Merge branch 'main' into falcon_new_semconv

Author: Filip Nikolovski

.readthedocs.yml

@@ -12,4 +12,4 @@ build:
 
 python:
   install:
-    - requirements: docs-requirements.txt
+    - requirements: rtd-requirements.txt

CHANGELOG.md

@@ -48,6 +48,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   ([#3022](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/3022))
 - Replace all instrumentor unit test `assertEqualSpanInstrumentationInfo` calls with `assertEqualSpanInstrumentationScope` calls
   ([#3037](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/3037))
+- `opentelemetry-instrumentation-sqlalchemy` Fixes engines from `sqlalchemy.engine_from_config` not being fully instrumented
+  ([#2816](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/2816))
 - `opentelemetry-instrumentation-sqlalchemy`: Fix a remaining memory leak in EngineTracer
   ([#3053](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/3053))
 

CONTRIBUTING.md

@@ -272,7 +272,11 @@ Below is a checklist of things to be mindful of when implementing a new instrume
 - Isolate sync and async test
   - For synchronous tests, the typical test case class is inherited from `opentelemetry.test.test_base.TestBase`. However, if you want to write asynchronous tests, the test case class should inherit also from `IsolatedAsyncioTestCase`. Adding asynchronous tests to a common test class can lead to tests passing without actually running, which can be misleading.
   - ex. <https://github.com/open-telemetry/opentelemetry-python-contrib/blob/60fb936b7e5371b3e5587074906c49fb873cbd76/instrumentation/opentelemetry-instrumentation-grpc/tests/test_aio_server_interceptor.py#L84>
-- All instrumentations have the same version. If you are going to develop a new instrumentation it would probably have `X.Y.dev` version and depends on `opentelemetry-instrumentation` and `opentelemetry-semantic-conventions` for the same version. That means that if you want to install your instrumentation you need to install its dependencies from this repo and the core repo also from git.
+- Most of the instrumentations have the same version. If you are going to develop a new instrumentation it would probably have `X.Y.dev` version and depends on `opentelemetry-instrumentation` and `opentelemetry-semantic-conventions` for a [compatible version](https://peps.python.org/pep-0440/#compatible-release). That means that you may need to install the instrumentation dependencies from this repo and the core repo from git.
+- Documentation
+  - When adding a new instrumentation remember to add an entry in `docs/instrumentation/` named `<instrumentation>/<instrumentation>.rst` to have the instrumentation documentation referenced from the index. You can use the entry template available [here](./_template/autodoc_entry.rst)
+- Testing
+  - When adding a new instrumentation remember to update `tox.ini` adding appropriate rules in `envlist`, `command_pre` and `commands` sections
 
 ## Guideline for GenAI instrumentations
 

_template/autodoc_entry.rst

@@ -0,0 +1,7 @@
+.. include:: ../../../instrumentation/opentelemetry-instrumentation-<REPLACE ME>/README.rst
+    :end-before: References
+
+.. automodule:: opentelemetry.instrumentation.<REPLACE ME>
+    :members:
+    :undoc-members:
+    :show-inheritance:

docs-requirements.txt

@@ -2,14 +2,6 @@ sphinx==7.1.2
 sphinx-rtd-theme==2.0.0rc4
 sphinx-autodoc-typehints==1.25.2
 
-# Need to install the api/sdk in the venv for autodoc. Modifying sys.path
-# doesn't work for pkg_resources.
--e "git+https://github.com/open-telemetry/opentelemetry-python.git#egg=opentelemetry-api&subdirectory=opentelemetry-api"
--e "git+https://github.com/open-telemetry/opentelemetry-python.git#egg=opentelemetry-semantic-conventions&subdirectory=opentelemetry-semantic-conventions"
--e "git+https://github.com/open-telemetry/opentelemetry-python.git#egg=opentelemetry-sdk&subdirectory=opentelemetry-sdk"
--e "git+https://github.com/open-telemetry/opentelemetry-python-contrib.git#egg=opentelemetry-util-http&subdirectory=util/opentelemetry-util-http"
-./opentelemetry-instrumentation
-
 # Required by opentelemetry-instrumentation
 fastapi>=0.65.2
 pymemcache~=1.3

instrumentation-genai/opentelemetry-instrumentation-openai-v2/CHANGELOG.md

@@ -7,6 +7,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## Unreleased
 
+- Coerce openai response_format to semconv format
+  ([#3073](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/3073))
 - Add example to `opentelemetry-instrumentation-openai-v2`
   ([#3006](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/3006))
 - Support for `AsyncOpenAI/AsyncCompletions` ([#2984](https://github.com/open-telemetry/opentelemetry-python-contrib/pull/2984))

instrumentation-genai/opentelemetry-instrumentation-openai-v2/src/opentelemetry/instrumentation/openai_v2/utils.py

@@ -13,7 +13,7 @@
 # limitations under the License.
 
 from os import environ
-from typing import Optional, Union
+from typing import Mapping, Optional, Union
 from urllib.parse import urlparse
 
 from httpx import URL
@@ -202,12 +202,23 @@ def get_llm_request_attributes(
         GenAIAttributes.GEN_AI_REQUEST_FREQUENCY_PENALTY: kwargs.get(
             "frequency_penalty"
         ),
-        GenAIAttributes.GEN_AI_OPENAI_REQUEST_RESPONSE_FORMAT: kwargs.get(
-            "response_format"
-        ),
         GenAIAttributes.GEN_AI_OPENAI_REQUEST_SEED: kwargs.get("seed"),
     }
 
+    if (response_format := kwargs.get("response_format")) is not None:
+        # response_format may be string or object with a string in the `type` key
+        if isinstance(response_format, Mapping):
+            if (
+                response_format_type := response_format.get("type")
+            ) is not None:
+                attributes[
+                    GenAIAttributes.GEN_AI_OPENAI_REQUEST_RESPONSE_FORMAT
+                ] = response_format_type
+        else:
+            attributes[
+                GenAIAttributes.GEN_AI_OPENAI_REQUEST_RESPONSE_FORMAT
+            ] = response_format
+
     set_server_address_and_port(client_instance, attributes)
     service_tier = kwargs.get("service_tier")
     attributes[GenAIAttributes.GEN_AI_OPENAI_RESPONSE_SERVICE_TIER] = (

instrumentation-genai/opentelemetry-instrumentation-openai-v2/tests/cassettes/test_async_chat_completion_extra_params.yaml

@@ -10,6 +10,9 @@ interactions:
         ],
         "model": "gpt-4o-mini",
         "max_tokens": 50,
+        "response_format": {
+          "type": "text"
+        },
         "seed": 42,
         "stream": false,
         "temperature": 0.5,
@@ -25,7 +28,7 @@ interactions:
       connection:
       - keep-alive
       content-length:
-      - '183'
+      - '220'
       content-type:
       - application/json
       host:
@@ -45,16 +48,16 @@ interactions:
       x-stainless-runtime:
       - CPython
       x-stainless-runtime-version:
-      - 3.12.5
+      - 3.12.7
     method: POST
     uri: https://api.openai.com/v1/chat/completions
   response:
     body:
       string: |-
         {
-          "id": "chatcmpl-ASv9WMTAMZY4O1EImv3csZa6Ch7KI",
+          "id": "chatcmpl-AbMH3rR6OBMN9hG5w0TRrezuiHLMr",
           "object": "chat.completion",
-          "created": 1731456242,
+          "created": 1733467121,
           "model": "gpt-4o-mini-2024-07-18",
           "choices": [
             {
@@ -84,19 +87,19 @@ interactions:
             }
           },
           "service_tier": "default",
-          "system_fingerprint": "fp_0ba0d124f1"
+          "system_fingerprint": "fp_bba3c8e70b"
         }
     headers:
       CF-Cache-Status:
       - DYNAMIC
       CF-RAY:
-      - 8e1a8088f867e167-MRS
+      - 8eda4640ead3e535-KUL
       Connection:
       - keep-alive
       Content-Type:
       - application/json
       Date:
-      - Wed, 13 Nov 2024 00:04:02 GMT
+      - Fri, 06 Dec 2024 06:38:42 GMT
       Server:
       - cloudflare
       Set-Cookie: test_set_cookie
@@ -112,25 +115,25 @@ interactions:
       - '825'
       openai-organization: test_openai_org_id
       openai-processing-ms:
-      - '488'
+      - '835'
       openai-version:
       - '2020-10-01'
       strict-transport-security:
       - max-age=31536000; includeSubDomains; preload
       x-ratelimit-limit-requests:
-      - '30000'
+      - '10000'
       x-ratelimit-limit-tokens:
-      - '150000000'
+      - '200000'
       x-ratelimit-remaining-requests:
-      - '29999'
+      - '9999'
       x-ratelimit-remaining-tokens:
-      - '149999943'
+      - '199943'
       x-ratelimit-reset-requests:
-      - 2ms
+      - 8.64s
       x-ratelimit-reset-tokens:
-      - 0s
+      - 16ms
       x-request-id:
-      - req_6df08d6267415e8f5db3628a6757edad
+      - req_fea877c0a861ff92a6a5217247681f24
     status:
       code: 200
       message: OK

instrumentation-genai/opentelemetry-instrumentation-openai-v2/tests/cassettes/test_chat_completion_extra_params.yaml

