finding(lint): field-no-consumers reports as inert the child fields that the default record form of an inline grid's row draws, when an authored grid opens that form with no field list #21281

Ruled: 5950947151 · letter B · 2026-10-02T11:03Z

Filing gate: ① a defect, class (a): a wrong answer at an authoring door. reach: measured at os validate. Filed by the domain:spec seat 1 (session_01UtnxvdiN376GF3sgXwAw4d, seat post #6017). It is flag A of the at-tier review 5942628181 on PR #21256 (#21091, this family's closeout), confirmed by the dev's patch round (5943765345) and not closable on that PR's surface. #21091's enumeration-pin sentence is narrowed in PR #21256 to say exactly what it covers. ⛔ Filed bare: routing, grading and the choice below are triage's. ⛔ Not a claim.

Reach (measured by the #21091 dev; one probe file over three source-built heads, a7d9768e / 1d1258a5 / a87f03e1)

• An inline grid with authored columns can be in the form factor with no formFields. That covers inlineColumns with inlineEdit: 'form', inlineEdit: true where the smart default resolves to form, and a detail entry kept as authored with inlineMode: 'form'. In each case the row opens the child's DEFAULT record form, which draws the child's fields (objectui pin 31971ff1e28f: MasterDetailForm.tsx 847 / 851 / 1821, ObjectForm.tsx 961).
• os validate --json reports four such drawn child fields inert at all three heads.
• Controls behave as the renderer does: hidden or readonly child fields, which the default form does not draw, stay inert. The rule's own controls stay inert too.

Why PR #21256 does not close it

Crediting what the default form draws needs two spec-owned derivations, each with a differential against objectui and an objectui consumer:

• the default object form's field set (ObjectForm.tsx 961, through filterSystemFields);
• the inline form-factor smart default (resolveInlineMode: the form-only types, and the 2-rich and 8-field thresholds).

The triage direction for this family forbids copying objectui's rule into the lint.

The choice (for triage, not a ruling)

• A: promote both derivations to @objectstack/spec and credit them. The cost is two new cross-repo contracts. It also leaves an asymmetry: the same default form opened as the child's own create / edit form stays uncredited by the rule's existing untitled-bucket posture.

• B: keep the default form uncounted everywhere, as a stated boundary. PR fix(lint,spec): field-no-consumers reads an inline relationship's join key, its per-row expand form, a detail entry's formFields and a record:line_items block against the child (#21091) #21256 already states it in the module note and pins it with a test. The card would close as that boundary. The cost is that an author whose form-mode grid has authored inlineColumns sees those child fields reported.

• The dev's recommendation, which the seat finds well argued: B.

  • No producer is measured: the only inlineColumns in examples/ is the showcase invoice in grid mode, and no example authors inlineMode.
  • A would make the spec own the default record form for one door only.
  • A form-mode toggle that credited every visible child field would mask genuinely unused fields, which is the failure this rule exists to surface.
  • A adds two contracts with no pull.

  If triage takes B, this is a won't-fix candidate, and it goes to the decision box per the protocol rather than being closed by a seat.

Dedupe

454 objectstack issues and PRs (the open set plus the 300 most recently updated closed) were listed over REST and grepped locally for default ObjectForm, resolveInlineMode, inlineEdit … form and row form. The hits were PR #21256 and #21091 (the source) and #20951 (the family's earlier card, closed). As a control, field-no-consumers answers 6 hits, so the scan was live.

Dedupe words: field-no-consumers default ObjectForm row form inlineColumns form mode · inlineEdit true smart default resolveInlineMode · formFields absent inlineMode form