This workflow should be called when a new release branch is created. When invoked, it does the following:
update specified base images in the Dockerfile to use the right release tags
update specified golang deps to track the right release branch
commits the results back to the originating branch
Before doing the actual changes, it checks if the originating branch matches the pattern release-X.Y where X and Y are integers representing major and minor versions for the release.
steps:
- uses: konveyor/release-tools/.github/workflows/prep-release.yaml@main
with:
# This is the ref of the branch. Only branches of format `refs/heads/release-X.Y`
# will enable the workflow to do the actual replacement of tags in Dockerfile.
# All other branches are ignored. `${{ github.ref }}` in the original repo
# should be used to get the value for this variable.
# Required
branch_ref: ${{ github.ref }}
# This is a list of images in the Dockerfile for which you want to update
# the tags. It should be a JSON array string e.g.
# `'["quay.io/konveyor/operator"]'`.
# Default: '[]'
images_to_update: '["quay.io/konveyor/jdtls-server-base"]'
# This is a list of go deps in the go.mod file for which you want to update
# the branches. It should be a JSON array string e.g.
# `'["github.com/konveyor/analyzer-lsp"]'`.
# Default: '[]'
go_deps_to_update: '["github.com/my/dependencies"]'
# This is the relative path to the Dockerfile in the repo.
# Default: ./Dockerfile
dockerfile: ./my.Dockerfile
Only https://github.com/konveyor/release-tools/blob/main/.github/workflows/prep-release.yaml action is referenced in README. I would suggest creating a folder (like
docs/) and adding a file for each action, then reference each file in the README. The format of the files could be something like this: