From d08aa7c5d8f2e5c3cb6978a9c30d27b96898e3db Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fran=C3=A7ois=20S=C3=A9guin?=
 <162352622+francois-pass-culture@users.noreply.github.com>
Date: Tue, 11 Feb 2025 10:50:22 +0100
Subject: [PATCH] (PC-34429)[API] ci: use cached pytest duration file

---
 .../workflows/dev_on_workflow_tests_api.yml   | 125 +++++++++++++++++-
 1 file changed, 124 insertions(+), 1 deletion(-)

diff --git a/.github/workflows/dev_on_workflow_tests_api.yml b/.github/workflows/dev_on_workflow_tests_api.yml
index 169faeb91b4..e8047658ac7 100644
--- a/.github/workflows/dev_on_workflow_tests_api.yml
+++ b/.github/workflows/dev_on_workflow_tests_api.yml
@@ -393,6 +393,53 @@ jobs:
         env:
           SLACK_BOT_TOKEN: ${{ steps.secrets.outputs.SLACK_BOT_TOKEN }}
 
+  restore-test-durations:
+    name: Restore pytest durations
+    runs-on: ubuntu-22.04
+    steps:
+#      # http://man7.org/linux/man-pages/man1/date.1.html
+#      - name: Get Date
+#        id: get-date
+#        run: |
+#          echo "date=$(/bin/date -u "+%Y%m%d")" >> $GITHUB_OUTPUT
+#        shell: bash
+#     Then we'll use tests-durations-${{ steps.get-date.outputs.date }}
+
+    # It's mandatory to use the exact same path when saving/restoring cache, otherwise it won't work
+    # (the same key is not enough - see documentation:
+    # https://github.com/actions/cache/blob/main/README.md#cache-version).
+    # I went with `/tests/.test_durations`.
+      - name: Restore test durations
+        id: restore-test-durations
+        uses: actions/cache/restore@v4
+        with:
+          path: /tests/.test_durations
+          key: tests-durations-${{ github.sha }}
+          restore-keys: |
+            tests-durations-${{ github.sha }}-
+            tests-durations-
+          fail-on-cache-miss: false
+
+      # Then we upload the restored test durations as an artifact. This way, each matrix job will download
+      # it when it starts. When a matrix job will be manually retried, it will also reuse the artifact (to
+      # retry the exact same tests, even if the cache has been updated in the meantime).
+      - name: Upload test durations
+        if: steps.restore-test-durations.outputs.cache-hit != ''
+        uses: actions/upload-artifact@v4
+        with:
+          name: test-durations-before
+          path: /tests/.test_durations
+
+    outputs:
+      # This output will be used to know if we had a cache hit (exact match or not), or no cache hit at all.
+      # See documentation about the `cache-hit` output:
+      # https://github.com/actions/cache/blob/main/README.md#outputs
+      # > cache-hit - A string value to indicate an exact match was found for the key.
+      # > If there's a cache hit, this will be 'true' or 'false' to indicate if there's an exact match
+      # > for key.
+      # > If there's a cache miss, this will be an empty string.
+      restored: ${{ steps.restore-test-durations.outputs.cache-hit == '' && 'false' || 'true' }}
+
   pytest:
     name: "Pytest ${{ matrix.pytest_args.description }} : ${{ matrix.group }}/4"
     env:
@@ -490,6 +537,7 @@ jobs:
             alembic upgrade pre@head
             alembic upgrade post@head
             flask install_data
+
       - name: "Mount a Volume with pcapi rights"
         uses: addnab/docker-run-action@v3
         with:
@@ -499,6 +547,25 @@ jobs:
           run: |
             echo "Changing owner and group fort directory test"
             chown -R pcapi:pcapi /tests
+
+      # These two steps will be executed only when there IS a cache hit (exact match or not). When a matrix
+      # job is retried, it will reuse the same artifact, to execute the exact same split.
+      - name: Download test durations
+        if: needs.restore-test-durations.outputs.restored == 'true'
+        uses: actions/download-artifact@v4
+        with:
+          name: test-durations-before
