientOptions( + client_cert_source=mock_client_cert_source, api_endpoint=api_override + ) + client = client_class( + client_options=options, + credentials=ga_credentials.AnonymousCredentials(), + ) + assert client.api_endpoint == api_override + + # If ClientOptions.api_endpoint is not set and GOOGLE_API_USE_MTLS_ENDPOINT="never", + # use the _DEFAULT_ENDPOINT_TEMPLATE populated with GDU as the api endpoint. + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): + client = client_class(credentials=ga_credentials.AnonymousCredentials()) + assert client.api_endpoint == default_endpoint + + # If ClientOptions.api_endpoint is not set and GOOGLE_API_USE_MTLS_ENDPOINT="always", + # use the DEFAULT_MTLS_ENDPOINT as the api endpoint. + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "always"}): + client = client_class(credentials=ga_credentials.AnonymousCredentials()) + assert client.api_endpoint == client_class.DEFAULT_MTLS_ENDPOINT + + # If ClientOptions.api_endpoint is not set, GOOGLE_API_USE_MTLS_ENDPOINT="auto" (default), + # GOOGLE_API_USE_CLIENT_CERTIFICATE="false" (default), default cert source doesn't exist, + # and ClientOptions.universe_domain="bar.com", + # use the _DEFAULT_ENDPOINT_TEMPLATE populated with universe domain as the api endpoint. + options = client_options.ClientOptions() + universe_exists = hasattr(options, "universe_domain") + if universe_exists: + options = client_options.ClientOptions(universe_domain=mock_universe) + client = client_class( + client_options=options, credentials=ga_credentials.AnonymousCredentials() + ) + else: + client = client_class( + client_options=options, credentials=ga_credentials.AnonymousCredentials() + ) + assert client.api_endpoint == ( + mock_endpoint if universe_exists else default_endpoint + ) + assert client.universe_domain == ( + mock_universe if universe_exists else default_universe + ) + + # If ClientOptions does not have a universe domain attribute and GOOGLE_API_USE_MTLS_ENDPOINT="never", + # use the _DEFAULT_ENDPOINT_TEMPLATE populated with GDU as the api endpoint. + options = client_options.ClientOptions() + if hasattr(options, "universe_domain"): + delattr(options, "universe_domain") + with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): + client = client_class( + client_options=options, credentials=ga_credentials.AnonymousCredentials() + ) + assert client.api_endpoint == default_endpoint + + +@pytest.mark.parametrize( + "client_class,transport_class,transport_name", + [ + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartGrpcTransport, + "grpc", + ), + ( + GkeInferenceQuickstartAsyncClient, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + "grpc_asyncio", + ), + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartRestTransport, + "rest", + ), + ], +) +def test_gke_inference_quickstart_client_client_options_scopes( + client_class, transport_class, transport_name +): + # Check the case scopes are provided. + options = client_options.ClientOptions( + scopes=["1", "2"], + ) + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(client_options=options, transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file=None, + host=client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ), + scopes=["1", "2"], + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + +@pytest.mark.parametrize( + "client_class,transport_class,transport_name,grpc_helpers", + [ + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartGrpcTransport, + "grpc", + grpc_helpers, + ), + ( + GkeInferenceQuickstartAsyncClient, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + "grpc_asyncio", + grpc_helpers_async, + ), + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartRestTransport, + "rest", + None, + ), + ], +) +def test_gke_inference_quickstart_client_client_options_credentials_file( + client_class, transport_class, transport_name, grpc_helpers +): + # Check the case credentials file is provided. + options = client_options.ClientOptions(credentials_file="credentials.json") + + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(client_options=options, transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file="credentials.json", + host=client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ), + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + +def test_gke_inference_quickstart_client_client_options_from_dict(): + with mock.patch( + "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.GkeInferenceQuickstartGrpcTransport.__init__" + ) as grpc_transport: + grpc_transport.return_value = None + client = GkeInferenceQuickstartClient( + client_options={"api_endpoint": "squid.clam.whelk"} + ) + grpc_transport.assert_called_once_with( + credentials=None, + credentials_file=None, + host="squid.clam.whelk", + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + +@pytest.mark.parametrize( + "client_class,transport_class,transport_name,grpc_helpers", + [ + ( + GkeInferenceQuickstartClient, + transports.GkeInferenceQuickstartGrpcTransport, + "grpc", + grpc_helpers, + ), + ( + GkeInferenceQuickstartAsyncClient, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + "grpc_asyncio", + grpc_helpers_async, + ), + ], +) +def test_gke_inference_quickstart_client_create_channel_credentials_file( + client_class, transport_class, transport_name, grpc_helpers +): + # Check the case credentials file is provided. + options = client_options.ClientOptions(credentials_file="credentials.json") + + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(client_options=options, transport=transport_name) + patched.assert_called_once_with( + credentials=None, + credentials_file="credentials.json", + host=client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ), + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) + + # test that the credentials from file are saved and used as the credentials. + with mock.patch.object( + google.auth, "load_credentials_from_file", autospec=True + ) as load_creds, mock.patch.object( + google.auth, "default", autospec=True + ) as adc, mock.patch.object( + grpc_helpers, "create_channel" + ) as create_channel: + creds = ga_credentials.AnonymousCredentials() + file_creds = ga_credentials.AnonymousCredentials() + load_creds.return_value = (file_creds, None) + adc.return_value = (creds, None) + client = client_class(client_options=options, transport=transport_name) + create_channel.assert_called_with( + "gkerecommender.googleapis.com:443", + credentials=file_creds, + credentials_file=None, + quota_project_id=None, + default_scopes=("https://www.googleapis.com/auth/cloud-platform",), + scopes=None, + default_host="gkerecommender.googleapis.com", + ssl_credentials=None, + options=[ + ("grpc.max_send_message_length", -1), + ("grpc.max_receive_message_length", -1), + ], + ) + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchModelsRequest, + dict, + ], +) +def test_fetch_models(request_type, transport: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: + # Designate an appropriate return value for the call. + call.return_value = gkerecommender.FetchModelsResponse( + models=["models_value"], + next_page_token="next_page_token_value", + ) + response = client.fetch_models(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) == 1 + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchModelsRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelsPager) + assert response.models == ["models_value"] + assert response.next_page_token == "next_page_token_value" + + +def test_fetch_models_non_empty_request_with_auto_populated_field(): + # This test is a coverage failsafe to make sure that UUID4 fields are + # automatically populated, according to AIP-4235, with non-empty requests. + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Populate all string fields in the request which are not UUID4 + # since we want to check that UUID4 are populated automatically + # if they meet the requirements of AIP 4235. + request = gkerecommender.FetchModelsRequest( + page_token="page_token_value", + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: + call.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client.fetch_models(request=request) + call.assert_called() + _, args, _ = call.mock_calls[0] + assert args[0] == gkerecommender.FetchModelsRequest( + page_token="page_token_value", + ) + + +def test_fetch_models_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert client._transport.fetch_models in client._transport._wrapped_methods + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[client._transport.fetch_models] = mock_rpc + request = {} + client.fetch_models(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_models(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_models_async_use_cached_wrapped_rpc( + transport: str = "grpc_asyncio", +): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._client._transport.fetch_models + in client._client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.AsyncMock() + mock_rpc.return_value = mock.Mock() + client._client._transport._wrapped_methods[ + client._client._transport.fetch_models + ] = mock_rpc + + request = {} + await client.fetch_models(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + await client.fetch_models(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_models_async( + transport: str = "grpc_asyncio", request_type=gkerecommender.FetchModelsRequest +): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchModelsResponse( + models=["models_value"], + next_page_token="next_page_token_value", + ) + ) + response = await client.fetch_models(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchModelsRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelsAsyncPager) + assert response.models == ["models_value"] + assert response.next_page_token == "next_page_token_value" + + +@pytest.mark.asyncio +async def test_fetch_models_async_from_dict(): + await test_fetch_models_async(request_type=dict) + + +def test_fetch_models_pager(transport_name: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport_name, + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelsResponse( + models=[], + next_page_token="def", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + ], + ), + RuntimeError, + ) + + expected_metadata = () + retry = retries.Retry() + timeout = 5 + pager = client.fetch_models(request={}, retry=retry, timeout=timeout) + + assert pager._metadata == expected_metadata + assert pager._retry == retry + assert pager._timeout == timeout + + results = list(pager) + assert len(results) == 6 + assert all(isinstance(i, str) for i in results) + + +def test_fetch_models_pages(transport_name: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport_name, + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelsResponse( + models=[], + next_page_token="def", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + ], + ), + RuntimeError, + ) + pages = list(client.fetch_models(request={}).pages) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +@pytest.mark.asyncio +async def test_fetch_models_async_pager(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_models), "__call__", new_callable=mock.AsyncMock + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelsResponse( + models=[], + next_page_token="def", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + ], + ), + RuntimeError, + ) + async_pager = await client.fetch_models( + request={}, + ) + assert async_pager.next_page_token == "abc" + responses = [] + async for response in async_pager: # pragma: no branch + responses.append(response) + + assert len(responses) == 6 + assert all(isinstance(i, str) for i in responses) + + +@pytest.mark.asyncio +async def test_fetch_models_async_pages(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_models), "__call__", new_callable=mock.AsyncMock + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelsResponse( + models=[], + next_page_token="def", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + ], + ), + RuntimeError, + ) + pages = [] + # Workaround issue in python 3.9 related to code coverage by adding `# pragma: no branch` + # See https://github.com/googleapis/gapic-generator-python/pull/1174#issuecomment-1025132372 + async for page_ in ( # pragma: no branch + await client.fetch_models(request={}) + ).pages: + pages.append(page_) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchModelServersRequest, + dict, + ], +) +def test_fetch_model_servers(request_type, transport: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = gkerecommender.FetchModelServersResponse( + model_servers=["model_servers_value"], + next_page_token="next_page_token_value", + ) + response = client.fetch_model_servers(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) == 1 + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchModelServersRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelServersPager) + assert response.model_servers == ["model_servers_value"] + assert response.next_page_token == "next_page_token_value" + + +def test_fetch_model_servers_non_empty_request_with_auto_populated_field(): + # This test is a coverage failsafe to make sure that UUID4 fields are + # automatically populated, according to AIP-4235, with non-empty requests. + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Populate all string fields in the request which are not UUID4 + # since we want to check that UUID4 are populated automatically + # if they meet the requirements of AIP 4235. + request = gkerecommender.FetchModelServersRequest( + model="model_value", + page_token="page_token_value", + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), "__call__" + ) as call: + call.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client.fetch_model_servers(request=request) + call.assert_called() + _, args, _ = call.mock_calls[0] + assert args[0] == gkerecommender.FetchModelServersRequest( + model="model_value", + page_token="page_token_value", + ) + + +def test_fetch_model_servers_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._transport.fetch_model_servers in client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[ + client._transport.fetch_model_servers + ] = mock_rpc + request = {} + client.fetch_model_servers(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_model_servers(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_model_servers_async_use_cached_wrapped_rpc( + transport: str = "grpc_asyncio", +): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._client._transport.fetch_model_servers + in client._client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.AsyncMock() + mock_rpc.return_value = mock.Mock() + client._client._transport._wrapped_methods[ + client._client._transport.fetch_model_servers + ] = mock_rpc + + request = {} + await client.fetch_model_servers(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + await client.fetch_model_servers(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_model_servers_async( + transport: str = "grpc_asyncio", + request_type=gkerecommender.FetchModelServersRequest, +): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchModelServersResponse( + model_servers=["model_servers_value"], + next_page_token="next_page_token_value", + ) + ) + response = await client.fetch_model_servers(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchModelServersRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelServersAsyncPager) + assert response.model_servers == ["model_servers_value"] + assert response.next_page_token == "next_page_token_value" + + +@pytest.mark.asyncio +async def test_fetch_model_servers_async_from_dict(): + await test_fetch_model_servers_async(request_type=dict) + + +def test_fetch_model_servers_pager(transport_name: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport_name, + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), "__call__" + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[], + next_page_token="def", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + ], + ), + RuntimeError, + ) + + expected_metadata = () + retry = retries.Retry() + timeout = 5 + pager = client.fetch_model_servers(request={}, retry=retry, timeout=timeout) + + assert pager._metadata == expected_metadata + assert pager._retry == retry + assert pager._timeout == timeout + + results = list(pager) + assert len(results) == 6 + assert all(isinstance(i, str) for i in results) + + +def test_fetch_model_servers_pages(transport_name: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport_name, + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), "__call__" + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[], + next_page_token="def", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + ], + ), + RuntimeError, + ) + pages = list(client.fetch_model_servers(request={}).pages) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +@pytest.mark.asyncio +async def test_fetch_model_servers_async_pager(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), + "__call__", + new_callable=mock.AsyncMock, + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[], + next_page_token="def", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + ], + ), + RuntimeError, + ) + async_pager = await client.fetch_model_servers( + request={}, + ) + assert async_pager.next_page_token == "abc" + responses = [] + async for response in async_pager: # pragma: no branch + responses.append(response) + + assert len(responses) == 6 + assert all(isinstance(i, str) for i in responses) + + +@pytest.mark.asyncio +async def test_fetch_model_servers_async_pages(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), + "__call__", + new_callable=mock.AsyncMock, + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[], + next_page_token="def", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + ], + ), + RuntimeError, + ) + pages = [] + # Workaround issue in python 3.9 related to code coverage by adding `# pragma: no branch` + # See https://github.com/googleapis/gapic-generator-python/pull/1174#issuecomment-1025132372 + async for page_ in ( # pragma: no branch + await client.fetch_model_servers(request={}) + ).pages: + pages.append(page_) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchModelServerVersionsRequest, + dict, + ], +) +def test_fetch_model_server_versions(request_type, transport: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=["model_server_versions_value"], + next_page_token="next_page_token_value", + ) + response = client.fetch_model_server_versions(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) == 1 + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchModelServerVersionsRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelServerVersionsPager) + assert response.model_server_versions == ["model_server_versions_value"] + assert response.next_page_token == "next_page_token_value" + + +def test_fetch_model_server_versions_non_empty_request_with_auto_populated_field(): + # This test is a coverage failsafe to make sure that UUID4 fields are + # automatically populated, according to AIP-4235, with non-empty requests. + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Populate all string fields in the request which are not UUID4 + # since we want to check that UUID4 are populated automatically + # if they meet the requirements of AIP 4235. + request = gkerecommender.FetchModelServerVersionsRequest( + model="model_value", + model_server="model_server_value", + page_token="page_token_value", + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), "__call__" + ) as call: + call.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client.fetch_model_server_versions(request=request) + call.assert_called() + _, args, _ = call.mock_calls[0] + assert args[0] == gkerecommender.FetchModelServerVersionsRequest( + model="model_value", + model_server="model_server_value", + page_token="page_token_value", + ) + + +def test_fetch_model_server_versions_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._transport.fetch_model_server_versions + in client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[ + client._transport.fetch_model_server_versions + ] = mock_rpc + request = {} + client.fetch_model_server_versions(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_model_server_versions(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_model_server_versions_async_use_cached_wrapped_rpc( + transport: str = "grpc_asyncio", +): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._client._transport.fetch_model_server_versions + in client._client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.AsyncMock() + mock_rpc.return_value = mock.Mock() + client._client._transport._wrapped_methods[ + client._client._transport.fetch_model_server_versions + ] = mock_rpc + + request = {} + await client.fetch_model_server_versions(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + await client.fetch_model_server_versions(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_model_server_versions_async( + transport: str = "grpc_asyncio", + request_type=gkerecommender.FetchModelServerVersionsRequest, +): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=["model_server_versions_value"], + next_page_token="next_page_token_value", + ) + ) + response = await client.fetch_model_server_versions(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchModelServerVersionsRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelServerVersionsAsyncPager) + assert response.model_server_versions == ["model_server_versions_value"] + assert response.next_page_token == "next_page_token_value" + + +@pytest.mark.asyncio +async def test_fetch_model_server_versions_async_from_dict(): + await test_fetch_model_server_versions_async(request_type=dict) + + +def test_fetch_model_server_versions_pager(transport_name: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport_name, + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), "__call__" + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[], + next_page_token="def", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + ], + ), + RuntimeError, + ) + + expected_metadata = () + retry = retries.Retry() + timeout = 5 + pager = client.fetch_model_server_versions( + request={}, retry=retry, timeout=timeout + ) + + assert pager._metadata == expected_metadata + assert pager._retry == retry + assert pager._timeout == timeout + + results = list(pager) + assert len(results) == 6 + assert all(isinstance(i, str) for i in results) + + +def test_fetch_model_server_versions_pages(transport_name: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport_name, + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), "__call__" + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[], + next_page_token="def", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + ], + ), + RuntimeError, + ) + pages = list(client.fetch_model_server_versions(request={}).pages) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +@pytest.mark.asyncio +async def test_fetch_model_server_versions_async_pager(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), + "__call__", + new_callable=mock.AsyncMock, + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[], + next_page_token="def", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + ], + ), + RuntimeError, + ) + async_pager = await client.fetch_model_server_versions( + request={}, + ) + assert async_pager.next_page_token == "abc" + responses = [] + async for response in async_pager: # pragma: no branch + responses.append(response) + + assert len(responses) == 6 + assert all(isinstance(i, str) for i in responses) + + +@pytest.mark.asyncio +async def test_fetch_model_server_versions_async_pages(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), + "__call__", + new_callable=mock.AsyncMock, + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[], + next_page_token="def", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + ], + ), + RuntimeError, + ) + pages = [] + # Workaround issue in python 3.9 related to code coverage by adding `# pragma: no branch` + # See https://github.com/googleapis/gapic-generator-python/pull/1174#issuecomment-1025132372 + async for page_ in ( # pragma: no branch + await client.fetch_model_server_versions(request={}) + ).pages: + pages.append(page_) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchProfilesRequest, + dict, + ], +) +def test_fetch_profiles(request_type, transport: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: + # Designate an appropriate return value for the call. + call.return_value = gkerecommender.FetchProfilesResponse( + comments="comments_value", + next_page_token="next_page_token_value", + ) + response = client.fetch_profiles(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) == 1 + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchProfilesRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchProfilesPager) + assert response.comments == "comments_value" + assert response.next_page_token == "next_page_token_value" + + +def test_fetch_profiles_non_empty_request_with_auto_populated_field(): + # This test is a coverage failsafe to make sure that UUID4 fields are + # automatically populated, according to AIP-4235, with non-empty requests. + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Populate all string fields in the request which are not UUID4 + # since we want to check that UUID4 are populated automatically + # if they meet the requirements of AIP 4235. + request = gkerecommender.FetchProfilesRequest( + model="model_value", + model_server="model_server_value", + model_server_version="model_server_version_value", + page_token="page_token_value", + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: + call.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client.fetch_profiles(request=request) + call.assert_called() + _, args, _ = call.mock_calls[0] + assert args[0] == gkerecommender.FetchProfilesRequest( + model="model_value", + model_server="model_server_value", + model_server_version="model_server_version_value", + page_token="page_token_value", + ) + + +def test_fetch_profiles_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert client._transport.fetch_profiles in client._transport._wrapped_methods + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[client._transport.fetch_profiles] = mock_rpc + request = {} + client.fetch_profiles(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_profiles(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_profiles_async_use_cached_wrapped_rpc( + transport: str = "grpc_asyncio", +): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._client._transport.fetch_profiles + in client._client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.AsyncMock() + mock_rpc.return_value = mock.Mock() + client._client._transport._wrapped_methods[ + client._client._transport.fetch_profiles + ] = mock_rpc + + request = {} + await client.fetch_profiles(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + await client.fetch_profiles(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_profiles_async( + transport: str = "grpc_asyncio", request_type=gkerecommender.FetchProfilesRequest +): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchProfilesResponse( + comments="comments_value", + next_page_token="next_page_token_value", + ) + ) + response = await client.fetch_profiles(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchProfilesRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchProfilesAsyncPager) + assert response.comments == "comments_value" + assert response.next_page_token == "next_page_token_value" + + +@pytest.mark.asyncio +async def test_fetch_profiles_async_from_dict(): + await test_fetch_profiles_async(request_type=dict) + + +def test_fetch_profiles_pager(transport_name: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport_name, + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + next_page_token="abc", + ), + gkerecommender.FetchProfilesResponse( + profile=[], + next_page_token="def", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + ), + RuntimeError, + ) + + expected_metadata = () + retry = retries.Retry() + timeout = 5 + pager = client.fetch_profiles(request={}, retry=retry, timeout=timeout) + + assert pager._metadata == expected_metadata + assert pager._retry == retry + assert pager._timeout == timeout + + results = list(pager) + assert len(results) == 6 + assert all(isinstance(i, gkerecommender.Profile) for i in results) + + +def test_fetch_profiles_pages(transport_name: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport_name, + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + next_page_token="abc", + ), + gkerecommender.FetchProfilesResponse( + profile=[], + next_page_token="def", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + ), + RuntimeError, + ) + pages = list(client.fetch_profiles(request={}).pages) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +@pytest.mark.asyncio +async def test_fetch_profiles_async_pager(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_profiles), "__call__", new_callable=mock.AsyncMock + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + next_page_token="abc", + ), + gkerecommender.FetchProfilesResponse( + profile=[], + next_page_token="def", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + ), + RuntimeError, + ) + async_pager = await client.fetch_profiles( + request={}, + ) + assert async_pager.next_page_token == "abc" + responses = [] + async for response in async_pager: # pragma: no branch + responses.append(response) + + assert len(responses) == 6 + assert all(isinstance(i, gkerecommender.Profile) for i in responses) + + +@pytest.mark.asyncio +async def test_fetch_profiles_async_pages(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_profiles), "__call__", new_callable=mock.AsyncMock + ) as call: + # Set the response to a series of pages. + call.side_effect = ( + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + next_page_token="abc", + ), + gkerecommender.FetchProfilesResponse( + profile=[], + next_page_token="def", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + ), + RuntimeError, + ) + pages = [] + # Workaround issue in python 3.9 related to code coverage by adding `# pragma: no branch` + # See https://github.com/googleapis/gapic-generator-python/pull/1174#issuecomment-1025132372 + async for page_ in ( # pragma: no branch + await client.fetch_profiles(request={}) + ).pages: + pages.append(page_) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.GenerateOptimizedManifestRequest, + dict, + ], +) +def test_generate_optimized_manifest(request_type, transport: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.generate_optimized_manifest), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = gkerecommender.GenerateOptimizedManifestResponse( + comments=["comments_value"], + manifest_version="manifest_version_value", + ) + response = client.generate_optimized_manifest(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) == 1 + _, args, _ = call.mock_calls[0] + request = gkerecommender.GenerateOptimizedManifestRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, gkerecommender.GenerateOptimizedManifestResponse) + assert response.comments == ["comments_value"] + assert response.manifest_version == "manifest_version_value" + + +def test_generate_optimized_manifest_non_empty_request_with_auto_populated_field(): + # This test is a coverage failsafe to make sure that UUID4 fields are + # automatically populated, according to AIP-4235, with non-empty requests. + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Populate all string fields in the request which are not UUID4 + # since we want to check that UUID4 are populated automatically + # if they meet the requirements of AIP 4235. + request = gkerecommender.GenerateOptimizedManifestRequest( + accelerator_type="accelerator_type_value", + kubernetes_namespace="kubernetes_namespace_value", + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.generate_optimized_manifest), "__call__" + ) as call: + call.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client.generate_optimized_manifest(request=request) + call.assert_called() + _, args, _ = call.mock_calls[0] + assert args[0] == gkerecommender.GenerateOptimizedManifestRequest( + accelerator_type="accelerator_type_value", + kubernetes_namespace="kubernetes_namespace_value", + ) + + +def test_generate_optimized_manifest_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._transport.generate_optimized_manifest + in client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[ + client._transport.generate_optimized_manifest + ] = mock_rpc + request = {} + client.generate_optimized_manifest(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.generate_optimized_manifest(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_generate_optimized_manifest_async_use_cached_wrapped_rpc( + transport: str = "grpc_asyncio", +): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._client._transport.generate_optimized_manifest + in client._client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.AsyncMock() + mock_rpc.return_value = mock.Mock() + client._client._transport._wrapped_methods[ + client._client._transport.generate_optimized_manifest + ] = mock_rpc + + request = {} + await client.generate_optimized_manifest(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + await client.generate_optimized_manifest(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_generate_optimized_manifest_async( + transport: str = "grpc_asyncio", + request_type=gkerecommender.GenerateOptimizedManifestRequest, +): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.generate_optimized_manifest), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.GenerateOptimizedManifestResponse( + comments=["comments_value"], + manifest_version="manifest_version_value", + ) + ) + response = await client.generate_optimized_manifest(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) + _, args, _ = call.mock_calls[0] + request = gkerecommender.GenerateOptimizedManifestRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, gkerecommender.GenerateOptimizedManifestResponse) + assert response.comments == ["comments_value"] + assert response.manifest_version == "manifest_version_value" + + +@pytest.mark.asyncio +async def test_generate_optimized_manifest_async_from_dict(): + await test_generate_optimized_manifest_async(request_type=dict) + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchBenchmarkingDataRequest, + dict, + ], +) +def test_fetch_benchmarking_data(request_type, transport: str = "grpc"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_benchmarking_data), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = gkerecommender.FetchBenchmarkingDataResponse() + response = client.fetch_benchmarking_data(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) == 1 + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchBenchmarkingDataRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, gkerecommender.FetchBenchmarkingDataResponse) + + +def test_fetch_benchmarking_data_non_empty_request_with_auto_populated_field(): + # This test is a coverage failsafe to make sure that UUID4 fields are + # automatically populated, according to AIP-4235, with non-empty requests. + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Populate all string fields in the request which are not UUID4 + # since we want to check that UUID4 are populated automatically + # if they meet the requirements of AIP 4235. + request = gkerecommender.FetchBenchmarkingDataRequest( + instance_type="instance_type_value", + pricing_model="pricing_model_value", + ) + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_benchmarking_data), "__call__" + ) as call: + call.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client.fetch_benchmarking_data(request=request) + call.assert_called() + _, args, _ = call.mock_calls[0] + assert args[0] == gkerecommender.FetchBenchmarkingDataRequest( + instance_type="instance_type_value", + pricing_model="pricing_model_value", + ) + + +def test_fetch_benchmarking_data_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._transport.fetch_benchmarking_data + in client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[ + client._transport.fetch_benchmarking_data + ] = mock_rpc + request = {} + client.fetch_benchmarking_data(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_benchmarking_data(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_benchmarking_data_async_use_cached_wrapped_rpc( + transport: str = "grpc_asyncio", +): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._client._transport.fetch_benchmarking_data + in client._client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.AsyncMock() + mock_rpc.return_value = mock.Mock() + client._client._transport._wrapped_methods[ + client._client._transport.fetch_benchmarking_data + ] = mock_rpc + + request = {} + await client.fetch_benchmarking_data(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + await client.fetch_benchmarking_data(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +@pytest.mark.asyncio +async def test_fetch_benchmarking_data_async( + transport: str = "grpc_asyncio", + request_type=gkerecommender.FetchBenchmarkingDataRequest, +): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport=transport, + ) + + # Everything is optional in proto3 as far as the runtime is concerned, + # and we are mocking out the actual API, so just send an empty request. + request = request_type() + + # Mock the actual call within the gRPC stub, and fake the request. + with mock.patch.object( + type(client.transport.fetch_benchmarking_data), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchBenchmarkingDataResponse() + ) + response = await client.fetch_benchmarking_data(request) + + # Establish that the underlying gRPC stub method was called. + assert len(call.mock_calls) + _, args, _ = call.mock_calls[0] + request = gkerecommender.FetchBenchmarkingDataRequest() + assert args[0] == request + + # Establish that the response is the type that we expect. + assert isinstance(response, gkerecommender.FetchBenchmarkingDataResponse) + + +@pytest.mark.asyncio +async def test_fetch_benchmarking_data_async_from_dict(): + await test_fetch_benchmarking_data_async(request_type=dict) + + +def test_fetch_models_rest_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert client._transport.fetch_models in client._transport._wrapped_methods + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[client._transport.fetch_models] = mock_rpc + + request = {} + client.fetch_models(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_models(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +def test_fetch_models_rest_pager(transport: str = "rest"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(Session, "request") as req: + # TODO(kbandes): remove this mock unless there's a good reason for it. + # with mock.patch.object(path_template, 'transcode') as transcode: + # Set the response as a series of pages + response = ( + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelsResponse( + models=[], + next_page_token="def", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelsResponse( + models=[ + str(), + str(), + ], + ), + ) + # Two responses for two calls + response = response + response + + # Wrap the values into proper Response objs + response = tuple( + gkerecommender.FetchModelsResponse.to_json(x) for x in response + ) + return_values = tuple(Response() for i in response) + for return_val, response_val in zip(return_values, response): + return_val._content = response_val.encode("UTF-8") + return_val.status_code = 200 + req.side_effect = return_values + + sample_request = {} + + pager = client.fetch_models(request=sample_request) + + results = list(pager) + assert len(results) == 6 + assert all(isinstance(i, str) for i in results) + + pages = list(client.fetch_models(request=sample_request).pages) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +def test_fetch_model_servers_rest_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._transport.fetch_model_servers in client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[ + client._transport.fetch_model_servers + ] = mock_rpc + + request = {} + client.fetch_model_servers(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_model_servers(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +def test_fetch_model_servers_rest_required_fields( + request_type=gkerecommender.FetchModelServersRequest, +): + transport_class = transports.GkeInferenceQuickstartRestTransport + + request_init = {} + request_init["model"] = "" + request = request_type(**request_init) + pb_request = request_type.pb(request) + jsonified_request = json.loads( + json_format.MessageToJson(pb_request, use_integers_for_enums=False) + ) + + # verify fields with default values are dropped + assert "model" not in jsonified_request + + unset_fields = transport_class( + credentials=ga_credentials.AnonymousCredentials() + ).fetch_model_servers._get_unset_required_fields(jsonified_request) + jsonified_request.update(unset_fields) + + # verify required fields with default values are now present + assert "model" in jsonified_request + assert jsonified_request["model"] == request_init["model"] + + jsonified_request["model"] = "model_value" + + unset_fields = transport_class( + credentials=ga_credentials.AnonymousCredentials() + ).fetch_model_servers._get_unset_required_fields(jsonified_request) + # Check that path parameters and body parameters are not mixing in. + assert not set(unset_fields) - set( + ( + "model", + "page_size", + "page_token", + ) + ) + jsonified_request.update(unset_fields) + + # verify required fields with non-default values are left alone + assert "model" in jsonified_request + assert jsonified_request["model"] == "model_value" + + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + request = request_type(**request_init) + + # Designate an appropriate value for the returned response. + return_value = gkerecommender.FetchModelServersResponse() + # Mock the http request call within the method and fake a response. + with mock.patch.object(Session, "request") as req: + # We need to mock transcode() because providing default values + # for required fields will fail the real version if the http_options + # expect actual values for those fields. + with mock.patch.object(path_template, "transcode") as transcode: + # A uri without fields and an empty body will force all the + # request fields to show up in the query_params. + pb_request = request_type.pb(request) + transcode_result = { + "uri": "v1/sample_method", + "method": "get", + "query_params": pb_request, + } + transcode.return_value = transcode_result + + response_value = Response() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.FetchModelServersResponse.pb(return_value) + json_return_value = json_format.MessageToJson(return_value) + + response_value._content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + + response = client.fetch_model_servers(request) + + expected_params = [ + ( + "model", + "", + ), + ("$alt", "json;enum-encoding=int"), + ] + actual_params = req.call_args.kwargs["params"] + assert expected_params == actual_params + + +def test_fetch_model_servers_rest_unset_required_fields(): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials + ) + + unset_fields = transport.fetch_model_servers._get_unset_required_fields({}) + assert set(unset_fields) == ( + set( + ( + "model", + "pageSize", + "pageToken", + ) + ) + & set(("model",)) + ) + + +def test_fetch_model_servers_rest_pager(transport: str = "rest"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(Session, "request") as req: + # TODO(kbandes): remove this mock unless there's a good reason for it. + # with mock.patch.object(path_template, 'transcode') as transcode: + # Set the response as a series of pages + response = ( + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[], + next_page_token="def", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServersResponse( + model_servers=[ + str(), + str(), + ], + ), + ) + # Two responses for two calls + response = response + response + + # Wrap the values into proper Response objs + response = tuple( + gkerecommender.FetchModelServersResponse.to_json(x) for x in response + ) + return_values = tuple(Response() for i in response) + for return_val, response_val in zip(return_values, response): + return_val._content = response_val.encode("UTF-8") + return_val.status_code = 200 + req.side_effect = return_values + + sample_request = {} + + pager = client.fetch_model_servers(request=sample_request) + + results = list(pager) + assert len(results) == 6 + assert all(isinstance(i, str) for i in results) + + pages = list(client.fetch_model_servers(request=sample_request).pages) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +def test_fetch_model_server_versions_rest_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._transport.fetch_model_server_versions + in client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[ + client._transport.fetch_model_server_versions + ] = mock_rpc + + request = {} + client.fetch_model_server_versions(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_model_server_versions(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +def test_fetch_model_server_versions_rest_required_fields( + request_type=gkerecommender.FetchModelServerVersionsRequest, +): + transport_class = transports.GkeInferenceQuickstartRestTransport + + request_init = {} + request_init["model"] = "" + request_init["model_server"] = "" + request = request_type(**request_init) + pb_request = request_type.pb(request) + jsonified_request = json.loads( + json_format.MessageToJson(pb_request, use_integers_for_enums=False) + ) + + # verify fields with default values are dropped + assert "model" not in jsonified_request + assert "modelServer" not in jsonified_request + + unset_fields = transport_class( + credentials=ga_credentials.AnonymousCredentials() + ).fetch_model_server_versions._get_unset_required_fields(jsonified_request) + jsonified_request.update(unset_fields) + + # verify required fields with default values are now present + assert "model" in jsonified_request + assert jsonified_request["model"] == request_init["model"] + assert "modelServer" in jsonified_request + assert jsonified_request["modelServer"] == request_init["model_server"] + + jsonified_request["model"] = "model_value" + jsonified_request["modelServer"] = "model_server_value" + + unset_fields = transport_class( + credentials=ga_credentials.AnonymousCredentials() + ).fetch_model_server_versions._get_unset_required_fields(jsonified_request) + # Check that path parameters and body parameters are not mixing in. + assert not set(unset_fields) - set( + ( + "model", + "model_server", + "page_size", + "page_token", + ) + ) + jsonified_request.update(unset_fields) + + # verify required fields with non-default values are left alone + assert "model" in jsonified_request + assert jsonified_request["model"] == "model_value" + assert "modelServer" in jsonified_request + assert jsonified_request["modelServer"] == "model_server_value" + + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + request = request_type(**request_init) + + # Designate an appropriate value for the returned response. + return_value = gkerecommender.FetchModelServerVersionsResponse() + # Mock the http request call within the method and fake a response. + with mock.patch.object(Session, "request") as req: + # We need to mock transcode() because providing default values + # for required fields will fail the real version if the http_options + # expect actual values for those fields. + with mock.patch.object(path_template, "transcode") as transcode: + # A uri without fields and an empty body will force all the + # request fields to show up in the query_params. + pb_request = request_type.pb(request) + transcode_result = { + "uri": "v1/sample_method", + "method": "get", + "query_params": pb_request, + } + transcode.return_value = transcode_result + + response_value = Response() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.FetchModelServerVersionsResponse.pb( + return_value + ) + json_return_value = json_format.MessageToJson(return_value) + + response_value._content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + + response = client.fetch_model_server_versions(request) + + expected_params = [ + ( + "model", + "", + ), + ( + "modelServer", + "", + ), + ("$alt", "json;enum-encoding=int"), + ] + actual_params = req.call_args.kwargs["params"] + assert expected_params == actual_params + + +def test_fetch_model_server_versions_rest_unset_required_fields(): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials + ) + + unset_fields = transport.fetch_model_server_versions._get_unset_required_fields({}) + assert set(unset_fields) == ( + set( + ( + "model", + "modelServer", + "pageSize", + "pageToken", + ) + ) + & set( + ( + "model", + "modelServer", + ) + ) + ) + + +def test_fetch_model_server_versions_rest_pager(transport: str = "rest"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(Session, "request") as req: + # TODO(kbandes): remove this mock unless there's a good reason for it. + # with mock.patch.object(path_template, 'transcode') as transcode: + # Set the response as a series of pages + response = ( + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + str(), + ], + next_page_token="abc", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[], + next_page_token="def", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=[ + str(), + str(), + ], + ), + ) + # Two responses for two calls + response = response + response + + # Wrap the values into proper Response objs + response = tuple( + gkerecommender.FetchModelServerVersionsResponse.to_json(x) for x in response + ) + return_values = tuple(Response() for i in response) + for return_val, response_val in zip(return_values, response): + return_val._content = response_val.encode("UTF-8") + return_val.status_code = 200 + req.side_effect = return_values + + sample_request = {} + + pager = client.fetch_model_server_versions(request=sample_request) + + results = list(pager) + assert len(results) == 6 + assert all(isinstance(i, str) for i in results) + + pages = list(client.fetch_model_server_versions(request=sample_request).pages) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +def test_fetch_profiles_rest_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert client._transport.fetch_profiles in client._transport._wrapped_methods + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[client._transport.fetch_profiles] = mock_rpc + + request = {} + client.fetch_profiles(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_profiles(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +def test_fetch_profiles_rest_pager(transport: str = "rest"): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(Session, "request") as req: + # TODO(kbandes): remove this mock unless there's a good reason for it. + # with mock.patch.object(path_template, 'transcode') as transcode: + # Set the response as a series of pages + response = ( + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + next_page_token="abc", + ), + gkerecommender.FetchProfilesResponse( + profile=[], + next_page_token="def", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + ], + next_page_token="ghi", + ), + gkerecommender.FetchProfilesResponse( + profile=[ + gkerecommender.Profile(), + gkerecommender.Profile(), + ], + ), + ) + # Two responses for two calls + response = response + response + + # Wrap the values into proper Response objs + response = tuple( + gkerecommender.FetchProfilesResponse.to_json(x) for x in response + ) + return_values = tuple(Response() for i in response) + for return_val, response_val in zip(return_values, response): + return_val._content = response_val.encode("UTF-8") + return_val.status_code = 200 + req.side_effect = return_values + + sample_request = {} + + pager = client.fetch_profiles(request=sample_request) + + results = list(pager) + assert len(results) == 6 + assert all(isinstance(i, gkerecommender.Profile) for i in results) + + pages = list(client.fetch_profiles(request=sample_request).pages) + for page_, token in zip(pages, ["abc", "def", "ghi", ""]): + assert page_.raw_page.next_page_token == token + + +def test_generate_optimized_manifest_rest_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._transport.generate_optimized_manifest + in client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[ + client._transport.generate_optimized_manifest + ] = mock_rpc + + request = {} + client.generate_optimized_manifest(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.generate_optimized_manifest(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +def test_generate_optimized_manifest_rest_required_fields( + request_type=gkerecommender.GenerateOptimizedManifestRequest, +): + transport_class = transports.GkeInferenceQuickstartRestTransport + + request_init = {} + request_init["accelerator_type"] = "" + request = request_type(**request_init) + pb_request = request_type.pb(request) + jsonified_request = json.loads( + json_format.MessageToJson(pb_request, use_integers_for_enums=False) + ) + + # verify fields with default values are dropped + + unset_fields = transport_class( + credentials=ga_credentials.AnonymousCredentials() + ).generate_optimized_manifest._get_unset_required_fields(jsonified_request) + jsonified_request.update(unset_fields) + + # verify required fields with default values are now present + + jsonified_request["acceleratorType"] = "accelerator_type_value" + + unset_fields = transport_class( + credentials=ga_credentials.AnonymousCredentials() + ).generate_optimized_manifest._get_unset_required_fields(jsonified_request) + jsonified_request.update(unset_fields) + + # verify required fields with non-default values are left alone + assert "acceleratorType" in jsonified_request + assert jsonified_request["acceleratorType"] == "accelerator_type_value" + + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + request = request_type(**request_init) + + # Designate an appropriate value for the returned response. + return_value = gkerecommender.GenerateOptimizedManifestResponse() + # Mock the http request call within the method and fake a response. + with mock.patch.object(Session, "request") as req: + # We need to mock transcode() because providing default values + # for required fields will fail the real version if the http_options + # expect actual values for those fields. + with mock.patch.object(path_template, "transcode") as transcode: + # A uri without fields and an empty body will force all the + # request fields to show up in the query_params. + pb_request = request_type.pb(request) + transcode_result = { + "uri": "v1/sample_method", + "method": "post", + "query_params": pb_request, + } + transcode_result["body"] = pb_request + transcode.return_value = transcode_result + + response_value = Response() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.GenerateOptimizedManifestResponse.pb( + return_value + ) + json_return_value = json_format.MessageToJson(return_value) + + response_value._content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + + response = client.generate_optimized_manifest(request) + + expected_params = [("$alt", "json;enum-encoding=int")] + actual_params = req.call_args.kwargs["params"] + assert expected_params == actual_params + + +def test_generate_optimized_manifest_rest_unset_required_fields(): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials + ) + + unset_fields = transport.generate_optimized_manifest._get_unset_required_fields({}) + assert set(unset_fields) == ( + set(()) + & set( + ( + "modelServerInfo", + "acceleratorType", + ) + ) + ) + + +def test_fetch_benchmarking_data_rest_use_cached_wrapped_rpc(): + # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, + # instead of constructing them on each call + with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Should wrap all calls on client creation + assert wrapper_fn.call_count > 0 + wrapper_fn.reset_mock() + + # Ensure method has been cached + assert ( + client._transport.fetch_benchmarking_data + in client._transport._wrapped_methods + ) + + # Replace cached wrapped function with mock + mock_rpc = mock.Mock() + mock_rpc.return_value.name = ( + "foo" # operation_request.operation in compute client(s) expect a string. + ) + client._transport._wrapped_methods[ + client._transport.fetch_benchmarking_data + ] = mock_rpc + + request = {} + client.fetch_benchmarking_data(request) + + # Establish that the underlying gRPC stub method was called. + assert mock_rpc.call_count == 1 + + client.fetch_benchmarking_data(request) + + # Establish that a new wrapper was not created for this call + assert wrapper_fn.call_count == 0 + assert mock_rpc.call_count == 2 + + +def test_fetch_benchmarking_data_rest_required_fields( + request_type=gkerecommender.FetchBenchmarkingDataRequest, +): + transport_class = transports.GkeInferenceQuickstartRestTransport + + request_init = {} + request = request_type(**request_init) + pb_request = request_type.pb(request) + jsonified_request = json.loads( + json_format.MessageToJson(pb_request, use_integers_for_enums=False) + ) + + # verify fields with default values are dropped + + unset_fields = transport_class( + credentials=ga_credentials.AnonymousCredentials() + ).fetch_benchmarking_data._get_unset_required_fields(jsonified_request) + jsonified_request.update(unset_fields) + + # verify required fields with default values are now present + + unset_fields = transport_class( + credentials=ga_credentials.AnonymousCredentials() + ).fetch_benchmarking_data._get_unset_required_fields(jsonified_request) + jsonified_request.update(unset_fields) + + # verify required fields with non-default values are left alone + + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + request = request_type(**request_init) + + # Designate an appropriate value for the returned response. + return_value = gkerecommender.FetchBenchmarkingDataResponse() + # Mock the http request call within the method and fake a response. + with mock.patch.object(Session, "request") as req: + # We need to mock transcode() because providing default values + # for required fields will fail the real version if the http_options + # expect actual values for those fields. + with mock.patch.object(path_template, "transcode") as transcode: + # A uri without fields and an empty body will force all the + # request fields to show up in the query_params. + pb_request = request_type.pb(request) + transcode_result = { + "uri": "v1/sample_method", + "method": "post", + "query_params": pb_request, + } + transcode_result["body"] = pb_request + transcode.return_value = transcode_result + + response_value = Response() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.FetchBenchmarkingDataResponse.pb(return_value) + json_return_value = json_format.MessageToJson(return_value) + + response_value._content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + + response = client.fetch_benchmarking_data(request) + + expected_params = [("$alt", "json;enum-encoding=int")] + actual_params = req.call_args.kwargs["params"] + assert expected_params == actual_params + + +def test_fetch_benchmarking_data_rest_unset_required_fields(): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials + ) + + unset_fields = transport.fetch_benchmarking_data._get_unset_required_fields({}) + assert set(unset_fields) == (set(()) & set(("modelServerInfo",))) + + +def test_credentials_transport_error(): + # It is an error to provide credentials and a transport instance. + transport = transports.GkeInferenceQuickstartGrpcTransport( + credentials=ga_credentials.AnonymousCredentials(), + ) + with pytest.raises(ValueError): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport=transport, + ) + + # It is an error to provide a credentials file and a transport instance. + transport = transports.GkeInferenceQuickstartGrpcTransport( + credentials=ga_credentials.AnonymousCredentials(), + ) + with pytest.raises(ValueError): + client = GkeInferenceQuickstartClient( + client_options={"credentials_file": "credentials.json"}, + transport=transport, + ) + + # It is an error to provide an api_key and a transport instance. + transport = transports.GkeInferenceQuickstartGrpcTransport( + credentials=ga_credentials.AnonymousCredentials(), + ) + options = client_options.ClientOptions() + options.api_key = "api_key" + with pytest.raises(ValueError): + client = GkeInferenceQuickstartClient( + client_options=options, + transport=transport, + ) + + # It is an error to provide an api_key and a credential. + options = client_options.ClientOptions() + options.api_key = "api_key" + with pytest.raises(ValueError): + client = GkeInferenceQuickstartClient( + client_options=options, credentials=ga_credentials.AnonymousCredentials() + ) + + # It is an error to provide scopes and a transport instance. + transport = transports.GkeInferenceQuickstartGrpcTransport( + credentials=ga_credentials.AnonymousCredentials(), + ) + with pytest.raises(ValueError): + client = GkeInferenceQuickstartClient( + client_options={"scopes": ["1", "2"]}, + transport=transport, + ) + + +def test_transport_instance(): + # A client may be instantiated with a custom transport instance. + transport = transports.GkeInferenceQuickstartGrpcTransport( + credentials=ga_credentials.AnonymousCredentials(), + ) + client = GkeInferenceQuickstartClient(transport=transport) + assert client.transport is transport + + +def test_transport_get_channel(): + # A client may be instantiated with a custom transport instance. + transport = transports.GkeInferenceQuickstartGrpcTransport( + credentials=ga_credentials.AnonymousCredentials(), + ) + channel = transport.grpc_channel + assert channel + + transport = transports.GkeInferenceQuickstartGrpcAsyncIOTransport( + credentials=ga_credentials.AnonymousCredentials(), + ) + channel = transport.grpc_channel + assert channel + + +@pytest.mark.parametrize( + "transport_class", + [ + transports.GkeInferenceQuickstartGrpcTransport, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + transports.GkeInferenceQuickstartRestTransport, + ], +) +def test_transport_adc(transport_class): + # Test default credentials are used if not provided. + with mock.patch.object(google.auth, "default") as adc: + adc.return_value = (ga_credentials.AnonymousCredentials(), None) + transport_class() + adc.assert_called_once() + + +def test_transport_kind_grpc(): + transport = GkeInferenceQuickstartClient.get_transport_class("grpc")( + credentials=ga_credentials.AnonymousCredentials() + ) + assert transport.kind == "grpc" + + +def test_initialize_client_w_grpc(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="grpc" + ) + assert client is not None + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_models_empty_call_grpc(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: + call.return_value = gkerecommender.FetchModelsResponse() + client.fetch_models(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelsRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_model_servers_empty_call_grpc(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), "__call__" + ) as call: + call.return_value = gkerecommender.FetchModelServersResponse() + client.fetch_model_servers(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelServersRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_model_server_versions_empty_call_grpc(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), "__call__" + ) as call: + call.return_value = gkerecommender.FetchModelServerVersionsResponse() + client.fetch_model_server_versions(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelServerVersionsRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_profiles_empty_call_grpc(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: + call.return_value = gkerecommender.FetchProfilesResponse() + client.fetch_profiles(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchProfilesRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_generate_optimized_manifest_empty_call_grpc(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.generate_optimized_manifest), "__call__" + ) as call: + call.return_value = gkerecommender.GenerateOptimizedManifestResponse() + client.generate_optimized_manifest(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.GenerateOptimizedManifestRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_benchmarking_data_empty_call_grpc(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="grpc", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_benchmarking_data), "__call__" + ) as call: + call.return_value = gkerecommender.FetchBenchmarkingDataResponse() + client.fetch_benchmarking_data(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchBenchmarkingDataRequest() + + assert args[0] == request_msg + + +def test_transport_kind_grpc_asyncio(): + transport = GkeInferenceQuickstartAsyncClient.get_transport_class("grpc_asyncio")( + credentials=async_anonymous_credentials() + ) + assert transport.kind == "grpc_asyncio" + + +def test_initialize_client_w_grpc_asyncio(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), transport="grpc_asyncio" + ) + assert client is not None + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +@pytest.mark.asyncio +async def test_fetch_models_empty_call_grpc_asyncio(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport="grpc_asyncio", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchModelsResponse( + models=["models_value"], + next_page_token="next_page_token_value", + ) + ) + await client.fetch_models(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelsRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +@pytest.mark.asyncio +async def test_fetch_model_servers_empty_call_grpc_asyncio(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport="grpc_asyncio", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchModelServersResponse( + model_servers=["model_servers_value"], + next_page_token="next_page_token_value", + ) + ) + await client.fetch_model_servers(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelServersRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +@pytest.mark.asyncio +async def test_fetch_model_server_versions_empty_call_grpc_asyncio(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport="grpc_asyncio", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=["model_server_versions_value"], + next_page_token="next_page_token_value", + ) + ) + await client.fetch_model_server_versions(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelServerVersionsRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +@pytest.mark.asyncio +async def test_fetch_profiles_empty_call_grpc_asyncio(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport="grpc_asyncio", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchProfilesResponse( + comments="comments_value", + next_page_token="next_page_token_value", + ) + ) + await client.fetch_profiles(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchProfilesRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +@pytest.mark.asyncio +async def test_generate_optimized_manifest_empty_call_grpc_asyncio(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport="grpc_asyncio", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.generate_optimized_manifest), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.GenerateOptimizedManifestResponse( + comments=["comments_value"], + manifest_version="manifest_version_value", + ) + ) + await client.generate_optimized_manifest(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.GenerateOptimizedManifestRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +@pytest.mark.asyncio +async def test_fetch_benchmarking_data_empty_call_grpc_asyncio(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), + transport="grpc_asyncio", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_benchmarking_data), "__call__" + ) as call: + # Designate an appropriate return value for the call. + call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( + gkerecommender.FetchBenchmarkingDataResponse() + ) + await client.fetch_benchmarking_data(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchBenchmarkingDataRequest() + + assert args[0] == request_msg + + +def test_transport_kind_rest(): + transport = GkeInferenceQuickstartClient.get_transport_class("rest")( + credentials=ga_credentials.AnonymousCredentials() + ) + assert transport.kind == "rest" + + +def test_fetch_models_rest_bad_request(request_type=gkerecommender.FetchModelsRequest): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a BadRequest error. + with mock.patch.object(Session, "request") as req, pytest.raises( + core_exceptions.BadRequest + ): + # Wrap the value into a proper Response obj + response_value = mock.Mock() + json_return_value = "" + response_value.json = mock.Mock(return_value={}) + response_value.status_code = 400 + response_value.request = mock.Mock() + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + client.fetch_models(request) + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchModelsRequest, + dict, + ], +) +def test_fetch_models_rest_call_success(request_type): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(type(client.transport._session), "request") as req: + # Designate an appropriate value for the returned response. + return_value = gkerecommender.FetchModelsResponse( + models=["models_value"], + next_page_token="next_page_token_value", + ) + + # Wrap the value into a proper Response obj + response_value = mock.Mock() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.FetchModelsResponse.pb(return_value) + json_return_value = json_format.MessageToJson(return_value) + response_value.content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + response = client.fetch_models(request) + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelsPager) + assert response.models == ["models_value"] + assert response.next_page_token == "next_page_token_value" + + +@pytest.mark.parametrize("null_interceptor", [True, False]) +def test_fetch_models_rest_interceptors(null_interceptor): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials(), + interceptor=None + if null_interceptor + else transports.GkeInferenceQuickstartRestInterceptor(), + ) + client = GkeInferenceQuickstartClient(transport=transport) + + with mock.patch.object( + type(client.transport._session), "request" + ) as req, mock.patch.object( + path_template, "transcode" + ) as transcode, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, "post_fetch_models" + ) as post, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "post_fetch_models_with_metadata", + ) as post_with_metadata, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, "pre_fetch_models" + ) as pre: + pre.assert_not_called() + post.assert_not_called() + post_with_metadata.assert_not_called() + pb_message = gkerecommender.FetchModelsRequest.pb( + gkerecommender.FetchModelsRequest() + ) + transcode.return_value = { + "method": "post", + "uri": "my_uri", + "body": pb_message, + "query_params": pb_message, + } + + req.return_value = mock.Mock() + req.return_value.status_code = 200 + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + return_value = gkerecommender.FetchModelsResponse.to_json( + gkerecommender.FetchModelsResponse() + ) + req.return_value.content = return_value + + request = gkerecommender.FetchModelsRequest() + metadata = [ + ("key", "val"), + ("cephalopod", "squid"), + ] + pre.return_value = request, metadata + post.return_value = gkerecommender.FetchModelsResponse() + post_with_metadata.return_value = gkerecommender.FetchModelsResponse(), metadata + + client.fetch_models( + request, + metadata=[ + ("key", "val"), + ("cephalopod", "squid"), + ], + ) + + pre.assert_called_once() + post.assert_called_once() + post_with_metadata.assert_called_once() + + +def test_fetch_model_servers_rest_bad_request( + request_type=gkerecommender.FetchModelServersRequest, +): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a