Conversation
…uffix Found during v4.6.0's post-release surface verification: brew list --versions reports the Cellar directory name, which carries a trailing _N revision suffix whenever the formula was rebuilt without a version bump (e.g. "4.6.0_1"). resolve_brew() returned that raw string, and add_leg()'s exact-equality check against plugin.json's bare "4.6.0" then blocked every release with a false MISMATCH -- this would have recurred on every future release, not just this one. Strip the suffix in resolve_brew() (both the real brew path and the SURFACES_BREW_VERSION test override) so the comparison is apples to apples. New positive-control test confirms it fails without the fix (exit 1, false BLOCKED) and passes with it (exit 0, ALIGNED). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
….yml) Post-release audit found the exact prose-drift bug class this release just hardened craft's own tooling against, live in craft's own docs: - docs/index.md's "Latest" info-box and README.md's top callout both had their version NUMBER bumped by bump-version.sh, but the accompanying HEADLINE/description text still described v4.5.0's content (repo-triage skill) rather than v4.6.0's -- bump-version.sh only touches version/count strings, never semantic headline prose, so this goes stale every release unless someone catches it by hand. - mkdocs.yml's site_description had ridden along even longer: its version number kept getting bumped mechanically across 4 releases while the described feature (/craft:finish rename) is actually from v4.2.0. All three now describe v4.6.0's real content. Neither docs-staleness- check.sh nor pre-release-check.sh catches this class (they check counts/dates, not headline-vs-version content matching) -- a real gap, not something this commit claims to close. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Post-release audit of v4.6.0 found two issues, both fixed and verified:
README.md,docs/index.md, andmkdocs.yml— the exactprose-drift bug class v4.6.0 just hardened craft's own tooling against, found live in craft's
own docs.
bump-version.shbumps version numbers and counts, never semantic headline text, sothe "Latest" callouts kept describing v4.5.0's content (or, in mkdocs.yml's case, v4.2.0's)
after each mechanical version bump. All three now describe v4.6.0's real content.
verify-surfaces.shfalse-positived on Homebrew's Cellar revision suffix —brew list --versionsreports4.6.0_1(a rebuild-without-bump suffix), compared via exact stringequality against plugin.json's bare
4.6.0. Would have blocked every future release's surfaceverification. Fixed with a new positive-control test (fails without the fix, passes with it).
Test plan
docs-staleness-check.sh— GREENmkdocs build --strict— cleantests/test_verify_surfaces.sh— 88/88 passed (86 pre-existing + 2 new), including the newCellar-suffix regression test verified via revert-and-retest (fails without the fix for the
documented reason, passes with it)
🤖 Generated with Claude Code