No reviewers
Labels
No labels
bump
major
bump
minor
bump
patch
kind
bug
kind
chore
kind
docs
kind
epic
kind
feature
loop/paused
mirror-drift
priority
critical
priority
high
priority
low
priority
medium
size
L
size
M
size
S
size
XL
status
blocked
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
frankenbit/release-toolkit!1522
Loading…
Reference in a new issue
No description provided.
Delete branch "claude/1499-recover-pr-body-1173"
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?
Removes the stale "Tagged-adopter dependency" paragraph from the recovery (B) PR body, and corrects one word in
docs/integration.md.Refs #1499
What changes
.forgejo/workflows/reusable-recover-pending-cut.ymland its.giteatwin: delete the oneprintfthat told the PR's reviewer #1173 had not landed. Nothing else in the body builder moves. The$REMAININGand could-not-grade disclosures after it are untouched.docs/integration.md: toolkit-self's@mainpin goes through the #456 floating-pin override, not a "local" invocation (that arm handles a./.forgejo/...uses line). The sentence now says so.changelog.d/1499-recover-pr-body-1173.fixed.md: a one-line### Fixednote, because adopters' recovery PRs now read differently.Remove or rewrite
rt. Measured, every one does:@maintakes the #456 override;bootstrap rtsince #1514.docs/integration.md(sincedc9df38), the page that tells adopters what to pin. Whoever reads this PR body is already running a version past the floor.Measured
#1173 is closed (2026-09-05, forge read-back). The file is in
canonicalFiles(internal/bake/marker.go:26) and ingates.DefaultComposeScripts.Baked ref per tag, read at the forge's tag SHAs on 2026-10-02, both twins:
git ls-remotematched the local tags one-for-one, and the contents API atv0.58.0returned blob5cb836d1, the same blob as the local tag.On this head (
8333455), in agolang:1.25container on an LF clone:rt gitea-twin --check:PASS — 11 twin(s) match their source.go test ./...: rc=0, no failing packagebats --print-output-on-failure tests/: 221 tests, rc=0FRAGMENT_CHECK_COVERAGE=content rt fragment-check changelog.d: rc=0, density PASSrt check-self-bootstrap:floating-pin (release-toolkit#456): main matches main by construction, rc=0git grep -n 1173 -- .forgejo .gitea cmd internalon this head returns onlyinternal/bake/marker.go:151andinternal/bake/marker_test.go:102,140, all historical.The workflow diff is two deletions and no additions, so the ASCII-only rule (AGENTS.md section 11) holds.
Self-bootstrap (AGENTS.md section 2)
This touches a file in
gates.DefaultComposeScripts, and section 2 predicts a redcheck-self-bootstrapplus a re-pin. Measured instead: every toolkit-self wrapper pins@main, so the check takes the #456 floating-pin exit and passes (output above). No re-pin is needed. Section 2's text predates the@mainpin; that gap is noted on #1499, not changed here.Review request: verify at source (AGENTS.md section 6)
@bosun, you reviewed #1182, so this one is yours. Please check against the tree, not this body:
printfis the only change in either twin, and the twins still match (rt gitea-twin --check).git show v0.64.0:.forgejo/workflows/reusable-recover-pending-cut.yml | grep build-ref);composite/bootstrap-rt/action.yml:9-11.docs/integration.mdsentence, against the resolve step'sfloating-pin-override (release-toolkit#456)arm.Pre-submit, pin the review to head
8333455d68a3a9cb6945e594e4625a94be661b46withrt review-submit-check(AGENTS.md section 6.2).🤖 Generated with Claude Code
Withdrawn as a duplicate of #1504 (issue #1500). That PR made the same removal first (11:31), and it also leaves a comment in the run block explaining why the body carries no caveat.
This PR's one extra item, the "local
@main" wording indocs/integration.md, is offered on #1504 as a suggestion. The branchclaude/1499-recover-pr-body-1173stays in place for now.Pull request closed