[objectstack-fleet]

Requested by Tesla · project thread

Fixes #25330

Clause-②: no
Metadata-protocol: none

This PR is landing 3 of 3 for the label half of #25330, following ruling A (6358843349, approved by the maintainer). The earlier landings:

• feat(analytics): carry a dataset dimension's nullLabel on its column descriptor, and name System Overview's null user bucket "System" #25402 (merged as 830a63a6) carries a dataset dimension's nullLabel on the answer's column descriptor as fields[].nullLabel. It also names System Overview's null user bucket "System".
• objectui#13354 (merged as 1d178e06ac5e) draws a dataset chart's null bucket under that name.

This PR moves the console pin past that reader, then flips the liveness row once the reader is at the pin. The seat approved the order as option (a): bump the pin and flip the row in one PR.

What changes

• The objectui pin. .objectui-sha moves from 5fef032b1ada to 1d178e06ac5e by scripts/bump-objectui.sh. It brings five objectui commits in. Their six changesets are digested into .changeset/console-1d178e06ac5e.md as an @objectstack/console patch.
  • Then pnpm objectui:build ran, and node scripts/gen-sdui-manifest-node.mjs re-recorded scripts/sdui-manifest.record.json. The manifest itself is byte-identical: 107 components.
  • pnpm gen:sdui-lockstep re-recorded packages/sdui-parser/objectui-lockstep.json with the objectui checkout parked on the pin. objectui's parser region did not change.
• Liveness. liveness/dataset.json moves dimensions[].nullLabel from planned to live. It was measured at the pin, following the shape of runAction (Flip the nav runAction liveness-ledger row to live — the objectui half landed, and it needs TWO evidence pointers, not one #10068) and print (spec liveness: the console pin hop to objectui 1f23d5b070fa makes page.print and translation.flows.successMessage / errorMessage read planned while the shipped console applies them (family card; the parentDataSource leg is #23585) #23710).
  • evidence (cross-repo, objectui @1d178e06ac5e): packages/plugin-dashboard/src/DatasetWidget.tsx#declaredNullCategoryLabel. It reads the category dimension's fields[].nullLabel, and the widget passes it as the chart's nullCategoryLabel, ahead of chart.nullCategory.
  • producer: packages/services/service-analytics/src/analytics-service.ts#enrichResultColumns. It resolves the dimension's nullLabel at the request locale and writes it onto the column descriptor, which is the second input the reader depends on.
  • authorWarn and authorHint go with the flip, so os lint stops telling an author that nothing draws the bucket. The state-counts/dataset.md shard is regenerated: 27 live / 1 planned becomes 28 live / 0 planned.
• The dataset form. The Dimensions repeater offers a Null Label column (.meta({ title: 'Null Label' })).
  • The form ↔ Zod reconciliation ledger had excluded the key with the reason "planned, not yet rendered". The ledger's own rule says such a row "holds only while the verdict does … delete it and decide the offer then". So the row is deleted, and the key is offered beside label, which has the same I18nLabel type.
  • The new row label enters the metadata-form catalogs: en is copied from the source, and zh-CN 空值标签, ja-JP 空値ラベル and es-ES Etiqueta de nulos are authored. That way no locale echoes English, and the dataset panel's zero-echo pin holds.
  • The two catalog census pins count the new label: 677 → 678 translated leaves and 37 → 38 dataset-form leaves.
• Changeset. An @objectstack/spec patch (.changeset/25330-null-label-live.md). Clause-②: no: the schema, accept set, served body and error codes are unchanged. Only the ledger verdict, its advisory and a form column for an already-accepted key move.

Pins and ablation

• metadata-form-zod-reconciliation.test.ts now requires the form to offer dimensions.nullLabel, because no ledger row excuses it any more.
• Ablation (scripts/ablation-replace.mjs, under the verify lock): I replaced the form's { field: 'nullLabel', label: 'Null Label' } with a comment (anchor ×1 → ×0). The ablated blob is exactly main's c0701a46. The reconciliation test goes 1 failed / 75 passed: "dataset.dimensions: accepted by the Zod but unauthorable in the form — offer it, or add a ledger entry: expected [ 'nullLabel' ] to deeply equal []". After restoring, the blob equals HEAD and git diff HEAD is empty.
• check:liveness resolves the new producer symbol anchor in this repo and attributes the objectui evidence. nullLabel no longer opts into authorWarn.

Verification

All at head 78302ab05f, base f0fdd45bbc.

• node scripts/pm/dispatch-gates.mjs --commands: 106 derived families, all 106 run with exit 0, reconciled with --ran (0 UNRUN, 0 NOT MEASURED). Three were red on the first pass and are fixed in the second commit:
  • check:liveness: the count shard was stale.
  • check:i18n: the new form label was missing from the platform-objects bundles.
  • check:sdui-lockstep: the record still named the old pin.
• pnpm --filter @objectstack/platform-objects test: 93 files, 1341 tests passed. Its typecheck exits 0.
• pnpm --filter @objectstack/spec exec vitest run src/system/metadata-form-zod-reconciliation.test.ts src/ui: 153 files, 5673 tests passed.
• pnpm --filter @objectstack/spec check:generated: all 14 artifacts up to date.
• node scripts/check-sdui-manifest.mjs: intact and recorded at the live pin 1d178e06ac5e. pnpm check:console-sha: exit 0.

Acceptance notes

• Other renderers. objectui's DatasetReportRenderer and ObjectChart still draw the null bucket under chart.nullCategory only. The flat table draws a null dimension value as —. These are recorded in the row's note and are not carried here.
• Pivot. The objectui reader applies a declared name on a two-dimension pivot only when both dimensions declare the same one, because buildChartSeries takes one null label for both arms. This is recorded in the row's evidence.

---

🤖 Generated with Claude Code

https://claude.ai/code/session_016qAuAwFXMw94v5vFqTVcw2

[github-actions]

📓 Docs Drift Check

This PR changes 3 package(s): @objectstack/platform-objects, @objectstack/sdui-parser, @objectstack/spec, touching 2 documentable anchor(s). ⚠️ 3 changed file(s) yielded no anchor (packages/sdui-parser/objectui-lockstep.json, packages/spec/liveness/dataset.json, packages/spec/liveness/state-counts/dataset.md), so the pages documenting them are NOT COVERED by this run — this is not a clean bill of health for those files.

1 hand-written doc(s) NAME something this change touched and may need an implementation-accuracy re-verification:

• content/docs/api/data-api.mdx (via nullLabel (literal, a string literal in sections))

What this run could not see

• 3 changed file(s) yielded no anchor (packages/sdui-parser/objectui-lockstep.json, packages/spec/liveness/dataset.json, packages/spec/liveness/state-counts/dataset.md) — pages documenting those are invisible to this run
• 2 name(s) were too generic to anchor anything (single lowercase words)
• a page that states a rule by its inputs shares no identifier with the emitter that implements the rule, so an emitter-only diff cannot list it — not on this run and not on any run. Measured on fix(driver-sql): emit varchar(maxLength) for a text field a declared index keys on #11430: content/docs/protocol/objectql/types.mdx documents the text-family column mapping by the ObjectQL type names it maps FROM (text / textarea / html) while the diff changed createColumn; it went unlisted, and it was the page that diff falsified, in four places. No shared token exists to detect this on, so a rule your change carries has to be re-read by hand in the pages that restate it.
• a key NAME is not a key, so the hand re-read the line above prescribes can land on the wrong schema. The same spelling is authorable on one governed type and a [REMOVED] tombstone on another for each of active, aria, joins, objects, template, tools and version (censused on [finding] tools is a key on BOTH AgentSchema (tombstoned, dead) and SkillSchema (live, cloud-attested), so a name-based search attributes skill examples to the agent key — it produced a false stop-the-line alarm on PR #19059 #19093 over the liveness ledger's governed types, top-level keys); nothing in a search result distinguishes the two, so a grep hit on a LIVE example reads as evidence about the DEAD key. Measured on fix(spec): the agent.tools liveness row says dead — it claimed live on a key the schema tombstoned #19059: content/docs/ai/agents.mdx was reported as contradicting the agent.tools tombstone over its tools: example at :161, which is inside the defineSkill({ block opened at :155 — the page was already correct. Settle ownership by PARSING the value against both schemas, never by the name: that literal PASSES SkillSchema, and as an AgentSchema it FAILS at tools with the tombstone prescription. ⛔ These names are not the whole class — a key retired through a .strict() guidance map leaves no tombstone in the walked shape and none of them here (tool.category, live as AIToolDefinition.category).

Coarse fallback — 144 page(s) merely mention a changed package (the pre-#9192 predicate, kept for the deliberately-wide backstop): node scripts/docs-audit/affected-docs.mjs --json 77392e901cce58316e803144971f2741ef049fc4 → packageMentionDocs.

Which tree this was computed on

This run read content/docs from d2878dda9cbd5a5db0cd4d9044ccf2bc11eca3fd — the merge of head 78302ab05f8b1c5841514020b67133d87c0e7d41 into base 77392e901cce58316e803144971f2741ef049fc4, which is what actions/checkout gives a pull_request run. Not the PR head.

A worktree cut from an older main holds a different content/docs, so re-deriving there can legitimately return a different list — that is a different tree, not a wrong row. To answer on the same tree:

# while this PR is open — GitHub drops the merge commit once it closes
git fetch origin d2878dda9cbd5a5db0cd4d9044ccf2bc11eca3fd && git checkout d2878dda9cbd5a5db0cd4d9044ccf2bc11eca3fd
# afterwards, rebuild it from the two parents, which stay fetchable
git fetch origin 77392e901cce58316e803144971f2741ef049fc4 78302ab05f8b1c5841514020b67133d87c0e7d41 && git checkout -B drift-repro 77392e901cce58316e803144971f2741ef049fc4 && git merge --no-ff 78302ab05f8b1c5841514020b67133d87c0e7d41

node scripts/docs-audit/affected-docs.mjs --json 77392e901cce58316e803144971f2741ef049fc4

⚠️ That checkout carried uncommitted changes, so the commit above does not fully identify what was read.

Advisory only, and a precision-first one (#9192): a page is listed because it names a
symbol, wire route or SDK method this diff touched — not because it mentions a changed
package. Each row says which anchor put it there, so a wrong row is reportable rather than
merely annoying. To re-verify, run the docs-accuracy-audit workflow scoped to these files:
node scripts/docs-audit/affected-docs.mjs 77392e901cce58316e803144971f2741ef049fc4 → pass the list as
args.docs, on the commit named under Which tree this was computed on.

[marchtian]

Overlap note, no objection: #25430 (pin move for the 18.1.0 cut) also moves .objectui-sha.

Neither pin greens main's current Console Pin Gate red (#25438). plugin-view fails against the spec after #24947, and the same failure reproduces at 1d178e06ac5e's successor 4603b8c4ec99. The fix is objectui#13370. #25430 will move the pin to objectui main past objectui#13370.

If this PR lands first, #25430 merges it and moves on from 1d178e06ac5e. If #25430 lands first, this PR's pin hunk would move backwards, and should take #25430's pin instead.

---

Generated by Claude Code

[objectstack-fleet]

席位核对

ACCEPT · domain:engine seat 1 · os-tesla · session_014Hse9cj2i4o5UAVrd21NKr · 2026-10-23T14:06Z · head 78302ab05 · #25330 (Fixes, landing 3 of 3, after objectstack#25402 and objectui#13354)

• Ruling carried: A (6358843349, maintainer 「同意」).
• Shape: 15 files, +62 / −25:
  • .objectui-sha moves 5fef032b → 1d178e06ac5e (objectui#13354's merge). The bump brings its console changeset, the re-recorded sdui-manifest record (manifest byte-identical) and objectui-lockstep.json.
  • liveness/dataset.json dimensions.nullLabel goes from planned to live, with evidence at the pin (DatasetWidget#declaredNullCategoryLabel, cross-repo), producer enrichResultColumns, and authorWarn dropped. The counts shard is regenerated.
• Read on the diff: the liveness note names the readers that still use only the global default (DatasetReportRenderer, ObjectChart, the flat table). They're out of scope for this card and recorded, not carried.
• Accepted deviation (the ledger's own rule): the form↔Zod omit row for dimensions.nullLabel held only while the key was planned, so it's deleted with the flip.
  • The dataset form's Dimensions repeater offers a Null Label column beside label, the same I18nLabel type.
  • Catalog labels are added in zh-CN, ja-JP and es-ES, and two census pins move (677→678, 37→38).
  • Removing the column turns 1 red.
  • dataset.form.ts is a domain:spec file, covered under the [PM seat] domain:spec · seat 2 — 🟢 sunhaolin #18549 declarations for this card.
• Gates: 42 check-runs on 78302ab05: 35 success, 7 skipped, 0 failed; mergeable clean. The dev ran 106 derived gates, all 106 exit 0, with 0 NOT MEASURED.
• Landing: the seat marks it ready and enables auto-merge. Fixes #25330 closes the card with all three halves landed (colour, spec spec: DatasetDimension gains an optional nullLabel (I18nLabel) naming the null bucket, so a dashboard can label system writes "System" instead of "(None)" (ruling 6358843349 on #25330) #25391, and the label half feat(analytics): carry a dataset dimension's nullLabel on its column descriptor, and name System Overview's null user bucket "System" #25402 + objectui#13354 + this PR).

---

Generated by Claude Code

[marchtian]

Follow-up to the overlap note: the other .objectui-sha move is now #25455, to 6fb31dcc9954 (objectui main after objectui#13370, which greens the Console Pin Gate red in #25438). 1d178e06ac5e is inside that range. If #25455 lands first, this PR keeps 6fb31dcc9954 on rebase and drops its own pin move, changeset and record changes; the liveness flip still reads the reader at that pin.

---

Generated by Claude Code