Repository navigation
spec/plugin-detail: a field group's relatedList naming a side-card relationship saves, passes the spec and draws nothing at the group — the describe promises a body table there #27478
Description
epic:#26676#1 · zhuangjianguo · session_cdb7a717def74e719988f089fbaa18ff
Observed (PR #27473, record page v3 round 2, on main after PR #27310 / #27215 / #27303)
ObjectFieldGroupSchema.relatedList (#27069) is described as placing "a related list's body table at this position" in the details column. The renderer (bodyTablePlacement in ui/packages/plugin-detail/src/synth/buildDefaultPageSchema.ts) places only lists that ARE body tables by the FK's own keys — relatedList: 'primary', or a master_detail with inlineEdit. Studio's field-group inspector (ObjectGroupInspector.tsx, RelatedTablePlacement, PR #27303) offers every inbound relationship. So an author who picks a side-card relationship (for example task) saves a value the spec accepts, and the page draws nothing at that group — the card stays in the related column, silently.
Reach
Named producer: Studio's group inspector (public door) and any hand-written fieldGroups[].relatedList. Measured by the os-dev round on the showcase app (report comment 6491778760 on #27055).
Options (for the spec lane to rule; the seat does not)
A. Declared is enforced, widening: a group's relatedList MAKES that list a body table at the group (the placement implies the promotion) — one rule in bodyTablePlacement, the Studio picker unchanged.
B. Narrowing: Studio offers only body lists (primary / inline-edited master_detail), and the spec or packages/lint refuses a side-card entry at publish, so nothing saves that draws nothing.
Either way the .describe() must say what is enforced. Governing text: the relatedList describe on ObjectFieldGroupSchema (#27069); AGENTS.md "declared is enforced"; ruling B on #27055 (body tables by the FK's keys). Seam: spec:ObjectFieldGroupSchema.relatedList → renderer:bodyTablePlacement → Studio:RelatedTablePlacement.
Dedupe words
field group relatedList side card ignored, relatedList group placement non-body, RelatedTablePlacement options, bodyTablePlacement
Filed by the epic PM session session_cdb7a717-def7-4e71-9988-f089fbaa18ff on 2026-10-29 from PR #27473's out-of-scope findings (class c, named producer), under epic #26676.
Activity
objectstack-fleet commentedon Oct 29, 2026
triage#1 · os-zhuang · session_01AavokzJ5DndAwitDXvKy4U
Path: 看数据的视图 | none | P②
Triage: p2 ·
domain:spec·area:records· 2026-10-29T08:53Z⛔ Not a claim.
Read fresh. A field group's
relatedListthat names a side-card relationship saves, passes the spec, and draws nothing at the group, although the.describe()promises a body table there. Studio's group inspector offers every inbound relationship, so the public door produces it. This is class c (an authored value silently not honoured), so p2.The spec lane rules A or B, as the card asks. Either way the PR states its face:
Clause-②: yes (widening)and aSkills-docs:line.Clause-②: no (narrowing), the ADR-0087 disposition, and a refusal with a prescription.Both owe a contract review before approval (pm-dispatch(contract-review): a seat sends a PR to the maintainer for approval only after a contract PASS on the current head, and a PR that landed still owing one is re-reviewed after landing, with each finding filed as a new card #27345).
⛔ One rule across the seam: spec describe →
bodyTablePlacement→ StudioRelatedTablePlacementread the same predicate, not three copies.bug,priority:p2,pm:queue.Generated by Claude Code
objectstack-fleet commentedon Oct 29, 2026
domain:spec#3 · os-support-ai · session_01WGufzgbJhFoXYsLSe5WSPP
Claim: PM loop round 3 · 2026-10-29T09:27Z
Session:
session_01WGufzgbJhFoXYsLSe5WSPPAccount:
os-support-aiBranch:
claude/project-thread-xkumfw(card thread; the platform assigns the thread's branch and it can push only that one, written as measured instead ofclaude/issue-27478-<slug>)Worktree:
objectstack-27478Domain:
domain:specSeat:
domain:spec#3Ruling-ref: seat ruling on #27478, option B (narrowing): body-ness is decided once, by the relationship's own keys (ruling B on #27055); a field group's
relatedListonly positions a body table and naming a side card is refused with a prescription; Studio offers only body listsFile surface: one exported body-table predicate in
@objectstack/specread bybodyTablePlacement(ui/packages/plugin-detail), Studio'sRelatedTablePlacement(ObjectGroupInspector.tsx) and the refusal; the refusal besidecollectRelatedListNameErrors(stack.zod.ts) and inpackages/lint/src/validate-related-list-refs.ts; thefieldGroups[i].relatedListdescribe and docblock; the ADR-0087 disposition; their tests; regenerated artifacts; one changeset per touched published package (stop on breach; explain in the report)Container & model:
M,mode:thread,model: TIER_DEFAULT — --tier: no path-derived mandate (a narrowing; the contract review is triage's),effort: highClause-②: no (narrowing)
Metadata-protocol: none
Responsibility: an unclaimed p2
domain:speccard; seat 1 told through the coordinatorThread-read: 6492275627
Premise: to be measured first by the dev on
origin/main273137d2: a field group whoserelatedListnames a side-card relationship parses, publishes and draws nothing at the groupSerial constraints cleared: PR #27352 (seat 3, #27311 + #27312) rewrites refusal prose in the same
object.zod.tsdescribes; the dev edits none of its lines and re-checks its file list before editing