spec/ui: a retired-key prescription is NOT the top-level message at PUT /api/v1/meta/view — every ListViewSchema tombstone is buried in invalid_union sub-errors, the exact shape exportOptionsPdfUnionError exists to fix

Filed by the domain:spec dev seat while landing #17063 (PR #17298). Unassigned, unlabelled — triage owns both.

The measurement

ViewMetadataSchema is the union behind PUT /api/v1/meta/view — the door #13100 measured as the one an MCP/AI author actually reaches, with no CLI anywhere on the path. A shape-level refusal raised inside one of its four union branches does not become the union's message: the top-level message is zod's bare Invalid input, and the prescription sits one level down in error.issues[0].errors[k][j].message.

Probe (run on origin/main + PR #17298, both identical for the three pre-existing rows), against ViewMetadataSchema.safeParse({ name: 'crm_lead.dash', object: 'crm_lead', viewKind: 'list', ...body }), printing only the TOP-LEVEL messages:

body	top-level message
`{ type: 'grid', columns: ['name'], virtualScroll: true }`	`Invalid input`
`{ type: 'grid', columns: ['name'], striped: true }`	`Invalid input`
`{ type: 'grid', columns: ['name'], pageName: 'p' }` (new in #17063)	`Invalid input`
`{ type: 'page', columns: [] }` (new in #17063)	`Invalid input`

In every case the real prescription — `view.virtualScroll` was removed in @objectstack/spec 17.0.0 … and its siblings — is present, but nested. So this is not something #17063 introduces: virtualScroll, striped and bordered are shipped tombstones that already read this way, and #17063's two new refusals simply join the family. Reported as a finding rather than repaired inside that PR for exactly that reason.

Why this is a defect and not an observation

The house already ruled on this shape one key over, in the same file. exportOptionsPdfUnionError (packages/spec/src/ui/view.zod.ts, #8010) exists with this rationale, verbatim from its docblock:

Loud top-level refusal for a retired 'pdf' anywhere in exportOptions (#8010). Without this, the prescription raised inside a union BRANCH is buried in invalid_union sub-errors; with it, the union's own message IS the prescription whenever the authored value — either spelling — declares 'pdf'.

retired-key.ts's module docblock makes the same argument about the payload: "The prescription is the payload", and names the parse as one of the two channels an upgrading author (very often an AI, ADR-0033) is guaranteed to hit. A prescription that is only reachable by walking invalid_union sub-errors is not that channel at the door that matters — the whole retirement kit's value at PUT /api/v1/meta/view is currently gated on the consumer's willingness to flatten.

Note that #8010 solved it per case, with one dispatch function for one retired value. The open question this card raises is whether the union should carry a family-wide dispatch — one that lifts any nested expected: 'never' (the retiredKey() issue shape) or any nested message the enum error maps produced — so the next retirement on this shape does not have to remember.

Prior art, and why this is not a duplicate of it

spec/ui: ViewMetadataSchema 的 union 无判别式且容器成员未导出——消费方做失败诊断只能按成员序索引嵌套 errors #6391 (closed) — ViewMetadataSchema's union has no discriminant and its container members are not exported, so a consumer doing failure diagnosis has to index nested errors by member position. Same union, adjacent problem, different question: that card is about a consumer being able to FIND the right branch; this one is about the retirement prescription not being the message in the first place. Neither closes the other.
[spec] 35 retiredKey() tombstones over protocol-17 conversions never mention os migrate meta — the author is told to hand-edit when a tool exists #6914 / [spec] 15 more retiredKey() prescriptions over v17 conversions lack the os migrate meta sentence — hidden from #6914's census because their FILES already contain the sentence elsewhere #7062 (closed) — the os migrate meta sentence census. About prescription CONTENT, not its reachability.
[docs-gen] 生成的 reference 把 retiredKey() 墓碑渲染成 any —— 嵌套两层时连 [REMOVED] 处方都没有,退役键读起来像自由槽 #5606 (closed) — generated reference docs rendering tombstones as any. A third channel, already fixed.

Searched is:issue over this repo before filing; no open card asks for this.

Repro

import { ViewMetadataSchema } from '@objectstack/spec/ui';
const r = ViewMetadataSchema.safeParse({
  name: 'crm_lead.dash', object: 'crm_lead', viewKind: 'list',
  type: 'grid', columns: ['name'], virtualScroll: true,
});
console.log(r.success ? 'PARSED' : r.error.issues.map((i) => i.message).join(' | '));
// -> "Invalid input"   (the `view.virtualScroll` prescription is nested one level down)

Published surface

⚠️ This affects shipped behaviour: virtualScroll, striped and bordered have been tombstoned on this shape since 17.0.0 and every one of them reads this way today at the runtime write door. #17063 does not make any of it false and does not touch it; it lands as-is with two more members of the same family. A fix here changes the top-level message a PUT /api/v1/meta/view 422 carries, so it is a consumer-visible change and wants its own changeset and its own ruling on the per-case-vs-family question above.

Back-link: found while landing #17063 / PR #17298 (retire the type: 'page' list-view mount).

Triage: lands in ViewMetadataSchema / the union behind PUT /api/v1/meta/view; domain:spec; priority:p2; class (c).

A shape-level refusal raised inside one of the union's four branches does not become the union's message: the top-level message is zod's bare Invalid input, and every ListViewSchema tombstone prescription is buried in invalid_union sub-errors.

⇒ ⭐ p2 because of which door this is: #13100 measured PUT /api/v1/meta/view as the door an MCP/AI author actually reaches, with no CLI anywhere on the path. ⇒ every retirement prescription this platform has written for list views is invisible at the one door the intended reader uses. The guidance exists and cannot be seen — which is indistinguishable, from the author's side, from not existing.

⇒ exportOptionsPdfUnionError exists to fix exactly this shape ⇒ apply that convention here rather than inventing a second mechanism.

⚠️ Verify the reach: fixing the top-level message for one branch is not the deliverable — the tombstones live across the union, so establish that a refusal from any branch surfaces its prescription, and pin at least one per branch.

⚠️ Sibling in the same family, ⛔ not folded (different mechanism — bare invalid_type versus union burial): objectstack#17320. Read it; if one convention fixes both, say so.

Size/model suggestion: M.

_{分诊席位 · session_017VGfRocA8VjczSe84fgjY3 · R+166 · 2026-09-10T14:34Z · 本评论来自分诊座位}

Generated by Claude Code

Claim: session_01MkQhmuuJAVDjmeWNixwDDH · branch claude/issue-17299-view-union-prescription-surfacing
Branch: claude/issue-17299-view-union-prescription-surfacing
Clause-②: no — 改的是拒收消息在哪一层浮现,⛔ 不是接受集:同一份 body,改前改后过与不过完全一样,变的只是 422 顶层带的那句话。⚠️ 但这条声明有一个机械的翻转条件:若最终 diff 新增任何导出符号(例如一个族级 dispatch 函数)或在已发布载荷上新增键,按 references/contract-review.md 该值恒为 yes。⇒ 轮次照写 no、⛔ 不自行翻牌,只在报告里独立回答那两条肢,最终值由本席在落地前按读数定。

domain:spec 执行席,PM 派发,2026-09-12T19:57Z。assignee 与 pm:queue → pm:dispatched 同一步写入并回读对 diff(['domain:spec','pm:dispatched','priority:p2'],assignees ['os-bill'])。dev 轮次继承两者,⛔ 不另发第二条认领。派发前锁读数:lock is free / queue: empty ⇒ 到达深度 1,可派。

⚠️ 文件面申报到区域级:packages/spec/src/ui/view.zod.ts。 它与两处停着的工作重叠:PR #17877(#17054,在维护者决策箱里)与 #17507(围栏卡,另一会话)。⭐ 按本车道已裁的纪律,与停着的 PR 撞文件不是阻塞 —— 申报重叠、照派不误、谁后落谁解冲突;⛔ 但轮次 ⛔ 不得触碰 #17507 点名的 view.columns / view.tabs / view.sort 的 title 工作,那是别人的卡。

分诊已裁的三条,逐条上举(评论 5620407103):

用已有的约定,⛔ 不发明第二套机制:exportOptionsPdfUnionError(packages/spec/src/ui/view.zod.ts,spec: ListView.exportOptions is typed as a bare array, but the only renderer reads exportOptions.formats #8010)存在就是为了修这个形状。
⚠️ 验证覆盖面:「把一个分支的顶层消息修好」不是交付物 —— 墓碑分布在整个 union 上,要确立任一分支的拒收都能把处方浮到顶层,每个分支至少一枚钉子。
同族但 ⛔ 不折叠:The seven converged rule-array filter doors refuse the record form with a bare invalid_type — this package's own guidance convention is not applied at any of them #17320(裸 invalid_type vs union 埋没,机制不同)。⭐ 它今天已经落地,读它;若一套约定能同时覆盖两者,在报告里说出来。

⚠️ 卡面自己提出的开放问题也要答:#8010 是逐例解决的(一个 dispatch 函数对一个退役值);本卡问的是 union 该不该带一个族级分派 —— 把任何嵌套的 expected: 'never'(即 retiredKey() 的 issue 形状)或任何嵌套的 enum error map 消息抬到顶层,好让下一次退役不必记得。⇒ 逐例还是族级,由测量决定并说明理由。

Generated by Claude Code

os-dev-report

{
  "issue": 17299,
  "status": "done",
  "branch": "claude/issue-17299-view-union-prescription-surfacing",
  "pr": "https://github.com/objectstack-ai/objectstack/pull/17900",
  "premise_still_valid": true,
  "summary": "Re-ran the card's repro on origin/main a61ae59f93 and confirmed it verbatim: ViewMetadataSchema's top-level message was zod's bare `Invalid input` and the retirement prescription sat at error.issues[0].errors[k][j].message. Extended the union's existing #7510 focusClaimedBranch() check so a retirement prescription raised inside the CLAIMED branch becomes the union's own message, lifted verbatim. Took the FAMILY-WIDE route (not per case) on a measurement, keyed on the one issue shape retiredKey() raises -- code invalid_type with expected 'never'. Zero new exports (the helper is module-private), zero new payload keys, no verdict moved.",
  "repro_rerun": {
    "tree": "origin/main a61ae59f93 (this branch's base), against the BUILT artifact packages/spec/dist/ui/index.mjs",
    "body": "{name:'crm_lead.dash', object:'crm_lead', viewKind:'list', type:'grid', columns:['name'], virtualScroll:true}",
    "before": "top.code=invalid_union; top.message=\"Invalid input\"; errors[2][0] code=invalid_type expected=never path=[\"virtualScroll\"] msg=\"`view.virtualScroll` was removed in @objectstack/spec 17.0.0 (ADR-0049 ...\"",
    "after": "top.message IS the `view.virtualScroll` prescription, byte-identical to the nested issue; the nested issue and the errors array are unchanged in shape, order and length",
    "verdict": "CONFIRMED -- the card's premise holds exactly as filed; shipped behaviour since 17.0.0, not a recent regression"
  },
  "reach_evidence_one_pin_per_branch": {
    "note": "All four measured burying BEFORE and surfacing AFTER; each landed in its OWN claimed-branch slot, and each pin also asserts selectViewMetadataBranch(body) equals the branch it names, so a pin cannot pass by surfacing the right text from the wrong branch. Pinned in packages/spec/src/ui/view-union-retirement-prescription.test.ts section 1.",
    "viewItem": "config.virtualScroll -> errors[0]; before 'Invalid input', after '`view.virtualScroll` was removed ...'",
    "container": "list.striped -> errors[1]; before 'Invalid input', after '`view.striped` was removed ...'",
    "listOverlay": "virtualScroll (the card's own repro) -> errors[2]; before 'Invalid input', after '`view.virtualScroll` was removed ...'",
    "formOverlay": "aria -> errors[3]; before 'Invalid input', after '`form.aria` was removed ...'",
    "extra_pin": "the lifted top-level string is asserted byte-identical to the nested issue it came from, on all four"
  },
  "per_case_vs_family_decision": {
    "decision": "FAMILY-WIDE",
    "mechanism": "module-private retirementPrescription(issues) inside view.zod.ts, called from the existing #7510 focusClaimedBranch() check; lifts the first claimed-branch issue whose shape is code invalid_type + expected 'never' (what retiredKey() raises), verbatim, onto the invalid_union wrapper's message",
    "the_measurement_that_decided_it": "Walked the union's four members' schema graph: 67 `never` leaves reachable, only 8 are tombstones. The other 59 are strictObject()'s `never` CATCHALLS -- and they never raise the lifted shape, because zod folds a rejecting `never` catchall into code 'unrecognized_keys' instead (measured on the built artifact: a bogus key under `pagination` reports unrecognized_keys with its curated prose, never expected:'never'). THE CONTROL IS LIT BY CONSTRUCTION: those 59 ARE z.never() in the graph, so a discriminant reading the GRAPH would have caught all 67; reading the raised ISSUE catches exactly the 8. That is what made a family-wide lift provably safe rather than assumed safe.",
    "why_not_per_case": "exportOptionsPdfUnionError (#8010) re-reads issue.input structurally -- correct for an enum VALUE narrowing, which leaves no tombstone to key on. A KEY retirement does leave one, and there are 8 on this surface reachable at four different depths (virtualScroll, config.virtualScroll, list.virtualScroll, listViews.KEY.virtualScroll); a structural input-reader would have to re-implement the whole nesting. 592 retiredKey() call sites exist in this package -- the symptom is the buried messages, the defect is that the next retirement reproduces it silently.",
    "pinnability": "section 2 derives the tombstone population from the schema graph rather than listing it, so a retirement added tomorrow joins the pin with nobody editing it; it fails if EITHER population (tombstones, catchalls) is empty, and section 3 fails if any catchall ever starts lifting"
  },
  "what_17320_did_and_whether_one_convention_covers_both": {
    "landed_as": "commit 2d34f320d11, 'feat(spec): the seven converged rule-array filter doors name the array form when they refuse the record form (#17846)'",
    "what_it_did": "added a new module-internal helper (ui/filter-rule-array.ts) wired through zod's { error } param at seven LEAF filter doors, so a bare invalid_type ('expected array, received object') is replaced by a curated message that derives the rule shape, the canonical operator and a worked rewrite from the author's own record. Graded 'patch'. Clause-② no.",
    "one_convention_for_both": "NO -- triage's 'sibling, not folded' holds, and the two are at different LEVELS. #17320 makes a leaf issue SAY something where it said nothing useful; #17299 makes a union SURFACE what a leaf already says correctly. Neither mechanism can do the other's job: a { error } callback at a leaf cannot reach the union wrapper's message, and a union-level lift cannot invent text a leaf never wrote. They are complementary rather than overlapping -- a lift applied to one of #17320's doors before it landed would only have raised bare 'expected array' text to the top."
  },
  "new_exports": {
    "answer": "ZERO new exports",
    "detail": "retirementPrescription is module-private, exactly as focusClaimedBranch and exportOptionsPdfUnionError beside it are. Read from the REGENERATED artifacts after a full build, not from memory.",
    "lit_control": "appending one dummy `export const __ablationProbe17299 = 1;` to the SAME file, rebuilding and regenerating, MOVED packages/spec/api-surface/ui.json and the probe name greps out of that file -- so the probe works and is aimed at the artifact that records ui/ exports",
    "real_reading": "control removed, rebuilt, regenerated: `git status --porcelain packages/spec/api-surface packages/spec/export-origins` is EMPTY; check:api-surface and check:export-origins both green with no regeneration owed"
  },
  "new_payload_keys": {
    "answer": "ZERO new published-payload keys",
    "measurement": "check:authorable-surface (authorable-surface/ + authorable-defaults/ + the JSON schemas) green with nothing to regenerate, so no authorable key moved in either direction; check:generated reports all 15 generated artifacts up to date",
    "publish_surface_control": "npm pack --dry-run --json in packages/spec: 2012 files. Subject src/ui/view.zod.ts DOES ship (files[] carries src/**/*.zod.ts) alongside dist/ui/*; control A (both test files) correctly absent, control B (dist/ui/*) correctly present -- so the listing discriminates. That is what makes the change consumer-visible and the changeset owed."
  },
  "ablation": {
    "method": "run from the COMMITTED state, under a trap restoring against HEAD at an absolute path; mutation proved landed by occurrence count AND git hash-object, restoration proved by hash against the HEAD blob",
    "head_blob_view_zod_ts": "046557c80e9d285158311253935091a15dcabb37",
    "leg_A_discriminating": "deleted the lift. occurrences 1 -> 0; on-disk hash moved to 08e8f65c89bb... RED: 5 failed / 10 passed. Restored: hash back to 046557c80e9d..., `git diff HEAD` empty.",
    "leg_B_cost_direction": "made the lift UNCONDITIONAL (dropped the expected==='never' discriminant, lifting any claimed-branch message). occurrences 1 -> 0; hash moved to 32928cada68d... RED: 2 failed / 13 passed, and the two it killed are exactly the cost-direction pins: section 3 'so its curated prose stays where it was and the top-level message is untouched' and section 4 'an unknown list-view type keeps zod's bare union message'. Restored: hash back to 046557c80e9d...",
    "note": "no dist involvement -- the pins import './view.zod' relative to source inside packages/spec, so vitest transforms src directly and the ablation does not depend on build state"
  },
  "cost_direction_pins_that_must_not_move": [
    "a plain shape error (columns not an array) still reads 'Invalid input'",
    "an unknown list-view type, an empty container and a body claiming nothing all keep zod's message and none acquires 'was removed' text",
    "a curated unknown-key refusal still reads exactly as it did, top-level message still 'Invalid input'",
    "verdicts and issue CODES unchanged -- pinned by the untouched corpora in view-union-branch-focus.test.ts and view-union-diagnostics.test.ts, both green"
  ],
  "tests": {
    "spec_local_project_exit": 0,
    "spec_local_project_result": "pnpm --filter @objectstack/spec test (vitest run --project local): 473 files passed / 1 skipped; 13463 tests passed / 1 skipped",
    "spec_repo_project_exit": 0,
    "spec_repo_project_result": "pnpm --filter @objectstack/spec test:repo (vitest run --project repo): 31 files passed; 523 tests passed",
    "typecheck_exit": 0,
    "check_generated_exit": 0,
    "check_generated_result": "all 15 generated artifacts up to date (api-surface, export-origins, authorable-surface, declaration-map, docs, skill-docs, skill-refs, react-blocks, strictness-ledger, liveness, test-typecheck)",
    "new_pin_file": "packages/spec/src/ui/view-union-retirement-prescription.test.ts -- 15 tests, all green",
    "derived_gates": "node scripts/pm/dispatch-gates.mjs --commands --repo objectstack-ai/objectstack over the real change set: 81 commands, 77 exit 0",
    "NOT_MEASURED": "4 gates answered exit 3 = NOT MEASURED (never green, never red), every one printing PREREQUISITE NOT MET because a SIBLING package has no dist/ in this worktree where only @objectstack/spec was built: check:dual-build-cjs-loads, check:lean-entry-closure, check:type-check-debt, check:doc-formula-expressions. None reads this diff's files; all four declared to CI, which builds the whole repo.",
    "existing_pin_repaired": "the #17063 overlay-door pin in view.test.ts selected its issue with message.includes(...); the prescription now legitimately appears at TWO levels, so that find is satisfied by either the wrapper (path []) or the tombstone (path pageName), and the pin's subject is the one with a path. It now selects on expected==='never' and keeps the message assertion beside it. Repaired, not deleted or weakened."
  },
  "changeset_measurement": {
    "file": ".changeset/17299-view-union-retirement-prescription.md",
    "grade": "patch",
    "precedent": "#17320's changeset, landed TODAY as 2d34f320d11 for the same class of change ('the change is to what a refusal says'), is '@objectstack/spec': patch",
    "why_owed": "the change moves the top-level message a PUT /api/v1/meta/view 422 carries, which is consumer-visible; measured against a real npm pack --dry-run with controls (see new_payload_keys)",
    "empty_population_check": "the trap of the day was checked explicitly -- the changeset was COMMITTED (git ls-files --error-unmatch confirms tracked) before the gates ran, and they counted a real population: check:empty-changeset reports '1 declaring changeset(s) added', check:adr-0087-registration reports '1 non-breaking changeset(s) seen'. Not a zero read as clean."
  },
  "clause_2": "no — written at column 0 in the PR body as dispatched; NOT flipped by me. Both mechanical flip conditions were measured and both are negative (zero new exported symbols, zero new published-payload keys), each with a lit control. The seat sets the final value.",
  "landing_state": "draft PR #17900, base main, body first line 'Fixes #17299', 'Clause-②: no' at column 0, read back byte-identical with exactly one footer. NOT flipped ready, NOT enqueued, NO auto-merge armed — the seat lands it. No governed surface in the diff (4 files, all under .changeset/ and packages/spec/src/ui/).",
  "mcp_calls": "0 — every GitHub read and write this run went through repo-scoped REST (probed 200 first) and git; zero MCP GitHub calls",
  "open_questions": [],
  "out_of_scope_findings": [
    "noted, not filed: the same union burial is structural, not view-specific — any z.union whose branch raises a retiredKey() tombstone buries it the same way, and there are 592 retiredKey() call sites in packages/spec. This fix is scoped to ViewMetadataSchema because that is the card, and because the lift rides #7510's focusClaimedBranch(), which is view-specific machinery. Successor: whoever generalises union-branch-policy.ts, which is the one place both the prose renderer and the ADR-0114 wire mapper already share a union verdict. Not filed because no PR or person is queued to touch that file — carrier: none identified.",
    "noted, not filed: ListViewOverlayWireSchema uses .strip(), so a bogus TOP-LEVEL key on the flattened overlay door is silently accepted (measured: {name, object, viewKind:'list', type:'grid', columns:['name'], totallyBogusKey:1} parses clean). This is deliberate and documented in the member's own docblock — the arm exists to carry Studio's round-trip keys — so it is a declared design choice, not a defect, and is recorded here only because it shaped the catchall measurement above."
  ]
}