BadRequest error. + with mock.patch.object(Session, "request") as req, pytest.raises( + core_exceptions.BadRequest + ): + # Wrap the value into a proper Response obj + response_value = mock.Mock() + json_return_value = "" + response_value.json = mock.Mock(return_value={}) + response_value.status_code = 400 + response_value.request = mock.Mock() + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + client.fetch_model_servers(request) + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchModelServersRequest, + dict, + ], +) +def test_fetch_model_servers_rest_call_success(request_type): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(type(client.transport._session), "request") as req: + # Designate an appropriate value for the returned response. + return_value = gkerecommender.FetchModelServersResponse( + model_servers=["model_servers_value"], + next_page_token="next_page_token_value", + ) + + # Wrap the value into a proper Response obj + response_value = mock.Mock() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.FetchModelServersResponse.pb(return_value) + json_return_value = json_format.MessageToJson(return_value) + response_value.content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + response = client.fetch_model_servers(request) + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelServersPager) + assert response.model_servers == ["model_servers_value"] + assert response.next_page_token == "next_page_token_value" + + +@pytest.mark.parametrize("null_interceptor", [True, False]) +def test_fetch_model_servers_rest_interceptors(null_interceptor): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials(), + interceptor=None + if null_interceptor + else transports.GkeInferenceQuickstartRestInterceptor(), + ) + client = GkeInferenceQuickstartClient(transport=transport) + + with mock.patch.object( + type(client.transport._session), "request" + ) as req, mock.patch.object( + path_template, "transcode" + ) as transcode, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, "post_fetch_model_servers" + ) as post, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "post_fetch_model_servers_with_metadata", + ) as post_with_metadata, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, "pre_fetch_model_servers" + ) as pre: + pre.assert_not_called() + post.assert_not_called() + post_with_metadata.assert_not_called() + pb_message = gkerecommender.FetchModelServersRequest.pb( + gkerecommender.FetchModelServersRequest() + ) + transcode.return_value = { + "method": "post", + "uri": "my_uri", + "body": pb_message, + "query_params": pb_message, + } + + req.return_value = mock.Mock() + req.return_value.status_code = 200 + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + return_value = gkerecommender.FetchModelServersResponse.to_json( + gkerecommender.FetchModelServersResponse() + ) + req.return_value.content = return_value + + request = gkerecommender.FetchModelServersRequest() + metadata = [ + ("key", "val"), + ("cephalopod", "squid"), + ] + pre.return_value = request, metadata + post.return_value = gkerecommender.FetchModelServersResponse() + post_with_metadata.return_value = ( + gkerecommender.FetchModelServersResponse(), + metadata, + ) + + client.fetch_model_servers( + request, + metadata=[ + ("key", "val"), + ("cephalopod", "squid"), + ], + ) + + pre.assert_called_once() + post.assert_called_once() + post_with_metadata.assert_called_once() + + +def test_fetch_model_server_versions_rest_bad_request( + request_type=gkerecommender.FetchModelServerVersionsRequest, +): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a BadRequest error. + with mock.patch.object(Session, "request") as req, pytest.raises( + core_exceptions.BadRequest + ): + # Wrap the value into a proper Response obj + response_value = mock.Mock() + json_return_value = "" + response_value.json = mock.Mock(return_value={}) + response_value.status_code = 400 + response_value.request = mock.Mock() + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + client.fetch_model_server_versions(request) + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchModelServerVersionsRequest, + dict, + ], +) +def test_fetch_model_server_versions_rest_call_success(request_type): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(type(client.transport._session), "request") as req: + # Designate an appropriate value for the returned response. + return_value = gkerecommender.FetchModelServerVersionsResponse( + model_server_versions=["model_server_versions_value"], + next_page_token="next_page_token_value", + ) + + # Wrap the value into a proper Response obj + response_value = mock.Mock() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.FetchModelServerVersionsResponse.pb(return_value) + json_return_value = json_format.MessageToJson(return_value) + response_value.content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + response = client.fetch_model_server_versions(request) + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchModelServerVersionsPager) + assert response.model_server_versions == ["model_server_versions_value"] + assert response.next_page_token == "next_page_token_value" + + +@pytest.mark.parametrize("null_interceptor", [True, False]) +def test_fetch_model_server_versions_rest_interceptors(null_interceptor): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials(), + interceptor=None + if null_interceptor + else transports.GkeInferenceQuickstartRestInterceptor(), + ) + client = GkeInferenceQuickstartClient(transport=transport) + + with mock.patch.object( + type(client.transport._session), "request" + ) as req, mock.patch.object( + path_template, "transcode" + ) as transcode, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "post_fetch_model_server_versions", + ) as post, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "post_fetch_model_server_versions_with_metadata", + ) as post_with_metadata, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "pre_fetch_model_server_versions", + ) as pre: + pre.assert_not_called() + post.assert_not_called() + post_with_metadata.assert_not_called() + pb_message = gkerecommender.FetchModelServerVersionsRequest.pb( + gkerecommender.FetchModelServerVersionsRequest() + ) + transcode.return_value = { + "method": "post", + "uri": "my_uri", + "body": pb_message, + "query_params": pb_message, + } + + req.return_value = mock.Mock() + req.return_value.status_code = 200 + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + return_value = gkerecommender.FetchModelServerVersionsResponse.to_json( + gkerecommender.FetchModelServerVersionsResponse() + ) + req.return_value.content = return_value + + request = gkerecommender.FetchModelServerVersionsRequest() + metadata = [ + ("key", "val"), + ("cephalopod", "squid"), + ] + pre.return_value = request, metadata + post.return_value = gkerecommender.FetchModelServerVersionsResponse() + post_with_metadata.return_value = ( + gkerecommender.FetchModelServerVersionsResponse(), + metadata, + ) + + client.fetch_model_server_versions( + request, + metadata=[ + ("key", "val"), + ("cephalopod", "squid"), + ], + ) + + pre.assert_called_once() + post.assert_called_once() + post_with_metadata.assert_called_once() + + +def test_fetch_profiles_rest_bad_request( + request_type=gkerecommender.FetchProfilesRequest, +): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a BadRequest error. + with mock.patch.object(Session, "request") as req, pytest.raises( + core_exceptions.BadRequest + ): + # Wrap the value into a proper Response obj + response_value = mock.Mock() + json_return_value = "" + response_value.json = mock.Mock(return_value={}) + response_value.status_code = 400 + response_value.request = mock.Mock() + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + client.fetch_profiles(request) + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchProfilesRequest, + dict, + ], +) +def test_fetch_profiles_rest_call_success(request_type): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(type(client.transport._session), "request") as req: + # Designate an appropriate value for the returned response. + return_value = gkerecommender.FetchProfilesResponse( + comments="comments_value", + next_page_token="next_page_token_value", + ) + + # Wrap the value into a proper Response obj + response_value = mock.Mock() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.FetchProfilesResponse.pb(return_value) + json_return_value = json_format.MessageToJson(return_value) + response_value.content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + response = client.fetch_profiles(request) + + # Establish that the response is the type that we expect. + assert isinstance(response, pagers.FetchProfilesPager) + assert response.comments == "comments_value" + assert response.next_page_token == "next_page_token_value" + + +@pytest.mark.parametrize("null_interceptor", [True, False]) +def test_fetch_profiles_rest_interceptors(null_interceptor): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials(), + interceptor=None + if null_interceptor + else transports.GkeInferenceQuickstartRestInterceptor(), + ) + client = GkeInferenceQuickstartClient(transport=transport) + + with mock.patch.object( + type(client.transport._session), "request" + ) as req, mock.patch.object( + path_template, "transcode" + ) as transcode, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, "post_fetch_profiles" + ) as post, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "post_fetch_profiles_with_metadata", + ) as post_with_metadata, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, "pre_fetch_profiles" + ) as pre: + pre.assert_not_called() + post.assert_not_called() + post_with_metadata.assert_not_called() + pb_message = gkerecommender.FetchProfilesRequest.pb( + gkerecommender.FetchProfilesRequest() + ) + transcode.return_value = { + "method": "post", + "uri": "my_uri", + "body": pb_message, + "query_params": pb_message, + } + + req.return_value = mock.Mock() + req.return_value.status_code = 200 + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + return_value = gkerecommender.FetchProfilesResponse.to_json( + gkerecommender.FetchProfilesResponse() + ) + req.return_value.content = return_value + + request = gkerecommender.FetchProfilesRequest() + metadata = [ + ("key", "val"), + ("cephalopod", "squid"), + ] + pre.return_value = request, metadata + post.return_value = gkerecommender.FetchProfilesResponse() + post_with_metadata.return_value = ( + gkerecommender.FetchProfilesResponse(), + metadata, + ) + + client.fetch_profiles( + request, + metadata=[ + ("key", "val"), + ("cephalopod", "squid"), + ], + ) + + pre.assert_called_once() + post.assert_called_once() + post_with_metadata.assert_called_once() + + +def test_generate_optimized_manifest_rest_bad_request( + request_type=gkerecommender.GenerateOptimizedManifestRequest, +): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a BadRequest error. + with mock.patch.object(Session, "request") as req, pytest.raises( + core_exceptions.BadRequest + ): + # Wrap the value into a proper Response obj + response_value = mock.Mock() + json_return_value = "" + response_value.json = mock.Mock(return_value={}) + response_value.status_code = 400 + response_value.request = mock.Mock() + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + client.generate_optimized_manifest(request) + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.GenerateOptimizedManifestRequest, + dict, + ], +) +def test_generate_optimized_manifest_rest_call_success(request_type): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(type(client.transport._session), "request") as req: + # Designate an appropriate value for the returned response. + return_value = gkerecommender.GenerateOptimizedManifestResponse( + comments=["comments_value"], + manifest_version="manifest_version_value", + ) + + # Wrap the value into a proper Response obj + response_value = mock.Mock() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.GenerateOptimizedManifestResponse.pb(return_value) + json_return_value = json_format.MessageToJson(return_value) + response_value.content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + response = client.generate_optimized_manifest(request) + + # Establish that the response is the type that we expect. + assert isinstance(response, gkerecommender.GenerateOptimizedManifestResponse) + assert response.comments == ["comments_value"] + assert response.manifest_version == "manifest_version_value" + + +@pytest.mark.parametrize("null_interceptor", [True, False]) +def test_generate_optimized_manifest_rest_interceptors(null_interceptor): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials(), + interceptor=None + if null_interceptor + else transports.GkeInferenceQuickstartRestInterceptor(), + ) + client = GkeInferenceQuickstartClient(transport=transport) + + with mock.patch.object( + type(client.transport._session), "request" + ) as req, mock.patch.object( + path_template, "transcode" + ) as transcode, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "post_generate_optimized_manifest", + ) as post, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "post_generate_optimized_manifest_with_metadata", + ) as post_with_metadata, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "pre_generate_optimized_manifest", + ) as pre: + pre.assert_not_called() + post.assert_not_called() + post_with_metadata.assert_not_called() + pb_message = gkerecommender.GenerateOptimizedManifestRequest.pb( + gkerecommender.GenerateOptimizedManifestRequest() + ) + transcode.return_value = { + "method": "post", + "uri": "my_uri", + "body": pb_message, + "query_params": pb_message, + } + + req.return_value = mock.Mock() + req.return_value.status_code = 200 + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + return_value = gkerecommender.GenerateOptimizedManifestResponse.to_json( + gkerecommender.GenerateOptimizedManifestResponse() + ) + req.return_value.content = return_value + + request = gkerecommender.GenerateOptimizedManifestRequest() + metadata = [ + ("key", "val"), + ("cephalopod", "squid"), + ] + pre.return_value = request, metadata + post.return_value = gkerecommender.GenerateOptimizedManifestResponse() + post_with_metadata.return_value = ( + gkerecommender.GenerateOptimizedManifestResponse(), + metadata, + ) + + client.generate_optimized_manifest( + request, + metadata=[ + ("key", "val"), + ("cephalopod", "squid"), + ], + ) + + pre.assert_called_once() + post.assert_called_once() + post_with_metadata.assert_called_once() + + +def test_fetch_benchmarking_data_rest_bad_request( + request_type=gkerecommender.FetchBenchmarkingDataRequest, +): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a BadRequest error. + with mock.patch.object(Session, "request") as req, pytest.raises( + core_exceptions.BadRequest + ): + # Wrap the value into a proper Response obj + response_value = mock.Mock() + json_return_value = "" + response_value.json = mock.Mock(return_value={}) + response_value.status_code = 400 + response_value.request = mock.Mock() + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + client.fetch_benchmarking_data(request) + + +@pytest.mark.parametrize( + "request_type", + [ + gkerecommender.FetchBenchmarkingDataRequest, + dict, + ], +) +def test_fetch_benchmarking_data_rest_call_success(request_type): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + + # send a request that will satisfy transcoding + request_init = {} + request = request_type(**request_init) + + # Mock the http request call within the method and fake a response. + with mock.patch.object(type(client.transport._session), "request") as req: + # Designate an appropriate value for the returned response. + return_value = gkerecommender.FetchBenchmarkingDataResponse() + + # Wrap the value into a proper Response obj + response_value = mock.Mock() + response_value.status_code = 200 + + # Convert return value to protobuf type + return_value = gkerecommender.FetchBenchmarkingDataResponse.pb(return_value) + json_return_value = json_format.MessageToJson(return_value) + response_value.content = json_return_value.encode("UTF-8") + req.return_value = response_value + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + response = client.fetch_benchmarking_data(request) + + # Establish that the response is the type that we expect. + assert isinstance(response, gkerecommender.FetchBenchmarkingDataResponse) + + +@pytest.mark.parametrize("null_interceptor", [True, False]) +def test_fetch_benchmarking_data_rest_interceptors(null_interceptor): + transport = transports.GkeInferenceQuickstartRestTransport( + credentials=ga_credentials.AnonymousCredentials(), + interceptor=None + if null_interceptor + else transports.GkeInferenceQuickstartRestInterceptor(), + ) + client = GkeInferenceQuickstartClient(transport=transport) + + with mock.patch.object( + type(client.transport._session), "request" + ) as req, mock.patch.object( + path_template, "transcode" + ) as transcode, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, "post_fetch_benchmarking_data" + ) as post, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, + "post_fetch_benchmarking_data_with_metadata", + ) as post_with_metadata, mock.patch.object( + transports.GkeInferenceQuickstartRestInterceptor, "pre_fetch_benchmarking_data" + ) as pre: + pre.assert_not_called() + post.assert_not_called() + post_with_metadata.assert_not_called() + pb_message = gkerecommender.FetchBenchmarkingDataRequest.pb( + gkerecommender.FetchBenchmarkingDataRequest() + ) + transcode.return_value = { + "method": "post", + "uri": "my_uri", + "body": pb_message, + "query_params": pb_message, + } + + req.return_value = mock.Mock() + req.return_value.status_code = 200 + req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} + return_value = gkerecommender.FetchBenchmarkingDataResponse.to_json( + gkerecommender.FetchBenchmarkingDataResponse() + ) + req.return_value.content = return_value + + request = gkerecommender.FetchBenchmarkingDataRequest() + metadata = [ + ("key", "val"), + ("cephalopod", "squid"), + ] + pre.return_value = request, metadata + post.return_value = gkerecommender.FetchBenchmarkingDataResponse() + post_with_metadata.return_value = ( + gkerecommender.FetchBenchmarkingDataResponse(), + metadata, + ) + + client.fetch_benchmarking_data( + request, + metadata=[ + ("key", "val"), + ("cephalopod", "squid"), + ], + ) + + pre.assert_called_once() + post.assert_called_once() + post_with_metadata.assert_called_once() + + +def test_initialize_client_w_rest(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + assert client is not None + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_models_empty_call_rest(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: + client.fetch_models(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelsRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_model_servers_empty_call_rest(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_servers), "__call__" + ) as call: + client.fetch_model_servers(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelServersRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_model_server_versions_empty_call_rest(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_model_server_versions), "__call__" + ) as call: + client.fetch_model_server_versions(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchModelServerVersionsRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_profiles_empty_call_rest(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: + client.fetch_profiles(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchProfilesRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_generate_optimized_manifest_empty_call_rest(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.generate_optimized_manifest), "__call__" + ) as call: + client.generate_optimized_manifest(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.GenerateOptimizedManifestRequest() + + assert args[0] == request_msg + + +# This test is a coverage failsafe to make sure that totally empty calls, +# i.e. request == None and no flattened fields passed, work. +def test_fetch_benchmarking_data_empty_call_rest(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + transport="rest", + ) + + # Mock the actual call, and fake the request. + with mock.patch.object( + type(client.transport.fetch_benchmarking_data), "__call__" + ) as call: + client.fetch_benchmarking_data(request=None) + + # Establish that the underlying stub method was called. + call.assert_called() + _, args, _ = call.mock_calls[0] + request_msg = gkerecommender.FetchBenchmarkingDataRequest() + + assert args[0] == request_msg + + +def test_transport_grpc_default(): + # A client should use the gRPC transport by default. + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + ) + assert isinstance( + client.transport, + transports.GkeInferenceQuickstartGrpcTransport, + ) + + +def test_gke_inference_quickstart_base_transport_error(): + # Passing both a credentials object and credentials_file should raise an error + with pytest.raises(core_exceptions.DuplicateCredentialArgs): + transport = transports.GkeInferenceQuickstartTransport( + credentials=ga_credentials.AnonymousCredentials(), + credentials_file="credentials.json", + ) + + +def test_gke_inference_quickstart_base_transport(): + # Instantiate the base transport. + with mock.patch( + "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.GkeInferenceQuickstartTransport.__init__" + ) as Transport: + Transport.return_value = None + transport = transports.GkeInferenceQuickstartTransport( + credentials=ga_credentials.AnonymousCredentials(), + ) + + # Every method on the transport should just blindly + # raise NotImplementedError. + methods = ( + "fetch_models", + "fetch_model_servers", + "fetch_model_server_versions", + "fetch_profiles", + "generate_optimized_manifest", + "fetch_benchmarking_data", + ) + for method in methods: + with pytest.raises(NotImplementedError): + getattr(transport, method)(request=object()) + + with pytest.raises(NotImplementedError): + transport.close() + + # Catch all for all remaining methods and properties + remainder = [ + "kind", + ] + for r in remainder: + with pytest.raises(NotImplementedError): + getattr(transport, r)() + + +def test_gke_inference_quickstart_base_transport_with_credentials_file(): + # Instantiate the base transport with a credentials file + with mock.patch.object( + google.auth, "load_credentials_from_file", autospec=True + ) as load_creds, mock.patch( + "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.GkeInferenceQuickstartTransport._prep_wrapped_messages" + ) as Transport: + Transport.return_value = None + load_creds.return_value = (ga_credentials.AnonymousCredentials(), None) + transport = transports.GkeInferenceQuickstartTransport( + credentials_file="credentials.json", + quota_project_id="octopus", + ) + load_creds.assert_called_once_with( + "credentials.json", + scopes=None, + default_scopes=("https://www.googleapis.com/auth/cloud-platform",), + quota_project_id="octopus", + ) + + +def test_gke_inference_quickstart_base_transport_with_adc(): + # Test the default credentials are used if credentials and credentials_file are None. + with mock.patch.object(google.auth, "default", autospec=True) as adc, mock.patch( + "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.GkeInferenceQuickstartTransport._prep_wrapped_messages" + ) as Transport: + Transport.return_value = None + adc.return_value = (ga_credentials.AnonymousCredentials(), None) + transport = transports.GkeInferenceQuickstartTransport() + adc.assert_called_once() + + +def test_gke_inference_quickstart_auth_adc(): + # If no credentials are provided, we should use ADC credentials. + with mock.patch.object(google.auth, "default", autospec=True) as adc: + adc.return_value = (ga_credentials.AnonymousCredentials(), None) + GkeInferenceQuickstartClient() + adc.assert_called_once_with( + scopes=None, + default_scopes=("https://www.googleapis.com/auth/cloud-platform",), + quota_project_id=None, + ) + + +@pytest.mark.parametrize( + "transport_class", + [ + transports.GkeInferenceQuickstartGrpcTransport, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + ], +) +def test_gke_inference_quickstart_transport_auth_adc(transport_class): + # If credentials and host are not provided, the transport class should use + # ADC credentials. + with mock.patch.object(google.auth, "default", autospec=True) as adc: + adc.return_value = (ga_credentials.AnonymousCredentials(), None) + transport_class(quota_project_id="octopus", scopes=["1", "2"]) + adc.assert_called_once_with( + scopes=["1", "2"], + default_scopes=("https://www.googleapis.com/auth/cloud-platform",), + quota_project_id="octopus", + ) + + +@pytest.mark.parametrize( + "transport_class", + [ + transports.GkeInferenceQuickstartGrpcTransport, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + transports.GkeInferenceQuickstartRestTransport, + ], +) +def test_gke_inference_quickstart_transport_auth_gdch_credentials(transport_class): + host = "https://language.com" + api_audience_tests = [None, "https://language2.com"] + api_audience_expect = [host, "https://language2.com"] + for t, e in zip(api_audience_tests, api_audience_expect): + with mock.patch.object(google.auth, "default", autospec=True) as adc: + gdch_mock = mock.MagicMock() + type(gdch_mock).with_gdch_audience = mock.PropertyMock( + return_value=gdch_mock + ) + adc.return_value = (gdch_mock, None) + transport_class(host=host, api_audience=t) + gdch_mock.with_gdch_audience.assert_called_once_with(e) + + +@pytest.mark.parametrize( + "transport_class,grpc_helpers", + [ + (transports.GkeInferenceQuickstartGrpcTransport, grpc_helpers), + (transports.GkeInferenceQuickstartGrpcAsyncIOTransport, grpc_helpers_async), + ], +) +def test_gke_inference_quickstart_transport_create_channel( + transport_class, grpc_helpers +): + # If credentials and host are not provided, the transport class should use + # ADC credentials. + with mock.patch.object( + google.auth, "default", autospec=True + ) as adc, mock.patch.object( + grpc_helpers, "create_channel", autospec=True + ) as create_channel: + creds = ga_credentials.AnonymousCredentials() + adc.return_value = (creds, None) + transport_class(quota_project_id="octopus", scopes=["1", "2"]) + + create_channel.assert_called_with( + "gkerecommender.googleapis.com:443", + credentials=creds, + credentials_file=None, + quota_project_id="octopus", + default_scopes=("https://www.googleapis.com/auth/cloud-platform",), + scopes=["1", "2"], + default_host="gkerecommender.googleapis.com", + ssl_credentials=None, + options=[ + ("grpc.max_send_message_length", -1), + ("grpc.max_receive_message_length", -1), + ], + ) + + +@pytest.mark.parametrize( + "transport_class", + [ + transports.GkeInferenceQuickstartGrpcTransport, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + ], +) +def test_gke_inference_quickstart_grpc_transport_client_cert_source_for_mtls( + transport_class, +): + cred = ga_credentials.AnonymousCredentials() + + # Check ssl_channel_credentials is used if provided. + with mock.patch.object(transport_class, "create_channel") as mock_create_channel: + mock_ssl_channel_creds = mock.Mock() + transport_class( + host="squid.clam.whelk", + credentials=cred, + ssl_channel_credentials=mock_ssl_channel_creds, + ) + mock_create_channel.assert_called_once_with( + "squid.clam.whelk:443", + credentials=cred, + credentials_file=None, + scopes=None, + ssl_credentials=mock_ssl_channel_creds, + quota_project_id=None, + options=[ + ("grpc.max_send_message_length", -1), + ("grpc.max_receive_message_length", -1), + ], + ) + + # Check if ssl_channel_credentials is not provided, then client_cert_source_for_mtls + # is used. + with mock.patch.object(transport_class, "create_channel", return_value=mock.Mock()): + with mock.patch("grpc.ssl_channel_credentials") as mock_ssl_cred: + transport_class( + credentials=cred, + client_cert_source_for_mtls=client_cert_source_callback, + ) + expected_cert, expected_key = client_cert_source_callback() + mock_ssl_cred.assert_called_once_with( + certificate_chain=expected_cert, private_key=expected_key + ) + + +def test_gke_inference_quickstart_http_transport_client_cert_source_for_mtls(): + cred = ga_credentials.AnonymousCredentials() + with mock.patch( + "google.auth.transport.requests.AuthorizedSession.configure_mtls_channel" + ) as mock_configure_mtls_channel: + transports.GkeInferenceQuickstartRestTransport( + credentials=cred, client_cert_source_for_mtls=client_cert_source_callback + ) + mock_configure_mtls_channel.assert_called_once_with(client_cert_source_callback) + + +@pytest.mark.parametrize( + "transport_name", + [ + "grpc", + "grpc_asyncio", + "rest", + ], +) +def test_gke_inference_quickstart_host_no_port(transport_name): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + client_options=client_options.ClientOptions( + api_endpoint="gkerecommender.googleapis.com" + ), + transport=transport_name, + ) + assert client.transport._host == ( + "gkerecommender.googleapis.com:443" + if transport_name in ["grpc", "grpc_asyncio"] + else "https://gkerecommender.googleapis.com" + ) + + +@pytest.mark.parametrize( + "transport_name", + [ + "grpc", + "grpc_asyncio", + "rest", + ], +) +def test_gke_inference_quickstart_host_with_port(transport_name): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + client_options=client_options.ClientOptions( + api_endpoint="gkerecommender.googleapis.com:8000" + ), + transport=transport_name, + ) + assert client.transport._host == ( + "gkerecommender.googleapis.com:8000" + if transport_name in ["grpc", "grpc_asyncio"] + else "https://gkerecommender.googleapis.com:8000" + ) + + +@pytest.mark.parametrize( + "transport_name", + [ + "rest", + ], +) +def test_gke_inference_quickstart_client_transport_session_collision(transport_name): + creds1 = ga_credentials.AnonymousCredentials() + creds2 = ga_credentials.AnonymousCredentials() + client1 = GkeInferenceQuickstartClient( + credentials=creds1, + transport=transport_name, + ) + client2 = GkeInferenceQuickstartClient( + credentials=creds2, + transport=transport_name, + ) + session1 = client1.transport.fetch_models._session + session2 = client2.transport.fetch_models._session + assert session1 != session2 + session1 = client1.transport.fetch_model_servers._session + session2 = client2.transport.fetch_model_servers._session + assert session1 != session2 + session1 = client1.transport.fetch_model_server_versions._session + session2 = client2.transport.fetch_model_server_versions._session + assert session1 != session2 + session1 = client1.transport.fetch_profiles._session + session2 = client2.transport.fetch_profiles._session + assert session1 != session2 + session1 = client1.transport.generate_optimized_manifest._session + session2 = client2.transport.generate_optimized_manifest._session + assert session1 != session2 + session1 = client1.transport.fetch_benchmarking_data._session + session2 = client2.transport.fetch_benchmarking_data._session + assert session1 != session2 + + +def test_gke_inference_quickstart_grpc_transport_channel(): + channel = grpc.secure_channel("http://localhost/", grpc.local_channel_credentials()) + + # Check that channel is used if provided. + transport = transports.GkeInferenceQuickstartGrpcTransport( + host="squid.clam.whelk", + channel=channel, + ) + assert transport.grpc_channel == channel + assert transport._host == "squid.clam.whelk:443" + assert transport._ssl_channel_credentials == None + + +def test_gke_inference_quickstart_grpc_asyncio_transport_channel(): + channel = aio.secure_channel("http://localhost/", grpc.local_channel_credentials()) + + # Check that channel is used if provided. + transport = transports.GkeInferenceQuickstartGrpcAsyncIOTransport( + host="squid.clam.whelk", + channel=channel, + ) + assert transport.grpc_channel == channel + assert transport._host == "squid.clam.whelk:443" + assert transport._ssl_channel_credentials == None + + +# Remove this test when deprecated arguments (api_mtls_endpoint, client_cert_source) are +# removed from grpc/grpc_asyncio transport constructor. +@pytest.mark.parametrize( + "transport_class", + [ + transports.GkeInferenceQuickstartGrpcTransport, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + ], +) +def test_gke_inference_quickstart_transport_channel_mtls_with_client_cert_source( + transport_class, +): + with mock.patch( + "grpc.ssl_channel_credentials", autospec=True + ) as grpc_ssl_channel_cred: + with mock.patch.object( + transport_class, "create_channel" + ) as grpc_create_channel: + mock_ssl_cred = mock.Mock() + grpc_ssl_channel_cred.return_value = mock_ssl_cred + + mock_grpc_channel = mock.Mock() + grpc_create_channel.return_value = mock_grpc_channel + + cred = ga_credentials.AnonymousCredentials() + with pytest.warns(DeprecationWarning): + with mock.patch.object(google.auth, "default") as adc: + adc.return_value = (cred, None) + transport = transport_class( + host="squid.clam.whelk", + api_mtls_endpoint="mtls.squid.clam.whelk", + client_cert_source=client_cert_source_callback, + ) + adc.assert_called_once() + + grpc_ssl_channel_cred.assert_called_once_with( + certificate_chain=b"cert bytes", private_key=b"key bytes" + ) + grpc_create_channel.assert_called_once_with( + "mtls.squid.clam.whelk:443", + credentials=cred, + credentials_file=None, + scopes=None, + ssl_credentials=mock_ssl_cred, + quota_project_id=None, + options=[ + ("grpc.max_send_message_length", -1), + ("grpc.max_receive_message_length", -1), + ], + ) + assert transport.grpc_channel == mock_grpc_channel + assert transport._ssl_channel_credentials == mock_ssl_cred + + +# Remove this test when deprecated arguments (api_mtls_endpoint, client_cert_source) are +# removed from grpc/grpc_asyncio transport constructor. +@pytest.mark.parametrize( + "transport_class", + [ + transports.GkeInferenceQuickstartGrpcTransport, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + ], +) +def test_gke_inference_quickstart_transport_channel_mtls_with_adc(transport_class): + mock_ssl_cred = mock.Mock() + with mock.patch.multiple( + "google.auth.transport.grpc.SslCredentials", + __init__=mock.Mock(return_value=None), + ssl_credentials=mock.PropertyMock(return_value=mock_ssl_cred), + ): + with mock.patch.object( + transport_class, "create_channel" + ) as grpc_create_channel: + mock_grpc_channel = mock.Mock() + grpc_create_channel.return_value = mock_grpc_channel + mock_cred = mock.Mock() + + with pytest.warns(DeprecationWarning): + transport = transport_class( + host="squid.clam.whelk", + credentials=mock_cred, + api_mtls_endpoint="mtls.squid.clam.whelk", + client_cert_source=None, + ) + + grpc_create_channel.assert_called_once_with( + "mtls.squid.clam.whelk:443", + credentials=mock_cred, + credentials_file=None, + scopes=None, + ssl_credentials=mock_ssl_cred, + quota_project_id=None, + options=[ + ("grpc.max_send_message_length", -1), + ("grpc.max_receive_message_length", -1), + ], + ) + assert transport.grpc_channel == mock_grpc_channel + + +def test_common_billing_account_path(): + billing_account = "squid" + expected = "billingAccounts/{billing_account}".format( + billing_account=billing_account, + ) + actual = GkeInferenceQuickstartClient.common_billing_account_path(billing_account) + assert expected == actual + + +def test_parse_common_billing_account_path(): + expected = { + "billing_account": "clam", + } + path = GkeInferenceQuickstartClient.common_billing_account_path(**expected) + + # Check that the path construction is reversible. + actual = GkeInferenceQuickstartClient.parse_common_billing_account_path(path) + assert expected == actual + + +def test_common_folder_path(): + folder = "whelk" + expected = "folders/{folder}".format( + folder=folder, + ) + actual = GkeInferenceQuickstartClient.common_folder_path(folder) + assert expected == actual + + +def test_parse_common_folder_path(): + expected = { + "folder": "octopus", + } + path = GkeInferenceQuickstartClient.common_folder_path(**expected) + + # Check that the path construction is reversible. + actual = GkeInferenceQuickstartClient.parse_common_folder_path(path) + assert expected == actual + + +def test_common_organization_path(): + organization = "oyster" + expected = "organizations/{organization}".format( + organization=organization, + ) + actual = GkeInferenceQuickstartClient.common_organization_path(organization) + assert expected == actual + + +def test_parse_common_organization_path(): + expected = { + "organization": "nudibranch", + } + path = GkeInferenceQuickstartClient.common_organization_path(**expected) + + # Check that the path construction is reversible. + actual = GkeInferenceQuickstartClient.parse_common_organization_path(path) + assert expected == actual + + +def test_common_project_path(): + project = "cuttlefish" + expected = "projects/{project}".format( + project=project, + ) + actual = GkeInferenceQuickstartClient.common_project_path(project) + assert expected == actual + + +def test_parse_common_project_path(): + expected = { + "project": "mussel", + } + path = GkeInferenceQuickstartClient.common_project_path(**expected) + + # Check that the path construction is reversible. + actual = GkeInferenceQuickstartClient.parse_common_project_path(path) + assert expected == actual + + +def test_common_location_path(): + project = "winkle" + location = "nautilus" + expected = "projects/{project}/locations/{location}".format( + project=project, + location=location, + ) + actual = GkeInferenceQuickstartClient.common_location_path(project, location) + assert expected == actual + + +def test_parse_common_location_path(): + expected = { + "project": "scallop", + "location": "abalone", + } + path = GkeInferenceQuickstartClient.common_location_path(**expected) + + # Check that the path construction is reversible. + actual = GkeInferenceQuickstartClient.parse_common_location_path(path) + assert expected == actual + + +def test_client_with_default_client_info(): + client_info = gapic_v1.client_info.ClientInfo() + + with mock.patch.object( + transports.GkeInferenceQuickstartTransport, "_prep_wrapped_messages" + ) as prep: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), + client_info=client_info, + ) + prep.assert_called_once_with(client_info) + + with mock.patch.object( + transports.GkeInferenceQuickstartTransport, "_prep_wrapped_messages" + ) as prep: + transport_class = GkeInferenceQuickstartClient.get_transport_class() + transport = transport_class( + credentials=ga_credentials.AnonymousCredentials(), + client_info=client_info, + ) + prep.assert_called_once_with(client_info) + + +def test_transport_close_grpc(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="grpc" + ) + with mock.patch.object( + type(getattr(client.transport, "_grpc_channel")), "close" + ) as close: + with client: + close.assert_not_called() + close.assert_called_once() + + +@pytest.mark.asyncio +async def test_transport_close_grpc_asyncio(): + client = GkeInferenceQuickstartAsyncClient( + credentials=async_anonymous_credentials(), transport="grpc_asyncio" + ) + with mock.patch.object( + type(getattr(client.transport, "_grpc_channel")), "close" + ) as close: + async with client: + close.assert_not_called() + close.assert_called_once() + + +def test_transport_close_rest(): + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport="rest" + ) + with mock.patch.object( + type(getattr(client.transport, "_session")), "close" + ) as close: + with client: + close.assert_not_called() + close.assert_called_once() + + +def test_client_ctx(): + transports = [ + "rest", + "grpc", + ] + for transport in transports: + client = GkeInferenceQuickstartClient( + credentials=ga_credentials.AnonymousCredentials(), transport=transport + ) + # Test client calls underlying transport. + with mock.patch.object(type(client.transport), "close") as close: + close.assert_not_called() + with client: + pass + close.assert_called() + + +@pytest.mark.parametrize( + "client_class,transport_class", + [ + (GkeInferenceQuickstartClient, transports.GkeInferenceQuickstartGrpcTransport), + ( + GkeInferenceQuickstartAsyncClient, + transports.GkeInferenceQuickstartGrpcAsyncIOTransport, + ), + ], +) +def test_api_key_credentials(client_class, transport_class): + with mock.patch.object( + google.auth._default, "get_api_key_credentials", create=True + ) as get_api_key_credentials: + mock_cred = mock.Mock() + get_api_key_credentials.return_value = mock_cred + options = client_options.ClientOptions() + options.api_key = "api_key" + with mock.patch.object(transport_class, "__init__") as patched: + patched.return_value = None + client = client_class(client_options=options) + patched.assert_called_once_with( + credentials=mock_cred, + credentials_file=None, + host=client._DEFAULT_ENDPOINT_TEMPLATE.format( + UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE + ), + scopes=None, + client_cert_source_for_mtls=None, + quota_project_id=None, + client_info=transports.base.DEFAULT_CLIENT_INFO, + always_use_jwt_access=True, + api_audience=None, + ) From c82ca9157f4b0ad117b7a77bf9fc4569b93775cc Mon Sep 17 00:00:00 2001 From: ohmayr Date: Thu, 25 Sep 2025 18:31:13 +0000 Subject: [PATCH 2/4] remove the generated package --- .../google-cloud-gkerecommender/.coveragerc | 13 - packages/google-cloud-gkerecommender/.flake8 | 34 - .../.repo-metadata.json | 14 - packages/google-cloud-gkerecommender/LICENSE | 202 - .../google-cloud-gkerecommender/MANIFEST.in | 20 - .../google-cloud-gkerecommender/README.rst | 197 - .../docs/_static/custom.css | 20 - .../docs/_templates/layout.html | 50 - .../google-cloud-gkerecommender/docs/conf.py | 385 -- .../gke_inference_quickstart.rst | 10 - .../docs/gkerecommender_v1/services_.rst | 6 - .../docs/gkerecommender_v1/types_.rst | 6 - .../docs/index.rst | 10 - .../docs/multiprocessing.rst | 7 - .../docs/summary_overview.md | 22 - .../google/cloud/gkerecommender/__init__.py | 81 - .../cloud/gkerecommender/gapic_version.py | 16 - .../google/cloud/gkerecommender/py.typed | 2 - .../cloud/gkerecommender_v1/__init__.py | 79 - .../gkerecommender_v1/gapic_metadata.json | 118 - .../cloud/gkerecommender_v1/gapic_version.py | 16 - .../google/cloud/gkerecommender_v1/py.typed | 2 - .../gkerecommender_v1/services/__init__.py | 15 - .../gke_inference_quickstart/__init__.py | 22 - .../gke_inference_quickstart/async_client.py | 894 --- .../gke_inference_quickstart/client.py | 1296 ---- .../gke_inference_quickstart/pagers.py | 669 -- .../transports/README.rst | 9 - .../transports/__init__.py | 41 - .../transports/base.py | 253 - .../transports/grpc.py | 536 -- .../transports/grpc_asyncio.py | 586 -- .../transports/rest.py | 1532 ----- .../transports/rest_base.py | 378 - .../cloud/gkerecommender_v1/types/__init__.py | 68 - .../gkerecommender_v1/types/gkerecommender.py | 983 --- packages/google-cloud-gkerecommender/mypy.ini | 3 - .../google-cloud-gkerecommender/noxfile.py | 592 -- ...uickstart_fetch_benchmarking_data_async.py | 56 - ...quickstart_fetch_benchmarking_data_sync.py | 56 - ...start_fetch_model_server_versions_async.py | 54 - ...kstart_fetch_model_server_versions_sync.py | 54 - ...ce_quickstart_fetch_model_servers_async.py | 53 - ...nce_quickstart_fetch_model_servers_sync.py | 53 - ...inference_quickstart_fetch_models_async.py | 52 - ..._inference_quickstart_fetch_models_sync.py | 52 - ...ference_quickstart_fetch_profiles_async.py | 52 - ...nference_quickstart_fetch_profiles_sync.py | 52 - ...start_generate_optimized_manifest_async.py | 57 - ...kstart_generate_optimized_manifest_sync.py | 57 - ...tadata_google.cloud.gkerecommender.v1.json | 933 --- .../fixup_gkerecommender_v1_keywords.py | 181 - packages/google-cloud-gkerecommender/setup.py | 99 - .../testing/constraints-3.10.txt | 6 - .../testing/constraints-3.11.txt | 6 - .../testing/constraints-3.12.txt | 6 - .../testing/constraints-3.13.txt | 11 - .../testing/constraints-3.7.txt | 10 - .../testing/constraints-3.8.txt | 6 - .../testing/constraints-3.9.txt | 6 - .../tests/__init__.py | 15 - .../tests/unit/__init__.py | 15 - .../tests/unit/gapic/__init__.py | 15 - .../unit/gapic/gkerecommender_v1/__init__.py | 15 - .../test_gke_inference_quickstart.py | 6051 ----------------- 65 files changed, 17210 deletions(-) delete mode 100644 packages/google-cloud-gkerecommender/.coveragerc delete mode 100644 packages/google-cloud-gkerecommender/.flake8 delete mode 100644 packages/google-cloud-gkerecommender/.repo-metadata.json delete mode 100644 packages/google-cloud-gkerecommender/LICENSE delete mode 100644 packages/google-cloud-gkerecommender/MANIFEST.in delete mode 100644 packages/google-cloud-gkerecommender/README.rst delete mode 100644 packages/google-cloud-gkerecommender/docs/_static/custom.css delete mode 100644 packages/google-cloud-gkerecommender/docs/_templates/layout.html delete mode 100644 packages/google-cloud-gkerecommender/docs/conf.py delete mode 100644 packages/google-cloud-gkerecommender/docs/gkerecommender_v1/gke_inference_quickstart.rst delete mode 100644 packages/google-cloud-gkerecommender/docs/gkerecommender_v1/services_.rst delete mode 100644 packages/google-cloud-gkerecommender/docs/gkerecommender_v1/types_.rst delete mode 100644 packages/google-cloud-gkerecommender/docs/index.rst delete mode 100644 packages/google-cloud-gkerecommender/docs/multiprocessing.rst delete mode 100644 packages/google-cloud-gkerecommender/docs/summary_overview.md delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender/gapic_version.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender/py.typed delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_metadata.json delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_version.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/py.typed delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/async_client.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/client.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/pagers.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/README.rst delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/base.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc_asyncio.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest_base.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/gkerecommender.py delete mode 100644 packages/google-cloud-gkerecommender/mypy.ini delete mode 100644 packages/google-cloud-gkerecommender/noxfile.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py delete mode 100644 packages/google-cloud-gkerecommender/samples/generated_samples/snippet_metadata_google.cloud.gkerecommender.v1.json delete mode 100644 packages/google-cloud-gkerecommender/scripts/fixup_gkerecommender_v1_keywords.py delete mode 100644 packages/google-cloud-gkerecommender/setup.py delete mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.10.txt delete mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.11.txt delete mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.12.txt delete mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.13.txt delete mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.7.txt delete mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.8.txt delete mode 100644 packages/google-cloud-gkerecommender/testing/constraints-3.9.txt delete mode 100644 packages/google-cloud-gkerecommender/tests/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/tests/unit/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/tests/unit/gapic/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/__init__.py delete mode 100644 packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/test_gke_inference_quickstart.py diff --git a/packages/google-cloud-gkerecommender/.coveragerc b/packages/google-cloud-gkerecommender/.coveragerc deleted file mode 100644 index 61f8e57597d7..000000000000 --- a/packages/google-cloud-gkerecommender/.coveragerc +++ /dev/null @@ -1,13 +0,0 @@ -[run] -branch = True - -[report] -show_missing = True -omit = - google/cloud/gkerecommender/__init__.py - google/cloud/gkerecommender/gapic_version.py -exclude_lines = - # Re-enable the standard pragma - pragma: NO COVER - # Ignore debug-only repr - def __repr__ diff --git a/packages/google-cloud-gkerecommender/.flake8 b/packages/google-cloud-gkerecommender/.flake8 deleted file mode 100644 index 90316de21489..000000000000 --- a/packages/google-cloud-gkerecommender/.flake8 +++ /dev/null @@ -1,34 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -[flake8] -# TODO(https://github.com/googleapis/gapic-generator-python/issues/2333): -# Resolve flake8 lint issues -ignore = E203, E231, E266, E501, W503 -exclude = - # TODO(https://github.com/googleapis/gapic-generator-python/issues/2333): - # Ensure that generated code passes flake8 lint - **/gapic/** - **/services/** - **/types/** - # Exclude Protobuf gencode - *_pb2.py - - # Standard linting exemptions. - **/.nox/** - __pycache__, - .git, - *.pyc, - conf.py diff --git a/packages/google-cloud-gkerecommender/.repo-metadata.json b/packages/google-cloud-gkerecommender/.repo-metadata.json deleted file mode 100644 index 4834c4122181..000000000000 --- a/packages/google-cloud-gkerecommender/.repo-metadata.json +++ /dev/null @@ -1,14 +0,0 @@ -{ - "api_shortname": "gkerecommender", - "name_pretty": "GKE Recommender API", - "product_documentation": "https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart", - "api_description": "GKE Recommender API", - "client_documentation": "https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest", - "issue_tracker": "https://github.com/googleapis/google-cloud-python/issues", - "release_level": "preview", - "language": "python", - "library_type": "GAPIC_AUTO", - "repo": "googleapis/google-cloud-python", - "distribution_name": "google-cloud-gkerecommender", - "api_id": "gkerecommender.googleapis.com" -} diff --git a/packages/google-cloud-gkerecommender/LICENSE b/packages/google-cloud-gkerecommender/LICENSE deleted file mode 100644 index d64569567334..000000000000 --- a/packages/google-cloud-gkerecommender/LICENSE +++ /dev/null @@ -1,202 +0,0 @@ - - Apache License - Version 2.0, January 2004 - http://www.apache.org/licenses/ - - TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION - - 1. Definitions. - - "License" shall mean the terms and conditions for use, reproduction, - and distribution as defined by Sections 1 through 9 of this document. - - "Licensor" shall mean the copyright owner or entity authorized by - the copyright owner that is granting the License. - - "Legal Entity" shall mean the union of the acting entity and all - other entities that control, are controlled by, or are under common - control with that entity. For the purposes of this definition, - "control" means (i) the power, direct or indirect, to cause the - direction or management of such entity, whether by contract or - otherwise, or (ii) ownership of fifty percent (50%) or more of the - outstanding shares, or (iii) beneficial ownership of such entity. - - "You" (or "Your") shall mean an individual or Legal Entity - exercising permissions granted by this License. - - "Source" form shall mean the preferred form for making modifications, - including but not limited to software source code, documentation - source, and configuration files. - - "Object" form shall mean any form resulting from mechanical - transformation or translation of a Source form, including but - not limited to compiled object code, generated documentation, - and conversions to other media types. - - "Work" shall mean the work of authorship, whether in Source or - Object form, made available under the License, as indicated by a - copyright notice that is included in or attached to the work - (an example is provided in the Appendix below). - - "Derivative Works" shall mean any work, whether in Source or Object - form, that is based on (or derived from) the Work and for which the - editorial revisions, annotations, elaborations, or other modifications - represent, as a whole, an original work of authorship. For the purposes - of this License, Derivative Works shall not include works that remain - separable from, or merely link (or bind by name) to the interfaces of, - the Work and Derivative Works thereof. - - "Contribution" shall mean any work of authorship, including - the original version of the Work and any modifications or additions - to that Work or Derivative Works thereof, that is intentionally - submitted to Licensor for inclusion in the Work by the copyright owner - or by an individual or Legal Entity authorized to submit on behalf of - the copyright owner. For the purposes of this definition, "submitted" - means any form of electronic, verbal, or written communication sent - to the Licensor or its representatives, including but not limited to - communication on electronic mailing lists, source code control systems, - and issue tracking systems that are managed by, or on behalf of, the - Licensor for the purpose of discussing and improving the Work, but - excluding communication that is conspicuously marked or otherwise - designated in writing by the copyright owner as "Not a Contribution." - - "Contributor" shall mean Licensor and any individual or Legal Entity - on behalf of whom a Contribution has been received by Licensor and - subsequently incorporated within the Work. - - 2. Grant of Copyright License. Subject to the terms and conditions of - this License, each Contributor hereby grants to You a perpetual, - worldwide, non-exclusive, no-charge, royalty-free, irrevocable - copyright license to reproduce, prepare Derivative Works of, - publicly display, publicly perform, sublicense, and distribute the - Work and such Derivative Works in Source or Object form. - - 3. Grant of Patent License. Subject to the terms and conditions of - this License, each Contributor hereby grants to You a perpetual, - worldwide, non-exclusive, no-charge, royalty-free, irrevocable - (except as stated in this section) patent license to make, have made, - use, offer to sell, sell, import, and otherwise transfer the Work, - where such license applies only to those patent claims licensable - by such Contributor that are necessarily infringed by their - Contribution(s) alone or by combination of their Contribution(s) - with the Work to which such Contribution(s) was submitted. If You - institute patent litigation against any entity (including a - cross-claim or counterclaim in a lawsuit) alleging that the Work - or a Contribution incorporated within the Work constitutes direct - or contributory patent infringement, then any patent licenses - granted to You under this License for that Work shall terminate - as of the date such litigation is filed. - - 4. Redistribution. You may reproduce and distribute copies of the - Work or Derivative Works thereof in any medium, with or without - modifications, and in Source or Object form, provided that You - meet the following conditions: - - (a) You must give any other recipients of the Work or - Derivative Works a copy of this License; and - - (b) You must cause any modified files to carry prominent notices - stating that You changed the files; and - - (c) You must retain, in the Source form of any Derivative Works - that You distribute, all copyright, patent, trademark, and - attribution notices from the Source form of the Work, - excluding those notices that do not pertain to any part of - the Derivative Works; and - - (d) If the Work includes a "NOTICE" text file as part of its - distribution, then any Derivative Works that You distribute must - include a readable copy of the attribution notices contained - within such NOTICE file, excluding those notices that do not - pertain to any part of the Derivative Works, in at least one - of the following places: within a NOTICE text file distributed - as part of the Derivative Works; within the Source form or - documentation, if provided along with the Derivative Works; or, - within a display generated by the Derivative Works, if and - wherever such third-party notices normally appear. The contents - of the NOTICE file are for informational purposes only and - do not modify the License. You may add Your own attribution - notices within Derivative Works that You distribute, alongside - or as an addendum to the NOTICE text from the Work, provided - that such additional attribution notices cannot be construed - as modifying the License. - - You may add Your own copyright statement to Your modifications and - may provide additional or different license terms and conditions - for use, reproduction, or distribution of Your modifications, or - for any such Derivative Works as a whole, provided Your use, - reproduction, and distribution of the Work otherwise complies with - the conditions stated in this License. - - 5. Submission of Contributions. Unless You explicitly state otherwise, - any Contribution intentionally submitted for inclusion in the Work - by You to the Licensor shall be under the terms and conditions of - this License, without any additional terms or conditions. - Notwithstanding the above, nothing herein shall supersede or modify - the terms of any separate license agreement you may have executed - with Licensor regarding such Contributions. - - 6. Trademarks. This License does not grant permission to use the trade - names, trademarks, service marks, or product names of the Licensor, - except as required for reasonable and customary use in describing the - origin of the Work and reproducing the content of the NOTICE file. - - 7. Disclaimer of Warranty. Unless required by applicable law or - agreed to in writing, Licensor provides the Work (and each - Contributor provides its Contributions) on an "AS IS" BASIS, - WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or - implied, including, without limitation, any warranties or conditions - of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A - PARTICULAR PURPOSE. You are solely responsible for determining the - appropriateness of using or redistributing the Work and assume any - risks associated with Your exercise of permissions under this License. - - 8. Limitation of Liability. In no event and under no legal theory, - whether in tort (including negligence), contract, or otherwise, - unless required by applicable law (such as deliberate and grossly - negligent acts) or agreed to in writing, shall any Contributor be - liable to You for damages, including any direct, indirect, special, - incidental, or consequential damages of any character arising as a - result of this License or out of the use or inability to use the - Work (including but not limited to damages for loss of goodwill, - work stoppage, computer failure or malfunction, or any and all - other commercial damages or losses), even if such Contributor - has been advised of the possibility of such damages. - - 9. Accepting Warranty or Additional Liability. While redistributing - the Work or Derivative Works thereof, You may choose to offer, - and charge a fee for, acceptance of support, warranty, indemnity, - or other liability obligations and/or rights consistent with this - License. However, in accepting such obligations, You may act only - on Your own behalf and on Your sole responsibility, not on behalf - of any other Contributor, and only if You agree to indemnify, - defend, and hold each Contributor harmless for any liability - incurred by, or claims asserted against, such Contributor by reason - of your accepting any such warranty or additional liability. - - END OF TERMS AND CONDITIONS - - APPENDIX: How to apply the Apache License to your work. - - To apply the Apache License to your work, attach the following - boilerplate notice, with the fields enclosed by brackets "[]" - replaced with your own identifying information. (Don't include - the brackets!) The text should be enclosed in the appropriate - comment syntax for the file format. We also recommend that a - file or class name and description of purpose be included on the - same "printed page" as the copyright notice for easier - identification within third-party archives. - - Copyright [yyyy] [name of copyright owner] - - Licensed under the Apache License, Version 2.0 (the "License"); - you may not use this file except in compliance with the License. - You may obtain a copy of the License at - - http://www.apache.org/licenses/LICENSE-2.0 - - Unless required by applicable law or agreed to in writing, software - distributed under the License is distributed on an "AS IS" BASIS, - WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - See the License for the specific language governing permissions and - limitations under the License. diff --git a/packages/google-cloud-gkerecommender/MANIFEST.in b/packages/google-cloud-gkerecommender/MANIFEST.in deleted file mode 100644 index dae249ec8976..000000000000 --- a/packages/google-cloud-gkerecommender/MANIFEST.in +++ /dev/null @@ -1,20 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -include README.rst LICENSE -recursive-include google *.py *.pyi *.json *.proto py.typed -recursive-include tests * -global-exclude *.py[co] -global-exclude __pycache__ diff --git a/packages/google-cloud-gkerecommender/README.rst b/packages/google-cloud-gkerecommender/README.rst deleted file mode 100644 index dd6d4b294bb3..000000000000 --- a/packages/google-cloud-gkerecommender/README.rst +++ /dev/null @@ -1,197 +0,0 @@ -Python Client for GKE Recommender API -===================================== - -|preview| |pypi| |versions| - -`GKE Recommender API`_: GKE Recommender API - -- `Client Library Documentation`_ -- `Product Documentation`_ - -.. |preview| image:: https://img.shields.io/badge/support-preview-orange.svg - :target: https://github.com/googleapis/google-cloud-python/blob/main/README.rst#stability-levels -.. |pypi| image:: https://img.shields.io/pypi/v/google-cloud-gkerecommender.svg - :target: https://pypi.org/project/google-cloud-gkerecommender/ -.. |versions| image:: https://img.shields.io/pypi/pyversions/google-cloud-gkerecommender.svg - :target: https://pypi.org/project/google-cloud-gkerecommender/ -.. _GKE Recommender API: https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart -.. _Client Library Documentation: https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest/summary_overview -.. _Product Documentation: https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart - -Quick Start ------------ - -In order to use this library, you first need to go through the following steps: - -1. `Select or create a Cloud Platform project.`_ -2. `Enable billing for your project.`_ -3. `Enable the GKE Recommender API.`_ -4. `Set up Authentication.`_ - -.. _Select or create a Cloud Platform project.: https://console.cloud.google.com/project -.. _Enable billing for your project.: https://cloud.google.com/billing/docs/how-to/modify-project#enable_billing_for_a_project -.. _Enable the GKE Recommender API.: https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart -.. _Set up Authentication.: https://googleapis.dev/python/google-api-core/latest/auth.html - -Installation -~~~~~~~~~~~~ - -Install this library in a virtual environment using `venv`_. `venv`_ is a tool that -creates isolated Python environments. These isolated environments can have separate -versions of Python packages, which allows you to isolate one project's dependencies -from the dependencies of other projects. - -With `venv`_, it's possible to install this library without needing system -install permissions, and without clashing with the installed system -dependencies. - -.. _`venv`: https://docs.python.org/3/library/venv.html - - -Code samples and snippets -~~~~~~~~~~~~~~~~~~~~~~~~~ - -Code samples and snippets live in the `samples/`_ folder. - -.. _samples/: https://github.com/googleapis/google-cloud-python/tree/main/packages/google-cloud-gkerecommender/samples - - -Supported Python Versions -^^^^^^^^^^^^^^^^^^^^^^^^^ -Our client libraries are compatible with all current `active`_ and `maintenance`_ versions of -Python. - -Python >= 3.7 - -.. _active: https://devguide.python.org/devcycle/#in-development-main-branch -.. _maintenance: https://devguide.python.org/devcycle/#maintenance-branches - -Unsupported Python Versions -^^^^^^^^^^^^^^^^^^^^^^^^^^^ -Python <= 3.6 - -If you are using an `end-of-life`_ -version of Python, we recommend that you update as soon as possible to an actively supported version. - -.. _end-of-life: https://devguide.python.org/devcycle/#end-of-life-branches - -Mac/Linux -^^^^^^^^^ - -.. code-block:: console - - python3 -m venv - source /bin/activate - pip install google-cloud-gkerecommender - - -Windows -^^^^^^^ - -.. code-block:: console - - py -m venv - .\\Scripts\activate - pip install google-cloud-gkerecommender - -Next Steps -~~~~~~~~~~ - -- Read the `Client Library Documentation`_ for GKE Recommender API - to see other available methods on the client. -- Read the `GKE Recommender API Product documentation`_ to learn - more about the product and see How-to Guides. -- View this `README`_ to see the full list of Cloud - APIs that we cover. - -.. _GKE Recommender API Product documentation: https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference-quickstart -.. _README: https://github.com/googleapis/google-cloud-python/blob/main/README.rst - -Logging -------- - -This library uses the standard Python :code:`logging` functionality to log some RPC events that could be of interest for debugging and monitoring purposes. -Note the following: - -#. Logs may contain sensitive information. Take care to **restrict access to the logs** if they are saved, whether it be on local storage or on Google Cloud Logging. -#. Google may refine the occurrence, level, and content of various log messages in this library without flagging such changes as breaking. **Do not depend on immutability of the logging events**. -#. By default, the logging events from this library are not handled. You must **explicitly configure log handling** using one of the mechanisms below. - -Simple, environment-based configuration -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - -To enable logging for this library without any changes in your code, set the :code:`GOOGLE_SDK_PYTHON_LOGGING_SCOPE` environment variable to a valid Google -logging scope. This configures handling of logging events (at level :code:`logging.DEBUG` or higher) from this library in a default manner, emitting the logged -messages in a structured format. It does not currently allow customizing the logging levels captured nor the handlers, formatters, etc. used for any logging -event. - -A logging scope is a period-separated namespace that begins with :code:`google`, identifying the Python module or package to log. - -- Valid logging scopes: :code:`google`, :code:`google.cloud.asset.v1`, :code:`google.api`, :code:`google.auth`, etc. -- Invalid logging scopes: :code:`foo`, :code:`123`, etc. - -**NOTE**: If the logging scope is invalid, the library does not set up any logging handlers. - -Environment-Based Examples -^^^^^^^^^^^^^^^^^^^^^^^^^^ - -- Enabling the default handler for all Google-based loggers - -.. code-block:: console - - export GOOGLE_SDK_PYTHON_LOGGING_SCOPE=google - -- Enabling the default handler for a specific Google module (for a client library called :code:`library_v1`): - -.. code-block:: console - - export GOOGLE_SDK_PYTHON_LOGGING_SCOPE=google.cloud.library_v1 - - -Advanced, code-based configuration -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - -You can also configure a valid logging scope using Python's standard `logging` mechanism. - -Code-Based Examples -^^^^^^^^^^^^^^^^^^^ - -- Configuring a handler for all Google-based loggers - -.. code-block:: python - - import logging - - from google.cloud import library_v1 - - base_logger = logging.getLogger("google") - base_logger.addHandler(logging.StreamHandler()) - base_logger.setLevel(logging.DEBUG) - -- Configuring a handler for a specific Google module (for a client library called :code:`library_v1`): - -.. code-block:: python - - import logging - - from google.cloud import library_v1 - - base_logger = logging.getLogger("google.cloud.library_v1") - base_logger.addHandler(logging.StreamHandler()) - base_logger.setLevel(logging.DEBUG) - -Logging details -~~~~~~~~~~~~~~~ - -#. Regardless of which of the mechanisms above you use to configure logging for this library, by default logging events are not propagated up to the root - logger from the `google`-level logger. If you need the events to be propagated to the root logger, you must explicitly set - :code:`logging.getLogger("google").propagate = True` in your code. -#. You can mix the different logging configurations above for different Google modules. For example, you may want use a code-based logging configuration for - one library, but decide you need to also set up environment-based logging configuration for another library. - - #. If you attempt to use both code-based and environment-based configuration for the same module, the environment-based configuration will be ineffectual - if the code -based configuration gets applied first. - -#. The Google-specific logging configurations (default handlers for environment-based configuration; not propagating logging events to the root logger) get - executed the first time *any* client library is instantiated in your application, and only if the affected loggers have not been previously configured. - (This is the reason for 2.i. above.) diff --git a/packages/google-cloud-gkerecommender/docs/_static/custom.css b/packages/google-cloud-gkerecommender/docs/_static/custom.css deleted file mode 100644 index b0a295464b23..000000000000 --- a/packages/google-cloud-gkerecommender/docs/_static/custom.css +++ /dev/null @@ -1,20 +0,0 @@ -div#python2-eol { - border-color: red; - border-width: medium; -} - -/* Ensure minimum width for 'Parameters' / 'Returns' column */ -dl.field-list > dt { - min-width: 100px -} - -/* Insert space between methods for readability */ -dl.method { - padding-top: 10px; - padding-bottom: 10px -} - -/* Insert empty space between classes */ -dl.class { - padding-bottom: 50px -} diff --git a/packages/google-cloud-gkerecommender/docs/_templates/layout.html b/packages/google-cloud-gkerecommender/docs/_templates/layout.html deleted file mode 100644 index 95e9c77fcfe1..000000000000 --- a/packages/google-cloud-gkerecommender/docs/_templates/layout.html +++ /dev/null @@ -1,50 +0,0 @@ - -{% extends "!layout.html" %} -{%- block content %} -{%- if theme_fixed_sidebar|lower == 'true' %} -
- {{ sidebar() }} - {%- block document %} -
- {%- if render_sidebar %} -
- {%- endif %} - - {%- block relbar_top %} - {%- if theme_show_relbar_top|tobool %} -
-   - {{- rellink_markup () }} -
- {%- endif %} - {% endblock %} - -
-
- As of January 1, 2020 this library no longer supports Python 2 on the latest released version. - Library versions released prior to that date will continue to be available. For more information please - visit Python 2 support on Google Cloud. -
- {% block body %} {% endblock %} -
- - {%- block relbar_bottom %} - {%- if theme_show_relbar_bottom|tobool %} -
-   - {{- rellink_markup () }} -
- {%- endif %} - {% endblock %} - - {%- if render_sidebar %} -
- {%- endif %} -
- {%- endblock %} -
-
-{%- else %} -{{ super() }} -{%- endif %} -{%- endblock %} diff --git a/packages/google-cloud-gkerecommender/docs/conf.py b/packages/google-cloud-gkerecommender/docs/conf.py deleted file mode 100644 index 77dcf0d97d44..000000000000 --- a/packages/google-cloud-gkerecommender/docs/conf.py +++ /dev/null @@ -1,385 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# -# google-cloud-gkerecommender documentation build configuration file -# -# This file is execfile()d with the current directory set to its -# containing dir. -# -# Note that not all possible configuration values are present in this -# autogenerated file. -# -# All configuration values have a default; values that are commented out -# serve to show the default. - -import os -import shlex -import sys - -# If extensions (or modules to document with autodoc) are in another directory, -# add these directories to sys.path here. If the directory is relative to the -# documentation root, use os.path.abspath to make it absolute, like shown here. -sys.path.insert(0, os.path.abspath("..")) - -# For plugins that can not read conf.py. -# See also: https://github.com/docascode/sphinx-docfx-yaml/issues/85 -sys.path.insert(0, os.path.abspath(".")) - -__version__ = "" - -# -- General configuration ------------------------------------------------ - -# If your documentation needs a minimal Sphinx version, state it here. -needs_sphinx = "4.5.0" - -# Add any Sphinx extension module names here, as strings. They can be -# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom -# ones. -extensions = [ - "sphinx.ext.autodoc", - "sphinx.ext.autosummary", - "sphinx.ext.intersphinx", - "sphinx.ext.coverage", - "sphinx.ext.doctest", - "sphinx.ext.napoleon", - "sphinx.ext.todo", - "sphinx.ext.viewcode", - "recommonmark", -] - -# autodoc/autosummary flags -autoclass_content = "both" -autodoc_default_options = {"members": True} -autosummary_generate = True - - -# Add any paths that contain templates here, relative to this directory. -templates_path = ["_templates"] - -# The suffix(es) of source filenames. -# You can specify multiple suffix as a list of string: -# source_suffix = ['.rst', '.md'] -source_suffix = [".rst", ".md"] - -# The encoding of source files. -# source_encoding = 'utf-8-sig' - -# The root toctree document. -root_doc = "index" - -# General information about the project. -project = "google-cloud-gkerecommender" -copyright = "2025, Google, LLC" -author = "Google APIs" - -# The version info for the project you're documenting, acts as replacement for -# |version| and |release|, also used in various other places throughout the -# built documents. -# -# The full version, including alpha/beta/rc tags. -release = __version__ -# The short X.Y version. -version = ".".join(release.split(".")[0:2]) - -# The language for content autogenerated by Sphinx. Refer to documentation -# for a list of supported languages. -# -# This is also used if you do content translation via gettext catalogs. -# Usually you set "language" from the command line for these cases. -language = None - -# There are two options for replacing |today|: either, you set today to some -# non-false value, then it is used: -# today = '' -# Else, today_fmt is used as the format for a strftime call. -# today_fmt = '%B %d, %Y' - -# List of patterns, relative to source directory, that match files and -# directories to ignore when looking for source files. -exclude_patterns = [ - "_build", - "**/.nox/**/*", - "samples/AUTHORING_GUIDE.md", - "samples/CONTRIBUTING.md", - "samples/snippets/README.rst", -] - -# The reST default role (used for this markup: `text`) to use for all -# documents. -# default_role = None - -# If true, '()' will be appended to :func: etc. cross-reference text. -# add_function_parentheses = True - -# If true, the current module name will be prepended to all description -# unit titles (such as .. function::). -# add_module_names = True - -# If true, sectionauthor and moduleauthor directives will be shown in the -# output. They are ignored by default. -# show_authors = False - -# The name of the Pygments (syntax highlighting) style to use. -pygments_style = "sphinx" - -# A list of ignored prefixes for module index sorting. -# modindex_common_prefix = [] - -# If true, keep warnings as "system message" paragraphs in the built documents. -# keep_warnings = False - -# If true, `todo` and `todoList` produce output, else they produce nothing. -todo_include_todos = True - - -# -- Options for HTML output ---------------------------------------------- - -# The theme to use for HTML and HTML Help pages. See the documentation for -# a list of builtin themes. -html_theme = "alabaster" - -# Theme options are theme-specific and customize the look and feel of a theme -# further. For a list of options available for each theme, see the -# documentation. -html_theme_options = { - "description": "Google Cloud Client Libraries for google-cloud-gkerecommender", - "github_user": "googleapis", - "github_repo": "google-cloud-python", - "github_banner": True, - "font_family": "'Roboto', Georgia, sans", - "head_font_family": "'Roboto', Georgia, serif", - "code_font_family": "'Roboto Mono', 'Consolas', monospace", -} - -# Add any paths that contain custom themes here, relative to this directory. -# html_theme_path = [] - -# The name for this set of Sphinx documents. If None, it defaults to -# " v documentation". -# html_title = None - -# A shorter title for the navigation bar. Default is the same as html_title. -# html_short_title = None - -# The name of an image file (relative to this directory) to place at the top -# of the sidebar. -# html_logo = None - -# The name of an image file (within the static path) to use as favicon of the -# docs. This file should be a Windows icon file (.ico) being 16x16 or 32x32 -# pixels large. -# html_favicon = None - -# Add any paths that contain custom static files (such as style sheets) here, -# relative to this directory. They are copied after the builtin static files, -# so a file named "default.css" will overwrite the builtin "default.css". -html_static_path = ["_static"] - -# Add any extra paths that contain custom files (such as robots.txt or -# .htaccess) here, relative to this directory. These files are copied -# directly to the root of the documentation. -# html_extra_path = [] - -# If not '', a 'Last updated on:' timestamp is inserted at every page bottom, -# using the given strftime format. -# html_last_updated_fmt = '%b %d, %Y' - -# If true, SmartyPants will be used to convert quotes and dashes to -# typographically correct entities. -# html_use_smartypants = True - -# Custom sidebar templates, maps document names to template names. -# html_sidebars = {} - -# Additional templates that should be rendered to pages, maps page names to -# template names. -# html_additional_pages = {} - -# If false, no module index is generated. -# html_domain_indices = True - -# If false, no index is generated. -# html_use_index = True - -# If true, the index is split into individual pages for each letter. -# html_split_index = False - -# If true, links to the reST sources are added to the pages. -# html_show_sourcelink = True - -# If true, "Created using Sphinx" is shown in the HTML footer. Default is True. -# html_show_sphinx = True - -# If true, "(C) Copyright ..." is shown in the HTML footer. Default is True. -# html_show_copyright = True - -# If true, an OpenSearch description file will be output, and all pages will -# contain a tag referring to it. The value of this option must be the -# base URL from which the finished HTML is served. -# html_use_opensearch = '' - -# This is the file name suffix for HTML files (e.g. ".xhtml"). -# html_file_suffix = None - -# Language to be used for generating the HTML full-text search index. -# Sphinx supports the following languages: -# 'da', 'de', 'en', 'es', 'fi', 'fr', 'hu', 'it', 'ja' -# 'nl', 'no', 'pt', 'ro', 'ru', 'sv', 'tr' -# html_search_language = 'en' - -# A dictionary with options for the search language support, empty by default. -# Now only 'ja' uses this config value -# html_search_options = {'type': 'default'} - -# The name of a javascript file (relative to the configuration directory) that -# implements a search results scorer. If empty, the default will be used. -# html_search_scorer = 'scorer.js' - -# Output file base name for HTML help builder. -htmlhelp_basename = "google-cloud-gkerecommender-doc" - -# -- Options for warnings ------------------------------------------------------ - - -suppress_warnings = [ - # Temporarily suppress this to avoid "more than one target found for - # cross-reference" warning, which are intractable for us to avoid while in - # a mono-repo. - # See https://github.com/sphinx-doc/sphinx/blob - # /2a65ffeef5c107c19084fabdd706cdff3f52d93c/sphinx/domains/python.py#L843 - "ref.python" -] - -# -- Options for LaTeX output --------------------------------------------- - -latex_elements = { - # The paper size ('letterpaper' or 'a4paper'). - # 'papersize': 'letterpaper', - # The font size ('10pt', '11pt' or '12pt'). - # 'pointsize': '10pt', - # Additional stuff for the LaTeX preamble. - # 'preamble': '', - # Latex figure (float) alignment - # 'figure_align': 'htbp', -} - -# Grouping the document tree into LaTeX files. List of tuples -# (source start file, target name, title, -# author, documentclass [howto, manual, or own class]). -latex_documents = [ - ( - root_doc, - "google-cloud-gkerecommender.tex", - "google-cloud-gkerecommender Documentation", - author, - "manual", - ) -] - -# The name of an image file (relative to this directory) to place at the top of -# the title page. -# latex_logo = None - -# For "manual" documents, if this is true, then toplevel headings are parts, -# not chapters. -# latex_use_parts = False - -# If true, show page references after internal links. -# latex_show_pagerefs = False - -# If true, show URL addresses after external links. -# latex_show_urls = False - -# Documents to append as an appendix to all manuals. -# latex_appendices = [] - -# If false, no module index is generated. -# latex_domain_indices = True - - -# -- Options for manual page output --------------------------------------- - -# One entry per manual page. List of tuples -# (source start file, name, description, authors, manual section). -man_pages = [ - ( - root_doc, - "google-cloud-gkerecommender", - "google-cloud-gkerecommender Documentation", - [author], - 1, - ) -] - -# If true, show URL addresses after external links. -# man_show_urls = False - - -# -- Options for Texinfo output ------------------------------------------- - -# Grouping the document tree into Texinfo files. List of tuples -# (source start file, target name, title, author, -# dir menu entry, description, category) -texinfo_documents = [ - ( - root_doc, - "google-cloud-gkerecommender", - "google-cloud-gkerecommender Documentation", - author, - "google-cloud-gkerecommender", - "google-cloud-gkerecommender Library", - "APIs", - ) -] - -# Documents to append as an appendix to all manuals. -# texinfo_appendices = [] - -# If false, no module index is generated. -# texinfo_domain_indices = True - -# How to display URL addresses: 'footnote', 'no', or 'inline'. -# texinfo_show_urls = 'footnote' - -# If true, do not generate a @detailmenu in the "Top" node's menu. -# texinfo_no_detailmenu = False - - -# Example configuration for intersphinx: refer to the Python standard library. -intersphinx_mapping = { - "python": ("https://python.readthedocs.org/en/latest/", None), - "google-auth": ("https://googleapis.dev/python/google-auth/latest/", None), - "google.api_core": ( - "https://googleapis.dev/python/google-api-core/latest/", - None, - ), - "grpc": ("https://grpc.github.io/grpc/python/", None), - "proto-plus": ("https://proto-plus-python.readthedocs.io/en/latest/", None), - "protobuf": ("https://googleapis.dev/python/protobuf/latest/", None), -} - - -# Napoleon settings -napoleon_google_docstring = True -napoleon_numpy_docstring = True -napoleon_include_private_with_doc = False -napoleon_include_special_with_doc = True -napoleon_use_admonition_for_examples = False -napoleon_use_admonition_for_notes = False -napoleon_use_admonition_for_references = False -napoleon_use_ivar = False -napoleon_use_param = True -napoleon_use_rtype = True diff --git a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/gke_inference_quickstart.rst b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/gke_inference_quickstart.rst deleted file mode 100644 index 60b290775d65..000000000000 --- a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/gke_inference_quickstart.rst +++ /dev/null @@ -1,10 +0,0 @@ -GkeInferenceQuickstart ----------------------------------------- - -.. automodule:: google.cloud.gkerecommender_v1.services.gke_inference_quickstart - :members: - :inherited-members: - -.. automodule:: google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers - :members: - :inherited-members: diff --git a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/services_.rst b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/services_.rst deleted file mode 100644 index 6db73968996f..000000000000 --- a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/services_.rst +++ /dev/null @@ -1,6 +0,0 @@ -Services for Google Cloud Gkerecommender v1 API -=============================================== -.. toctree:: - :maxdepth: 2 - - gke_inference_quickstart diff --git a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/types_.rst b/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/types_.rst deleted file mode 100644 index eb4cb3310dc5..000000000000 --- a/packages/google-cloud-gkerecommender/docs/gkerecommender_v1/types_.rst +++ /dev/null @@ -1,6 +0,0 @@ -Types for Google Cloud Gkerecommender v1 API -============================================ - -.. automodule:: google.cloud.gkerecommender_v1.types - :members: - :show-inheritance: diff --git a/packages/google-cloud-gkerecommender/docs/index.rst b/packages/google-cloud-gkerecommender/docs/index.rst deleted file mode 100644 index 8220e3992d15..000000000000 --- a/packages/google-cloud-gkerecommender/docs/index.rst +++ /dev/null @@ -1,10 +0,0 @@ -.. include:: multiprocessing.rst - - -API Reference -------------- -.. toctree:: - :maxdepth: 2 - - gkerecommender_v1/services_ - gkerecommender_v1/types_ diff --git a/packages/google-cloud-gkerecommender/docs/multiprocessing.rst b/packages/google-cloud-gkerecommender/docs/multiprocessing.rst deleted file mode 100644 index 536d17b2ea65..000000000000 --- a/packages/google-cloud-gkerecommender/docs/multiprocessing.rst +++ /dev/null @@ -1,7 +0,0 @@ -.. note:: - - Because this client uses :mod:`grpc` library, it is safe to - share instances across threads. In multiprocessing scenarios, the best - practice is to create client instances *after* the invocation of - :func:`os.fork` by :class:`multiprocessing.pool.Pool` or - :class:`multiprocessing.Process`. diff --git a/packages/google-cloud-gkerecommender/docs/summary_overview.md b/packages/google-cloud-gkerecommender/docs/summary_overview.md deleted file mode 100644 index 5878774fd11e..000000000000 --- a/packages/google-cloud-gkerecommender/docs/summary_overview.md +++ /dev/null @@ -1,22 +0,0 @@ -[ -This is a templated file. Adding content to this file may result in it being -reverted. Instead, if you want to place additional content, create an -"overview_content.md" file in `docs/` directory. The Sphinx tool will -pick up on the content and merge the content. -]: # - -# GKE Recommender API API - -Overview of the APIs available for GKE Recommender API API. - -## All entries - -Classes, methods and properties & attributes for -GKE Recommender API API. - -[classes](https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest/summary_class.html) - -[methods](https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest/summary_method.html) - -[properties and -attributes](https://cloud.google.com/python/docs/reference/google-cloud-gkerecommender/latest/summary_property.html) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/__init__.py deleted file mode 100644 index 937bc88d6b72..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/__init__.py +++ /dev/null @@ -1,81 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from google.cloud.gkerecommender import gapic_version as package_version - -__version__ = package_version.__version__ - - -from google.cloud.gkerecommender_v1.services.gke_inference_quickstart.async_client import ( - GkeInferenceQuickstartAsyncClient, -) -from google.cloud.gkerecommender_v1.services.gke_inference_quickstart.client import ( - GkeInferenceQuickstartClient, -) -from google.cloud.gkerecommender_v1.types.gkerecommender import ( - Amount, - Cost, - FetchBenchmarkingDataRequest, - FetchBenchmarkingDataResponse, - FetchModelServersRequest, - FetchModelServersResponse, - FetchModelServerVersionsRequest, - FetchModelServerVersionsResponse, - FetchModelsRequest, - FetchModelsResponse, - FetchProfilesRequest, - FetchProfilesResponse, - GenerateOptimizedManifestRequest, - GenerateOptimizedManifestResponse, - KubernetesManifest, - MillisecondRange, - ModelServerInfo, - PerformanceRange, - PerformanceRequirements, - PerformanceStats, - Profile, - ResourcesUsed, - StorageConfig, - TokensPerSecondRange, -) - -__all__ = ( - "GkeInferenceQuickstartClient", - "GkeInferenceQuickstartAsyncClient", - "Amount", - "Cost", - "FetchBenchmarkingDataRequest", - "FetchBenchmarkingDataResponse", - "FetchModelServersRequest", - "FetchModelServersResponse", - "FetchModelServerVersionsRequest", - "FetchModelServerVersionsResponse", - "FetchModelsRequest", - "FetchModelsResponse", - "FetchProfilesRequest", - "FetchProfilesResponse", - "GenerateOptimizedManifestRequest", - "GenerateOptimizedManifestResponse", - "KubernetesManifest", - "MillisecondRange", - "ModelServerInfo", - "PerformanceRange", - "PerformanceRequirements", - "PerformanceStats", - "Profile", - "ResourcesUsed", - "StorageConfig", - "TokensPerSecondRange", -) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/gapic_version.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/gapic_version.py deleted file mode 100644 index 20a9cd975b02..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/gapic_version.py +++ /dev/null @@ -1,16 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -__version__ = "0.0.0" # {x-release-please-version} diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/py.typed b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/py.typed deleted file mode 100644 index ebf2186dedbf..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender/py.typed +++ /dev/null @@ -1,2 +0,0 @@ -# Marker file for PEP 561. -# The google-cloud-gkerecommender package uses inline types. diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/__init__.py deleted file mode 100644 index 97b37003c197..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/__init__.py +++ /dev/null @@ -1,79 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from google.cloud.gkerecommender_v1 import gapic_version as package_version - -__version__ = package_version.__version__ - - -from .services.gke_inference_quickstart import ( - GkeInferenceQuickstartAsyncClient, - GkeInferenceQuickstartClient, -) -from .types.gkerecommender import ( - Amount, - Cost, - FetchBenchmarkingDataRequest, - FetchBenchmarkingDataResponse, - FetchModelServersRequest, - FetchModelServersResponse, - FetchModelServerVersionsRequest, - FetchModelServerVersionsResponse, - FetchModelsRequest, - FetchModelsResponse, - FetchProfilesRequest, - FetchProfilesResponse, - GenerateOptimizedManifestRequest, - GenerateOptimizedManifestResponse, - KubernetesManifest, - MillisecondRange, - ModelServerInfo, - PerformanceRange, - PerformanceRequirements, - PerformanceStats, - Profile, - ResourcesUsed, - StorageConfig, - TokensPerSecondRange, -) - -__all__ = ( - "GkeInferenceQuickstartAsyncClient", - "Amount", - "Cost", - "FetchBenchmarkingDataRequest", - "FetchBenchmarkingDataResponse", - "FetchModelServerVersionsRequest", - "FetchModelServerVersionsResponse", - "FetchModelServersRequest", - "FetchModelServersResponse", - "FetchModelsRequest", - "FetchModelsResponse", - "FetchProfilesRequest", - "FetchProfilesResponse", - "GenerateOptimizedManifestRequest", - "GenerateOptimizedManifestResponse", - "GkeInferenceQuickstartClient", - "KubernetesManifest", - "MillisecondRange", - "ModelServerInfo", - "PerformanceRange", - "PerformanceRequirements", - "PerformanceStats", - "Profile", - "ResourcesUsed", - "StorageConfig", - "TokensPerSecondRange", -) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_metadata.json b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_metadata.json deleted file mode 100644 index 277d36460b3b..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_metadata.json +++ /dev/null @@ -1,118 +0,0 @@ - { - "comment": "This file maps proto services/RPCs to the corresponding library clients/methods", - "language": "python", - "libraryPackage": "google.cloud.gkerecommender_v1", - "protoPackage": "google.cloud.gkerecommender.v1", - "schema": "1.0", - "services": { - "GkeInferenceQuickstart": { - "clients": { - "grpc": { - "libraryClient": "GkeInferenceQuickstartClient", - "rpcs": { - "FetchBenchmarkingData": { - "methods": [ - "fetch_benchmarking_data" - ] - }, - "FetchModelServerVersions": { - "methods": [ - "fetch_model_server_versions" - ] - }, - "FetchModelServers": { - "methods": [ - "fetch_model_servers" - ] - }, - "FetchModels": { - "methods": [ - "fetch_models" - ] - }, - "FetchProfiles": { - "methods": [ - "fetch_profiles" - ] - }, - "GenerateOptimizedManifest": { - "methods": [ - "generate_optimized_manifest" - ] - } - } - }, - "grpc-async": { - "libraryClient": "GkeInferenceQuickstartAsyncClient", - "rpcs": { - "FetchBenchmarkingData": { - "methods": [ - "fetch_benchmarking_data" - ] - }, - "FetchModelServerVersions": { - "methods": [ - "fetch_model_server_versions" - ] - }, - "FetchModelServers": { - "methods": [ - "fetch_model_servers" - ] - }, - "FetchModels": { - "methods": [ - "fetch_models" - ] - }, - "FetchProfiles": { - "methods": [ - "fetch_profiles" - ] - }, - "GenerateOptimizedManifest": { - "methods": [ - "generate_optimized_manifest" - ] - } - } - }, - "rest": { - "libraryClient": "GkeInferenceQuickstartClient", - "rpcs": { - "FetchBenchmarkingData": { - "methods": [ - "fetch_benchmarking_data" - ] - }, - "FetchModelServerVersions": { - "methods": [ - "fetch_model_server_versions" - ] - }, - "FetchModelServers": { - "methods": [ - "fetch_model_servers" - ] - }, - "FetchModels": { - "methods": [ - "fetch_models" - ] - }, - "FetchProfiles": { - "methods": [ - "fetch_profiles" - ] - }, - "GenerateOptimizedManifest": { - "methods": [ - "generate_optimized_manifest" - ] - } - } - } - } - } - } -} diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_version.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_version.py deleted file mode 100644 index 20a9cd975b02..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/gapic_version.py +++ /dev/null @@ -1,16 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -__version__ = "0.0.0" # {x-release-please-version} diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/py.typed b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/py.typed deleted file mode 100644 index ebf2186dedbf..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/py.typed +++ /dev/null @@ -1,2 +0,0 @@ -# Marker file for PEP 561. -# The google-cloud-gkerecommender package uses inline types. diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/__init__.py deleted file mode 100644 index cbf94b283c70..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/__init__.py +++ /dev/null @@ -1,15 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/__init__.py deleted file mode 100644 index 2f911bff8338..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/__init__.py +++ /dev/null @@ -1,22 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from .async_client import GkeInferenceQuickstartAsyncClient -from .client import GkeInferenceQuickstartClient - -__all__ = ( - "GkeInferenceQuickstartClient", - "GkeInferenceQuickstartAsyncClient", -) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/async_client.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/async_client.py deleted file mode 100644 index 19182edb0c6c..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/async_client.py +++ /dev/null @@ -1,894 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from collections import OrderedDict -import logging as std_logging -import re -from typing import ( - Callable, - Dict, - Mapping, - MutableMapping, - MutableSequence, - Optional, - Sequence, - Tuple, - Type, - Union, -) - -from google.api_core import exceptions as core_exceptions -from google.api_core import gapic_v1 -from google.api_core import retry_async as retries -from google.api_core.client_options import ClientOptions -from google.auth import credentials as ga_credentials # type: ignore -from google.oauth2 import service_account # type: ignore -import google.protobuf - -from google.cloud.gkerecommender_v1 import gapic_version as package_version - -try: - OptionalRetry = Union[retries.AsyncRetry, gapic_v1.method._MethodDefault, None] -except AttributeError: # pragma: NO COVER - OptionalRetry = Union[retries.AsyncRetry, object, None] # type: ignore - -from google.cloud.gkerecommender_v1.services.gke_inference_quickstart import pagers -from google.cloud.gkerecommender_v1.types import gkerecommender - -from .client import GkeInferenceQuickstartClient -from .transports.base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport -from .transports.grpc_asyncio import GkeInferenceQuickstartGrpcAsyncIOTransport - -try: - from google.api_core import client_logging # type: ignore - - CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER -except ImportError: # pragma: NO COVER - CLIENT_LOGGING_SUPPORTED = False - -_LOGGER = std_logging.getLogger(__name__) - - -class GkeInferenceQuickstartAsyncClient: - """GKE Inference Quickstart (GIQ) service provides profiles with - performance metrics for popular models and model servers across - multiple accelerators. These profiles help generate optimized - best practices for running inference on GKE. - """ - - _client: GkeInferenceQuickstartClient - - # Copy defaults from the synchronous client for use here. - # Note: DEFAULT_ENDPOINT is deprecated. Use _DEFAULT_ENDPOINT_TEMPLATE instead. - DEFAULT_ENDPOINT = GkeInferenceQuickstartClient.DEFAULT_ENDPOINT - DEFAULT_MTLS_ENDPOINT = GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT - _DEFAULT_ENDPOINT_TEMPLATE = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE - _DEFAULT_UNIVERSE = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE - - common_billing_account_path = staticmethod( - GkeInferenceQuickstartClient.common_billing_account_path - ) - parse_common_billing_account_path = staticmethod( - GkeInferenceQuickstartClient.parse_common_billing_account_path - ) - common_folder_path = staticmethod(GkeInferenceQuickstartClient.common_folder_path) - parse_common_folder_path = staticmethod( - GkeInferenceQuickstartClient.parse_common_folder_path - ) - common_organization_path = staticmethod( - GkeInferenceQuickstartClient.common_organization_path - ) - parse_common_organization_path = staticmethod( - GkeInferenceQuickstartClient.parse_common_organization_path - ) - common_project_path = staticmethod(GkeInferenceQuickstartClient.common_project_path) - parse_common_project_path = staticmethod( - GkeInferenceQuickstartClient.parse_common_project_path - ) - common_location_path = staticmethod( - GkeInferenceQuickstartClient.common_location_path - ) - parse_common_location_path = staticmethod( - GkeInferenceQuickstartClient.parse_common_location_path - ) - - @classmethod - def from_service_account_info(cls, info: dict, *args, **kwargs): - """Creates an instance of this client using the provided credentials - info. - - Args: - info (dict): The service account private key info. - args: Additional arguments to pass to the constructor. - kwargs: Additional arguments to pass to the constructor. - - Returns: - GkeInferenceQuickstartAsyncClient: The constructed client. - """ - return GkeInferenceQuickstartClient.from_service_account_info.__func__(GkeInferenceQuickstartAsyncClient, info, *args, **kwargs) # type: ignore - - @classmethod - def from_service_account_file(cls, filename: str, *args, **kwargs): - """Creates an instance of this client using the provided credentials - file. - - Args: - filename (str): The path to the service account private key json - file. - args: Additional arguments to pass to the constructor. - kwargs: Additional arguments to pass to the constructor. - - Returns: - GkeInferenceQuickstartAsyncClient: The constructed client. - """ - return GkeInferenceQuickstartClient.from_service_account_file.__func__(GkeInferenceQuickstartAsyncClient, filename, *args, **kwargs) # type: ignore - - from_service_account_json = from_service_account_file - - @classmethod - def get_mtls_endpoint_and_cert_source( - cls, client_options: Optional[ClientOptions] = None - ): - """Return the API endpoint and client cert source for mutual TLS. - - The client cert source is determined in the following order: - (1) if `GOOGLE_API_USE_CLIENT_CERTIFICATE` environment variable is not "true", the - client cert source is None. - (2) if `client_options.client_cert_source` is provided, use the provided one; if the - default client cert source exists, use the default one; otherwise the client cert - source is None. - - The API endpoint is determined in the following order: - (1) if `client_options.api_endpoint` if provided, use the provided one. - (2) if `GOOGLE_API_USE_CLIENT_CERTIFICATE` environment variable is "always", use the - default mTLS endpoint; if the environment variable is "never", use the default API - endpoint; otherwise if client cert source exists, use the default mTLS endpoint, otherwise - use the default API endpoint. - - More details can be found at https://google.aip.dev/auth/4114. - - Args: - client_options (google.api_core.client_options.ClientOptions): Custom options for the - client. Only the `api_endpoint` and `client_cert_source` properties may be used - in this method. - - Returns: - Tuple[str, Callable[[], Tuple[bytes, bytes]]]: returns the API endpoint and the - client cert source to use. - - Raises: - google.auth.exceptions.MutualTLSChannelError: If any errors happen. - """ - return GkeInferenceQuickstartClient.get_mtls_endpoint_and_cert_source(client_options) # type: ignore - - @property - def transport(self) -> GkeInferenceQuickstartTransport: - """Returns the transport used by the client instance. - - Returns: - GkeInferenceQuickstartTransport: The transport used by the client instance. - """ - return self._client.transport - - @property - def api_endpoint(self): - """Return the API endpoint used by the client instance. - - Returns: - str: The API endpoint used by the client instance. - """ - return self._client._api_endpoint - - @property - def universe_domain(self) -> str: - """Return the universe domain used by the client instance. - - Returns: - str: The universe domain used - by the client instance. - """ - return self._client._universe_domain - - get_transport_class = GkeInferenceQuickstartClient.get_transport_class - - def __init__( - self, - *, - credentials: Optional[ga_credentials.Credentials] = None, - transport: Optional[ - Union[ - str, - GkeInferenceQuickstartTransport, - Callable[..., GkeInferenceQuickstartTransport], - ] - ] = "grpc_asyncio", - client_options: Optional[ClientOptions] = None, - client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, - ) -> None: - """Instantiates the gke inference quickstart async client. - - Args: - credentials (Optional[google.auth.credentials.Credentials]): The - authorization credentials to attach to requests. These - credentials identify the application to the service; if none - are specified, the client will attempt to ascertain the - credentials from the environment. - transport (Optional[Union[str,GkeInferenceQuickstartTransport,Callable[..., GkeInferenceQuickstartTransport]]]): - The transport to use, or a Callable that constructs and returns a new transport to use. - If a Callable is given, it will be called with the same set of initialization - arguments as used in the GkeInferenceQuickstartTransport constructor. - If set to None, a transport is chosen automatically. - client_options (Optional[Union[google.api_core.client_options.ClientOptions, dict]]): - Custom options for the client. - - 1. The ``api_endpoint`` property can be used to override the - default endpoint provided by the client when ``transport`` is - not explicitly provided. Only if this property is not set and - ``transport`` was not explicitly provided, the endpoint is - determined by the GOOGLE_API_USE_MTLS_ENDPOINT environment - variable, which have one of the following values: - "always" (always use the default mTLS endpoint), "never" (always - use the default regular endpoint) and "auto" (auto-switch to the - default mTLS endpoint if client certificate is present; this is - the default value). - - 2. If the GOOGLE_API_USE_CLIENT_CERTIFICATE environment variable - is "true", then the ``client_cert_source`` property can be used - to provide a client certificate for mTLS transport. If - not provided, the default SSL client certificate will be used if - present. If GOOGLE_API_USE_CLIENT_CERTIFICATE is "false" or not - set, no client certificate will be used. - - 3. The ``universe_domain`` property can be used to override the - default "googleapis.com" universe. Note that ``api_endpoint`` - property still takes precedence; and ``universe_domain`` is - currently not supported for mTLS. - - client_info (google.api_core.gapic_v1.client_info.ClientInfo): - The client info used to send a user-agent string along with - API requests. If ``None``, then default info will be used. - Generally, you only need to set this if you're developing - your own client library. - - Raises: - google.auth.exceptions.MutualTlsChannelError: If mutual TLS transport - creation failed for any reason. - """ - self._client = GkeInferenceQuickstartClient( - credentials=credentials, - transport=transport, - client_options=client_options, - client_info=client_info, - ) - - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - std_logging.DEBUG - ): # pragma: NO COVER - _LOGGER.debug( - "Created client `google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient`.", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "universeDomain": getattr( - self._client._transport._credentials, "universe_domain", "" - ), - "credentialsType": f"{type(self._client._transport._credentials).__module__}.{type(self._client._transport._credentials).__qualname__}", - "credentialsInfo": getattr( - self.transport._credentials, "get_cred_info", lambda: None - )(), - } - if hasattr(self._client._transport, "_credentials") - else { - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "credentialsType": None, - }, - ) - - async def fetch_models( - self, - request: Optional[Union[gkerecommender.FetchModelsRequest, dict]] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> pagers.FetchModelsAsyncPager: - r"""Fetches available models. Open-source models follow the - Huggingface Hub ``owner/model_name`` format. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - async def sample_fetch_models(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelsRequest( - ) - - # Make the request - page_result = client.fetch_models(request=request) - - # Handle the response - async for response in page_result: - print(response) - - Args: - request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchModelsRequest, dict]]): - The request object. Request message for - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. - retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsAsyncPager: - Response message for - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. - - Iterating over this object will yield results and - resolve additional pages automatically. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchModelsRequest): - request = gkerecommender.FetchModelsRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._client._transport._wrapped_methods[ - self._client._transport.fetch_models - ] - - # Validate the universe domain. - self._client._validate_universe_domain() - - # Send the request. - response = await rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # This method is paged; wrap the response in a pager, which provides - # an `__aiter__` convenience method. - response = pagers.FetchModelsAsyncPager( - method=rpc, - request=request, - response=response, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - async def fetch_model_servers( - self, - request: Optional[Union[gkerecommender.FetchModelServersRequest, dict]] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> pagers.FetchModelServersAsyncPager: - r"""Fetches available model servers. Open-source model servers use - simplified, lowercase names (e.g., ``vllm``). - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - async def sample_fetch_model_servers(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelServersRequest( - model="model_value", - ) - - # Make the request - page_result = client.fetch_model_servers(request=request) - - # Handle the response - async for response in page_result: - print(response) - - Args: - request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchModelServersRequest, dict]]): - The request object. Request message for - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. - retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersAsyncPager: - Response message for - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. - - Iterating over this object will yield results and - resolve additional pages automatically. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchModelServersRequest): - request = gkerecommender.FetchModelServersRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._client._transport._wrapped_methods[ - self._client._transport.fetch_model_servers - ] - - # Validate the universe domain. - self._client._validate_universe_domain() - - # Send the request. - response = await rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # This method is paged; wrap the response in a pager, which provides - # an `__aiter__` convenience method. - response = pagers.FetchModelServersAsyncPager( - method=rpc, - request=request, - response=response, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - async def fetch_model_server_versions( - self, - request: Optional[ - Union[gkerecommender.FetchModelServerVersionsRequest, dict] - ] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> pagers.FetchModelServerVersionsAsyncPager: - r"""Fetches available model server versions. Open-source servers use - their own versioning schemas (e.g., ``vllm`` uses semver like - ``v1.0.0``). - - Some model servers have different versioning schemas depending - on the accelerator. For example, ``vllm`` uses semver on GPUs, - but returns nightly build tags on TPUs. All available versions - will be returned when different schemas are present. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - async def sample_fetch_model_server_versions(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelServerVersionsRequest( - model="model_value", - model_server="model_server_value", - ) - - # Make the request - page_result = client.fetch_model_server_versions(request=request) - - # Handle the response - async for response in page_result: - print(response) - - Args: - request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest, dict]]): - The request object. Request message for - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. - retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsAsyncPager: - Response message for - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. - - Iterating over this object will yield results and - resolve additional pages automatically. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchModelServerVersionsRequest): - request = gkerecommender.FetchModelServerVersionsRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._client._transport._wrapped_methods[ - self._client._transport.fetch_model_server_versions - ] - - # Validate the universe domain. - self._client._validate_universe_domain() - - # Send the request. - response = await rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # This method is paged; wrap the response in a pager, which provides - # an `__aiter__` convenience method. - response = pagers.FetchModelServerVersionsAsyncPager( - method=rpc, - request=request, - response=response, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - async def fetch_profiles( - self, - request: Optional[Union[gkerecommender.FetchProfilesRequest, dict]] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> pagers.FetchProfilesAsyncPager: - r"""Fetches available profiles. A profile contains performance - metrics and cost information for a specific model server setup. - Profiles can be filtered by parameters. If no filters are - provided, all profiles are returned. - - Profiles display a single value per performance metric based on - the provided performance requirements. If no requirements are - given, the metrics represent the inflection point. See `Run best - practice inference with GKE Inference Quickstart - recipes `__ - for details. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - async def sample_fetch_profiles(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchProfilesRequest( - ) - - # Make the request - page_result = client.fetch_profiles(request=request) - - # Handle the response - async for response in page_result: - print(response) - - Args: - request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchProfilesRequest, dict]]): - The request object. Request message for - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesAsyncPager: - Response message for - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - - Iterating over this object will yield results and - resolve additional pages automatically. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchProfilesRequest): - request = gkerecommender.FetchProfilesRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._client._transport._wrapped_methods[ - self._client._transport.fetch_profiles - ] - - # Validate the universe domain. - self._client._validate_universe_domain() - - # Send the request. - response = await rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # This method is paged; wrap the response in a pager, which provides - # an `__aiter__` convenience method. - response = pagers.FetchProfilesAsyncPager( - method=rpc, - request=request, - response=response, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - async def generate_optimized_manifest( - self, - request: Optional[ - Union[gkerecommender.GenerateOptimizedManifestRequest, dict] - ] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.GenerateOptimizedManifestResponse: - r"""Generates an optimized deployment manifest for a given model and - model server, based on the specified accelerator, performance - targets, and configurations. See `Run best practice inference - with GKE Inference Quickstart - recipes `__ - for deployment details. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - async def sample_generate_optimized_manifest(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - model_server_info = gkerecommender_v1.ModelServerInfo() - model_server_info.model = "model_value" - model_server_info.model_server = "model_server_value" - - request = gkerecommender_v1.GenerateOptimizedManifestRequest( - model_server_info=model_server_info, - accelerator_type="accelerator_type_value", - ) - - # Make the request - response = await client.generate_optimized_manifest(request=request) - - # Handle the response - print(response) - - Args: - request (Optional[Union[google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest, dict]]): - The request object. Request message for - [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. - retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse: - Response message for - [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.GenerateOptimizedManifestRequest): - request = gkerecommender.GenerateOptimizedManifestRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._client._transport._wrapped_methods[ - self._client._transport.generate_optimized_manifest - ] - - # Validate the universe domain. - self._client._validate_universe_domain() - - # Send the request. - response = await rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - async def fetch_benchmarking_data( - self, - request: Optional[ - Union[gkerecommender.FetchBenchmarkingDataRequest, dict] - ] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.FetchBenchmarkingDataResponse: - r"""Fetches all of the benchmarking data available for a - profile. Benchmarking data returns all of the - performance metrics available for a given model server - setup on a given instance type. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - async def sample_fetch_benchmarking_data(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - model_server_info = gkerecommender_v1.ModelServerInfo() - model_server_info.model = "model_value" - model_server_info.model_server = "model_server_value" - - request = gkerecommender_v1.FetchBenchmarkingDataRequest( - model_server_info=model_server_info, - ) - - # Make the request - response = await client.fetch_benchmarking_data(request=request) - - # Handle the response - print(response) - - Args: - request (Optional[Union[google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest, dict]]): - The request object. Request message for - [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. - retry (google.api_core.retry_async.AsyncRetry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse: - Response message for - [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchBenchmarkingDataRequest): - request = gkerecommender.FetchBenchmarkingDataRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._client._transport._wrapped_methods[ - self._client._transport.fetch_benchmarking_data - ] - - # Validate the universe domain. - self._client._validate_universe_domain() - - # Send the request. - response = await rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - async def __aenter__(self) -> "GkeInferenceQuickstartAsyncClient": - return self - - async def __aexit__(self, exc_type, exc, tb): - await self.transport.close() - - -DEFAULT_CLIENT_INFO = gapic_v1.client_info.ClientInfo( - gapic_version=package_version.__version__ -) - -if hasattr(DEFAULT_CLIENT_INFO, "protobuf_runtime_version"): # pragma: NO COVER - DEFAULT_CLIENT_INFO.protobuf_runtime_version = google.protobuf.__version__ - - -__all__ = ("GkeInferenceQuickstartAsyncClient",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/client.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/client.py deleted file mode 100644 index cb2284d0744c..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/client.py +++ /dev/null @@ -1,1296 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from collections import OrderedDict -from http import HTTPStatus -import json -import logging as std_logging -import os -import re -from typing import ( - Callable, - Dict, - Mapping, - MutableMapping, - MutableSequence, - Optional, - Sequence, - Tuple, - Type, - Union, - cast, -) -import warnings - -from google.api_core import client_options as client_options_lib -from google.api_core import exceptions as core_exceptions -from google.api_core import gapic_v1 -from google.api_core import retry as retries -from google.auth import credentials as ga_credentials # type: ignore -from google.auth.exceptions import MutualTLSChannelError # type: ignore -from google.auth.transport import mtls # type: ignore -from google.auth.transport.grpc import SslCredentials # type: ignore -from google.oauth2 import service_account # type: ignore -import google.protobuf - -from google.cloud.gkerecommender_v1 import gapic_version as package_version - -try: - OptionalRetry = Union[retries.Retry, gapic_v1.method._MethodDefault, None] -except AttributeError: # pragma: NO COVER - OptionalRetry = Union[retries.Retry, object, None] # type: ignore - -try: - from google.api_core import client_logging # type: ignore - - CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER -except ImportError: # pragma: NO COVER - CLIENT_LOGGING_SUPPORTED = False - -_LOGGER = std_logging.getLogger(__name__) - -from google.cloud.gkerecommender_v1.services.gke_inference_quickstart import pagers -from google.cloud.gkerecommender_v1.types import gkerecommender - -from .transports.base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport -from .transports.grpc import GkeInferenceQuickstartGrpcTransport -from .transports.grpc_asyncio import GkeInferenceQuickstartGrpcAsyncIOTransport -from .transports.rest import GkeInferenceQuickstartRestTransport - - -class GkeInferenceQuickstartClientMeta(type): - """Metaclass for the GkeInferenceQuickstart client. - - This provides class-level methods for building and retrieving - support objects (e.g. transport) without polluting the client instance - objects. - """ - - _transport_registry = ( - OrderedDict() - ) # type: Dict[str, Type[GkeInferenceQuickstartTransport]] - _transport_registry["grpc"] = GkeInferenceQuickstartGrpcTransport - _transport_registry["grpc_asyncio"] = GkeInferenceQuickstartGrpcAsyncIOTransport - _transport_registry["rest"] = GkeInferenceQuickstartRestTransport - - def get_transport_class( - cls, - label: Optional[str] = None, - ) -> Type[GkeInferenceQuickstartTransport]: - """Returns an appropriate transport class. - - Args: - label: The name of the desired transport. If none is - provided, then the first transport in the registry is used. - - Returns: - The transport class to use. - """ - # If a specific transport is requested, return that one. - if label: - return cls._transport_registry[label] - - # No transport is requested; return the default (that is, the first one - # in the dictionary). - return next(iter(cls._transport_registry.values())) - - -class GkeInferenceQuickstartClient(metaclass=GkeInferenceQuickstartClientMeta): - """GKE Inference Quickstart (GIQ) service provides profiles with - performance metrics for popular models and model servers across - multiple accelerators. These profiles help generate optimized - best practices for running inference on GKE. - """ - - @staticmethod - def _get_default_mtls_endpoint(api_endpoint): - """Converts api endpoint to mTLS endpoint. - - Convert "*.sandbox.googleapis.com" and "*.googleapis.com" to - "*.mtls.sandbox.googleapis.com" and "*.mtls.googleapis.com" respectively. - Args: - api_endpoint (Optional[str]): the api endpoint to convert. - Returns: - str: converted mTLS api endpoint. - """ - if not api_endpoint: - return api_endpoint - - mtls_endpoint_re = re.compile( - r"(?P[^.]+)(?P\.mtls)?(?P\.sandbox)?(?P\.googleapis\.com)?" - ) - - m = mtls_endpoint_re.match(api_endpoint) - name, mtls, sandbox, googledomain = m.groups() - if mtls or not googledomain: - return api_endpoint - - if sandbox: - return api_endpoint.replace( - "sandbox.googleapis.com", "mtls.sandbox.googleapis.com" - ) - - return api_endpoint.replace(".googleapis.com", ".mtls.googleapis.com") - - # Note: DEFAULT_ENDPOINT is deprecated. Use _DEFAULT_ENDPOINT_TEMPLATE instead. - DEFAULT_ENDPOINT = "gkerecommender.googleapis.com" - DEFAULT_MTLS_ENDPOINT = _get_default_mtls_endpoint.__func__( # type: ignore - DEFAULT_ENDPOINT - ) - - _DEFAULT_ENDPOINT_TEMPLATE = "gkerecommender.{UNIVERSE_DOMAIN}" - _DEFAULT_UNIVERSE = "googleapis.com" - - @classmethod - def from_service_account_info(cls, info: dict, *args, **kwargs): - """Creates an instance of this client using the provided credentials - info. - - Args: - info (dict): The service account private key info. - args: Additional arguments to pass to the constructor. - kwargs: Additional arguments to pass to the constructor. - - Returns: - GkeInferenceQuickstartClient: The constructed client. - """ - credentials = service_account.Credentials.from_service_account_info(info) - kwargs["credentials"] = credentials - return cls(*args, **kwargs) - - @classmethod - def from_service_account_file(cls, filename: str, *args, **kwargs): - """Creates an instance of this client using the provided credentials - file. - - Args: - filename (str): The path to the service account private key json - file. - args: Additional arguments to pass to the constructor. - kwargs: Additional arguments to pass to the constructor. - - Returns: - GkeInferenceQuickstartClient: The constructed client. - """ - credentials = service_account.Credentials.from_service_account_file(filename) - kwargs["credentials"] = credentials - return cls(*args, **kwargs) - - from_service_account_json = from_service_account_file - - @property - def transport(self) -> GkeInferenceQuickstartTransport: - """Returns the transport used by the client instance. - - Returns: - GkeInferenceQuickstartTransport: The transport used by the client - instance. - """ - return self._transport - - @staticmethod - def common_billing_account_path( - billing_account: str, - ) -> str: - """Returns a fully-qualified billing_account string.""" - return "billingAccounts/{billing_account}".format( - billing_account=billing_account, - ) - - @staticmethod - def parse_common_billing_account_path(path: str) -> Dict[str, str]: - """Parse a billing_account path into its component segments.""" - m = re.match(r"^billingAccounts/(?P.+?)$", path) - return m.groupdict() if m else {} - - @staticmethod - def common_folder_path( - folder: str, - ) -> str: - """Returns a fully-qualified folder string.""" - return "folders/{folder}".format( - folder=folder, - ) - - @staticmethod - def parse_common_folder_path(path: str) -> Dict[str, str]: - """Parse a folder path into its component segments.""" - m = re.match(r"^folders/(?P.+?)$", path) - return m.groupdict() if m else {} - - @staticmethod - def common_organization_path( - organization: str, - ) -> str: - """Returns a fully-qualified organization string.""" - return "organizations/{organization}".format( - organization=organization, - ) - - @staticmethod - def parse_common_organization_path(path: str) -> Dict[str, str]: - """Parse a organization path into its component segments.""" - m = re.match(r"^organizations/(?P.+?)$", path) - return m.groupdict() if m else {} - - @staticmethod - def common_project_path( - project: str, - ) -> str: - """Returns a fully-qualified project string.""" - return "projects/{project}".format( - project=project, - ) - - @staticmethod - def parse_common_project_path(path: str) -> Dict[str, str]: - """Parse a project path into its component segments.""" - m = re.match(r"^projects/(?P.+?)$", path) - return m.groupdict() if m else {} - - @staticmethod - def common_location_path( - project: str, - location: str, - ) -> str: - """Returns a fully-qualified location string.""" - return "projects/{project}/locations/{location}".format( - project=project, - location=location, - ) - - @staticmethod - def parse_common_location_path(path: str) -> Dict[str, str]: - """Parse a location path into its component segments.""" - m = re.match(r"^projects/(?P.+?)/locations/(?P.+?)$", path) - return m.groupdict() if m else {} - - @classmethod - def get_mtls_endpoint_and_cert_source( - cls, client_options: Optional[client_options_lib.ClientOptions] = None - ): - """Deprecated. Return the API endpoint and client cert source for mutual TLS. - - The client cert source is determined in the following order: - (1) if `GOOGLE_API_USE_CLIENT_CERTIFICATE` environment variable is not "true", the - client cert source is None. - (2) if `client_options.client_cert_source` is provided, use the provided one; if the - default client cert source exists, use the default one; otherwise the client cert - source is None. - - The API endpoint is determined in the following order: - (1) if `client_options.api_endpoint` if provided, use the provided one. - (2) if `GOOGLE_API_USE_CLIENT_CERTIFICATE` environment variable is "always", use the - default mTLS endpoint; if the environment variable is "never", use the default API - endpoint; otherwise if client cert source exists, use the default mTLS endpoint, otherwise - use the default API endpoint. - - More details can be found at https://google.aip.dev/auth/4114. - - Args: - client_options (google.api_core.client_options.ClientOptions): Custom options for the - client. Only the `api_endpoint` and `client_cert_source` properties may be used - in this method. - - Returns: - Tuple[str, Callable[[], Tuple[bytes, bytes]]]: returns the API endpoint and the - client cert source to use. - - Raises: - google.auth.exceptions.MutualTLSChannelError: If any errors happen. - """ - - warnings.warn( - "get_mtls_endpoint_and_cert_source is deprecated. Use the api_endpoint property instead.", - DeprecationWarning, - ) - if client_options is None: - client_options = client_options_lib.ClientOptions() - use_client_cert = os.getenv("GOOGLE_API_USE_CLIENT_CERTIFICATE", "false") - use_mtls_endpoint = os.getenv("GOOGLE_API_USE_MTLS_ENDPOINT", "auto") - if use_client_cert not in ("true", "false"): - raise ValueError( - "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" - ) - if use_mtls_endpoint not in ("auto", "never", "always"): - raise MutualTLSChannelError( - "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" - ) - - # Figure out the client cert source to use. - client_cert_source = None - if use_client_cert == "true": - if client_options.client_cert_source: - client_cert_source = client_options.client_cert_source - elif mtls.has_default_client_cert_source(): - client_cert_source = mtls.default_client_cert_source() - - # Figure out which api endpoint to use. - if client_options.api_endpoint is not None: - api_endpoint = client_options.api_endpoint - elif use_mtls_endpoint == "always" or ( - use_mtls_endpoint == "auto" and client_cert_source - ): - api_endpoint = cls.DEFAULT_MTLS_ENDPOINT - else: - api_endpoint = cls.DEFAULT_ENDPOINT - - return api_endpoint, client_cert_source - - @staticmethod - def _read_environment_variables(): - """Returns the environment variables used by the client. - - Returns: - Tuple[bool, str, str]: returns the GOOGLE_API_USE_CLIENT_CERTIFICATE, - GOOGLE_API_USE_MTLS_ENDPOINT, and GOOGLE_CLOUD_UNIVERSE_DOMAIN environment variables. - - Raises: - ValueError: If GOOGLE_API_USE_CLIENT_CERTIFICATE is not - any of ["true", "false"]. - google.auth.exceptions.MutualTLSChannelError: If GOOGLE_API_USE_MTLS_ENDPOINT - is not any of ["auto", "never", "always"]. - """ - use_client_cert = os.getenv( - "GOOGLE_API_USE_CLIENT_CERTIFICATE", "false" - ).lower() - use_mtls_endpoint = os.getenv("GOOGLE_API_USE_MTLS_ENDPOINT", "auto").lower() - universe_domain_env = os.getenv("GOOGLE_CLOUD_UNIVERSE_DOMAIN") - if use_client_cert not in ("true", "false"): - raise ValueError( - "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" - ) - if use_mtls_endpoint not in ("auto", "never", "always"): - raise MutualTLSChannelError( - "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" - ) - return use_client_cert == "true", use_mtls_endpoint, universe_domain_env - - @staticmethod - def _get_client_cert_source(provided_cert_source, use_cert_flag): - """Return the client cert source to be used by the client. - - Args: - provided_cert_source (bytes): The client certificate source provided. - use_cert_flag (bool): A flag indicating whether to use the client certificate. - - Returns: - bytes or None: The client cert source to be used by the client. - """ - client_cert_source = None - if use_cert_flag: - if provided_cert_source: - client_cert_source = provided_cert_source - elif mtls.has_default_client_cert_source(): - client_cert_source = mtls.default_client_cert_source() - return client_cert_source - - @staticmethod - def _get_api_endpoint( - api_override, client_cert_source, universe_domain, use_mtls_endpoint - ): - """Return the API endpoint used by the client. - - Args: - api_override (str): The API endpoint override. If specified, this is always - the return value of this function and the other arguments are not used. - client_cert_source (bytes): The client certificate source used by the client. - universe_domain (str): The universe domain used by the client. - use_mtls_endpoint (str): How to use the mTLS endpoint, which depends also on the other parameters. - Possible values are "always", "auto", or "never". - - Returns: - str: The API endpoint to be used by the client. - """ - if api_override is not None: - api_endpoint = api_override - elif use_mtls_endpoint == "always" or ( - use_mtls_endpoint == "auto" and client_cert_source - ): - _default_universe = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE - if universe_domain != _default_universe: - raise MutualTLSChannelError( - f"mTLS is not supported in any universe other than {_default_universe}." - ) - api_endpoint = GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT - else: - api_endpoint = ( - GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=universe_domain - ) - ) - return api_endpoint - - @staticmethod - def _get_universe_domain( - client_universe_domain: Optional[str], universe_domain_env: Optional[str] - ) -> str: - """Return the universe domain used by the client. - - Args: - client_universe_domain (Optional[str]): The universe domain configured via the client options. - universe_domain_env (Optional[str]): The universe domain configured via the "GOOGLE_CLOUD_UNIVERSE_DOMAIN" environment variable. - - Returns: - str: The universe domain to be used by the client. - - Raises: - ValueError: If the universe domain is an empty string. - """ - universe_domain = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE - if client_universe_domain is not None: - universe_domain = client_universe_domain - elif universe_domain_env is not None: - universe_domain = universe_domain_env - if len(universe_domain.strip()) == 0: - raise ValueError("Universe Domain cannot be an empty string.") - return universe_domain - - def _validate_universe_domain(self): - """Validates client's and credentials' universe domains are consistent. - - Returns: - bool: True iff the configured universe domain is valid. - - Raises: - ValueError: If the configured universe domain is not valid. - """ - - # NOTE (b/349488459): universe validation is disabled until further notice. - return True - - def _add_cred_info_for_auth_errors( - self, error: core_exceptions.GoogleAPICallError - ) -> None: - """Adds credential info string to error details for 401/403/404 errors. - - Args: - error (google.api_core.exceptions.GoogleAPICallError): The error to add the cred info. - """ - if error.code not in [ - HTTPStatus.UNAUTHORIZED, - HTTPStatus.FORBIDDEN, - HTTPStatus.NOT_FOUND, - ]: - return - - cred = self._transport._credentials - - # get_cred_info is only available in google-auth>=2.35.0 - if not hasattr(cred, "get_cred_info"): - return - - # ignore the type check since pypy test fails when get_cred_info - # is not available - cred_info = cred.get_cred_info() # type: ignore - if cred_info and hasattr(error._details, "append"): - error._details.append(json.dumps(cred_info)) - - @property - def api_endpoint(self): - """Return the API endpoint used by the client instance. - - Returns: - str: The API endpoint used by the client instance. - """ - return self._api_endpoint - - @property - def universe_domain(self) -> str: - """Return the universe domain used by the client instance. - - Returns: - str: The universe domain used by the client instance. - """ - return self._universe_domain - - def __init__( - self, - *, - credentials: Optional[ga_credentials.Credentials] = None, - transport: Optional[ - Union[ - str, - GkeInferenceQuickstartTransport, - Callable[..., GkeInferenceQuickstartTransport], - ] - ] = None, - client_options: Optional[Union[client_options_lib.ClientOptions, dict]] = None, - client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, - ) -> None: - """Instantiates the gke inference quickstart client. - - Args: - credentials (Optional[google.auth.credentials.Credentials]): The - authorization credentials to attach to requests. These - credentials identify the application to the service; if none - are specified, the client will attempt to ascertain the - credentials from the environment. - transport (Optional[Union[str,GkeInferenceQuickstartTransport,Callable[..., GkeInferenceQuickstartTransport]]]): - The transport to use, or a Callable that constructs and returns a new transport. - If a Callable is given, it will be called with the same set of initialization - arguments as used in the GkeInferenceQuickstartTransport constructor. - If set to None, a transport is chosen automatically. - client_options (Optional[Union[google.api_core.client_options.ClientOptions, dict]]): - Custom options for the client. - - 1. The ``api_endpoint`` property can be used to override the - default endpoint provided by the client when ``transport`` is - not explicitly provided. Only if this property is not set and - ``transport`` was not explicitly provided, the endpoint is - determined by the GOOGLE_API_USE_MTLS_ENDPOINT environment - variable, which have one of the following values: - "always" (always use the default mTLS endpoint), "never" (always - use the default regular endpoint) and "auto" (auto-switch to the - default mTLS endpoint if client certificate is present; this is - the default value). - - 2. If the GOOGLE_API_USE_CLIENT_CERTIFICATE environment variable - is "true", then the ``client_cert_source`` property can be used - to provide a client certificate for mTLS transport. If - not provided, the default SSL client certificate will be used if - present. If GOOGLE_API_USE_CLIENT_CERTIFICATE is "false" or not - set, no client certificate will be used. - - 3. The ``universe_domain`` property can be used to override the - default "googleapis.com" universe. Note that the ``api_endpoint`` - property still takes precedence; and ``universe_domain`` is - currently not supported for mTLS. - - client_info (google.api_core.gapic_v1.client_info.ClientInfo): - The client info used to send a user-agent string along with - API requests. If ``None``, then default info will be used. - Generally, you only need to set this if you're developing - your own client library. - - Raises: - google.auth.exceptions.MutualTLSChannelError: If mutual TLS transport - creation failed for any reason. - """ - self._client_options = client_options - if isinstance(self._client_options, dict): - self._client_options = client_options_lib.from_dict(self._client_options) - if self._client_options is None: - self._client_options = client_options_lib.ClientOptions() - self._client_options = cast( - client_options_lib.ClientOptions, self._client_options - ) - - universe_domain_opt = getattr(self._client_options, "universe_domain", None) - - ( - self._use_client_cert, - self._use_mtls_endpoint, - self._universe_domain_env, - ) = GkeInferenceQuickstartClient._read_environment_variables() - self._client_cert_source = GkeInferenceQuickstartClient._get_client_cert_source( - self._client_options.client_cert_source, self._use_client_cert - ) - self._universe_domain = GkeInferenceQuickstartClient._get_universe_domain( - universe_domain_opt, self._universe_domain_env - ) - self._api_endpoint = None # updated below, depending on `transport` - - # Initialize the universe domain validation. - self._is_universe_domain_valid = False - - if CLIENT_LOGGING_SUPPORTED: # pragma: NO COVER - # Setup logging. - client_logging.initialize_logging() - - api_key_value = getattr(self._client_options, "api_key", None) - if api_key_value and credentials: - raise ValueError( - "client_options.api_key and credentials are mutually exclusive" - ) - - # Save or instantiate the transport. - # Ordinarily, we provide the transport, but allowing a custom transport - # instance provides an extensibility point for unusual situations. - transport_provided = isinstance(transport, GkeInferenceQuickstartTransport) - if transport_provided: - # transport is a GkeInferenceQuickstartTransport instance. - if credentials or self._client_options.credentials_file or api_key_value: - raise ValueError( - "When providing a transport instance, " - "provide its credentials directly." - ) - if self._client_options.scopes: - raise ValueError( - "When providing a transport instance, provide its scopes " - "directly." - ) - self._transport = cast(GkeInferenceQuickstartTransport, transport) - self._api_endpoint = self._transport.host - - self._api_endpoint = ( - self._api_endpoint - or GkeInferenceQuickstartClient._get_api_endpoint( - self._client_options.api_endpoint, - self._client_cert_source, - self._universe_domain, - self._use_mtls_endpoint, - ) - ) - - if not transport_provided: - import google.auth._default # type: ignore - - if api_key_value and hasattr( - google.auth._default, "get_api_key_credentials" - ): - credentials = google.auth._default.get_api_key_credentials( - api_key_value - ) - - transport_init: Union[ - Type[GkeInferenceQuickstartTransport], - Callable[..., GkeInferenceQuickstartTransport], - ] = ( - GkeInferenceQuickstartClient.get_transport_class(transport) - if isinstance(transport, str) or transport is None - else cast(Callable[..., GkeInferenceQuickstartTransport], transport) - ) - # initialize with the provided callable or the passed in class - self._transport = transport_init( - credentials=credentials, - credentials_file=self._client_options.credentials_file, - host=self._api_endpoint, - scopes=self._client_options.scopes, - client_cert_source_for_mtls=self._client_cert_source, - quota_project_id=self._client_options.quota_project_id, - client_info=client_info, - always_use_jwt_access=True, - api_audience=self._client_options.api_audience, - ) - - if "async" not in str(self._transport): - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - std_logging.DEBUG - ): # pragma: NO COVER - _LOGGER.debug( - "Created client `google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient`.", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "universeDomain": getattr( - self._transport._credentials, "universe_domain", "" - ), - "credentialsType": f"{type(self._transport._credentials).__module__}.{type(self._transport._credentials).__qualname__}", - "credentialsInfo": getattr( - self.transport._credentials, "get_cred_info", lambda: None - )(), - } - if hasattr(self._transport, "_credentials") - else { - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "credentialsType": None, - }, - ) - - def fetch_models( - self, - request: Optional[Union[gkerecommender.FetchModelsRequest, dict]] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> pagers.FetchModelsPager: - r"""Fetches available models. Open-source models follow the - Huggingface Hub ``owner/model_name`` format. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - def sample_fetch_models(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelsRequest( - ) - - # Make the request - page_result = client.fetch_models(request=request) - - # Handle the response - for response in page_result: - print(response) - - Args: - request (Union[google.cloud.gkerecommender_v1.types.FetchModelsRequest, dict]): - The request object. Request message for - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsPager: - Response message for - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. - - Iterating over this object will yield results and - resolve additional pages automatically. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchModelsRequest): - request = gkerecommender.FetchModelsRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._transport._wrapped_methods[self._transport.fetch_models] - - # Validate the universe domain. - self._validate_universe_domain() - - # Send the request. - response = rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # This method is paged; wrap the response in a pager, which provides - # an `__iter__` convenience method. - response = pagers.FetchModelsPager( - method=rpc, - request=request, - response=response, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - def fetch_model_servers( - self, - request: Optional[Union[gkerecommender.FetchModelServersRequest, dict]] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> pagers.FetchModelServersPager: - r"""Fetches available model servers. Open-source model servers use - simplified, lowercase names (e.g., ``vllm``). - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - def sample_fetch_model_servers(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelServersRequest( - model="model_value", - ) - - # Make the request - page_result = client.fetch_model_servers(request=request) - - # Handle the response - for response in page_result: - print(response) - - Args: - request (Union[google.cloud.gkerecommender_v1.types.FetchModelServersRequest, dict]): - The request object. Request message for - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersPager: - Response message for - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. - - Iterating over this object will yield results and - resolve additional pages automatically. