on:
  workflow_call:
    inputs:
      push_coverage:
        type: boolean
        description: "If true, the coverage results are pushed to codecov.io."
        default: true
      run_lint:
        type: boolean
        description: "If true, run the full lint job in this workflow."
        default: true
      force_all_checks:
        type: boolean
        description: "If true, run all test/example/package lanes regardless of changed-path filters."
        default: false
      run_package_compat:
        type: boolean
        description: "If true, run the cross-version package compatibility lanes."
        default: false
      run_windows:
        type: boolean
        description: "If true, run the Windows smoke test lane on GitHub-hosted runners."
        default: false
      run_macos:
        type: boolean
        description: "If true, run the macOS smoke test lane on GitHub-hosted runners."
        default: false
      use_tach:
        type: boolean
        description: "If true, enable Tach-aware test skipping in the cheap CI lanes."
        default: false
      tach_base_ref:
        type: string
        description: "Base git ref or commit used for Tach impact analysis."
        default: ""
      python_versions:
        type: string
        description: 'JSON array of Python versions to use for multi-version jobs, e.g. ["3.10", "3.12"].'
        default: '["3.10", "3.11", "3.12", "3.13", "3.14"]'
    secrets:
      CODECOV_TOKEN:
        required: false

env:
  FORCE_JAVASCRIPT_ACTIONS_TO_NODE24: "true"
  HF_HUB_DOWNLOAD_TIMEOUT: "90"
  HF_HUB_ETAG_TIMEOUT: "90"
  UV_FROZEN: "1"
  EXAMPLES_HEAVY: '^(batch_convert|chart_extraction|granite_vision_table_structure|minimal_asr_pipeline|minimal_vlm_pipeline)\.py$'
  EXAMPLES_UNSUPPORTED_IN_CI: '^(compare_vlm_models|custom_convert|demo_layout_vlm|develop_picture_enrichment|export_multimodal|gpu_standard_pipeline|gpu_vlm_pipeline|granitedocling_repetition_stopping|minimal|mlx_whisper_example|offline_convert|pictures_description|pictures_description_api|post_process_ocr_with_vlm|rapidocr_with_custom_models|run_with_formats_html_rendered|run_with_formats_html_rendered_mp|suryaocr_with_custom_models|vlm_pipeline_api_model)\.py$|xbrl_conversion\.ipynb$'

jobs:
  changes:
    runs-on: ubuntu-latest
    timeout-minutes: 5
    permissions:
      contents: read
      pull-requests: read
    outputs:
      run_tests: ${{ steps.force.outputs.run_tests || steps.filter.outputs.run_tests }}
      ml_suites: ${{ steps.force.outputs.ml_suites || steps.ml-matrix.outputs.ml_suites || '[]' }}
      run_examples_all: ${{ steps.force.outputs.run_examples_all || steps.filter.outputs.run_examples_all }}
      changed_example_scripts: ${{ steps.force.outputs.changed_example_scripts || steps.filter.outputs.changed_example_scripts }}
      changed_example_scripts_files: ${{ steps.force.outputs.changed_example_scripts_files || steps.filter.outputs.changed_example_scripts_files || '[]' }}
      run_package: ${{ steps.force.outputs.run_package || steps.filter.outputs.run_package }}
      run_tach: ${{ steps.force.outputs.run_tach || steps.filter.outputs.run_tach }}
    steps:
      - name: Force all lanes
        if: ${{ inputs.force_all_checks }}
        id: force
        run: |
          echo "run_tests=true" >> "$GITHUB_OUTPUT"
          echo 'ml_suites=["ocr","pdf-model","vlm","asr"]' >> "$GITHUB_OUTPUT"
          echo "run_examples_all=true" >> "$GITHUB_OUTPUT"
          echo "changed_example_scripts=true" >> "$GITHUB_OUTPUT"
          echo "changed_example_scripts_files=[]" >> "$GITHUB_OUTPUT"
          echo "run_package=true" >> "$GITHUB_OUTPUT"
          echo "run_tach=true" >> "$GITHUB_OUTPUT"

      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
        if: ${{ !inputs.force_all_checks }}
        with:
          fetch-depth: 0

      - name: Detect changed paths
        if: ${{ !inputs.force_all_checks }}
        id: filter
        uses: dorny/paths-filter@fbd0ab8f3e69293af611ebaee6363fc25e6d187d # v4.0.1
        with:
          list-files: json
          filters: |
            # Default catch-all for regular tests. The core pytest command
            # ignores modules marked with ML pytest markers; newly added tests
            # still run here unless intentionally marked for an ML lane.
            run_tests:
              - ".github/workflows/**"
              - ".github/scripts/**"
              - ".github/actions/**"
              - "pyproject.toml"
              - "uv.lock"
              - "Dockerfile"
              - "docling/**"
              - "tests/**"
            run_examples_all:
              - ".github/workflows/**"
              - ".github/scripts/**"
              - ".github/actions/**"
              - "pyproject.toml"
              - "uv.lock"
              - "Dockerfile"
              - "docling/**"
            changed_example_scripts:
              - added|modified: "docs/examples/*.py"
            run_package:
              - ".github/workflows/**"
              - ".github/scripts/**"
              - ".github/actions/**"
              - "pyproject.toml"
              - "uv.lock"
              - "README.md"
              - "LICENSE"
              - "docling/**"
            run_tach:
              - ".github/workflows/**"
              - ".github/scripts/**"
              - ".github/actions/**"
              - ".pre-commit-config.yaml"
              - "pyproject.toml"
              - "scripts/check_tach_module_coverage.py"
              - "tach.toml"
              - "uv.lock"
              - "docling/**"

      - name: Build ML suite matrix
        if: ${{ !inputs.force_all_checks }}
        id: ml-matrix
        env:
          # ML lanes share the regular test trigger; run_tach adds Tach and
          # pre-commit config changes that can affect test selection.
          RUN_ML: ${{ steps.filter.outputs.run_tests == 'true' || steps.filter.outputs.run_tach == 'true' }}
        run: |
          python3 .github/scripts/pytest_marker_selection.py matrix \
            --run-all-ml "$RUN_ML"

  lint:
    if: ${{ inputs.run_lint }}
    runs-on: ubuntu-latest
    timeout-minutes: 20
    strategy:
      fail-fast: false
      matrix:
        python-version: ["3.10"]
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
      - uses: ./.github/actions/setup-ubuntu-ci
        with:
          python_version: ${{ matrix.python-version }}
          uv_sync_args: --frozen --group dev --all-extras --all-packages --no-group docs --no-group examples

      - name: Set prek cache key
        run: echo "PY=$(python -VV | sha256sum | cut -d' ' -f1)" >> "$GITHUB_ENV"

      - name: Cache prek environments
        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
        with:
          path: ~/.cache/prek
          key: prek|${{ env.PY }}|${{ hashFiles('.pre-commit-config.yaml', '.github/dprint.json') }}
          restore-keys: |
            prek|${{ env.PY }}|

      - name: Check style
        run: |
          echo "--- Running prek style checks ---"
          uv run prek run --all-files

  tach:
    needs: changes
    if: ${{ needs.changes.outputs.run_tach == 'true' }}
    runs-on: ubuntu-latest
    timeout-minutes: 10
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
      - uses: ./.github/actions/setup-ubuntu-ci
        with:
          python_version: "3.10"
          uv_sync_args: --frozen --only-group dev --no-install-project

      - name: Check module coverage
        run: python3 scripts/check_tach_module_coverage.py

      - name: Check module boundaries
        run: uv run --no-sync tach check

  run-tests-core:
    needs: changes
    if: ${{ needs.changes.outputs.run_tests == 'true' }}
    runs-on: ubuntu-latest
    timeout-minutes: 45
    strategy:
      fail-fast: false
      matrix:
        python-version: ${{ fromJSON(inputs.python_versions) }}
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
        with:
          fetch-depth: 0
      - uses: ./.github/actions/setup-ubuntu-ci
        with:
          python_version: ${{ matrix.python-version }}
          uv_sync_args: --frozen --group dev --all-extras --all-packages --no-group docs --no-group examples
          install_system_deps: "true"
          cache_models: "true"

      - name: Run core test suite
        env:
          TACH_BASE_REF: ${{ inputs.tach_base_ref }}
        run: |
          echo "--- Running core tests ---"
          mapfile -t ML_IGNORE_ARGS < <(uv run --no-sync python .github/scripts/pytest_marker_selection.py core-ignore-args)
          TACH_ARGS=""
          if [ "${{ inputs.use_tach }}" = "true" ] && [ -n "$TACH_BASE_REF" ] && [ "$TACH_BASE_REF" != "0000000000000000000000000000000000000000" ]; then
            TACH_ARGS="--tach-base $TACH_BASE_REF"
          fi
          echo "Running core test suite"
          uv run pytest -v --durations=0 --cov=docling --cov-report=xml --cov-context=test $TACH_ARGS "${ML_IGNORE_ARGS[@]}"

