Skip to content

Release.md

Latest commit

 

History

History
36 lines (26 loc) · 2.45 KB

Release.md

File metadata and controls

36 lines (26 loc) · 2.45 KB

Generating a Release and Artifact:

Release Creation Process

  • We follow GitHub's standard process for creating a release, as described in GitHub's documentation. We use the UI to create a release with the Create new tag option, which synchronizes the release creation with tag creation.

NOTE: Each release should initially be marked as Set as a pre-release to allow for verification of the release notes and artifacts. If the release workflow fails, you can fix the issues and create a new release before publishing the final release.

Release Notes

For release notes, you can either:

  • Manually specify changes based on commits
  • Use the Generate release notes feature to automatically compile commit history from the main branch

Branch and Pull Request Workflow

  • All releases should be based on the main branch. For each release, changes to the main branch are aggregated through pull requests.

  • Each pull request triggers a build workflow that:

    • Verifies documentation structure
    • Builds the current branch (as latest) to identify navigation or Markdown issues

Deployment Process

Each push to the main branch triggers a deploy workflow that:

  • Builds the latest documentation
  • Pushes the latest build to the S3 bucket for live serving

Submodule Management

Some pages are fetched from the wire-server repository via submodules. We expect to submit pull requests for updating submodule pointers. In the future, we plan to automate this process to keep the wire-docs repository synchronized with the latest commits.

Note: For each commit to /docs in the wire-server repository, a build workflow verifies the document structure.

Release Workflow

Creating a release triggers a release workflow that:

  1. Builds documentation for the current tag
  2. Creates a tar file containing all documentation web pages
  3. Pushes the tag (ref_name) to the S3 bucket for live serving

The tar file created during the release can be used to serve documentation via standard web servers such as Apache, Nginx, or Python's http.server.