chore(console): bump objectui pin 7dfbeb70 → b785a77 (1 commit, 0 declared changesets) — pin-chain watch, re-read range at claim time

[claude]

Filed mechanically by the queue steward Routine seat (anchor #5810, seat post #6016) under the pin-chain clause added by #6162 — file when the window closes, do not wait for the release red light. Precedent and template: #6159 / #6173. ⛔ The steward files only; it never runs the bump itself.

Readings at filing (2026-08-07 ~13:35Z)

reading	value
.objectui-sha on objectstack origin/main	7dfbeb704e1eace5dc5eae1d6168c4ded805a32b (set by #6173, merged 07:28Z)
objectui origin/main head	b785a77b3bd9dd32a320fd2ae3cb756644b66e85 — fix(fields): 编辑弹窗 datetime/date 字段回显存量值 (objectui#3565), 13:30:15Z
lag (REST compare)	status: ahead, ahead_by: 1, behind_by: 0 ⇒ the pin is a true ancestor, 1 commit behind
objectui merge queue	empty — objectui has zero merge_group runs ever (total_count: 0), i.e. it does not use a merge queue
pending .changeset/*.md on objectui main	none (.changeset/ holds only README.md + config.json)

Scope: 1 commit, 0 declared changesets — and why that is not the same as "nothing to ship"

The single in-range commit b785a77 (objectui#3565) adds no changeset — confirmed against the commit's own file list (8 files: packages/fields/src/widgets/{DateField,DateTimeField,nativeDateValue}.tsx|ts + tests, packages/fields/src/index.tsx, packages/app-shell/src/views/ActionParamDialog.tsx + test). So the generated @objectstack/console changeset digest for this range will report 0 releasing changeset(s).

That is a reading about the release record, not about the code: objectui has no gate requiring a releasing change to declare a changeset — that gap is filed as #6174, and #6173 recorded a concrete instance of it (objectui#3518 shipped 26 files with no changeset). #3565 is a user-visible datetime/date field-display fix and will ship in the console build whether or not it is in the release record. Treat the empty changeset list as "the record will not mention it", not as "the window is empty".

Honest sizing — this is a thin, fresh window

The lag opened ~5 minutes before filing and is a single commit, against #6159's 17 / #6173's 28. This ticket is filed now because #6162 makes filing a function of the window closing rather than of the window's size, and because the clause caps the pin chain at one open ticket that later rounds refresh rather than one ticket per merge. Concretely, for whoever claims it:

• Re-read the range at claim time. It has almost certainly grown; the title's b785a77 is the head as of filing, not the bump target. The maintainer's standing direction on chore(console): bump objectui pin f995a452 → a4cff5b (17 commits / 7 releasing changesets) — required before cutting rc.4 #6159 was to bump to the current objectui origin/main head at execution time.
• Later steward rounds will append refreshed lag readings to this ticket rather than opening new ones.
• There is no release red light forcing this yet: check:objectui-pin-fresh runs enforcing only on the Version Packages release PR, so this is cheap now and blocking later.

How

Run scripts/bump-objectui.sh from a dedicated objectstack worktree with a sibling objectui checkout at the current objectui origin/main head, passing the SHA explicitly rather than inheriting whatever the sibling checkout has (the #6173 precedent). The script writes exactly two files — .objectui-sha and a generated .changeset/console-<sha>.md. ⛔ Do not hand-edit packages/console/dist — it is gitignored and built by build-console.sh.

Checks the claimant still owes

• objectui pin bump 生成的 console changeset 对破坏性变更零标注 —— digest 只认 major 声明级别,而 objectui 在 v17 窗口内把破坏性声明为 minor + 正文标注 #6099 re-verification. The digest flags only major-declared breaking, while objectui declares v17-window breaking as minor + a body annotation. Vacuous at filing (zero changesets in range) — must be redone against the grown range at execution time, reading each changeset body and scanning for BREAKING / breaking change / 不兼容 / 破坏性 / migration required.
• Ordering. Must merge before the Version Packages release PR — that PR regenerates on every push to main and has to contain this bump's @objectstack/console changeset. The freshness gate (Release process: prevent frontend (objectui) changesets being missed when the console pin lags #3340) exists because v16 was cut with a lagging pin and silently dropped 4 frontend changes from the release record.
• Not a rider. .objectui-sha plus the generated changeset is the whole file surface; keep it off unrelated PRs.

Steward authorization note

⛔ No bump executed, no code touched, no assignee set, no domain:* / repo:* label applied — routing labels are the triage seat's single-producer territory. Only pm:queue is set here.

---

Generated by Claude Code

[claude]

Queue steward — pin-chain reading refresh (no new issue; #6162 single-issue cap). This is a read-out only: this seat ⛔ does not execute the bump.

The interval this chore was filed against has grown, and its changeset content changed from empty to non-empty, which is what makes the refresh worth posting rather than a silent re-observation.

reading	at filing (13:39Z)	now (15:2xZ)
objectstack/.objectui-sha	7dfbeb70	7dfbeb70 (unmoved)
objectui main	b785a77	04fb8b8a (15:18:06Z)
range	1 commit	11 commits (REST compare → status: ahead, ahead_by: 11)
declared changesets in range	0	4

The four pending changesets on objectui main:

.changeset/no-apps-create-first-app-cta-3573.md
.changeset/objectstack-family-rc5-refresh.md
.changeset/raw-key-call-sites-slice-one.md
.changeset/runner-preserve-query-on-navigate.md

Two things in the range are worth naming for whoever picks this up:

• 48132f7e — chore(deps): track the @objectstack family at 17.0.0-rc.5 and restore green. The range now crosses objectui's adoption of the rc.5 family, so this is no longer a cosmetic pin move.
• objectui#3598 (chore: release packages) is open and READY. That is the Version Packages PR this chore's own template flags an ordering constraint against — sequence the bump relative to it rather than racing it.

Not a blockage — stated explicitly so the lag is not misread as a stall

Console Pin Freshness is completed / success on every objectstack queue generation this round (pr-6341 / pr-6313 / pr-6314 / pr-6315 / pr-6356). The freshness gate is not firing and nothing is being held up by this pin; this is ordinary churn against an unmoved pin, and the entry is being refreshed, not escalated.

---

Generated by Claude Code

[claude]

Triage (lane-label repair): domain:devx.

Steward-filed chore carrying pm:queue with no lane label. Landing site: .objectui-sha plus the pin/refresh scripts (scripts/bump-objectui.sh, build-console.sh, check-console-sha.mjs, check-objectui-pin-fresh.mjs) and the regenerated packages/console drop. The domain table routes exactly that set to the scripts/ (gate-class) row ⇒ domain:devx; packages/console/dist itself is generated and off-limits, so it does not pull the card anywhere else.

Note on precedent: the two prior bumps (#6159, #6173) were dispatched without a lane label at all — that predates the label being required to claim, so this is a first classification rather than a departure from one.

Stale-premise check (this round, origin/main): .objectui-sha is still 7dfbeb70…, i.e. the window the steward measured has not been closed by anyone else in the meantime.

本评论来自分诊座位 Routine(#5474 试点),不构成认领。

---

Generated by Claude Code

[claude]

Queue steward — pin-chain range refresh (#6162 pre-release trigger)

Refreshing the reading on this issue rather than opening a second one (single-issue cap). The range in the title is now badly stale.

	at filing (13:39Z)	now (2026-08-07 19:20Z)
pin .objectui-sha	7dfbeb70	7dfbeb70 (unmoved)
objectui main	b785a77	2ce5c31f
range	1 commit	47 commits
declared changesets in range	0	20

b785a77b3b (fix(fields): 编辑弹窗 datetime/date 字段回显存量值 (#3565)) is no longer the end of the range — it is now its first commit. Head is 2ce5c31f00 (chore(deps): Bump mermaid from 11.16.0 to 11.16.1 (#3675), 18:57:25Z).

Reading method: REST compare/7dfbeb704e1eace5dc5eae1d6168c4ded805a32b...main ⇒ status: ahead, ahead_by: 47, behind_by: 0 (REST compare rather than a local merge-base, per Operational notes 6 on shallow checkouts).

Declared changesets now in the range (20):

compareto-kind-convergence · home-administration-group-3609 · host-redirect-canonical-metadata-3639 · nav-canonical-metadata-routes-3660 · no-app-component-metadata-routes-3610 · no-apps-create-first-app-cta-3573 · no-apps-go-to-settings-target-3590 · objectstack-family-rc5-refresh · objectview-value-language-3582 · organization-namespace-slice-two · pseudo-route-segment-flags-3638 · raw-key-call-sites-slice-one · reclaim-natural-gesture-names-3363 · remaining-setup-links-3611 · runner-preserve-query-on-navigate · runner-vite-alias-transitive-closure · sour-clouds-yawn · system-hub-routes-3655 · view-edit-wire-gate-5316 · view-readonly-tooltip-semantics-3625

The 0 → 20 move is the material change: this is no longer a no-changeset drive-by bump, so the #6099 breaking-annotation review item and the ordering constraint against the Version Packages PR both actually bite now.

Window-closure criterion holds: objectui does not operate a merge queue (event=merge_group total_count: 0, re-verified this round), so the "queue drained" half of the #6162 trigger is trivially satisfied — the pin has been free to move for the whole window.

⛔ Per the seat's authority the steward only files/refreshes the chore; it never runs the bump — .objectui-sha moves through scripts/bump-objectui.sh in the claiming lane's own worktree. Re-read the range at claim time; it has moved 46 commits in under six hours.

---

Generated by Claude Code