      - name: Upload coverage to Codecov
        if: inputs.push_coverage
        uses: codecov/codecov-action@57e3a136b779b570ffcdbf80b3bdc90e7fab3de2 # v6.0.0
        with:
          token: ${{ secrets.CODECOV_TOKEN }}
          files: ./coverage.xml
          flags: run-tests-core

  run-tests-ml:
    needs: changes
    if: ${{ needs.changes.outputs.ml_suites != '[]' }}
    runs-on: ubuntu-latest
    timeout-minutes: 60
    strategy:
      fail-fast: false
      matrix:
        python-version: ${{ fromJSON(inputs.python_versions) }}
        suite: ${{ fromJSON(needs.changes.outputs.ml_suites) }}
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
        with:
          fetch-depth: 0
      - uses: ./.github/actions/setup-ubuntu-ci
        with:
          python_version: ${{ matrix.python-version }}
          uv_sync_args: --frozen --group dev --all-extras --all-packages --no-group docs --no-group examples
          install_system_deps: "true"
          cache_models: "true"

      - name: Pre-download OCR models
        if: ${{ matrix.suite == 'ocr' }}
        run: uv run python -c "import easyocr; reader = easyocr.Reader(['en', 'fr', 'de', 'es'])"

      - name: Run ML test suite
        env:
          TACH_BASE_REF: ${{ inputs.tach_base_ref }}
        run: |
          echo "--- Running ML suite: ${{ matrix.suite }} ---"
          SUITE_MARKER=$(uv run --no-sync python .github/scripts/pytest_marker_selection.py suite-marker "${{ matrix.suite }}")
          mapfile -t SUITE_ARGS < <(uv run --no-sync python .github/scripts/pytest_marker_selection.py suite-args "${{ matrix.suite }}")
          if [ ${#SUITE_ARGS[@]} -eq 0 ]; then
            echo "No tests are marked for ${{ matrix.suite }}" >&2
            exit 1
          fi
          TACH_ARGS=""
          if [ "${{ inputs.use_tach }}" = "true" ] && [ -n "$TACH_BASE_REF" ] && [ "$TACH_BASE_REF" != "0000000000000000000000000000000000000000" ]; then
            TACH_ARGS="--tach-base $TACH_BASE_REF"
          fi
          uv run pytest -v --durations=0 --cov=docling --cov-report=xml --cov-context=test $TACH_ARGS -m "$SUITE_MARKER" "${SUITE_ARGS[@]}"

      - name: Upload coverage to Codecov
        if: inputs.push_coverage
        uses: codecov/codecov-action@57e3a136b779b570ffcdbf80b3bdc90e7fab3de2 # v6.0.0
        with:
          token: ${{ secrets.CODECOV_TOKEN }}
          files: ./coverage.xml
          flags: run-tests-ml-${{ matrix.suite }}

  run-examples-light:
    needs: changes
    if: >-
      ${{
        needs.changes.outputs.run_examples_all == 'true' ||
        needs.changes.outputs.changed_example_scripts == 'true'
      }}
    runs-on: ubuntu-latest
    timeout-minutes: 60
    strategy:
      fail-fast: false
      matrix:
        python-version: ["3.10"]
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
      - uses: ./.github/actions/setup-ubuntu-ci
        with:
          python_version: ${{ matrix.python-version }}
          uv_sync_args: --frozen --group examples --all-extras --all-packages --no-group docs --no-group dev
          install_system_deps: "true"
          cache_models: "true"

      - name: Free up disk space
        run: |
          df -h
          sudo rm -rf /usr/share/dotnet
          sudo rm -rf /usr/local/lib/android
          sudo rm -rf /opt/ghc
          sudo apt-get clean
          df -h

      - name: Run selected light examples
        env:
          RUN_ALL_EXAMPLES: ${{ needs.changes.outputs.run_examples_all }}
          CHANGED_EXAMPLE_FILES: ${{ needs.changes.outputs.changed_example_scripts_files }}
        run: |
          args=(
            --examples-dir docs/examples
            --changed-files-json "$CHANGED_EXAMPLE_FILES"
            --exclude-pattern "${EXAMPLES_HEAVY}|${EXAMPLES_UNSUPPORTED_IN_CI}"
          )
          if [ "$RUN_ALL_EXAMPLES" = "true" ]; then
            args+=(--run-all)
          fi
          uv run --no-sync python .github/scripts/run_selected_examples.py "${args[@]}"

  run-tests-windows:
    needs: changes
    if: ${{ inputs.run_windows }}
    runs-on: windows-latest
    timeout-minutes: 30
    env:
      PYTHONUTF8: "1"
      UV_PYTHON_PREFERENCE: only-system
    strategy:
      fail-fast: false
      matrix:
        python-version: ["3.10"]
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      - name: Set up Python ${{ matrix.python-version }}
        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
        with:
          python-version: ${{ matrix.python-version }}

