chore(release): v0.61.1 #1303
No reviewers
Labels
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
No milestone
No project
No assignees
2 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
frankenbit/release-toolkit!1303
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "release-prep/rolling"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Changelog density — clean
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
mainon every compose.Added
None.
Changed
None.
Fixed
Removed
None.
Deprecated
None.
Upgrade
None.
Internal
toolkit-selfgate jobs are consolidated into one (#1253).12c2b6c61e13d358adcdHOLD — same shape as
v0.61.0, one release later. Comment, not aREQUEST_CHANGES.Reviewed at
12c2b6c61e64bab86c4b34b269fb457fc3f0c2c6. The cut's own content is correct. It is the base that has moved.#1297and#1298merged twenty minutes after this branch was built. Merged as-is,v0.61.1ships both changes with their release notes deferred to the next version.✅ Content check on what the cut does contain, all clean:
Why a comment again
internal/prep/pr.go:80skips regeneration while a liveREQUEST_CHANGESstands (#1183). The base moved, so the bot's own response is to regenerate and consume both fragments — a rejection would freeze the branch at the state I am objecting to. Nothing is blocking it today; keep it that way.🔑 The part that is new: this is the second consecutive cut with the same defect
v0.61.0had it (1287.fixed.md), was held, regenerated, and shipped complete. This one has it again, with two fragments instead of one.⚠️ It is not a bug in the bot — regeneration is working. It is that NOTHING CHECKS the condition, so whether a release ships complete notes depends on a reviewer noticing.
fragment-checkgrades the fragments in the PR; a fragment that exists only on the base is outside every required context. 28/28 green, complement empty, control fires — twice in a row.✅ The check is mechanical and cheap to state: for a cut PR, every
changelog.d/*entry on the BASE must appear in the set the cut deletes. Empty complement or refuse. I am asking @bosun to file it rather than filing it myself.📌 Until then the disposition is the same as last time and it costs one merge cycle: let the base settle, let the bot regenerate, and I stamp the regenerated head once.
APPROVE —
13d358adcd2eea6d30c36f4f5b4fcd6a6839ea3fRe-reviewed after regeneration. My hold is resolved. I held
12c2b6c6because#1297and#1298merged twenty minutes after that branch was built, leaving1253and1295unconsumed on the base.#1296… sorry — the base settling did it:Content check
⚠️ This stamp asserts the CONTENT is correct at this tree. It does NOT assert CI is green — three
ac-closure-checkcontexts were still running when I filed, and withenable_status_check=trueover 27 required contexts the gate is the enforcement there. Read the required set at merge time.📌 One thing I predicted on the last cut, now confirmed — not a blocker
On
#1286I noted thatdocs/integration.md:28would age:The newest tag is now
v0.61.0, and after this cut it will bev0.61.1. The line is one release behind today and two after merge.✅ The cut is right to leave it alone —
doc_version_refs.gorewritesReplace `vX.Y.Z`and@vX.Y.Z, and this is neither. The distinction it teaches still holds (newest is still a real release; highest-sorting is stillv1.0.0-alpha.0), so only the literal drifts.📌 It is the exact case
#1283's ILLUSTRATIVE disposition exists for, and it has now aged twice in two hours, which is a better argument than the one I made in prose. A one-line marker on that block would end it. Separate change; nothing for this cut to do.📌 Tidying one garbled sentence in
6613above — a stray#1296reference and a visible self-correction survived into the published body. What resolved the hold was the base settling and the bot regenerating, not#1296(which merged earlier and is unrelated). The table under it is correct as written:No conclusion changes; the stamp stands.
13d358adcdc472cbe22fAPPROVE —
c472cbe22fbe75a338eda98c566317a19693b942(re-stamp;6613came unbound when the bot regenerated)The cut's own content is unchanged from the head I graded at
13d358ad. The regeneration rebased onto a main that had gained#1300's ADR amendment; the delta between the two heads is that file and nothing else.📌 This is the second re-stamp on this branch and the third time today a bot regeneration has spent one of mine. Not a complaint — the mechanism is working and each regeneration made the cut more correct. It is the cost of the rule we adopted this morning (grade the content, let the gate grade the contexts), and it is cheaper than the alternative: five stamps died waiting for CI on
#1230before we changed it.⚠️ Same disclosure as before: this asserts the CONTENT at this tree, not that CI is green. Read the required set at merge time.