@@ -10,6 +10,9 @@ interactions:
         ],
         "model": "gpt-4o-mini",
         "max_tokens": 50,
+        "response_format": {
+          "type": "text"
+        },
         "seed": 42,
         "stream": false,
         "temperature": 0.5,
@@ -25,13 +28,13 @@ interactions:
       connection:
       - keep-alive
       content-length:
-      - '183'
+      - '220'
       content-type:
       - application/json
       host:
       - api.openai.com
       user-agent:
-      - OpenAI/Python 1.54.3
+      - OpenAI/Python 1.26.0
       x-stainless-arch:
       - arm64
       x-stainless-async:
@@ -41,22 +44,20 @@ interactions:
       x-stainless-os:
       - MacOS
       x-stainless-package-version:
-      - 1.54.3
-      x-stainless-retry-count:
-      - '0'
+      - 1.26.0
       x-stainless-runtime:
       - CPython
       x-stainless-runtime-version:
-      - 3.12.6
+      - 3.12.7
     method: POST
     uri: https://api.openai.com/v1/chat/completions
   response:
     body:
       string: |-
         {
-          "id": "chatcmpl-ASYMT7913Sp58qhZqQgY7g7Ia2J4M",
+          "id": "chatcmpl-AbMH70fQA9lMPIClvBPyBSjqJBm9F",
           "object": "chat.completion",
-          "created": 1731368633,
+          "created": 1733467125,
           "model": "gpt-4o-mini-2024-07-18",
           "choices": [
             {
@@ -86,19 +87,17 @@ interactions:
             }
           },
           "service_tier": "default",
-          "system_fingerprint": "fp_0ba0d124f1"
+          "system_fingerprint": "fp_0705bf87c0"
         }
     headers:
-      CF-Cache-Status:
-      - DYNAMIC
       CF-RAY:
-      - 8e1225a3f8e9ce65-SIN
+      - 8eda465e8fe9e58c-KUL
       Connection:
       - keep-alive
       Content-Type:
       - application/json
       Date:
-      - Mon, 11 Nov 2024 23:43:53 GMT
+      - Fri, 06 Dec 2024 06:38:46 GMT
       Server:
       - cloudflare
       Set-Cookie: test_set_cookie
@@ -110,11 +109,13 @@ interactions:
       - X-Request-ID
       alt-svc:
       - h3=":443"; ma=86400
+      cf-cache-status:
+      - DYNAMIC
       content-length:
       - '825'
       openai-organization: test_openai_org_id
       openai-processing-ms:
-      - '431'
+      - '558'
       openai-version:
       - '2020-10-01'
       strict-transport-security:
@@ -128,11 +129,11 @@ interactions:
       x-ratelimit-remaining-tokens:
       - '199943'
       x-ratelimit-reset-requests:
-      - 14.746s
+      - 12.967s
       x-ratelimit-reset-tokens:
       - 16ms
       x-request-id:
-      - req_81e29a8992ea8001c0240bd990acf0ab
+      - req_22ff608d47a299f0780f52360631eabb
     status:
       code: 200
       message: OK

instrumentation-genai/opentelemetry-instrumentation-openai-v2/tests/test_async_chat_completions.py

@@ -158,6 +158,7 @@ async def test_async_chat_completion_extra_params(
         max_tokens=50,
         stream=False,
         extra_body={"service_tier": "default"},
+        response_format={"type": "text"},
     )
 
     spans = span_exporter.get_finished_spans()
@@ -173,6 +174,12 @@ async def test_async_chat_completion_extra_params(
         spans[0].attributes[GenAIAttributes.GEN_AI_OPENAI_REQUEST_SERVICE_TIER]
         == "default"
     )
+    assert (
+        spans[0].attributes[
+            GenAIAttributes.GEN_AI_OPENAI_REQUEST_RESPONSE_FORMAT
+        ]
+        == "text"
+    )
 
 
 @pytest.mark.vcr()