      - name: Install uv
        uses: astral-sh/setup-uv@v7
        with:
          enable-cache: true

      - name: Install dependencies
        shell: bash
        run: |
          uv python pin "${{ matrix.python-version }}"
          uv sync --frozen --group dev --all-packages --no-group docs --no-group examples

      - name: Check import and CLI wiring
        shell: bash
        run: |
          uv run python -c "import docling; from docling.document_converter import DocumentConverter; DocumentConverter(); print('Docling import smoke passed')"
          uv run docling --help

      - name: Run cross-platform smoke tests
        shell: bash
        run: |
          mapfile -t cross_platform_args < <(uv run --no-sync python .github/scripts/pytest_marker_selection.py marker-args cross_platform)
          if [ "${#cross_platform_args[@]}" -eq 0 ]; then
            echo "No tests are marked with pytest.mark.cross_platform" >&2
            exit 1
          fi
          uv run pytest -p no:tach -m cross_platform "${cross_platform_args[@]}" --durations=10

  run-tests-macos:
    needs: changes
    if: ${{ inputs.run_macos }}
    runs-on: macos-latest
    timeout-minutes: 30
    env:
      UV_PYTHON_PREFERENCE: only-system
    strategy:
      fail-fast: false
      matrix:
        python-version: ["3.10"]
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      - name: Set up Python ${{ matrix.python-version }}
        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
        with:
          python-version: ${{ matrix.python-version }}

      - name: Install uv
        uses: astral-sh/setup-uv@v7
        with:
          enable-cache: true

      - name: Install dependencies
        shell: bash
        run: |
          uv python pin "${{ matrix.python-version }}"
          uv sync --frozen --group dev --all-packages --no-group docs --no-group examples

      - name: Check import and CLI wiring
        shell: bash
        run: |
          uv run python -c "import docling; from docling.document_converter import DocumentConverter; DocumentConverter(); print('Docling import smoke passed')"
          uv run docling --help

      - name: Run cross-platform smoke tests
        shell: bash
        run: |
          mapfile -t cross_platform_args < <(uv run --no-sync python .github/scripts/pytest_marker_selection.py marker-args cross_platform)
          if [ "${#cross_platform_args[@]}" -eq 0 ]; then
            echo "No tests are marked with pytest.mark.cross_platform" >&2
            exit 1
          fi
          uv run pytest -p no:tach -m cross_platform "${cross_platform_args[@]}" --durations=10

  test-pip-install-no-lock:
    needs: changes
    if: ${{ inputs.run_package_compat && needs.changes.outputs.run_package == 'true' }}
    runs-on: ubuntu-latest
    timeout-minutes: 30
    env:
      SELECTED_PYTHON_VERSIONS: ${{ join(fromJSON(inputs.python_versions), ' ') }}
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      - name: Install uv
        uses: astral-sh/setup-uv@v7

      - name: Test pip install across selected Python versions
        run: |
          for py_version in $SELECTED_PYTHON_VERSIONS; do
            echo "=========================================="
            echo "Testing Python $py_version"
            echo "=========================================="

            # Create virtual environment with uv
            uv venv /tmp/venv-${py_version} --python=${py_version}
            source /tmp/venv-${py_version}/bin/activate

            # Install docling-slim with pip (no lock file)
            uv pip install --torch-backend=cpu -e .[all]

            # Run basic import test
            python -c "import docling; print('Import successful for Python ${py_version}')"

            # Cleanup
            deactivate
            rm -rf /tmp/venv-${py_version}

            echo "Python $py_version: PASSED"
            echo ""
          done

  test-pip-install-no-dev-headers:
    needs: changes
    if: ${{ inputs.run_package_compat && needs.changes.outputs.run_package == 'true' }}
    runs-on: ubuntu-latest
    timeout-minutes: 30
    env:
      SELECTED_PYTHON_VERSIONS: ${{ join(fromJSON(inputs.python_versions), ' ') }}
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      - name: Install uv
        uses: astral-sh/setup-uv@v7

      - name: Test pip install without dev headers across selected Python versions
        run: |
          for py_version in $SELECTED_PYTHON_VERSIONS; do
            echo "=========================================="
            echo "Testing Python $py_version (no dev headers)"
            echo "=========================================="

            # Create virtual environment with uv
            uv venv /tmp/venv-nodev-${py_version} --python=${py_version}
            source /tmp/venv-nodev-${py_version}/bin/activate

