Skip to content

Added a workflow to parallelise the E2E tests #8

Added a workflow to parallelise the E2E tests

Added a workflow to parallelise the E2E tests #8

name: Parallel Code Coverage
permissions:
contents: read
on: [pull_request, workflow_dispatch]
jobs:
discover-tests:
runs-on: ubuntu-latest
outputs:
test-files: ${{ steps.discover.outputs.test-files }}
steps:
- name: Check out repository
uses: actions/checkout@v4
with:
fetch-depth: 0
ref: ${{ github.event.pull_request.head.ref || github.ref_name }}
repository: ${{ github.event.pull_request.head.repo.full_name || github.repository }}
- name: Discover test files
id: discover
run: |
# Find all test files in e2e directory and create JSON array (excluding test_driver.py and test_parameterized_queries.py for now)
TEST_FILES=$(find tests/e2e -name "test_*.py" -type f | grep -v -E "(test_driver\.py|test_parameterized_queries\.py)" | sort | jq -R -s -c 'split("\n")[:-1]')
echo "test-files=$TEST_FILES" >> $GITHUB_OUTPUT
echo "Discovered test files: $TEST_FILES"
e2e-tests:
runs-on: ubuntu-latest
environment: azure-prod
needs: discover-tests
strategy:
matrix:
test_file: ${{ fromJson(needs.discover-tests.outputs.test-files) }}
mode: ["thrift", "sea"]
env:
DATABRICKS_SERVER_HOSTNAME: ${{ secrets.DATABRICKS_HOST }}
DATABRICKS_HTTP_PATH: ${{ secrets.TEST_PECO_WAREHOUSE_HTTP_PATH }}
DATABRICKS_TOKEN: ${{ secrets.DATABRICKS_TOKEN }}
DATABRICKS_CATALOG: peco
DATABRICKS_USER: ${{ secrets.TEST_PECO_SP_ID }}
steps:
- name: Check out repository
uses: actions/checkout@v4
with:
fetch-depth: 0
ref: ${{ github.event.pull_request.head.ref || github.ref_name }}
repository: ${{ github.event.pull_request.head.repo.full_name || github.repository }}
- name: Set up python
id: setup-python
uses: actions/setup-python@v5
with:
python-version: "3.10"
- name: Install Poetry
uses: snok/install-poetry@v1
with:
virtualenvs-create: true
virtualenvs-in-project: true
installer-parallel: true
- name: Load cached venv
id: cached-poetry-dependencies
uses: actions/cache@v4
with:
path: .venv
key: venv-${{ runner.os }}-${{ steps.setup-python.outputs.python-version }}-${{ github.event.repository.name }}-${{ hashFiles('**/poetry.lock') }}
- name: Install dependencies
if: steps.cached-poetry-dependencies.outputs.cache-hit != 'true'
run: poetry install --no-interaction --no-root
- name: Install library
run: poetry install --no-interaction --all-extras
- name: Run ${{ matrix.mode }} tests for ${{ matrix.test_file }}
run: |
echo "Running ${{ matrix.mode }} tests for ${{ matrix.test_file }}"
if [ "${{ matrix.mode }}" = "sea" ]; then
TEST_FILTER="-k"
TEST_EXPRESSION="extra_params1 or extra_params2"
else
TEST_FILTER="-k"
TEST_EXPRESSION="extra_params0 or not extra_params"
fi
TEST_NAME=$(basename "${{ matrix.test_file }}" .py)
COVERAGE_FILE="coverage-${TEST_NAME}-${{ matrix.mode }}.xml"
COVERAGE_DATA=".coverage-${TEST_NAME}-${{ matrix.mode }}"
echo "TEST_NAME=$TEST_NAME" >> $GITHUB_ENV
# Clean any previous coverage data to avoid conflicts
rm -f .coverage*
echo "πŸ§ͺ Running pytest with coverage..."
poetry run pytest "${{ matrix.test_file }}" "$TEST_FILTER" "$TEST_EXPRESSION" \
--cov=src --cov-report=xml:$COVERAGE_FILE --cov-report=term -v || [ $? -eq 5 ]
echo "πŸ” DEBUG: Post-test file analysis..."
echo "Current directory contents:"
ls -la
echo "Coverage-related files:"
ls -la .coverage* coverage-* 2>/dev/null || echo "No coverage files found"
if [ -f ".coverage" ]; then
COVERAGE_SIZE=$(stat -c%s .coverage 2>/dev/null || stat -f%z .coverage)
echo "βœ… .coverage file found (${COVERAGE_SIZE} bytes)"
mv .coverage "$COVERAGE_DATA"
echo "βœ… Saved coverage data as $COVERAGE_DATA"
else
echo "⚠️ No .coverage generated, creating minimal coverage file"
# Create a minimal but valid coverage file using coverage.py
poetry run coverage erase
poetry run coverage run --source=src -m pytest --version > /dev/null 2>&1 || true
if [ -f ".coverage" ]; then
mv .coverage "$COVERAGE_DATA"
echo "βœ… Created minimal coverage file as $COVERAGE_DATA"
else
# Fallback: create empty file (will be handled gracefully in merge)
touch "$COVERAGE_DATA"
echo "⚠️ Created empty placeholder as $COVERAGE_DATA"
fi
fi
if [ -f "$COVERAGE_FILE" ]; then
XML_SIZE=$(stat -c%s "$COVERAGE_FILE" 2>/dev/null || stat -f%z "$COVERAGE_FILE")
echo "βœ… XML coverage file found: $COVERAGE_FILE (${XML_SIZE} bytes)"
echo "XML preview:"
head -3 "$COVERAGE_FILE" 2>/dev/null || echo "Cannot read XML file"
else
echo "❌ No XML coverage file generated: $COVERAGE_FILE"
fi
echo "πŸ“ Final files available for upload:"
ls -la .coverage* coverage-* 2>/dev/null || echo "No coverage files found"
echo "πŸ” Specifically checking for files to upload:"
echo " .coverage-* files:"
ls -la .coverage-* 2>/dev/null || echo " None found"
echo " coverage-*-${{ matrix.mode }}.xml files:"
ls -la coverage-*-${{ matrix.mode }}.xml 2>/dev/null || echo " None found"
echo "πŸ” CRITICAL DEBUG: Checking exact upload patterns:"
echo " Pattern '.coverage-*' matches:"
find . -maxdepth 1 -name ".coverage-*" -type f 2>/dev/null || echo " No matches"
echo " Pattern 'coverage-*-${{ matrix.mode }}.xml' matches:"
find . -maxdepth 1 -name "coverage-*-${{ matrix.mode }}.xml" -type f 2>/dev/null || echo " No matches"
echo "πŸ” Working directory: $(pwd)"
echo "πŸ” All files in current directory:"
ls -la
- name: Upload coverage artifact
uses: actions/upload-artifact@v4
if: always()
with:
name: coverage-${{ env.TEST_NAME }}-${{ matrix.mode }}
path: |
.coverage-*
coverage-*-${{ matrix.mode }}.xml
if-no-files-found: warn
merge-coverage:
runs-on: ubuntu-latest
needs: [e2e-tests]
steps:
- name: Check out repository
uses: actions/checkout@v4
- name: Set up python
id: setup-python
uses: actions/setup-python@v5
with:
python-version: "3.10"
- name: Install Poetry
uses: snok/install-poetry@v1
with:
virtualenvs-create: true
virtualenvs-in-project: true
installer-parallel: true
- name: Load cached venv
id: cached-poetry-dependencies
uses: actions/cache@v4
with:
path: .venv
key: venv-${{ runner.os }}-${{ steps.setup-python.outputs.python-version }}-${{ github.event.repository.name }}-${{ hashFiles('**/poetry.lock') }}
- name: Install dependencies
if: steps.cached-poetry-dependencies.outputs.cache-hit != 'true'
run: poetry install --no-interaction --no-root
- name: Install library
run: poetry install --no-interaction --all-extras
- name: Download all coverage artifacts
uses: actions/download-artifact@v4
with:
path: coverage_files
- name: Merge coverage
run: |
echo "πŸ”§ Installing xmllint..."
