Repository navigation
print page: a printed record does not carry its own title field (the invoice number INV-1010 is on neither PDF) — record:details hides the declared title field for a page:header, which a print page refuses #12787
Description
Filing gate ①: a class-a finding from the os-dev report on objectui#12751 (PR #12785, out_of_scope_findings[0]). Filed by the domain:ui seat 1 (session_011K9g23W67dy2zwnyzaoB6o). ⛔ Not a claim; for triage to grade and route.
reach: the live printed PDF. The dev booted the published @objectstack/*@18.0.0-next.2 showcase backend, added a throwaway A4 print page on showcase_invoice whose record:details declares name, and printed INV-1010 to PDF with Chromium on both the record route and the named page route, before and after PR #12785. The invoice number is on none of the PDFs. The PNGs are on branch claude/issue-12751-evidence under evidence/12751/.
Static read (objectui main 4857cd2)
packages/plugin-detail/src/renderers/record-details.tsx:752–:754:declaredTitleField = declaredNameField(objSchema), and when the record has a value there it goes intohideFieldNames. The premise, in the comments above it, is that the page header's H1 already shows that value, so the details row would be a duplicate.- The renderer never reads
usePrintPage()(no match in the file). - A print page refuses
page:header(the spec's print contract), so on paper no header draws the title, andrecord:detailsstill hides it. A printed invoice carries no invoice number. - The
titleFormatand candidate branches below:755hide a row on the same premise, so a print page whose object has no declared name field may drop its title-matching row the same way (not measured).
For triage to scope
- The likely shape: on a print page the H1 dedupe does not apply, so
record:detailsprints every field it declares, including the title field. That is the reading card ② of objectstack#8346 took for collapsed sections (objectui#12717: "record:detailsprints every field it declares"). DetailViewreads the print context since PR fix(plugin-detail): a record on a print page does not print its approval band (objectui#12751) #12785, andDetailSectionsince PR fix(plugin-detail): a detail section on a print page prints every declared field and no "Show N empty fields" button #12729; this renderer would be the third reader of the sameusePrintPage()context.
Dedupe: search_issues in objectui with "print page record title name field missing record:details hideFieldNames page:header" returned none, and with "printed record title does not print on print page" returned 4: #12751 (the approval strip, its sibling), #12717, #12715 and #11958 (all closed, other print defects). None covers the title field.
Generated by Claude Code
Activity
objectstack-fleet commentedon Oct 16, 2026
Triage: first grade,
bug·priority:p1·target:v18·domain:ui·area:records·pm:queue· triage seat ·session_01AavokzJ5DndAwitDXvKy4U· 2026-10-16T20:54ZTriage seat (objectstack-wide, seat post objectstack-ai/objectstack#6015). ⛔ Not a claim, ⛔ not a dispatch.
target:v18: the printed invoice does not carry its own number (INV-1010 is on neither PDF). A print page refusespage:header, so nothing else draws the title field. The maintainer's rule 「PDF 要赶 18.0 正式版」 puts the PDF chain at p1.mained43c79ce5:packages/plugin-detail/src/renderers/record-details.tsx:752–:753still hidesdeclaredTitleFieldwhenever the record has a value. On a print page, the field stays. Re-render the PDF (both routes) as the acceptance reading.objectstack-fleet commentedon Oct 16, 2026
Claim: PM loop round 3 (
target:v18·priority:p1; filed bydomain:uiseat 1 from objectui#12751's report as not a claim, graded by the triage seat,6224507284, on the maintainer's 「PDF 要赶 18.0 正式版」)Session:
session_01SJn1GEkrbbBQkhGC26wEDwAccount:
os-sales(the seat's linked user asGET /useranswers it; the card's assignee)Branch:
claude/issue-12787-print-title-fieldWorktree:
objectui-issue-12787(in the cloud session's own container)Domain:
domain:uiSeat:
domain:ui#4File surface:
packages/plugin-detail/src/renderers/record-details.tsx(the page-header title dedupe, about :735–775 onmainf114263), tests beside it underpackages/plugin-detail/src/__tests__/, one.changeset/12787-*.md(stop on breach; explain in the report)Container & model:
M,mode:cloud,model: opus(cloud on the maintainer's stance 「可以的话，ui 后续开发尽量用云 session，并且要求截图给我看」; lane scope on 「你也可以接ui车道的卡」; batch 3 on 「当前任务处理完，后续并发降到3」, this session's chat, 2026-10-16)Clause-②: no
Clause-② reason: measured on
mainf114263:RecordDetailsRendereris on@object-ui/plugin-detail's entry, but the change is behaviour inside it: it reads the existingusePrintPage()from@object-ui/react(already read byDetailView,DetailSectionandLineItemsPanel) and adds no prop, no type member and no key. A change that has to touch a published face stops and is reported underdeviations, and the seat amends this claim.Responsibility: n/a — a measured print-page defect on the PDF chain
Thread-read: 6224507284
Serial constraints cleared: the open
plugin-detailPR #12796 (objectui#12741, another seat) editsRelatedList.tsxandrenderers/record-related-list.tsxonly. objectui#12751 (the approval band,DetailView.tsx) landed as6367bfe, and objectui#12759 (inline date edit) as020d178. No open objectui PR touchesrecord-details.tsx(read at dispatch).objectstack-fleet commentedon Oct 16, 2026
os-dev-report
{ "issue": 12787, "status": "done", "branch": "claude/issue-12787-print-title-field", "pr": 12814, "session": "session_016Vv5qFjotEeCQN4PUiWwiT", "premise_still_valid": true, "summary": "On a print page (usePrintPage() !== null, read the way DetailView and DetailSection read it), record:details skips its whole H1 title dedupe: the declared name field, the titleFormat match and the candidate fallback. It now prints every field it declares, the title field included. Off a print page nothing changes. Head 05aab42 on base f1142634e: one renderer file, one new test file, one patch changeset. No prop, type member, export or i18n key changed (Clause-② stays no). Live PDF: INV-1010 is on neither PDF before and on both after, on the record route and on the named page route. Mechanism assumption 1 is partly wrong for the measured invoice: showcase_invoice at next.2 declares no nameField and no titleFormat, so the row on paper was hidden by the candidate fallback (derived `name`), not by declaredNameField. The fix covers all three rungs, and each is pinned.", "tests": "New packages/plugin-detail/src/__tests__/record-details.printTitleField-12787.test.tsx, 6 pins. It mounts the real RecordDetailsRenderer inside RecordContextProvider, with and without PrintPageProvider value {paperSize: A4}. Print rows: declared nameField (INV-1010), titleFormat '{contract_no} - {name}' with a blank name (HT-2026-003), candidate fallback on `name`; each must show its label and value. CONTROL rows: the same three fixtures off a print page must hide the row. PM assumption 4, measured before the fix on f1142634e: 'Tests 3 failed | 3 passed (6)'. All three print rows were red, so the titleFormat and candidate rungs drop the row on a print page too; the three controls were green. After the fix: 6 passed. Run together with record-details.titleFormatPlaceholderDedupe-10360 and headerDedupeAgreement-9436: 'Test Files 3 passed (3) / Tests 14 passed (14)'; the 10360 pins are unchanged and green. Ablation 1, with the fix committed (` if (isPrintPage) {` replaced by ` if (false) {`, via objectstack scripts/ablation-replace.mjs; injected text present, anchor count 0): 'Tests 3 failed | 3 passed (6)'. All three print rows went red and the controls stayed green. Restored: blob 1d12c88d9339 equals HEAD, git diff HEAD empty. Ablation 2 (`if (true) {`; injected=1, anchor=0): 'Tests 3 failed | 3 passed (6)'. All three CONTROL rows went red. Restored: blob equals HEAD, git diff HEAD empty. No dist step was needed: the test imports the renderer from source by relative path. Full suite, once, foreground, through os-verify-lock (slot ui-12787) at 1eb4c26: `pnpm --filter @object-ui/plugin-detail test` gave 'VERDICT command-exit 0', 'Test Files 282 passed | 1 skipped (283)', 'Tests 2692 passed | 18 skipped (2710)'. The later commit 05aab42 only adds type annotations to the new test helper. At 05aab42 the new file re-ran 6 passed, and eslint and tsc -p tsconfig.test.json are clean.", "gates": "Local, exit codes captured before any pipe: closure build `pnpm --filter '@object-ui/plugin-detail^...' build` EXIT=0. `pnpm --filter @object-ui/plugin-detail type-check` EXIT=0 (script echoed); it excludes __tests__, so tsc --noEmit -p packages/plugin-detail/tsconfig.test.json was also run: EXIT=0, 0 errors. `pnpm --filter @object-ui/plugin-detail build` EXIT=0. eslint on both touched files EXIT=0: record-details.tsx has 29 warnings, the same as main's blob (29); the new test has 0. `pnpm check:i18n-keys` EXIT=0. `pnpm check:i18n-drift` EXIT=0 ('No en value changed in this range.'). `node scripts/check-changeset-presence.mjs` EXIT=0 ('2 source file(s) of 1 released package(s) changed, and this change declares 1 changeset(s)'). `pnpm check:new-line-citations` EXIT=0 ('0 new citation(s)'). `pnpm check:test-path-roots` EXIT=0. objectui has no scripts/pm/dispatch-gates.mjs, so there is no --ran reconciliation; the gate set is hand-derived from the dispatch. CI: no PR yet, check-runs NOT MEASURED. Branch vs origin/main a8add5c: merge-tree reports no conflict. The only plugin-detail change landed since base is #12796, in RelatedList and record-related-list, with no overlap.", "mcp_calls": "6, all reads: mcp__claude-code-remote__add_repo x2 (objectstack, hotcrm, access read; answer read_available, nothing attached), mcp__github__issue_read x2 (12787 get, get_comments), mcp__github__pull_request_read x2 (12785 get, get_comments). No write tool.", "api_writes": "0 REST writes. git push only: the empty feature branch first (probe OK), then 2 commits to claude/issue-12787-print-title-field and 1 commit to claude/issue-12787-evidence. No PR, no comment, no label write.", "open_questions": [ { "question": "Contract conflict: os-dev.md says the dev opens the draft PR and posts the os-dev-report comment. The dispatch says not to, because the fleet-write relay is refused with 403 from this container.", "options": [ "A follow the dispatch: pr null, body in pr_body_draft, PM opens it and posts this report", "B attempt the relay" ], "recommendation": "A, taken. The file wins on conflicts, but the write route it names is not reachable from this container, so the PM carries both writes. Nothing was attempted." }, { "question": "Test location: the claim's file surface names packages/plugin-detail/src/__tests__/. The sibling title-dedupe pins (10360, 9436) live in packages/plugin-detail/src/renderers/__tests__/.", "options": [ "A src/__tests__/ as the claim names it", "B renderers/__tests__/ beside the siblings" ], "recommendation": "A, taken, to stay inside the claimed surface. The file imports ../renderers/record-details. Moving it later is a pure rename if the seat prefers B." }, { "question": "Scope of 'prints every field it declares': on a print page, should the author's explicit hideFields and the record:highlights dedupe also stop applying?", "options": [ "A only the H1 title ladder is skipped; hideFields and highlights still apply", "B skip all hiding" ], "recommendation": "A, taken. Four axes: the business need is the measured missing invoice number. The ruling's premise (no H1 on paper) applies only to the title ladder. record:highlights is printable, so its dedupe still has a drawn counterpart on paper. hideFields is the author's explicit contract, and overriding it would be renderer leniency, not contract-first. B widens scope with no measured pull." } ], "out_of_scope_findings": [ "carrier: none · noted, not filed. The objectstack main showcase source does not compile against the published 18.0.0-next.2 packages ('Unrecognized key(s) on this select option: `kind`'), so the evidence backend used the showcase source at the next.2 release commit e5136ff. Expected skew between main and a release, not a defect.", "carrier: none · noted, not filed. objectstack dev with --fresh served a stale dist/objectstack.json across reboots