            # Find and remove Python.h from the Python installation
            echo "Removing Python development headers from Python installation..."
            python_include_dir=$(python -c "import sysconfig; print(sysconfig.get_path('include'))")
            echo "Python include directory: $python_include_dir"

            if [ -f "$python_include_dir/Python.h" ]; then
              echo "Found Python.h, removing it and other headers..."
              # Use sudo if the directory is system-owned
              if [ -w "$python_include_dir" ]; then
                rm -rf "$python_include_dir"/*
              else
                sudo rm -rf "$python_include_dir"/*
              fi
              echo "✓ Headers removed"
            else
              echo "Warning: Python.h not found at expected location"
            fi

            # Verify that compilation fails without dev headers
            # Try to install numpy from source (sdist) - this should fail
            echo "Verifying compilation fails without dev headers..."
            set +e  # Temporarily allow command to fail
            uv pip install --no-binary=:all: numpy==1.26.4 > /tmp/numpy-install-${py_version}.log 2>&1
            numpy_exit_code=$?
            set -e  # Re-enable exit on error

            if [ $numpy_exit_code -eq 0 ]; then
              echo "ERROR: numpy installation from source succeeded, but it should have failed without dev headers!"
              cat /tmp/numpy-install-${py_version}.log
              exit 1
            else
              echo "✓ Compilation correctly failed without dev headers (expected behavior)"
              # Check that the error mentions missing Python.h or similar
              if grep -qi "Python.h\|fatal error.*\.h" /tmp/numpy-install-${py_version}.log; then
                echo "✓ Error message confirms missing development headers"
              else
                echo "Warning: Error message doesn't explicitly mention missing headers, but compilation failed as expected"
              fi
            fi

            # Install docling-slim with pip (no lock file, no compilation)
            echo "grpcio>=1.71.0" > override-grpcio.txt
            uv pip install --torch-backend=cpu -e ".[all]" --overrides override-grpcio.txt

            # Run basic import test
            python -c "import docling; print('Import successful for Python ${py_version} without dev headers')"

            # Cleanup
            deactivate
            rm -rf /tmp/venv-nodev-${py_version}

            echo "Python $py_version (no dev headers): PASSED"
            echo ""
          done

  build-package:
    needs: changes
    if: ${{ needs.changes.outputs.run_package == 'true' }}
    runs-on: ubuntu-latest
    timeout-minutes: 15
    strategy:
      matrix:
        python-version: ["3.10"]
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      - name: Install uv and set the python version
        uses: astral-sh/setup-uv@v7
        with:
          python-version: ${{ matrix.python-version }}
          enable-cache: true

      - name: Install dependencies
        run: uv sync --all-extras

      - name: Build packages
        run: bash .github/scripts/build-packages.sh

      - name: Check content of wheels
        run: |
          for whl in dist/*/*.whl; do
            echo "=== $whl ==="
            unzip -l "$whl"
          done

      - name: Store the distribution packages
        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
        with:
          name: python-package-distributions
          path: dist/

  test-package:
    needs:
      - changes
      - build-package
    if: ${{ needs.changes.outputs.run_package == 'true' }}
    runs-on: ubuntu-latest
    timeout-minutes: 15
    strategy:
      matrix:
        python-version: ["3.10"]
    steps:
      - name: Download all the dists
        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
        with:
          name: python-package-distributions
          path: dist/

      - name: Install uv and set the python version
        uses: astral-sh/setup-uv@v7
        with:
          python-version: ${{ matrix.python-version }}
          activate-environment: true
          enable-cache: false

      - name: Install package
        run: |
          uv pip install --find-links dist/docling-slim/ dist/docling/docling-*.whl

      - name: Run docling
        run: uv run docling --help

  check:
    if: ${{ always() }}
    needs:
      - changes
      - lint
      - tach
      - run-tests-core
      - run-tests-ml
      - run-examples-light
      - run-tests-windows
      - run-tests-macos
      - test-pip-install-no-lock
      - test-pip-install-no-dev-headers
      - build-package
      - test-package
    runs-on: ubuntu-latest
    timeout-minutes: 5
    permissions:
      contents: read
    steps:
      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      - name: All Green
        env:
          NEEDS_JSON: ${{ toJSON(needs) }}
          ALLOWED_SKIPS: >-
            lint tach run-tests-core run-tests-ml run-examples-light run-tests-windows
            run-tests-macos test-pip-install-no-lock test-pip-install-no-dev-headers
            build-package test-package
        run: >-
          python3 .github/scripts/check_needs_results.py
          --needs-json "$NEEDS_JSON"
          --allowed-skips "$ALLOWED_SKIPS"