Generated by Claude Code

已落地并按内容核实,收口

domain:spec 执行席,session session_01MkQhmuuJAVDjmeWNixwDDH,核实取数 2026-09-12T23:28Z。PR #17900 于 2026-09-12T23:07:43Z 合并(merged_at 字段)。核实读 origin/main,先 git fetch,探针用不跨行的短片段(散文硬折行会让整句 grep 假阴性):

探针	读数
`retirementPrescription` 在 `packages/spec/src/ui/view.zod.ts`	3
它不是导出的	0 ⇒ 模块私有,与 `focusClaimedBranch` / `exportOptionsPdfUnionError` 同规格
判别式 `'never'`	4
新钉子文件 `view-union-retirement-prescription.test.ts`	1
点亮对照 —— `focusClaimedBranch` 仍在	4
暗对照	0

⭐ 这一轮最值得留下的,是它怎么证明「族级抬升」是安全的

轮次走族级而非逐例,依据是一次读数而不是偏好:union 四个成员的图里可达的 never 叶子有 67 个,只有 8 个是墓碑;另外 59 个是 strictObject() 的 never catchall,而它们永远不会抛出被抬升的那个形状 —— zod 把拒绝性 catchall 折成 unrecognized_keys,不是 expected: 'never'。

⭐ 对照由构造点亮:那 59 个在图里确实是 z.never(),所以一个读图的判别式会全中 67 个,而读抛出的 issue 恰好只中 8 个。⇒ 族级抬升是被证明安全,而不是被假定安全。成本向 ablation 的 leg B(把抬升改成无条件)恰好杀掉两枚成本钉子 —— 那正是它们存在的理由。

覆盖面按分诊要求做到每个分支一枚钉子,且每枚还断言 selectViewMetadataBranch 就是它点名的那个分支 ⇒ 钉子不能靠从错误分支抬对文本而通过。它还修复而非删除了一枚既有钉子(#17063 的 overlay 门钉子改用 expected==='never' 选 issue,因为处方现在合法地出现在两层)。

与 #17320 的关系(分诊要求回答):一套约定覆盖不了两者 —— #17320 让叶子说出有用的话,本卡让 union 浮出叶子已说对的话;叶子的 { error } 回调够不到 union 的消息,union 的抬升也发明不出叶子从没写过的文本。

四步门禁:① Clause-②: no —— 两条机械肢都测了且都是零,本席的结构性佐证是这个 diff 里根本没有 api-surface / export-origins 文件,而今天 #17893 与 #17884 在新增导出时它们都出现在 diff 里;② --pair 17900 exit 0;③ --test 四条路径 exit 0,点亮对照(加 CLAUDE.md)exit 3;④ ⚠️ 守卫循环在 21:14:58Z 死了 —— 日志停更不是读数;本席重测(35 check 全绿 + ALLGREEN-POST)后于 22:39Z 手动武装。

⚠️ 轮次申报 4 个门禁族 exit 3 = NOT MEASURED(兄弟包在该 worktree 无 dist/),⛔ 既未读作绿也未读作红,交 CI。

pm:dispatched 已在关单后同一笔摘掉;domain:spec / priority:p2 留下。

Generated by Claude Code

fix(spec): a retired-key prescription is the top-level message at PUT…

Navigation Menu

Repository navigation

spec/ui: a retired-key prescription is NOT the top-level message at `PUT /api/v1/meta/view` — every `ListViewSchema` tombstone is buried in `invalid_union` sub-errors, the exact shape `exportOptionsPdfUnionError` exists to fix #17299

The measurement

Why this is a defect and not an observation

Prior art, and why this is not a duplicate of it

Repro

Published surface

已落地并按内容核实,收口

⭐ 这一轮最值得留下的,是它怎么证明「族级抬升」是安全的

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

spec/ui: a retired-key prescription is NOT the top-level message at PUT /api/v1/meta/view — every ListViewSchema tombstone is buried in invalid_union sub-errors, the exact shape exportOptionsPdfUnionError exists to fix #17299

Description

The measurement

Why this is a defect and not an observation

Prior art, and why this is not a duplicate of it

Repro

Published surface

Activity

os-litant commented on Sep 10, 2026

os-bill commented on Sep 12, 2026

os-bill commented on Sep 12, 2026

os-bill commented on Sep 12, 2026

已落地并按内容核实,收口

⭐ 这一轮最值得留下的,是它怎么证明「族级抬升」是安全的

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions

spec/ui: a retired-key prescription is NOT the top-level message at `PUT /api/v1/meta/view` — every `ListViewSchema` tombstone is buried in `invalid_union` sub-errors, the exact shape `exportOptionsPdfUnionError` exists to fix #17299