+
+      - name: Use cached test durations
+        if: needs.restore-test-durations.outputs.restored == 'true'
+        run: mv .test_durations ${{ runner.workspace }}/pass-culture-main/api/tests/.test_durations
+
+        # When running pytest, we write the new test durations using options
+        # `--store-durations --clean-durations`.
+        # Option `--clean-durations` is undocumented but you can check its implementation here:
+        # https://github.com/jerry-git/pytest-split/blob/fb9af7e0122c18a96a7c01ca734c4ab01027f8d9/src/pytest_split/plugin.py#L68-L76
+        # > Removes the test duration info for tests which are not present while running the suite with
+        # > '--store-durations'.
       - name: "Run pytest"
         uses: addnab/docker-run-action@v3
         with:
@@ -506,7 +573,22 @@ jobs:
           shell: bash
           options: -e RUN_ENV -e DATABASE_URL_TEST -e REDIS_URL -e SQLALCHEMY_WARN_20 -v ${{ runner.workspace }}/pass-culture-main/api/tests/:/tests
           run: |
-            pytest ${{ matrix.pytest_args.collection }} --splits 4 --group ${{ matrix.group }} --durations=10 --junitxml='/tests/junit.xml' -q --color=yes
+            pytest ${{ matrix.pytest_args.collection }} \
+            --splits 4 --group ${{ matrix.group }} \
+            --store-durations --durations-path /tests/.test_durations --clean-durations \
+            --durations=10 --junitxml='/tests/junit.xml' \
+            -q --color=yes
+
+      - name: Upload test durations
+#        if: github.run_attempt == 1
+        if: always()
+        uses: actions/upload-artifact@v4
+        with:
+          name: "test-durations-after-partial-${{ matrix.pytest_args.description }}-${{ matrix.group }}"
+          path: "${{ runner.workspace }}/pass-culture-main/api/tests/.test_durations"
+          if-no-files-found: warn
+          include-hidden-files: true
+
       - name: "Publish Test Report"
         uses: mikepenz/action-junit-report@v5
         if: always() # always run even if the previous step fails
@@ -540,3 +622,44 @@ jobs:
             }
         env:
           SLACK_BOT_TOKEN: ${{ steps.secrets.outputs.SLACK_BOT_TOKEN }}
+
+  cache-test-durations:
+    name: Cache test durations
+    needs: pytest
+    if: github.run_attempt == 1 && (success() || failure())
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Download all partial test durations
+        uses: actions/download-artifact@v4
+        with:
+          pattern: test-durations-after-partial-*
+          path: ${{ runner.temp }}
+
+      # This step regroups the 8 partial files and sorts keys alphabetically:
+      - name: Merge all partial test durations
+        # overrides workflow default
+        working-directory: /
+        run: |
+          jq -s 'add' ${{ runner.temp }}/test-durations-after-partial-*/.test_durations \
+          | jq 'to_entries | sort_by(.key) | from_entries' \
+          > ${{ runner.temp }}/.test_durations
+
+      # This step uploads the final file as an artifact. You can then download it from the Github GUI,
+      # and use it to manually commit file `.test_durations_fallback` from time to time,
+      # to keep an up-to-date fallback:
+      - name: Upload final test durations
+        uses: actions/upload-artifact@v4
+        with:
+          name: test-durations-after
+          path: ${{ runner.temp }}/.test_durations
+          if-no-files-found: warn
+          include-hidden-files: true
+
+      # Finally, we cache the new test durations. This file will be restored in next CI execution
+      # (see step "Restore test durations" above).
+      - name: Cache final test durations
+        uses: actions/cache/save@v4
+        with:
+          path: ${{ runner.temp }}/.test_durations
+          key: tests-durations-${{ github.sha }}
+#          key: tests-durations-${{ steps.get-date.outputs.date }}}-