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchModelServersRequest): - request = gkerecommender.FetchModelServersRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._transport._wrapped_methods[self._transport.fetch_model_servers] - - # Validate the universe domain. - self._validate_universe_domain() - - # Send the request. - response = rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # This method is paged; wrap the response in a pager, which provides - # an `__iter__` convenience method. - response = pagers.FetchModelServersPager( - method=rpc, - request=request, - response=response, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - def fetch_model_server_versions( - self, - request: Optional[ - Union[gkerecommender.FetchModelServerVersionsRequest, dict] - ] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> pagers.FetchModelServerVersionsPager: - r"""Fetches available model server versions. Open-source servers use - their own versioning schemas (e.g., ``vllm`` uses semver like - ``v1.0.0``). - - Some model servers have different versioning schemas depending - on the accelerator. For example, ``vllm`` uses semver on GPUs, - but returns nightly build tags on TPUs. All available versions - will be returned when different schemas are present. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - def sample_fetch_model_server_versions(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelServerVersionsRequest( - model="model_value", - model_server="model_server_value", - ) - - # Make the request - page_result = client.fetch_model_server_versions(request=request) - - # Handle the response - for response in page_result: - print(response) - - Args: - request (Union[google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest, dict]): - The request object. Request message for - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsPager: - Response message for - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. - - Iterating over this object will yield results and - resolve additional pages automatically. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchModelServerVersionsRequest): - request = gkerecommender.FetchModelServerVersionsRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._transport._wrapped_methods[ - self._transport.fetch_model_server_versions - ] - - # Validate the universe domain. - self._validate_universe_domain() - - # Send the request. - response = rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # This method is paged; wrap the response in a pager, which provides - # an `__iter__` convenience method. - response = pagers.FetchModelServerVersionsPager( - method=rpc, - request=request, - response=response, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - def fetch_profiles( - self, - request: Optional[Union[gkerecommender.FetchProfilesRequest, dict]] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> pagers.FetchProfilesPager: - r"""Fetches available profiles. A profile contains performance - metrics and cost information for a specific model server setup. - Profiles can be filtered by parameters. If no filters are - provided, all profiles are returned. - - Profiles display a single value per performance metric based on - the provided performance requirements. If no requirements are - given, the metrics represent the inflection point. See `Run best - practice inference with GKE Inference Quickstart - recipes `__ - for details. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - def sample_fetch_profiles(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchProfilesRequest( - ) - - # Make the request - page_result = client.fetch_profiles(request=request) - - # Handle the response - for response in page_result: - print(response) - - Args: - request (Union[google.cloud.gkerecommender_v1.types.FetchProfilesRequest, dict]): - The request object. Request message for - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesPager: - Response message for - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - - Iterating over this object will yield results and - resolve additional pages automatically. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchProfilesRequest): - request = gkerecommender.FetchProfilesRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._transport._wrapped_methods[self._transport.fetch_profiles] - - # Validate the universe domain. - self._validate_universe_domain() - - # Send the request. - response = rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # This method is paged; wrap the response in a pager, which provides - # an `__iter__` convenience method. - response = pagers.FetchProfilesPager( - method=rpc, - request=request, - response=response, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - def generate_optimized_manifest( - self, - request: Optional[ - Union[gkerecommender.GenerateOptimizedManifestRequest, dict] - ] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.GenerateOptimizedManifestResponse: - r"""Generates an optimized deployment manifest for a given model and - model server, based on the specified accelerator, performance - targets, and configurations. See `Run best practice inference - with GKE Inference Quickstart - recipes `__ - for deployment details. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - def sample_generate_optimized_manifest(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - model_server_info = gkerecommender_v1.ModelServerInfo() - model_server_info.model = "model_value" - model_server_info.model_server = "model_server_value" - - request = gkerecommender_v1.GenerateOptimizedManifestRequest( - model_server_info=model_server_info, - accelerator_type="accelerator_type_value", - ) - - # Make the request - response = client.generate_optimized_manifest(request=request) - - # Handle the response - print(response) - - Args: - request (Union[google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest, dict]): - The request object. Request message for - [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse: - Response message for - [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.GenerateOptimizedManifestRequest): - request = gkerecommender.GenerateOptimizedManifestRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._transport._wrapped_methods[ - self._transport.generate_optimized_manifest - ] - - # Validate the universe domain. - self._validate_universe_domain() - - # Send the request. - response = rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - def fetch_benchmarking_data( - self, - request: Optional[ - Union[gkerecommender.FetchBenchmarkingDataRequest, dict] - ] = None, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.FetchBenchmarkingDataResponse: - r"""Fetches all of the benchmarking data available for a - profile. Benchmarking data returns all of the - performance metrics available for a given model server - setup on a given instance type. - - .. code-block:: python - - # This snippet has been automatically generated and should be regarded as a - # code template only. - # It will require modifications to work: - # - It may require correct/in-range values for request initialization. - # - It may require specifying regional endpoints when creating the service - # client as shown in: - # https://googleapis.dev/python/google-api-core/latest/client_options.html - from google.cloud import gkerecommender_v1 - - def sample_fetch_benchmarking_data(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - model_server_info = gkerecommender_v1.ModelServerInfo() - model_server_info.model = "model_value" - model_server_info.model_server = "model_server_value" - - request = gkerecommender_v1.FetchBenchmarkingDataRequest( - model_server_info=model_server_info, - ) - - # Make the request - response = client.fetch_benchmarking_data(request=request) - - # Handle the response - print(response) - - Args: - request (Union[google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest, dict]): - The request object. Request message for - [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse: - Response message for - [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. - - """ - # Create or coerce a protobuf request object. - # - Use the request object if provided (there's no risk of modifying the input as - # there are no flattened fields), or create one. - if not isinstance(request, gkerecommender.FetchBenchmarkingDataRequest): - request = gkerecommender.FetchBenchmarkingDataRequest(request) - - # Wrap the RPC method; this adds retry and timeout information, - # and friendly error handling. - rpc = self._transport._wrapped_methods[self._transport.fetch_benchmarking_data] - - # Validate the universe domain. - self._validate_universe_domain() - - # Send the request. - response = rpc( - request, - retry=retry, - timeout=timeout, - metadata=metadata, - ) - - # Done; return the response. - return response - - def __enter__(self) -> "GkeInferenceQuickstartClient": - return self - - def __exit__(self, type, value, traceback): - """Releases underlying transport's resources. - - .. warning:: - ONLY use as a context manager if the transport is NOT shared - with other clients! Exiting the with block will CLOSE the transport - and may cause errors in other clients! - """ - self.transport.close() - - -DEFAULT_CLIENT_INFO = gapic_v1.client_info.ClientInfo( - gapic_version=package_version.__version__ -) - -if hasattr(DEFAULT_CLIENT_INFO, "protobuf_runtime_version"): # pragma: NO COVER - DEFAULT_CLIENT_INFO.protobuf_runtime_version = google.protobuf.__version__ - -__all__ = ("GkeInferenceQuickstartClient",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/pagers.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/pagers.py deleted file mode 100644 index 6a59cefc936c..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/pagers.py +++ /dev/null @@ -1,669 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from typing import ( - Any, - AsyncIterator, - Awaitable, - Callable, - Iterator, - Optional, - Sequence, - Tuple, - Union, -) - -from google.api_core import gapic_v1 -from google.api_core import retry as retries -from google.api_core import retry_async as retries_async - -try: - OptionalRetry = Union[retries.Retry, gapic_v1.method._MethodDefault, None] - OptionalAsyncRetry = Union[ - retries_async.AsyncRetry, gapic_v1.method._MethodDefault, None - ] -except AttributeError: # pragma: NO COVER - OptionalRetry = Union[retries.Retry, object, None] # type: ignore - OptionalAsyncRetry = Union[retries_async.AsyncRetry, object, None] # type: ignore - -from google.cloud.gkerecommender_v1.types import gkerecommender - - -class FetchModelsPager: - """A pager for iterating through ``fetch_models`` requests. - - This class thinly wraps an initial - :class:`google.cloud.gkerecommender_v1.types.FetchModelsResponse` object, and - provides an ``__iter__`` method to iterate through its - ``models`` field. - - If there are more pages, the ``__iter__`` method will make additional - ``FetchModels`` requests and continue to iterate - through the ``models`` field on the - corresponding responses. - - All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelsResponse` - attributes are available on the pager. If multiple requests are made, only - the most recent response is retained, and thus used for attribute lookup. - """ - - def __init__( - self, - method: Callable[..., gkerecommender.FetchModelsResponse], - request: gkerecommender.FetchModelsRequest, - response: gkerecommender.FetchModelsResponse, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = () - ): - """Instantiate the pager. - - Args: - method (Callable): The method that was originally called, and - which instantiated this pager. - request (google.cloud.gkerecommender_v1.types.FetchModelsRequest): - The initial request object. - response (google.cloud.gkerecommender_v1.types.FetchModelsResponse): - The initial response object. - retry (google.api_core.retry.Retry): Designation of what errors, - if any, should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - """ - self._method = method - self._request = gkerecommender.FetchModelsRequest(request) - self._response = response - self._retry = retry - self._timeout = timeout - self._metadata = metadata - - def __getattr__(self, name: str) -> Any: - return getattr(self._response, name) - - @property - def pages(self) -> Iterator[gkerecommender.FetchModelsResponse]: - yield self._response - while self._response.next_page_token: - self._request.page_token = self._response.next_page_token - self._response = self._method( - self._request, - retry=self._retry, - timeout=self._timeout, - metadata=self._metadata, - ) - yield self._response - - def __iter__(self) -> Iterator[str]: - for page in self.pages: - yield from page.models - - def __repr__(self) -> str: - return "{0}<{1!r}>".format(self.__class__.__name__, self._response) - - -class FetchModelsAsyncPager: - """A pager for iterating through ``fetch_models`` requests. - - This class thinly wraps an initial - :class:`google.cloud.gkerecommender_v1.types.FetchModelsResponse` object, and - provides an ``__aiter__`` method to iterate through its - ``models`` field. - - If there are more pages, the ``__aiter__`` method will make additional - ``FetchModels`` requests and continue to iterate - through the ``models`` field on the - corresponding responses. - - All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelsResponse` - attributes are available on the pager. If multiple requests are made, only - the most recent response is retained, and thus used for attribute lookup. - """ - - def __init__( - self, - method: Callable[..., Awaitable[gkerecommender.FetchModelsResponse]], - request: gkerecommender.FetchModelsRequest, - response: gkerecommender.FetchModelsResponse, - *, - retry: OptionalAsyncRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = () - ): - """Instantiates the pager. - - Args: - method (Callable): The method that was originally called, and - which instantiated this pager. - request (google.cloud.gkerecommender_v1.types.FetchModelsRequest): - The initial request object. - response (google.cloud.gkerecommender_v1.types.FetchModelsResponse): - The initial response object. - retry (google.api_core.retry.AsyncRetry): Designation of what errors, - if any, should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - """ - self._method = method - self._request = gkerecommender.FetchModelsRequest(request) - self._response = response - self._retry = retry - self._timeout = timeout - self._metadata = metadata - - def __getattr__(self, name: str) -> Any: - return getattr(self._response, name) - - @property - async def pages(self) -> AsyncIterator[gkerecommender.FetchModelsResponse]: - yield self._response - while self._response.next_page_token: - self._request.page_token = self._response.next_page_token - self._response = await self._method( - self._request, - retry=self._retry, - timeout=self._timeout, - metadata=self._metadata, - ) - yield self._response - - def __aiter__(self) -> AsyncIterator[str]: - async def async_generator(): - async for page in self.pages: - for response in page.models: - yield response - - return async_generator() - - def __repr__(self) -> str: - return "{0}<{1!r}>".format(self.__class__.__name__, self._response) - - -class FetchModelServersPager: - """A pager for iterating through ``fetch_model_servers`` requests. - - This class thinly wraps an initial - :class:`google.cloud.gkerecommender_v1.types.FetchModelServersResponse` object, and - provides an ``__iter__`` method to iterate through its - ``model_servers`` field. - - If there are more pages, the ``__iter__`` method will make additional - ``FetchModelServers`` requests and continue to iterate - through the ``model_servers`` field on the - corresponding responses. - - All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelServersResponse` - attributes are available on the pager. If multiple requests are made, only - the most recent response is retained, and thus used for attribute lookup. - """ - - def __init__( - self, - method: Callable[..., gkerecommender.FetchModelServersResponse], - request: gkerecommender.FetchModelServersRequest, - response: gkerecommender.FetchModelServersResponse, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = () - ): - """Instantiate the pager. - - Args: - method (Callable): The method that was originally called, and - which instantiated this pager. - request (google.cloud.gkerecommender_v1.types.FetchModelServersRequest): - The initial request object. - response (google.cloud.gkerecommender_v1.types.FetchModelServersResponse): - The initial response object. - retry (google.api_core.retry.Retry): Designation of what errors, - if any, should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - """ - self._method = method - self._request = gkerecommender.FetchModelServersRequest(request) - self._response = response - self._retry = retry - self._timeout = timeout - self._metadata = metadata - - def __getattr__(self, name: str) -> Any: - return getattr(self._response, name) - - @property - def pages(self) -> Iterator[gkerecommender.FetchModelServersResponse]: - yield self._response - while self._response.next_page_token: - self._request.page_token = self._response.next_page_token - self._response = self._method( - self._request, - retry=self._retry, - timeout=self._timeout, - metadata=self._metadata, - ) - yield self._response - - def __iter__(self) -> Iterator[str]: - for page in self.pages: - yield from page.model_servers - - def __repr__(self) -> str: - return "{0}<{1!r}>".format(self.__class__.__name__, self._response) - - -class FetchModelServersAsyncPager: - """A pager for iterating through ``fetch_model_servers`` requests. - - This class thinly wraps an initial - :class:`google.cloud.gkerecommender_v1.types.FetchModelServersResponse` object, and - provides an ``__aiter__`` method to iterate through its - ``model_servers`` field. - - If there are more pages, the ``__aiter__`` method will make additional - ``FetchModelServers`` requests and continue to iterate - through the ``model_servers`` field on the - corresponding responses. - - All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelServersResponse` - attributes are available on the pager. If multiple requests are made, only - the most recent response is retained, and thus used for attribute lookup. - """ - - def __init__( - self, - method: Callable[..., Awaitable[gkerecommender.FetchModelServersResponse]], - request: gkerecommender.FetchModelServersRequest, - response: gkerecommender.FetchModelServersResponse, - *, - retry: OptionalAsyncRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = () - ): - """Instantiates the pager. - - Args: - method (Callable): The method that was originally called, and - which instantiated this pager. - request (google.cloud.gkerecommender_v1.types.FetchModelServersRequest): - The initial request object. - response (google.cloud.gkerecommender_v1.types.FetchModelServersResponse): - The initial response object. - retry (google.api_core.retry.AsyncRetry): Designation of what errors, - if any, should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - """ - self._method = method - self._request = gkerecommender.FetchModelServersRequest(request) - self._response = response - self._retry = retry - self._timeout = timeout - self._metadata = metadata - - def __getattr__(self, name: str) -> Any: - return getattr(self._response, name) - - @property - async def pages(self) -> AsyncIterator[gkerecommender.FetchModelServersResponse]: - yield self._response - while self._response.next_page_token: - self._request.page_token = self._response.next_page_token - self._response = await self._method( - self._request, - retry=self._retry, - timeout=self._timeout, - metadata=self._metadata, - ) - yield self._response - - def __aiter__(self) -> AsyncIterator[str]: - async def async_generator(): - async for page in self.pages: - for response in page.model_servers: - yield response - - return async_generator() - - def __repr__(self) -> str: - return "{0}<{1!r}>".format(self.__class__.__name__, self._response) - - -class FetchModelServerVersionsPager: - """A pager for iterating through ``fetch_model_server_versions`` requests. - - This class thinly wraps an initial - :class:`google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse` object, and - provides an ``__iter__`` method to iterate through its - ``model_server_versions`` field. - - If there are more pages, the ``__iter__`` method will make additional - ``FetchModelServerVersions`` requests and continue to iterate - through the ``model_server_versions`` field on the - corresponding responses. - - All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse` - attributes are available on the pager. If multiple requests are made, only - the most recent response is retained, and thus used for attribute lookup. - """ - - def __init__( - self, - method: Callable[..., gkerecommender.FetchModelServerVersionsResponse], - request: gkerecommender.FetchModelServerVersionsRequest, - response: gkerecommender.FetchModelServerVersionsResponse, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = () - ): - """Instantiate the pager. - - Args: - method (Callable): The method that was originally called, and - which instantiated this pager. - request (google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest): - The initial request object. - response (google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse): - The initial response object. - retry (google.api_core.retry.Retry): Designation of what errors, - if any, should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - """ - self._method = method - self._request = gkerecommender.FetchModelServerVersionsRequest(request) - self._response = response - self._retry = retry - self._timeout = timeout - self._metadata = metadata - - def __getattr__(self, name: str) -> Any: - return getattr(self._response, name) - - @property - def pages(self) -> Iterator[gkerecommender.FetchModelServerVersionsResponse]: - yield self._response - while self._response.next_page_token: - self._request.page_token = self._response.next_page_token - self._response = self._method( - self._request, - retry=self._retry, - timeout=self._timeout, - metadata=self._metadata, - ) - yield self._response - - def __iter__(self) -> Iterator[str]: - for page in self.pages: - yield from page.model_server_versions - - def __repr__(self) -> str: - return "{0}<{1!r}>".format(self.__class__.__name__, self._response) - - -class FetchModelServerVersionsAsyncPager: - """A pager for iterating through ``fetch_model_server_versions`` requests. - - This class thinly wraps an initial - :class:`google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse` object, and - provides an ``__aiter__`` method to iterate through its - ``model_server_versions`` field. - - If there are more pages, the ``__aiter__`` method will make additional - ``FetchModelServerVersions`` requests and continue to iterate - through the ``model_server_versions`` field on the - corresponding responses. - - All the usual :class:`google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse` - attributes are available on the pager. If multiple requests are made, only - the most recent response is retained, and thus used for attribute lookup. - """ - - def __init__( - self, - method: Callable[ - ..., Awaitable[gkerecommender.FetchModelServerVersionsResponse] - ], - request: gkerecommender.FetchModelServerVersionsRequest, - response: gkerecommender.FetchModelServerVersionsResponse, - *, - retry: OptionalAsyncRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = () - ): - """Instantiates the pager. - - Args: - method (Callable): The method that was originally called, and - which instantiated this pager. - request (google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest): - The initial request object. - response (google.cloud.gkerecommender_v1.types.FetchModelServerVersionsResponse): - The initial response object. - retry (google.api_core.retry.AsyncRetry): Designation of what errors, - if any, should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - """ - self._method = method - self._request = gkerecommender.FetchModelServerVersionsRequest(request) - self._response = response - self._retry = retry - self._timeout = timeout - self._metadata = metadata - - def __getattr__(self, name: str) -> Any: - return getattr(self._response, name) - - @property - async def pages( - self, - ) -> AsyncIterator[gkerecommender.FetchModelServerVersionsResponse]: - yield self._response - while self._response.next_page_token: - self._request.page_token = self._response.next_page_token - self._response = await self._method( - self._request, - retry=self._retry, - timeout=self._timeout, - metadata=self._metadata, - ) - yield self._response - - def __aiter__(self) -> AsyncIterator[str]: - async def async_generator(): - async for page in self.pages: - for response in page.model_server_versions: - yield response - - return async_generator() - - def __repr__(self) -> str: - return "{0}<{1!r}>".format(self.__class__.__name__, self._response) - - -class FetchProfilesPager: - """A pager for iterating through ``fetch_profiles`` requests. - - This class thinly wraps an initial - :class:`google.cloud.gkerecommender_v1.types.FetchProfilesResponse` object, and - provides an ``__iter__`` method to iterate through its - ``profile`` field. - - If there are more pages, the ``__iter__`` method will make additional - ``FetchProfiles`` requests and continue to iterate - through the ``profile`` field on the - corresponding responses. - - All the usual :class:`google.cloud.gkerecommender_v1.types.FetchProfilesResponse` - attributes are available on the pager. If multiple requests are made, only - the most recent response is retained, and thus used for attribute lookup. - """ - - def __init__( - self, - method: Callable[..., gkerecommender.FetchProfilesResponse], - request: gkerecommender.FetchProfilesRequest, - response: gkerecommender.FetchProfilesResponse, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = () - ): - """Instantiate the pager. - - Args: - method (Callable): The method that was originally called, and - which instantiated this pager. - request (google.cloud.gkerecommender_v1.types.FetchProfilesRequest): - The initial request object. - response (google.cloud.gkerecommender_v1.types.FetchProfilesResponse): - The initial response object. - retry (google.api_core.retry.Retry): Designation of what errors, - if any, should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - """ - self._method = method - self._request = gkerecommender.FetchProfilesRequest(request) - self._response = response - self._retry = retry - self._timeout = timeout - self._metadata = metadata - - def __getattr__(self, name: str) -> Any: - return getattr(self._response, name) - - @property - def pages(self) -> Iterator[gkerecommender.FetchProfilesResponse]: - yield self._response - while self._response.next_page_token: - self._request.page_token = self._response.next_page_token - self._response = self._method( - self._request, - retry=self._retry, - timeout=self._timeout, - metadata=self._metadata, - ) - yield self._response - - def __iter__(self) -> Iterator[gkerecommender.Profile]: - for page in self.pages: - yield from page.profile - - def __repr__(self) -> str: - return "{0}<{1!r}>".format(self.__class__.__name__, self._response) - - -class FetchProfilesAsyncPager: - """A pager for iterating through ``fetch_profiles`` requests. - - This class thinly wraps an initial - :class:`google.cloud.gkerecommender_v1.types.FetchProfilesResponse` object, and - provides an ``__aiter__`` method to iterate through its - ``profile`` field. - - If there are more pages, the ``__aiter__`` method will make additional - ``FetchProfiles`` requests and continue to iterate - through the ``profile`` field on the - corresponding responses. - - All the usual :class:`google.cloud.gkerecommender_v1.types.FetchProfilesResponse` - attributes are available on the pager. If multiple requests are made, only - the most recent response is retained, and thus used for attribute lookup. - """ - - def __init__( - self, - method: Callable[..., Awaitable[gkerecommender.FetchProfilesResponse]], - request: gkerecommender.FetchProfilesRequest, - response: gkerecommender.FetchProfilesResponse, - *, - retry: OptionalAsyncRetry = gapic_v1.method.DEFAULT, - timeout: Union[float, object] = gapic_v1.method.DEFAULT, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = () - ): - """Instantiates the pager. - - Args: - method (Callable): The method that was originally called, and - which instantiated this pager. - request (google.cloud.gkerecommender_v1.types.FetchProfilesRequest): - The initial request object. - response (google.cloud.gkerecommender_v1.types.FetchProfilesResponse): - The initial response object. - retry (google.api_core.retry.AsyncRetry): Designation of what errors, - if any, should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - """ - self._method = method - self._request = gkerecommender.FetchProfilesRequest(request) - self._response = response - self._retry = retry - self._timeout = timeout - self._metadata = metadata - - def __getattr__(self, name: str) -> Any: - return getattr(self._response, name) - - @property - async def pages(self) -> AsyncIterator[gkerecommender.FetchProfilesResponse]: - yield self._response - while self._response.next_page_token: - self._request.page_token = self._response.next_page_token - self._response = await self._method( - self._request, - retry=self._retry, - timeout=self._timeout, - metadata=self._metadata, - ) - yield self._response - - def __aiter__(self) -> AsyncIterator[gkerecommender.Profile]: - async def async_generator(): - async for page in self.pages: - for response in page.profile: - yield response - - return async_generator() - - def __repr__(self) -> str: - return "{0}<{1!r}>".format(self.__class__.__name__, self._response) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/README.rst b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/README.rst deleted file mode 100644 index bbcb4a6af96c..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/README.rst +++ /dev/null @@ -1,9 +0,0 @@ - -transport inheritance structure -_______________________________ - -`GkeInferenceQuickstartTransport` is the ABC for all transports. -- public child `GkeInferenceQuickstartGrpcTransport` for sync gRPC transport (defined in `grpc.py`). -- public child `GkeInferenceQuickstartGrpcAsyncIOTransport` for async gRPC transport (defined in `grpc_asyncio.py`). -- private child `_BaseGkeInferenceQuickstartRestTransport` for base REST transport with inner classes `_BaseMETHOD` (defined in `rest_base.py`). -- public child `GkeInferenceQuickstartRestTransport` for sync REST transport with inner classes `METHOD` derived from the parent's corresponding `_BaseMETHOD` classes (defined in `rest.py`). diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/__init__.py deleted file mode 100644 index 4163ed5a0753..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/__init__.py +++ /dev/null @@ -1,41 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from collections import OrderedDict -from typing import Dict, Type - -from .base import GkeInferenceQuickstartTransport -from .grpc import GkeInferenceQuickstartGrpcTransport -from .grpc_asyncio import GkeInferenceQuickstartGrpcAsyncIOTransport -from .rest import ( - GkeInferenceQuickstartRestInterceptor, - GkeInferenceQuickstartRestTransport, -) - -# Compile a registry of transports. -_transport_registry = ( - OrderedDict() -) # type: Dict[str, Type[GkeInferenceQuickstartTransport]] -_transport_registry["grpc"] = GkeInferenceQuickstartGrpcTransport -_transport_registry["grpc_asyncio"] = GkeInferenceQuickstartGrpcAsyncIOTransport -_transport_registry["rest"] = GkeInferenceQuickstartRestTransport - -__all__ = ( - "GkeInferenceQuickstartTransport", - "GkeInferenceQuickstartGrpcTransport", - "GkeInferenceQuickstartGrpcAsyncIOTransport", - "GkeInferenceQuickstartRestTransport", - "GkeInferenceQuickstartRestInterceptor", -) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/base.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/base.py deleted file mode 100644 index 405081650eee..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/base.py +++ /dev/null @@ -1,253 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import abc -from typing import Awaitable, Callable, Dict, Optional, Sequence, Union - -import google.api_core -from google.api_core import exceptions as core_exceptions -from google.api_core import gapic_v1 -from google.api_core import retry as retries -import google.auth # type: ignore -from google.auth import credentials as ga_credentials # type: ignore -from google.oauth2 import service_account # type: ignore -import google.protobuf - -from google.cloud.gkerecommender_v1 import gapic_version as package_version -from google.cloud.gkerecommender_v1.types import gkerecommender - -DEFAULT_CLIENT_INFO = gapic_v1.client_info.ClientInfo( - gapic_version=package_version.__version__ -) - -if hasattr(DEFAULT_CLIENT_INFO, "protobuf_runtime_version"): # pragma: NO COVER - DEFAULT_CLIENT_INFO.protobuf_runtime_version = google.protobuf.__version__ - - -class GkeInferenceQuickstartTransport(abc.ABC): - """Abstract transport class for GkeInferenceQuickstart.""" - - AUTH_SCOPES = ("https://www.googleapis.com/auth/cloud-platform",) - - DEFAULT_HOST: str = "gkerecommender.googleapis.com" - - def __init__( - self, - *, - host: str = DEFAULT_HOST, - credentials: Optional[ga_credentials.Credentials] = None, - credentials_file: Optional[str] = None, - scopes: Optional[Sequence[str]] = None, - quota_project_id: Optional[str] = None, - client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, - always_use_jwt_access: Optional[bool] = False, - api_audience: Optional[str] = None, - **kwargs, - ) -> None: - """Instantiate the transport. - - Args: - host (Optional[str]): - The hostname to connect to (default: 'gkerecommender.googleapis.com'). - credentials (Optional[google.auth.credentials.Credentials]): The - authorization credentials to attach to requests. These - credentials identify the application to the service; if none - are specified, the client will attempt to ascertain the - credentials from the environment. - credentials_file (Optional[str]): A file with credentials that can - be loaded with :func:`google.auth.load_credentials_from_file`. - This argument is mutually exclusive with credentials. - scopes (Optional[Sequence[str]]): A list of scopes. - quota_project_id (Optional[str]): An optional project to use for billing - and quota. - client_info (google.api_core.gapic_v1.client_info.ClientInfo): - The client info used to send a user-agent string along with - API requests. If ``None``, then default info will be used. - Generally, you only need to set this if you're developing - your own client library. - always_use_jwt_access (Optional[bool]): Whether self signed JWT should - be used for service account credentials. - """ - - scopes_kwargs = {"scopes": scopes, "default_scopes": self.AUTH_SCOPES} - - # Save the scopes. - self._scopes = scopes - if not hasattr(self, "_ignore_credentials"): - self._ignore_credentials: bool = False - - # If no credentials are provided, then determine the appropriate - # defaults. - if credentials and credentials_file: - raise core_exceptions.DuplicateCredentialArgs( - "'credentials_file' and 'credentials' are mutually exclusive" - ) - - if credentials_file is not None: - credentials, _ = google.auth.load_credentials_from_file( - credentials_file, **scopes_kwargs, quota_project_id=quota_project_id - ) - elif credentials is None and not self._ignore_credentials: - credentials, _ = google.auth.default( - **scopes_kwargs, quota_project_id=quota_project_id - ) - # Don't apply audience if the credentials file passed from user. - if hasattr(credentials, "with_gdch_audience"): - credentials = credentials.with_gdch_audience( - api_audience if api_audience else host - ) - - # If the credentials are service account credentials, then always try to use self signed JWT. - if ( - always_use_jwt_access - and isinstance(credentials, service_account.Credentials) - and hasattr(service_account.Credentials, "with_always_use_jwt_access") - ): - credentials = credentials.with_always_use_jwt_access(True) - - # Save the credentials. - self._credentials = credentials - - # Save the hostname. Default to port 443 (HTTPS) if none is specified. - if ":" not in host: - host += ":443" - self._host = host - - @property - def host(self): - return self._host - - def _prep_wrapped_messages(self, client_info): - # Precompute the wrapped methods. - self._wrapped_methods = { - self.fetch_models: gapic_v1.method.wrap_method( - self.fetch_models, - default_timeout=60.0, - client_info=client_info, - ), - self.fetch_model_servers: gapic_v1.method.wrap_method( - self.fetch_model_servers, - default_timeout=60.0, - client_info=client_info, - ), - self.fetch_model_server_versions: gapic_v1.method.wrap_method( - self.fetch_model_server_versions, - default_timeout=60.0, - client_info=client_info, - ), - self.fetch_profiles: gapic_v1.method.wrap_method( - self.fetch_profiles, - default_timeout=60.0, - client_info=client_info, - ), - self.generate_optimized_manifest: gapic_v1.method.wrap_method( - self.generate_optimized_manifest, - default_timeout=60.0, - client_info=client_info, - ), - self.fetch_benchmarking_data: gapic_v1.method.wrap_method( - self.fetch_benchmarking_data, - default_timeout=60.0, - client_info=client_info, - ), - } - - def close(self): - """Closes resources associated with the transport. - - .. warning:: - Only call this method if the transport is NOT shared - with other clients - this may cause errors in other clients! - """ - raise NotImplementedError() - - @property - def fetch_models( - self, - ) -> Callable[ - [gkerecommender.FetchModelsRequest], - Union[ - gkerecommender.FetchModelsResponse, - Awaitable[gkerecommender.FetchModelsResponse], - ], - ]: - raise NotImplementedError() - - @property - def fetch_model_servers( - self, - ) -> Callable[ - [gkerecommender.FetchModelServersRequest], - Union[ - gkerecommender.FetchModelServersResponse, - Awaitable[gkerecommender.FetchModelServersResponse], - ], - ]: - raise NotImplementedError() - - @property - def fetch_model_server_versions( - self, - ) -> Callable[ - [gkerecommender.FetchModelServerVersionsRequest], - Union[ - gkerecommender.FetchModelServerVersionsResponse, - Awaitable[gkerecommender.FetchModelServerVersionsResponse], - ], - ]: - raise NotImplementedError() - - @property - def fetch_profiles( - self, - ) -> Callable[ - [gkerecommender.FetchProfilesRequest], - Union[ - gkerecommender.FetchProfilesResponse, - Awaitable[gkerecommender.FetchProfilesResponse], - ], - ]: - raise NotImplementedError() - - @property - def generate_optimized_manifest( - self, - ) -> Callable[ - [gkerecommender.GenerateOptimizedManifestRequest], - Union[ - gkerecommender.GenerateOptimizedManifestResponse, - Awaitable[gkerecommender.GenerateOptimizedManifestResponse], - ], - ]: - raise NotImplementedError() - - @property - def fetch_benchmarking_data( - self, - ) -> Callable[ - [gkerecommender.FetchBenchmarkingDataRequest], - Union[ - gkerecommender.FetchBenchmarkingDataResponse, - Awaitable[gkerecommender.FetchBenchmarkingDataResponse], - ], - ]: - raise NotImplementedError() - - @property - def kind(self) -> str: - raise NotImplementedError() - - -__all__ = ("GkeInferenceQuickstartTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc.py deleted file mode 100644 index f36abcfd5789..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc.py +++ /dev/null @@ -1,536 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import json -import logging as std_logging -import pickle -from typing import Callable, Dict, Optional, Sequence, Tuple, Union -import warnings - -from google.api_core import gapic_v1, grpc_helpers -import google.auth # type: ignore -from google.auth import credentials as ga_credentials # type: ignore -from google.auth.transport.grpc import SslCredentials # type: ignore -from google.protobuf.json_format import MessageToJson -import google.protobuf.message -import grpc # type: ignore -import proto # type: ignore - -from google.cloud.gkerecommender_v1.types import gkerecommender - -from .base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport - -try: - from google.api_core import client_logging # type: ignore - - CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER -except ImportError: # pragma: NO COVER - CLIENT_LOGGING_SUPPORTED = False - -_LOGGER = std_logging.getLogger(__name__) - - -class _LoggingClientInterceptor(grpc.UnaryUnaryClientInterceptor): # pragma: NO COVER - def intercept_unary_unary(self, continuation, client_call_details, request): - logging_enabled = CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - std_logging.DEBUG - ) - if logging_enabled: # pragma: NO COVER - request_metadata = client_call_details.metadata - if isinstance(request, proto.Message): - request_payload = type(request).to_json(request) - elif isinstance(request, google.protobuf.message.Message): - request_payload = MessageToJson(request) - else: - request_payload = f"{type(request).__name__}: {pickle.dumps(request)}" - - request_metadata = { - key: value.decode("utf-8") if isinstance(value, bytes) else value - for key, value in request_metadata - } - grpc_request = { - "payload": request_payload, - "requestMethod": "grpc", - "metadata": dict(request_metadata), - } - _LOGGER.debug( - f"Sending request for {client_call_details.method}", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": str(client_call_details.method), - "request": grpc_request, - "metadata": grpc_request["metadata"], - }, - ) - response = continuation(client_call_details, request) - if logging_enabled: # pragma: NO COVER - response_metadata = response.trailing_metadata() - # Convert gRPC metadata `` to list of tuples - metadata = ( - dict([(k, str(v)) for k, v in response_metadata]) - if response_metadata - else None - ) - result = response.result() - if isinstance(result, proto.Message): - response_payload = type(result).to_json(result) - elif isinstance(result, google.protobuf.message.Message): - response_payload = MessageToJson(result) - else: - response_payload = f"{type(result).__name__}: {pickle.dumps(result)}" - grpc_response = { - "payload": response_payload, - "metadata": metadata, - "status": "OK", - } - _LOGGER.debug( - f"Received response for {client_call_details.method}.", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": client_call_details.method, - "response": grpc_response, - "metadata": grpc_response["metadata"], - }, - ) - return response - - -class GkeInferenceQuickstartGrpcTransport(GkeInferenceQuickstartTransport): - """gRPC backend transport for GkeInferenceQuickstart. - - GKE Inference Quickstart (GIQ) service provides profiles with - performance metrics for popular models and model servers across - multiple accelerators. These profiles help generate optimized - best practices for running inference on GKE. - - This class defines the same methods as the primary client, so the - primary client can load the underlying transport implementation - and call it. - - It sends protocol buffers over the wire using gRPC (which is built on - top of HTTP/2); the ``grpcio`` package must be installed. - """ - - _stubs: Dict[str, Callable] - - def __init__( - self, - *, - host: str = "gkerecommender.googleapis.com", - credentials: Optional[ga_credentials.Credentials] = None, - credentials_file: Optional[str] = None, - scopes: Optional[Sequence[str]] = None, - channel: Optional[Union[grpc.Channel, Callable[..., grpc.Channel]]] = None, - api_mtls_endpoint: Optional[str] = None, - client_cert_source: Optional[Callable[[], Tuple[bytes, bytes]]] = None, - ssl_channel_credentials: Optional[grpc.ChannelCredentials] = None, - client_cert_source_for_mtls: Optional[Callable[[], Tuple[bytes, bytes]]] = None, - quota_project_id: Optional[str] = None, - client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, - always_use_jwt_access: Optional[bool] = False, - api_audience: Optional[str] = None, - ) -> None: - """Instantiate the transport. - - Args: - host (Optional[str]): - The hostname to connect to (default: 'gkerecommender.googleapis.com'). - credentials (Optional[google.auth.credentials.Credentials]): The - authorization credentials to attach to requests. These - credentials identify the application to the service; if none - are specified, the client will attempt to ascertain the - credentials from the environment. - This argument is ignored if a ``channel`` instance is provided. - credentials_file (Optional[str]): A file with credentials that can - be loaded with :func:`google.auth.load_credentials_from_file`. - This argument is ignored if a ``channel`` instance is provided. - scopes (Optional(Sequence[str])): A list of scopes. This argument is - ignored if a ``channel`` instance is provided. - channel (Optional[Union[grpc.Channel, Callable[..., grpc.Channel]]]): - A ``Channel`` instance through which to make calls, or a Callable - that constructs and returns one. If set to None, ``self.create_channel`` - is used to create the channel. If a Callable is given, it will be called - with the same arguments as used in ``self.create_channel``. - api_mtls_endpoint (Optional[str]): Deprecated. The mutual TLS endpoint. - If provided, it overrides the ``host`` argument and tries to create - a mutual TLS channel with client SSL credentials from - ``client_cert_source`` or application default SSL credentials. - client_cert_source (Optional[Callable[[], Tuple[bytes, bytes]]]): - Deprecated. A callback to provide client SSL certificate bytes and - private key bytes, both in PEM format. It is ignored if - ``api_mtls_endpoint`` is None. - ssl_channel_credentials (grpc.ChannelCredentials): SSL credentials - for the grpc channel. It is ignored if a ``channel`` instance is provided. - client_cert_source_for_mtls (Optional[Callable[[], Tuple[bytes, bytes]]]): - A callback to provide client certificate bytes and private key bytes, - both in PEM format. It is used to configure a mutual TLS channel. It is - ignored if a ``channel`` instance or ``ssl_channel_credentials`` is provided. - quota_project_id (Optional[str]): An optional project to use for billing - and quota. - client_info (google.api_core.gapic_v1.client_info.ClientInfo): - The client info used to send a user-agent string along with - API requests. If ``None``, then default info will be used. - Generally, you only need to set this if you're developing - your own client library. - always_use_jwt_access (Optional[bool]): Whether self signed JWT should - be used for service account credentials. - - Raises: - google.auth.exceptions.MutualTLSChannelError: If mutual TLS transport - creation failed for any reason. - google.api_core.exceptions.DuplicateCredentialArgs: If both ``credentials`` - and ``credentials_file`` are passed. - """ - self._grpc_channel = None - self._ssl_channel_credentials = ssl_channel_credentials - self._stubs: Dict[str, Callable] = {} - - if api_mtls_endpoint: - warnings.warn("api_mtls_endpoint is deprecated", DeprecationWarning) - if client_cert_source: - warnings.warn("client_cert_source is deprecated", DeprecationWarning) - - if isinstance(channel, grpc.Channel): - # Ignore credentials if a channel was passed. - credentials = None - self._ignore_credentials = True - # If a channel was explicitly provided, set it. - self._grpc_channel = channel - self._ssl_channel_credentials = None - - else: - if api_mtls_endpoint: - host = api_mtls_endpoint - - # Create SSL credentials with client_cert_source or application - # default SSL credentials. - if client_cert_source: - cert, key = client_cert_source() - self._ssl_channel_credentials = grpc.ssl_channel_credentials( - certificate_chain=cert, private_key=key - ) - else: - self._ssl_channel_credentials = SslCredentials().ssl_credentials - - else: - if client_cert_source_for_mtls and not ssl_channel_credentials: - cert, key = client_cert_source_for_mtls() - self._ssl_channel_credentials = grpc.ssl_channel_credentials( - certificate_chain=cert, private_key=key - ) - - # The base transport sets the host, credentials and scopes - super().__init__( - host=host, - credentials=credentials, - credentials_file=credentials_file, - scopes=scopes, - quota_project_id=quota_project_id, - client_info=client_info, - always_use_jwt_access=always_use_jwt_access, - api_audience=api_audience, - ) - - if not self._grpc_channel: - # initialize with the provided callable or the default channel - channel_init = channel or type(self).create_channel - self._grpc_channel = channel_init( - self._host, - # use the credentials which are saved - credentials=self._credentials, - # Set ``credentials_file`` to ``None`` here as - # the credentials that we saved earlier should be used. - credentials_file=None, - scopes=self._scopes, - ssl_credentials=self._ssl_channel_credentials, - quota_project_id=quota_project_id, - options=[ - ("grpc.max_send_message_length", -1), - ("grpc.max_receive_message_length", -1), - ], - ) - - self._interceptor = _LoggingClientInterceptor() - self._logged_channel = grpc.intercept_channel( - self._grpc_channel, self._interceptor - ) - - # Wrap messages. This must be done after self._logged_channel exists - self._prep_wrapped_messages(client_info) - - @classmethod - def create_channel( - cls, - host: str = "gkerecommender.googleapis.com", - credentials: Optional[ga_credentials.Credentials] = None, - credentials_file: Optional[str] = None, - scopes: Optional[Sequence[str]] = None, - quota_project_id: Optional[str] = None, - **kwargs, - ) -> grpc.Channel: - """Create and return a gRPC channel object. - Args: - host (Optional[str]): The host for the channel to use. - credentials (Optional[~.Credentials]): The - authorization credentials to attach to requests. These - credentials identify this application to the service. If - none are specified, the client will attempt to ascertain - the credentials from the environment. - credentials_file (Optional[str]): A file with credentials that can - be loaded with :func:`google.auth.load_credentials_from_file`. - This argument is mutually exclusive with credentials. - scopes (Optional[Sequence[str]]): A optional list of scopes needed for this - service. These are only used when credentials are not specified and - are passed to :func:`google.auth.default`. - quota_project_id (Optional[str]): An optional project to use for billing - and quota. - kwargs (Optional[dict]): Keyword arguments, which are passed to the - channel creation. - Returns: - grpc.Channel: A gRPC channel object. - - Raises: - google.api_core.exceptions.DuplicateCredentialArgs: If both ``credentials`` - and ``credentials_file`` are passed. - """ - - return grpc_helpers.create_channel( - host, - credentials=credentials, - credentials_file=credentials_file, - quota_project_id=quota_project_id, - default_scopes=cls.AUTH_SCOPES, - scopes=scopes, - default_host=cls.DEFAULT_HOST, - **kwargs, - ) - - @property - def grpc_channel(self) -> grpc.Channel: - """Return the channel designed to connect to this service.""" - return self._grpc_channel - - @property - def fetch_models( - self, - ) -> Callable[ - [gkerecommender.FetchModelsRequest], gkerecommender.FetchModelsResponse - ]: - r"""Return a callable for the fetch models method over gRPC. - - Fetches available models. Open-source models follow the - Huggingface Hub ``owner/model_name`` format. - - Returns: - Callable[[~.FetchModelsRequest], - ~.FetchModelsResponse]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_models" not in self._stubs: - self._stubs["fetch_models"] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModels", - request_serializer=gkerecommender.FetchModelsRequest.serialize, - response_deserializer=gkerecommender.FetchModelsResponse.deserialize, - ) - return self._stubs["fetch_models"] - - @property - def fetch_model_servers( - self, - ) -> Callable[ - [gkerecommender.FetchModelServersRequest], - gkerecommender.FetchModelServersResponse, - ]: - r"""Return a callable for the fetch model servers method over gRPC. - - Fetches available model servers. Open-source model servers use - simplified, lowercase names (e.g., ``vllm``). - - Returns: - Callable[[~.FetchModelServersRequest], - ~.FetchModelServersResponse]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_model_servers" not in self._stubs: - self._stubs["fetch_model_servers"] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModelServers", - request_serializer=gkerecommender.FetchModelServersRequest.serialize, - response_deserializer=gkerecommender.FetchModelServersResponse.deserialize, - ) - return self._stubs["fetch_model_servers"] - - @property - def fetch_model_server_versions( - self, - ) -> Callable[ - [gkerecommender.FetchModelServerVersionsRequest], - gkerecommender.FetchModelServerVersionsResponse, - ]: - r"""Return a callable for the fetch model server versions method over gRPC. - - Fetches available model server versions. Open-source servers use - their own versioning schemas (e.g., ``vllm`` uses semver like - ``v1.0.0``). - - Some model servers have different versioning schemas depending - on the accelerator. For example, ``vllm`` uses semver on GPUs, - but returns nightly build tags on TPUs. All available versions - will be returned when different schemas are present. - - Returns: - Callable[[~.FetchModelServerVersionsRequest], - ~.FetchModelServerVersionsResponse]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_model_server_versions" not in self._stubs: - self._stubs[ - "fetch_model_server_versions" - ] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModelServerVersions", - request_serializer=gkerecommender.FetchModelServerVersionsRequest.serialize, - response_deserializer=gkerecommender.FetchModelServerVersionsResponse.deserialize, - ) - return self._stubs["fetch_model_server_versions"] - - @property - def fetch_profiles( - self, - ) -> Callable[ - [gkerecommender.FetchProfilesRequest], gkerecommender.FetchProfilesResponse - ]: - r"""Return a callable for the fetch profiles method over gRPC. - - Fetches available profiles. A profile contains performance - metrics and cost information for a specific model server setup. - Profiles can be filtered by parameters. If no filters are - provided, all profiles are returned. - - Profiles display a single value per performance metric based on - the provided performance requirements. If no requirements are - given, the metrics represent the inflection point. See `Run best - practice inference with GKE Inference Quickstart - recipes `__ - for details. - - Returns: - Callable[[~.FetchProfilesRequest], - ~.FetchProfilesResponse]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_profiles" not in self._stubs: - self._stubs["fetch_profiles"] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchProfiles", - request_serializer=gkerecommender.FetchProfilesRequest.serialize, - response_deserializer=gkerecommender.FetchProfilesResponse.deserialize, - ) - return self._stubs["fetch_profiles"] - - @property - def generate_optimized_manifest( - self, - ) -> Callable[ - [gkerecommender.GenerateOptimizedManifestRequest], - gkerecommender.GenerateOptimizedManifestResponse, - ]: - r"""Return a callable for the generate optimized manifest method over gRPC. - - Generates an optimized deployment manifest for a given model and - model server, based on the specified accelerator, performance - targets, and configurations. See `Run best practice inference - with GKE Inference Quickstart - recipes `__ - for deployment details. - - Returns: - Callable[[~.GenerateOptimizedManifestRequest], - ~.GenerateOptimizedManifestResponse]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "generate_optimized_manifest" not in self._stubs: - self._stubs[ - "generate_optimized_manifest" - ] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/GenerateOptimizedManifest", - request_serializer=gkerecommender.GenerateOptimizedManifestRequest.serialize, - response_deserializer=gkerecommender.GenerateOptimizedManifestResponse.deserialize, - ) - return self._stubs["generate_optimized_manifest"] - - @property - def fetch_benchmarking_data( - self, - ) -> Callable[ - [gkerecommender.FetchBenchmarkingDataRequest], - gkerecommender.FetchBenchmarkingDataResponse, - ]: - r"""Return a callable for the fetch benchmarking data method over gRPC. - - Fetches all of the benchmarking data available for a - profile. Benchmarking data returns all of the - performance metrics available for a given model server - setup on a given instance type. - - Returns: - Callable[[~.FetchBenchmarkingDataRequest], - ~.FetchBenchmarkingDataResponse]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_benchmarking_data" not in self._stubs: - self._stubs["fetch_benchmarking_data"] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchBenchmarkingData", - request_serializer=gkerecommender.FetchBenchmarkingDataRequest.serialize, - response_deserializer=gkerecommender.FetchBenchmarkingDataResponse.deserialize, - ) - return self._stubs["fetch_benchmarking_data"] - - def close(self): - self._logged_channel.close() - - @property - def kind(self) -> str: - return "grpc" - - -__all__ = ("GkeInferenceQuickstartGrpcTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc_asyncio.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc_asyncio.py deleted file mode 100644 index 3a719e36d2fc..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/grpc_asyncio.py +++ /dev/null @@ -1,586 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import inspect -import json -import logging as std_logging -import pickle -from typing import Awaitable, Callable, Dict, Optional, Sequence, Tuple, Union -import warnings - -from google.api_core import exceptions as core_exceptions -from google.api_core import gapic_v1, grpc_helpers_async -from google.api_core import retry_async as retries -from google.auth import credentials as ga_credentials # type: ignore -from google.auth.transport.grpc import SslCredentials # type: ignore -from google.protobuf.json_format import MessageToJson -import google.protobuf.message -import grpc # type: ignore -from grpc.experimental import aio # type: ignore -import proto # type: ignore - -from google.cloud.gkerecommender_v1.types import gkerecommender - -from .base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport -from .grpc import GkeInferenceQuickstartGrpcTransport - -try: - from google.api_core import client_logging # type: ignore - - CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER -except ImportError: # pragma: NO COVER - CLIENT_LOGGING_SUPPORTED = False - -_LOGGER = std_logging.getLogger(__name__) - - -class _LoggingClientAIOInterceptor( - grpc.aio.UnaryUnaryClientInterceptor -): # pragma: NO COVER - async def intercept_unary_unary(self, continuation, client_call_details, request): - logging_enabled = CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - std_logging.DEBUG - ) - if logging_enabled: # pragma: NO COVER - request_metadata = client_call_details.metadata - if isinstance(request, proto.Message): - request_payload = type(request).to_json(request) - elif isinstance(request, google.protobuf.message.Message): - request_payload = MessageToJson(request) - else: - request_payload = f"{type(request).__name__}: {pickle.dumps(request)}" - - request_metadata = { - key: value.decode("utf-8") if isinstance(value, bytes) else value - for key, value in request_metadata - } - grpc_request = { - "payload": request_payload, - "requestMethod": "grpc", - "metadata": dict(request_metadata), - } - _LOGGER.debug( - f"Sending request for {client_call_details.method}", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": str(client_call_details.method), - "request": grpc_request, - "metadata": grpc_request["metadata"], - }, - ) - response = await continuation(client_call_details, request) - if logging_enabled: # pragma: NO COVER - response_metadata = await response.trailing_metadata() - # Convert gRPC metadata `` to list of tuples - metadata = ( - dict([(k, str(v)) for k, v in response_metadata]) - if response_metadata - else None - ) - result = await response - if isinstance(result, proto.Message): - response_payload = type(result).to_json(result) - elif isinstance(result, google.protobuf.message.Message): - response_payload = MessageToJson(result) - else: - response_payload = f"{type(result).__name__}: {pickle.dumps(result)}" - grpc_response = { - "payload": response_payload, - "metadata": metadata, - "status": "OK", - } - _LOGGER.debug( - f"Received response to rpc {client_call_details.method}.", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": str(client_call_details.method), - "response": grpc_response, - "metadata": grpc_response["metadata"], - }, - ) - return response - - -class GkeInferenceQuickstartGrpcAsyncIOTransport(GkeInferenceQuickstartTransport): - """gRPC AsyncIO backend transport for GkeInferenceQuickstart. - - GKE Inference Quickstart (GIQ) service provides profiles with - performance metrics for popular models and model servers across - multiple accelerators. These profiles help generate optimized - best practices for running inference on GKE. - - This class defines the same methods as the primary client, so the - primary client can load the underlying transport implementation - and call it. - - It sends protocol buffers over the wire using gRPC (which is built on - top of HTTP/2); the ``grpcio`` package must be installed. - """ - - _grpc_channel: aio.Channel - _stubs: Dict[str, Callable] = {} - - @classmethod - def create_channel( - cls, - host: str = "gkerecommender.googleapis.com", - credentials: Optional[ga_credentials.Credentials] = None, - credentials_file: Optional[str] = None, - scopes: Optional[Sequence[str]] = None, - quota_project_id: Optional[str] = None, - **kwargs, - ) -> aio.Channel: - """Create and return a gRPC AsyncIO channel object. - Args: - host (Optional[str]): The host for the channel to use. - credentials (Optional[~.Credentials]): The - authorization credentials to attach to requests. These - credentials identify this application to the service. If - none are specified, the client will attempt to ascertain - the credentials from the environment. - credentials_file (Optional[str]): A file with credentials that can - be loaded with :func:`google.auth.load_credentials_from_file`. - scopes (Optional[Sequence[str]]): A optional list of scopes needed for this - service. These are only used when credentials are not specified and - are passed to :func:`google.auth.default`. - quota_project_id (Optional[str]): An optional project to use for billing - and quota. - kwargs (Optional[dict]): Keyword arguments, which are passed to the - channel creation. - Returns: - aio.Channel: A gRPC AsyncIO channel object. - """ - - return grpc_helpers_async.create_channel( - host, - credentials=credentials, - credentials_file=credentials_file, - quota_project_id=quota_project_id, - default_scopes=cls.AUTH_SCOPES, - scopes=scopes, - default_host=cls.DEFAULT_HOST, - **kwargs, - ) - - def __init__( - self, - *, - host: str = "gkerecommender.googleapis.com", - credentials: Optional[ga_credentials.Credentials] = None, - credentials_file: Optional[str] = None, - scopes: Optional[Sequence[str]] = None, - channel: Optional[Union[aio.Channel, Callable[..., aio.Channel]]] = None, - api_mtls_endpoint: Optional[str] = None, - client_cert_source: Optional[Callable[[], Tuple[bytes, bytes]]] = None, - ssl_channel_credentials: Optional[grpc.ChannelCredentials] = None, - client_cert_source_for_mtls: Optional[Callable[[], Tuple[bytes, bytes]]] = None, - quota_project_id: Optional[str] = None, - client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, - always_use_jwt_access: Optional[bool] = False, - api_audience: Optional[str] = None, - ) -> None: - """Instantiate the transport. - - Args: - host (Optional[str]): - The hostname to connect to (default: 'gkerecommender.googleapis.com'). - credentials (Optional[google.auth.credentials.Credentials]): The - authorization credentials to attach to requests. These - credentials identify the application to the service; if none - are specified, the client will attempt to ascertain the - credentials from the environment. - This argument is ignored if a ``channel`` instance is provided. - credentials_file (Optional[str]): A file with credentials that can - be loaded with :func:`google.auth.load_credentials_from_file`. - This argument is ignored if a ``channel`` instance is provided. - scopes (Optional[Sequence[str]]): A optional list of scopes needed for this - service. These are only used when credentials are not specified and - are passed to :func:`google.auth.default`. - channel (Optional[Union[aio.Channel, Callable[..., aio.Channel]]]): - A ``Channel`` instance through which to make calls, or a Callable - that constructs and returns one. If set to None, ``self.create_channel`` - is used to create the channel. If a Callable is given, it will be called - with the same arguments as used in ``self.create_channel``. - api_mtls_endpoint (Optional[str]): Deprecated. The mutual TLS endpoint. - If provided, it overrides the ``host`` argument and tries to create - a mutual TLS channel with client SSL credentials from - ``client_cert_source`` or application default SSL credentials. - client_cert_source (Optional[Callable[[], Tuple[bytes, bytes]]]): - Deprecated. A callback to provide client SSL certificate bytes and - private key bytes, both in PEM format. It is ignored if - ``api_mtls_endpoint`` is None. - ssl_channel_credentials (grpc.ChannelCredentials): SSL credentials - for the grpc channel. It is ignored if a ``channel`` instance is provided. - client_cert_source_for_mtls (Optional[Callable[[], Tuple[bytes, bytes]]]): - A callback to provide client certificate bytes and private key bytes, - both in PEM format. It is used to configure a mutual TLS channel. It is - ignored if a ``channel`` instance or ``ssl_channel_credentials`` is provided. - quota_project_id (Optional[str]): An optional project to use for billing - and quota. - client_info (google.api_core.gapic_v1.client_info.ClientInfo): - The client info used to send a user-agent string along with - API requests. If ``None``, then default info will be used. - Generally, you only need to set this if you're developing - your own client library. - always_use_jwt_access (Optional[bool]): Whether self signed JWT should - be used for service account credentials. - - Raises: - google.auth.exceptions.MutualTlsChannelError: If mutual TLS transport - creation failed for any reason. - google.api_core.exceptions.DuplicateCredentialArgs: If both ``credentials`` - and ``credentials_file`` are passed. - """ - self._grpc_channel = None - self._ssl_channel_credentials = ssl_channel_credentials - self._stubs: Dict[str, Callable] = {} - - if api_mtls_endpoint: - warnings.warn("api_mtls_endpoint is deprecated", DeprecationWarning) - if client_cert_source: - warnings.warn("client_cert_source is deprecated", DeprecationWarning) - - if isinstance(channel, aio.Channel): - # Ignore credentials if a channel was passed. - credentials = None - self._ignore_credentials = True - # If a channel was explicitly provided, set it. - self._grpc_channel = channel - self._ssl_channel_credentials = None - else: - if api_mtls_endpoint: - host = api_mtls_endpoint - - # Create SSL credentials with client_cert_source or application - # default SSL credentials. - if client_cert_source: - cert, key = client_cert_source() - self._ssl_channel_credentials = grpc.ssl_channel_credentials( - certificate_chain=cert, private_key=key - ) - else: - self._ssl_channel_credentials = SslCredentials().ssl_credentials - - else: - if client_cert_source_for_mtls and not ssl_channel_credentials: - cert, key = client_cert_source_for_mtls() - self._ssl_channel_credentials = grpc.ssl_channel_credentials( - certificate_chain=cert, private_key=key - ) - - # The base transport sets the host, credentials and scopes - super().