From d07dc5aad632b6d8a69c0bac317ffd562ba8678c Mon Sep 17 00:00:00 2001 From: jochen Date: Fri, 20 Nov 2026 23:17:07 +0100 Subject: [PATCH] Design 47 H: H-A-212, the re-pin revision carries entries; the runner keeps starting a publish's pipelines --- 03-DESIGN/01-to-be/47-milestones/H-lead-answers.md | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/03-DESIGN/01-to-be/47-milestones/H-lead-answers.md b/03-DESIGN/01-to-be/47-milestones/H-lead-answers.md index f239e7ce..03132565 100644 --- a/03-DESIGN/01-to-be/47-milestones/H-lead-answers.md +++ b/03-DESIGN/01-to-be/47-milestones/H-lead-answers.md @@ -1151,3 +1151,13 @@ Where two preparers disagreed, the chosen one is named. revision" by the revision number, never by the source (a no-op where they were equal); the take-in again of H-A-210 is kept the same way (its call "take-in again of "). H-A-210's "no revision" is amended so. P1 (one truth: every fingerprint has its entry in a kept revision), P4. +- **H-A-212 (the re-pin revision carries entries; the runner keeps starting a publish's pipelines; H-A-211 amended, + 23:17)** Met in the build of H-A-211: the runner already starts a publish's pipelines itself on package-published + (start.go Published, movers read from its catalog half, keyed by branch, source, artifact and version), and keys a + take-in's pipeline by source. So the indexer says no take-in.taken for a re-pin: it keeps the re-pinned entries as + revision N+1 of the same source S on every watched branch, with the revision's call naming the publish + ("publish of "), and says catalog-revision; the runner's Published reads its movers from the half at + a revision whose call names that publish (it waits for it as it waits for a take-in's revision), so the pipeline + carries the re-pinned fingerprints; pipeline keys stay as they are. The take-in again (H-A-210) keeps its revision + the same way with the call "take-in again of " and the runner treats it as a publish's revision for + every mover whose fingerprint moved. P1, P4, P7.