From 3b01956af74883a777bbcafcf6eed91eebd85543 Mon Sep 17 00:00:00 2001 From: RJ Hollberg Date: Sun, 20 Sep 2026 14:40:17 -0600 Subject: [PATCH] refactor: point draft-release.yml at release/draft@v1 Companion to TaffarelJr/.actions#39, which moved draft-release into release/. draft-release.yml itself (the workflow file name and its identical copy in every repo) is unchanged - only the action path it calls, and the two docs/comments naming that action directly. Verified locally: YAML parses; Common-GitHub's own test suite (the file whose string this touches) still passes, 68/68. Co-Authored-By: Claude Sonnet 5 --- .github/workflows/draft-release.yml | 6 +++--- docs/ReleaseProcess.md | 4 ++-- scripts/Common-GitHub.psm1 | 2 +- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/.github/workflows/draft-release.yml b/.github/workflows/draft-release.yml index 8bea261..259c605 100644 --- a/.github/workflows/draft-release.yml +++ b/.github/workflows/draft-release.yml @@ -4,7 +4,7 @@ # See docs/ReleaseProcess.md for why this is the only way in and why a # release is a deliberate decision, not something every merge does. # -# Nothing is rebuilt here - it just triggers; draft-release takes the +# Nothing is rebuilt here - it just triggers; release/draft takes the # artifacts CI already produced, for this commit or for a named version. # # 'uses' can't take an expression, so bumping the major version is an edit @@ -41,12 +41,12 @@ jobs: fetch-depth: 0 # The changelog needs all the history and tags - name: Draft the release - uses: TaffarelJr/.actions/draft-release@v1 + uses: TaffarelJr/.actions/release/draft@v1 with: version: ${{ inputs.version }} env: # Must be set here, at the workflow level - an env var set on a - # *nested* composite action's own step (draft-release calling + # *nested* composite action's own step (release/draft calling # ai-inference) does not reliably reach the action it calls; only # one set by the calling workflow cascades down through every # layer. This is what lets the Copilot CLI, several layers deep, diff --git a/docs/ReleaseProcess.md b/docs/ReleaseProcess.md index d32416d..85d37ea 100644 --- a/docs/ReleaseProcess.md +++ b/docs/ReleaseProcess.md @@ -291,7 +291,7 @@ store, or rotate. The permission alone is not enough, though: `draft-release.yml` still passes `GH_TOKEN: ${{ github.token }}` explicitly, at the workflow level. An environment variable set on a *nested* composite action's own step — -`draft-release` calling `ai-inference` — does not reliably reach the +`release/draft` calling `ai-inference` — does not reliably reach the action it calls; only one set by the calling workflow cascades down through every layer, which is what actually lets the Copilot CLI, several layers deep, see the token. @@ -314,7 +314,7 @@ and the release drafting — lives in the separate [`TaffarelJr/.actions`][actionsRepo] repo as composite actions, and is shared by every consumer, `.github` included. This repo's own `draft-release.yml` is a shell: -it checks out the history and calls the `draft-release` action, +it checks out the history and calls the `release/draft` action, and is identical in every repo in the chain. A layer contributes two things, both in its own workflows: diff --git a/scripts/Common-GitHub.psm1 b/scripts/Common-GitHub.psm1 index ea52c6f..b85e0ad 100644 --- a/scripts/Common-GitHub.psm1 +++ b/scripts/Common-GitHub.psm1 @@ -756,7 +756,7 @@ function Get-ManualCiWorkflow { Steps = @( "Build and test this repo's own code." "Produce a version.txt and upload a 'packages' artifact -" - "or whatever names draft-release's build-workflow/artifact-name." + "or whatever names release/draft's build-workflow/artifact-name." 'Via a short-lived branch and PR; rulesets require it.' ) }