__init__( - host=host, - credentials=credentials, - credentials_file=credentials_file, - scopes=scopes, - quota_project_id=quota_project_id, - client_info=client_info, - always_use_jwt_access=always_use_jwt_access, - api_audience=api_audience, - ) - - if not self._grpc_channel: - # initialize with the provided callable or the default channel - channel_init = channel or type(self).create_channel - self._grpc_channel = channel_init( - self._host, - # use the credentials which are saved - credentials=self._credentials, - # Set ``credentials_file`` to ``None`` here as - # the credentials that we saved earlier should be used. - credentials_file=None, - scopes=self._scopes, - ssl_credentials=self._ssl_channel_credentials, - quota_project_id=quota_project_id, - options=[ - ("grpc.max_send_message_length", -1), - ("grpc.max_receive_message_length", -1), - ], - ) - - self._interceptor = _LoggingClientAIOInterceptor() - self._grpc_channel._unary_unary_interceptors.append(self._interceptor) - self._logged_channel = self._grpc_channel - self._wrap_with_kind = ( - "kind" in inspect.signature(gapic_v1.method_async.wrap_method).parameters - ) - # Wrap messages. This must be done after self._logged_channel exists - self._prep_wrapped_messages(client_info) - - @property - def grpc_channel(self) -> aio.Channel: - """Create the channel designed to connect to this service. - - This property caches on the instance; repeated calls return - the same channel. - """ - # Return the channel from cache. - return self._grpc_channel - - @property - def fetch_models( - self, - ) -> Callable[ - [gkerecommender.FetchModelsRequest], - Awaitable[gkerecommender.FetchModelsResponse], - ]: - r"""Return a callable for the fetch models method over gRPC. - - Fetches available models. Open-source models follow the - Huggingface Hub ``owner/model_name`` format. - - Returns: - Callable[[~.FetchModelsRequest], - Awaitable[~.FetchModelsResponse]]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_models" not in self._stubs: - self._stubs["fetch_models"] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModels", - request_serializer=gkerecommender.FetchModelsRequest.serialize, - response_deserializer=gkerecommender.FetchModelsResponse.deserialize, - ) - return self._stubs["fetch_models"] - - @property - def fetch_model_servers( - self, - ) -> Callable[ - [gkerecommender.FetchModelServersRequest], - Awaitable[gkerecommender.FetchModelServersResponse], - ]: - r"""Return a callable for the fetch model servers method over gRPC. - - Fetches available model servers. Open-source model servers use - simplified, lowercase names (e.g., ``vllm``). - - Returns: - Callable[[~.FetchModelServersRequest], - Awaitable[~.FetchModelServersResponse]]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_model_servers" not in self._stubs: - self._stubs["fetch_model_servers"] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModelServers", - request_serializer=gkerecommender.FetchModelServersRequest.serialize, - response_deserializer=gkerecommender.FetchModelServersResponse.deserialize, - ) - return self._stubs["fetch_model_servers"] - - @property - def fetch_model_server_versions( - self, - ) -> Callable[ - [gkerecommender.FetchModelServerVersionsRequest], - Awaitable[gkerecommender.FetchModelServerVersionsResponse], - ]: - r"""Return a callable for the fetch model server versions method over gRPC. - - Fetches available model server versions. Open-source servers use - their own versioning schemas (e.g., ``vllm`` uses semver like - ``v1.0.0``). - - Some model servers have different versioning schemas depending - on the accelerator. For example, ``vllm`` uses semver on GPUs, - but returns nightly build tags on TPUs. All available versions - will be returned when different schemas are present. - - Returns: - Callable[[~.FetchModelServerVersionsRequest], - Awaitable[~.FetchModelServerVersionsResponse]]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_model_server_versions" not in self._stubs: - self._stubs[ - "fetch_model_server_versions" - ] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchModelServerVersions", - request_serializer=gkerecommender.FetchModelServerVersionsRequest.serialize, - response_deserializer=gkerecommender.FetchModelServerVersionsResponse.deserialize, - ) - return self._stubs["fetch_model_server_versions"] - - @property - def fetch_profiles( - self, - ) -> Callable[ - [gkerecommender.FetchProfilesRequest], - Awaitable[gkerecommender.FetchProfilesResponse], - ]: - r"""Return a callable for the fetch profiles method over gRPC. - - Fetches available profiles. A profile contains performance - metrics and cost information for a specific model server setup. - Profiles can be filtered by parameters. If no filters are - provided, all profiles are returned. - - Profiles display a single value per performance metric based on - the provided performance requirements. If no requirements are - given, the metrics represent the inflection point. See `Run best - practice inference with GKE Inference Quickstart - recipes `__ - for details. - - Returns: - Callable[[~.FetchProfilesRequest], - Awaitable[~.FetchProfilesResponse]]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_profiles" not in self._stubs: - self._stubs["fetch_profiles"] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchProfiles", - request_serializer=gkerecommender.FetchProfilesRequest.serialize, - response_deserializer=gkerecommender.FetchProfilesResponse.deserialize, - ) - return self._stubs["fetch_profiles"] - - @property - def generate_optimized_manifest( - self, - ) -> Callable[ - [gkerecommender.GenerateOptimizedManifestRequest], - Awaitable[gkerecommender.GenerateOptimizedManifestResponse], - ]: - r"""Return a callable for the generate optimized manifest method over gRPC. - - Generates an optimized deployment manifest for a given model and - model server, based on the specified accelerator, performance - targets, and configurations. See `Run best practice inference - with GKE Inference Quickstart - recipes `__ - for deployment details. - - Returns: - Callable[[~.GenerateOptimizedManifestRequest], - Awaitable[~.GenerateOptimizedManifestResponse]]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "generate_optimized_manifest" not in self._stubs: - self._stubs[ - "generate_optimized_manifest" - ] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/GenerateOptimizedManifest", - request_serializer=gkerecommender.GenerateOptimizedManifestRequest.serialize, - response_deserializer=gkerecommender.GenerateOptimizedManifestResponse.deserialize, - ) - return self._stubs["generate_optimized_manifest"] - - @property - def fetch_benchmarking_data( - self, - ) -> Callable[ - [gkerecommender.FetchBenchmarkingDataRequest], - Awaitable[gkerecommender.FetchBenchmarkingDataResponse], - ]: - r"""Return a callable for the fetch benchmarking data method over gRPC. - - Fetches all of the benchmarking data available for a - profile. Benchmarking data returns all of the - performance metrics available for a given model server - setup on a given instance type. - - Returns: - Callable[[~.FetchBenchmarkingDataRequest], - Awaitable[~.FetchBenchmarkingDataResponse]]: - A function that, when called, will call the underlying RPC - on the server. - """ - # Generate a "stub function" on-the-fly which will actually make - # the request. - # gRPC handles serialization and deserialization, so we just need - # to pass in the functions for each. - if "fetch_benchmarking_data" not in self._stubs: - self._stubs["fetch_benchmarking_data"] = self._logged_channel.unary_unary( - "/google.cloud.gkerecommender.v1.GkeInferenceQuickstart/FetchBenchmarkingData", - request_serializer=gkerecommender.FetchBenchmarkingDataRequest.serialize, - response_deserializer=gkerecommender.FetchBenchmarkingDataResponse.deserialize, - ) - return self._stubs["fetch_benchmarking_data"] - - def _prep_wrapped_messages(self, client_info): - """Precompute the wrapped methods, overriding the base class method to use async wrappers.""" - self._wrapped_methods = { - self.fetch_models: self._wrap_method( - self.fetch_models, - default_timeout=60.0, - client_info=client_info, - ), - self.fetch_model_servers: self._wrap_method( - self.fetch_model_servers, - default_timeout=60.0, - client_info=client_info, - ), - self.fetch_model_server_versions: self._wrap_method( - self.fetch_model_server_versions, - default_timeout=60.0, - client_info=client_info, - ), - self.fetch_profiles: self._wrap_method( - self.fetch_profiles, - default_timeout=60.0, - client_info=client_info, - ), - self.generate_optimized_manifest: self._wrap_method( - self.generate_optimized_manifest, - default_timeout=60.0, - client_info=client_info, - ), - self.fetch_benchmarking_data: self._wrap_method( - self.fetch_benchmarking_data, - default_timeout=60.0, - client_info=client_info, - ), - } - - def _wrap_method(self, func, *args, **kwargs): - if self._wrap_with_kind: # pragma: NO COVER - kwargs["kind"] = self.kind - return gapic_v1.method_async.wrap_method(func, *args, **kwargs) - - def close(self): - return self._logged_channel.close() - - @property - def kind(self) -> str: - return "grpc_asyncio" - - -__all__ = ("GkeInferenceQuickstartGrpcAsyncIOTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest.py deleted file mode 100644 index e59cde7649f9..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest.py +++ /dev/null @@ -1,1532 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import dataclasses -import json # type: ignore -import logging -from typing import Any, Callable, Dict, List, Optional, Sequence, Tuple, Union -import warnings - -from google.api_core import exceptions as core_exceptions -from google.api_core import gapic_v1, rest_helpers, rest_streaming -from google.api_core import retry as retries -from google.auth import credentials as ga_credentials # type: ignore -from google.auth.transport.requests import AuthorizedSession # type: ignore -import google.protobuf -from google.protobuf import json_format -from requests import __version__ as requests_version - -from google.cloud.gkerecommender_v1.types import gkerecommender - -from .base import DEFAULT_CLIENT_INFO as BASE_DEFAULT_CLIENT_INFO -from .rest_base import _BaseGkeInferenceQuickstartRestTransport - -try: - OptionalRetry = Union[retries.Retry, gapic_v1.method._MethodDefault, None] -except AttributeError: # pragma: NO COVER - OptionalRetry = Union[retries.Retry, object, None] # type: ignore - -try: - from google.api_core import client_logging # type: ignore - - CLIENT_LOGGING_SUPPORTED = True # pragma: NO COVER -except ImportError: # pragma: NO COVER - CLIENT_LOGGING_SUPPORTED = False - -_LOGGER = logging.getLogger(__name__) - -DEFAULT_CLIENT_INFO = gapic_v1.client_info.ClientInfo( - gapic_version=BASE_DEFAULT_CLIENT_INFO.gapic_version, - grpc_version=None, - rest_version=f"requests@{requests_version}", -) - -if hasattr(DEFAULT_CLIENT_INFO, "protobuf_runtime_version"): # pragma: NO COVER - DEFAULT_CLIENT_INFO.protobuf_runtime_version = google.protobuf.__version__ - - -class GkeInferenceQuickstartRestInterceptor: - """Interceptor for GkeInferenceQuickstart. - - Interceptors are used to manipulate requests, request metadata, and responses - in arbitrary ways. - Example use cases include: - * Logging - * Verifying requests according to service or custom semantics - * Stripping extraneous information from responses - - These use cases and more can be enabled by injecting an - instance of a custom subclass when constructing the GkeInferenceQuickstartRestTransport. - - .. code-block:: python - class MyCustomGkeInferenceQuickstartInterceptor(GkeInferenceQuickstartRestInterceptor): - def pre_fetch_benchmarking_data(self, request, metadata): - logging.log(f"Received request: {request}") - return request, metadata - - def post_fetch_benchmarking_data(self, response): - logging.log(f"Received response: {response}") - return response - - def pre_fetch_models(self, request, metadata): - logging.log(f"Received request: {request}") - return request, metadata - - def post_fetch_models(self, response): - logging.log(f"Received response: {response}") - return response - - def pre_fetch_model_servers(self, request, metadata): - logging.log(f"Received request: {request}") - return request, metadata - - def post_fetch_model_servers(self, response): - logging.log(f"Received response: {response}") - return response - - def pre_fetch_model_server_versions(self, request, metadata): - logging.log(f"Received request: {request}") - return request, metadata - - def post_fetch_model_server_versions(self, response): - logging.log(f"Received response: {response}") - return response - - def pre_fetch_profiles(self, request, metadata): - logging.log(f"Received request: {request}") - return request, metadata - - def post_fetch_profiles(self, response): - logging.log(f"Received response: {response}") - return response - - def pre_generate_optimized_manifest(self, request, metadata): - logging.log(f"Received request: {request}") - return request, metadata - - def post_generate_optimized_manifest(self, response): - logging.log(f"Received response: {response}") - return response - - transport = GkeInferenceQuickstartRestTransport(interceptor=MyCustomGkeInferenceQuickstartInterceptor()) - client = GkeInferenceQuickstartClient(transport=transport) - - - """ - - def pre_fetch_benchmarking_data( - self, - request: gkerecommender.FetchBenchmarkingDataRequest, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchBenchmarkingDataRequest, - Sequence[Tuple[str, Union[str, bytes]]], - ]: - """Pre-rpc interceptor for fetch_benchmarking_data - - Override in a subclass to manipulate the request or metadata - before they are sent to the GkeInferenceQuickstart server. - """ - return request, metadata - - def post_fetch_benchmarking_data( - self, response: gkerecommender.FetchBenchmarkingDataResponse - ) -> gkerecommender.FetchBenchmarkingDataResponse: - """Post-rpc interceptor for fetch_benchmarking_data - - DEPRECATED. Please use the `post_fetch_benchmarking_data_with_metadata` - interceptor instead. - - Override in a subclass to read or manipulate the response - after it is returned by the GkeInferenceQuickstart server but before - it is returned to user code. This `post_fetch_benchmarking_data` interceptor runs - before the `post_fetch_benchmarking_data_with_metadata` interceptor. - """ - return response - - def post_fetch_benchmarking_data_with_metadata( - self, - response: gkerecommender.FetchBenchmarkingDataResponse, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchBenchmarkingDataResponse, - Sequence[Tuple[str, Union[str, bytes]]], - ]: - """Post-rpc interceptor for fetch_benchmarking_data - - Override in a subclass to read or manipulate the response or metadata after it - is returned by the GkeInferenceQuickstart server but before it is returned to user code. - - We recommend only using this `post_fetch_benchmarking_data_with_metadata` - interceptor in new development instead of the `post_fetch_benchmarking_data` interceptor. - When both interceptors are used, this `post_fetch_benchmarking_data_with_metadata` interceptor runs after the - `post_fetch_benchmarking_data` interceptor. The (possibly modified) response returned by - `post_fetch_benchmarking_data` will be passed to - `post_fetch_benchmarking_data_with_metadata`. - """ - return response, metadata - - def pre_fetch_models( - self, - request: gkerecommender.FetchModelsRequest, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchModelsRequest, Sequence[Tuple[str, Union[str, bytes]]] - ]: - """Pre-rpc interceptor for fetch_models - - Override in a subclass to manipulate the request or metadata - before they are sent to the GkeInferenceQuickstart server. - """ - return request, metadata - - def post_fetch_models( - self, response: gkerecommender.FetchModelsResponse - ) -> gkerecommender.FetchModelsResponse: - """Post-rpc interceptor for fetch_models - - DEPRECATED. Please use the `post_fetch_models_with_metadata` - interceptor instead. - - Override in a subclass to read or manipulate the response - after it is returned by the GkeInferenceQuickstart server but before - it is returned to user code. This `post_fetch_models` interceptor runs - before the `post_fetch_models_with_metadata` interceptor. - """ - return response - - def post_fetch_models_with_metadata( - self, - response: gkerecommender.FetchModelsResponse, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchModelsResponse, Sequence[Tuple[str, Union[str, bytes]]] - ]: - """Post-rpc interceptor for fetch_models - - Override in a subclass to read or manipulate the response or metadata after it - is returned by the GkeInferenceQuickstart server but before it is returned to user code. - - We recommend only using this `post_fetch_models_with_metadata` - interceptor in new development instead of the `post_fetch_models` interceptor. - When both interceptors are used, this `post_fetch_models_with_metadata` interceptor runs after the - `post_fetch_models` interceptor. The (possibly modified) response returned by - `post_fetch_models` will be passed to - `post_fetch_models_with_metadata`. - """ - return response, metadata - - def pre_fetch_model_servers( - self, - request: gkerecommender.FetchModelServersRequest, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchModelServersRequest, Sequence[Tuple[str, Union[str, bytes]]] - ]: - """Pre-rpc interceptor for fetch_model_servers - - Override in a subclass to manipulate the request or metadata - before they are sent to the GkeInferenceQuickstart server. - """ - return request, metadata - - def post_fetch_model_servers( - self, response: gkerecommender.FetchModelServersResponse - ) -> gkerecommender.FetchModelServersResponse: - """Post-rpc interceptor for fetch_model_servers - - DEPRECATED. Please use the `post_fetch_model_servers_with_metadata` - interceptor instead. - - Override in a subclass to read or manipulate the response - after it is returned by the GkeInferenceQuickstart server but before - it is returned to user code. This `post_fetch_model_servers` interceptor runs - before the `post_fetch_model_servers_with_metadata` interceptor. - """ - return response - - def post_fetch_model_servers_with_metadata( - self, - response: gkerecommender.FetchModelServersResponse, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchModelServersResponse, - Sequence[Tuple[str, Union[str, bytes]]], - ]: - """Post-rpc interceptor for fetch_model_servers - - Override in a subclass to read or manipulate the response or metadata after it - is returned by the GkeInferenceQuickstart server but before it is returned to user code. - - We recommend only using this `post_fetch_model_servers_with_metadata` - interceptor in new development instead of the `post_fetch_model_servers` interceptor. - When both interceptors are used, this `post_fetch_model_servers_with_metadata` interceptor runs after the - `post_fetch_model_servers` interceptor. The (possibly modified) response returned by - `post_fetch_model_servers` will be passed to - `post_fetch_model_servers_with_metadata`. - """ - return response, metadata - - def pre_fetch_model_server_versions( - self, - request: gkerecommender.FetchModelServerVersionsRequest, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchModelServerVersionsRequest, - Sequence[Tuple[str, Union[str, bytes]]], - ]: - """Pre-rpc interceptor for fetch_model_server_versions - - Override in a subclass to manipulate the request or metadata - before they are sent to the GkeInferenceQuickstart server. - """ - return request, metadata - - def post_fetch_model_server_versions( - self, response: gkerecommender.FetchModelServerVersionsResponse - ) -> gkerecommender.FetchModelServerVersionsResponse: - """Post-rpc interceptor for fetch_model_server_versions - - DEPRECATED. Please use the `post_fetch_model_server_versions_with_metadata` - interceptor instead. - - Override in a subclass to read or manipulate the response - after it is returned by the GkeInferenceQuickstart server but before - it is returned to user code. This `post_fetch_model_server_versions` interceptor runs - before the `post_fetch_model_server_versions_with_metadata` interceptor. - """ - return response - - def post_fetch_model_server_versions_with_metadata( - self, - response: gkerecommender.FetchModelServerVersionsResponse, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchModelServerVersionsResponse, - Sequence[Tuple[str, Union[str, bytes]]], - ]: - """Post-rpc interceptor for fetch_model_server_versions - - Override in a subclass to read or manipulate the response or metadata after it - is returned by the GkeInferenceQuickstart server but before it is returned to user code. - - We recommend only using this `post_fetch_model_server_versions_with_metadata` - interceptor in new development instead of the `post_fetch_model_server_versions` interceptor. - When both interceptors are used, this `post_fetch_model_server_versions_with_metadata` interceptor runs after the - `post_fetch_model_server_versions` interceptor. The (possibly modified) response returned by - `post_fetch_model_server_versions` will be passed to - `post_fetch_model_server_versions_with_metadata`. - """ - return response, metadata - - def pre_fetch_profiles( - self, - request: gkerecommender.FetchProfilesRequest, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchProfilesRequest, Sequence[Tuple[str, Union[str, bytes]]] - ]: - """Pre-rpc interceptor for fetch_profiles - - Override in a subclass to manipulate the request or metadata - before they are sent to the GkeInferenceQuickstart server. - """ - return request, metadata - - def post_fetch_profiles( - self, response: gkerecommender.FetchProfilesResponse - ) -> gkerecommender.FetchProfilesResponse: - """Post-rpc interceptor for fetch_profiles - - DEPRECATED. Please use the `post_fetch_profiles_with_metadata` - interceptor instead. - - Override in a subclass to read or manipulate the response - after it is returned by the GkeInferenceQuickstart server but before - it is returned to user code. This `post_fetch_profiles` interceptor runs - before the `post_fetch_profiles_with_metadata` interceptor. - """ - return response - - def post_fetch_profiles_with_metadata( - self, - response: gkerecommender.FetchProfilesResponse, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.FetchProfilesResponse, Sequence[Tuple[str, Union[str, bytes]]] - ]: - """Post-rpc interceptor for fetch_profiles - - Override in a subclass to read or manipulate the response or metadata after it - is returned by the GkeInferenceQuickstart server but before it is returned to user code. - - We recommend only using this `post_fetch_profiles_with_metadata` - interceptor in new development instead of the `post_fetch_profiles` interceptor. - When both interceptors are used, this `post_fetch_profiles_with_metadata` interceptor runs after the - `post_fetch_profiles` interceptor. The (possibly modified) response returned by - `post_fetch_profiles` will be passed to - `post_fetch_profiles_with_metadata`. - """ - return response, metadata - - def pre_generate_optimized_manifest( - self, - request: gkerecommender.GenerateOptimizedManifestRequest, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.GenerateOptimizedManifestRequest, - Sequence[Tuple[str, Union[str, bytes]]], - ]: - """Pre-rpc interceptor for generate_optimized_manifest - - Override in a subclass to manipulate the request or metadata - before they are sent to the GkeInferenceQuickstart server. - """ - return request, metadata - - def post_generate_optimized_manifest( - self, response: gkerecommender.GenerateOptimizedManifestResponse - ) -> gkerecommender.GenerateOptimizedManifestResponse: - """Post-rpc interceptor for generate_optimized_manifest - - DEPRECATED. Please use the `post_generate_optimized_manifest_with_metadata` - interceptor instead. - - Override in a subclass to read or manipulate the response - after it is returned by the GkeInferenceQuickstart server but before - it is returned to user code. This `post_generate_optimized_manifest` interceptor runs - before the `post_generate_optimized_manifest_with_metadata` interceptor. - """ - return response - - def post_generate_optimized_manifest_with_metadata( - self, - response: gkerecommender.GenerateOptimizedManifestResponse, - metadata: Sequence[Tuple[str, Union[str, bytes]]], - ) -> Tuple[ - gkerecommender.GenerateOptimizedManifestResponse, - Sequence[Tuple[str, Union[str, bytes]]], - ]: - """Post-rpc interceptor for generate_optimized_manifest - - Override in a subclass to read or manipulate the response or metadata after it - is returned by the GkeInferenceQuickstart server but before it is returned to user code. - - We recommend only using this `post_generate_optimized_manifest_with_metadata` - interceptor in new development instead of the `post_generate_optimized_manifest` interceptor. - When both interceptors are used, this `post_generate_optimized_manifest_with_metadata` interceptor runs after the - `post_generate_optimized_manifest` interceptor. The (possibly modified) response returned by - `post_generate_optimized_manifest` will be passed to - `post_generate_optimized_manifest_with_metadata`. - """ - return response, metadata - - -@dataclasses.dataclass -class GkeInferenceQuickstartRestStub: - _session: AuthorizedSession - _host: str - _interceptor: GkeInferenceQuickstartRestInterceptor - - -class GkeInferenceQuickstartRestTransport(_BaseGkeInferenceQuickstartRestTransport): - """REST backend synchronous transport for GkeInferenceQuickstart. - - GKE Inference Quickstart (GIQ) service provides profiles with - performance metrics for popular models and model servers across - multiple accelerators. These profiles help generate optimized - best practices for running inference on GKE. - - This class defines the same methods as the primary client, so the - primary client can load the underlying transport implementation - and call it. - - It sends JSON representations of protocol buffers over HTTP/1.1 - """ - - def __init__( - self, - *, - host: str = "gkerecommender.googleapis.com", - credentials: Optional[ga_credentials.Credentials] = None, - credentials_file: Optional[str] = None, - scopes: Optional[Sequence[str]] = None, - client_cert_source_for_mtls: Optional[Callable[[], Tuple[bytes, bytes]]] = None, - quota_project_id: Optional[str] = None, - client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, - always_use_jwt_access: Optional[bool] = False, - url_scheme: str = "https", - interceptor: Optional[GkeInferenceQuickstartRestInterceptor] = None, - api_audience: Optional[str] = None, - ) -> None: - """Instantiate the transport. - - Args: - host (Optional[str]): - The hostname to connect to (default: 'gkerecommender.googleapis.com'). - credentials (Optional[google.auth.credentials.Credentials]): The - authorization credentials to attach to requests. These - credentials identify the application to the service; if none - are specified, the client will attempt to ascertain the - credentials from the environment. - - credentials_file (Optional[str]): A file with credentials that can - be loaded with :func:`google.auth.load_credentials_from_file`. - This argument is ignored if ``channel`` is provided. - scopes (Optional(Sequence[str])): A list of scopes. This argument is - ignored if ``channel`` is provided. - client_cert_source_for_mtls (Callable[[], Tuple[bytes, bytes]]): Client - certificate to configure mutual TLS HTTP channel. It is ignored - if ``channel`` is provided. - quota_project_id (Optional[str]): An optional project to use for billing - and quota. - client_info (google.api_core.gapic_v1.client_info.ClientInfo): - The client info used to send a user-agent string along with - API requests. If ``None``, then default info will be used. - Generally, you only need to set this if you are developing - your own client library. - always_use_jwt_access (Optional[bool]): Whether self signed JWT should - be used for service account credentials. - url_scheme: the protocol scheme for the API endpoint. Normally - "https", but for testing or local servers, - "http" can be specified. - """ - # Run the base constructor - # TODO(yon-mg): resolve other ctor params i.e. scopes, quota, etc. - # TODO: When custom host (api_endpoint) is set, `scopes` must *also* be set on the - # credentials object - super().__init__( - host=host, - credentials=credentials, - client_info=client_info, - always_use_jwt_access=always_use_jwt_access, - url_scheme=url_scheme, - api_audience=api_audience, - ) - self._session = AuthorizedSession( - self._credentials, default_host=self.DEFAULT_HOST - ) - if client_cert_source_for_mtls: - self._session.configure_mtls_channel(client_cert_source_for_mtls) - self._interceptor = interceptor or GkeInferenceQuickstartRestInterceptor() - self._prep_wrapped_messages(client_info) - - class _FetchBenchmarkingData( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData, - GkeInferenceQuickstartRestStub, - ): - def __hash__(self): - return hash("GkeInferenceQuickstartRestTransport.FetchBenchmarkingData") - - @staticmethod - def _get_response( - host, - metadata, - query_params, - session, - timeout, - transcoded_request, - body=None, - ): - uri = transcoded_request["uri"] - method = transcoded_request["method"] - headers = dict(metadata) - headers["Content-Type"] = "application/json" - response = getattr(session, method)( - "{host}{uri}".format(host=host, uri=uri), - timeout=timeout, - headers=headers, - params=rest_helpers.flatten_query_params(query_params, strict=True), - data=body, - ) - return response - - def __call__( - self, - request: gkerecommender.FetchBenchmarkingDataRequest, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Optional[float] = None, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.FetchBenchmarkingDataResponse: - r"""Call the fetch benchmarking data method over HTTP. - - Args: - request (~.gkerecommender.FetchBenchmarkingDataRequest): - The request object. Request message for - [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - ~.gkerecommender.FetchBenchmarkingDataResponse: - Response message for - [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. - - """ - - http_options = ( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_http_options() - ) - - request, metadata = self._interceptor.pre_fetch_benchmarking_data( - request, metadata - ) - transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_transcoded_request( - http_options, request - ) - - body = _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_request_body_json( - transcoded_request - ) - - # Jsonify the query params - query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_query_params_json( - transcoded_request - ) - - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - request_url = "{host}{uri}".format( - host=self._host, uri=transcoded_request["uri"] - ) - method = transcoded_request["method"] - try: - request_payload = type(request).to_json(request) - except: - request_payload = None - http_request = { - "payload": request_payload, - "requestMethod": method, - "requestUrl": request_url, - "headers": dict(metadata), - } - _LOGGER.debug( - f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchBenchmarkingData", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchBenchmarkingData", - "httpRequest": http_request, - "metadata": http_request["headers"], - }, - ) - - # Send the request - response = GkeInferenceQuickstartRestTransport._FetchBenchmarkingData._get_response( - self._host, - metadata, - query_params, - self._session, - timeout, - transcoded_request, - body, - ) - - # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception - # subclass. - if response.status_code >= 400: - raise core_exceptions.from_http_response(response) - - # Return the response - resp = gkerecommender.FetchBenchmarkingDataResponse() - pb_resp = gkerecommender.FetchBenchmarkingDataResponse.pb(resp) - - json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) - - resp = self._interceptor.post_fetch_benchmarking_data(resp) - response_metadata = [(k, str(v)) for k, v in response.headers.items()] - resp, _ = self._interceptor.post_fetch_benchmarking_data_with_metadata( - resp, response_metadata - ) - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - try: - response_payload = ( - gkerecommender.FetchBenchmarkingDataResponse.to_json(response) - ) - except: - response_payload = None - http_response = { - "payload": response_payload, - "headers": dict(response.headers), - "status": response.status_code, - } - _LOGGER.debug( - "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_benchmarking_data", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchBenchmarkingData", - "metadata": http_response["headers"], - "httpResponse": http_response, - }, - ) - return resp - - class _FetchModels( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchModels, - GkeInferenceQuickstartRestStub, - ): - def __hash__(self): - return hash("GkeInferenceQuickstartRestTransport.FetchModels") - - @staticmethod - def _get_response( - host, - metadata, - query_params, - session, - timeout, - transcoded_request, - body=None, - ): - uri = transcoded_request["uri"] - method = transcoded_request["method"] - headers = dict(metadata) - headers["Content-Type"] = "application/json" - response = getattr(session, method)( - "{host}{uri}".format(host=host, uri=uri), - timeout=timeout, - headers=headers, - params=rest_helpers.flatten_query_params(query_params, strict=True), - ) - return response - - def __call__( - self, - request: gkerecommender.FetchModelsRequest, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Optional[float] = None, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.FetchModelsResponse: - r"""Call the fetch models method over HTTP. - - Args: - request (~.gkerecommender.FetchModelsRequest): - The request object. Request message for - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - ~.gkerecommender.FetchModelsResponse: - Response message for - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. - - """ - - http_options = ( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchModels._get_http_options() - ) - - request, metadata = self._interceptor.pre_fetch_models(request, metadata) - transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModels._get_transcoded_request( - http_options, request - ) - - # Jsonify the query params - query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModels._get_query_params_json( - transcoded_request - ) - - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - request_url = "{host}{uri}".format( - host=self._host, uri=transcoded_request["uri"] - ) - method = transcoded_request["method"] - try: - request_payload = type(request).to_json(request) - except: - request_payload = None - http_request = { - "payload": request_payload, - "requestMethod": method, - "requestUrl": request_url, - "headers": dict(metadata), - } - _LOGGER.debug( - f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchModels", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchModels", - "httpRequest": http_request, - "metadata": http_request["headers"], - }, - ) - - # Send the request - response = GkeInferenceQuickstartRestTransport._FetchModels._get_response( - self._host, - metadata, - query_params, - self._session, - timeout, - transcoded_request, - ) - - # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception - # subclass. - if response.status_code >= 400: - raise core_exceptions.from_http_response(response) - - # Return the response - resp = gkerecommender.FetchModelsResponse() - pb_resp = gkerecommender.FetchModelsResponse.pb(resp) - - json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) - - resp = self._interceptor.post_fetch_models(resp) - response_metadata = [(k, str(v)) for k, v in response.headers.items()] - resp, _ = self._interceptor.post_fetch_models_with_metadata( - resp, response_metadata - ) - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - try: - response_payload = gkerecommender.FetchModelsResponse.to_json( - response - ) - except: - response_payload = None - http_response = { - "payload": response_payload, - "headers": dict(response.headers), - "status": response.status_code, - } - _LOGGER.debug( - "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_models", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchModels", - "metadata": http_response["headers"], - "httpResponse": http_response, - }, - ) - return resp - - class _FetchModelServers( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers, - GkeInferenceQuickstartRestStub, - ): - def __hash__(self): - return hash("GkeInferenceQuickstartRestTransport.FetchModelServers") - - @staticmethod - def _get_response( - host, - metadata, - query_params, - session, - timeout, - transcoded_request, - body=None, - ): - uri = transcoded_request["uri"] - method = transcoded_request["method"] - headers = dict(metadata) - headers["Content-Type"] = "application/json" - response = getattr(session, method)( - "{host}{uri}".format(host=host, uri=uri), - timeout=timeout, - headers=headers, - params=rest_helpers.flatten_query_params(query_params, strict=True), - ) - return response - - def __call__( - self, - request: gkerecommender.FetchModelServersRequest, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Optional[float] = None, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.FetchModelServersResponse: - r"""Call the fetch model servers method over HTTP. - - Args: - request (~.gkerecommender.FetchModelServersRequest): - The request object. Request message for - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - ~.gkerecommender.FetchModelServersResponse: - Response message for - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. - - """ - - http_options = ( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers._get_http_options() - ) - - request, metadata = self._interceptor.pre_fetch_model_servers( - request, metadata - ) - transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers._get_transcoded_request( - http_options, request - ) - - # Jsonify the query params - query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers._get_query_params_json( - transcoded_request - ) - - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - request_url = "{host}{uri}".format( - host=self._host, uri=transcoded_request["uri"] - ) - method = transcoded_request["method"] - try: - request_payload = type(request).to_json(request) - except: - request_payload = None - http_request = { - "payload": request_payload, - "requestMethod": method, - "requestUrl": request_url, - "headers": dict(metadata), - } - _LOGGER.debug( - f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchModelServers", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchModelServers", - "httpRequest": http_request, - "metadata": http_request["headers"], - }, - ) - - # Send the request - response = ( - GkeInferenceQuickstartRestTransport._FetchModelServers._get_response( - self._host, - metadata, - query_params, - self._session, - timeout, - transcoded_request, - ) - ) - - # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception - # subclass. - if response.status_code >= 400: - raise core_exceptions.from_http_response(response) - - # Return the response - resp = gkerecommender.FetchModelServersResponse() - pb_resp = gkerecommender.FetchModelServersResponse.pb(resp) - - json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) - - resp = self._interceptor.post_fetch_model_servers(resp) - response_metadata = [(k, str(v)) for k, v in response.headers.items()] - resp, _ = self._interceptor.post_fetch_model_servers_with_metadata( - resp, response_metadata - ) - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - try: - response_payload = gkerecommender.FetchModelServersResponse.to_json( - response - ) - except: - response_payload = None - http_response = { - "payload": response_payload, - "headers": dict(response.headers), - "status": response.status_code, - } - _LOGGER.debug( - "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_model_servers", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchModelServers", - "metadata": http_response["headers"], - "httpResponse": http_response, - }, - ) - return resp - - class _FetchModelServerVersions( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions, - GkeInferenceQuickstartRestStub, - ): - def __hash__(self): - return hash("GkeInferenceQuickstartRestTransport.FetchModelServerVersions") - - @staticmethod - def _get_response( - host, - metadata, - query_params, - session, - timeout, - transcoded_request, - body=None, - ): - uri = transcoded_request["uri"] - method = transcoded_request["method"] - headers = dict(metadata) - headers["Content-Type"] = "application/json" - response = getattr(session, method)( - "{host}{uri}".format(host=host, uri=uri), - timeout=timeout, - headers=headers, - params=rest_helpers.flatten_query_params(query_params, strict=True), - ) - return response - - def __call__( - self, - request: gkerecommender.FetchModelServerVersionsRequest, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Optional[float] = None, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.FetchModelServerVersionsResponse: - r"""Call the fetch model server - versions method over HTTP. - - Args: - request (~.gkerecommender.FetchModelServerVersionsRequest): - The request object. Request message for - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - ~.gkerecommender.FetchModelServerVersionsResponse: - Response message for - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. - - """ - - http_options = ( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions._get_http_options() - ) - - request, metadata = self._interceptor.pre_fetch_model_server_versions( - request, metadata - ) - transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions._get_transcoded_request( - http_options, request - ) - - # Jsonify the query params - query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions._get_query_params_json( - transcoded_request - ) - - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - request_url = "{host}{uri}".format( - host=self._host, uri=transcoded_request["uri"] - ) - method = transcoded_request["method"] - try: - request_payload = type(request).to_json(request) - except: - request_payload = None - http_request = { - "payload": request_payload, - "requestMethod": method, - "requestUrl": request_url, - "headers": dict(metadata), - } - _LOGGER.debug( - f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchModelServerVersions", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchModelServerVersions", - "httpRequest": http_request, - "metadata": http_request["headers"], - }, - ) - - # Send the request - response = GkeInferenceQuickstartRestTransport._FetchModelServerVersions._get_response( - self._host, - metadata, - query_params, - self._session, - timeout, - transcoded_request, - ) - - # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception - # subclass. - if response.status_code >= 400: - raise core_exceptions.from_http_response(response) - - # Return the response - resp = gkerecommender.FetchModelServerVersionsResponse() - pb_resp = gkerecommender.FetchModelServerVersionsResponse.pb(resp) - - json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) - - resp = self._interceptor.post_fetch_model_server_versions(resp) - response_metadata = [(k, str(v)) for k, v in response.headers.items()] - resp, _ = self._interceptor.post_fetch_model_server_versions_with_metadata( - resp, response_metadata - ) - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - try: - response_payload = ( - gkerecommender.FetchModelServerVersionsResponse.to_json( - response - ) - ) - except: - response_payload = None - http_response = { - "payload": response_payload, - "headers": dict(response.headers), - "status": response.status_code, - } - _LOGGER.debug( - "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_model_server_versions", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchModelServerVersions", - "metadata": http_response["headers"], - "httpResponse": http_response, - }, - ) - return resp - - class _FetchProfiles( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles, - GkeInferenceQuickstartRestStub, - ): - def __hash__(self): - return hash("GkeInferenceQuickstartRestTransport.FetchProfiles") - - @staticmethod - def _get_response( - host, - metadata, - query_params, - session, - timeout, - transcoded_request, - body=None, - ): - uri = transcoded_request["uri"] - method = transcoded_request["method"] - headers = dict(metadata) - headers["Content-Type"] = "application/json" - response = getattr(session, method)( - "{host}{uri}".format(host=host, uri=uri), - timeout=timeout, - headers=headers, - params=rest_helpers.flatten_query_params(query_params, strict=True), - data=body, - ) - return response - - def __call__( - self, - request: gkerecommender.FetchProfilesRequest, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Optional[float] = None, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.FetchProfilesResponse: - r"""Call the fetch profiles method over HTTP. - - Args: - request (~.gkerecommender.FetchProfilesRequest): - The request object. Request message for - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - ~.gkerecommender.FetchProfilesResponse: - Response message for - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - - """ - - http_options = ( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles._get_http_options() - ) - - request, metadata = self._interceptor.pre_fetch_profiles(request, metadata) - transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles._get_transcoded_request( - http_options, request - ) - - body = _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles._get_request_body_json( - transcoded_request - ) - - # Jsonify the query params - query_params = _BaseGkeInferenceQuickstartRestTransport._BaseFetchProfiles._get_query_params_json( - transcoded_request - ) - - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - request_url = "{host}{uri}".format( - host=self._host, uri=transcoded_request["uri"] - ) - method = transcoded_request["method"] - try: - request_payload = type(request).to_json(request) - except: - request_payload = None - http_request = { - "payload": request_payload, - "requestMethod": method, - "requestUrl": request_url, - "headers": dict(metadata), - } - _LOGGER.debug( - f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.FetchProfiles", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchProfiles", - "httpRequest": http_request, - "metadata": http_request["headers"], - }, - ) - - # Send the request - response = GkeInferenceQuickstartRestTransport._FetchProfiles._get_response( - self._host, - metadata, - query_params, - self._session, - timeout, - transcoded_request, - body, - ) - - # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception - # subclass. - if response.status_code >= 400: - raise core_exceptions.from_http_response(response) - - # Return the response - resp = gkerecommender.FetchProfilesResponse() - pb_resp = gkerecommender.FetchProfilesResponse.pb(resp) - - json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) - - resp = self._interceptor.post_fetch_profiles(resp) - response_metadata = [(k, str(v)) for k, v in response.headers.items()] - resp, _ = self._interceptor.post_fetch_profiles_with_metadata( - resp, response_metadata - ) - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - try: - response_payload = gkerecommender.FetchProfilesResponse.to_json( - response - ) - except: - response_payload = None - http_response = { - "payload": response_payload, - "headers": dict(response.headers), - "status": response.status_code, - } - _LOGGER.debug( - "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_profiles", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "FetchProfiles", - "metadata": http_response["headers"], - "httpResponse": http_response, - }, - ) - return resp - - class _GenerateOptimizedManifest( - _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest, - GkeInferenceQuickstartRestStub, - ): - def __hash__(self): - return hash("GkeInferenceQuickstartRestTransport.GenerateOptimizedManifest") - - @staticmethod - def _get_response( - host, - metadata, - query_params, - session, - timeout, - transcoded_request, - body=None, - ): - uri = transcoded_request["uri"] - method = transcoded_request["method"] - headers = dict(metadata) - headers["Content-Type"] = "application/json" - response = getattr(session, method)( - "{host}{uri}".format(host=host, uri=uri), - timeout=timeout, - headers=headers, - params=rest_helpers.flatten_query_params(query_params, strict=True), - data=body, - ) - return response - - def __call__( - self, - request: gkerecommender.GenerateOptimizedManifestRequest, - *, - retry: OptionalRetry = gapic_v1.method.DEFAULT, - timeout: Optional[float] = None, - metadata: Sequence[Tuple[str, Union[str, bytes]]] = (), - ) -> gkerecommender.GenerateOptimizedManifestResponse: - r"""Call the generate optimized - manifest method over HTTP. - - Args: - request (~.gkerecommender.GenerateOptimizedManifestRequest): - The request object. Request message for - [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. - retry (google.api_core.retry.Retry): Designation of what errors, if any, - should be retried. - timeout (float): The timeout for this request. - metadata (Sequence[Tuple[str, Union[str, bytes]]]): Key/value pairs which should be - sent along with the request as metadata. Normally, each value must be of type `str`, - but for metadata keys ending with the suffix `-bin`, the corresponding values must - be of type `bytes`. - - Returns: - ~.gkerecommender.GenerateOptimizedManifestResponse: - Response message for - [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. - - """ - - http_options = ( - _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_http_options() - ) - - request, metadata = self._interceptor.pre_generate_optimized_manifest( - request, metadata - ) - transcoded_request = _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_transcoded_request( - http_options, request - ) - - body = _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_request_body_json( - transcoded_request - ) - - # Jsonify the query params - query_params = _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_query_params_json( - transcoded_request - ) - - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - request_url = "{host}{uri}".format( - host=self._host, uri=transcoded_request["uri"] - ) - method = transcoded_request["method"] - try: - request_payload = type(request).to_json(request) - except: - request_payload = None - http_request = { - "payload": request_payload, - "requestMethod": method, - "requestUrl": request_url, - "headers": dict(metadata), - } - _LOGGER.debug( - f"Sending request for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.GenerateOptimizedManifest", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "GenerateOptimizedManifest", - "httpRequest": http_request, - "metadata": http_request["headers"], - }, - ) - - # Send the request - response = GkeInferenceQuickstartRestTransport._GenerateOptimizedManifest._get_response( - self._host, - metadata, - query_params, - self._session, - timeout, - transcoded_request, - body, - ) - - # In case of error, raise the appropriate core_exceptions.GoogleAPICallError exception - # subclass. - if response.status_code >= 400: - raise core_exceptions.from_http_response(response) - - # Return the response - resp = gkerecommender.GenerateOptimizedManifestResponse() - pb_resp = gkerecommender.GenerateOptimizedManifestResponse.pb(resp) - - json_format.Parse(response.content, pb_resp, ignore_unknown_fields=True) - - resp = self._interceptor.post_generate_optimized_manifest(resp) - response_metadata = [(k, str(v)) for k, v in response.headers.items()] - resp, _ = self._interceptor.post_generate_optimized_manifest_with_metadata( - resp, response_metadata - ) - if CLIENT_LOGGING_SUPPORTED and _LOGGER.isEnabledFor( - logging.DEBUG - ): # pragma: NO COVER - try: - response_payload = ( - gkerecommender.GenerateOptimizedManifestResponse.to_json( - response - ) - ) - except: - response_payload = None - http_response = { - "payload": response_payload, - "headers": dict(response.headers), - "status": response.status_code, - } - _LOGGER.debug( - "Received response for google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.generate_optimized_manifest", - extra={ - "serviceName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "rpcName": "GenerateOptimizedManifest", - "metadata": http_response["headers"], - "httpResponse": http_response, - }, - ) - return resp - - @property - def fetch_benchmarking_data( - self, - ) -> Callable[ - [gkerecommender.FetchBenchmarkingDataRequest], - gkerecommender.FetchBenchmarkingDataResponse, - ]: - # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. - # In C++ this would require a dynamic_cast - return self._FetchBenchmarkingData(self._session, self._host, self._interceptor) # type: ignore - - @property - def fetch_models( - self, - ) -> Callable[ - [gkerecommender.FetchModelsRequest], gkerecommender.FetchModelsResponse - ]: - # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. - # In C++ this would require a dynamic_cast - return self._FetchModels(self._session, self._host, self._interceptor) # type: ignore - - @property - def fetch_model_servers( - self, - ) -> Callable[ - [gkerecommender.FetchModelServersRequest], - gkerecommender.FetchModelServersResponse, - ]: - # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. - # In C++ this would require a dynamic_cast - return self._FetchModelServers(self._session, self._host, self._interceptor) # type: ignore - - @property - def fetch_model_server_versions( - self, - ) -> Callable[ - [gkerecommender.FetchModelServerVersionsRequest], - gkerecommender.FetchModelServerVersionsResponse, - ]: - # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. - # In C++ this would require a dynamic_cast - return self._FetchModelServerVersions(self._session, self._host, self._interceptor) # type: ignore - - @property - def fetch_profiles( - self, - ) -> Callable[ - [gkerecommender.FetchProfilesRequest], gkerecommender.FetchProfilesResponse - ]: - # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. - # In C++ this would require a dynamic_cast - return self._FetchProfiles(self._session, self._host, self._interceptor) # type: ignore - - @property - def generate_optimized_manifest( - self, - ) -> Callable[ - [gkerecommender.GenerateOptimizedManifestRequest], - gkerecommender.GenerateOptimizedManifestResponse, - ]: - # The return type is fine, but mypy isn't sophisticated enough to determine what's going on here. - # In C++ this would require a dynamic_cast - return self._GenerateOptimizedManifest(self._session, self._host, self._interceptor) # type: ignore - - @property - def kind(self) -> str: - return "rest" - - def close(self): - self._session.close() - - -__all__ = ("GkeInferenceQuickstartRestTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest_base.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest_base.py deleted file mode 100644 index 07ac486ec876..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/services/gke_inference_quickstart/transports/rest_base.py +++ /dev/null @@ -1,378 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import json # type: ignore -import re -from typing import Any, Callable, Dict, List, Optional, Sequence, Tuple, Union - -from google.api_core import gapic_v1, path_template -from google.protobuf import json_format - -from google.cloud.gkerecommender_v1.types import gkerecommender - -from .base import DEFAULT_CLIENT_INFO, GkeInferenceQuickstartTransport - - -class _BaseGkeInferenceQuickstartRestTransport(GkeInferenceQuickstartTransport): - """Base REST backend transport for GkeInferenceQuickstart. - - Note: This class is not meant to be used directly. Use its sync and - async sub-classes instead. - - This class defines the same methods as the primary client, so the - primary client can load the underlying transport implementation - and call it. - - It sends JSON representations of protocol buffers over HTTP/1.1 - """ - - def __init__( - self, - *, - host: str = "gkerecommender.googleapis.com", - credentials: Optional[Any] = None, - client_info: gapic_v1.client_info.ClientInfo = DEFAULT_CLIENT_INFO, - always_use_jwt_access: Optional[bool] = False, - url_scheme: str = "https", - api_audience: Optional[str] = None, - ) -> None: - """Instantiate the transport. - Args: - host (Optional[str]): - The hostname to connect to (default: 'gkerecommender.googleapis.com'). - credentials (Optional[Any]): The - authorization credentials to attach to requests. These - credentials identify the application to the service; if none - are specified, the client will attempt to ascertain the - credentials from the environment. - client_info (google.api_core.gapic_v1.client_info.ClientInfo): - The client info used to send a user-agent string along with - API requests. If ``None``, then default info will be used. - Generally, you only need to set this if you are developing - your own client library. - always_use_jwt_access (Optional[bool]): Whether self signed JWT should - be used for service account credentials. - url_scheme: the protocol scheme for the API endpoint. Normally - "https", but for testing or local servers, - "http" can be specified. - """ - # Run the base constructor - maybe_url_match = re.match("^(?Phttp(?:s)?://)?(?P.*)$", host) - if maybe_url_match is None: - raise ValueError( - f"Unexpected hostname structure: {host}" - ) # pragma: NO COVER - - url_match_items = maybe_url_match.groupdict() - - host = f"{url_scheme}://{host}" if not url_match_items["scheme"] else host - - super().__init__( - host=host, - credentials=credentials, - client_info=client_info, - always_use_jwt_access=always_use_jwt_access, - api_audience=api_audience, - ) - - class _BaseFetchBenchmarkingData: - def __hash__(self): # pragma: NO COVER - return NotImplementedError("__hash__ must be implemented.") - - __REQUIRED_FIELDS_DEFAULT_VALUES: Dict[str, Any] = {} - - @classmethod - def _get_unset_required_fields(cls, message_dict): - return { - k: v - for k, v in cls.__REQUIRED_FIELDS_DEFAULT_VALUES.items() - if k not in message_dict - } - - @staticmethod - def _get_http_options(): - http_options: List[Dict[str, str]] = [ - { - "method": "post", - "uri": "/v1/benchmarkingData:fetch", - "body": "*", - }, - ] - return http_options - - @staticmethod - def _get_transcoded_request(http_options, request): - pb_request = gkerecommender.FetchBenchmarkingDataRequest.pb(request) - transcoded_request = path_template.transcode(http_options, pb_request) - return transcoded_request - - @staticmethod - def _get_request_body_json(transcoded_request): - # Jsonify the request body - - body = json_format.MessageToJson( - transcoded_request["body"], use_integers_for_enums=True - ) - return body - - @staticmethod - def _get_query_params_json(transcoded_request): - query_params = json.loads( - json_format.MessageToJson( - transcoded_request["query_params"], - use_integers_for_enums=True, - ) - ) - query_params.update( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchBenchmarkingData._get_unset_required_fields( - query_params - ) - ) - - query_params["$alt"] = "json;enum-encoding=int" - return query_params - - class _BaseFetchModels: - def __hash__(self): # pragma: NO COVER - return NotImplementedError("__hash__ must be implemented.") - - @staticmethod - def _get_http_options(): - http_options: List[Dict[str, str]] = [ - { - "method": "get", - "uri": "/v1/models:fetch", - }, - ] - return http_options - - @staticmethod - def _get_transcoded_request(http_options, request): - pb_request = gkerecommender.FetchModelsRequest.pb(request) - transcoded_request = path_template.transcode(http_options, pb_request) - return transcoded_request - - @staticmethod - def _get_query_params_json(transcoded_request): - query_params = json.loads( - json_format.MessageToJson( - transcoded_request["query_params"], - use_integers_for_enums=True, - ) - ) - - query_params["$alt"] = "json;enum-encoding=int" - return query_params - - class _BaseFetchModelServers: - def __hash__(self): # pragma: NO COVER - return NotImplementedError("__hash__ must be implemented.") - - __REQUIRED_FIELDS_DEFAULT_VALUES: Dict[str, Any] = { - "model": "", - } - - @classmethod - def _get_unset_required_fields(cls, message_dict): - return { - k: v - for k, v in cls.__REQUIRED_FIELDS_DEFAULT_VALUES.items() - if k not in message_dict - } - - @staticmethod - def _get_http_options(): - http_options: List[Dict[str, str]] = [ - { - "method": "get", - "uri": "/v1/modelServers:fetch", - }, - ] - return http_options - - @staticmethod - def _get_transcoded_request(http_options, request): - pb_request = gkerecommender.FetchModelServersRequest.pb(request) - transcoded_request = path_template.transcode(http_options, pb_request) - return transcoded_request - - @staticmethod - def _get_query_params_json(transcoded_request): - query_params = json.loads( - json_format.MessageToJson( - transcoded_request["query_params"], - use_integers_for_enums=True, - ) - ) - query_params.update( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServers._get_unset_required_fields( - query_params - ) - ) - - query_params["$alt"] = "json;enum-encoding=int" - return query_params - - class _BaseFetchModelServerVersions: - def __hash__(self): # pragma: NO COVER - return NotImplementedError("__hash__ must be implemented.") - - __REQUIRED_FIELDS_DEFAULT_VALUES: Dict[str, Any] = { - "model": "", - "modelServer": "", - } - - @classmethod - def _get_unset_required_fields(cls, message_dict): - return { - k: v - for k, v in cls.__REQUIRED_FIELDS_DEFAULT_VALUES.items() - if k not in message_dict - } - - @staticmethod - def _get_http_options(): - http_options: List[Dict[str, str]] = [ - { - "method": "get", - "uri": "/v1/modelServerVersions:fetch", - }, - ] - return http_options - - @staticmethod - def _get_transcoded_request(http_options, request): - pb_request = gkerecommender.FetchModelServerVersionsRequest.pb(request) - transcoded_request = path_template.transcode(http_options, pb_request) - return transcoded_request - - @staticmethod - def _get_query_params_json(transcoded_request): - query_params = json.loads( - json_format.MessageToJson( - transcoded_request["query_params"], - use_integers_for_enums=True, - ) - ) - query_params.update( - _BaseGkeInferenceQuickstartRestTransport._BaseFetchModelServerVersions._get_unset_required_fields( - query_params - ) - ) - - query_params["$alt"] = "json;enum-encoding=int" - return query_params - - class _BaseFetchProfiles: - def __hash__(self): # pragma: NO COVER - return NotImplementedError("__hash__ must be implemented.") - - @staticmethod - def _get_http_options(): - http_options: List[Dict[str, str]] = [ - { - "method": "post", - "uri": "/v1/profiles:fetch", - "body": "*", - }, - ] - return http_options - - @staticmethod - def _get_transcoded_request(http_options, request): - pb_request = gkerecommender.FetchProfilesRequest.pb(request) - transcoded_request = path_template.transcode(http_options, pb_request) - return transcoded_request - - @staticmethod - def _get_request_body_json(transcoded_request): - # Jsonify the request body - - body = json_format.MessageToJson( - transcoded_request["body"], use_integers_for_enums=True - ) - return body - - @staticmethod - def _get_query_params_json(transcoded_request): - query_params = json.loads( - json_format.MessageToJson( - transcoded_request["query_params"], - use_integers_for_enums=True, - ) - ) - - query_params["$alt"] = "json;enum-encoding=int" - return query_params - - class _BaseGenerateOptimizedManifest: - def __hash__(self): # pragma: NO COVER - return NotImplementedError("__hash__ must be implemented.") - - __REQUIRED_FIELDS_DEFAULT_VALUES: Dict[str, Any] = {} - - @classmethod - def _get_unset_required_fields(cls, message_dict): - return { - k: v - for k, v in cls.__REQUIRED_FIELDS_DEFAULT_VALUES.items() - if k not in message_dict - } - - @staticmethod - def _get_http_options(): - http_options: List[Dict[str, str]] = [ - { - "method": "post", - "uri": "/v1/optimizedManifest:generate", - "body": "*", - }, - ] - return http_options - - @staticmethod - def _get_transcoded_request(http_options, request): - pb_request = gkerecommender.GenerateOptimizedManifestRequest.pb(request) - transcoded_request = path_template.transcode(http_options, pb_request) - return transcoded_request - - @staticmethod - def _get_request_body_json(transcoded_request): - # Jsonify the request body - - body = json_format.MessageToJson( - transcoded_request["body"], use_integers_for_enums=True - ) - return body - - @staticmethod - def _get_query_params_json(transcoded_request): - query_params = json.loads( - json_format.MessageToJson( - transcoded_request["query_params"], - use_integers_for_enums=True, - ) - ) - query_params.update( - _BaseGkeInferenceQuickstartRestTransport._BaseGenerateOptimizedManifest._get_unset_required_fields( - query_params - ) - ) - - query_params["$alt"] = "json;enum-encoding=int" - return query_params - - -__all__ = ("_BaseGkeInferenceQuickstartRestTransport",) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/__init__.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/__init__.py deleted file mode 100644 index bf87f8b02a3e..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/__init__.py +++ /dev/null @@ -1,68 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from .gkerecommender import ( - Amount, - Cost, - FetchBenchmarkingDataRequest, - FetchBenchmarkingDataResponse, - FetchModelServersRequest, - FetchModelServersResponse, - FetchModelServerVersionsRequest, - FetchModelServerVersionsResponse, - FetchModelsRequest, - FetchModelsResponse, - FetchProfilesRequest, - FetchProfilesResponse, - GenerateOptimizedManifestRequest, - GenerateOptimizedManifestResponse, - KubernetesManifest, - MillisecondRange, - ModelServerInfo, - PerformanceRange, - PerformanceRequirements, - PerformanceStats, - Profile, - ResourcesUsed, - StorageConfig, - TokensPerSecondRange, -) - -__all__ = ( - "Amount", - "Cost", - "FetchBenchmarkingDataRequest", - "FetchBenchmarkingDataResponse", - "FetchModelServersRequest", - "FetchModelServersResponse", - "FetchModelServerVersionsRequest", - "FetchModelServerVersionsResponse", - "FetchModelsRequest", - "FetchModelsResponse", - "FetchProfilesRequest", - "FetchProfilesResponse", - "GenerateOptimizedManifestRequest", - "GenerateOptimizedManifestResponse", - "KubernetesManifest", - "MillisecondRange", - "ModelServerInfo", - "PerformanceRange", - "PerformanceRequirements", - "PerformanceStats", - "Profile", - "ResourcesUsed", - "StorageConfig", - "TokensPerSecondRange", -) diff --git a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/gkerecommender.py b/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/gkerecommender.py deleted file mode 100644 index cff65826fa75..000000000000 --- a/packages/google-cloud-gkerecommender/google/cloud/gkerecommender_v1/types/gkerecommender.py +++ /dev/null @@ -1,983 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -from __future__ import annotations - -from typing import MutableMapping, MutableSequence - -import proto # type: ignore - -__protobuf__ = proto.module( - package="google.cloud.gkerecommender.v1", - manifest={ - "FetchModelsRequest", - "FetchModelsResponse", - "FetchModelServersRequest", - "FetchModelServersResponse", - "FetchModelServerVersionsRequest", - "FetchModelServerVersionsResponse", - "FetchBenchmarkingDataRequest", - "FetchBenchmarkingDataResponse", - "FetchProfilesRequest", - "PerformanceRequirements", - "Amount", - "Cost", - "TokensPerSecondRange", - "MillisecondRange", - "PerformanceRange", - "FetchProfilesResponse", - "ModelServerInfo", - "ResourcesUsed", - "PerformanceStats", - "Profile", - "GenerateOptimizedManifestRequest", - "KubernetesManifest", - "GenerateOptimizedManifestResponse", - "StorageConfig", - }, -) - - -class FetchModelsRequest(proto.Message): - r"""Request message for - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. - - - .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields - - Attributes: - page_size (int): - Optional. The target number of results to return in a single - response. If not specified, a default value will be chosen - by the service. Note that the response may include a partial - list and a caller should only rely on the response's - [next_page_token][google.cloud.gkerecommender.v1.FetchModelsResponse.next_page_token] - to determine if there are more instances left to be queried. - - This field is a member of `oneof`_ ``_page_size``. - page_token (str): - Optional. The value of - [next_page_token][google.cloud.gkerecommender.v1.FetchModelsResponse.next_page_token] - received from a previous ``FetchModelsRequest`` call. - Provide this to retrieve the subsequent page in a multi-page - list of results. When paginating, all other parameters - provided to ``FetchModelsRequest`` must match the call that - provided the page token. - - This field is a member of `oneof`_ ``_page_token``. - """ - - page_size: int = proto.Field( - proto.INT32, - number=1, - optional=True, - ) - page_token: str = proto.Field( - proto.STRING, - number=2, - optional=True, - ) - - -class FetchModelsResponse(proto.Message): - r"""Response message for - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels]. - - Attributes: - models (MutableSequence[str]): - Output only. List of available models. Open-source models - follow the Huggingface Hub ``owner/model_name`` format. - next_page_token (str): - Output only. A token which may be sent as - [page_token][FetchModelsResponse.page_token] in a subsequent - ``FetchModelsResponse`` call to retrieve the next page of - results. If this field is omitted or empty, then there are - no more results to return. - """ - - @property - def raw_page(self): - return self - - models: MutableSequence[str] = proto.RepeatedField( - proto.STRING, - number=1, - ) - next_page_token: str = proto.Field( - proto.STRING, - number=2, - ) - - -class FetchModelServersRequest(proto.Message): - r"""Request message for - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. - - - .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields - - Attributes: - model (str): - Required. The model for which to list model servers. - Open-source models follow the Huggingface Hub - ``owner/model_name`` format. Use - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels] - to find available models. - page_size (int): - Optional. The target number of results to return in a single - response. If not specified, a default value will be chosen - by the service. Note that the response may include a partial - list and a caller should only rely on the response's - [next_page_token][google.cloud.gkerecommender.v1.FetchModelServersResponse.next_page_token] - to determine if there are more instances left to be queried. - - This field is a member of `oneof`_ ``_page_size``. - page_token (str): - Optional. The value of - [next_page_token][google.cloud.gkerecommender.v1.FetchModelServersResponse.next_page_token] - received from a previous ``FetchModelServersRequest`` call. - Provide this to retrieve the subsequent page in a multi-page - list of results. When paginating, all other parameters - provided to ``FetchModelServersRequest`` must match the call - that provided the page token. - - This field is a member of `oneof`_ ``_page_token``. - """ - - model: str = proto.Field( - proto.STRING, - number=1, - ) - page_size: int = proto.Field( - proto.INT32, - number=2, - optional=True, - ) - page_token: str = proto.Field( - proto.STRING, - number=3, - optional=True, - ) - - -class FetchModelServersResponse(proto.Message): - r"""Response message for - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers]. - - Attributes: - model_servers (MutableSequence[str]): - Output only. List of available model servers. Open-source - model servers use simplified, lowercase names (e.g., - ``vllm``). - next_page_token (str): - Output only. A token which may be sent as - [page_token][FetchModelServersResponse.page_token] in a - subsequent ``FetchModelServersResponse`` call to retrieve - the next page of results. If this field is omitted or empty, - then there are no more results to return. - """ - - @property - def raw_page(self): - return self - - model_servers: MutableSequence[str] = proto.RepeatedField( - proto.STRING, - number=1, - ) - next_page_token: str = proto.Field( - proto.STRING, - number=2, - ) - - -class FetchModelServerVersionsRequest(proto.Message): - r"""Request message for - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. - - - .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields - - Attributes: - model (str): - Required. The model for which to list model server versions. - Open-source models follow the Huggingface Hub - ``owner/model_name`` format. Use - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels] - to find available models. - model_server (str): - Required. The model server for which to list versions. - Open-source model servers use simplified, lowercase names - (e.g., ``vllm``). Use - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers] - to find available model servers. - page_size (int): - Optional. The target number of results to return in a single - response. If not specified, a default value will be chosen - by the service. Note that the response may include a partial - list and a caller should only rely on the response's - [next_page_token][google.cloud.gkerecommender.v1.FetchModelServerVersionsResponse.next_page_token] - to determine if there are more instances left to be queried. - - This field is a member of `oneof`_ ``_page_size``. - page_token (str): - Optional. The value of - [next_page_token][google.cloud.gkerecommender.v1.FetchModelServerVersionsResponse.next_page_token] - received from a previous ``FetchModelServerVersionsRequest`` - call. Provide this to retrieve the subsequent page in a - multi-page list of results. When paginating, all other - parameters provided to ``FetchModelServerVersionsRequest`` - must match the call that provided the page token. - - This field is a member of `oneof`_ ``_page_token``. - """ - - model: str = proto.Field( - proto.STRING, - number=1, - ) - model_server: str = proto.Field( - proto.STRING, - number=2, - ) - page_size: int = proto.Field( - proto.INT32, - number=3, - optional=True, - ) - page_token: str = proto.Field( - proto.STRING, - number=4, - optional=True, - ) - - -class FetchModelServerVersionsResponse(proto.Message): - r"""Response message for - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions]. - - Attributes: - model_server_versions (MutableSequence[str]): - Output only. A list of available model server - versions. - next_page_token (str): - Output only. A token which may be sent as - [page_token][FetchModelServerVersionsResponse.page_token] in - a subsequent ``FetchModelServerVersionsResponse`` call to - retrieve the next page of results. If this field is omitted - or empty, then there are no more results to return. - """ - - @property - def raw_page(self): - return self - - model_server_versions: MutableSequence[str] = proto.RepeatedField( - proto.STRING, - number=1, - ) - next_page_token: str = proto.Field( - proto.STRING, - number=2, - ) - - -class FetchBenchmarkingDataRequest(proto.Message): - r"""Request message for - [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. - - Attributes: - model_server_info (google.cloud.gkerecommender_v1.types.ModelServerInfo): - Required. The model server configuration to get benchmarking - data for. Use - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] - to find valid configurations. - instance_type (str): - Optional. The instance type to filter benchmarking data. - Instance types are in the format ``a2-highgpu-1g``. If not - provided, all instance types for the given profile's - ``model_server_info`` will be returned. Use - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] - to find available instance types. - pricing_model (str): - Optional. The pricing model to use for the benchmarking - data. Defaults to ``spot``. - """ - - model_server_info: "ModelServerInfo" = proto.Field( - proto.MESSAGE, - number=1, - message="ModelServerInfo", - ) - instance_type: str = proto.Field( - proto.STRING, - number=3, - ) - pricing_model: str = proto.Field( - proto.STRING, - number=4, - ) - - -class FetchBenchmarkingDataResponse(proto.Message): - r"""Response message for - [GkeInferenceQuickstart.FetchBenchmarkingData][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData]. - - Attributes: - profile (MutableSequence[google.cloud.gkerecommender_v1.types.Profile]): - Output only. List of profiles containing - their respective benchmarking data. - """ - - profile: MutableSequence["Profile"] = proto.RepeatedField( - proto.MESSAGE, - number=1, - message="Profile", - ) - - -class FetchProfilesRequest(proto.Message): - r"""Request message for - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - - - .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields - - Attributes: - model (str): - Optional. The model to filter profiles by. Open-source - models follow the Huggingface Hub ``owner/model_name`` - format. If not provided, all models are returned. Use - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels] - to find available models. - model_server (str): - Optional. The model server to filter profiles by. If not - provided, all model servers are returned. Use - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers] - to find available model servers for a given model. - model_server_version (str): - Optional. The model server version to filter profiles by. If - not provided, all model server versions are returned. Use - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions] - to find available versions for a given model and server. - performance_requirements (google.cloud.gkerecommender_v1.types.PerformanceRequirements): - Optional. The performance requirements to - filter profiles. Profiles that do not meet these - requirements are filtered out. If not provided, - all profiles are returned. - page_size (int): - Optional. The target number of results to return in a single - response. If not specified, a default value will be chosen - by the service. Note that the response may include a partial - list and a caller should only rely on the response's - [next_page_token][google.cloud.gkerecommender.v1.FetchProfilesResponse.next_page_token] - to determine if there are more instances left to be queried. - - This field is a member of `oneof`_ ``_page_size``. - page_token (str): - Optional. The value of - [next_page_token][google.cloud.gkerecommender.v1.FetchProfilesResponse.next_page_token] - received from a previous ``FetchProfilesRequest`` call. - Provide this to retrieve the subsequent page in a multi-page - list of results. When paginating, all other parameters - provided to ``FetchProfilesRequest`` must match the call - that provided the page token. - - This field is a member of `oneof`_ ``_page_token``. - """ - - model: str = proto.Field( - proto.STRING, - number=1, - ) - model_server: str = proto.Field( - proto.STRING, - number=2, - ) - model_server_version: str = proto.Field( - proto.STRING, - number=3, - ) - performance_requirements: "PerformanceRequirements" = proto.Field( - proto.MESSAGE, - number=4, - message="PerformanceRequirements", - ) - page_size: int = proto.Field( - proto.INT32, - number=5, - optional=True, - ) - page_token: str = proto.Field( - proto.STRING, - number=6, - optional=True, - ) - - -class PerformanceRequirements(proto.Message): - r"""Performance requirements for a profile and or model - deployment. - - - .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields - - Attributes: - target_ntpot_milliseconds (int): - Optional. The target Normalized Time Per Output Token - (NTPOT) in milliseconds. NTPOT is calculated as - ``request_latency / total_output_tokens``. If not provided, - this target will not be enforced. - - This field is a member of `oneof`_ ``_target_ntpot_milliseconds``. - target_ttft_milliseconds (int): - Optional. The target Time To First Token - (TTFT) in milliseconds. TTFT is the time it - takes to generate the first token for a request. - If not provided, this target will not be - enforced. - - This field is a member of `oneof`_ ``_target_ttft_milliseconds``. - target_cost (google.cloud.gkerecommender_v1.types.Cost): - Optional. The target cost for running a - profile's model server. If not provided, this - requirement will not be enforced. - """ - - target_ntpot_milliseconds: int = proto.Field( - proto.INT32, - number=1, - optional=True, - ) - target_ttft_milliseconds: int = proto.Field( - proto.INT32, - number=2, - optional=True, - ) - target_cost: "Cost" = proto.Field( - proto.MESSAGE, - number=3, - message="Cost", - ) - - -class Amount(proto.Message): - r"""Represents an amount of money in a specific currency. - - Attributes: - units (int): - Output only. The whole units of the amount. For example if - ``currencyCode`` is ``"USD"``, then 1 unit is one US dollar. - nanos (int): - Output only. Number of nano (10^-9) units of the amount. The - value must be between -999,999,999 and +999,999,999 - inclusive. If ``units`` is positive, ``nanos`` must be - positive or zero. If ``units`` is zero, ``nanos`` can be - positive, zero, or negative. If ``units`` is negative, - ``nanos`` must be negative or zero. For example $-1.75 is - represented as ``units``\ =-1 and ``nanos``\ =-750,000,000. - """ - - units: int = proto.Field( - proto.INT64, - number=1, - ) - nanos: int = proto.Field( - proto.INT32, - number=2, - ) - - -class Cost(proto.Message): - r"""Cost for running a model deployment on a given instance type. - Currently, only USD currency code is supported. - - - .. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields - - Attributes: - cost_per_million_output_tokens (google.cloud.gkerecommender_v1.types.Amount): - Optional. The cost per million output tokens, calculated as: - $/output token = GPU $/s / (1/output-to-input-cost-ratio \* - input tokens/s + output tokens/s) - cost_per_million_input_tokens (google.cloud.gkerecommender_v1.types.Amount): - Optional. The cost per million input tokens. - $/input token = ($/output token) / - output-to-input-cost-ratio. - pricing_model (str): - Optional. The pricing model used to calculate the cost. Can - be one of: ``3-years-cud``, ``1-year-cud``, ``on-demand``, - ``spot``. If not provided, ``spot`` will be used. - output_input_cost_ratio (float): - Optional. The output-to-input cost ratio. This determines - how the total GPU cost is split between input and output - tokens. If not provided, ``4.0`` is used, assuming a 4:1 - output:input cost ratio. - - This field is a member of `oneof`_ ``_output_input_cost_ratio``. - """ - - cost_per_million_output_tokens: "Amount" = proto.Field( - proto.MESSAGE, - number=1, - message="Amount", - ) - cost_per_million_input_tokens: "Amount" = proto.Field( - proto.MESSAGE, - number=2, - message="Amount", - ) - pricing_model: str = proto.Field( - proto.STRING, - number=3, - ) - output_input_cost_ratio: float = proto.Field( - proto.FLOAT, - number=4, - optional=True, - ) - - -class TokensPerSecondRange(proto.Message): - r"""Represents a range of throughput values in tokens per second. - - Attributes: - min_ (int): - Output only. The minimum value of the range. - max_ (int): - Output only. The maximum value of the range. - """ - - min_: int = proto.Field( - proto.INT32, - number=1, - ) - max_: int = proto.Field( - proto.INT32, - number=2, - ) - - -class MillisecondRange(proto.Message): - r"""Represents a range of latency values in milliseconds. - - Attributes: - min_ (int): - Output only. The minimum value of the range. - max_ (int): - Output only. The maximum value of the range. - """ - - min_: int = proto.Field( - proto.INT32, - number=1, - ) - max_: int = proto.Field( - proto.INT32, - number=2, - ) - - -class PerformanceRange(proto.Message): - r"""Performance range for a model deployment. - - Attributes: - throughput_output_range (google.cloud.gkerecommender_v1.types.TokensPerSecondRange): - Output only. The range of throughput in output tokens per - second. This is measured as - total_output_tokens_generated_by_server / - elapsed_time_in_seconds. - ttft_range (google.cloud.gkerecommender_v1.types.MillisecondRange): - Output only. The range of TTFT (Time To First - Token) in milliseconds. TTFT is the time it - takes to generate the first token for a request. - ntpot_range (google.cloud.gkerecommender_v1.types.MillisecondRange): - Output only. The range of NTPOT (Normalized Time Per Output - Token) in milliseconds. NTPOT is the request latency - normalized by the number of output tokens, measured as - request_latency / total_output_tokens. - """ - - throughput_output_range: "TokensPerSecondRange" = proto.Field( - proto.MESSAGE, - number=1, - message="TokensPerSecondRange", - ) - ttft_range: "MillisecondRange" = proto.Field( - proto.MESSAGE, - number=2, - message="MillisecondRange", - ) - ntpot_range: "MillisecondRange" = proto.Field( - proto.MESSAGE, - number=3, - message="MillisecondRange", - ) - - -class FetchProfilesResponse(proto.Message): - r"""Response message for - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - - Attributes: - profile (MutableSequence[google.cloud.gkerecommender_v1.types.Profile]): - Output only. List of profiles that match the - given model server info and performance - requirements (if provided). - performance_range (google.cloud.gkerecommender_v1.types.PerformanceRange): - Output only. The combined range of - performance values observed across all profiles - in this response. - comments (str): - Output only. Additional comments related to - the response. - next_page_token (str): - Output only. A token which may be sent as - [page_token][FetchProfilesResponse.page_token] in a - subsequent ``FetchProfilesResponse`` call to retrieve the - next page of results. If this field is omitted or empty, - then there are no more results to return. - """ - - @property - def raw_page(self): - return self - - profile: MutableSequence["Profile"] = proto.RepeatedField( - proto.MESSAGE, - number=1, - message="Profile", - ) - performance_range: "PerformanceRange" = proto.Field( - proto.MESSAGE, - number=2, - message="PerformanceRange", - ) - comments: str = proto.Field( - proto.STRING, - number=3, - ) - next_page_token: str = proto.Field( - proto.STRING, - number=4, - ) - - -class ModelServerInfo(proto.Message): - r"""Model server information gives. Valid model server info combinations - can be found using - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles]. - - Attributes: - model (str): - Required. The model. Open-source models follow the - Huggingface Hub ``owner/model_name`` format. Use - [GkeInferenceQuickstart.FetchModels][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels] - to find available models. - model_server (str): - Required. The model server. Open-source model servers use - simplified, lowercase names (e.g., ``vllm``). Use - [GkeInferenceQuickstart.FetchModelServers][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers] - to find available servers. - model_server_version (str): - Optional. The model server version. Use - [GkeInferenceQuickstart.FetchModelServerVersions][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions] - to find available versions. If not provided, the latest - available version is used. - """ - - model: str = proto.Field( - proto.STRING, - number=1, - ) - model_server: str = proto.Field( - proto.STRING, - number=2, - ) - model_server_version: str = proto.Field( - proto.STRING, - number=3, - ) - - -class ResourcesUsed(proto.Message): - r"""Resources used by a model deployment. - - Attributes: - accelerator_count (int): - Output only. The number of accelerators - (e.g., GPUs or TPUs) used by the model - deployment on the Kubernetes node. - """ - - accelerator_count: int = proto.Field( - proto.INT32, - number=1, - ) - - -class PerformanceStats(proto.Message): - r"""Performance statistics for a model deployment. - - Attributes: - queries_per_second (float): - Output only. The number of queries per - second. Note: This metric can vary widely based - on context length and may not be a reliable - measure of LLM throughput. - output_tokens_per_second (int): - Output only. The number of output tokens per second. This is - the throughput measured as - total_output_tokens_generated_by_server / - elapsed_time_in_seconds. - ntpot_milliseconds (int): - Output only. The Normalized Time Per Output Token (NTPOT) in - milliseconds. This is the request latency normalized by the - number of output tokens, measured as request_latency / - total_output_tokens. - ttft_milliseconds (int): - Output only. The Time To First Token (TTFT) - in milliseconds. This is the time it takes to - generate the first token for a request. - cost (MutableSequence[google.cloud.gkerecommender_v1.types.Cost]): - Output only. The cost of running the model - deployment. - """ - - queries_per_second: float = proto.Field( - proto.FLOAT, - number=1, - ) - output_tokens_per_second: int = proto.Field( - proto.INT32, - number=2, - ) - ntpot_milliseconds: int = proto.Field( - proto.INT32, - number=3, - ) - ttft_milliseconds: int = proto.Field( - proto.INT32, - number=4, - ) - cost: MutableSequence["Cost"] = proto.RepeatedField( - proto.MESSAGE, - number=5, - message="Cost", - ) - - -class Profile(proto.Message): - r"""A profile containing information about a model deployment. - - Attributes: - model_server_info (google.cloud.gkerecommender_v1.types.ModelServerInfo): - Output only. The model server configuration. Use - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] - to find valid configurations. - accelerator_type (str): - Output only. The accelerator type. Expected format: - ``nvidia-h100-80gb``. - tpu_topology (str): - Output only. The TPU topology (if - applicable). - instance_type (str): - Output only. The instance type. Expected format: - ``a2-highgpu-1g``. - resources_used (google.cloud.gkerecommender_v1.types.ResourcesUsed): - Output only. The resources used by the model - deployment. - performance_stats (MutableSequence[google.cloud.gkerecommender_v1.types.PerformanceStats]): - Output only. The performance statistics for - this profile. - """ - - model_server_info: "ModelServerInfo" = proto.Field( - proto.MESSAGE, - number=1, - message="ModelServerInfo", - ) - accelerator_type: str = proto.Field( - proto.STRING, - number=2, - ) - tpu_topology: str = proto.Field( - proto.STRING, - number=3, - ) - instance_type: str = proto.Field( - proto.STRING, - number=4, - ) - resources_used: "ResourcesUsed" = proto.Field( - proto.MESSAGE, - number=5, - message="ResourcesUsed", - ) - performance_stats: MutableSequence["PerformanceStats"] = proto.RepeatedField( - proto.MESSAGE, - number=6, - message="PerformanceStats", - ) - - -class GenerateOptimizedManifestRequest(proto.Message): - r"""Request message for - [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. - - Attributes: - model_server_info (google.cloud.gkerecommender_v1.types.ModelServerInfo): - Required. The model server configuration to generate the - manifest for. Use - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] - to find valid configurations. - accelerator_type (str): - Required. The accelerator type. Use - [GkeInferenceQuickstart.FetchProfiles][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles] - to find valid accelerators for a given - ``model_server_info``. - kubernetes_namespace (str): - Optional. The kubernetes namespace to deploy - the manifests in. - performance_requirements (google.cloud.gkerecommender_v1.types.PerformanceRequirements): - Optional. The performance requirements to use - for generating Horizontal Pod Autoscaler (HPA) - resources. If provided, the manifest includes - HPA resources to adjust the model server replica - count to maintain the specified targets (e.g., - NTPOT, TTFT) at a P50 latency. Cost targets are - not currently supported for HPA generation. If - the specified targets are not achievable, the - HPA manifest will not be generated. - storage_config (google.cloud.gkerecommender_v1.types.StorageConfig): - Optional. The storage configuration for the - model. If not provided, the model is loaded from - Huggingface. - """ - - model_server_info: "ModelServerInfo" = proto.Field( - proto.MESSAGE, - number=1, - message="ModelServerInfo", - ) - accelerator_type: str = proto.Field( - proto.STRING, - number=2, - ) - kubernetes_namespace: str = proto.Field( - proto.STRING, - number=3, - ) - performance_requirements: "PerformanceRequirements" = proto.Field( - proto.MESSAGE, - number=4, - message="PerformanceRequirements", - ) - storage_config: "StorageConfig" = proto.Field( - proto.MESSAGE, - number=5, - message="StorageConfig", - ) - - -class KubernetesManifest(proto.Message): - r"""A Kubernetes manifest. - - Attributes: - kind (str): - Output only. Kubernetes resource kind. - api_version (str): - Output only. Kubernetes API version. - content (str): - Output only. YAML content. - """ - - kind: str = proto.Field( - proto.STRING, - number=1, - ) - api_version: str = proto.Field( - proto.STRING, - number=2, - ) - content: str = proto.Field( - proto.STRING, - number=3, - ) - - -class GenerateOptimizedManifestResponse(proto.Message): - r"""Response message for - [GkeInferenceQuickstart.GenerateOptimizedManifest][google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest]. - - Attributes: - kubernetes_manifests (MutableSequence[google.cloud.gkerecommender_v1.types.KubernetesManifest]): - Output only. A list of generated Kubernetes - manifests. - comments (MutableSequence[str]): - Output only. Comments related to deploying - the generated manifests. - manifest_version (str): - Output only. Additional information about the versioned - dependencies used to generate the manifests. See `Run best - practice inference with GKE Inference Quickstart - recipes `__ - for details. - """ - - kubernetes_manifests: MutableSequence["KubernetesManifest"] = proto.RepeatedField( - proto.MESSAGE, - number=1, - message="KubernetesManifest", - ) - comments: MutableSequence[str] = proto.RepeatedField( - proto.STRING, - number=2, - ) - manifest_version: str = proto.Field( - proto.STRING, - number=3, - ) - - -class StorageConfig(proto.Message): - r"""Storage configuration for a model deployment. - - Attributes: - model_bucket_uri (str): - Optional. The Google Cloud Storage bucket URI to load the - model from. This URI must point to the directory containing - the model's config file (``config.json``) and model weights. - A tuned GCSFuse setup can improve LLM Pod startup time by - more than 7x. Expected format: - ``gs:///``. - xla_cache_bucket_uri (str): - Optional. The URI for the GCS bucket containing the XLA - compilation cache. If using TPUs, the XLA cache will be - written to the same path as ``model_bucket_uri``. This can - speed up vLLM model preparation for repeated deployments. - """ - - model_bucket_uri: str = proto.Field( - proto.STRING, - number=1, - ) - xla_cache_bucket_uri: str = proto.Field( - proto.STRING, - number=2, - ) - - -__all__ = tuple(sorted(__protobuf__.manifest)) diff --git a/packages/google-cloud-gkerecommender/mypy.ini b/packages/google-cloud-gkerecommender/mypy.ini deleted file mode 100644 index 574c5aed394b..000000000000 --- a/packages/google-cloud-gkerecommender/mypy.ini +++ /dev/null @@ -1,3 +0,0 @@ -[mypy] -python_version = 3.7 -namespace_packages = True diff --git a/packages/google-cloud-gkerecommender/noxfile.py b/packages/google-cloud-gkerecommender/noxfile.py deleted file mode 100644 index abcdd0db1687..000000000000 --- a/packages/google-cloud-gkerecommender/noxfile.py +++ /dev/null @@ -1,592 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import os -import pathlib -import re -import shutil -from typing import Dict, List -import warnings - -import nox - -BLACK_VERSION = "black[jupyter]==23.7.0" -ISORT_VERSION = "isort==5.11.0" - -LINT_PATHS = ["docs", "google", "tests", "noxfile.py", "setup.py"] - -ALL_PYTHON = [ - "3.7", - "3.8", - "3.9", - "3.10", - "3.11", - "3.12", - "3.13", -] - -DEFAULT_PYTHON_VERSION = ALL_PYTHON[-1] - -CURRENT_DIRECTORY = pathlib.Path(__file__).parent.absolute() - -LOWER_BOUND_CONSTRAINTS_FILE = CURRENT_DIRECTORY / "constraints.txt" -PACKAGE_NAME = "google-cloud-gkerecommender" - -UNIT_TEST_STANDARD_DEPENDENCIES = [ - "mock", - "asyncmock", - "pytest", - "pytest-cov", - "pytest-asyncio", -] -UNIT_TEST_EXTERNAL_DEPENDENCIES: List[str] = [] -UNIT_TEST_LOCAL_DEPENDENCIES: List[str] = [] -UNIT_TEST_DEPENDENCIES: List[str] = [] -UNIT_TEST_EXTRAS: List[str] = [] -UNIT_TEST_EXTRAS_BY_PYTHON: Dict[str, List[str]] = {} - -SYSTEM_TEST_PYTHON_VERSIONS: List[str] = ["3.8", "3.9", "3.10", "3.11", "3.12", "3.13"] -SYSTEM_TEST_STANDARD_DEPENDENCIES = [ - "mock", - "pytest", - "google-cloud-testutils", -] -SYSTEM_TEST_EXTERNAL_DEPENDENCIES: List[str] = [] -SYSTEM_TEST_LOCAL_DEPENDENCIES: List[str] = [] -SYSTEM_TEST_DEPENDENCIES: List[str] = [] -SYSTEM_TEST_EXTRAS: List[str] = [] -SYSTEM_TEST_EXTRAS_BY_PYTHON: Dict[str, List[str]] = {} - -nox.options.sessions = [ - "unit", - "system", - "cover", - "lint", - "lint_setup_py", - "blacken", - "docs", -] - -# Error if a python version is missing -nox.options.error_on_missing_interpreters = True - - -@nox.session(python=ALL_PYTHON) -def mypy(session): - """Run the type checker.""" - session.install( - # TODO(https://github.com/googleapis/gapic-generator-python/issues/2410): Use the latest version of mypy - "mypy<1.16.0", - "types-requests", - "types-protobuf", - ) - session.install(".") - session.run( - "mypy", - "-p", - "google", - ) - - -@nox.session -def update_lower_bounds(session): - """Update lower bounds in constraints.txt to match setup.py""" - session.install("google-cloud-testutils") - session.install(".") - - session.run( - "lower-bound-checker", - "update", - "--package-name", - PACKAGE_NAME, - "--constraints-file", - str(LOWER_BOUND_CONSTRAINTS_FILE), - ) - - -@nox.session -def check_lower_bounds(session): - """Check lower bounds in setup.py are reflected in constraints file""" - session.install("google-cloud-testutils") - session.install(".") - - session.run( - "lower-bound-checker", - "check", - "--package-name", - PACKAGE_NAME, - "--constraints-file", - str(LOWER_BOUND_CONSTRAINTS_FILE), - ) - - -@nox.session(python=DEFAULT_PYTHON_VERSION) -def lint(session): - """Run linters. - - Returns a failure if the linters find linting errors or sufficiently - serious code quality issues. - """ - session.install("flake8", BLACK_VERSION) - session.run( - "black", - "--check", - *LINT_PATHS, - ) - - session.run("flake8", "google", "tests") - - -@nox.session(python=DEFAULT_PYTHON_VERSION) -def blacken(session): - """Run black. Format code to uniform standard.""" - session.install(BLACK_VERSION) - session.run( - "black", - *LINT_PATHS, - ) - - -@nox.session(python=DEFAULT_PYTHON_VERSION) -def format(session): - """ - Run isort to sort imports. Then run black - to format code to uniform standard. - """ - session.install(BLACK_VERSION, ISORT_VERSION) - # Use the --fss option to sort imports using strict alphabetical order. - # See https://pycqa.github.io/isort/docs/configuration/options.html#force-sort-within-sections - session.run( - "isort", - "--fss", - *LINT_PATHS, - ) - session.run( - "black", - *LINT_PATHS, - ) - - -@nox.session(python=DEFAULT_PYTHON_VERSION) -def lint_setup_py(session): - """Verify that setup.py is valid (including RST check).""" - session.install("setuptools", "docutils", "pygments") - session.run("python", "setup.py", "check", "--restructuredtext", "--strict") - - -def install_unittest_dependencies(session, *constraints): - standard_deps = UNIT_TEST_STANDARD_DEPENDENCIES + UNIT_TEST_DEPENDENCIES - session.install(*standard_deps, *constraints) - - if UNIT_TEST_EXTERNAL_DEPENDENCIES: - warnings.warn( - "'unit_test_external_dependencies' is deprecated. Instead, please " - "use 'unit_test_dependencies' or 'unit_test_local_dependencies'.", - DeprecationWarning, - ) - session.install(*UNIT_TEST_EXTERNAL_DEPENDENCIES, *constraints) - - if UNIT_TEST_LOCAL_DEPENDENCIES: - session.install(*UNIT_TEST_LOCAL_DEPENDENCIES, *constraints) - - if UNIT_TEST_EXTRAS_BY_PYTHON: - extras = UNIT_TEST_EXTRAS_BY_PYTHON.get(session.python, []) - elif UNIT_TEST_EXTRAS: - extras = UNIT_TEST_EXTRAS - else: - extras = [] - - if extras: - session.install("-e", f".[{','.join(extras)}]", *constraints) - else: - session.install("-e", ".", *constraints) - - -@nox.session(python=ALL_PYTHON) -@nox.parametrize( - "protobuf_implementation", - ["python", "upb", "cpp"], -) -def unit(session, protobuf_implementation): - # Install all test dependencies, then install this package in-place. - - if protobuf_implementation == "cpp" and session.python in ("3.11", "3.12", "3.13"): - session.skip("cpp implementation is not supported in python 3.11+") - - constraints_path = str( - CURRENT_DIRECTORY / "testing" / f"constraints-{session.python}.txt" - ) - install_unittest_dependencies(session, "-c", constraints_path) - - # TODO(https://github.com/googleapis/synthtool/issues/1976): - # Remove the 'cpp' implementation once support for Protobuf 3.x is dropped. - # The 'cpp' implementation requires Protobuf<4. - if protobuf_implementation == "cpp": - session.install("protobuf<4") - - # Run py.test against the unit tests. - session.run( - "py.test", - "--quiet", - f"--junitxml=unit_{session.python}_sponge_log.xml", - "--cov=google", - "--cov=tests/unit", - "--cov-append", - "--cov-config=.coveragerc", - "--cov-report=", - "--cov-fail-under=0", - os.path.join("tests", "unit"), - *session.posargs, - env={ - "PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION": protobuf_implementation, - }, - ) - - -def install_systemtest_dependencies(session, *constraints): - session.install("--pre", "grpcio") - - session.install(*SYSTEM_TEST_STANDARD_DEPENDENCIES, *constraints) - - if SYSTEM_TEST_EXTERNAL_DEPENDENCIES: - session.install(*SYSTEM_TEST_EXTERNAL_DEPENDENCIES, *constraints) - - if SYSTEM_TEST_LOCAL_DEPENDENCIES: - session.install("-e", *SYSTEM_TEST_LOCAL_DEPENDENCIES, *constraints) - - if SYSTEM_TEST_DEPENDENCIES: - session.install("-e", *SYSTEM_TEST_DEPENDENCIES, *constraints) - - if SYSTEM_TEST_EXTRAS_BY_PYTHON: - extras = SYSTEM_TEST_EXTRAS_BY_PYTHON.get(session.python, []) - elif SYSTEM_TEST_EXTRAS: - extras = SYSTEM_TEST_EXTRAS - else: - extras = [] - - if extras: - session.install("-e", f".[{','.join(extras)}]", *constraints) - else: - session.install("-e", ".", *constraints) - - -@nox.session(python=SYSTEM_TEST_PYTHON_VERSIONS) -def system(session): - """Run the system test suite.""" - constraints_path = str( - CURRENT_DIRECTORY / "testing" / f"constraints-{session.python}.txt" - ) - system_test_path = os.path.join("tests", "system.py") - system_test_folder_path = os.path.join("tests", "system") - - # Check the value of `RUN_SYSTEM_TESTS` env var. It defaults to true. - if os.environ.get("RUN_SYSTEM_TESTS", "true") == "false": - session.skip("RUN_SYSTEM_TESTS is set to false, skipping") - # Install pyopenssl for mTLS testing. - if os.environ.get("GOOGLE_API_USE_CLIENT_CERTIFICATE", "false") == "true": - session.install("pyopenssl") - - system_test_exists = os.path.exists(system_test_path) - system_test_folder_exists = os.path.exists(system_test_folder_path) - # Sanity check: only run tests if found. - if not system_test_exists and not system_test_folder_exists: - session.skip("System tests were not found") - - install_systemtest_dependencies(session, "-c", constraints_path) - - # Run py.test against the system tests. - if system_test_exists: - session.run( - "py.test", - "--quiet", - f"--junitxml=system_{session.python}_sponge_log.xml", - system_test_path, - *session.posargs, - ) - if system_test_folder_exists: - session.run( - "py.test", - "--quiet", - f"--junitxml=system_{session.python}_sponge_log.xml", - system_test_folder_path, - *session.posargs, - ) - - -@nox.session(python=DEFAULT_PYTHON_VERSION) -def cover(session): - """Run the final coverage report. - - This outputs the coverage report aggregating coverage from the unit - test runs (not system test runs), and then erases coverage data. - """ - session.install("coverage", "pytest-cov") - session.run("coverage", "report", "--show-missing", "--fail-under=100") - - session.run("coverage", "erase") - - -@nox.session(python="3.10") -def docs(session): - """Build the docs for this library.""" - - session.install("-e", ".") - session.install( - # We need to pin to specific versions of the `sphinxcontrib-*` packages - # which still support sphinx 4.x. - # See https://github.com/googleapis/sphinx-docfx-yaml/issues/344 - # and https://github.com/googleapis/sphinx-docfx-yaml/issues/345. - "sphinxcontrib-applehelp==1.0.4", - "sphinxcontrib-devhelp==1.0.2", - "sphinxcontrib-htmlhelp==2.0.1", - "sphinxcontrib-qthelp==1.0.3", - "sphinxcontrib-serializinghtml==1.1.5", - "sphinx==4.5.0", - "alabaster", - "recommonmark", - ) - - shutil.rmtree(os.path.join("docs", "_build"), ignore_errors=True) - session.run( - "sphinx-build", - "-W", # warnings as errors - "-T", # show full traceback on exception - "-N", # no colors - "-b", - "html", # builder - "-d", - os.path.join("docs", "_build", "doctrees", ""), # cache directory - # paths to build: - os.path.join("docs", ""), - os.path.join("docs", "_build", "html", ""), - ) - - -@nox.session(python="3.10") -def docfx(session): - """Build the docfx yaml files for this library.""" - - session.install("-e", ".") - session.install( - # We need to pin to specific versions of the `sphinxcontrib-*` packages - # which still support sphinx 4.x. - # See https://github.com/googleapis/sphinx-docfx-yaml/issues/344 - # and https://github.com/googleapis/sphinx-docfx-yaml/issues/345. - "sphinxcontrib-applehelp==1.0.4", - "sphinxcontrib-devhelp==1.0.2", - "sphinxcontrib-htmlhelp==2.0.1", - "sphinxcontrib-qthelp==1.0.3", - "sphinxcontrib-serializinghtml==1.1.5", - "gcp-sphinx-docfx-yaml", - "alabaster", - "recommonmark", - ) - - shutil.rmtree(os.path.join("docs", "_build"), ignore_errors=True) - session.run( - "sphinx-build", - "-T", # show full traceback on exception - "-N", # no colors - "-D", - ( - "extensions=sphinx.ext.autodoc," - "sphinx.ext.autosummary," - "docfx_yaml.extension," - "sphinx.ext.intersphinx," - "sphinx.ext.coverage," - "sphinx.ext.napoleon," - "sphinx.ext.todo," - "sphinx.ext.viewcode," - "recommonmark" - ), - "-b", - "html", - "-d", - os.path.join("docs", "_build", "doctrees", ""), - os.path.join("docs", ""), - os.path.join("docs", "_build", "html", ""), - ) - - -@nox.session(python=DEFAULT_PYTHON_VERSION) -@nox.parametrize( - "protobuf_implementation", - ["python", "upb", "cpp"], -) -def prerelease_deps(session, protobuf_implementation): - """ - Run all tests with pre-release versions of dependencies installed - rather than the standard non pre-release versions. - Pre-release versions can be installed using - `pip install --pre `. - """ - - if protobuf_implementation == "cpp" and session.python in ("3.11", "3.12", "3.13"): - session.skip("cpp implementation is not supported in python 3.11+") - - # Install all dependencies - session.install("-e", ".") - - # Install dependencies for the unit test environment - unit_deps_all = UNIT_TEST_STANDARD_DEPENDENCIES + UNIT_TEST_EXTERNAL_DEPENDENCIES - session.install(*unit_deps_all) - - # Install dependencies for the system test environment - system_deps_all = ( - SYSTEM_TEST_STANDARD_DEPENDENCIES - + SYSTEM_TEST_EXTERNAL_DEPENDENCIES - + SYSTEM_TEST_EXTRAS - ) - session.install(*system_deps_all) - - # Because we test minimum dependency versions on the minimum Python - # version, the first version we test with in the unit tests sessions has a - # constraints file containing all dependencies and extras. - with open( - CURRENT_DIRECTORY / "testing" / f"constraints-{ALL_PYTHON[0]}.txt", - encoding="utf-8", - ) as constraints_file: - constraints_text = constraints_file.read() - - # Ignore leading whitespace and comment lines. - constraints_deps = [ - match.group(1) - for match in re.finditer( - r"^\s*(\S+)(?===\S+)", constraints_text, flags=re.MULTILINE - ) - ] - - # Install dependencies specified in `testing/constraints-X.txt`. - session.install(*constraints_deps) - - # Note: If a dependency is added to the `prerel_deps` list, - # the `core_dependencies_from_source` list in the `core_deps_from_source` - # nox session should also be updated. - prerel_deps = [ - "googleapis-common-protos", - "google-api-core", - "google-auth", - "grpc-google-iam-v1", - "grpcio", - "grpcio-status", - "protobuf", - "proto-plus", - ] - - for dep in prerel_deps: - session.install("--pre", "--no-deps", "--ignore-installed", dep) - # TODO(https://github.com/grpc/grpc/issues/38965): Add `grpcio-status`` - # to the dictionary below once this bug is fixed. - # TODO(https://github.com/googleapis/google-cloud-python/issues/13643): Add - # `googleapis-common-protos` and `grpc-google-iam-v1` to the dictionary below - # once this bug is fixed. - package_namespaces = { - "google-api-core": "google.api_core", - "google-auth": "google.auth", - "grpcio": "grpc", - "protobuf": "google.protobuf", - "proto-plus": "proto", - } - - version_namespace = package_namespaces.get(dep) - - print(f"Installed {dep}") - if version_namespace: - session.run( - "python", - "-c", - f"import {version_namespace}; print({version_namespace}.__version__)", - ) - - session.run( - "py.test", - "tests/unit", - env={ - "PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION": protobuf_implementation, - }, - ) - - -@nox.session(python=DEFAULT_PYTHON_VERSION) -@nox.parametrize( - "protobuf_implementation", - ["python", "upb"], -) -def core_deps_from_source(session, protobuf_implementation): - """Run all tests with core dependencies installed from source - rather than pulling the dependencies from PyPI. - """ - - # Install all dependencies - session.install("-e", ".") - - # Install dependencies for the unit test environment - unit_deps_all = UNIT_TEST_STANDARD_DEPENDENCIES + UNIT_TEST_EXTERNAL_DEPENDENCIES - session.install(*unit_deps_all) - - # Install dependencies for the system test environment - system_deps_all = ( - SYSTEM_TEST_STANDARD_DEPENDENCIES - + SYSTEM_TEST_EXTERNAL_DEPENDENCIES - + SYSTEM_TEST_EXTRAS - ) - session.install(*system_deps_all) - - # Because we test minimum dependency versions on the minimum Python - # version, the first version we test with in the unit tests sessions has a - # constraints file containing all dependencies and extras. - with open( - CURRENT_DIRECTORY / "testing" / f"constraints-{ALL_PYTHON[0]}.txt", - encoding="utf-8", - ) as constraints_file: - constraints_text = constraints_file.read() - - # Ignore leading whitespace and comment lines. - constraints_deps = [ - match.group(1) - for match in re.finditer( - r"^\s*(\S+)(?===\S+)", constraints_text, flags=re.MULTILINE - ) - ] - - # Install dependencies specified in `testing/constraints-X.txt`. - session.install(*constraints_deps) - - # TODO(https://github.com/googleapis/gapic-generator-python/issues/2358): `grpcio` and - # `grpcio-status` should be added to the list below so that they are installed from source, - # rather than PyPI. - # TODO(https://github.com/googleapis/gapic-generator-python/issues/2357): `protobuf` should be - # added to the list below so that it is installed from source, rather than PyPI - # Note: If a dependency is added to the `core_dependencies_from_source` list, - # the `prerel_deps` list in the `prerelease_deps` nox session should also be updated. - core_dependencies_from_source = [ - "googleapis-common-protos @ git+https://github.com/googleapis/google-cloud-python#egg=googleapis-common-protos&subdirectory=packages/googleapis-common-protos", - "google-api-core @ git+https://github.com/googleapis/python-api-core.git", - "google-auth @ git+https://github.com/googleapis/google-auth-library-python.git", - "grpc-google-iam-v1 @ git+https://github.com/googleapis/google-cloud-python#egg=grpc-google-iam-v1&subdirectory=packages/grpc-google-iam-v1", - "proto-plus @ git+https://github.com/googleapis/proto-plus-python.git", - ] - - for dep in core_dependencies_from_source: - session.install(dep, "--no-deps", "--ignore-installed") - print(f"Installed {dep}") - - session.run( - "py.test", - "tests/unit", - env={ - "PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION": protobuf_implementation, - }, - ) diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py deleted file mode 100644 index f10a415865e0..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py +++ /dev/null @@ -1,56 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchBenchmarkingData -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_async] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -async def sample_fetch_benchmarking_data(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - model_server_info = gkerecommender_v1.ModelServerInfo() - model_server_info.model = "model_value" - model_server_info.model_server = "model_server_value" - - request = gkerecommender_v1.FetchBenchmarkingDataRequest( - model_server_info=model_server_info, - ) - - # Make the request - response = await client.fetch_benchmarking_data(request=request) - - # Handle the response - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py deleted file mode 100644 index dab6bce72582..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py +++ /dev/null @@ -1,56 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchBenchmarkingData -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_sync] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -def sample_fetch_benchmarking_data(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - model_server_info = gkerecommender_v1.ModelServerInfo() - model_server_info.model = "model_value" - model_server_info.model_server = "model_server_value" - - request = gkerecommender_v1.FetchBenchmarkingDataRequest( - model_server_info=model_server_info, - ) - - # Make the request - response = client.fetch_benchmarking_data(request=request) - - # Handle the response - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py deleted file mode 100644 index 81463976c4c9..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py +++ /dev/null @@ -1,54 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchModelServerVersions -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_async] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -async def sample_fetch_model_server_versions(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelServerVersionsRequest( - model="model_value", - model_server="model_server_value", - ) - - # Make the request - page_result = client.fetch_model_server_versions(request=request) - - # Handle the response - async for response in page_result: - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py deleted file mode 100644 index 9bc645591a20..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py +++ /dev/null @@ -1,54 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchModelServerVersions -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_sync] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -def sample_fetch_model_server_versions(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelServerVersionsRequest( - model="model_value", - model_server="model_server_value", - ) - - # Make the request - page_result = client.fetch_model_server_versions(request=request) - - # Handle the response - for response in page_result: - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py deleted file mode 100644 index 648669d45067..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py +++ /dev/null @@ -1,53 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchModelServers -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_async] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -async def sample_fetch_model_servers(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelServersRequest( - model="model_value", - ) - - # Make the request - page_result = client.fetch_model_servers(request=request) - - # Handle the response - async for response in page_result: - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py deleted file mode 100644 index ddbf9d98c5b7..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py +++ /dev/null @@ -1,53 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchModelServers -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_sync] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -def sample_fetch_model_servers(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelServersRequest( - model="model_value", - ) - - # Make the request - page_result = client.fetch_model_servers(request=request) - - # Handle the response - for response in page_result: - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py deleted file mode 100644 index 651f2483904f..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py +++ /dev/null @@ -1,52 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchModels -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_async] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -async def sample_fetch_models(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelsRequest( - ) - - # Make the request - page_result = client.fetch_models(request=request) - - # Handle the response - async for response in page_result: - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py deleted file mode 100644 index 357104b82387..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py +++ /dev/null @@ -1,52 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchModels -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_sync] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -def sample_fetch_models(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchModelsRequest( - ) - - # Make the request - page_result = client.fetch_models(request=request) - - # Handle the response - for response in page_result: - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py deleted file mode 100644 index e218622fe416..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py +++ /dev/null @@ -1,52 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchProfiles -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_async] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -async def sample_fetch_profiles(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchProfilesRequest( - ) - - # Make the request - page_result = client.fetch_profiles(request=request) - - # Handle the response - async for response in page_result: - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py deleted file mode 100644 index e9a6dea821f2..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py +++ /dev/null @@ -1,52 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for FetchProfiles -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_sync] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -def sample_fetch_profiles(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - request = gkerecommender_v1.FetchProfilesRequest( - ) - - # Make the request - page_result = client.fetch_profiles(request=request) - - # Handle the response - for response in page_result: - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py deleted file mode 100644 index 286c4c22a63c..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py +++ /dev/null @@ -1,57 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for GenerateOptimizedManifest -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_async] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -async def sample_generate_optimized_manifest(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartAsyncClient() - - # Initialize request argument(s) - model_server_info = gkerecommender_v1.ModelServerInfo() - model_server_info.model = "model_value" - model_server_info.model_server = "model_server_value" - - request = gkerecommender_v1.GenerateOptimizedManifestRequest( - model_server_info=model_server_info, - accelerator_type="accelerator_type_value", - ) - - # Make the request - response = await client.generate_optimized_manifest(request=request) - - # Handle the response - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_async] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py b/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py deleted file mode 100644 index 4d1362ea3b17..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py +++ /dev/null @@ -1,57 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -# Generated code. DO NOT EDIT! -# -# Snippet for GenerateOptimizedManifest -# NOTE: This snippet has been automatically generated for illustrative purposes only. -# It may require modifications to work in your environment. - -# To install the latest published package dependency, execute the following: -# python3 -m pip install google-cloud-gkerecommender - - -# [START gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_sync] -# This snippet has been automatically generated and should be regarded as a -# code template only. -# It will require modifications to work: -# - It may require correct/in-range values for request initialization. -# - It may require specifying regional endpoints when creating the service -# client as shown in: -# https://googleapis.dev/python/google-api-core/latest/client_options.html -from google.cloud import gkerecommender_v1 - - -def sample_generate_optimized_manifest(): - # Create a client - client = gkerecommender_v1.GkeInferenceQuickstartClient() - - # Initialize request argument(s) - model_server_info = gkerecommender_v1.ModelServerInfo() - model_server_info.model = "model_value" - model_server_info.model_server = "model_server_value" - - request = gkerecommender_v1.GenerateOptimizedManifestRequest( - model_server_info=model_server_info, - accelerator_type="accelerator_type_value", - ) - - # Make the request - response = client.generate_optimized_manifest(request=request) - - # Handle the response - print(response) - -# [END gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_sync] diff --git a/packages/google-cloud-gkerecommender/samples/generated_samples/snippet_metadata_google.cloud.gkerecommender.v1.json b/packages/google-cloud-gkerecommender/samples/generated_samples/snippet_metadata_google.cloud.gkerecommender.v1.json deleted file mode 100644 index d3eacb5ac8dd..000000000000 --- a/packages/google-cloud-gkerecommender/samples/generated_samples/snippet_metadata_google.cloud.gkerecommender.v1.json +++ /dev/null @@ -1,933 +0,0 @@ -{ - "clientLibrary": { - "apis": [ - { - "id": "google.cloud.gkerecommender.v1", - "version": "v1" - } - ], - "language": "PYTHON", - "name": "google-cloud-gkerecommender", - "version": "0.0.0" - }, - "snippets": [ - { - "canonical": true, - "clientMethod": { - "async": true, - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", - "shortName": "GkeInferenceQuickstartAsyncClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_benchmarking_data", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchBenchmarkingData" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse", - "shortName": "fetch_benchmarking_data" - }, - "description": "Sample for FetchBenchmarkingData", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_async", - "segments": [ - { - "end": 55, - "start": 27, - "type": "FULL" - }, - { - "end": 55, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 49, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 52, - "start": 50, - "type": "REQUEST_EXECUTION" - }, - { - "end": 56, - "start": 53, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_async.py" - }, - { - "canonical": true, - "clientMethod": { - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", - "shortName": "GkeInferenceQuickstartClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_benchmarking_data", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchBenchmarkingData", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchBenchmarkingData" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse", - "shortName": "fetch_benchmarking_data" - }, - "description": "Sample for FetchBenchmarkingData", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchBenchmarkingData_sync", - "segments": [ - { - "end": 55, - "start": 27, - "type": "FULL" - }, - { - "end": 55, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 49, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 52, - "start": 50, - "type": "REQUEST_EXECUTION" - }, - { - "end": 56, - "start": 53, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_benchmarking_data_sync.py" - }, - { - "canonical": true, - "clientMethod": { - "async": true, - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", - "shortName": "GkeInferenceQuickstartAsyncClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_model_server_versions", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchModelServerVersions" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsAsyncPager", - "shortName": "fetch_model_server_versions" - }, - "description": "Sample for FetchModelServerVersions", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_async", - "segments": [ - { - "end": 53, - "start": 27, - "type": "FULL" - }, - { - "end": 53, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 46, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 49, - "start": 47, - "type": "REQUEST_EXECUTION" - }, - { - "end": 54, - "start": 50, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_async.py" - }, - { - "canonical": true, - "clientMethod": { - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", - "shortName": "GkeInferenceQuickstartClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_model_server_versions", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServerVersions", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchModelServerVersions" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsPager", - "shortName": "fetch_model_server_versions" - }, - "description": "Sample for FetchModelServerVersions", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServerVersions_sync", - "segments": [ - { - "end": 53, - "start": 27, - "type": "FULL" - }, - { - "end": 53, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 46, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 49, - "start": 47, - "type": "REQUEST_EXECUTION" - }, - { - "end": 54, - "start": 50, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_server_versions_sync.py" - }, - { - "canonical": true, - "clientMethod": { - "async": true, - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", - "shortName": "GkeInferenceQuickstartAsyncClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_model_servers", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchModelServers" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchModelServersRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersAsyncPager", - "shortName": "fetch_model_servers" - }, - "description": "Sample for FetchModelServers", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_async", - "segments": [ - { - "end": 52, - "start": 27, - "type": "FULL" - }, - { - "end": 52, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 45, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 48, - "start": 46, - "type": "REQUEST_EXECUTION" - }, - { - "end": 53, - "start": 49, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_async.py" - }, - { - "canonical": true, - "clientMethod": { - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", - "shortName": "GkeInferenceQuickstartClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_model_servers", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModelServers", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchModelServers" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchModelServersRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersPager", - "shortName": "fetch_model_servers" - }, - "description": "Sample for FetchModelServers", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModelServers_sync", - "segments": [ - { - "end": 52, - "start": 27, - "type": "FULL" - }, - { - "end": 52, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 45, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 48, - "start": 46, - "type": "REQUEST_EXECUTION" - }, - { - "end": 53, - "start": 49, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_model_servers_sync.py" - }, - { - "canonical": true, - "clientMethod": { - "async": true, - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", - "shortName": "GkeInferenceQuickstartAsyncClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_models", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchModels" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchModelsRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsAsyncPager", - "shortName": "fetch_models" - }, - "description": "Sample for FetchModels", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_async", - "segments": [ - { - "end": 51, - "start": 27, - "type": "FULL" - }, - { - "end": 51, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 44, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 47, - "start": 45, - "type": "REQUEST_EXECUTION" - }, - { - "end": 52, - "start": 48, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_async.py" - }, - { - "canonical": true, - "clientMethod": { - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", - "shortName": "GkeInferenceQuickstartClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_models", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchModels", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchModels" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchModelsRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsPager", - "shortName": "fetch_models" - }, - "description": "Sample for FetchModels", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchModels_sync", - "segments": [ - { - "end": 51, - "start": 27, - "type": "FULL" - }, - { - "end": 51, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 44, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 47, - "start": 45, - "type": "REQUEST_EXECUTION" - }, - { - "end": 52, - "start": 48, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_models_sync.py" - }, - { - "canonical": true, - "clientMethod": { - "async": true, - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", - "shortName": "GkeInferenceQuickstartAsyncClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.fetch_profiles", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchProfiles" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchProfilesRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesAsyncPager", - "shortName": "fetch_profiles" - }, - "description": "Sample for FetchProfiles", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_async", - "segments": [ - { - "end": 51, - "start": 27, - "type": "FULL" - }, - { - "end": 51, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 44, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 47, - "start": 45, - "type": "REQUEST_EXECUTION" - }, - { - "end": 52, - "start": 48, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_async.py" - }, - { - "canonical": true, - "clientMethod": { - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", - "shortName": "GkeInferenceQuickstartClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.fetch_profiles", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.FetchProfiles", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "FetchProfiles" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.FetchProfilesRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesPager", - "shortName": "fetch_profiles" - }, - "description": "Sample for FetchProfiles", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_FetchProfiles_sync", - "segments": [ - { - "end": 51, - "start": 27, - "type": "FULL" - }, - { - "end": 51, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 44, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 47, - "start": 45, - "type": "REQUEST_EXECUTION" - }, - { - "end": 52, - "start": 48, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_fetch_profiles_sync.py" - }, - { - "canonical": true, - "clientMethod": { - "async": true, - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient", - "shortName": "GkeInferenceQuickstartAsyncClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartAsyncClient.generate_optimized_manifest", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "GenerateOptimizedManifest" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse", - "shortName": "generate_optimized_manifest" - }, - "description": "Sample for GenerateOptimizedManifest", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_async", - "segments": [ - { - "end": 56, - "start": 27, - "type": "FULL" - }, - { - "end": 56, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 50, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 53, - "start": 51, - "type": "REQUEST_EXECUTION" - }, - { - "end": 57, - "start": 54, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_async.py" - }, - { - "canonical": true, - "clientMethod": { - "client": { - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient", - "shortName": "GkeInferenceQuickstartClient" - }, - "fullName": "google.cloud.gkerecommender_v1.GkeInferenceQuickstartClient.generate_optimized_manifest", - "method": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart.GenerateOptimizedManifest", - "service": { - "fullName": "google.cloud.gkerecommender.v1.GkeInferenceQuickstart", - "shortName": "GkeInferenceQuickstart" - }, - "shortName": "GenerateOptimizedManifest" - }, - "parameters": [ - { - "name": "request", - "type": "google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest" - }, - { - "name": "retry", - "type": "google.api_core.retry.Retry" - }, - { - "name": "timeout", - "type": "float" - }, - { - "name": "metadata", - "type": "Sequence[Tuple[str, Union[str, bytes]]]" - } - ], - "resultType": "google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse", - "shortName": "generate_optimized_manifest" - }, - "description": "Sample for GenerateOptimizedManifest", - "file": "gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py", - "language": "PYTHON", - "origin": "API_DEFINITION", - "regionTag": "gkerecommender_v1_generated_GkeInferenceQuickstart_GenerateOptimizedManifest_sync", - "segments": [ - { - "end": 56, - "start": 27, - "type": "FULL" - }, - { - "end": 56, - "start": 27, - "type": "SHORT" - }, - { - "end": 40, - "start": 38, - "type": "CLIENT_INITIALIZATION" - }, - { - "end": 50, - "start": 41, - "type": "REQUEST_INITIALIZATION" - }, - { - "end": 53, - "start": 51, - "type": "REQUEST_EXECUTION" - }, - { - "end": 57, - "start": 54, - "type": "RESPONSE_HANDLING" - } - ], - "title": "gkerecommender_v1_generated_gke_inference_quickstart_generate_optimized_manifest_sync.py" - } - ] -} diff --git a/packages/google-cloud-gkerecommender/scripts/fixup_gkerecommender_v1_keywords.py b/packages/google-cloud-gkerecommender/scripts/fixup_gkerecommender_v1_keywords.py deleted file mode 100644 index 4c43993a2015..000000000000 --- a/packages/google-cloud-gkerecommender/scripts/fixup_gkerecommender_v1_keywords.py +++ /dev/null @@ -1,181 +0,0 @@ -#! /usr/bin/env python3 -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import argparse -import os -import libcst as cst -import pathlib -import sys -from typing import (Any, Callable, Dict, List, Sequence, Tuple) - - -def partition( - predicate: Callable[[Any], bool], - iterator: Sequence[Any] -) -> Tuple[List[Any], List[Any]]: - """A stable, out-of-place partition.""" - results = ([], []) - - for i in iterator: - results[int(predicate(i))].append(i) - - # Returns trueList, falseList - return results[1], results[0] - - -class gkerecommenderCallTransformer(cst.CSTTransformer): - CTRL_PARAMS: Tuple[str] = ('retry', 'timeout', 'metadata') - METHOD_TO_PARAMS: Dict[str, Tuple[str]] = { - 'fetch_benchmarking_data': ('model_server_info', 'instance_type', 'pricing_model', ), - 'fetch_models': ('page_size', 'page_token', ), - 'fetch_model_servers': ('model', 'page_size', 'page_token', ), - 'fetch_model_server_versions': ('model', 'model_server', 'page_size', 'page_token', ), - 'fetch_profiles': ('model', 'model_server', 'model_server_version', 'performance_requirements', 'page_size', 'page_token', ), - 'generate_optimized_manifest': ('model_server_info', 'accelerator_type', 'kubernetes_namespace', 'performance_requirements', 'storage_config', ), - } - - def leave_Call(self, original: cst.Call, updated: cst.Call) -> cst.CSTNode: - try: - key = original.func.attr.value - kword_params = self.METHOD_TO_PARAMS[key] - except (AttributeError, KeyError): - # Either not a method from the API or too convoluted to be sure. - return updated - - # If the existing code is valid, keyword args come after positional args. - # Therefore, all positional args must map to the first parameters. - args, kwargs = partition(lambda a: not bool(a.keyword), updated.args) - if any(k.keyword.value == "request" for k in kwargs): - # We've already fixed this file, don't fix it again. - return updated - - kwargs, ctrl_kwargs = partition( - lambda a: a.keyword.value not in self.CTRL_PARAMS, - kwargs - ) - - args, ctrl_args = args[:len(kword_params)], args[len(kword_params):] - ctrl_kwargs.extend(cst.Arg(value=a.value, keyword=cst.Name(value=ctrl)) - for a, ctrl in zip(ctrl_args, self.CTRL_PARAMS)) - - request_arg = cst.Arg( - value=cst.Dict([ - cst.DictElement( - cst.SimpleString("'{}'".format(name)), -cst.Element(value=arg.value) - ) - # Note: the args + kwargs looks silly, but keep in mind that - # the control parameters had to be stripped out, and that - # those could have been passed positionally or by keyword. - for name, arg in zip(kword_params, args + kwargs)]), - keyword=cst.Name("request") - ) - - return updated.with_changes( - args=[request_arg] + ctrl_kwargs - ) - - -def fix_files( - in_dir: pathlib.Path, - out_dir: pathlib.Path, - *, - transformer=gkerecommenderCallTransformer(), -): - """Duplicate the input dir to the output dir, fixing file method calls. - - Preconditions: - * in_dir is a real directory - * out_dir is a real, empty directory - """ - pyfile_gen = ( - pathlib.Path(os.path.join(root, f)) - for root, _, files in os.walk(in_dir) - for f in files if os.path.splitext(f)[1] == ".py" - ) - - for fpath in pyfile_gen: - with open(fpath, 'r') as f: - src = f.read() - - # Parse the code and insert method call fixes. - tree = cst.parse_module(src) - updated = tree.visit(transformer) - - # Create the path and directory structure for the new file. - updated_path = out_dir.joinpath(fpath.relative_to(in_dir)) - updated_path.parent.mkdir(parents=True, exist_ok=True) - - # Generate the updated source file at the corresponding path. - with open(updated_path, 'w') as f: - f.write(updated.code) - - -if __name__ == '__main__': - parser = argparse.ArgumentParser( - description="""Fix up source that uses the gkerecommender client library. - -The existing sources are NOT overwritten but are copied to output_dir with changes made. - -Note: This tool operates at a best-effort level at converting positional - parameters in client method calls to keyword based parameters. - Cases where it WILL FAIL include - A) * or ** expansion in a method call. - B) Calls via function or method alias (includes free function calls) - C) Indirect or dispatched calls (e.g. the method is looked up dynamically) - - These all constitute false negatives. The tool will also detect false - positives when an API method shares a name with another method. -""") - parser.add_argument( - '-d', - '--input-directory', - required=True, - dest='input_dir', - help='the input directory to walk for python files to fix up', - ) - parser.add_argument( - '-o', - '--output-directory', - required=True, - dest='output_dir', - help='the directory to output files fixed via un-flattening', - ) - args = parser.parse_args() - input_dir = pathlib.Path(args.input_dir) - output_dir = pathlib.Path(args.output_dir) - if not input_dir.is_dir(): - print( - f"input directory '{input_dir}' does not exist or is not a directory", - file=sys.stderr, - ) - sys.exit(-1) - - if not output_dir.is_dir(): - print( - f"output directory '{output_dir}' does not exist or is not a directory", - file=sys.stderr, - ) - sys.exit(-1) - - if os.listdir(output_dir): - print( - f"output directory '{output_dir}' is not empty", - file=sys.stderr, - ) - sys.exit(-1) - - fix_files(input_dir, output_dir) diff --git a/packages/google-cloud-gkerecommender/setup.py b/packages/google-cloud-gkerecommender/setup.py deleted file mode 100644 index b442b627cd44..000000000000 --- a/packages/google-cloud-gkerecommender/setup.py +++ /dev/null @@ -1,99 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import io -import os -import re - -import setuptools # type: ignore - -package_root = os.path.abspath(os.path.dirname(__file__)) - -name = "google-cloud-gkerecommender" - - -description = "Google Cloud Gkerecommender API client library" - -version = None - -with open( - os.path.join(package_root, "google/cloud/gkerecommender/gapic_version.py") -) as fp: - version_candidates = re.findall(r"(?<=\")\d+.\d+.\d+(?=\")", fp.read()) - assert len(version_candidates) == 1 - version = version_candidates[0] - -if version[0] == "0": - release_status = "Development Status :: 4 - Beta" -else: - release_status = "Development Status :: 5 - Production/Stable" - -dependencies = [ - "google-api-core[grpc] >= 1.34.1, <3.0.0,!=2.0.*,!=2.1.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,!=2.10.*", - # Exclude incompatible versions of `google-auth` - # See https://github.com/googleapis/google-cloud-python/issues/12364 - "google-auth >= 2.14.1, <3.0.0,!=2.24.0,!=2.25.0", - "proto-plus >= 1.22.3, <2.0.0", - "proto-plus >= 1.25.0, <2.0.0; python_version >= '3.13'", - "protobuf>=3.20.2,<7.0.0,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5", -] -extras = {} -url = "https://github.com/googleapis/google-cloud-python/tree/main/packages/google-cloud-gkerecommender" - -package_root = os.path.abspath(os.path.dirname(__file__)) - -readme_filename = os.path.join(package_root, "README.rst") -with io.open(readme_filename, encoding="utf-8") as readme_file: - readme = readme_file.read() - -packages = [ - package - for package in setuptools.find_namespace_packages() - if package.startswith("google") -] - -setuptools.setup( - name=name, - version=version, - description=description, - long_description=readme, - author="Google LLC", - author_email="googleapis-packages@google.com", - license="Apache 2.0", - url=url, - classifiers=[ - release_status, - "Intended Audience :: Developers", - "License :: OSI Approved :: Apache Software License", - "Programming Language :: Python", - "Programming Language :: Python :: 3", - "Programming Language :: Python :: 3.7", - "Programming Language :: Python :: 3.8", - "Programming Language :: Python :: 3.9", - "Programming Language :: Python :: 3.10", - "Programming Language :: Python :: 3.11", - "Programming Language :: Python :: 3.12", - "Programming Language :: Python :: 3.13", - "Operating System :: OS Independent", - "Topic :: Internet", - ], - platforms="Posix; MacOS X; Windows", - packages=packages, - python_requires=">=3.7", - install_requires=dependencies, - extras_require=extras, - include_package_data=True, - zip_safe=False, -) diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.10.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.10.txt deleted file mode 100644 index ed7f9aed2559..000000000000 --- a/packages/google-cloud-gkerecommender/testing/constraints-3.10.txt +++ /dev/null @@ -1,6 +0,0 @@ -# -*- coding: utf-8 -*- -# This constraints file is required for unit tests. -# List all library dependencies and extras in this file. -google-api-core -proto-plus -protobuf diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.11.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.11.txt deleted file mode 100644 index ed7f9aed2559..000000000000 --- a/packages/google-cloud-gkerecommender/testing/constraints-3.11.txt +++ /dev/null @@ -1,6 +0,0 @@ -# -*- coding: utf-8 -*- -# This constraints file is required for unit tests. -# List all library dependencies and extras in this file. -google-api-core -proto-plus -protobuf diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.12.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.12.txt deleted file mode 100644 index ed7f9aed2559..000000000000 --- a/packages/google-cloud-gkerecommender/testing/constraints-3.12.txt +++ /dev/null @@ -1,6 +0,0 @@ -# -*- coding: utf-8 -*- -# This constraints file is required for unit tests. -# List all library dependencies and extras in this file. -google-api-core -proto-plus -protobuf diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.13.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.13.txt deleted file mode 100644 index c20a77817caa..000000000000 --- a/packages/google-cloud-gkerecommender/testing/constraints-3.13.txt +++ /dev/null @@ -1,11 +0,0 @@ -# We use the constraints file for the latest Python version -# (currently this file) to check that the latest -# major versions of dependencies are supported in setup.py. -# List all library dependencies and extras in this file. -# Require the latest major version be installed for each dependency. -# e.g., if setup.py has "google-cloud-foo >= 1.14.0, < 2.0.0", -# Then this file should have google-cloud-foo>=1 -google-api-core>=2 -google-auth>=2 -proto-plus>=1 -protobuf>=6 diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.7.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.7.txt deleted file mode 100644 index a77f12bc13e4..000000000000 --- a/packages/google-cloud-gkerecommender/testing/constraints-3.7.txt +++ /dev/null @@ -1,10 +0,0 @@ -# This constraints file is used to check that lower bounds -# are correct in setup.py -# List all library dependencies and extras in this file. -# Pin the version to the lower bound. -# e.g., if setup.py has "google-cloud-foo >= 1.14.0, < 2.0.0", -# Then this file should have google-cloud-foo==1.14.0 -google-api-core==1.34.1 -google-auth==2.14.1 -proto-plus==1.22.3 -protobuf==3.20.2 diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.8.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.8.txt deleted file mode 100644 index ed7f9aed2559..000000000000 --- a/packages/google-cloud-gkerecommender/testing/constraints-3.8.txt +++ /dev/null @@ -1,6 +0,0 @@ -# -*- coding: utf-8 -*- -# This constraints file is required for unit tests. -# List all library dependencies and extras in this file. -google-api-core -proto-plus -protobuf diff --git a/packages/google-cloud-gkerecommender/testing/constraints-3.9.txt b/packages/google-cloud-gkerecommender/testing/constraints-3.9.txt deleted file mode 100644 index ed7f9aed2559..000000000000 --- a/packages/google-cloud-gkerecommender/testing/constraints-3.9.txt +++ /dev/null @@ -1,6 +0,0 @@ -# -*- coding: utf-8 -*- -# This constraints file is required for unit tests. -# List all library dependencies and extras in this file. -google-api-core -proto-plus -protobuf diff --git a/packages/google-cloud-gkerecommender/tests/__init__.py b/packages/google-cloud-gkerecommender/tests/__init__.py deleted file mode 100644 index cbf94b283c70..000000000000 --- a/packages/google-cloud-gkerecommender/tests/__init__.py +++ /dev/null @@ -1,15 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# diff --git a/packages/google-cloud-gkerecommender/tests/unit/__init__.py b/packages/google-cloud-gkerecommender/tests/unit/__init__.py deleted file mode 100644 index cbf94b283c70..000000000000 --- a/packages/google-cloud-gkerecommender/tests/unit/__init__.py +++ /dev/null @@ -1,15 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# diff --git a/packages/google-cloud-gkerecommender/tests/unit/gapic/__init__.py b/packages/google-cloud-gkerecommender/tests/unit/gapic/__init__.py deleted file mode 100644 index cbf94b283c70..000000000000 --- a/packages/google-cloud-gkerecommender/tests/unit/gapic/__init__.py +++ /dev/null @@ -1,15 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# diff --git a/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/__init__.py b/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/__init__.py deleted file mode 100644 index cbf94b283c70..000000000000 --- a/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/__init__.py +++ /dev/null @@ -1,15 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# diff --git a/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/test_gke_inference_quickstart.py b/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/test_gke_inference_quickstart.py deleted file mode 100644 index 417137074417..000000000000 --- a/packages/google-cloud-gkerecommender/tests/unit/gapic/gkerecommender_v1/test_gke_inference_quickstart.py +++ /dev/null @@ -1,6051 +0,0 @@ -# -*- coding: utf-8 -*- -# Copyright 2025 Google LLC -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# -import os - -# try/except added for compatibility with python < 3.8 -try: - from unittest import mock - from unittest.mock import AsyncMock # pragma: NO COVER -except ImportError: # pragma: NO COVER - import mock - -from collections.abc import AsyncIterable, Iterable -import json -import math - -from google.api_core import api_core_version -from google.protobuf import json_format -import grpc -from grpc.experimental import aio -from proto.marshal.rules import wrappers -from proto.marshal.rules.dates import DurationRule, TimestampRule -import pytest -from requests import PreparedRequest, Request, Response -from requests.sessions import Session - -try: - from google.auth.aio import credentials as ga_credentials_async - - HAS_GOOGLE_AUTH_AIO = True -except ImportError: # pragma: NO COVER - HAS_GOOGLE_AUTH_AIO = False - -from google.api_core import gapic_v1, grpc_helpers, grpc_helpers_async, path_template -from google.api_core import client_options -from google.api_core import exceptions as core_exceptions -from google.api_core import retry as retries -import google.auth -from google.auth import credentials as ga_credentials -from google.auth.exceptions import MutualTLSChannelError -from google.oauth2 import service_account - -from google.cloud.gkerecommender_v1.services.gke_inference_quickstart import ( - GkeInferenceQuickstartAsyncClient, - GkeInferenceQuickstartClient, - pagers, - transports, -) -from google.cloud.gkerecommender_v1.types import gkerecommender - -CRED_INFO_JSON = { - "credential_source": "/path/to/file", - "credential_type": "service account credentials", - "principal": "service-account@example.com", -} -CRED_INFO_STRING = json.dumps(CRED_INFO_JSON) - - -async def mock_async_gen(data, chunk_size=1): - for i in range(0, len(data)): # pragma: NO COVER - chunk = data[i : i + chunk_size] - yield chunk.encode("utf-8") - - -def client_cert_source_callback(): - return b"cert bytes", b"key bytes" - - -# TODO: use async auth anon credentials by default once the minimum version of google-auth is upgraded. -# See related issue: https://github.com/googleapis/gapic-generator-python/issues/2107. -def async_anonymous_credentials(): - if HAS_GOOGLE_AUTH_AIO: - return ga_credentials_async.AnonymousCredentials() - return ga_credentials.AnonymousCredentials() - - -# If default endpoint is localhost, then default mtls endpoint will be the same. -# This method modifies the default endpoint so the client can produce a different -# mtls endpoint for endpoint testing purposes. -def modify_default_endpoint(client): - return ( - "foo.googleapis.com" - if ("localhost" in client.DEFAULT_ENDPOINT) - else client.DEFAULT_ENDPOINT - ) - - -# If default endpoint template is localhost, then default mtls endpoint will be the same. -# This method modifies the default endpoint template so the client can produce a different -# mtls endpoint for endpoint testing purposes. -def modify_default_endpoint_template(client): - return ( - "test.{UNIVERSE_DOMAIN}" - if ("localhost" in client._DEFAULT_ENDPOINT_TEMPLATE) - else client._DEFAULT_ENDPOINT_TEMPLATE - ) - - -def test__get_default_mtls_endpoint(): - api_endpoint = "example.googleapis.com" - api_mtls_endpoint = "example.mtls.googleapis.com" - sandbox_endpoint = "example.sandbox.googleapis.com" - sandbox_mtls_endpoint = "example.mtls.sandbox.googleapis.com" - non_googleapi = "api.example.com" - - assert GkeInferenceQuickstartClient._get_default_mtls_endpoint(None) is None - assert ( - GkeInferenceQuickstartClient._get_default_mtls_endpoint(api_endpoint) - == api_mtls_endpoint - ) - assert ( - GkeInferenceQuickstartClient._get_default_mtls_endpoint(api_mtls_endpoint) - == api_mtls_endpoint - ) - assert ( - GkeInferenceQuickstartClient._get_default_mtls_endpoint(sandbox_endpoint) - == sandbox_mtls_endpoint - ) - assert ( - GkeInferenceQuickstartClient._get_default_mtls_endpoint(sandbox_mtls_endpoint) - == sandbox_mtls_endpoint - ) - assert ( - GkeInferenceQuickstartClient._get_default_mtls_endpoint(non_googleapi) - == non_googleapi - ) - - -def test__read_environment_variables(): - assert GkeInferenceQuickstartClient._read_environment_variables() == ( - False, - "auto", - None, - ) - - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): - assert GkeInferenceQuickstartClient._read_environment_variables() == ( - True, - "auto", - None, - ) - - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "false"}): - assert GkeInferenceQuickstartClient._read_environment_variables() == ( - False, - "auto", - None, - ) - - with mock.patch.dict( - os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "Unsupported"} - ): - with pytest.raises(ValueError) as excinfo: - GkeInferenceQuickstartClient._read_environment_variables() - assert ( - str(excinfo.value) - == "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" - ) - - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): - assert GkeInferenceQuickstartClient._read_environment_variables() == ( - False, - "never", - None, - ) - - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "always"}): - assert GkeInferenceQuickstartClient._read_environment_variables() == ( - False, - "always", - None, - ) - - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "auto"}): - assert GkeInferenceQuickstartClient._read_environment_variables() == ( - False, - "auto", - None, - ) - - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "Unsupported"}): - with pytest.raises(MutualTLSChannelError) as excinfo: - GkeInferenceQuickstartClient._read_environment_variables() - assert ( - str(excinfo.value) - == "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" - ) - - with mock.patch.dict(os.environ, {"GOOGLE_CLOUD_UNIVERSE_DOMAIN": "foo.com"}): - assert GkeInferenceQuickstartClient._read_environment_variables() == ( - False, - "auto", - "foo.com", - ) - - -def test__get_client_cert_source(): - mock_provided_cert_source = mock.Mock() - mock_default_cert_source = mock.Mock() - - assert GkeInferenceQuickstartClient._get_client_cert_source(None, False) is None - assert ( - GkeInferenceQuickstartClient._get_client_cert_source( - mock_provided_cert_source, False - ) - is None - ) - assert ( - GkeInferenceQuickstartClient._get_client_cert_source( - mock_provided_cert_source, True - ) - == mock_provided_cert_source - ) - - with mock.patch( - "google.auth.transport.mtls.has_default_client_cert_source", return_value=True - ): - with mock.patch( - "google.auth.transport.mtls.default_client_cert_source", - return_value=mock_default_cert_source, - ): - assert ( - GkeInferenceQuickstartClient._get_client_cert_source(None, True) - is mock_default_cert_source - ) - assert ( - GkeInferenceQuickstartClient._get_client_cert_source( - mock_provided_cert_source, "true" - ) - is mock_provided_cert_source - ) - - -@mock.patch.object( - GkeInferenceQuickstartClient, - "_DEFAULT_ENDPOINT_TEMPLATE", - modify_default_endpoint_template(GkeInferenceQuickstartClient), -) -@mock.patch.object( - GkeInferenceQuickstartAsyncClient, - "_DEFAULT_ENDPOINT_TEMPLATE", - modify_default_endpoint_template(GkeInferenceQuickstartAsyncClient), -) -def test__get_api_endpoint(): - api_override = "foo.com" - mock_client_cert_source = mock.Mock() - default_universe = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE - default_endpoint = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=default_universe - ) - mock_universe = "bar.com" - mock_endpoint = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=mock_universe - ) - - assert ( - GkeInferenceQuickstartClient._get_api_endpoint( - api_override, mock_client_cert_source, default_universe, "always" - ) - == api_override - ) - assert ( - GkeInferenceQuickstartClient._get_api_endpoint( - None, mock_client_cert_source, default_universe, "auto" - ) - == GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT - ) - assert ( - GkeInferenceQuickstartClient._get_api_endpoint( - None, None, default_universe, "auto" - ) - == default_endpoint - ) - assert ( - GkeInferenceQuickstartClient._get_api_endpoint( - None, None, default_universe, "always" - ) - == GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT - ) - assert ( - GkeInferenceQuickstartClient._get_api_endpoint( - None, mock_client_cert_source, default_universe, "always" - ) - == GkeInferenceQuickstartClient.DEFAULT_MTLS_ENDPOINT - ) - assert ( - GkeInferenceQuickstartClient._get_api_endpoint( - None, None, mock_universe, "never" - ) - == mock_endpoint - ) - assert ( - GkeInferenceQuickstartClient._get_api_endpoint( - None, None, default_universe, "never" - ) - == default_endpoint - ) - - with pytest.raises(MutualTLSChannelError) as excinfo: - GkeInferenceQuickstartClient._get_api_endpoint( - None, mock_client_cert_source, mock_universe, "auto" - ) - assert ( - str(excinfo.value) - == "mTLS is not supported in any universe other than googleapis.com." - ) - - -def test__get_universe_domain(): - client_universe_domain = "foo.com" - universe_domain_env = "bar.com" - - assert ( - GkeInferenceQuickstartClient._get_universe_domain( - client_universe_domain, universe_domain_env - ) - == client_universe_domain - ) - assert ( - GkeInferenceQuickstartClient._get_universe_domain(None, universe_domain_env) - == universe_domain_env - ) - assert ( - GkeInferenceQuickstartClient._get_universe_domain(None, None) - == GkeInferenceQuickstartClient._DEFAULT_UNIVERSE - ) - - with pytest.raises(ValueError) as excinfo: - GkeInferenceQuickstartClient._get_universe_domain("", None) - assert str(excinfo.value) == "Universe Domain cannot be an empty string." - - -@pytest.mark.parametrize( - "error_code,cred_info_json,show_cred_info", - [ - (401, CRED_INFO_JSON, True), - (403, CRED_INFO_JSON, True), - (404, CRED_INFO_JSON, True), - (500, CRED_INFO_JSON, False), - (401, None, False), - (403, None, False), - (404, None, False), - (500, None, False), - ], -) -def test__add_cred_info_for_auth_errors(error_code, cred_info_json, show_cred_info): - cred = mock.Mock(["get_cred_info"]) - cred.get_cred_info = mock.Mock(return_value=cred_info_json) - client = GkeInferenceQuickstartClient(credentials=cred) - client._transport._credentials = cred - - error = core_exceptions.GoogleAPICallError("message", details=["foo"]) - error.code = error_code - - client._add_cred_info_for_auth_errors(error) - if show_cred_info: - assert error.details == ["foo", CRED_INFO_STRING] - else: - assert error.details == ["foo"] - - -@pytest.mark.parametrize("error_code", [401, 403, 404, 500]) -def test__add_cred_info_for_auth_errors_no_get_cred_info(error_code): - cred = mock.Mock([]) - assert not hasattr(cred, "get_cred_info") - client = GkeInferenceQuickstartClient(credentials=cred) - client._transport._credentials = cred - - error = core_exceptions.GoogleAPICallError("message", details=[]) - error.code = error_code - - client._add_cred_info_for_auth_errors(error) - assert error.details == [] - - -@pytest.mark.parametrize( - "client_class,transport_name", - [ - (GkeInferenceQuickstartClient, "grpc"), - (GkeInferenceQuickstartAsyncClient, "grpc_asyncio"), - (GkeInferenceQuickstartClient, "rest"), - ], -) -def test_gke_inference_quickstart_client_from_service_account_info( - client_class, transport_name -): - creds = ga_credentials.AnonymousCredentials() - with mock.patch.object( - service_account.Credentials, "from_service_account_info" - ) as factory: - factory.return_value = creds - info = {"valid": True} - client = client_class.from_service_account_info(info, transport=transport_name) - assert client.transport._credentials == creds - assert isinstance(client, client_class) - - assert client.transport._host == ( - "gkerecommender.googleapis.com:443" - if transport_name in ["grpc", "grpc_asyncio"] - else "https://gkerecommender.googleapis.com" - ) - - -@pytest.mark.parametrize( - "transport_class,transport_name", - [ - (transports.GkeInferenceQuickstartGrpcTransport, "grpc"), - (transports.GkeInferenceQuickstartGrpcAsyncIOTransport, "grpc_asyncio"), - (transports.GkeInferenceQuickstartRestTransport, "rest"), - ], -) -def test_gke_inference_quickstart_client_service_account_always_use_jwt( - transport_class, transport_name -): - with mock.patch.object( - service_account.Credentials, "with_always_use_jwt_access", create=True - ) as use_jwt: - creds = service_account.Credentials(None, None, None) - transport = transport_class(credentials=creds, always_use_jwt_access=True) - use_jwt.assert_called_once_with(True) - - with mock.patch.object( - service_account.Credentials, "with_always_use_jwt_access", create=True - ) as use_jwt: - creds = service_account.Credentials(None, None, None) - transport = transport_class(credentials=creds, always_use_jwt_access=False) - use_jwt.assert_not_called() - - -@pytest.mark.parametrize( - "client_class,transport_name", - [ - (GkeInferenceQuickstartClient, "grpc"), - (GkeInferenceQuickstartAsyncClient, "grpc_asyncio"), - (GkeInferenceQuickstartClient, "rest"), - ], -) -def test_gke_inference_quickstart_client_from_service_account_file( - client_class, transport_name -): - creds = ga_credentials.AnonymousCredentials() - with mock.patch.object( - service_account.Credentials, "from_service_account_file" - ) as factory: - factory.return_value = creds - client = client_class.from_service_account_file( - "dummy/file/path.json", transport=transport_name - ) - assert client.transport._credentials == creds - assert isinstance(client, client_class) - - client = client_class.from_service_account_json( - "dummy/file/path.json", transport=transport_name - ) - assert client.transport._credentials == creds - assert isinstance(client, client_class) - - assert client.transport._host == ( - "gkerecommender.googleapis.com:443" - if transport_name in ["grpc", "grpc_asyncio"] - else "https://gkerecommender.googleapis.com" - ) - - -def test_gke_inference_quickstart_client_get_transport_class(): - transport = GkeInferenceQuickstartClient.get_transport_class() - available_transports = [ - transports.GkeInferenceQuickstartGrpcTransport, - transports.GkeInferenceQuickstartRestTransport, - ] - assert transport in available_transports - - transport = GkeInferenceQuickstartClient.get_transport_class("grpc") - assert transport == transports.GkeInferenceQuickstartGrpcTransport - - -@pytest.mark.parametrize( - "client_class,transport_class,transport_name", - [ - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartGrpcTransport, - "grpc", - ), - ( - GkeInferenceQuickstartAsyncClient, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - "grpc_asyncio", - ), - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartRestTransport, - "rest", - ), - ], -) -@mock.patch.object( - GkeInferenceQuickstartClient, - "_DEFAULT_ENDPOINT_TEMPLATE", - modify_default_endpoint_template(GkeInferenceQuickstartClient), -) -@mock.patch.object( - GkeInferenceQuickstartAsyncClient, - "_DEFAULT_ENDPOINT_TEMPLATE", - modify_default_endpoint_template(GkeInferenceQuickstartAsyncClient), -) -def test_gke_inference_quickstart_client_client_options( - client_class, transport_class, transport_name -): - # Check that if channel is provided we won't create a new one. - with mock.patch.object(GkeInferenceQuickstartClient, "get_transport_class") as gtc: - transport = transport_class(credentials=ga_credentials.AnonymousCredentials()) - client = client_class(transport=transport) - gtc.assert_not_called() - - # Check that if channel is provided via str we will create a new one. - with mock.patch.object(GkeInferenceQuickstartClient, "get_transport_class") as gtc: - client = client_class(transport=transport_name) - gtc.assert_called() - - # Check the case api_endpoint is provided. - options = client_options.ClientOptions(api_endpoint="squid.clam.whelk") - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(transport=transport_name, client_options=options) - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host="squid.clam.whelk", - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - # Check the case api_endpoint is not provided and GOOGLE_API_USE_MTLS_ENDPOINT is - # "never". - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host=client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ), - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - # Check the case api_endpoint is not provided and GOOGLE_API_USE_MTLS_ENDPOINT is - # "always". - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "always"}): - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host=client.DEFAULT_MTLS_ENDPOINT, - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - # Check the case api_endpoint is not provided and GOOGLE_API_USE_MTLS_ENDPOINT has - # unsupported value. - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "Unsupported"}): - with pytest.raises(MutualTLSChannelError) as excinfo: - client = client_class(transport=transport_name) - assert ( - str(excinfo.value) - == "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" - ) - - # Check the case GOOGLE_API_USE_CLIENT_CERTIFICATE has unsupported value. - with mock.patch.dict( - os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "Unsupported"} - ): - with pytest.raises(ValueError) as excinfo: - client = client_class(transport=transport_name) - assert ( - str(excinfo.value) - == "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" - ) - - # Check the case quota_project_id is provided - options = client_options.ClientOptions(quota_project_id="octopus") - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(client_options=options, transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host=client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ), - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id="octopus", - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - # Check the case api_endpoint is provided - options = client_options.ClientOptions( - api_audience="https://language.googleapis.com" - ) - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(client_options=options, transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host=client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ), - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience="https://language.googleapis.com", - ) - - -@pytest.mark.parametrize( - "client_class,transport_class,transport_name,use_client_cert_env", - [ - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartGrpcTransport, - "grpc", - "true", - ), - ( - GkeInferenceQuickstartAsyncClient, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - "grpc_asyncio", - "true", - ), - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartGrpcTransport, - "grpc", - "false", - ), - ( - GkeInferenceQuickstartAsyncClient, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - "grpc_asyncio", - "false", - ), - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartRestTransport, - "rest", - "true", - ), - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartRestTransport, - "rest", - "false", - ), - ], -) -@mock.patch.object( - GkeInferenceQuickstartClient, - "_DEFAULT_ENDPOINT_TEMPLATE", - modify_default_endpoint_template(GkeInferenceQuickstartClient), -) -@mock.patch.object( - GkeInferenceQuickstartAsyncClient, - "_DEFAULT_ENDPOINT_TEMPLATE", - modify_default_endpoint_template(GkeInferenceQuickstartAsyncClient), -) -@mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "auto"}) -def test_gke_inference_quickstart_client_mtls_env_auto( - client_class, transport_class, transport_name, use_client_cert_env -): - # This tests the endpoint autoswitch behavior. Endpoint is autoswitched to the default - # mtls endpoint, if GOOGLE_API_USE_CLIENT_CERTIFICATE is "true" and client cert exists. - - # Check the case client_cert_source is provided. Whether client cert is used depends on - # GOOGLE_API_USE_CLIENT_CERTIFICATE value. - with mock.patch.dict( - os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": use_client_cert_env} - ): - options = client_options.ClientOptions( - client_cert_source=client_cert_source_callback - ) - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(client_options=options, transport=transport_name) - - if use_client_cert_env == "false": - expected_client_cert_source = None - expected_host = client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ) - else: - expected_client_cert_source = client_cert_source_callback - expected_host = client.DEFAULT_MTLS_ENDPOINT - - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host=expected_host, - scopes=None, - client_cert_source_for_mtls=expected_client_cert_source, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - # Check the case ADC client cert is provided. Whether client cert is used depends on - # GOOGLE_API_USE_CLIENT_CERTIFICATE value. - with mock.patch.dict( - os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": use_client_cert_env} - ): - with mock.patch.object(transport_class, "__init__") as patched: - with mock.patch( - "google.auth.transport.mtls.has_default_client_cert_source", - return_value=True, - ): - with mock.patch( - "google.auth.transport.mtls.default_client_cert_source", - return_value=client_cert_source_callback, - ): - if use_client_cert_env == "false": - expected_host = client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ) - expected_client_cert_source = None - else: - expected_host = client.DEFAULT_MTLS_ENDPOINT - expected_client_cert_source = client_cert_source_callback - - patched.return_value = None - client = client_class(transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host=expected_host, - scopes=None, - client_cert_source_for_mtls=expected_client_cert_source, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - # Check the case client_cert_source and ADC client cert are not provided. - with mock.patch.dict( - os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": use_client_cert_env} - ): - with mock.patch.object(transport_class, "__init__") as patched: - with mock.patch( - "google.auth.transport.mtls.has_default_client_cert_source", - return_value=False, - ): - patched.return_value = None - client = client_class(transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host=client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ), - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - -@pytest.mark.parametrize( - "client_class", [GkeInferenceQuickstartClient, GkeInferenceQuickstartAsyncClient] -) -@mock.patch.object( - GkeInferenceQuickstartClient, - "DEFAULT_ENDPOINT", - modify_default_endpoint(GkeInferenceQuickstartClient), -) -@mock.patch.object( - GkeInferenceQuickstartAsyncClient, - "DEFAULT_ENDPOINT", - modify_default_endpoint(GkeInferenceQuickstartAsyncClient), -) -def test_gke_inference_quickstart_client_get_mtls_endpoint_and_cert_source( - client_class, -): - mock_client_cert_source = mock.Mock() - - # Test the case GOOGLE_API_USE_CLIENT_CERTIFICATE is "true". - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): - mock_api_endpoint = "foo" - options = client_options.ClientOptions( - client_cert_source=mock_client_cert_source, api_endpoint=mock_api_endpoint - ) - api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source( - options - ) - assert api_endpoint == mock_api_endpoint - assert cert_source == mock_client_cert_source - - # Test the case GOOGLE_API_USE_CLIENT_CERTIFICATE is "false". - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "false"}): - mock_client_cert_source = mock.Mock() - mock_api_endpoint = "foo" - options = client_options.ClientOptions( - client_cert_source=mock_client_cert_source, api_endpoint=mock_api_endpoint - ) - api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source( - options - ) - assert api_endpoint == mock_api_endpoint - assert cert_source is None - - # Test the case GOOGLE_API_USE_MTLS_ENDPOINT is "never". - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): - api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source() - assert api_endpoint == client_class.DEFAULT_ENDPOINT - assert cert_source is None - - # Test the case GOOGLE_API_USE_MTLS_ENDPOINT is "always". - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "always"}): - api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source() - assert api_endpoint == client_class.DEFAULT_MTLS_ENDPOINT - assert cert_source is None - - # Test the case GOOGLE_API_USE_MTLS_ENDPOINT is "auto" and default cert doesn't exist. - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): - with mock.patch( - "google.auth.transport.mtls.has_default_client_cert_source", - return_value=False, - ): - api_endpoint, cert_source = client_class.get_mtls_endpoint_and_cert_source() - assert api_endpoint == client_class.DEFAULT_ENDPOINT - assert cert_source is None - - # Test the case GOOGLE_API_USE_MTLS_ENDPOINT is "auto" and default cert exists. - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): - with mock.patch( - "google.auth.transport.mtls.has_default_client_cert_source", - return_value=True, - ): - with mock.patch( - "google.auth.transport.mtls.default_client_cert_source", - return_value=mock_client_cert_source, - ): - ( - api_endpoint, - cert_source, - ) = client_class.get_mtls_endpoint_and_cert_source() - assert api_endpoint == client_class.DEFAULT_MTLS_ENDPOINT - assert cert_source == mock_client_cert_source - - # Check the case api_endpoint is not provided and GOOGLE_API_USE_MTLS_ENDPOINT has - # unsupported value. - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "Unsupported"}): - with pytest.raises(MutualTLSChannelError) as excinfo: - client_class.get_mtls_endpoint_and_cert_source() - - assert ( - str(excinfo.value) - == "Environment variable `GOOGLE_API_USE_MTLS_ENDPOINT` must be `never`, `auto` or `always`" - ) - - # Check the case GOOGLE_API_USE_CLIENT_CERTIFICATE has unsupported value. - with mock.patch.dict( - os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "Unsupported"} - ): - with pytest.raises(ValueError) as excinfo: - client_class.get_mtls_endpoint_and_cert_source() - - assert ( - str(excinfo.value) - == "Environment variable `GOOGLE_API_USE_CLIENT_CERTIFICATE` must be either `true` or `false`" - ) - - -@pytest.mark.parametrize( - "client_class", [GkeInferenceQuickstartClient, GkeInferenceQuickstartAsyncClient] -) -@mock.patch.object( - GkeInferenceQuickstartClient, - "_DEFAULT_ENDPOINT_TEMPLATE", - modify_default_endpoint_template(GkeInferenceQuickstartClient), -) -@mock.patch.object( - GkeInferenceQuickstartAsyncClient, - "_DEFAULT_ENDPOINT_TEMPLATE", - modify_default_endpoint_template(GkeInferenceQuickstartAsyncClient), -) -def test_gke_inference_quickstart_client_client_api_endpoint(client_class): - mock_client_cert_source = client_cert_source_callback - api_override = "foo.com" - default_universe = GkeInferenceQuickstartClient._DEFAULT_UNIVERSE - default_endpoint = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=default_universe - ) - mock_universe = "bar.com" - mock_endpoint = GkeInferenceQuickstartClient._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=mock_universe - ) - - # If ClientOptions.api_endpoint is set and GOOGLE_API_USE_CLIENT_CERTIFICATE="true", - # use ClientOptions.api_endpoint as the api endpoint regardless. - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_CLIENT_CERTIFICATE": "true"}): - with mock.patch( - "google.auth.transport.requests.AuthorizedSession.configure_mtls_channel" - ): - options = client_options.ClientOptions( - client_cert_source=mock_client_cert_source, api_endpoint=api_override - ) - client = client_class( - client_options=options, - credentials=ga_credentials.AnonymousCredentials(), - ) - assert client.api_endpoint == api_override - - # If ClientOptions.api_endpoint is not set and GOOGLE_API_USE_MTLS_ENDPOINT="never", - # use the _DEFAULT_ENDPOINT_TEMPLATE populated with GDU as the api endpoint. - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): - client = client_class(credentials=ga_credentials.AnonymousCredentials()) - assert client.api_endpoint == default_endpoint - - # If ClientOptions.api_endpoint is not set and GOOGLE_API_USE_MTLS_ENDPOINT="always", - # use the DEFAULT_MTLS_ENDPOINT as the api endpoint. - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "always"}): - client = client_class(credentials=ga_credentials.AnonymousCredentials()) - assert client.api_endpoint == client_class.DEFAULT_MTLS_ENDPOINT - - # If ClientOptions.api_endpoint is not set, GOOGLE_API_USE_MTLS_ENDPOINT="auto" (default), - # GOOGLE_API_USE_CLIENT_CERTIFICATE="false" (default), default cert source doesn't exist, - # and ClientOptions.universe_domain="bar.com", - # use the _DEFAULT_ENDPOINT_TEMPLATE populated with universe domain as the api endpoint. - options = client_options.ClientOptions() - universe_exists = hasattr(options, "universe_domain") - if universe_exists: - options = client_options.ClientOptions(universe_domain=mock_universe) - client = client_class( - client_options=options, credentials=ga_credentials.AnonymousCredentials() - ) - else: - client = client_class( - client_options=options, credentials=ga_credentials.AnonymousCredentials() - ) - assert client.api_endpoint == ( - mock_endpoint if universe_exists else default_endpoint - ) - assert client.universe_domain == ( - mock_universe if universe_exists else default_universe - ) - - # If ClientOptions does not have a universe domain attribute and GOOGLE_API_USE_MTLS_ENDPOINT="never", - # use the _DEFAULT_ENDPOINT_TEMPLATE populated with GDU as the api endpoint. - options = client_options.ClientOptions() - if hasattr(options, "universe_domain"): - delattr(options, "universe_domain") - with mock.patch.dict(os.environ, {"GOOGLE_API_USE_MTLS_ENDPOINT": "never"}): - client = client_class( - client_options=options, credentials=ga_credentials.AnonymousCredentials() - ) - assert client.api_endpoint == default_endpoint - - -@pytest.mark.parametrize( - "client_class,transport_class,transport_name", - [ - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartGrpcTransport, - "grpc", - ), - ( - GkeInferenceQuickstartAsyncClient, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - "grpc_asyncio", - ), - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartRestTransport, - "rest", - ), - ], -) -def test_gke_inference_quickstart_client_client_options_scopes( - client_class, transport_class, transport_name -): - # Check the case scopes are provided. - options = client_options.ClientOptions( - scopes=["1", "2"], - ) - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(client_options=options, transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file=None, - host=client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ), - scopes=["1", "2"], - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - -@pytest.mark.parametrize( - "client_class,transport_class,transport_name,grpc_helpers", - [ - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartGrpcTransport, - "grpc", - grpc_helpers, - ), - ( - GkeInferenceQuickstartAsyncClient, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - "grpc_asyncio", - grpc_helpers_async, - ), - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartRestTransport, - "rest", - None, - ), - ], -) -def test_gke_inference_quickstart_client_client_options_credentials_file( - client_class, transport_class, transport_name, grpc_helpers -): - # Check the case credentials file is provided. - options = client_options.ClientOptions(credentials_file="credentials.json") - - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(client_options=options, transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file="credentials.json", - host=client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ), - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - -def test_gke_inference_quickstart_client_client_options_from_dict(): - with mock.patch( - "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.GkeInferenceQuickstartGrpcTransport.__init__" - ) as grpc_transport: - grpc_transport.return_value = None - client = GkeInferenceQuickstartClient( - client_options={"api_endpoint": "squid.clam.whelk"} - ) - grpc_transport.assert_called_once_with( - credentials=None, - credentials_file=None, - host="squid.clam.whelk", - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - -@pytest.mark.parametrize( - "client_class,transport_class,transport_name,grpc_helpers", - [ - ( - GkeInferenceQuickstartClient, - transports.GkeInferenceQuickstartGrpcTransport, - "grpc", - grpc_helpers, - ), - ( - GkeInferenceQuickstartAsyncClient, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - "grpc_asyncio", - grpc_helpers_async, - ), - ], -) -def test_gke_inference_quickstart_client_create_channel_credentials_file( - client_class, transport_class, transport_name, grpc_helpers -): - # Check the case credentials file is provided. - options = client_options.ClientOptions(credentials_file="credentials.json") - - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(client_options=options, transport=transport_name) - patched.assert_called_once_with( - credentials=None, - credentials_file="credentials.json", - host=client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ), - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) - - # test that the credentials from file are saved and used as the credentials. - with mock.patch.object( - google.auth, "load_credentials_from_file", autospec=True - ) as load_creds, mock.patch.object( - google.auth, "default", autospec=True - ) as adc, mock.patch.object( - grpc_helpers, "create_channel" - ) as create_channel: - creds = ga_credentials.AnonymousCredentials() - file_creds = ga_credentials.AnonymousCredentials() - load_creds.return_value = (file_creds, None) - adc.return_value = (creds, None) - client = client_class(client_options=options, transport=transport_name) - create_channel.assert_called_with( - "gkerecommender.googleapis.com:443", - credentials=file_creds, - credentials_file=None, - quota_project_id=None, - default_scopes=("https://www.googleapis.com/auth/cloud-platform",), - scopes=None, - default_host="gkerecommender.googleapis.com", - ssl_credentials=None, - options=[ - ("grpc.max_send_message_length", -1), - ("grpc.max_receive_message_length", -1), - ], - ) - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchModelsRequest, - dict, - ], -) -def test_fetch_models(request_type, transport: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: - # Designate an appropriate return value for the call. - call.return_value = gkerecommender.FetchModelsResponse( - models=["models_value"], - next_page_token="next_page_token_value", - ) - response = client.fetch_models(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) == 1 - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchModelsRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelsPager) - assert response.models == ["models_value"] - assert response.next_page_token == "next_page_token_value" - - -def test_fetch_models_non_empty_request_with_auto_populated_field(): - # This test is a coverage failsafe to make sure that UUID4 fields are - # automatically populated, according to AIP-4235, with non-empty requests. - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Populate all string fields in the request which are not UUID4 - # since we want to check that UUID4 are populated automatically - # if they meet the requirements of AIP 4235. - request = gkerecommender.FetchModelsRequest( - page_token="page_token_value", - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: - call.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client.fetch_models(request=request) - call.assert_called() - _, args, _ = call.mock_calls[0] - assert args[0] == gkerecommender.FetchModelsRequest( - page_token="page_token_value", - ) - - -def test_fetch_models_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert client._transport.fetch_models in client._transport._wrapped_methods - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[client._transport.fetch_models] = mock_rpc - request = {} - client.fetch_models(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_models(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_models_async_use_cached_wrapped_rpc( - transport: str = "grpc_asyncio", -): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._client._transport.fetch_models - in client._client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.AsyncMock() - mock_rpc.return_value = mock.Mock() - client._client._transport._wrapped_methods[ - client._client._transport.fetch_models - ] = mock_rpc - - request = {} - await client.fetch_models(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - await client.fetch_models(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_models_async( - transport: str = "grpc_asyncio", request_type=gkerecommender.FetchModelsRequest -): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchModelsResponse( - models=["models_value"], - next_page_token="next_page_token_value", - ) - ) - response = await client.fetch_models(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchModelsRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelsAsyncPager) - assert response.models == ["models_value"] - assert response.next_page_token == "next_page_token_value" - - -@pytest.mark.asyncio -async def test_fetch_models_async_from_dict(): - await test_fetch_models_async(request_type=dict) - - -def test_fetch_models_pager(transport_name: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport_name, - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelsResponse( - models=[], - next_page_token="def", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - ], - ), - RuntimeError, - ) - - expected_metadata = () - retry = retries.Retry() - timeout = 5 - pager = client.fetch_models(request={}, retry=retry, timeout=timeout) - - assert pager._metadata == expected_metadata - assert pager._retry == retry - assert pager._timeout == timeout - - results = list(pager) - assert len(results) == 6 - assert all(isinstance(i, str) for i in results) - - -def test_fetch_models_pages(transport_name: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport_name, - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelsResponse( - models=[], - next_page_token="def", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - ], - ), - RuntimeError, - ) - pages = list(client.fetch_models(request={}).pages) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -@pytest.mark.asyncio -async def test_fetch_models_async_pager(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_models), "__call__", new_callable=mock.AsyncMock - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelsResponse( - models=[], - next_page_token="def", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - ], - ), - RuntimeError, - ) - async_pager = await client.fetch_models( - request={}, - ) - assert async_pager.next_page_token == "abc" - responses = [] - async for response in async_pager: # pragma: no branch - responses.append(response) - - assert len(responses) == 6 - assert all(isinstance(i, str) for i in responses) - - -@pytest.mark.asyncio -async def test_fetch_models_async_pages(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_models), "__call__", new_callable=mock.AsyncMock - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelsResponse( - models=[], - next_page_token="def", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - ], - ), - RuntimeError, - ) - pages = [] - # Workaround issue in python 3.9 related to code coverage by adding `# pragma: no branch` - # See https://github.com/googleapis/gapic-generator-python/pull/1174#issuecomment-1025132372 - async for page_ in ( # pragma: no branch - await client.fetch_models(request={}) - ).pages: - pages.append(page_) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchModelServersRequest, - dict, - ], -) -def test_fetch_model_servers(request_type, transport: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = gkerecommender.FetchModelServersResponse( - model_servers=["model_servers_value"], - next_page_token="next_page_token_value", - ) - response = client.fetch_model_servers(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) == 1 - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchModelServersRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelServersPager) - assert response.model_servers == ["model_servers_value"] - assert response.next_page_token == "next_page_token_value" - - -def test_fetch_model_servers_non_empty_request_with_auto_populated_field(): - # This test is a coverage failsafe to make sure that UUID4 fields are - # automatically populated, according to AIP-4235, with non-empty requests. - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Populate all string fields in the request which are not UUID4 - # since we want to check that UUID4 are populated automatically - # if they meet the requirements of AIP 4235. - request = gkerecommender.FetchModelServersRequest( - model="model_value", - page_token="page_token_value", - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), "__call__" - ) as call: - call.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client.fetch_model_servers(request=request) - call.assert_called() - _, args, _ = call.mock_calls[0] - assert args[0] == gkerecommender.FetchModelServersRequest( - model="model_value", - page_token="page_token_value", - ) - - -def test_fetch_model_servers_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._transport.fetch_model_servers in client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[ - client._transport.fetch_model_servers - ] = mock_rpc - request = {} - client.fetch_model_servers(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_model_servers(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_model_servers_async_use_cached_wrapped_rpc( - transport: str = "grpc_asyncio", -): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._client._transport.fetch_model_servers - in client._client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.AsyncMock() - mock_rpc.return_value = mock.Mock() - client._client._transport._wrapped_methods[ - client._client._transport.fetch_model_servers - ] = mock_rpc - - request = {} - await client.fetch_model_servers(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - await client.fetch_model_servers(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_model_servers_async( - transport: str = "grpc_asyncio", - request_type=gkerecommender.FetchModelServersRequest, -): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchModelServersResponse( - model_servers=["model_servers_value"], - next_page_token="next_page_token_value", - ) - ) - response = await client.fetch_model_servers(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchModelServersRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelServersAsyncPager) - assert response.model_servers == ["model_servers_value"] - assert response.next_page_token == "next_page_token_value" - - -@pytest.mark.asyncio -async def test_fetch_model_servers_async_from_dict(): - await test_fetch_model_servers_async(request_type=dict) - - -def test_fetch_model_servers_pager(transport_name: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport_name, - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), "__call__" - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[], - next_page_token="def", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - ], - ), - RuntimeError, - ) - - expected_metadata = () - retry = retries.Retry() - timeout = 5 - pager = client.fetch_model_servers(request={}, retry=retry, timeout=timeout) - - assert pager._metadata == expected_metadata - assert pager._retry == retry - assert pager._timeout == timeout - - results = list(pager) - assert len(results) == 6 - assert all(isinstance(i, str) for i in results) - - -def test_fetch_model_servers_pages(transport_name: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport_name, - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), "__call__" - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[], - next_page_token="def", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - ], - ), - RuntimeError, - ) - pages = list(client.fetch_model_servers(request={}).pages) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -@pytest.mark.asyncio -async def test_fetch_model_servers_async_pager(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), - "__call__", - new_callable=mock.AsyncMock, - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[], - next_page_token="def", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - ], - ), - RuntimeError, - ) - async_pager = await client.fetch_model_servers( - request={}, - ) - assert async_pager.next_page_token == "abc" - responses = [] - async for response in async_pager: # pragma: no branch - responses.append(response) - - assert len(responses) == 6 - assert all(isinstance(i, str) for i in responses) - - -@pytest.mark.asyncio -async def test_fetch_model_servers_async_pages(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), - "__call__", - new_callable=mock.AsyncMock, - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[], - next_page_token="def", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - ], - ), - RuntimeError, - ) - pages = [] - # Workaround issue in python 3.9 related to code coverage by adding `# pragma: no branch` - # See https://github.com/googleapis/gapic-generator-python/pull/1174#issuecomment-1025132372 - async for page_ in ( # pragma: no branch - await client.fetch_model_servers(request={}) - ).pages: - pages.append(page_) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchModelServerVersionsRequest, - dict, - ], -) -def test_fetch_model_server_versions(request_type, transport: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=["model_server_versions_value"], - next_page_token="next_page_token_value", - ) - response = client.fetch_model_server_versions(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) == 1 - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchModelServerVersionsRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelServerVersionsPager) - assert response.model_server_versions == ["model_server_versions_value"] - assert response.next_page_token == "next_page_token_value" - - -def test_fetch_model_server_versions_non_empty_request_with_auto_populated_field(): - # This test is a coverage failsafe to make sure that UUID4 fields are - # automatically populated, according to AIP-4235, with non-empty requests. - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Populate all string fields in the request which are not UUID4 - # since we want to check that UUID4 are populated automatically - # if they meet the requirements of AIP 4235. - request = gkerecommender.FetchModelServerVersionsRequest( - model="model_value", - model_server="model_server_value", - page_token="page_token_value", - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), "__call__" - ) as call: - call.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client.fetch_model_server_versions(request=request) - call.assert_called() - _, args, _ = call.mock_calls[0] - assert args[0] == gkerecommender.FetchModelServerVersionsRequest( - model="model_value", - model_server="model_server_value", - page_token="page_token_value", - ) - - -def test_fetch_model_server_versions_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._transport.fetch_model_server_versions - in client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[ - client._transport.fetch_model_server_versions - ] = mock_rpc - request = {} - client.fetch_model_server_versions(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_model_server_versions(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_model_server_versions_async_use_cached_wrapped_rpc( - transport: str = "grpc_asyncio", -): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._client._transport.fetch_model_server_versions - in client._client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.AsyncMock() - mock_rpc.return_value = mock.Mock() - client._client._transport._wrapped_methods[ - client._client._transport.fetch_model_server_versions - ] = mock_rpc - - request = {} - await client.fetch_model_server_versions(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - await client.fetch_model_server_versions(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_model_server_versions_async( - transport: str = "grpc_asyncio", - request_type=gkerecommender.FetchModelServerVersionsRequest, -): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=["model_server_versions_value"], - next_page_token="next_page_token_value", - ) - ) - response = await client.fetch_model_server_versions(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchModelServerVersionsRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelServerVersionsAsyncPager) - assert response.model_server_versions == ["model_server_versions_value"] - assert response.next_page_token == "next_page_token_value" - - -@pytest.mark.asyncio -async def test_fetch_model_server_versions_async_from_dict(): - await test_fetch_model_server_versions_async(request_type=dict) - - -def test_fetch_model_server_versions_pager(transport_name: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport_name, - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), "__call__" - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[], - next_page_token="def", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - ], - ), - RuntimeError, - ) - - expected_metadata = () - retry = retries.Retry() - timeout = 5 - pager = client.fetch_model_server_versions( - request={}, retry=retry, timeout=timeout - ) - - assert pager._metadata == expected_metadata - assert pager._retry == retry - assert pager._timeout == timeout - - results = list(pager) - assert len(results) == 6 - assert all(isinstance(i, str) for i in results) - - -def test_fetch_model_server_versions_pages(transport_name: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport_name, - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), "__call__" - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[], - next_page_token="def", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - ], - ), - RuntimeError, - ) - pages = list(client.fetch_model_server_versions(request={}).pages) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -@pytest.mark.asyncio -async def test_fetch_model_server_versions_async_pager(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), - "__call__", - new_callable=mock.AsyncMock, - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[], - next_page_token="def", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - ], - ), - RuntimeError, - ) - async_pager = await client.fetch_model_server_versions( - request={}, - ) - assert async_pager.next_page_token == "abc" - responses = [] - async for response in async_pager: # pragma: no branch - responses.append(response) - - assert len(responses) == 6 - assert all(isinstance(i, str) for i in responses) - - -@pytest.mark.asyncio -async def test_fetch_model_server_versions_async_pages(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), - "__call__", - new_callable=mock.AsyncMock, - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[], - next_page_token="def", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - ], - ), - RuntimeError, - ) - pages = [] - # Workaround issue in python 3.9 related to code coverage by adding `# pragma: no branch` - # See https://github.com/googleapis/gapic-generator-python/pull/1174#issuecomment-1025132372 - async for page_ in ( # pragma: no branch - await client.fetch_model_server_versions(request={}) - ).pages: - pages.append(page_) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchProfilesRequest, - dict, - ], -) -def test_fetch_profiles(request_type, transport: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: - # Designate an appropriate return value for the call. - call.return_value = gkerecommender.FetchProfilesResponse( - comments="comments_value", - next_page_token="next_page_token_value", - ) - response = client.fetch_profiles(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) == 1 - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchProfilesRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchProfilesPager) - assert response.comments == "comments_value" - assert response.next_page_token == "next_page_token_value" - - -def test_fetch_profiles_non_empty_request_with_auto_populated_field(): - # This test is a coverage failsafe to make sure that UUID4 fields are - # automatically populated, according to AIP-4235, with non-empty requests. - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Populate all string fields in the request which are not UUID4 - # since we want to check that UUID4 are populated automatically - # if they meet the requirements of AIP 4235. - request = gkerecommender.FetchProfilesRequest( - model="model_value", - model_server="model_server_value", - model_server_version="model_server_version_value", - page_token="page_token_value", - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: - call.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client.fetch_profiles(request=request) - call.assert_called() - _, args, _ = call.mock_calls[0] - assert args[0] == gkerecommender.FetchProfilesRequest( - model="model_value", - model_server="model_server_value", - model_server_version="model_server_version_value", - page_token="page_token_value", - ) - - -def test_fetch_profiles_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert client._transport.fetch_profiles in client._transport._wrapped_methods - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[client._transport.fetch_profiles] = mock_rpc - request = {} - client.fetch_profiles(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_profiles(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_profiles_async_use_cached_wrapped_rpc( - transport: str = "grpc_asyncio", -): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._client._transport.fetch_profiles - in client._client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.AsyncMock() - mock_rpc.return_value = mock.Mock() - client._client._transport._wrapped_methods[ - client._client._transport.fetch_profiles - ] = mock_rpc - - request = {} - await client.fetch_profiles(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - await client.fetch_profiles(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_profiles_async( - transport: str = "grpc_asyncio", request_type=gkerecommender.FetchProfilesRequest -): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchProfilesResponse( - comments="comments_value", - next_page_token="next_page_token_value", - ) - ) - response = await client.fetch_profiles(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchProfilesRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchProfilesAsyncPager) - assert response.comments == "comments_value" - assert response.next_page_token == "next_page_token_value" - - -@pytest.mark.asyncio -async def test_fetch_profiles_async_from_dict(): - await test_fetch_profiles_async(request_type=dict) - - -def test_fetch_profiles_pager(transport_name: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport_name, - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - next_page_token="abc", - ), - gkerecommender.FetchProfilesResponse( - profile=[], - next_page_token="def", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - ), - RuntimeError, - ) - - expected_metadata = () - retry = retries.Retry() - timeout = 5 - pager = client.fetch_profiles(request={}, retry=retry, timeout=timeout) - - assert pager._metadata == expected_metadata - assert pager._retry == retry - assert pager._timeout == timeout - - results = list(pager) - assert len(results) == 6 - assert all(isinstance(i, gkerecommender.Profile) for i in results) - - -def test_fetch_profiles_pages(transport_name: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport_name, - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - next_page_token="abc", - ), - gkerecommender.FetchProfilesResponse( - profile=[], - next_page_token="def", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - ), - RuntimeError, - ) - pages = list(client.fetch_profiles(request={}).pages) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -@pytest.mark.asyncio -async def test_fetch_profiles_async_pager(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_profiles), "__call__", new_callable=mock.AsyncMock - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - next_page_token="abc", - ), - gkerecommender.FetchProfilesResponse( - profile=[], - next_page_token="def", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - ), - RuntimeError, - ) - async_pager = await client.fetch_profiles( - request={}, - ) - assert async_pager.next_page_token == "abc" - responses = [] - async for response in async_pager: # pragma: no branch - responses.append(response) - - assert len(responses) == 6 - assert all(isinstance(i, gkerecommender.Profile) for i in responses) - - -@pytest.mark.asyncio -async def test_fetch_profiles_async_pages(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_profiles), "__call__", new_callable=mock.AsyncMock - ) as call: - # Set the response to a series of pages. - call.side_effect = ( - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - next_page_token="abc", - ), - gkerecommender.FetchProfilesResponse( - profile=[], - next_page_token="def", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - ), - RuntimeError, - ) - pages = [] - # Workaround issue in python 3.9 related to code coverage by adding `# pragma: no branch` - # See https://github.com/googleapis/gapic-generator-python/pull/1174#issuecomment-1025132372 - async for page_ in ( # pragma: no branch - await client.fetch_profiles(request={}) - ).pages: - pages.append(page_) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.GenerateOptimizedManifestRequest, - dict, - ], -) -def test_generate_optimized_manifest(request_type, transport: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.generate_optimized_manifest), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = gkerecommender.GenerateOptimizedManifestResponse( - comments=["comments_value"], - manifest_version="manifest_version_value", - ) - response = client.generate_optimized_manifest(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) == 1 - _, args, _ = call.mock_calls[0] - request = gkerecommender.GenerateOptimizedManifestRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, gkerecommender.GenerateOptimizedManifestResponse) - assert response.comments == ["comments_value"] - assert response.manifest_version == "manifest_version_value" - - -def test_generate_optimized_manifest_non_empty_request_with_auto_populated_field(): - # This test is a coverage failsafe to make sure that UUID4 fields are - # automatically populated, according to AIP-4235, with non-empty requests. - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Populate all string fields in the request which are not UUID4 - # since we want to check that UUID4 are populated automatically - # if they meet the requirements of AIP 4235. - request = gkerecommender.GenerateOptimizedManifestRequest( - accelerator_type="accelerator_type_value", - kubernetes_namespace="kubernetes_namespace_value", - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.generate_optimized_manifest), "__call__" - ) as call: - call.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client.generate_optimized_manifest(request=request) - call.assert_called() - _, args, _ = call.mock_calls[0] - assert args[0] == gkerecommender.GenerateOptimizedManifestRequest( - accelerator_type="accelerator_type_value", - kubernetes_namespace="kubernetes_namespace_value", - ) - - -def test_generate_optimized_manifest_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._transport.generate_optimized_manifest - in client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[ - client._transport.generate_optimized_manifest - ] = mock_rpc - request = {} - client.generate_optimized_manifest(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.generate_optimized_manifest(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_generate_optimized_manifest_async_use_cached_wrapped_rpc( - transport: str = "grpc_asyncio", -): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._client._transport.generate_optimized_manifest - in client._client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.AsyncMock() - mock_rpc.return_value = mock.Mock() - client._client._transport._wrapped_methods[ - client._client._transport.generate_optimized_manifest - ] = mock_rpc - - request = {} - await client.generate_optimized_manifest(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - await client.generate_optimized_manifest(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_generate_optimized_manifest_async( - transport: str = "grpc_asyncio", - request_type=gkerecommender.GenerateOptimizedManifestRequest, -): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.generate_optimized_manifest), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.GenerateOptimizedManifestResponse( - comments=["comments_value"], - manifest_version="manifest_version_value", - ) - ) - response = await client.generate_optimized_manifest(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) - _, args, _ = call.mock_calls[0] - request = gkerecommender.GenerateOptimizedManifestRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, gkerecommender.GenerateOptimizedManifestResponse) - assert response.comments == ["comments_value"] - assert response.manifest_version == "manifest_version_value" - - -@pytest.mark.asyncio -async def test_generate_optimized_manifest_async_from_dict(): - await test_generate_optimized_manifest_async(request_type=dict) - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchBenchmarkingDataRequest, - dict, - ], -) -def test_fetch_benchmarking_data(request_type, transport: str = "grpc"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_benchmarking_data), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = gkerecommender.FetchBenchmarkingDataResponse() - response = client.fetch_benchmarking_data(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) == 1 - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchBenchmarkingDataRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, gkerecommender.FetchBenchmarkingDataResponse) - - -def test_fetch_benchmarking_data_non_empty_request_with_auto_populated_field(): - # This test is a coverage failsafe to make sure that UUID4 fields are - # automatically populated, according to AIP-4235, with non-empty requests. - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Populate all string fields in the request which are not UUID4 - # since we want to check that UUID4 are populated automatically - # if they meet the requirements of AIP 4235. - request = gkerecommender.FetchBenchmarkingDataRequest( - instance_type="instance_type_value", - pricing_model="pricing_model_value", - ) - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_benchmarking_data), "__call__" - ) as call: - call.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client.fetch_benchmarking_data(request=request) - call.assert_called() - _, args, _ = call.mock_calls[0] - assert args[0] == gkerecommender.FetchBenchmarkingDataRequest( - instance_type="instance_type_value", - pricing_model="pricing_model_value", - ) - - -def test_fetch_benchmarking_data_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._transport.fetch_benchmarking_data - in client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[ - client._transport.fetch_benchmarking_data - ] = mock_rpc - request = {} - client.fetch_benchmarking_data(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_benchmarking_data(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_benchmarking_data_async_use_cached_wrapped_rpc( - transport: str = "grpc_asyncio", -): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method_async.