docling-parse: Add version 7.21.0 #78
Workflow file for this run
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
| # SPDX-FileCopyrightText: 2026 The RISE Project | |
| # SPDX-License-Identifier: MIT | |
| --- | |
| # Based on upstream's wheel builder: | |
| # https://github.com/docling-project/docling-parse/blob/main/.github/workflows/wheels.yml | |
| name: Build docling-parse wheels (riscv64) | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| version: | |
| description: 'Version glob to (re)build; empty builds every version of docs/packages/docling-parse.yaml not released yet' | |
| required: false | |
| default: '' | |
| pull_request: | |
| branches: [main] | |
| paths: | |
| - '.github/workflows/build-docling-parse.yml' | |
| - 'docs/packages/docling-parse.yaml' | |
| push: | |
| branches: [main] | |
| paths: | |
| - '.github/workflows/build-docling-parse.yml' | |
| - 'docs/packages/docling-parse.yaml' | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }} | |
| cancel-in-progress: true | |
| permissions: | |
| contents: read # to fetch code (actions/checkout) | |
| env: | |
| MANYLINUX_RISCV64_IMAGE: quay.io/pypa/manylinux_2_39_riscv64 | |
| jobs: | |
| setup: | |
| uses: $/.github/workflows/_setup.yml | |
| with: | |
| package: docling-parse | |
| version: ${{ inputs.version }} | |
| build_wheels: | |
| needs: [setup] | |
| if: needs.setup.outputs.versions != '[]' | |
| name: Build docling-parse ${{ matrix.version }} ${{ matrix.python }}-manylinux_riscv64 | |
| runs-on: ubuntu-24.04-riscv | |
| timeout-minutes: 720 | |
| strategy: | |
| fail-fast: false | |
| matrix: | |
| version: ${{ fromJSON(needs.setup.outputs.versions) }} | |
| python: ["cp312", "cp313", "cp314"] | |
| env: | |
| DOCLING_PARSE_VERSION: ${{ matrix.version }} | |
| steps: | |
| - name: Checkout docling-parse v${{ env.DOCLING_PARSE_VERSION }} | |
| uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1 | |
| with: | |
| repository: docling-project/docling-parse | |
| ref: v${{ env.DOCLING_PARSE_VERSION }} | |
| persist-credentials: false | |
| - uses: pypa/cibuildwheel@1828c10ab37f080699c7b81cea34097c684a7074 # v4.2.0 | |
| with: | |
| only: ${{ matrix.python }}-manylinux_riscv64 | |
| env: | |
| CIBW_MANYLINUX_RISCV64_IMAGE: ${{ env.MANYLINUX_RISCV64_IMAGE }} | |
| # pyproject pins cmake<4.0.0, which has no riscv64 wheel; the image's own | |
| # cmake is used instead, so the build requirements are preinstalled here. | |
| CIBW_BUILD_FRONTEND: "pip; args: --no-build-isolation" | |
| # The C++ dependencies are fetched and statically linked by the build, and | |
| # upstream ships none of their licences. | |
| CIBW_BEFORE_BUILD: | | |
| pip install "setuptools>=77.0.3" "pybind11>=2.13.6" "wheel>=0.43.0" | |
| curl -fsSL -o {project}/LICENSE.qpdf https://raw.githubusercontent.com/qpdf/qpdf/v12.2.0/LICENSE.txt | |
| curl -fsSL -o {project}/LICENSE.libjpeg-turbo https://raw.githubusercontent.com/libjpeg-turbo/libjpeg-turbo/3.1.1/LICENSE.md | |
| curl -fsSL -o {project}/LICENSE.openjpeg https://raw.githubusercontent.com/uclouvain/openjpeg/v2.5.3/LICENSE | |
| curl -fsSL -o {project}/LICENSE.lcms2 https://raw.githubusercontent.com/mm2/Little-CMS/lcms2.17/LICENSE | |
| curl -fsSL -o {project}/LICENSE.freetype https://raw.githubusercontent.com/freetype/freetype/VER-2-13-3/docs/FTL.TXT | |
