Repository navigation
spec rc.6 vacated WidgetManifest / WidgetSource — the Runtime… prefixes may be reclaimed #4164
Description
Recorded from the @objectstack/spec 17.0.0-rc.6 bump in objectstack#7100. Observation-class: nothing a user hits, no defect — an unlock that would otherwise be forgotten.
@object-ui/types carries RuntimeWidgetManifest and RuntimeWidgetSource under prefixes added purely to avoid colliding with spec exports of the same bare names (the objectui#3074 lesson, pinned in packages/types/src/__tests__/page-nav-misc-spec-parity.test.ts).
rc.6 retired the whole widget-manifest vocabulary from @objectstack/spec/ui — WidgetManifest, WidgetSource, WidgetProperty, WidgetEvent, WidgetLifecycle and their …Schema / …Parsed twins. The tripwire in that test fired exactly as designed, for the second time (objectui#3363 was the first, and reclaimed GestureType / GestureConfig on the same signal).
The reclaim is available, not required, so objectstack#7100's PR did not take it: both names are exported, so renaming them is a cross-package change with its own review, and that card is a global_nav retirement. What that PR did instead was move the two rows into the "spec no longer owns" list, so the vacancy stays pinned — if the spec ever re-publishes either name while objectui holds it, that becomes a live collision and the test goes red.
Whoever picks this up: the precedent to follow is objectui#3363's reclaim, and the reason a reclaim can be declined is written beside PWAOfflineConfig in the same file (there the name has a second claimant, so the prefix is still load-bearing — check that neither of these two does before renaming).
Activity
os-zhuang commentedon Aug 11, 2026
Findings triage (first grading): held (
findingkept).Premise verified @ objectui
origin/mainb42558a: the reclaim is genuinely available and genuinely untaken —RuntimeWidgetManifest/RuntimeWidgetSourcestill exported (packages/types/src/index.ts:1219-1220), and the parity test still carries both rows in the vacancy-pinning "spec no longer owns" section (page-nav-misc-spec-parity.test.ts:564-565) with their rationale entries at:646-647. The pin does exactly what the card says: if spec ever re-publishes either bare name while objectui holds it, the test goes red — so the vacancy cannot be silently lost while this card waits.Why held rather than queued: the reclaim is available, not required — a cross-package rename with its own review, zero user-facing effect, against a deep
pm:queuebacklog. The card + the pinned vacancy together already guarantee the unlock is not forgotten, which was the filing's stated purpose. Restart condition: the next deliberate naming/parity sweep in@object-ui/types, or any PR that touches these exports anyway — follow the objectui#3363 precedent and check neither name has a second claimant (thePWAOfflineConfignote in the same file) before renaming.Dup check: no other card tracks this reclaim (probes: RuntimeWidgetManifest / WidgetSource reclaim across open issues/PRs; objectui#3363 is the closed precedent, objectstack#7100 the source bump).
本评论来自分诊座位 Routine（#5474 试点），不构成认领。
Generated by Claude Code
os-zhuang commentedon Aug 11, 2026
Findings triage (re-verify @
origin/mainf812de6): HOLD —RuntimeWidgetManifest/RuntimeWidgetSourcestill exported (packages/types/src/index.ts:1227-1228, drifted a few lines from the 03:08Z anchor, content unchanged); reclaim still available-not-required, vacancy still pinned by the parity tripwire. Restart conditions from the first grade stand.findingstands.本评论来自分诊座位 Routine（#5474 试点），不构成认领。
Generated by Claude Code
claude commentedon Aug 11, 2026
Findings cadence re-check @
origin/main90e792e: HOLD (findingstays).packages/types/src/widget.ts:54/:80still carriesRuntimeWidgetManifest/RuntimeWidgetSourcewith the objectstack#4115 rename comments; no reclaim has happened. Rename-back is optional polish gated on nothing — stays an observation.本评论来自分诊座位 Routine（#5474 试点），不构成认领。
Generated by Claude Code
os-zhuang commentedon Aug 17, 2026
Findings state conversion (batch 3): closing not planned, type Task.
Grading history: held twice with restart condition "the next deliberate naming sweep" — a human-intent event, not a machine-checkable predicate, so
pm:on-holdhas no legal exit here. Closing per the maintainer's 2026-08-16 direction (work not planned for development closes rather than holds).Nothing is lost by closing: the vacancy is pinned IN-REPO —
page-nav-misc-spec-parity.test.tscarries both names in the "spec no longer owns" list and goes red if spec ever re-publishes them — and the reclaim recipe (objectui#3363 precedent; check thePWAOfflineConfigsecond-claimant note before renaming) is recorded on this card. Reopen is free when someone runs a naming sweep; the tripwire cannot be silently lost either way. Maintainer may veto.本评论来自分诊座位 Routine,不构成认领。
Generated by Claude Code
retire(studio): drop the global_nav designer surfaces with the rc.6 s…
fix(i18n): render I18nLabel objects at the 13 remaining sites (object…