sudachipy: add build-sudachipy.yml for riscv64 wheels

.github/workflows/build-sudachipy.yml

@@ -0,0 +1,107 @@
+# SPDX-FileCopyrightText: 2026 The RISE Project
+# SPDX-License-Identifier: MIT
+---
+# Based on the `build-wheels` job of
+# https://github.com/WorksApplications/sudachi.rs/blob/v0.6.11/.github/workflows/build-python-wheels.yml
+name: Build sudachipy wheels (riscv64)
+
+on:
+  workflow_dispatch:
+    inputs:
+      version:
+        description: 'sudachipy version to build (git tag, e.g. v0.6.11)'
+        required: true
+        default: 'v0.6.11'
+  pull_request:
+    paths:
+      - '.github/workflows/build-sudachipy.yml'
+      - 'patches/sudachipy/**'
+
+concurrency:
+  group: ${{ github.workflow }}-${{ inputs.version || 'v0.6.11' }}-${{ github.head_ref || github.run_id }}
+  cancel-in-progress: true
+
+permissions:
+  contents: read  # to fetch code (actions/checkout)
+
+env:
+  # `inputs.version` is empty on pull_request events; default to v0.6.11 there.
+  SUDACHIPY_VERSION: ${{ inputs.version || 'v0.6.11' }}
+  MANYLINUX_RISCV64_IMAGE: quay.io/pypa/manylinux_2_39_riscv64
+
+jobs:
+  setup:
+    uses: $/.github/workflows/_setup.yml
+
+  build_wheels:
+    needs: [setup]
+    name: Build sudachipy ${{ inputs.version || 'v0.6.11' }} ${{ matrix.python }}-manylinux_riscv64
+    runs-on: ubuntu-24.04-riscv
+    timeout-minutes: 90
+    strategy:
+      fail-fast: false
+      matrix:
+        # No abi3: pyo3's abi3 feature isn't enabled in python/Cargo.toml.
+        python: ["cp312", "cp313", "cp314", "cp314t"]
+        include:
+          # tokenizers has no free-threaded wheel anywhere (abi3-only); its sdist
+          # build is unproven here, so drop just the one test module that needs it.
+          - python: "cp314t"
+            drop_pretokenizer_test: "rm tests/test_pretokenizers.py &&"
+            test_requires: 'sudachidict_core'
+
+    steps:
+      - name: Checkout sudachi.rs ${{ env.SUDACHIPY_VERSION }}
+        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1  # v7.0.1
+        with:
+          repository: WorksApplications/sudachi.rs
+          ref: ${{ env.SUDACHIPY_VERSION }}
+          persist-credentials: false
+
+      - name: Checkout python-wheels
+        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1  # v7.0.1
+        with:
+          path: python-wheels
+          persist-credentials: false
+
+      - name: Patch sudachi.rs source
+        run: git apply python-wheels/patches/sudachipy/${{ env.SUDACHIPY_VERSION }}/*.patch
+
+      - name: Build and test wheel
+        uses: pypa/cibuildwheel@1828c10ab37f080699c7b81cea34097c684a7074  # v4.2.0
+        with:
+          package-dir: python
+        env:
+          CIBW_ARCHS: riscv64
+          CIBW_BUILD: ${{ matrix.python }}-manylinux_riscv64
+          CIBW_SKIP: '*-musllinux_*'  # rustup.rs has no riscv64 musl toolchain
+          CIBW_MANYLINUX_RISCV64_IMAGE: ${{ env.MANYLINUX_RISCV64_IMAGE }}
+          # No Rust in the manylinux image; install it and put cargo on PATH. Skips
+          # upstream's PGO before-all (external corpus download + release build of
+          # sudachi-cli), which would otherwise re-run once per matrix leg here.
+          CIBW_BEFORE_ALL_LINUX: >-
+            curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
+          CIBW_ENVIRONMENT: PATH="$PATH:$HOME/.cargo/bin" PIP_EXTRA_INDEX_URL=https://pypi.riseproject.dev/simple/
+          # pypi.riseproject.dev's newest tokenizers (0.22.2) is behind PyPI's
+          # (0.23.2, no riscv64 wheel); pin so pip resolves to ours.
+          CIBW_TEST_REQUIRES: ${{ matrix.test_requires || 'tokenizers==0.22.2 sudachidict_core' }}
+          # Upstream's own build_and_test.sh test command, run against {package}
+          # (gotcha 5) since cibuildwheel tests from an empty scratch directory.
+          CIBW_TEST_COMMAND: >-
+            bash -c "cd {package} && ${{ matrix.drop_pretokenizer_test }} python -m unittest discover -s tests -v"
+
+      - uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a  # v7.0.1
+        with:
+          name: sudachipy-${{ env.SUDACHIPY_VERSION }}-${{ matrix.python }}-manylinux_riscv64
+          path: ./wheelhouse/*.whl
+          if-no-files-found: error
+
+  publish:
+    name: Publish sudachipy ${{ inputs.version || 'v0.6.11' }}
+    needs: [setup, build_wheels]
+    permissions:
+      contents: write
+      pull-requests: write
+    uses: $/.github/workflows/_publish-wheel.yml
+    with:
+      artifact-pattern: sudachipy-${{ inputs.version || 'v0.6.11' }}-*-manylinux_riscv64

patches/sudachipy/v0.6.11/0001-python-pyproject-drop-the-removed-cpython-freethreading-enable-group.patch

@@ -0,0 +1,33 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Tue, 8 Sep 2026 16:00:00 +0200
+Subject: [PATCH] python/pyproject: drop the removed cpython-freethreading
+ enable group
+
+cibuildwheel 4.2.0 (the version this repo pins for riscv64 builds) no
+longer recognises "cpython-freethreading" as an enable group -- it
+errors "Failed to parse enable group" before any build starts, on
+every interpreter, since [tool.cibuildwheel] is parsed as a whole
+(same failure as patches/pi-heif/1.4.0). Free-threaded builds (cp314t)
+no longer need an opt-in.
+
+Upstream-Status: To upstream [not yet submitted; sudachi.rs's develop branch (checked at 2f5653eb, past the v0.6.11 tag) still carries the same enable list, so there is no existing upstream fix to backport]
+---
+ python/pyproject.toml | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/python/pyproject.toml b/python/pyproject.toml
+index 0d61a4f2..887464de 100644
+--- a/python/pyproject.toml
++++ b/python/pyproject.toml
+@@ -4,7 +4,7 @@ requires = ["setuptools", "wheel", "setuptools-rust"]
+ [tool.cibuildwheel]
+ build = "cp39-* cp310-* cp311-* cp312-* cp313-* cp313t-* cp314-* cp314t-*"
+ skip = "*t-win* *-win32 *-musllinux_*"
+-enable = ["cpython-prerelease", "cpython-freethreading"]
++enable = ["cpython-prerelease"]
+
+ manylinux-x86_64-image = "manylinux2014"
+ manylinux-aarch64-image = "manylinux2014"
+--
+2.43.0