chore: updated and linted #430
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: CI Python | |
| on: | |
| push: | |
| branches: [main] | |
| paths: | |
| - '.github/actions/**' | |
| - '.github/workflows/ci-python.yaml' | |
| - '.cargo/config.toml' | |
| - 'rust-toolchain.toml' | |
| - 'Cargo.toml' | |
| - 'Cargo.lock' | |
| - 'crates/kreuzberg/**' | |
| - 'crates/kreuzberg-py/**' | |
| - 'crates/kreuzberg-tesseract/**' | |
| - 'packages/python/**' | |
| - 'e2e/python/**' | |
| - 'test_documents/**' | |
| - 'fixtures/**' | |
| - 'tools/e2e-generator/**' | |
| - 'uv.lock' | |
| - 'pyproject.toml' | |
| - 'scripts/ci/python/**' | |
| - 'scripts/ci/cache/**' | |
| - 'scripts/ci/actions/**' | |
| pull_request: | |
| branches: [main] | |
| paths: | |
| - '.github/actions/**' | |
| - '.github/workflows/ci-python.yaml' | |
| - '.cargo/config.toml' | |
| - 'rust-toolchain.toml' | |
| - 'Cargo.toml' | |
| - 'Cargo.lock' | |
| - 'crates/kreuzberg/**' | |
| - 'crates/kreuzberg-py/**' | |
| - 'crates/kreuzberg-tesseract/**' | |
| - 'packages/python/**' | |
| - 'e2e/python/**' | |
| - 'test_documents/**' | |
| - 'fixtures/**' | |
| - 'tools/e2e-generator/**' | |
| - 'uv.lock' | |
| - 'pyproject.toml' | |
| - 'scripts/ci/python/**' | |
| - 'scripts/ci/cache/**' | |
| - 'scripts/ci/actions/**' | |
| concurrency: | |
| group: ci-python-${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} | |
| cancel-in-progress: true | |
| env: | |
| CARGO_TERM_COLOR: always | |
| CARGO_INCREMENTAL: 0 | |
| CARGO_PROFILE_DEV_DEBUG: 0 | |
| RUST_BACKTRACE: short | |
| RUST_MIN_STACK: 16777216 | |
| PDFIUM_VERSION: "7578" | |
| PDFIUM_STATIC_VERSION: "7442b" | |
| ORT_VERSION: "1.23.2" | |
| MACOSX_DEPLOYMENT_TARGET: "14.0" | |
| BUILD_PROFILE: "ci" | |
| jobs: | |
| build-and-smoke-python: | |
| name: Python Build + Smoke (${{ matrix.target }}) | |
| if: ${{ github.actor != 'dependabot[bot]' }} | |
| timeout-minutes: 180 | |
| strategy: | |
| fail-fast: true | |
| matrix: | |
| include: | |
| - os: ubuntu-latest | |
| target: x86_64-unknown-linux-gnu | |
| platform: linux-x86_64 | |
| - os: ubuntu-24.04-arm | |
| target: aarch64-unknown-linux-gnu | |
| platform: linux-arm64 | |
| - os: macos-latest | |
| target: aarch64-apple-darwin | |
| platform: macos-arm64 | |
| - os: windows-latest | |
| target: x86_64-pc-windows-msvc | |
| platform: windows-x86_64 | |
| runs-on: ${{ matrix.os }} | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Install system dependencies | |
| uses: ./.github/actions/install-system-deps | |
| - name: Free disk space before setup | |
| if: startsWith(matrix.os, 'ubuntu') | |
| uses: ./.github/actions/free-disk-space-linux | |
| with: | |
| show-initial: "true" | |
| show-final: "true" | |
| - name: Setup OpenSSL | |
| uses: ./.github/actions/setup-openssl | |
| - name: Setup Rust | |
| uses: ./.github/actions/setup-rust | |
| with: | |
| cache-key-prefix: python-${{ matrix.platform }} | |
| target: ${{ matrix.target }} | |
| - name: Setup Python | |
| uses: ./.github/actions/setup-python-env | |
| with: | |
| python-version: "3.13" | |
| cache-prefix: wheels-${{ matrix.platform }} | |
| - name: Cache PDFium | |
| uses: ./.github/actions/cache-pdfium | |
| with: | |
| pdfium-version: ${{ env.PDFIUM_VERSION }} | |
| - name: Download PDFium | |
| uses: ./.github/actions/download-pdfium | |
| with: | |
| pdfium-version: ${{ env.PDFIUM_VERSION }} | |
| - name: Stage PDFium runtime | |
| uses: ./.github/actions/stage-pdfium-runtime | |
| with: | |
| destination: target/release | |
| - name: Setup ONNX Runtime | |
| uses: ./.github/actions/setup-onnx-runtime | |
| with: | |
| ort-version: ${{ env.ORT_VERSION }} | |
| - name: Clean previous wheel artifacts | |
| shell: bash | |
| run: scripts/ci/python/clean-artifacts.sh | |
| - name: Install Task | |
| uses: ./.github/actions/install-task | |
| - name: Build CLI with features | |
| shell: bash | |
| run: | | |
| echo "=== Building CLI Binary with Features ===" | |
| cargo build --release --package kreuzberg-cli --features all | |
| mkdir -p packages/python/kreuzberg | |
| if [ "${{ runner.os }}" = "Windows" ]; then | |
| cp target/release/kreuzberg.exe packages/python/kreuzberg/ | |
| else | |
| cp target/release/kreuzberg packages/python/kreuzberg/ | |
| fi | |
| - name: Build FFI library | |
| id: ffi | |
| shell: bash | |
| run: | | |
| echo "=== Building FFI Library ===" | |
| echo "Building kreuzberg-py from source (no caching)" | |
| cargo build --release --package kreuzberg-py | |
| - name: Build Python wheels | |
| id: wheels | |
| shell: bash | |
| run: | | |
