Skip to content

ci: bring the lint gate configs back to current canon - #2

Merged
bryanfawcett merged 1 commit into
mainfrom
ci/normalise-lint-config
Sep 11, 2026
Merged

ci: bring the lint gate configs back to current canon#2
bryanfawcett merged 1 commit into
mainfrom
ci/normalise-lint-config

Conversation

@bryanfawcett

Copy link
Copy Markdown
Contributor

Config only — no source file and no lint result changes. The canonical
files in nyuchi/.github have moved since this repo adopted the gate, and
these are the files that had drifted:

  • .DS_Store
  • .yamllint.yaml

What changed upstream, and why each one matters here:

  • .yamllint.yaml now ignores lockfiles (pnpm-lock.yaml,
    package-lock.json, yarn.lock and their **/ forms) and Rails fixtures
    (test/fixtures/). Both are generated or templated rather than
    authored: pnpm copies registry deprecation notices into a lockfile
    verbatim at lengths nobody can rewrap, and a Rails fixture is an ERB
    template that is only valid YAML after Rails renders it.
  • .github/workflows/lint.yml now carries merge_group:, so required
    checks report on the gh-readonly-queue/** ref in any repo whose
    ruleset uses a merge queue — without it a queued PR stalls for the
    full check_response_timeout and nothing can land. It also widens the
    push trigger to branches: [main, master, scaffold], covering every
    default-branch name in the estate; a repo not listed there fires
    nothing on merge, so the five contexts never reach its
    default-branch head.

Neither applies to this repo today. They are adopted anyway because the
whole point of the canonical files is that they are byte-identical
everywhere — a repo that is "the same except" is how the estate drifted
in the first place.

Re-verified with CI's exact pinned versions (actionlint 1.7.12,
prettier 3.9.4, markdownlint-cli2 0.23.2, yamllint 1.38.0): all five
still clean.

Co-Authored-By: Claude Opus 5 (1M context) noreply@anthropic.com

Config only — no source file and no lint result changes. The canonical
files in nyuchi/.github have moved since this repo adopted the gate, and
these are the files that had drifted:

  - .DS_Store
  - .yamllint.yaml

What changed upstream, and why each one matters here:

  - .yamllint.yaml now ignores lockfiles (pnpm-lock.yaml,
    package-lock.json, yarn.lock and their **/ forms) and Rails fixtures
    (test/fixtures/). Both are generated or templated rather than
    authored: pnpm copies registry deprecation notices into a lockfile
    verbatim at lengths nobody can rewrap, and a Rails fixture is an ERB
    template that is only valid YAML after Rails renders it.
  - .github/workflows/lint.yml now carries `merge_group:`, so required
    checks report on the gh-readonly-queue/** ref in any repo whose
    ruleset uses a merge queue — without it a queued PR stalls for the
    full check_response_timeout and nothing can land. It also widens the
    push trigger to `branches: [main, master, scaffold]`, covering every
    default-branch name in the estate; a repo not listed there fires
    nothing on merge, so the five contexts never reach its
    default-branch head.

Neither applies to this repo today. They are adopted anyway because the
whole point of the canonical files is that they are byte-identical
everywhere — a repo that is "the same except" is how the estate drifted
in the first place.

Re-verified with CI's exact pinned versions (actionlint 1.7.12,
prettier 3.9.4, markdownlint-cli2 0.23.2, yamllint 1.38.0): all five
still clean.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@bryanfawcett
bryanfawcett enabled auto-merge (rebase) September 11, 2026 19:45
@bryanfawcett
bryanfawcett merged commit 5c7338f into main Sep 11, 2026
6 checks passed
@bryanfawcett
bryanfawcett deleted the ci/normalise-lint-config branch September 11, 2026 19:45
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant