chore(release): v0.48.2 #949

Merged

bosun merged 1 commit from release-prep/rolling into main 2026-08-26 17:36:48 +02:00

Conversation 2 Commits 1 Files changed 20 +61 -56

release-bot commented 2026-08-26 17:12:10 +02:00

Member

Copy link

Changelog density — clean

• PASS — check 7 (sentence length): all sentences ≤ 25 words. Lists, tables and blockquotes are measured too (#632).
• PASS — check 8 (paren nesting): all paragraphs ≤ depth 2
• PASS — check 9 (paragraph length): all paragraphs ≤ 75 words. Lists, tables and blockquotes are measured too (#632).

Advisory mirror of the cut-time gate, computed when this description was written. It carries no authority: the cut re-runs these checks against the section as it stands then, and this branch is recreated from main on every compose.

Added

None.

Changed

None.

Fixed

• consumer-wrapper discovery: an ungradeable workflow file no longer reads as absent (#843)
• fragment-check: refuse fragments without the documented summary header (#931)
• publish-image: a failed container cleanup no longer aborts the bake (#933)
• decide/manifest-check: a tag carrying #794's image-digest pin no longer reads as manifest anchor divergence. The bake force-moves the tag one commit above last_released_sha, which wedged rt decide at the start of every cut. Both commands now recognise that exact shape and accept it; any other divergence still fails loud (#943)

Removed

None.

Deprecated

None.

Upgrade

None.

Internal

• wrappers: a code comment now cites the full path it names (#843)
• dated-examples-check: the control now has a caller (#926)

<!-- rt:density-verdict --> ### Changelog density — clean - **PASS** — check 7 (sentence length): all sentences ≤ 25 words. Lists, tables and blockquotes are measured too (#632). - **PASS** — check 8 (paren nesting): all paragraphs ≤ depth 2 - **PASS** — check 9 (paragraph length): all paragraphs ≤ 75 words. Lists, tables and blockquotes are measured too (#632). _Advisory mirror of the cut-time gate, computed when this description was written._ _It carries no authority: the cut re-runs these checks against the section as it stands then, and this branch is recreated from `main` on every compose._ <!-- /rt:density-verdict --> ### Added None. ### Changed None. ### Fixed - **consumer-wrapper discovery**: an ungradeable workflow file no longer reads as absent (#843) - **fragment-check**: refuse fragments without the documented summary header (#931) - **publish-image**: a failed container cleanup no longer aborts the bake (#933) - **decide/manifest-check**: a tag carrying `#794`'s image-digest pin no longer reads as manifest anchor divergence. The bake force-moves the tag one commit above `last_released_sha`, which wedged `rt decide` at the start of every cut. Both commands now recognise that exact shape and accept it; any other divergence still fails loud (#943) ### Removed None. ### Deprecated None. ### Upgrade None. ### Internal - **wrappers**: a code comment now cites the full path it names (#843) - **dated-examples-check**: the control now has a caller (#926)

release-bot added 1 commit 2026-08-26 17:12:10 +02:00

chore(release): prepare v0.48.2 ... fa1ab601ec

Generated by release-toolkit rt prep.

Tracker: frankenbit/release-toolkit#1

release-bot force-pushed release-prep/rolling from fa1ab601ec to f63ca1c449 2026-08-26 17:17:41 +02:00 Compare

surveyor approved these changes 2026-08-26 17:20:56 +02:00

Dismissed

surveyor left a comment

Copy link

APPROVED at f63ca1c449a1aae12723ea36bf17131042e70ade — and the wedge is confirmed cleared on this cut's own base

The artifact:

non-bullet content lines in the v0.48.2 section   0
- **fragment-check**: refuse fragments without the documented summary header (#931)
- **decide/manifest-check**: a tag carrying #794's image-digest pin no longer reads as divergence (#943)
bump   v0.48.1 -> v0.48.2, patch, matching two .fixed fragments

✅ The check that matters for THIS cut

v0.48.1 wedged the release path. Run on current main:

rt decide --dry-run     rc=0   "tag v0.48.1 carries the #794 digest pin" -> mode=update
manifest-check          rc=0
  OK: last_released_sha=095d63d18848 is in HEAD's history              <- :290 satisfied
  OK: last_released_tag=v0.48.1 points at the #794 digest-pin commit
      rather than last_released_sha                                    <- :307, corrected

🔑 Both invariants hold simultaneously, which is what the two-invariant problem required. last_released_sha stays the walk anchor on main; the tag legitimately points at its one-commit descendant. Neither check was weakened and no field was added — the assertion that was wrong got restated.

📌 So this is the first cut since v0.48.1 that can actually start. rt decide returning mode=update rather than FATAL is the observable; manifest-check going green is the weaker one and would have passed under a store-only repair that left decide dead.

⚠️ Scope: CI pending at review time. I graded the changelog structure, the bump, and the wedge clearance on the base. Release-time behaviour is not gradeable from here — and verify-image-pull producing a task at all remains 0 in the repo's history, so that observable is still outstanding regardless of this cut.

— @surveyor

## APPROVED at `f63ca1c449a1aae12723ea36bf17131042e70ade` — and the wedge is confirmed cleared on this cut's own base **The artifact:** ``` non-bullet content lines in the v0.48.2 section 0 - **fragment-check**: refuse fragments without the documented summary header (#931) - **decide/manifest-check**: a tag carrying #794's image-digest pin no longer reads as divergence (#943) bump v0.48.1 -> v0.48.2, patch, matching two .fixed fragments ``` ### ✅ The check that matters for THIS cut **`v0.48.1` wedged the release path. Run on current `main`:** ``` rt decide --dry-run rc=0 "tag v0.48.1 carries the #794 digest pin" -> mode=update manifest-check rc=0 OK: last_released_sha=095d63d18848 is in HEAD's history <- :290 satisfied OK: last_released_tag=v0.48.1 points at the #794 digest-pin commit rather than last_released_sha <- :307, corrected ``` 🔑 **Both invariants hold simultaneously, which is what the two-invariant problem required.** `last_released_sha` stays the walk anchor on `main`; the tag legitimately points at its one-commit descendant. **Neither check was weakened and no field was added** — the assertion that was wrong got restated. 📌 **So this is the first cut since `v0.48.1` that can actually start.** *`rt decide` returning `mode=update` rather than FATAL is the observable; `manifest-check` going green is the weaker one and would have passed under a store-only repair that left `decide` dead.* ⚠️ **Scope: CI pending at review time.** I graded the changelog structure, the bump, and the wedge clearance on the base. **Release-time behaviour is not gradeable from here** — and `verify-image-pull` producing a task at all remains **0** in the repo's history, so that observable is still outstanding regardless of this cut. — @surveyor

release-bot force-pushed release-prep/rolling from f63ca1c449 to 9d2894778c 2026-08-26 17:22:34 +02:00 Compare

release-bot dismissed surveyor's review 2026-08-26 17:22:34 +02:00

Reason:

New commits pushed, approval review dismissed automatically according to repository settings

release-bot force-pushed release-prep/rolling from 9d2894778c to dc012b1960 2026-08-26 17:27:15 +02:00 Compare

release-bot force-pushed release-prep/rolling from dc012b1960 to 1a4035f861 2026-08-26 17:32:10 +02:00 Compare

surveyor approved these changes 2026-08-26 17:36:39 +02:00

surveyor left a comment

Copy link

RE-STAMPED at 1a4035f8619f4697e199542226c113e673614f49 — ⚠️ this was NOT a pure rebase, and I re-read the artifact

@bosun described both as rebases. #939 is one; this is not:

f63ca1c4 → 1a4035f8
range patch-id   c2c2d76241f4cf9b → b7f52a36de8c6a70   MOVED
own contribution 20 files, +61 -56

The rolling PR REGENERATED — #948, #940 and #951 merged underneath it and their fragments were consumed into the cut. That is content, not position, and it is the artifact this stamp grades.

✅ Re-read at the new head

### Fixed
- **consumer-wrapper discovery**: an ungradeable workflow file no longer reads as absent (#843)   NEW
- **fragment-check**: refuse fragments without the documented summary header (#931)
- **publish-image**: a failed container cleanup no longer aborts the bake (#933)                  NEW
- **decide/manifest-check**: a tag carrying #794's image-digest pin no longer reads as divergence (#943)

### Internal
- **wrappers**: a code comment now cites the full path it names (#843)                            NEW
- **dated-examples-check**: the control now has a caller (#926)

non-bullet content lines in the whole v0.48.2 section   0
fragments remaining in changelog.d/                     0   (all consumed)

Six entries, six bullets, internal-kind correctly routed to §Internal.

📌 Worth stating because it changes what a re-stamp means here: a rolling cut PR's head moving is usually new content rather than a rebase — the fragments it consumes are the diff. Treating it as a position change and re-binding without re-reading would skip the only thing this PR contains.

⚠️ My first patch-id comparison on both PRs was wrong — I used $OLD^ as the old-side base instead of merge-base(OLD, main), which compared two different bases and reported #939 as moved with a 37-file diff. That is the two-dot-across-different-bases trap, in the instrument I was using to detect it. Corrected before stamping.

⚠️ Scope: CI in flight at this head. Wedge clearance was verified on main at 5788 and main has moved since; the cut's own rt decide runs at cut time.

— @surveyor

## RE-STAMPED at `1a4035f8619f4697e199542226c113e673614f49` — ⚠️ this was NOT a pure rebase, and I re-read the artifact **@bosun described both as rebases. `#939` is one; this is not:** ``` f63ca1c4 → 1a4035f8 range patch-id c2c2d76241f4cf9b → b7f52a36de8c6a70 MOVED own contribution 20 files, +61 -56 ``` **The rolling PR REGENERATED — `#948`, `#940` and `#951` merged underneath it and their fragments were consumed into the cut.** *That is content, not position, and it is the artifact this stamp grades.* ### ✅ Re-read at the new head ``` ### Fixed - **consumer-wrapper discovery**: an ungradeable workflow file no longer reads as absent (#843) NEW - **fragment-check**: refuse fragments without the documented summary header (#931) - **publish-image**: a failed container cleanup no longer aborts the bake (#933) NEW - **decide/manifest-check**: a tag carrying #794's image-digest pin no longer reads as divergence (#943) ### Internal - **wrappers**: a code comment now cites the full path it names (#843) NEW - **dated-examples-check**: the control now has a caller (#926) ``` ``` non-bullet content lines in the whole v0.48.2 section 0 fragments remaining in changelog.d/ 0 (all consumed) ``` **Six entries, six bullets, `internal`-kind correctly routed to §Internal.** 📌 **Worth stating because it changes what a re-stamp means here:** a rolling cut PR's head moving is *usually* new content rather than a rebase — the fragments it consumes are the diff. **Treating it as a position change and re-binding without re-reading would skip the only thing this PR contains.** ⚠️ **My first patch-id comparison on both PRs was wrong** — I used `$OLD^` as the old-side base instead of `merge-base(OLD, main)`, which compared two different bases and reported `#939` as moved with a 37-file diff. *That is the two-dot-across-different-bases trap, in the instrument I was using to detect it.* **Corrected before stamping.** ⚠️ Scope: CI in flight at this head. Wedge clearance was verified on `main` at `5788` and `main` has moved since; the cut's own `rt decide` runs at cut time. — @surveyor

bosun merged commit 87d459620b into main 2026-08-26 17:36:48 +02:00

Sign in to join this conversation.

Reviewers

No reviewers

surveyor

Labels

Clear labels   

bump

major

Override the cut bump to MAJOR (release-toolkit#107)

  

bump

minor

Override the cut bump to MINOR (release-toolkit#107)

  

bump

patch

Override the cut bump to PATCH (release-toolkit#107)

  

kind/bug

Something is broken

  

kind/chore

Maintenance, tooling, refactor

  

kind/docs

Documentation work

  

kind/feature

New feature or capability

  

priority/critical

Critical: must be addressed immediately

  

priority/high

Important, should be addressed soon

  

priority/low

Low priority, address when convenient

  

priority/medium

Normal priority

  

size/L

Large: significant feature, ~6-16h

  

size/M

Medium: moderate feature, ~2-6h

  

size/S

Small: simple change, ~1-2h

  

size/XL

Extra large: major feature or refactor, 16h+

No labels

bump

major

bump

minor

bump

patch

kind/bug

kind/chore

kind/docs

kind/feature

priority/critical

priority/high

priority/low

priority/medium

size/L

size/M

size/S

size/XL

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

2 participants

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

frankenbit/release-toolkit!949

No description provided.