release-process.md
While our release schedule is flexible, our general strategy is to release several larger improvements inside each stable release. In order to provide faster access to fixes and enhancements between main releases we provide release candidates which are published on every merge into main.
While the versioning and publishing of our primitives is mostly automated via scripts, updates to our documentation website is currently a manual process. We are working to improve this but for now this outline should help contributors with the process.
We track versions during the pull request process. As features are added, modified or improved it's important to keep track of these via versioning.
Run pnpm changeset to mark the appropriate type of change for those packages. This is later consumed when publishing new versions. Be sure to check-in these files along with your code changes.
stable branch and pull the latest changes from mainstable. This will trigger the Changeset action, which will create a new release PR.Release candidates are automatically published when new changes are merged into main.
Our documentation is in a separate repository and updating it is a three step process:
Steps 2 and 3 are typically raised as separate pull requests to make changes easier to review.
This is as simple as duplicating the latest page and updating the version number to match the release. Some things to keep in mind: