test(spec): the first data/ file group's test titles state each cited decision in words instead of a tracker number (stage 15)

[objectstack-fleet]

Part of #20749
Clause-②: no

Stage 15 of this card: the next area of class (e), the test strings shipped under packages/spec/src, as ruled in 5902360492 on #20513. This stage takes the first name-ordered file group directly under packages/spec/src/data/: the 20 test files from aggregate-field-type-compatibility.test.ts to date-range-presets.test.ts. They carried 94 messages and 102 tracker ids, citing 60 records. Every one of those ids now either states what its record decided, in words (form D), or is dropped where the title already says it. Text only: no assertion, identifier, test count or code comment changes.

Census at the base (0a3480311a)

Instruments: census10.cjs (md5 9d08602ab972b4b8643c90d64d40fa41), census.cjs (md5 6e42a45a926d375013c32d62f16a296e), census-wide.cjs (md5 c98410a19529c439adb0afbfb00026a2) and dirtable.cjs (md5 dda605c54745b4a60cc14c9a686e4eff). They are byte-identical to the copies stages 10 to 14 used. A literal counts as a test title when its folded message is argument 0 of a describe / it / test call, .each / .skip / .only chains included. Everything else is an "other" string.

Both instruments read 1325 messages / 1406 ids in 282 files, the seat's reading at 0a3480311a (stage 14's head).

directory	files	messages / ids	titles	other
data/ (this PR: the first 20 files)	95	468 / 501	445 / 475	23 / 26
ui/	81	393 / 416	375 / 398	18 / 18
api/	40	189 / 201	181 / 193	8 / 8
system/	34	154 / 165	128 / 138	26 / 27
(files directly in src/)	30	118 / 120	117 / 119	1 / 1
ai/	1	2 / 2	0	2 / 2
contracts/	1	1 / 1	0	1 / 1
total	282	1325 / 1406	1246 / 1323	79 / 83

The group reads 94 messages / 102 ids in 20 files, the seat's figures, file for file:

file (under data/)	messages / ids	titles	other
aggregate-field-type-compatibility.test.ts	4 / 4	4 / 4	0
analytics-date-range-closed-vocabulary.test.ts	1 / 1	1 / 1	0
analytics-date-range-two-bound-window.test.ts	3 / 3	3 / 3	0
analytics-query-window-integer.test.ts	2 / 2	2 / 2	0
analytics-strictness-batchd.test.ts	9 / 9	9 / 9	0
analytics.test.ts	6 / 6	6 / 6	0
api-derivation.test.ts	6 / 6	6 / 6	0
api-methods-batch-conformance.test.ts	3 / 4	1 / 1	2 / 3
authoring-key-lint.test.ts	2 / 2	2 / 2	0
autonumber-format.test.ts	3 / 3	3 / 3	0
autonumber-unanchored-boundary.test.ts	3 / 4	3 / 4	0
bulk-write-hook-conformance.test.ts	2 / 2	2 / 2	0
calendar-day.test.ts	2 / 2	2 / 2	0
context-tokens.test.ts	1 / 1	1 / 1	0
currency-mode-family-closure.pin.test.ts	2 / 2	2 / 2	0
currency-precision-iso4217.test.ts	7 / 7	7 / 7	0
data-engine.test.ts	20 / 25	20 / 25	0
datasource-credential-redaction.test.ts	6 / 6	6 / 6	0
datasource.test.ts	10 / 10	10 / 10	0
date-range-presets.test.ts	2 / 3	2 / 3	0
20 files	94 / 102	92 / 99	2 / 3

• Controls. Lit, a title: data/document.test.ts reads 2 / 2 at the head. Lit, "other" strings: the two in data/external-lookup-retirement.test.ts (:89, :130) still read at the head. Dark: the file comment at data/analytics-strictness-batchd.test.ts:4 (it names the strictness batch by its number) reads 0. Planted in a scratch copy of the head data/calendar-day.test.ts: an id put into a title reads 1 / 1, and an id put into a comment reads 0.
• A wider pattern (any # plus digits) reads the same totals in 19 of the 20 files. In aggregate-field-type-compatibility.test.ts it reads one more, a decision-batch number at :150 that sits beside a cited record in the same literal. The gate's pattern needs three to five digits, so it is not counted there.
• At the head: 1231 messages / 1304 ids in 262 files. The 20 files read 0 / 0 on both patterns, and no other file moved.

How the area was chosen

data/ has no subdirectory to split by (449 ids directly under it, data/driver/ 52), so its stages take name-ordered file groups near the ~100-id bound, as stage 14's report proposed. This census reads the first group at exactly 102, the claim's figure, so the rule needed no re-cut.

Named for the next stages (re-cut from the head census, 1231 / 1304; data/ 374 / 399 left):

• data/ in four more stages, name-ordered:
  1. default-value-shape.test.ts to filter-comparand-shape.test.ts: 20 files, 94 messages / 100 ids;
  2. filter-comparand-type.test.ts to filter-view-operator-parity.test.ts: 20 files, 95 / 99;
  3. filter.test.ts to object.test.ts: 17 files, 103 / 114. object.test.ts alone carries 42, so no cut lands nearer the bound;
  4. query-transport.test.ts to validation.test.ts (11 files, 34 / 34) with data/driver/ (7 files, 48 / 52): 86 ids.
• ui/ 416, about four stages. api/ 201, two. system/ 165, two. The files directly in src/, 120, one.
• The three docblock needles (ai/build-progress.test.ts:236, :237, contracts/approval-service.test.ts:274), one stage with their docblocks.

What each id became

24 literals (28 ids) now state a decision in words. 2 literals (2 ids) get their subject back in words where the number stood in for it. 69 literals (72 ids) drop a number the title already explains. (95 literals in 94 messages: the sys_organization reason string is one message over two lines.)

Every cited record was read with its comments through REST: 55 answer 200. #6345, #8876, #9040, #10194 and #17014 answer 404, and their decisions were read from what landed: e2798fa (one driver vocabulary for start and migrate), d634e66 (the username half of the URL userinfo grammar), 2420641 (a credential in the mongo options passthrough is refused), 2306a76 (theme / analytics_cube validated at the /meta write door) and 80aef80 (a one-day window for the one-day presets), each with its CHANGELOG entry. No cross-repo record is cited in this group.

record(s)	literal (under data/)	now reads
#11152	aggregate-field-type-compatibility.test.ts:150	"accepts sum / avg / min / max over booleans — numbers on every backend, a ruling that outranks the refused-by-default rule". The maintainer ruled that booleans aggregate as numbers on every backend; decision batch 80 held that ruling over batch 59's blanket refusal of unnamed pairs. That batch number went with the id.
#4001 (3)	analytics-strictness-batchd.test.ts:83, :248, :306	"batch D, unknown keys refused — …" before "the doors the cube family is reachable through", "alias claims are true of the surfaces they point at" and "deliberate non-closures (re-verdicts, not omissions)". The campaign's decision: an unknown key is refused, not stripped.
#3878 (2)	analytics-strictness-batchd.test.ts:270, :297	"matching the dispatcher's bespoke hint at the /analytics entry" and "the retired-envelope tombstones still fire". The body is the bare AnalyticsQuery; the { cube, query } envelope was retired with tombstones, and the entry answers 400 with a hint at where.
#18612	analytics.test.ts:314	"a persisted cube heals at the door — the retired join sql / relationship are stripped (ADR-0087 D2)".
#3391	api-derivation.test.ts:16	"api-derivation — one table resolves the effective operations from six primitives". The server is the only adjudicator, through one derivation table.
#3543	api-derivation.test.ts:286	"vocabulary split — authors write six primitives, the wire speaks operations". The authored enum shrank; the wire vocabulary stayed byte-stable.
#15873	api-methods-batch-conformance.test.ts:221	A declared reason string: "(a ruling grants update; both are column-clamped per row by ADR-0092 D2)". Option (a), decision batch 64.
#3786	authoring-key-lint.test.ts:37	"lintAuthoredRecordKeys — an unknown authoring key is reported, not swallowed", the decision its source docblock records.
#6555	autonumber-format.test.ts:23	"DEFAULT_AUTONUMBER_FORMAT / resolveAutonumberFormat — one declared default both sides read". Route 3: {0000} became the contract default, and both fallbacks went away.
#5038	bulk-write-hook-conformance.test.ts:114	"records the after half as DELIVERED — the engine fires it once per row".
#5574	bulk-write-hook-conformance.test.ts:119	"records the before half as DELIVERED — the engine dispatches it per row too".
#20126	currency-mode-family-closure.pin.test.ts:348	"currency-mode family — the enumerating closure pin: defaultCurrency holds only under fixed".
#19992	currency-precision-iso4217.test.ts:163	"the removed currencyConfig.precision at rest: a stored row carrying the baked precision: 2 is served canonical".
#7918	currency-precision-iso4217.test.ts:224	"… where the ISO 4217 width check used to refuse it". That check was the record's option A, later reversed.
#3407, #6437	data-engine.test.ts:1185	"DroppedFieldsEventSchema.reason — why a write dropped submitted fields, widened past the readonly pair".
#6262, #6433, #6435	data-engine.test.ts:1198	"primary_key is the value the engine reports when it strips a payload id it ruled is not an identifier", the schema's own wording of that strip on the bulk and the by-id paths.
#8300	datasource-credential-redaction.test.ts:70	"(the drift guard on the one credential-key definition)".
#8876	datasource-credential-redaction.test.ts:232	"— the username half of the same alignment".
#8337	datasource-credential-redaction.test.ts:243	"redactUrlCredentialQueryParams — the read half: a credential query parameter is never served back".
#8153	datasource.test.ts:673	"— unchanged by the managed-row credentialsRef allowance". The ruling allowed external.credentialsRef, and only it, on managed rows.
#4614, #8793	date-range-presets.test.ts:14	"date-range preset vocabulary — one source of truth, read by both the UI and the data side".

Subject restored (2 ids): #20126 at currency-mode-family-closure.pin.test.ts:403 ("currency-mode closure controls — each rule can fail, and passes what it must") and #7918 at currency-precision-iso4217.test.ts:311 ("carries the measured anchors — 0 digits for JPY, 2 for USD, 3 for KWD"). That literal moved from double to single quotes, since it no longer holds an apostrophe.

Dropped only (72 ids): #1603, #2377, #3026, #3391, #3543, #3545, #3795 (9), #4001 (2), #4286, #4346 (2), #4538, #4583, #5586, #6345, #6555, #6560, #7178 (5), #7265, #7287 (2), #7802 (2), #8032, #8057 (2), #8153 (7), #8336, #8337, #9040, #10194, #10414, #13802, #16041, #16632, #17014, #17296, #17598 (2), #18278, #19992 (3), #20011, #20300 (2), #20550, #20600, #20808 (3), #21365 (2).

• Each of these titles already states the decision it pins: for example "empty array → deny-all (flipped semantics)" for 跟踪:UI 操作按钮与 apiMethods 白名单一致性契约落地(#3026 设计定稿) #3391, "accepts the BARE query string — the canonical ADR-0061 D1 spelling" for EngineQueryOptionsSchema.search rejects the bare query string that ADR-0061 D1 calls canonical — so every engine caller that wants it must as any, losing the whole query's checking #7178, or "currencyConfig.precision is removed: refused with the prescription, whatever its value" for [finding] currencyConfig.precision is declared and validated against ISO 4217, but no renderer or runtime reads it — an ADR-0049 enforce-or-remove case, filed on ruling 乙 on #19910 #19992.
• Small rewordings that carry no new claim: analytics-strictness-batchd.test.ts:307 reads "are CLOSED now" where it named the record; autonumber-unanchored-boundary.test.ts:51 reads "(ruled: mixed content is out of contract)"; datasource.test.ts:553 reads "(the happy path)". The circled part numbers after AnalyticsDateRangeSchema's array arm admits shapes every consumer now refuses, and the one shared refusal sentence is false for them #17598 went with the id.
• The two api-methods-batch-conformance.test.ts reason strings (sys_api_key, sys_organization) end "rather than hitting /batch." now. The table is read only through !== undefined, so no assertion reads their text.

Readers

• Test-name filters: none. A tracked-tree search for -t and --testNamePattern finds only packages/qa/dogfood/README.md:142 (-t "owner-scoped"), which is unrelated.
• Snapshots: none. No __snapshots__ directory exists under data/, and no .snap file is tracked under packages/spec.
• Projects: two touched files are listed in packages/spec/vitest.repo-tests.json: api-methods-batch-conformance.test.ts and currency-mode-family-closure.pin.test.ts. Both were run in the repo project at the base and at the head, and the other 18 in local.
• By substring: every old literal, plus a window around each id (289 needles), was searched across the tracked tree outside its own file. No gate, doc, filter, snapshot or scripts/check-*.mjs self-test reads one. The 14 hits are:
  • sibling titles in other lanes: service-analytics aggregate-nontemporal-measure-refusal.test.ts:344 and objectql engine-autonumber-default-format.test.ts:248;
  • this card's later data/ stage: data/driver/postgres.test.ts:169, the same "placeholders are not resolved here" title, already in the census;
  • comments, CHANGELOG, an audit ledger and liveness evidence: lint validate-dataset-measure-aggregates.test.ts:179, service-analytics dataset-compiler.ts:227, objectql engine.ts:6206 and :6272, analytics.zod.ts:1002, docs/audits/2026-07-unknown-key-strictness-ledger.md:728, two packages/spec/CHANGELOG.md entries and the liveness/field.json:218 evidence string, which quotes the engine.ts comment. None reads a test title.
• Same-text titles named in stage 14's ACCEPT ((#15680), (#5955), #3896 close-out): none falls in this group.

Text-only proof

Stage 10's scratch tool (textonly10.cjs, md5 d5e4801dbb4329ab1984da91e92fc47c) compares base and head file by file on three legs:

1. Skeleton: the full AST, with string pieces masked. It must be identical.
2. Comments: every comment, byte-equal.
3. Strings: each changed string leaf must sit in a test-call title position or on a declared line, must carry a tracker id before, and must carry no # plus digits after. The declared lines are the three reason-string leaves in api-methods-batch-conformance.test.ts.

• Result: 20 of 20 files SAME on all three legs, as predicted in writing before the run.
• Totals: 95 changed literals, 92 titles and 3 declared. The diff's + and - lines are exactly the 95 planned lines, and every file keeps its line count.
• Controls (10 of 10 as predicted, on scratch copies, each anchor hit once): identifier rename DIFF; numeric literal DIFF; comment edit COMMENT DIFF; a non-title string given an id VIOLATION; a rewritten title given a new id VIOLATION; a title that was id-free at base edited VIOLATION; one title reverted to base SAME; a declared string given a new id VIOLATION; an undeclared expect message changed VIOLATION; a title re-split into a + chain DIFF.

Test counts: the 20 files were run at the base, in a separate base worktree, and at the head, with --project local --project repo. Both sides read 553 / 553 passed, with the same count and status sequence per file in 20 of 20. 291 full test names change, and each equals the base name with the planned replacements applied (0 mismatches). No full name repeats on either side.

main merged in, once

#21800 (the console pin bump) landed while this branch was being verified, and it rewrites the comment block at :61-77 of api-methods-batch-conformance.test.ts. This PR edits only string literals in that file, more than 100 lines below the block, so origin/main (18c2ddc1ec, which also carries #21801) was merged in with a plain merge, no rebase, and no conflict. The PR's delta against main is still exactly the 20 files, +95 / -95. Every reading in this body was re-taken on the merged head bf16ad1190, against 18c2ddc1ec as the base: the census (1325 / 1406 there, 1231 / 1304 here, unchanged by the two commits), the text-only proof and its controls (the three declared lines now sit at :202, :230 and :235), the 20-file runs, the full build, the suite, the typecheck and the gates. Re-fetched just before this PR opened, origin/main was one commit further (75ddcd1b41, #21805, in cloud-connection, metadata-core and runtime). It touches no packages/spec path and no file here, so it was not merged.

