From 056da1af8a856d97746f5119c15b51916e97cc14 Mon Sep 17 00:00:00 2001
From: laughingman7743 <laughingman7743@gmail.com>
Date: Thu, 1 Oct 2026 09:25:53 +0900
Subject: [PATCH] Backport #906: Run SQLAlchemy and Spark tests when shared
 core modules change

The Test workflow's changes job now also selects the SQLAlchemy and Spark
tests when a pull request changes a top-level module of pyathena or
pyathena.aio, or a shared test fixture (the tests and tests/pyathena(/aio)
package initializers and conftest files, tests/pyathena/tables.py,
tests/pyathena/util.py, tests/resources/). Changes limited to other
subpackages still skip them. The selection now uses dorny/paths-filter
v4.0.3 (SHA-pinned) instead of gh api and grep; other events still run
every suite.

(cherry picked from commit 9b9421d8464fec020841c2ebda374caac8d6aad6)

The merge commit applies cleanly; the added and removed lines match #906
exactly. 3.x has no tests/pyathena/tables.py, so that pattern matches
nothing there.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---
 .github/workflows/test.yaml | 80 ++++++++++++++++++++++---------------
 1 file changed, 48 insertions(+), 32 deletions(-)

diff --git a/.github/workflows/test.yaml b/.github/workflows/test.yaml
index 4f2cf48a..d293b133 100644
--- a/.github/workflows/test.yaml
+++ b/.github/workflows/test.yaml
@@ -61,9 +61,10 @@ jobs:
   # requests run none. A ready pull request always runs the PyAthena suite; it
   # runs the SQLAlchemy tests (the compliance suites and the PyAthena suite's
   # SQLAlchemy tests) and the Spark tests only when their code, tests,
-  # dependencies, or this workflow change. Pull requests and the schedule test
-  # the newest Python version; a dispatch tests the requested versions or
-  # every version, and the Release workflow every version.
+  # dependencies, this workflow, the top-level modules of pyathena and
+  # pyathena.aio, or the shared test fixtures change. Pull requests and the
+  # schedule test the newest Python version; a dispatch tests the requested
+  # versions or every version, and the Release workflow every version.
   changes:
     if: >-
       github.event_name != 'pull_request' ||
@@ -73,16 +74,53 @@ jobs:
     permissions:
       pull-requests: read
     outputs:
-      sqla: ${{ steps.filter.outputs.sqla }}
-      spark: ${{ steps.filter.outputs.spark }}
-      python-versions: ${{ steps.filter.outputs.python-versions }}
+      sqla: ${{ github.event_name != 'pull_request' || steps.paths.outputs.sqla == 'true' }}
+      spark: ${{ github.event_name != 'pull_request' || steps.paths.outputs.spark == 'true' }}
+      python-versions: ${{ steps.versions.outputs.python-versions }}
     steps:
-      - id: filter
+      - id: paths
+        if: github.event_name == 'pull_request'
+        uses: dorny/paths-filter@ceb8a2b8f2d89434be7ff52d3de7ec3738c5cc9d # v4.0.3
+        with:
+          filters: |
+            shared: &shared
+              - .github/workflows/test.yaml
+              - .github/workflows/test-suite.yaml
+              - justfile
+              - pyproject.toml
+              - uv.lock
+            # The SQLAlchemy and Spark packages load the top-level modules,
+            # directly or transitively, and their tests use the shared fixtures.
+            core: &core
+              - pyathena/*.py
+              - pyathena/aio/*.py
+              - tests/__init__.py
+              - tests/pyathena/__init__.py
+              - tests/pyathena/conftest.py
+              - tests/pyathena/tables.py
+              - tests/pyathena/util.py
+              - tests/pyathena/aio/__init__.py
+              - tests/pyathena/aio/conftest.py
+              - tests/resources/**
+            sqla:
+              - *shared
+              - *core
+              - setup.cfg
+              - pyathena/sqlalchemy/**
+              - pyathena/aio/sqlalchemy/**
+              - tests/sqlalchemy/**
+              - tests/pyathena/sqlalchemy/**
+              - tests/pyathena/aio/sqlalchemy/**
+            spark:
+              - *shared
+              - *core
+              - pyathena/spark/**
+              - pyathena/aio/spark/**
+              - tests/pyathena/spark/**
+              - tests/pyathena/aio/spark/**
+      - id: versions
         env:
-          GH_TOKEN: ${{ github.token }}
           EVENT_NAME: ${{ github.event_name }}
-          REPO: ${{ github.repository }}
-          PR_NUMBER: ${{ github.event.pull_request.number }}
           REQUESTED_VERSIONS: ${{ inputs.python-versions }}
           # Every supported version, oldest first; keep in sync with the
           # pyproject.toml classifiers.
@@ -106,28 +144,6 @@ jobs:
               ;;
           esac
           echo "python-versions=$versions" >> "$GITHUB_OUTPUT"
-          if [[ "$EVENT_NAME" != "pull_request" ]]; then
-            {
-              echo "sqla=true"
-              echo "spark=true"
-            } >> "$GITHUB_OUTPUT"
-            exit 0
-          fi
-          files=$(gh api "repos/$REPO/pulls/$PR_NUMBER/files" --paginate --jq '.[].filename')
-          printf 'Changed files:\n%s\n' "$files"
-          shared='^(\.github/workflows/test(-suite)?\.yaml|justfile|pyproject\.toml|uv\.lock)$'
-          sqla="$shared|^pyathena/(aio/)?sqlalchemy/|^tests/sqlalchemy/|^tests/pyathena/(aio/)?sqlalchemy/|^setup\.cfg$"
-          spark="$shared|^pyathena/(aio/)?spark/|^tests/pyathena/(aio/)?spark/"
-          if grep -qE "$sqla" <<< "$files"; then
-            echo "sqla=true" >> "$GITHUB_OUTPUT"
-          else
-            echo "sqla=false" >> "$GITHUB_OUTPUT"
-          fi
-          if grep -qE "$spark" <<< "$files"; then
-            echo "spark=true" >> "$GITHUB_OUTPUT"
-          else
-            echo "spark=false" >> "$GITHUB_OUTPUT"
-          fi
 
   # The three suites create their own schemas and tables, so they run in
   # parallel; each is still a separate job for "Re-run failed jobs".