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._client._transport.fetch_benchmarking_data - in client._client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.AsyncMock() - mock_rpc.return_value = mock.Mock() - client._client._transport._wrapped_methods[ - client._client._transport.fetch_benchmarking_data - ] = mock_rpc - - request = {} - await client.fetch_benchmarking_data(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - await client.fetch_benchmarking_data(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -@pytest.mark.asyncio -async def test_fetch_benchmarking_data_async( - transport: str = "grpc_asyncio", - request_type=gkerecommender.FetchBenchmarkingDataRequest, -): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport=transport, - ) - - # Everything is optional in proto3 as far as the runtime is concerned, - # and we are mocking out the actual API, so just send an empty request. - request = request_type() - - # Mock the actual call within the gRPC stub, and fake the request. - with mock.patch.object( - type(client.transport.fetch_benchmarking_data), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchBenchmarkingDataResponse() - ) - response = await client.fetch_benchmarking_data(request) - - # Establish that the underlying gRPC stub method was called. - assert len(call.mock_calls) - _, args, _ = call.mock_calls[0] - request = gkerecommender.FetchBenchmarkingDataRequest() - assert args[0] == request - - # Establish that the response is the type that we expect. - assert isinstance(response, gkerecommender.FetchBenchmarkingDataResponse) - - -@pytest.mark.asyncio -async def test_fetch_benchmarking_data_async_from_dict(): - await test_fetch_benchmarking_data_async(request_type=dict) - - -def test_fetch_models_rest_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert client._transport.fetch_models in client._transport._wrapped_methods - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[client._transport.fetch_models] = mock_rpc - - request = {} - client.fetch_models(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_models(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -def test_fetch_models_rest_pager(transport: str = "rest"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(Session, "request") as req: - # TODO(kbandes): remove this mock unless there's a good reason for it. - # with mock.patch.object(path_template, 'transcode') as transcode: - # Set the response as a series of pages - response = ( - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelsResponse( - models=[], - next_page_token="def", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelsResponse( - models=[ - str(), - str(), - ], - ), - ) - # Two responses for two calls - response = response + response - - # Wrap the values into proper Response objs - response = tuple( - gkerecommender.FetchModelsResponse.to_json(x) for x in response - ) - return_values = tuple(Response() for i in response) - for return_val, response_val in zip(return_values, response): - return_val._content = response_val.encode("UTF-8") - return_val.status_code = 200 - req.side_effect = return_values - - sample_request = {} - - pager = client.fetch_models(request=sample_request) - - results = list(pager) - assert len(results) == 6 - assert all(isinstance(i, str) for i in results) - - pages = list(client.fetch_models(request=sample_request).pages) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -def test_fetch_model_servers_rest_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._transport.fetch_model_servers in client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[ - client._transport.fetch_model_servers - ] = mock_rpc - - request = {} - client.fetch_model_servers(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_model_servers(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -def test_fetch_model_servers_rest_required_fields( - request_type=gkerecommender.FetchModelServersRequest, -): - transport_class = transports.GkeInferenceQuickstartRestTransport - - request_init = {} - request_init["model"] = "" - request = request_type(**request_init) - pb_request = request_type.pb(request) - jsonified_request = json.loads( - json_format.MessageToJson(pb_request, use_integers_for_enums=False) - ) - - # verify fields with default values are dropped - assert "model" not in jsonified_request - - unset_fields = transport_class( - credentials=ga_credentials.AnonymousCredentials() - ).fetch_model_servers._get_unset_required_fields(jsonified_request) - jsonified_request.update(unset_fields) - - # verify required fields with default values are now present - assert "model" in jsonified_request - assert jsonified_request["model"] == request_init["model"] - - jsonified_request["model"] = "model_value" - - unset_fields = transport_class( - credentials=ga_credentials.AnonymousCredentials() - ).fetch_model_servers._get_unset_required_fields(jsonified_request) - # Check that path parameters and body parameters are not mixing in. - assert not set(unset_fields) - set( - ( - "model", - "page_size", - "page_token", - ) - ) - jsonified_request.update(unset_fields) - - # verify required fields with non-default values are left alone - assert "model" in jsonified_request - assert jsonified_request["model"] == "model_value" - - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - request = request_type(**request_init) - - # Designate an appropriate value for the returned response. - return_value = gkerecommender.FetchModelServersResponse() - # Mock the http request call within the method and fake a response. - with mock.patch.object(Session, "request") as req: - # We need to mock transcode() because providing default values - # for required fields will fail the real version if the http_options - # expect actual values for those fields. - with mock.patch.object(path_template, "transcode") as transcode: - # A uri without fields and an empty body will force all the - # request fields to show up in the query_params. - pb_request = request_type.pb(request) - transcode_result = { - "uri": "v1/sample_method", - "method": "get", - "query_params": pb_request, - } - transcode.return_value = transcode_result - - response_value = Response() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.FetchModelServersResponse.pb(return_value) - json_return_value = json_format.MessageToJson(return_value) - - response_value._content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - - response = client.fetch_model_servers(request) - - expected_params = [ - ( - "model", - "", - ), - ("$alt", "json;enum-encoding=int"), - ] - actual_params = req.call_args.kwargs["params"] - assert expected_params == actual_params - - -def test_fetch_model_servers_rest_unset_required_fields(): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials - ) - - unset_fields = transport.fetch_model_servers._get_unset_required_fields({}) - assert set(unset_fields) == ( - set( - ( - "model", - "pageSize", - "pageToken", - ) - ) - & set(("model",)) - ) - - -def test_fetch_model_servers_rest_pager(transport: str = "rest"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(Session, "request") as req: - # TODO(kbandes): remove this mock unless there's a good reason for it. - # with mock.patch.object(path_template, 'transcode') as transcode: - # Set the response as a series of pages - response = ( - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[], - next_page_token="def", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServersResponse( - model_servers=[ - str(), - str(), - ], - ), - ) - # Two responses for two calls - response = response + response - - # Wrap the values into proper Response objs - response = tuple( - gkerecommender.FetchModelServersResponse.to_json(x) for x in response - ) - return_values = tuple(Response() for i in response) - for return_val, response_val in zip(return_values, response): - return_val._content = response_val.encode("UTF-8") - return_val.status_code = 200 - req.side_effect = return_values - - sample_request = {} - - pager = client.fetch_model_servers(request=sample_request) - - results = list(pager) - assert len(results) == 6 - assert all(isinstance(i, str) for i in results) - - pages = list(client.fetch_model_servers(request=sample_request).pages) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -def test_fetch_model_server_versions_rest_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._transport.fetch_model_server_versions - in client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[ - client._transport.fetch_model_server_versions - ] = mock_rpc - - request = {} - client.fetch_model_server_versions(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_model_server_versions(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -def test_fetch_model_server_versions_rest_required_fields( - request_type=gkerecommender.FetchModelServerVersionsRequest, -): - transport_class = transports.GkeInferenceQuickstartRestTransport - - request_init = {} - request_init["model"] = "" - request_init["model_server"] = "" - request = request_type(**request_init) - pb_request = request_type.pb(request) - jsonified_request = json.loads( - json_format.MessageToJson(pb_request, use_integers_for_enums=False) - ) - - # verify fields with default values are dropped - assert "model" not in jsonified_request - assert "modelServer" not in jsonified_request - - unset_fields = transport_class( - credentials=ga_credentials.AnonymousCredentials() - ).fetch_model_server_versions._get_unset_required_fields(jsonified_request) - jsonified_request.update(unset_fields) - - # verify required fields with default values are now present - assert "model" in jsonified_request - assert jsonified_request["model"] == request_init["model"] - assert "modelServer" in jsonified_request - assert jsonified_request["modelServer"] == request_init["model_server"] - - jsonified_request["model"] = "model_value" - jsonified_request["modelServer"] = "model_server_value" - - unset_fields = transport_class( - credentials=ga_credentials.AnonymousCredentials() - ).fetch_model_server_versions._get_unset_required_fields(jsonified_request) - # Check that path parameters and body parameters are not mixing in. - assert not set(unset_fields) - set( - ( - "model", - "model_server", - "page_size", - "page_token", - ) - ) - jsonified_request.update(unset_fields) - - # verify required fields with non-default values are left alone - assert "model" in jsonified_request - assert jsonified_request["model"] == "model_value" - assert "modelServer" in jsonified_request - assert jsonified_request["modelServer"] == "model_server_value" - - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - request = request_type(**request_init) - - # Designate an appropriate value for the returned response. - return_value = gkerecommender.FetchModelServerVersionsResponse() - # Mock the http request call within the method and fake a response. - with mock.patch.object(Session, "request") as req: - # We need to mock transcode() because providing default values - # for required fields will fail the real version if the http_options - # expect actual values for those fields. - with mock.patch.object(path_template, "transcode") as transcode: - # A uri without fields and an empty body will force all the - # request fields to show up in the query_params. - pb_request = request_type.pb(request) - transcode_result = { - "uri": "v1/sample_method", - "method": "get", - "query_params": pb_request, - } - transcode.return_value = transcode_result - - response_value = Response() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.FetchModelServerVersionsResponse.pb( - return_value - ) - json_return_value = json_format.MessageToJson(return_value) - - response_value._content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - - response = client.fetch_model_server_versions(request) - - expected_params = [ - ( - "model", - "", - ), - ( - "modelServer", - "", - ), - ("$alt", "json;enum-encoding=int"), - ] - actual_params = req.call_args.kwargs["params"] - assert expected_params == actual_params - - -def test_fetch_model_server_versions_rest_unset_required_fields(): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials - ) - - unset_fields = transport.fetch_model_server_versions._get_unset_required_fields({}) - assert set(unset_fields) == ( - set( - ( - "model", - "modelServer", - "pageSize", - "pageToken", - ) - ) - & set( - ( - "model", - "modelServer", - ) - ) - ) - - -def test_fetch_model_server_versions_rest_pager(transport: str = "rest"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(Session, "request") as req: - # TODO(kbandes): remove this mock unless there's a good reason for it. - # with mock.patch.object(path_template, 'transcode') as transcode: - # Set the response as a series of pages - response = ( - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - str(), - ], - next_page_token="abc", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[], - next_page_token="def", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=[ - str(), - str(), - ], - ), - ) - # Two responses for two calls - response = response + response - - # Wrap the values into proper Response objs - response = tuple( - gkerecommender.FetchModelServerVersionsResponse.to_json(x) for x in response - ) - return_values = tuple(Response() for i in response) - for return_val, response_val in zip(return_values, response): - return_val._content = response_val.encode("UTF-8") - return_val.status_code = 200 - req.side_effect = return_values - - sample_request = {} - - pager = client.fetch_model_server_versions(request=sample_request) - - results = list(pager) - assert len(results) == 6 - assert all(isinstance(i, str) for i in results) - - pages = list(client.fetch_model_server_versions(request=sample_request).pages) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -def test_fetch_profiles_rest_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert client._transport.fetch_profiles in client._transport._wrapped_methods - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[client._transport.fetch_profiles] = mock_rpc - - request = {} - client.fetch_profiles(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_profiles(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -def test_fetch_profiles_rest_pager(transport: str = "rest"): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(Session, "request") as req: - # TODO(kbandes): remove this mock unless there's a good reason for it. - # with mock.patch.object(path_template, 'transcode') as transcode: - # Set the response as a series of pages - response = ( - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - next_page_token="abc", - ), - gkerecommender.FetchProfilesResponse( - profile=[], - next_page_token="def", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - ], - next_page_token="ghi", - ), - gkerecommender.FetchProfilesResponse( - profile=[ - gkerecommender.Profile(), - gkerecommender.Profile(), - ], - ), - ) - # Two responses for two calls - response = response + response - - # Wrap the values into proper Response objs - response = tuple( - gkerecommender.FetchProfilesResponse.to_json(x) for x in response - ) - return_values = tuple(Response() for i in response) - for return_val, response_val in zip(return_values, response): - return_val._content = response_val.encode("UTF-8") - return_val.status_code = 200 - req.side_effect = return_values - - sample_request = {} - - pager = client.fetch_profiles(request=sample_request) - - results = list(pager) - assert len(results) == 6 - assert all(isinstance(i, gkerecommender.Profile) for i in results) - - pages = list(client.fetch_profiles(request=sample_request).pages) - for page_, token in zip(pages, ["abc", "def", "ghi", ""]): - assert page_.raw_page.next_page_token == token - - -def test_generate_optimized_manifest_rest_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._transport.generate_optimized_manifest - in client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[ - client._transport.generate_optimized_manifest - ] = mock_rpc - - request = {} - client.generate_optimized_manifest(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.generate_optimized_manifest(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -def test_generate_optimized_manifest_rest_required_fields( - request_type=gkerecommender.GenerateOptimizedManifestRequest, -): - transport_class = transports.GkeInferenceQuickstartRestTransport - - request_init = {} - request_init["accelerator_type"] = "" - request = request_type(**request_init) - pb_request = request_type.pb(request) - jsonified_request = json.loads( - json_format.MessageToJson(pb_request, use_integers_for_enums=False) - ) - - # verify fields with default values are dropped - - unset_fields = transport_class( - credentials=ga_credentials.AnonymousCredentials() - ).generate_optimized_manifest._get_unset_required_fields(jsonified_request) - jsonified_request.update(unset_fields) - - # verify required fields with default values are now present - - jsonified_request["acceleratorType"] = "accelerator_type_value" - - unset_fields = transport_class( - credentials=ga_credentials.AnonymousCredentials() - ).generate_optimized_manifest._get_unset_required_fields(jsonified_request) - jsonified_request.update(unset_fields) - - # verify required fields with non-default values are left alone - assert "acceleratorType" in jsonified_request - assert jsonified_request["acceleratorType"] == "accelerator_type_value" - - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - request = request_type(**request_init) - - # Designate an appropriate value for the returned response. - return_value = gkerecommender.GenerateOptimizedManifestResponse() - # Mock the http request call within the method and fake a response. - with mock.patch.object(Session, "request") as req: - # We need to mock transcode() because providing default values - # for required fields will fail the real version if the http_options - # expect actual values for those fields. - with mock.patch.object(path_template, "transcode") as transcode: - # A uri without fields and an empty body will force all the - # request fields to show up in the query_params. - pb_request = request_type.pb(request) - transcode_result = { - "uri": "v1/sample_method", - "method": "post", - "query_params": pb_request, - } - transcode_result["body"] = pb_request - transcode.return_value = transcode_result - - response_value = Response() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.GenerateOptimizedManifestResponse.pb( - return_value - ) - json_return_value = json_format.MessageToJson(return_value) - - response_value._content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - - response = client.generate_optimized_manifest(request) - - expected_params = [("$alt", "json;enum-encoding=int")] - actual_params = req.call_args.kwargs["params"] - assert expected_params == actual_params - - -def test_generate_optimized_manifest_rest_unset_required_fields(): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials - ) - - unset_fields = transport.generate_optimized_manifest._get_unset_required_fields({}) - assert set(unset_fields) == ( - set(()) - & set( - ( - "modelServerInfo", - "acceleratorType", - ) - ) - ) - - -def test_fetch_benchmarking_data_rest_use_cached_wrapped_rpc(): - # Clients should use _prep_wrapped_messages to create cached wrapped rpcs, - # instead of constructing them on each call - with mock.patch("google.api_core.gapic_v1.method.wrap_method") as wrapper_fn: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Should wrap all calls on client creation - assert wrapper_fn.call_count > 0 - wrapper_fn.reset_mock() - - # Ensure method has been cached - assert ( - client._transport.fetch_benchmarking_data - in client._transport._wrapped_methods - ) - - # Replace cached wrapped function with mock - mock_rpc = mock.Mock() - mock_rpc.return_value.name = ( - "foo" # operation_request.operation in compute client(s) expect a string. - ) - client._transport._wrapped_methods[ - client._transport.fetch_benchmarking_data - ] = mock_rpc - - request = {} - client.fetch_benchmarking_data(request) - - # Establish that the underlying gRPC stub method was called. - assert mock_rpc.call_count == 1 - - client.fetch_benchmarking_data(request) - - # Establish that a new wrapper was not created for this call - assert wrapper_fn.call_count == 0 - assert mock_rpc.call_count == 2 - - -def test_fetch_benchmarking_data_rest_required_fields( - request_type=gkerecommender.FetchBenchmarkingDataRequest, -): - transport_class = transports.GkeInferenceQuickstartRestTransport - - request_init = {} - request = request_type(**request_init) - pb_request = request_type.pb(request) - jsonified_request = json.loads( - json_format.MessageToJson(pb_request, use_integers_for_enums=False) - ) - - # verify fields with default values are dropped - - unset_fields = transport_class( - credentials=ga_credentials.AnonymousCredentials() - ).fetch_benchmarking_data._get_unset_required_fields(jsonified_request) - jsonified_request.update(unset_fields) - - # verify required fields with default values are now present - - unset_fields = transport_class( - credentials=ga_credentials.AnonymousCredentials() - ).fetch_benchmarking_data._get_unset_required_fields(jsonified_request) - jsonified_request.update(unset_fields) - - # verify required fields with non-default values are left alone - - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - request = request_type(**request_init) - - # Designate an appropriate value for the returned response. - return_value = gkerecommender.FetchBenchmarkingDataResponse() - # Mock the http request call within the method and fake a response. - with mock.patch.object(Session, "request") as req: - # We need to mock transcode() because providing default values - # for required fields will fail the real version if the http_options - # expect actual values for those fields. - with mock.patch.object(path_template, "transcode") as transcode: - # A uri without fields and an empty body will force all the - # request fields to show up in the query_params. - pb_request = request_type.pb(request) - transcode_result = { - "uri": "v1/sample_method", - "method": "post", - "query_params": pb_request, - } - transcode_result["body"] = pb_request - transcode.return_value = transcode_result - - response_value = Response() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.FetchBenchmarkingDataResponse.pb(return_value) - json_return_value = json_format.MessageToJson(return_value) - - response_value._content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - - response = client.fetch_benchmarking_data(request) - - expected_params = [("$alt", "json;enum-encoding=int")] - actual_params = req.call_args.kwargs["params"] - assert expected_params == actual_params - - -def test_fetch_benchmarking_data_rest_unset_required_fields(): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials - ) - - unset_fields = transport.fetch_benchmarking_data._get_unset_required_fields({}) - assert set(unset_fields) == (set(()) & set(("modelServerInfo",))) - - -def test_credentials_transport_error(): - # It is an error to provide credentials and a transport instance. - transport = transports.GkeInferenceQuickstartGrpcTransport( - credentials=ga_credentials.AnonymousCredentials(), - ) - with pytest.raises(ValueError): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport=transport, - ) - - # It is an error to provide a credentials file and a transport instance. - transport = transports.GkeInferenceQuickstartGrpcTransport( - credentials=ga_credentials.AnonymousCredentials(), - ) - with pytest.raises(ValueError): - client = GkeInferenceQuickstartClient( - client_options={"credentials_file": "credentials.json"}, - transport=transport, - ) - - # It is an error to provide an api_key and a transport instance. - transport = transports.GkeInferenceQuickstartGrpcTransport( - credentials=ga_credentials.AnonymousCredentials(), - ) - options = client_options.ClientOptions() - options.api_key = "api_key" - with pytest.raises(ValueError): - client = GkeInferenceQuickstartClient( - client_options=options, - transport=transport, - ) - - # It is an error to provide an api_key and a credential. - options = client_options.ClientOptions() - options.api_key = "api_key" - with pytest.raises(ValueError): - client = GkeInferenceQuickstartClient( - client_options=options, credentials=ga_credentials.AnonymousCredentials() - ) - - # It is an error to provide scopes and a transport instance. - transport = transports.GkeInferenceQuickstartGrpcTransport( - credentials=ga_credentials.AnonymousCredentials(), - ) - with pytest.raises(ValueError): - client = GkeInferenceQuickstartClient( - client_options={"scopes": ["1", "2"]}, - transport=transport, - ) - - -def test_transport_instance(): - # A client may be instantiated with a custom transport instance. - transport = transports.GkeInferenceQuickstartGrpcTransport( - credentials=ga_credentials.AnonymousCredentials(), - ) - client = GkeInferenceQuickstartClient(transport=transport) - assert client.transport is transport - - -def test_transport_get_channel(): - # A client may be instantiated with a custom transport instance. - transport = transports.GkeInferenceQuickstartGrpcTransport( - credentials=ga_credentials.AnonymousCredentials(), - ) - channel = transport.grpc_channel - assert channel - - transport = transports.GkeInferenceQuickstartGrpcAsyncIOTransport( - credentials=ga_credentials.AnonymousCredentials(), - ) - channel = transport.grpc_channel - assert channel - - -@pytest.mark.parametrize( - "transport_class", - [ - transports.GkeInferenceQuickstartGrpcTransport, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - transports.GkeInferenceQuickstartRestTransport, - ], -) -def test_transport_adc(transport_class): - # Test default credentials are used if not provided. - with mock.patch.object(google.auth, "default") as adc: - adc.return_value = (ga_credentials.AnonymousCredentials(), None) - transport_class() - adc.assert_called_once() - - -def test_transport_kind_grpc(): - transport = GkeInferenceQuickstartClient.get_transport_class("grpc")( - credentials=ga_credentials.AnonymousCredentials() - ) - assert transport.kind == "grpc" - - -def test_initialize_client_w_grpc(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="grpc" - ) - assert client is not None - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_models_empty_call_grpc(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: - call.return_value = gkerecommender.FetchModelsResponse() - client.fetch_models(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelsRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_model_servers_empty_call_grpc(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), "__call__" - ) as call: - call.return_value = gkerecommender.FetchModelServersResponse() - client.fetch_model_servers(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelServersRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_model_server_versions_empty_call_grpc(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), "__call__" - ) as call: - call.return_value = gkerecommender.FetchModelServerVersionsResponse() - client.fetch_model_server_versions(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelServerVersionsRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_profiles_empty_call_grpc(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: - call.return_value = gkerecommender.FetchProfilesResponse() - client.fetch_profiles(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchProfilesRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_generate_optimized_manifest_empty_call_grpc(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.generate_optimized_manifest), "__call__" - ) as call: - call.return_value = gkerecommender.GenerateOptimizedManifestResponse() - client.generate_optimized_manifest(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.GenerateOptimizedManifestRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_benchmarking_data_empty_call_grpc(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="grpc", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_benchmarking_data), "__call__" - ) as call: - call.return_value = gkerecommender.FetchBenchmarkingDataResponse() - client.fetch_benchmarking_data(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchBenchmarkingDataRequest() - - assert args[0] == request_msg - - -def test_transport_kind_grpc_asyncio(): - transport = GkeInferenceQuickstartAsyncClient.get_transport_class("grpc_asyncio")( - credentials=async_anonymous_credentials() - ) - assert transport.kind == "grpc_asyncio" - - -def test_initialize_client_w_grpc_asyncio(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), transport="grpc_asyncio" - ) - assert client is not None - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -@pytest.mark.asyncio -async def test_fetch_models_empty_call_grpc_asyncio(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport="grpc_asyncio", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchModelsResponse( - models=["models_value"], - next_page_token="next_page_token_value", - ) - ) - await client.fetch_models(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelsRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -@pytest.mark.asyncio -async def test_fetch_model_servers_empty_call_grpc_asyncio(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport="grpc_asyncio", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchModelServersResponse( - model_servers=["model_servers_value"], - next_page_token="next_page_token_value", - ) - ) - await client.fetch_model_servers(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelServersRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -@pytest.mark.asyncio -async def test_fetch_model_server_versions_empty_call_grpc_asyncio(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport="grpc_asyncio", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=["model_server_versions_value"], - next_page_token="next_page_token_value", - ) - ) - await client.fetch_model_server_versions(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelServerVersionsRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -@pytest.mark.asyncio -async def test_fetch_profiles_empty_call_grpc_asyncio(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport="grpc_asyncio", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchProfilesResponse( - comments="comments_value", - next_page_token="next_page_token_value", - ) - ) - await client.fetch_profiles(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchProfilesRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -@pytest.mark.asyncio -async def test_generate_optimized_manifest_empty_call_grpc_asyncio(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport="grpc_asyncio", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.generate_optimized_manifest), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.GenerateOptimizedManifestResponse( - comments=["comments_value"], - manifest_version="manifest_version_value", - ) - ) - await client.generate_optimized_manifest(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.GenerateOptimizedManifestRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -@pytest.mark.asyncio -async def test_fetch_benchmarking_data_empty_call_grpc_asyncio(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), - transport="grpc_asyncio", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_benchmarking_data), "__call__" - ) as call: - # Designate an appropriate return value for the call. - call.return_value = grpc_helpers_async.FakeUnaryUnaryCall( - gkerecommender.FetchBenchmarkingDataResponse() - ) - await client.fetch_benchmarking_data(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchBenchmarkingDataRequest() - - assert args[0] == request_msg - - -def test_transport_kind_rest(): - transport = GkeInferenceQuickstartClient.get_transport_class("rest")( - credentials=ga_credentials.AnonymousCredentials() - ) - assert transport.kind == "rest" - - -def test_fetch_models_rest_bad_request(request_type=gkerecommender.FetchModelsRequest): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a BadRequest error. - with mock.patch.object(Session, "request") as req, pytest.raises( - core_exceptions.BadRequest - ): - # Wrap the value into a proper Response obj - response_value = mock.Mock() - json_return_value = "" - response_value.json = mock.Mock(return_value={}) - response_value.status_code = 400 - response_value.request = mock.Mock() - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - client.fetch_models(request) - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchModelsRequest, - dict, - ], -) -def test_fetch_models_rest_call_success(request_type): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(type(client.transport._session), "request") as req: - # Designate an appropriate value for the returned response. - return_value = gkerecommender.FetchModelsResponse( - models=["models_value"], - next_page_token="next_page_token_value", - ) - - # Wrap the value into a proper Response obj - response_value = mock.Mock() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.FetchModelsResponse.pb(return_value) - json_return_value = json_format.MessageToJson(return_value) - response_value.content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - response = client.fetch_models(request) - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelsPager) - assert response.models == ["models_value"] - assert response.next_page_token == "next_page_token_value" - - -@pytest.mark.parametrize("null_interceptor", [True, False]) -def test_fetch_models_rest_interceptors(null_interceptor): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials(), - interceptor=None - if null_interceptor - else transports.GkeInferenceQuickstartRestInterceptor(), - ) - client = GkeInferenceQuickstartClient(transport=transport) - - with mock.patch.object( - type(client.transport._session), "request" - ) as req, mock.patch.object( - path_template, "transcode" - ) as transcode, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, "post_fetch_models" - ) as post, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "post_fetch_models_with_metadata", - ) as post_with_metadata, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, "pre_fetch_models" - ) as pre: - pre.assert_not_called() - post.assert_not_called() - post_with_metadata.assert_not_called() - pb_message = gkerecommender.FetchModelsRequest.pb( - gkerecommender.FetchModelsRequest() - ) - transcode.return_value = { - "method": "post", - "uri": "my_uri", - "body": pb_message, - "query_params": pb_message, - } - - req.return_value = mock.Mock() - req.return_value.status_code = 200 - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - return_value = gkerecommender.FetchModelsResponse.to_json( - gkerecommender.FetchModelsResponse() - ) - req.return_value.content = return_value - - request = gkerecommender.FetchModelsRequest() - metadata = [ - ("key", "val"), - ("cephalopod", "squid"), - ] - pre.return_value = request, metadata - post.return_value = gkerecommender.FetchModelsResponse() - post_with_metadata.return_value = gkerecommender.FetchModelsResponse(), metadata - - client.fetch_models( - request, - metadata=[ - ("key", "val"), - ("cephalopod", "squid"), - ], - ) - - pre.assert_called_once() - post.assert_called_once() - post_with_metadata.assert_called_once() - - -def test_fetch_model_servers_rest_bad_request( - request_type=gkerecommender.FetchModelServersRequest, -): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a BadRequest error. - with mock.patch.object(Session, "request") as req, pytest.raises( - core_exceptions.BadRequest - ): - # Wrap the value into a proper Response obj - response_value = mock.Mock() - json_return_value = "" - response_value.json = mock.Mock(return_value={}) - response_value.status_code = 400 - response_value.request = mock.Mock() - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - client.fetch_model_servers(request) - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchModelServersRequest, - dict, - ], -) -def test_fetch_model_servers_rest_call_success(request_type): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(type(client.transport._session), "request") as req: - # Designate an appropriate value for the returned response. - return_value = gkerecommender.FetchModelServersResponse( - model_servers=["model_servers_value"], - next_page_token="next_page_token_value", - ) - - # Wrap the value into a proper Response obj - response_value = mock.Mock() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.FetchModelServersResponse.pb(return_value) - json_return_value = json_format.MessageToJson(return_value) - response_value.content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - response = client.fetch_model_servers(request) - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelServersPager) - assert response.model_servers == ["model_servers_value"] - assert response.next_page_token == "next_page_token_value" - - -@pytest.mark.parametrize("null_interceptor", [True, False]) -def test_fetch_model_servers_rest_interceptors(null_interceptor): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials(), - interceptor=None - if null_interceptor - else transports.GkeInferenceQuickstartRestInterceptor(), - ) - client = GkeInferenceQuickstartClient(transport=transport) - - with mock.patch.object( - type(client.transport._session), "request" - ) as req, mock.patch.object( - path_template, "transcode" - ) as transcode, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, "post_fetch_model_servers" - ) as post, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "post_fetch_model_servers_with_metadata", - ) as post_with_metadata, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, "pre_fetch_model_servers" - ) as pre: - pre.assert_not_called() - post.assert_not_called() - post_with_metadata.assert_not_called() - pb_message = gkerecommender.FetchModelServersRequest.pb( - gkerecommender.FetchModelServersRequest() - ) - transcode.return_value = { - "method": "post", - "uri": "my_uri", - "body": pb_message, - "query_params": pb_message, - } - - req.return_value = mock.Mock() - req.return_value.status_code = 200 - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - return_value = gkerecommender.FetchModelServersResponse.to_json( - gkerecommender.FetchModelServersResponse() - ) - req.return_value.content = return_value - - request = gkerecommender.FetchModelServersRequest() - metadata = [ - ("key", "val"), - ("cephalopod", "squid"), - ] - pre.return_value = request, metadata - post.return_value = gkerecommender.FetchModelServersResponse() - post_with_metadata.return_value = ( - gkerecommender.FetchModelServersResponse(), - metadata, - ) - - client.fetch_model_servers( - request, - metadata=[ - ("key", "val"), - ("cephalopod", "squid"), - ], - ) - - pre.assert_called_once() - post.assert_called_once() - post_with_metadata.assert_called_once() - - -def test_fetch_model_server_versions_rest_bad_request( - request_type=gkerecommender.FetchModelServerVersionsRequest, -): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a BadRequest error. - with mock.patch.object(Session, "request") as req, pytest.raises( - core_exceptions.BadRequest - ): - # Wrap the value into a proper Response obj - response_value = mock.Mock() - json_return_value = "" - response_value.json = mock.Mock(return_value={}) - response_value.status_code = 400 - response_value.request = mock.Mock() - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - client.fetch_model_server_versions(request) - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchModelServerVersionsRequest, - dict, - ], -) -def test_fetch_model_server_versions_rest_call_success(request_type): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(type(client.transport._session), "request") as req: - # Designate an appropriate value for the returned response. - return_value = gkerecommender.FetchModelServerVersionsResponse( - model_server_versions=["model_server_versions_value"], - next_page_token="next_page_token_value", - ) - - # Wrap the value into a proper Response obj - response_value = mock.Mock() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.FetchModelServerVersionsResponse.pb(return_value) - json_return_value = json_format.MessageToJson(return_value) - response_value.content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - response = client.fetch_model_server_versions(request) - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchModelServerVersionsPager) - assert response.model_server_versions == ["model_server_versions_value"] - assert response.next_page_token == "next_page_token_value" - - -@pytest.mark.parametrize("null_interceptor", [True, False]) -def test_fetch_model_server_versions_rest_interceptors(null_interceptor): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials(), - interceptor=None - if null_interceptor - else transports.GkeInferenceQuickstartRestInterceptor(), - ) - client = GkeInferenceQuickstartClient(transport=transport) - - with mock.patch.object( - type(client.transport._session), "request" - ) as req, mock.patch.object( - path_template, "transcode" - ) as transcode, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "post_fetch_model_server_versions", - ) as post, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "post_fetch_model_server_versions_with_metadata", - ) as post_with_metadata, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "pre_fetch_model_server_versions", - ) as pre: - pre.assert_not_called() - post.assert_not_called() - post_with_metadata.assert_not_called() - pb_message = gkerecommender.FetchModelServerVersionsRequest.pb( - gkerecommender.FetchModelServerVersionsRequest() - ) - transcode.return_value = { - "method": "post", - "uri": "my_uri", - "body": pb_message, - "query_params": pb_message, - } - - req.return_value = mock.Mock() - req.return_value.status_code = 200 - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - return_value = gkerecommender.FetchModelServerVersionsResponse.to_json( - gkerecommender.FetchModelServerVersionsResponse() - ) - req.return_value.content = return_value - - request = gkerecommender.FetchModelServerVersionsRequest() - metadata = [ - ("key", "val"), - ("cephalopod", "squid"), - ] - pre.return_value = request, metadata - post.return_value = gkerecommender.FetchModelServerVersionsResponse() - post_with_metadata.return_value = ( - gkerecommender.FetchModelServerVersionsResponse(), - metadata, - ) - - client.fetch_model_server_versions( - request, - metadata=[ - ("key", "val"), - ("cephalopod", "squid"), - ], - ) - - pre.assert_called_once() - post.assert_called_once() - post_with_metadata.assert_called_once() - - -def test_fetch_profiles_rest_bad_request( - request_type=gkerecommender.FetchProfilesRequest, -): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a BadRequest error. - with mock.patch.object(Session, "request") as req, pytest.raises( - core_exceptions.BadRequest - ): - # Wrap the value into a proper Response obj - response_value = mock.Mock() - json_return_value = "" - response_value.json = mock.Mock(return_value={}) - response_value.status_code = 400 - response_value.request = mock.Mock() - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - client.fetch_profiles(request) - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchProfilesRequest, - dict, - ], -) -def test_fetch_profiles_rest_call_success(request_type): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(type(client.transport._session), "request") as req: - # Designate an appropriate value for the returned response. - return_value = gkerecommender.FetchProfilesResponse( - comments="comments_value", - next_page_token="next_page_token_value", - ) - - # Wrap the value into a proper Response obj - response_value = mock.Mock() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.FetchProfilesResponse.pb(return_value) - json_return_value = json_format.MessageToJson(return_value) - response_value.content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - response = client.fetch_profiles(request) - - # Establish that the response is the type that we expect. - assert isinstance(response, pagers.FetchProfilesPager) - assert response.comments == "comments_value" - assert response.next_page_token == "next_page_token_value" - - -@pytest.mark.parametrize("null_interceptor", [True, False]) -def test_fetch_profiles_rest_interceptors(null_interceptor): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials(), - interceptor=None - if null_interceptor - else transports.GkeInferenceQuickstartRestInterceptor(), - ) - client = GkeInferenceQuickstartClient(transport=transport) - - with mock.patch.object( - type(client.transport._session), "request" - ) as req, mock.patch.object( - path_template, "transcode" - ) as transcode, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, "post_fetch_profiles" - ) as post, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "post_fetch_profiles_with_metadata", - ) as post_with_metadata, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, "pre_fetch_profiles" - ) as pre: - pre.assert_not_called() - post.assert_not_called() - post_with_metadata.assert_not_called() - pb_message = gkerecommender.FetchProfilesRequest.pb( - gkerecommender.FetchProfilesRequest() - ) - transcode.return_value = { - "method": "post", - "uri": "my_uri", - "body": pb_message, - "query_params": pb_message, - } - - req.return_value = mock.Mock() - req.return_value.status_code = 200 - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - return_value = gkerecommender.FetchProfilesResponse.to_json( - gkerecommender.FetchProfilesResponse() - ) - req.return_value.content = return_value - - request = gkerecommender.FetchProfilesRequest() - metadata = [ - ("key", "val"), - ("cephalopod", "squid"), - ] - pre.return_value = request, metadata - post.return_value = gkerecommender.FetchProfilesResponse() - post_with_metadata.return_value = ( - gkerecommender.FetchProfilesResponse(), - metadata, - ) - - client.fetch_profiles( - request, - metadata=[ - ("key", "val"), - ("cephalopod", "squid"), - ], - ) - - pre.assert_called_once() - post.assert_called_once() - post_with_metadata.assert_called_once() - - -def test_generate_optimized_manifest_rest_bad_request( - request_type=gkerecommender.GenerateOptimizedManifestRequest, -): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a BadRequest error. - with mock.patch.object(Session, "request") as req, pytest.raises( - core_exceptions.BadRequest - ): - # Wrap the value into a proper Response obj - response_value = mock.Mock() - json_return_value = "" - response_value.json = mock.Mock(return_value={}) - response_value.status_code = 400 - response_value.request = mock.Mock() - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - client.generate_optimized_manifest(request) - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.GenerateOptimizedManifestRequest, - dict, - ], -) -def test_generate_optimized_manifest_rest_call_success(request_type): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(type(client.transport._session), "request") as req: - # Designate an appropriate value for the returned response. - return_value = gkerecommender.GenerateOptimizedManifestResponse( - comments=["comments_value"], - manifest_version="manifest_version_value", - ) - - # Wrap the value into a proper Response obj - response_value = mock.Mock() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.GenerateOptimizedManifestResponse.pb(return_value) - json_return_value = json_format.MessageToJson(return_value) - response_value.content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - response = client.generate_optimized_manifest(request) - - # Establish that the response is the type that we expect. - assert isinstance(response, gkerecommender.GenerateOptimizedManifestResponse) - assert response.comments == ["comments_value"] - assert response.manifest_version == "manifest_version_value" - - -@pytest.mark.parametrize("null_interceptor", [True, False]) -def test_generate_optimized_manifest_rest_interceptors(null_interceptor): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials(), - interceptor=None - if null_interceptor - else transports.GkeInferenceQuickstartRestInterceptor(), - ) - client = GkeInferenceQuickstartClient(transport=transport) - - with mock.patch.object( - type(client.transport._session), "request" - ) as req, mock.patch.object( - path_template, "transcode" - ) as transcode, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "post_generate_optimized_manifest", - ) as post, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "post_generate_optimized_manifest_with_metadata", - ) as post_with_metadata, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "pre_generate_optimized_manifest", - ) as pre: - pre.assert_not_called() - post.assert_not_called() - post_with_metadata.assert_not_called() - pb_message = gkerecommender.GenerateOptimizedManifestRequest.pb( - gkerecommender.GenerateOptimizedManifestRequest() - ) - transcode.return_value = { - "method": "post", - "uri": "my_uri", - "body": pb_message, - "query_params": pb_message, - } - - req.return_value = mock.Mock() - req.return_value.status_code = 200 - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - return_value = gkerecommender.GenerateOptimizedManifestResponse.to_json( - gkerecommender.GenerateOptimizedManifestResponse() - ) - req.return_value.content = return_value - - request = gkerecommender.GenerateOptimizedManifestRequest() - metadata = [ - ("key", "val"), - ("cephalopod", "squid"), - ] - pre.return_value = request, metadata - post.return_value = gkerecommender.GenerateOptimizedManifestResponse() - post_with_metadata.return_value = ( - gkerecommender.GenerateOptimizedManifestResponse(), - metadata, - ) - - client.generate_optimized_manifest( - request, - metadata=[ - ("key", "val"), - ("cephalopod", "squid"), - ], - ) - - pre.assert_called_once() - post.assert_called_once() - post_with_metadata.assert_called_once() - - -def test_fetch_benchmarking_data_rest_bad_request( - request_type=gkerecommender.FetchBenchmarkingDataRequest, -): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a BadRequest error. - with mock.patch.object(Session, "request") as req, pytest.raises( - core_exceptions.BadRequest - ): - # Wrap the value into a proper Response obj - response_value = mock.Mock() - json_return_value = "" - response_value.json = mock.Mock(return_value={}) - response_value.status_code = 400 - response_value.request = mock.Mock() - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - client.fetch_benchmarking_data(request) - - -@pytest.mark.parametrize( - "request_type", - [ - gkerecommender.FetchBenchmarkingDataRequest, - dict, - ], -) -def test_fetch_benchmarking_data_rest_call_success(request_type): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - - # send a request that will satisfy transcoding - request_init = {} - request = request_type(**request_init) - - # Mock the http request call within the method and fake a response. - with mock.patch.object(type(client.transport._session), "request") as req: - # Designate an appropriate value for the returned response. - return_value = gkerecommender.FetchBenchmarkingDataResponse() - - # Wrap the value into a proper Response obj - response_value = mock.Mock() - response_value.status_code = 200 - - # Convert return value to protobuf type - return_value = gkerecommender.FetchBenchmarkingDataResponse.pb(return_value) - json_return_value = json_format.MessageToJson(return_value) - response_value.content = json_return_value.encode("UTF-8") - req.return_value = response_value - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - response = client.fetch_benchmarking_data(request) - - # Establish that the response is the type that we expect. - assert isinstance(response, gkerecommender.FetchBenchmarkingDataResponse) - - -@pytest.mark.parametrize("null_interceptor", [True, False]) -def test_fetch_benchmarking_data_rest_interceptors(null_interceptor): - transport = transports.GkeInferenceQuickstartRestTransport( - credentials=ga_credentials.AnonymousCredentials(), - interceptor=None - if null_interceptor - else transports.GkeInferenceQuickstartRestInterceptor(), - ) - client = GkeInferenceQuickstartClient(transport=transport) - - with mock.patch.object( - type(client.transport._session), "request" - ) as req, mock.patch.object( - path_template, "transcode" - ) as transcode, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, "post_fetch_benchmarking_data" - ) as post, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, - "post_fetch_benchmarking_data_with_metadata", - ) as post_with_metadata, mock.patch.object( - transports.GkeInferenceQuickstartRestInterceptor, "pre_fetch_benchmarking_data" - ) as pre: - pre.assert_not_called() - post.assert_not_called() - post_with_metadata.assert_not_called() - pb_message = gkerecommender.FetchBenchmarkingDataRequest.pb( - gkerecommender.FetchBenchmarkingDataRequest() - ) - transcode.return_value = { - "method": "post", - "uri": "my_uri", - "body": pb_message, - "query_params": pb_message, - } - - req.return_value = mock.Mock() - req.return_value.status_code = 200 - req.return_value.headers = {"header-1": "value-1", "header-2": "value-2"} - return_value = gkerecommender.FetchBenchmarkingDataResponse.to_json( - gkerecommender.FetchBenchmarkingDataResponse() - ) - req.return_value.content = return_value - - request = gkerecommender.FetchBenchmarkingDataRequest() - metadata = [ - ("key", "val"), - ("cephalopod", "squid"), - ] - pre.return_value = request, metadata - post.return_value = gkerecommender.FetchBenchmarkingDataResponse() - post_with_metadata.return_value = ( - gkerecommender.FetchBenchmarkingDataResponse(), - metadata, - ) - - client.fetch_benchmarking_data( - request, - metadata=[ - ("key", "val"), - ("cephalopod", "squid"), - ], - ) - - pre.assert_called_once() - post.assert_called_once() - post_with_metadata.assert_called_once() - - -def test_initialize_client_w_rest(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - assert client is not None - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_models_empty_call_rest(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object(type(client.transport.fetch_models), "__call__") as call: - client.fetch_models(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelsRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_model_servers_empty_call_rest(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_servers), "__call__" - ) as call: - client.fetch_model_servers(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelServersRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_model_server_versions_empty_call_rest(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_model_server_versions), "__call__" - ) as call: - client.fetch_model_server_versions(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchModelServerVersionsRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_profiles_empty_call_rest(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object(type(client.transport.fetch_profiles), "__call__") as call: - client.fetch_profiles(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchProfilesRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_generate_optimized_manifest_empty_call_rest(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.generate_optimized_manifest), "__call__" - ) as call: - client.generate_optimized_manifest(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.GenerateOptimizedManifestRequest() - - assert args[0] == request_msg - - -# This test is a coverage failsafe to make sure that totally empty calls, -# i.e. request == None and no flattened fields passed, work. -def test_fetch_benchmarking_data_empty_call_rest(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - transport="rest", - ) - - # Mock the actual call, and fake the request. - with mock.patch.object( - type(client.transport.fetch_benchmarking_data), "__call__" - ) as call: - client.fetch_benchmarking_data(request=None) - - # Establish that the underlying stub method was called. - call.assert_called() - _, args, _ = call.mock_calls[0] - request_msg = gkerecommender.FetchBenchmarkingDataRequest() - - assert args[0] == request_msg - - -def test_transport_grpc_default(): - # A client should use the gRPC transport by default. - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - ) - assert isinstance( - client.transport, - transports.GkeInferenceQuickstartGrpcTransport, - ) - - -def test_gke_inference_quickstart_base_transport_error(): - # Passing both a credentials object and credentials_file should raise an error - with pytest.raises(core_exceptions.DuplicateCredentialArgs): - transport = transports.GkeInferenceQuickstartTransport( - credentials=ga_credentials.AnonymousCredentials(), - credentials_file="credentials.json", - ) - - -def test_gke_inference_quickstart_base_transport(): - # Instantiate the base transport. - with mock.patch( - "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.GkeInferenceQuickstartTransport.__init__" - ) as Transport: - Transport.return_value = None - transport = transports.GkeInferenceQuickstartTransport( - credentials=ga_credentials.AnonymousCredentials(), - ) - - # Every method on the transport should just blindly - # raise NotImplementedError. - methods = ( - "fetch_models", - "fetch_model_servers", - "fetch_model_server_versions", - "fetch_profiles", - "generate_optimized_manifest", - "fetch_benchmarking_data", - ) - for method in methods: - with pytest.raises(NotImplementedError): - getattr(transport, method)(request=object()) - - with pytest.raises(NotImplementedError): - transport.close() - - # Catch all for all remaining methods and properties - remainder = [ - "kind", - ] - for r in remainder: - with pytest.raises(NotImplementedError): - getattr(transport, r)() - - -def test_gke_inference_quickstart_base_transport_with_credentials_file(): - # Instantiate the base transport with a credentials file - with mock.patch.object( - google.auth, "load_credentials_from_file", autospec=True - ) as load_creds, mock.patch( - "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.GkeInferenceQuickstartTransport._prep_wrapped_messages" - ) as Transport: - Transport.return_value = None - load_creds.return_value = (ga_credentials.AnonymousCredentials(), None) - transport = transports.GkeInferenceQuickstartTransport( - credentials_file="credentials.json", - quota_project_id="octopus", - ) - load_creds.assert_called_once_with( - "credentials.json", - scopes=None, - default_scopes=("https://www.googleapis.com/auth/cloud-platform",), - quota_project_id="octopus", - ) - - -def test_gke_inference_quickstart_base_transport_with_adc(): - # Test the default credentials are used if credentials and credentials_file are None. - with mock.patch.object(google.auth, "default", autospec=True) as adc, mock.patch( - "google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.GkeInferenceQuickstartTransport._prep_wrapped_messages" - ) as Transport: - Transport.return_value = None - adc.return_value = (ga_credentials.AnonymousCredentials(), None) - transport = transports.GkeInferenceQuickstartTransport() - adc.assert_called_once() - - -def test_gke_inference_quickstart_auth_adc(): - # If no credentials are provided, we should use ADC credentials. - with mock.patch.object(google.auth, "default", autospec=True) as adc: - adc.return_value = (ga_credentials.AnonymousCredentials(), None) - GkeInferenceQuickstartClient() - adc.assert_called_once_with( - scopes=None, - default_scopes=("https://www.googleapis.com/auth/cloud-platform",), - quota_project_id=None, - ) - - -@pytest.mark.parametrize( - "transport_class", - [ - transports.GkeInferenceQuickstartGrpcTransport, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - ], -) -def test_gke_inference_quickstart_transport_auth_adc(transport_class): - # If credentials and host are not provided, the transport class should use - # ADC credentials. - with mock.patch.object(google.auth, "default", autospec=True) as adc: - adc.return_value = (ga_credentials.AnonymousCredentials(), None) - transport_class(quota_project_id="octopus", scopes=["1", "2"]) - adc.assert_called_once_with( - scopes=["1", "2"], - default_scopes=("https://www.googleapis.com/auth/cloud-platform",), - quota_project_id="octopus", - ) - - -@pytest.mark.parametrize( - "transport_class", - [ - transports.GkeInferenceQuickstartGrpcTransport, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - transports.GkeInferenceQuickstartRestTransport, - ], -) -def test_gke_inference_quickstart_transport_auth_gdch_credentials(transport_class): - host = "https://language.com" - api_audience_tests = [None, "https://language2.com"] - api_audience_expect = [host, "https://language2.com"] - for t, e in zip(api_audience_tests, api_audience_expect): - with mock.patch.object(google.auth, "default", autospec=True) as adc: - gdch_mock = mock.MagicMock() - type(gdch_mock).with_gdch_audience = mock.PropertyMock( - return_value=gdch_mock - ) - adc.return_value = (gdch_mock, None) - transport_class(host=host, api_audience=t) - gdch_mock.with_gdch_audience.assert_called_once_with(e) - - -@pytest.mark.parametrize( - "transport_class,grpc_helpers", - [ - (transports.GkeInferenceQuickstartGrpcTransport, grpc_helpers), - (transports.GkeInferenceQuickstartGrpcAsyncIOTransport, grpc_helpers_async), - ], -) -def test_gke_inference_quickstart_transport_create_channel( - transport_class, grpc_helpers -): - # If credentials and host are not provided, the transport class should use - # ADC credentials. - with mock.patch.object( - google.auth, "default", autospec=True - ) as adc, mock.patch.object( - grpc_helpers, "create_channel", autospec=True - ) as create_channel: - creds = ga_credentials.AnonymousCredentials() - adc.return_value = (creds, None) - transport_class(quota_project_id="octopus", scopes=["1", "2"]) - - create_channel.assert_called_with( - "gkerecommender.googleapis.com:443", - credentials=creds, - credentials_file=None, - quota_project_id="octopus", - default_scopes=("https://www.googleapis.com/auth/cloud-platform",), - scopes=["1", "2"], - default_host="gkerecommender.googleapis.com", - ssl_credentials=None, - options=[ - ("grpc.max_send_message_length", -1), - ("grpc.max_receive_message_length", -1), - ], - ) - - -@pytest.mark.parametrize( - "transport_class", - [ - transports.GkeInferenceQuickstartGrpcTransport, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - ], -) -def test_gke_inference_quickstart_grpc_transport_client_cert_source_for_mtls( - transport_class, -): - cred = ga_credentials.AnonymousCredentials() - - # Check ssl_channel_credentials is used if provided. - with mock.patch.object(transport_class, "create_channel") as mock_create_channel: - mock_ssl_channel_creds = mock.Mock() - transport_class( - host="squid.clam.whelk", - credentials=cred, - ssl_channel_credentials=mock_ssl_channel_creds, - ) - mock_create_channel.assert_called_once_with( - "squid.clam.whelk:443", - credentials=cred, - credentials_file=None, - scopes=None, - ssl_credentials=mock_ssl_channel_creds, - quota_project_id=None, - options=[ - ("grpc.max_send_message_length", -1), - ("grpc.max_receive_message_length", -1), - ], - ) - - # Check if ssl_channel_credentials is not provided, then client_cert_source_for_mtls - # is used. - with mock.patch.object(transport_class, "create_channel", return_value=mock.Mock()): - with mock.patch("grpc.ssl_channel_credentials") as mock_ssl_cred: - transport_class( - credentials=cred, - client_cert_source_for_mtls=client_cert_source_callback, - ) - expected_cert, expected_key = client_cert_source_callback() - mock_ssl_cred.assert_called_once_with( - certificate_chain=expected_cert, private_key=expected_key - ) - - -def test_gke_inference_quickstart_http_transport_client_cert_source_for_mtls(): - cred = ga_credentials.AnonymousCredentials() - with mock.patch( - "google.auth.transport.requests.AuthorizedSession.configure_mtls_channel" - ) as mock_configure_mtls_channel: - transports.GkeInferenceQuickstartRestTransport( - credentials=cred, client_cert_source_for_mtls=client_cert_source_callback - ) - mock_configure_mtls_channel.assert_called_once_with(client_cert_source_callback) - - -@pytest.mark.parametrize( - "transport_name", - [ - "grpc", - "grpc_asyncio", - "rest", - ], -) -def test_gke_inference_quickstart_host_no_port(transport_name): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - client_options=client_options.ClientOptions( - api_endpoint="gkerecommender.googleapis.com" - ), - transport=transport_name, - ) - assert client.transport._host == ( - "gkerecommender.googleapis.com:443" - if transport_name in ["grpc", "grpc_asyncio"] - else "https://gkerecommender.googleapis.com" - ) - - -@pytest.mark.parametrize( - "transport_name", - [ - "grpc", - "grpc_asyncio", - "rest", - ], -) -def test_gke_inference_quickstart_host_with_port(transport_name): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - client_options=client_options.ClientOptions( - api_endpoint="gkerecommender.googleapis.com:8000" - ), - transport=transport_name, - ) - assert client.transport._host == ( - "gkerecommender.googleapis.com:8000" - if transport_name in ["grpc", "grpc_asyncio"] - else "https://gkerecommender.googleapis.com:8000" - ) - - -@pytest.mark.parametrize( - "transport_name", - [ - "rest", - ], -) -def test_gke_inference_quickstart_client_transport_session_collision(transport_name): - creds1 = ga_credentials.AnonymousCredentials() - creds2 = ga_credentials.AnonymousCredentials() - client1 = GkeInferenceQuickstartClient( - credentials=creds1, - transport=transport_name, - ) - client2 = GkeInferenceQuickstartClient( - credentials=creds2, - transport=transport_name, - ) - session1 = client1.transport.fetch_models._session - session2 = client2.transport.fetch_models._session - assert session1 != session2 - session1 = client1.transport.fetch_model_servers._session - session2 = client2.transport.fetch_model_servers._session - assert session1 != session2 - session1 = client1.transport.fetch_model_server_versions._session - session2 = client2.transport.fetch_model_server_versions._session - assert session1 != session2 - session1 = client1.transport.fetch_profiles._session - session2 = client2.transport.fetch_profiles._session - assert session1 != session2 - session1 = client1.transport.generate_optimized_manifest._session - session2 = client2.transport.generate_optimized_manifest._session - assert session1 != session2 - session1 = client1.transport.fetch_benchmarking_data._session - session2 = client2.transport.fetch_benchmarking_data._session - assert session1 != session2 - - -def test_gke_inference_quickstart_grpc_transport_channel(): - channel = grpc.secure_channel("http://localhost/", grpc.local_channel_credentials()) - - # Check that channel is used if provided. - transport = transports.GkeInferenceQuickstartGrpcTransport( - host="squid.clam.whelk", - channel=channel, - ) - assert transport.grpc_channel == channel - assert transport._host == "squid.clam.whelk:443" - assert transport._ssl_channel_credentials == None - - -def test_gke_inference_quickstart_grpc_asyncio_transport_channel(): - channel = aio.secure_channel("http://localhost/", grpc.local_channel_credentials()) - - # Check that channel is used if provided. - transport = transports.GkeInferenceQuickstartGrpcAsyncIOTransport( - host="squid.clam.whelk", - channel=channel, - ) - assert transport.grpc_channel == channel - assert transport._host == "squid.clam.whelk:443" - assert transport._ssl_channel_credentials == None - - -# Remove this test when deprecated arguments (api_mtls_endpoint, client_cert_source) are -# removed from grpc/grpc_asyncio transport constructor. -@pytest.mark.parametrize( - "transport_class", - [ - transports.GkeInferenceQuickstartGrpcTransport, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - ], -) -def test_gke_inference_quickstart_transport_channel_mtls_with_client_cert_source( - transport_class, -): - with mock.patch( - "grpc.ssl_channel_credentials", autospec=True - ) as grpc_ssl_channel_cred: - with mock.patch.object( - transport_class, "create_channel" - ) as grpc_create_channel: - mock_ssl_cred = mock.Mock() - grpc_ssl_channel_cred.return_value = mock_ssl_cred - - mock_grpc_channel = mock.Mock() - grpc_create_channel.return_value = mock_grpc_channel - - cred = ga_credentials.AnonymousCredentials() - with pytest.warns(DeprecationWarning): - with mock.patch.object(google.auth, "default") as adc: - adc.return_value = (cred, None) - transport = transport_class( - host="squid.clam.whelk", - api_mtls_endpoint="mtls.squid.clam.whelk", - client_cert_source=client_cert_source_callback, - ) - adc.assert_called_once() - - grpc_ssl_channel_cred.assert_called_once_with( - certificate_chain=b"cert bytes", private_key=b"key bytes" - ) - grpc_create_channel.assert_called_once_with( - "mtls.squid.clam.whelk:443", - credentials=cred, - credentials_file=None, - scopes=None, - ssl_credentials=mock_ssl_cred, - quota_project_id=None, - options=[ - ("grpc.max_send_message_length", -1), - ("grpc.max_receive_message_length", -1), - ], - ) - assert transport.grpc_channel == mock_grpc_channel - assert transport._ssl_channel_credentials == mock_ssl_cred - - -# Remove this test when deprecated arguments (api_mtls_endpoint, client_cert_source) are -# removed from grpc/grpc_asyncio transport constructor. -@pytest.mark.parametrize( - "transport_class", - [ - transports.GkeInferenceQuickstartGrpcTransport, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - ], -) -def test_gke_inference_quickstart_transport_channel_mtls_with_adc(transport_class): - mock_ssl_cred = mock.Mock() - with mock.patch.multiple( - "google.auth.transport.grpc.SslCredentials", - __init__=mock.Mock(return_value=None), - ssl_credentials=mock.PropertyMock(return_value=mock_ssl_cred), - ): - with mock.patch.object( - transport_class, "create_channel" - ) as grpc_create_channel: - mock_grpc_channel = mock.Mock() - grpc_create_channel.return_value = mock_grpc_channel - mock_cred = mock.Mock() - - with pytest.warns(DeprecationWarning): - transport = transport_class( - host="squid.clam.whelk", - credentials=mock_cred, - api_mtls_endpoint="mtls.squid.clam.whelk", - client_cert_source=None, - ) - - grpc_create_channel.assert_called_once_with( - "mtls.squid.clam.whelk:443", - credentials=mock_cred, - credentials_file=None, - scopes=None, - ssl_credentials=mock_ssl_cred, - quota_project_id=None, - options=[ - ("grpc.max_send_message_length", -1), - ("grpc.max_receive_message_length", -1), - ], - ) - assert transport.grpc_channel == mock_grpc_channel - - -def test_common_billing_account_path(): - billing_account = "squid" - expected = "billingAccounts/{billing_account}".format( - billing_account=billing_account, - ) - actual = GkeInferenceQuickstartClient.common_billing_account_path(billing_account) - assert expected == actual - - -def test_parse_common_billing_account_path(): - expected = { - "billing_account": "clam", - } - path = GkeInferenceQuickstartClient.common_billing_account_path(**expected) - - # Check that the path construction is reversible. - actual = GkeInferenceQuickstartClient.parse_common_billing_account_path(path) - assert expected == actual - - -def test_common_folder_path(): - folder = "whelk" - expected = "folders/{folder}".format( - folder=folder, - ) - actual = GkeInferenceQuickstartClient.common_folder_path(folder) - assert expected == actual - - -def test_parse_common_folder_path(): - expected = { - "folder": "octopus", - } - path = GkeInferenceQuickstartClient.common_folder_path(**expected) - - # Check that the path construction is reversible. - actual = GkeInferenceQuickstartClient.parse_common_folder_path(path) - assert expected == actual - - -def test_common_organization_path(): - organization = "oyster" - expected = "organizations/{organization}".format( - organization=organization, - ) - actual = GkeInferenceQuickstartClient.common_organization_path(organization) - assert expected == actual - - -def test_parse_common_organization_path(): - expected = { - "organization": "nudibranch", - } - path = GkeInferenceQuickstartClient.common_organization_path(**expected) - - # Check that the path construction is reversible. - actual = GkeInferenceQuickstartClient.parse_common_organization_path(path) - assert expected == actual - - -def test_common_project_path(): - project = "cuttlefish" - expected = "projects/{project}".format( - project=project, - ) - actual = GkeInferenceQuickstartClient.common_project_path(project) - assert expected == actual - - -def test_parse_common_project_path(): - expected = { - "project": "mussel", - } - path = GkeInferenceQuickstartClient.common_project_path(**expected) - - # Check that the path construction is reversible. - actual = GkeInferenceQuickstartClient.parse_common_project_path(path) - assert expected == actual - - -def test_common_location_path(): - project = "winkle" - location = "nautilus" - expected = "projects/{project}/locations/{location}".format( - project=project, - location=location, - ) - actual = GkeInferenceQuickstartClient.common_location_path(project, location) - assert expected == actual - - -def test_parse_common_location_path(): - expected = { - "project": "scallop", - "location": "abalone", - } - path = GkeInferenceQuickstartClient.common_location_path(**expected) - - # Check that the path construction is reversible. - actual = GkeInferenceQuickstartClient.parse_common_location_path(path) - assert expected == actual - - -def test_client_with_default_client_info(): - client_info = gapic_v1.client_info.ClientInfo() - - with mock.patch.object( - transports.GkeInferenceQuickstartTransport, "_prep_wrapped_messages" - ) as prep: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), - client_info=client_info, - ) - prep.assert_called_once_with(client_info) - - with mock.patch.object( - transports.GkeInferenceQuickstartTransport, "_prep_wrapped_messages" - ) as prep: - transport_class = GkeInferenceQuickstartClient.get_transport_class() - transport = transport_class( - credentials=ga_credentials.AnonymousCredentials(), - client_info=client_info, - ) - prep.assert_called_once_with(client_info) - - -def test_transport_close_grpc(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="grpc" - ) - with mock.patch.object( - type(getattr(client.transport, "_grpc_channel")), "close" - ) as close: - with client: - close.assert_not_called() - close.assert_called_once() - - -@pytest.mark.asyncio -async def test_transport_close_grpc_asyncio(): - client = GkeInferenceQuickstartAsyncClient( - credentials=async_anonymous_credentials(), transport="grpc_asyncio" - ) - with mock.patch.object( - type(getattr(client.transport, "_grpc_channel")), "close" - ) as close: - async with client: - close.assert_not_called() - close.assert_called_once() - - -def test_transport_close_rest(): - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport="rest" - ) - with mock.patch.object( - type(getattr(client.transport, "_session")), "close" - ) as close: - with client: - close.assert_not_called() - close.assert_called_once() - - -def test_client_ctx(): - transports = [ - "rest", - "grpc", - ] - for transport in transports: - client = GkeInferenceQuickstartClient( - credentials=ga_credentials.AnonymousCredentials(), transport=transport - ) - # Test client calls underlying transport. - with mock.patch.object(type(client.transport), "close") as close: - close.assert_not_called() - with client: - pass - close.assert_called() - - -@pytest.mark.parametrize( - "client_class,transport_class", - [ - (GkeInferenceQuickstartClient, transports.GkeInferenceQuickstartGrpcTransport), - ( - GkeInferenceQuickstartAsyncClient, - transports.GkeInferenceQuickstartGrpcAsyncIOTransport, - ), - ], -) -def test_api_key_credentials(client_class, transport_class): - with mock.patch.object( - google.auth._default, "get_api_key_credentials", create=True - ) as get_api_key_credentials: - mock_cred = mock.Mock() - get_api_key_credentials.return_value = mock_cred - options = client_options.ClientOptions() - options.api_key = "api_key" - with mock.patch.object(transport_class, "__init__") as patched: - patched.return_value = None - client = client_class(client_options=options) - patched.assert_called_once_with( - credentials=mock_cred, - credentials_file=None, - host=client._DEFAULT_ENDPOINT_TEMPLATE.format( - UNIVERSE_DOMAIN=client._DEFAULT_UNIVERSE - ), - scopes=None, - client_cert_source_for_mtls=None, - quota_project_id=None, - client_info=transports.base.DEFAULT_CLIENT_INFO, - always_use_jwt_access=True, - api_audience=None, - ) From 9ea565ac4ed4e46871b02074fb8fe80bafce6833 Mon Sep 17 00:00:00 2001 From: ohmayr Date: Thu, 25 Sep 2025 18:36:28 +0000 Subject: [PATCH 3/4] remove tar.gz from preserve regex --- .librarian/state.yaml | 11 ----------- 1 file changed, 11 deletions(-) diff --git a/.librarian/state.yaml b/.librarian/state.yaml index 936766195236..4fb806ecc009 100644 --- a/.librarian/state.yaml +++ b/.librarian/state.yaml @@ -13,7 +13,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -33,7 +32,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -53,7 +51,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -73,7 +70,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -101,7 +97,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -123,7 +118,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -145,7 +139,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -165,7 +158,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -185,7 +177,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -206,7 +197,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system @@ -226,7 +216,6 @@ libraries: - docs/CHANGELOG.md - docs/README.rst - samples/README.txt - - tar.gz - scripts/client-post-processing - samples/snippets/README.rst - tests/system From ceb349902de2b5a7a9c8b043a87562dfcad757a5 Mon Sep 17 00:00:00 2001 From: ohmayr Date: Thu, 25 Sep 2025 18:45:24 +0000 Subject: [PATCH 4/4] fix tag format --- .librarian/state.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.librarian/state.yaml b/.librarian/state.yaml index 4fb806ecc009..a86bfd28e49e 100644 --- a/.librarian/state.yaml +++ b/.librarian/state.yaml @@ -221,4 +221,4 @@ libraries: - tests/system remove_regex: - packages/google-cloud-gkerecommender - tag_format: '{{id}}-v{{version}}' + tag_format: '{id}-v{version}'