Releasing django-pgsync

Releases are fully automated via PyPI trusted publishing — no tokens, no local build tools. CI builds, twine checks and uploads the artifacts when a GitHub release is published.

The short version

1. Make sure CHANGELOG.md has a section for the new version:

   ## 0.2.0 (unreleased)
   
   - What changed...
   

2. Run:

   scripts/release.sh 0.2.0            # or --dry-run first to validate
   

That's it. The script bumps __version__, dates the changelog, commits, pushes, waits for tests, publishes the GitHub release (which triggers the PyPI upload), and verifies the new version is live on PyPI.

What the automation relies on

• .github/workflows/release.yml — builds and publishes on the release: published event. Run it manually from the Actions tab (workflow_dispatch) for a dry-run build that skips publishing.
• PyPI trusted publisher — registered on pypi.org for project django-pgsync, repo toluaina/django-pgsync, workflow release.yml, environment pypi. Nothing is stored in the repo.
• pypi GitHub environment — restricted to v* tag deployments, so only real releases can reach the publishing credentials.

Manual fallback

If the script is unavailable, the steps it automates:

# 1. bump __version__ in django_pgsync/__init__.py
# 2. change "## X.Y.Z (unreleased)" to "## X.Y.Z (YYYY-MM-DD)" in CHANGELOG.md
git commit -am "Release X.Y.Z" && git push
# 3. wait for CI, then:
gh release create vX.Y.Z --title "django-pgsync X.Y.Z" --notes "<changelog section>" --latest
# 4. the release workflow publishes to PyPI; verify:
curl -s https://pypi.org/pypi/django-pgsync/json | python3 -c "import json,sys; print(json.load(sys.stdin)['info']['version'])"

A failed publish burns nothing — no version lands on PyPI, and the workflow can be re-run from the Actions tab after fixing.