From c5a246c538d122bae808d7fe45873af76f0131ed Mon Sep 17 00:00:00 2001 From: Bill Denney Date: Sat, 22 Aug 2026 20:49:53 +0000 Subject: [PATCH] Waive the audit on the NEWS entry that records is_border_4()'s removal A new meta-repo audit trigger resolves every sibling-ecosystem `pkg::fn` reference against that package's committed sources, and reports one naming a function that exists nowhere. This NEWS entry names `thinr::is_border_4()` -- correctly, because the entry exists to record that it was removed from `src/thinr_common.h`. This is the canonical waiver case the meta CLAUDE.md describes: naming the problem IS the fix, so the line carries a token rather than being reworded. Rewording it to dodge the pattern would delete the only record of what was removed, which is the entry's whole purpose. The token goes on the line ABOVE, the position the check supports for exactly this reason: the reference sits mid-sentence in a wrapped bullet, and an inline HTML comment would break the bullet. A path escape was considered and rejected. The meta CLAUDE.md says the one existing path escape, `REVIEW-*.md`, "is not a precedent" -- and a NEWS.md exemption would silence every future false claim in this file, not this one true one. Co-Authored-By: Claude Opus 5 --- NEWS.md | 1 + 1 file changed, 1 insertion(+) diff --git a/NEWS.md b/NEWS.md index 44772f8..0debc10 100644 --- a/NEWS.md +++ b/NEWS.md @@ -2,6 +2,7 @@ * Development version bump following the 0.3.0 CRAN release. + * Internal, with no change to any output: the unused `thinr::is_border_4()` helper was removed from `src/thinr_common.h`. Its last callers were the stentiford and pavlidis kernels, dropped 2026-05-20; nothing has referenced