if ! command -v xmllint &> /dev/null; then
sudo apt-get update && sudo apt-get install -y libxml2-utils
fi
echo "πŸ“ DEBUG: Checking downloaded artifacts structure..."
find coverage_files -type f -name "*" | head -20
echo "Total files found: $(find coverage_files -type f | wc -l)"
echo "πŸ“ Checking for coverage files in artifacts..."
COVERAGE_FILES_FOUND=0
for artifact_dir in coverage_files/*/; do
if [ -d "$artifact_dir" ]; then
echo "πŸ” Artifact: $(basename "$artifact_dir")"
echo " Contents:"
ls -la "$artifact_dir" || echo " (empty or inaccessible)"
# Copy .coverage files
for cov_file in "$artifact_dir"/.coverage-*; do
if [ -f "$cov_file" ]; then
cp "$cov_file" .
COVERAGE_FILES_FOUND=$((COVERAGE_FILES_FOUND + 1))
echo " βœ… Copied $(basename "$cov_file") ($(stat -c%s "$cov_file" 2>/dev/null || stat -f%z "$cov_file") bytes)"
fi
done
# Copy XML files for debugging
for xml_file in "$artifact_dir"/coverage-*.xml; do
if [ -f "$xml_file" ]; then
cp "$xml_file" .
echo " πŸ“„ Copied $(basename "$xml_file") ($(stat -c%s "$xml_file" 2>/dev/null || stat -f%z "$xml_file") bytes)"
fi
done
fi
done
echo "πŸ“Š SUMMARY: Found $COVERAGE_FILES_FOUND .coverage files"
echo "Available .coverage files for merging:"
ls -la .coverage-* 2>/dev/null || echo "❌ No .coverage-* files found"
echo "Available XML files:"
ls -la coverage-*.xml 2>/dev/null || echo "❌ No XML files found"
if [ $COVERAGE_FILES_FOUND -gt 0 ]; then
echo "πŸ”„ Combining coverage data..."
poetry run coverage combine .coverage-* || {
echo "❌ Coverage combine failed, checking individual files:"
for f in .coverage-*; do
if [ -f "$f" ]; then
echo " File: $f ($(stat -c%s "$f" 2>/dev/null || stat -f%z "$f") bytes)"
file "$f" 2>/dev/null || echo " Cannot determine file type"
fi
done
echo "⚠️ Using fallback approach..."
touch .coverage
}
echo "πŸ“Š Generating XML report..."
poetry run coverage xml || {
echo "❌ XML generation failed, using fallback"
echo '<coverage lines-covered="0" lines-valid="1"></coverage>' > coverage.xml
}
echo "πŸ“‹ Generating text report..."
poetry run coverage report || echo "⚠️ Text report failed"
else
echo "⚠️ No coverage data files found, creating empty report"
echo '<coverage lines-covered="0" lines-valid="1"></coverage>' > coverage.xml
fi
echo "πŸ“„ Final coverage.xml preview:"
head -5 coverage.xml 2>/dev/null || echo "Cannot read coverage.xml"
- name: Report coverage percentage
run: |
COVERAGE_FILE="coverage.xml"
if [ ! -f "$COVERAGE_FILE" ]; then
echo "ERROR: Coverage file not found at $COVERAGE_FILE"
exit 1
fi
COVERED=$(xmllint --xpath "string(//coverage/@lines-covered)" "$COVERAGE_FILE")
TOTAL=$(xmllint --xpath "string(//coverage/@lines-valid)" "$COVERAGE_FILE")
PERCENTAGE=$(python3 -c "covered=${COVERED}; total=${TOTAL}; print(round((covered/total)*100, 2))")
echo "πŸ“Š Combined Coverage: ${PERCENTAGE}%"