build-kernels-data.yml

# SPDX-FileCopyrightText: 2026 The RISE Project
# SPDX-License-Identifier: MIT
---
# This workflow is based on the `linux` job of
# https://github.com/huggingface/kernels/blob/v0.16.1/.github/workflows/kernels_data_python_release.yaml
# and the test steps of its test_python.yaml.
name: Build kernels-data wheels (riscv64)

on:
  workflow_dispatch:
    inputs:
      version:
        description: 'Version glob to (re)build; empty builds every version of docs/packages/kernels-data.yaml not released yet'
        required: false
        default: ''
  pull_request:
    branches: [main]
    paths:
      - '.github/workflows/build-kernels-data.yml'
      - 'docs/packages/kernels-data.yaml'
  push:
    branches: [main]
    paths:
      - '.github/workflows/build-kernels-data.yml'
      - 'docs/packages/kernels-data.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: kernels-data
      version: ${{ inputs.version }}

  build_wheels:
    needs: [setup]
    if: needs.setup.outputs.versions != '[]'
    name: Build kernels-data ${{ matrix.version }} ${{ matrix.tag }}-manylinux_riscv64
    runs-on: ubuntu-24.04-riscv
    timeout-minutes: 120
    strategy:
      fail-fast: false
      matrix:
        version: ${{ fromJSON(needs.setup.outputs.versions) }}
        # `tag` is a real matrix dimension, not an include-only key: two includes
        # that add the same new key collapse into one job (gotcha 402).
        tag: [cp38-abi3, cp314t]
        include:
          # kernels-data/bindings/python/Cargo.toml declares pyo3 with the
          # abi3/abi3-py38 features unconditionally, so one abi3 wheel serves
          # every GIL-ful interpreter; pyo3 disables abi3 under Py_GIL_DISABLED,
          # giving the free-threaded build its own wheel, which is why upstream
          # runs a second `-i python3.14t` maturin step.
          - tag: cp38-abi3
            # 3.8 only names the tag's floor and is absent from the manylinux
            # image, so the wheel is built on this repo's cp312 floor and
            # re-tested on the newer interpreters.
            build: cp312-manylinux_riscv64 cp313-manylinux_riscv64 cp314-manylinux_riscv64
          - tag: cp314t
            build: cp314t-manylinux_riscv64

    env:
      KERNELS_DATA_VERSION: ${{ matrix.version }}

    steps:
      - name: Checkout kernels v${{ env.KERNELS_DATA_VERSION }}
        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1  # v7.0.1
        with:
          repository: huggingface/kernels
          ref: v${{ env.KERNELS_DATA_VERSION }}
          persist-credentials: false

      - name: Stage the licence beside kernels-data's pyproject.toml
        # maturin globs LICEN[CS]E* relative to the pyproject directory, which
        # is kernels-data/bindings/python/ in this workspace, and ignores the
        # `include ../../../LICENSE` in the MANIFEST.in next to it -- so
        # upstream's own wheels carry no licence file at all.
        run: cp LICENSE kernels-data/bindings/python/

      - name: Build wheels
        uses: pypa/cibuildwheel@1828c10ab37f080699c7b81cea34097c684a7074  # v4.2.0
        with:
          package-dir: kernels-data/bindings/python
          output-dir: wheelhouse/
        env:
          # musllinux is dropped: rustup.rs ships no riscv64 musl toolchain.
          CIBW_BUILD: ${{ matrix.build }}
          CIBW_MANYLINUX_RISCV64_IMAGE: ${{ env.MANYLINUX_RISCV64_IMAGE }}
          # kernels-data ships no [tool.cibuildwheel], so the Rust toolchain
          # its maturin backend needs is installed in-container here.
          CIBW_BEFORE_ALL_LINUX: >-
            curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
          # mypy, which upstream's stubtest check needs, publishes no riscv64
          # wheel on public PyPI; ours does.
          CIBW_ENVIRONMENT_LINUX: >-
            PATH="$PATH:$HOME/.cargo/bin"
            PIP_EXTRA_INDEX_URL=https://pypi.riseproject.dev/simple/
          CIBW_TEST_REQUIRES: pytest mypy
          CIBW_TEST_SOURCES: >-
            kernels-data/bindings/python/tests
            kernels-data/bindings/python/stubtest-allowlist.txt
          # maturin synthesises kernels_data/__init__.py as a re-export shim, so
          # the compiled module is the kernels_data.kernels_data submodule.
          CIBW_TEST_COMMAND: >-
            python -c "import kernels_data.kernels_data as m, importlib.metadata as md;
            assert m.__file__.endswith('.so'), m.__file__;
            files = md.files('kernels-data');
            assert any(str(p).endswith('licenses/LICENSE') for p in files), files" &&
            cd kernels-data/bindings/python &&
            python -m pytest -v tests &&
            python -m mypy.stubtest kernels_data --allowlist stubtest-allowlist.txt

      - uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a  # v7.0.1
        with:
          name: kernels-data-${{ env.KERNELS_DATA_VERSION }}-${{ matrix.tag }}-manylinux_riscv64
          path: wheelhouse/*.whl
          if-no-files-found: error

  publish:
    name: Publish kernels-data ${{ 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: ${{ secrets.RISEPROJECT_APP_PRIVATE_KEY }}
    with:
      artifact-pattern: kernels-data-${{ matrix.version }}-*-manylinux_riscv64