| echo "=== Building Python Wheels ===" | |
| echo "Building Python wheels from source (no caching)" | |
| task python:build:ci | |
| - name: Report cache statistics | |
| if: always() | |
| shell: bash | |
| run: | | |
| echo "=== Python Build Cache Statistics ===" | |
| echo "Platform: ${{ matrix.platform }}" | |
| echo "Target: ${{ matrix.target }}" | |
| if [ -d "target/wheels" ]; then | |
| echo "Wheels artifacts:" | |
| du -sh target/wheels || echo "target/wheels: empty" | |
| find target/wheels -maxdepth 1 -name "*.whl" -exec ls -lh {} \; 2>/dev/null | awk '{print $9, $5}' || echo "No wheel files found" | |
| fi | |
| if [ -d "packages/python/dist" ]; then | |
| echo "Python dist artifacts:" | |
| du -sh packages/python/dist || echo "packages/python/dist: empty" | |
| find packages/python/dist -maxdepth 1 -name "*.whl" -exec ls -lh {} \; 2>/dev/null | awk '{print $9, $5}' || echo "No dist wheel files found" | |
| fi | |
| - name: Upload wheels | |
| uses: actions/upload-artifact@v6 | |
| with: | |
| name: wheels-${{ matrix.platform }} | |
| path: target/wheels/*.whl | |
| retention-days: 7 | |
| - name: Cleanup Rust cache | |
| if: always() | |
| uses: ./.github/actions/cleanup-rust-cache | |
| - name: Smoke test wheel | |
| shell: bash | |
| run: scripts/ci/python/smoke-test-wheel.sh | |
| build-python-sdist: | |
| name: Python Sdist | |
| if: ${{ github.actor != 'dependabot[bot]' }} | |
| timeout-minutes: 180 | |
| runs-on: ubuntu-latest | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Setup Rust | |
| uses: ./.github/actions/setup-rust | |
| with: | |
| cache-key-prefix: python-sdist | |
| - name: Setup Python | |
| uses: ./.github/actions/setup-python-env | |
| with: | |
| python-version: "3.13" | |
| cache-prefix: sdist | |
| - name: Install Task | |
| uses: ./.github/actions/install-task | |
| - name: Build sdist | |
| shell: bash | |
| run: task python:build:sdist | |
| - name: Upload sdist | |
| uses: actions/upload-artifact@v6 | |
| with: | |
| name: python-sdist | |
| path: target/wheels/*.tar.gz | |
| retention-days: 7 | |
| test-python: | |
| name: Python Tests (${{ matrix.os }}) | |
| if: ${{ github.actor != 'dependabot[bot]' }} | |
| needs: build-and-smoke-python | |
| timeout-minutes: 180 | |
| strategy: | |
| fail-fast: true | |
| matrix: | |
| include: | |
| - os: ubuntu-latest | |
| coverage: true | |
| - os: ubuntu-24.04-arm | |
| coverage: false | |
| - os: macos-latest | |
| coverage: false | |
| - os: windows-latest | |
| coverage: false | |
| runs-on: ${{ matrix.os }} | |
| env: | |
| HF_HOME: ~/.cache/huggingface | |
| TRANSFORMERS_CACHE: ~/.cache/huggingface | |
| RUSTC_WRAPPER: '' | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Install system dependencies | |
| uses: ./.github/actions/install-system-deps | |
| - name: Setup Python | |
| uses: ./.github/actions/setup-python-env | |
| with: | |
| python-version: "3.13" | |
| cache-prefix: test-py-3.13 | |
| - name: Cache HuggingFace models | |
| uses: actions/cache@v5 | |
| with: | |
| path: ~/.cache/huggingface | |
| key: huggingface-${{ runner.os }}-py-3.13-${{ hashFiles('packages/python/pyproject.toml') || 'fallback' }} | |
| - name: Download wheels | |
| uses: actions/download-artifact@v7 | |
| with: | |
| name: ${{ matrix.os == 'ubuntu-latest' && 'wheels-linux-x86_64' || matrix.os == 'ubuntu-24.04-arm' && 'wheels-linux-arm64' || matrix.os == 'macos-latest' && 'wheels-macos-arm64' || 'wheels-windows-x86_64' }} | |
| path: dist/ | |
| - name: Install wheel | |
| shell: bash | |
| run: scripts/ci/python/install-wheel.sh | |
| - name: Install SpaCy model (Linux only) | |
| if: startsWith(matrix.os, 'ubuntu') | |
| run: scripts/ci/python/install-spacy-model.sh | |
| - name: Download PDFium | |
| uses: ./.github/actions/download-pdfium | |
| with: | |
| pdfium-version: ${{ env.PDFIUM_VERSION }} | |
| - name: Stage PDFium runtime | |
| uses: ./.github/actions/stage-pdfium-runtime | |
| with: | |
| destination: target/release | |
| - name: Setup ONNX Runtime | |
| uses: ./.github/actions/setup-onnx-runtime | |
| with: | |
| ort-version: ${{ env.ORT_VERSION }} | |
| - name: Install Task | |
| uses: ./.github/actions/install-task | |
| - name: Run Python tests | |
| shell: bash | |
| run: task python:test:ci | |
| - name: Upload coverage | |
| if: matrix.coverage | |
| uses: deepsourcelabs/test-coverage-action@master | |
| with: | |
| key: python | |
| coverage-file: packages/python/coverage.lcov | |
| dsn: ${{ secrets.DEEPSOURCE_DSN }} | |
| - name: Run E2E tests | |
| shell: bash | |
| env: | |
| PYTEST_TIMEOUT: 300 | |
| run: task python:e2e:test |