Release workflow (learning-at-home/hivemind)
The Release workflow from learning-at-home/hivemind, explained and optimized by Latchkey.
CI health: C - fair
Point runs-on at Latchkey and get caching, job timeouts, SHA-pinned actions, self-healing for flaky steps, and up to 58% lower cost, applied automatically.
What it does
This is the Release workflow from the learning-at-home/hivemind repository, a real project running GitHub Actions. It is shown here with attribution under its MIT license.
Below, Latchkey shows a faster, safer version produced by its optimization engine.
The workflow
name: Release
on:
release:
types: [ published ]
jobs:
build-wheels:
name: Build wheels for ${{ matrix.platform }}
runs-on: ubuntu-latest
strategy:
matrix:
platform: [ linux-amd64, linux-arm64, darwin-amd64, darwin-arm64 ]
steps:
- uses: actions/checkout@v4
- name: Set up Python
uses: actions/setup-python@v5
with:
python-version: '3.11'
- name: Cache dependencies
uses: actions/cache@v4
with:
path: ~/.cache/pip
key: Key-v1-3.11-build
- name: Install build dependencies
run: |
python -m pip install --upgrade pip build wheel
- name: Build wheel for ${{ matrix.platform }}
env:
HIVEMIND_TARGET_PLATFORM: ${{ matrix.platform }}
run: |
python -m build --wheel
- name: Retag wheel with platform-specific tag
run: |
case "${{ matrix.platform }}" in
linux-amd64)
wheel tags --python-tag=py3 --abi-tag=none --platform-tag=manylinux2014_x86_64 --remove dist/*.whl
;;
linux-arm64)
wheel tags --python-tag=py3 --abi-tag=none --platform-tag=manylinux2014_aarch64 --remove dist/*.whl
;;
darwin-amd64)
wheel tags --python-tag=py3 --abi-tag=none --platform-tag=macosx_10_9_x86_64 --remove dist/*.whl
;;
darwin-arm64)
wheel tags --python-tag=py3 --abi-tag=none --platform-tag=macosx_11_0_arm64 --remove dist/*.whl
;;
esac
- name: Verify wheel contents
run: |
python -m zipfile -l dist/*.whl | grep "hivemind_cli/p2pd"
python -m wheel unpack dist/*.whl
BINARY_PATH=$(echo hivemind-*/hivemind/hivemind_cli/p2pd)
FILE_OUTPUT=$(file "$BINARY_PATH")
echo "Binary info: $FILE_OUTPUT"
case "${{ matrix.platform }}" in
linux-amd64)
echo "$FILE_OUTPUT" | grep -q "ELF 64-bit.*x86-64"
;;
linux-arm64)
echo "$FILE_OUTPUT" | grep -q "ELF 64-bit.*aarch64"
;;
darwin-amd64)
echo "$FILE_OUTPUT" | grep -q "Mach-O.*x86_64"
;;
darwin-arm64)
echo "$FILE_OUTPUT" | grep -q "Mach-O.*arm64"
;;
esac
echo "Verified: p2pd binary format matches ${{ matrix.platform }}"
- name: Upload wheel artifacts
uses: actions/upload-artifact@v4
with:
name: wheel-${{ matrix.platform }}
path: dist/*.whl
retention-days: 1
build-sdist:
name: Build source distribution
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Set up Python
uses: actions/setup-python@v5
with:
python-version: '3.11'
- name: Cache dependencies
uses: actions/cache@v4
with:
path: ~/.cache/pip
key: Key-v1-3.11-build
- name: Install build dependencies
run: |
python -m pip install --upgrade pip build
- name: Build source distribution
run: |
python -m build --sdist
- name: Verify sdist contents
run: |
if tar -tzf dist/*.tar.gz | grep -q "hivemind_cli/p2pd"; then
echo "Error: p2pd binary found in sdist"
exit 1
fi
echo "p2pd binary not included in source distribution"
- name: Upload sdist artifacts
uses: actions/upload-artifact@v4
with:
name: sdist
path: dist/*.tar.gz
retention-days: 1
publish:
name: Publish to PyPI
needs: [ build-wheels, build-sdist ]
runs-on: ubuntu-latest
permissions:
id-token: write # Required for PyPI trusted publishing
steps:
- name: Download all artifacts
uses: actions/download-artifact@v4
with:
pattern: "*"
path: dist/
merge-multiple: true
- name: List distribution files
run: |
ls -la dist/
echo "Total files: $(ls dist/ | wc -l)"
- name: Publish to PyPI
uses: pypa/gh-action-pypi-publish@release/v1
with:
packages-dir: dist/
verbose: true
print-hash: true
The same workflow, on Latchkey
Estimated ~20% faster on cache hits, plus fewer wasted runs and a safer supply chain. Added and changed lines are highlighted.