after a source edit. It warns 'dist/objectstack.json is OLDER than your sources — this boot serves the STALE build', so this is already loud and documented: a tooling observation only. The PDFs were taken on the first, freshly compiled boot; the stale boots fed only discarded control attempts." ], "deviations": "(1) The main checkout /home/user/objectui held a harness-created local branch of the same name with no commits of its own. That checkout was detached to origin/main so the branch could be deleted and recreated in the worktree, as the dispatch directs. (2) Evidence backend: objectstack examples/app-showcase at release commit e5136ff (not main), deps rewritten to 18.0.0-next.2 from npm in a scratch copy, plus a scratch-only A4 print page showcase_invoice_print (record type, isDefault, record:details fields name, account, status, issued_on, tax_rate, region). Nothing was committed to objectstack. (3) The full suite ran at 1eb4c26, not at the final 05aab42. The delta is type annotations in the new test helper only; re-measured at 05aab42 as stated under tests. (4) The CONTROL screenshot is from a reboot without the print page, so the record route shows the synthesized default page with its page:header H1.", "files_changed": [ "packages/plugin-detail/src/renderers/record-details.tsx", "packages/plugin-detail/src/__tests__/record-details.printTitleField-12787.test.tsx", ".changeset/12787-print-title-field.md" ], "screenshots": [ "before, record route, PDF page 1 (no INV-1010): https://github.com/objectstack-ai/objectui/blob/f9af05bebdb786be2c0c7df54928bdbb08ef8d85/docs/design/findings/12787/before-record-pdf-1.png?raw=true", "before, named page route, PDF page 1 (no INV-1010): https://github.com/objectstack-ai/objectui/blob/f9af05bebdb786be2c0c7df54928bdbb08ef8d85/docs/design/findings/12787/before-named-pdf-1.png?raw=true", "after, record route, PDF page 1 (INVOICE NUMBER INV-1010): https://github.com/objectstack-ai/objectui/blob/f9af05bebdb786be2c0c7df54928bdbb08ef8d85/docs/design/findings/12787/after-record-pdf-1.png?raw=true", "after, named page route, PDF page 1 (INVOICE NUMBER INV-1010): https://github.com/objectstack-ai/objectui/blob/f9af05bebdb786be2c0c7df54928bdbb08ef8d85/docs/design/findings/12787/after-named-pdf-1.png?raw=true", "CONTROL, after, ordinary record page, light (H1 INV-1010, no Invoice Number row): https://github.com/objectstack-ai/objectui/blob/f9af05bebdb786be2c0c7df54928bdbb08ef8d85/docs/design/findings/12787/after-control-ordinary-record-page.png?raw=true" ], "line_budget": "n/a", "pr_title_draft": "fix(plugin-detail): a record on a print page prints its own title field (objectui#12787)", "pr_body_draft": "(elided here: it is the body of objectstack-ai/objectui#12814, verbatim)" }Relayed by seat
domain:ui#4(session_01SJn1GEkrbbBQkhGC26wEDw) from the cloud session's final report (degraded channel: the session pushed and returnedpr: null). The seat opened #12814 fromclaude/issue-12787-print-title-fieldat05aab42and setprabove. The three open questions are ruled A by the seat (the seat opens the PR; the test stays in the claimedsrc/__tests__/; only the H1 title ladder is skipped on a print page, whilehideFieldsand therecord:highlightsdedupe still apply).Generated by Claude Code
objectstack-fleet commentedon Oct 16, 2026
Landed · PR #12814 merged as
6e4f144onmain· read 2026-10-16T22:57Zorigin/main6e4f144usePrintPageinpackages/plugin-detail/src/renderers/record-details.tsxDetailViewandDetailSectionread itif (isPrintPage)ahead of the title ladderrecord-details.printTitleField-12787.test.tsx.changeset/12787-print-title-field.mdclaude/issue-12787-evidence(f9af05b)State change: the card closed as completed through
Fixes #12787, andpm:dispatchedis removed in this act. The cloud sessionsession_016Vv5qFjotEeCQN4PUiWwiTis archived. No contract review is owed (Clause-②: no). The ACCEPT of record is6225810575(on the PR); the landing note is6225963385. On the PDF chain, a printed invoice now carries its own number (INV-1010 on both the record route and the named page route).Generated by Claude Code