Add TorchData as hard dependency (#1985)

* Add TorchData as hard dependency

* Add install_torchdata.sh as pre-script

* Add conda constraint

* Fix tests

Author: ejguan

.circleci/config.yml

@@ -48,6 +48,10 @@ binary_common: &binary_common
       description: "PyTorch version to build against; by default, use a nightly"
       type: string
       default: ""
+    torchdata_version:
+      description: "TorchData version to build against; by default, use a nightly"
+      type: string
+      default: ""
     # Don't edit these
     python_version:
       description: "Python version to build against (e.g., 3.8)"
@@ -56,6 +60,7 @@ binary_common: &binary_common
     PYTHON_VERSION: << parameters.python_version >>
     BUILD_VERSION: << parameters.build_version >>
     PYTORCH_VERSION: << parameters.pytorch_version >>
+    TORCHDATA_VERSION: << parameters.torchdata_version >>
     CU_VERSION: cpu
     MACOSX_DEPLOYMENT_TARGET: 10.9
 

.circleci/config.yml.in

@@ -1,6 +1,7 @@
 #!/usr/bin/env bash
 
 unset PYTORCH_VERSION
+unset TORCHDATA_VERSION
 # For unittest, nightly PyTorch is used as the following section,
 # so no need to set PYTORCH_VERSION.
 # In fact, keeping PYTORCH_VERSION forces us to hardcode PyTorch version in config.

.circleci/unittest/linux/scripts/install.sh

@@ -1,6 +1,7 @@
 #!/usr/bin/env bash
 
 unset PYTORCH_VERSION
+unset TORCHDATA_VERSION
 # For unittest, nightly PyTorch is used as the following section,
 # so no need to set PYTORCH_VERSION.
 # In fact, keeping PYTORCH_VERSION forces us to hardcode PyTorch version in config.

.circleci/unittest/windows/scripts/install.sh

@@ -22,7 +22,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: ""
+            pre-script: packaging/install_torchdata.sh
             post-script: ""
             conda-package-directory: packaging/torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/build-conda-linux.yml

@@ -22,7 +22,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: ""
+            pre-script: packaging/install_torchdata.sh
             post-script: ""
             conda-package-directory: packaging/torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/build-conda-m1.yml

@@ -22,7 +22,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: ""
+            pre-script: packaging/install_torchdata.sh
             post-script: ""
             conda-package-directory: packaging/torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/build-conda-macos.yml

@@ -22,7 +22,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: ""
+            pre-script: packaging/install_torchdata.sh
             post-script: ""
             smoke-test-script: test/smoke_tests/smoke_tests.py
             package-name: torchtext

.github/workflows/build-wheels-linux.yml

@@ -22,7 +22,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: ""
+            pre-script: packaging/install_torchdata.sh
             post-script: ""
             package-name: torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py

.github/workflows/build-wheels-m1.yml

@@ -22,7 +22,7 @@ jobs:
       matrix:
         include:
           - repository: pytorch/text
-            pre-script: ""
+            pre-script: packaging/install_torchdata.sh
             post-script: ""
             package-name: torchtext
             smoke-test-script: test/smoke_tests/smoke_tests.py