Skip to main content

Publishing Workflow

Quantalytics ships with a PyPI-friendly layout. Use the following checklist when preparing a release.

1. Bump the Version​

Update the version field in pyproject.toml and document the change in CHANGELOG.md (optional).

2. Run Quality Gates​

pytest
ruff check .

3. Build Artifacts​

python -m build

This produces source and wheel distributions in dist/.

4. Trigger the Release Workflow​

  1. Commit the version bump.
  2. Create and push a tag that matches the version (e.g., git tag -a v0.2.0 -m "Release 0.2.0" && git push origin v0.2.0).

Pushing the tag runs the Release GitHub Actions workflow which:

  • Validates the tag matches pyproject.toml.
  • Builds source and wheel distributions.
  • Publishes to PyPI using the PYPI_API_TOKEN secret.
  • Drafts a GitHub release with the build artifacts attached.

Secrets: Configure PYPI_API_TOKEN with a PyPI API token that has upload rights for quantalytics.

Use workflow_dispatch on the Release workflow if you need to rerun the pipeline for a tag. For dry runs, keep the classic commands handy:

python -m build
twine upload --repository testpypi dist/*

5. Publish Documentation​

cd docs
npm install
npm run build

Deploy the generated build/ folder to your hosting provider (Netlify, Vercel, GitHub Pages, etc.).