| curl -fsSL -o {project}/LICENSE.blend2d https://raw.githubusercontent.com/blend2d/blend2d/6dbc2cefbc996379e07104e34519a440b49b15d7/LICENSE.md | |
| curl -fsSL -o {project}/LICENSE.utfcpp https://raw.githubusercontent.com/nemtrif/utfcpp/v4.0.6/LICENSE | |
| curl -fsSL -o {project}/LICENSE.json https://raw.githubusercontent.com/nlohmann/json/v3.12.0/LICENSE.MIT | |
| curl -fsSL -o {project}/LICENSE.loguru https://raw.githubusercontent.com/emilk/loguru/4adaa185883e3c04da25913579c451d3c32cfac1/LICENSE | |
| curl -fsSL -o {project}/LICENSE.cxxopts https://raw.githubusercontent.com/jarro2783/cxxopts/v3.3.1/LICENSE | |
| curl -fsSL -o {project}/LICENSE.pdfium https://raw.githubusercontent.com/chromium/pdfium/main/LICENSE | |
| CIBW_ENVIRONMENT: >- | |
| BUILD_THREADS=4 | |
| CMAKE_BUILD_TYPE=Release | |
| PIP_EXTRA_INDEX_URL=https://pypi.riseproject.dev/simple/ | |
| CIBW_TEST_SOURCES: tests | |
| CIBW_TEST_REQUIRES: pytest numpy huggingface_hub fonttools | |
| # Upstream's test runner installs the same static (CFF) Noto CJK face for the | |
| # non-embedded-font tests; the manylinux image ships no fonts at all. | |
| CIBW_BEFORE_TEST_LINUX: dnf -y install google-noto-sans-cjk-fonts | |
| CIBW_TEST_ENVIRONMENT: DOCLING_PARSE_CJK_FALLBACK_FONT=/usr/share/fonts/google-noto-sans-cjk-fonts/NotoSansCJK-Regular.ttc | |
| # Blend2D generates its compositing pipelines with asmjit, which has no riscv64 | |
| # backend, and its reference pipeline drops the Multiply blend mode — the same tests | |
| # fail on aarch64 built with BLEND2D_NO_JIT=ON. | |
| CIBW_TEST_COMMAND: python -m pytest -v tests/test_unit_*.py -k "not multiply" | |
| - name: Check wheel contents | |
| run: | | |
| python3 - ./wheelhouse/*.whl <<'EOF' | |
| import sys, zipfile | |
| names = zipfile.ZipFile(sys.argv[1]).namelist() | |
| assert any( | |
| n.startswith("docling_parse/pdf_parsers") and n.endswith(".so") for n in names | |
| ), names | |
| licences = { | |
| n.rsplit("/", 1)[1] for n in names if ".dist-info/licenses/" in n | |
| } - {""} | |
| expected = { | |
| "LICENSE", | |
| "LICENSE.blend2d", | |
| "LICENSE.cxxopts", | |
| "LICENSE.freetype", | |
| "LICENSE.json", | |
| "LICENSE.lcms2", | |
| "LICENSE.libjpeg-turbo", | |
| "LICENSE.loguru", | |
| "LICENSE.openjpeg", | |
| "LICENSE.pdfium", | |
| "LICENSE.qpdf", | |
| "LICENSE.utfcpp", | |
| } | |
| assert licences == expected, sorted(licences) | |
| EOF | |
| - name: Check Blend2D debug assertions are absent | |
| run: python3 scripts/check_no_blend2d_assertions.py ./wheelhouse/*.whl | |
| - uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1 | |
| with: | |
| name: docling-parse-${{ env.DOCLING_PARSE_VERSION }}-${{ matrix.python }}-manylinux_riscv64 | |
| path: ./wheelhouse/*.whl | |
| if-no-files-found: error | |
| publish: | |
| name: Publish docling-parse ${{ matrix.version }} | |
| needs: [setup, build_wheels] | |
| if: needs.setup.outputs.versions != '[]' | |
| strategy: | |
| fail-fast: false | |
| matrix: | |
| version: ${{ fromJSON(needs.setup.outputs.versions) }} | |
| permissions: | |
| contents: write | |
| pull-requests: write | |
| uses: $/.github/workflows/_publish-wheel.yml | |
| secrets: | |
| app-private-key: ${{ secrets.RISEPROJECT_APP_PRIVATE_KEY }} | |
| with: | |
| artifact-pattern: docling-parse-${{ matrix.version }}-*-manylinux_riscv64 |