name: Release on: release: types: [ published ] jobs: build-wheels: timeout-minutes: 30 name: Build wheels for ${{ matrix.platform }} runs-on: latchkey-small strategy: matrix: platform: [ linux-amd64, linux-arm64, darwin-amd64, darwin-arm64 ] steps: - uses: actions/checkout@v4 - name: Set up Python uses: actions/setup-python@v5 with: cache: 'pip' python-version: '3.11' - name: Cache dependencies uses: actions/cache@v4 with: path: ~/.cache/pip key: Key-v1-3.11-build - name: Install build dependencies run: | python -m pip install --upgrade pip build wheel - name: Build wheel for ${{ matrix.platform }} env: HIVEMIND_TARGET_PLATFORM: ${{ matrix.platform }} run: | python -m build --wheel - name: Retag wheel with platform-specific tag run: | case "${{ matrix.platform }}" in linux-amd64) wheel tags --python-tag=py3 --abi-tag=none --platform-tag=manylinux2014_x86_64 --remove dist/*.whl ;; linux-arm64) wheel tags --python-tag=py3 --abi-tag=none --platform-tag=manylinux2014_aarch64 --remove dist/*.whl ;; darwin-amd64) wheel tags --python-tag=py3 --abi-tag=none --platform-tag=macosx_10_9_x86_64 --remove dist/*.whl ;; darwin-arm64) wheel tags --python-tag=py3 --abi-tag=none --platform-tag=macosx_11_0_arm64 --remove dist/*.whl ;; esac - name: Verify wheel contents run: | python -m zipfile -l dist/*.whl | grep "hivemind_cli/p2pd" python -m wheel unpack dist/*.whl BINARY_PATH=$(echo hivemind-*/hivemind/hivemind_cli/p2pd) FILE_OUTPUT=$(file "$BINARY_PATH") echo "Binary info: $FILE_OUTPUT" case "${{ matrix.platform }}" in linux-amd64) echo "$FILE_OUTPUT" | grep -q "ELF 64-bit.*x86-64" ;; linux-arm64) echo "$FILE_OUTPUT" | grep -q "ELF 64-bit.*aarch64" ;; darwin-amd64) echo "$FILE_OUTPUT" | grep -q "Mach-O.*x86_64" ;; darwin-arm64) echo "$FILE_OUTPUT" | grep -q "Mach-O.*arm64" ;; esac echo "Verified: p2pd binary format matches ${{ matrix.platform }}" - name: Upload wheel artifacts uses: actions/upload-artifact@v4 with: name: wheel-${{ matrix.platform }} path: dist/*.whl retention-days: 1 build-sdist: timeout-minutes: 30 name: Build source distribution runs-on: latchkey-small steps: - uses: actions/checkout@v4 - name: Set up Python uses: actions/setup-python@v5 with: cache: 'pip' python-version: '3.11' - name: Cache dependencies uses: actions/cache@v4 with: path: ~/.cache/pip key: Key-v1-3.11-build - name: Install build dependencies run: | python -m pip install --upgrade pip build - name: Build source distribution run: | python -m build --sdist - name: Verify sdist contents run: | if tar -tzf dist/*.tar.gz | grep -q "hivemind_cli/p2pd"; then echo "Error: p2pd binary found in sdist" exit 1 fi echo "p2pd binary not included in source distribution" - name: Upload sdist artifacts uses: actions/upload-artifact@v4 with: name: sdist path: dist/*.tar.gz retention-days: 1 publish: timeout-minutes: 30 name: Publish to PyPI needs: [ build-wheels, build-sdist ] runs-on: latchkey-small permissions: id-token: write # Required for PyPI trusted publishing steps: - name: Download all artifacts uses: actions/download-artifact@v4 with: pattern: "*" path: dist/ merge-multiple: true - name: List distribution files run: | ls -la dist/ echo "Total files: $(ls dist/ | wc -l)" - name: Publish to PyPI uses: pypa/gh-action-pypi-publish@release/v1 with: packages-dir: dist/ verbose: true print-hash: true
What changed
- Run on Latchkey managed runners with one line (
runs-on), which apply the fixes below automatically and self-heal transient failures. This example useslatchkey-small; pick the runner size that fits the job. - Cache dependency installs on the setup step so they are served from cache.
- Add a job timeout so a hung step cannot burn hours of runner time.
1 third-party action is referenced by a movable tag. Pin it to the commit SHA (Latchkey resolves and applies this automatically) so a repointed tag cannot change what runs.
What Latchkey heals here
This workflow has steps that commonly fail on transient issues (network, registries, flaky browsers). On Latchkey managed runners they are detected, retried, and self-healed instead of failing your build:
- Dependency installs
This workflow runs 3 jobs (6 with the matrix expanded) per trigger. On Latchkey the same minutes cost up to 58% less than GitHub-hosted, with zero queue time.