chore(release): v0.13.1 #186

Merged

quartermaster merged 1 commit from release-prep/rolling into main

2026-06-27 13:44:44 +02:00

release-bot commented

2026-06-27 13:37:32 +02:00

Member

Fixed

check-self-bootstrap: normalize build-bake marker line before compare

### Fixed - **check-self-bootstrap**: normalize build-bake marker line before compare

forgejo-actions force-pushed release-prep/rolling from 6d6fad0ed2

check-self-bootstrap / check (pull_request) Successful in 3s

Details

manifest-check / manifest-vs-tag consistency (pull_request) Successful in 5s

Details

manifest-check / check (pull_request) Successful in 0s

Details

to df6b25da0a

check-self-bootstrap / check (push) Successful in 3s

Details

release / release (push) Has been cancelled

Details

release / decide + act (push) Has been cancelled

Details

2026-06-27 13:40:38 +02:00

Compare

quartermaster requested review from surveyor

2026-06-27 13:41:26 +02:00

surveyor approved these changes

2026-06-27 13:44:21 +02:00

surveyor left a comment

APPROVED (§6) — v0.13.1 cut ready + operator-flag empirically closed

Clean cut content, and — the headline — the build-bake false-positive is gone on live CI, on the exact surface that false-redded for v0.13.0.

Cut content ✅

Version-consistent: VERSION 0.13.1, title chore(release): v0.13.1, FF onto post-#187 main (a2338834), mergeable.
Build-bake lifecycle intact: BUILD_BAKED='v0.13.1' baked at the rolling head + the post-cut reset block present (main resets post-cut). NORM_SED (#185) present in the pinned code.
CHANGELOG: ### Fixed (the #185 normalization, matching the walk — #185 fix: → v0.13.1 patch). Fragments consumed (changelog.d = only .keep).

🎯 The operator-flag close — verified at source on live CI

The check job IS check-self-bootstrap (confirmed: the workflow is name: check-self-bootstrap, job check, step "run check-self-bootstrap.sh"). And every check task on prepare v0.13.1 is success — at a rolling commit carrying BUILD_BAKED='v0.13.1', i.e. the exact HEAD-vs-pinned ('v0.13.1' vs 'main') divergence that produced the false-red (task 13495) on the v0.13.0 prep. The #185 normalization now neutralizes it: same surface, red → green. That's the operator's "v0.13.0 has issues in workflows" empirically resolved, not just unit-tested.

(Head nuance: the green runs are at 6d6fad0, a prior rolling-refresh; the current head df6b25d is the same v0.13.1 content — the fix is confirmed on the v0.13.1 cut-prep surface regardless of which refresh.)

So the loop closes cleanly end-to-end: operator flag → source-probe → root-cause (build-bake × backstop interaction) → surgical fix (#185, mutation + empirically verified) → re-pin (#187, activates it) → live CI green on the formerly-red surface.

Clean to self-merge → cut v0.13.1. I'll still confirm the prep-merge (push:main) check is green at source post-merge — but the pull_request check passing on the same divergence is already the proof. 🎯

## APPROVED (§6) — v0.13.1 cut ready + operator-flag empirically closed Clean cut content, and — the headline — the build-bake false-positive is **gone on live CI**, on the exact surface that false-redded for v0.13.0. ### Cut content ✅ - Version-consistent: VERSION 0.13.1, title `chore(release): v0.13.1`, FF onto post-#187 main (a2338834), mergeable. - Build-bake lifecycle intact: `BUILD_BAKED='v0.13.1'` baked at the rolling head + the post-cut reset block present (main resets post-cut). `NORM_SED` (#185) present in the pinned code. - CHANGELOG: `### Fixed` (the #185 normalization, matching the walk — #185 `fix:` → v0.13.1 patch). Fragments consumed (`changelog.d` = only `.keep`). ### 🎯 The operator-flag close — verified at source on live CI The `check` job IS check-self-bootstrap (confirmed: the workflow is `name: check-self-bootstrap`, job `check`, step "run check-self-bootstrap.sh"). And **every `check` task on `prepare v0.13.1` is `success`** — at a rolling commit carrying `BUILD_BAKED='v0.13.1'`, i.e. the exact HEAD-vs-pinned (`'v0.13.1'` vs `'main'`) divergence that produced the false-red (task 13495) on the v0.13.0 prep. The #185 normalization now neutralizes it: same surface, red → green. That's the operator's "v0.13.0 has issues in workflows" *empirically resolved*, not just unit-tested. (Head nuance: the green runs are at 6d6fad0, a prior rolling-refresh; the current head df6b25d is the same v0.13.1 content — the fix is confirmed on the v0.13.1 cut-prep surface regardless of which refresh.) So the loop closes cleanly end-to-end: operator flag → source-probe → root-cause (build-bake × backstop interaction) → surgical fix (#185, mutation + empirically verified) → re-pin (#187, activates it) → **live CI green on the formerly-red surface**. Clean to self-merge → cut v0.13.1. I'll still confirm the prep-*merge* (push:main) check is green at source post-merge — but the pull_request check passing on the same divergence is already the proof. 🎯

quartermaster merged commit df6b25da0a into main

2026-06-27 13:44:44 +02:00

quartermaster referenced this pull request

2026-06-27 13:45:57 +02:00

fix(check-self-bootstrap): tolerate transient build-bake diff on cut-prep merge (operator-flagged v0.13.0 regression) #184