Changeset: skip-changeset

Measured, not assumed:

• npm pack --dry-run of @objectstack/spec lists 2068 files. 0 of the 20 touched files are in it, and no *.test.ts at all. Of src/, only the *.zod.ts sources ship: the controls src/data/analytics.zod.ts, src/data/data-engine.zod.ts and dist/data/index.js are in it.
• In the built dist/, five new phrases and four old literals each read in 0 files. The control Unrecognized key(s) on reads in 42.

So this PR publishes nothing, and no changeset is added.

Verification (at bf16ad1190)

• pnpm turbo run build over all packages: 71 / 71 (also 71 / 71 at the pre-merge head 89c4b300c2).
• @objectstack/spec:
  • vitest run --project local: 615 files, 18360 passed, 1 todo.
  • typecheck exit 0, including check:test-typecheck (52 files / 246 errors / 135 pinned signatures held). Its program holds all 20 touched files, counted with tsc --listFilesOnly -p tsconfig.test.json.
  • check:generated: all 15 generated artifacts up to date after the merge.
• Gates: dispatch-gates --commands derived 79 families, the same set as stages 13 and 14, and all 79 exit 0. --ran reconciles: 79 derived, 79 run, 0 NOT-MEASURED, 0 UNRUN.
  • The five roster families whose rosters sit under a touched directory were also run, and each exits 0: check:meta-url-spelling, check:spec-changes, check:authz-resolver, check:error-code-casing and check:filter-alias-parity.
• ESLint, a proven narrowing: --no-inline-config over the 20 files reads 0 errors and 0 warnings. The population comes from ESLint's own config: 20 configured, 0 ignored. No file sets parserOptions.project or projectService, so no untouched file's verdict can move.
• check-governed-merges --test: NOT governed, 190 changed lines.

Acceptance notes

• No needle in this group. Every id was a title or a declared reason string; no expected value of an assertion over a source docblock was found. The three known needles are untouched.
• Same-id test titles in other packages are their lanes' test-string shares. A search of describe / it / test lines outside packages/spec finds 156 lines citing ids this PR handled, in 79 files of 23 packages: objectql 73 (29 files), rest 18 (7), runtime 7 (4), plugin-security 6 (5), cli 6 (3), lint 6 (4), service-datasource 6 (3), driver-sql 4 (4), platform-objects 4 (2), plugin-approvals 3 (2), service-automation 3 (2), driver-mongodb 3 (1), plugin-auth 3 (1), service-analytics 3 (3), metadata-core 2 (1), plugin-hono-server 2 (1), and one each in client, triggers, core, metadata-protocol, qa/dogfood, types and driver-memory.
• Two spec test files outside src/ carry same-id titles: packages/spec/scripts/file-description.test.ts:66 and packages/spec/scripts/format-type.test.ts:85. They are outside class (e) as ruled ("the test strings shipped under src/").
• Numeric delivery fields: bulk-write-hook-conformance.test.ts:115-116 and :129-130 assert engineDeliveryIssue: 5038 / 5574, numbers in the source contract table. They are not strings, the gate's pattern cannot see them, and they are not this card's share.
• Code comments still carry ids in these files and their sources, for example the header of analytics-strictness-batchd.test.ts and the SINGLE_RECORD_WRITE_ONLY comments in api-methods-batch-conformance.test.ts. Comments are not this card's share, and none is touched here.

---

Generated by Claude Code

[github-actions]

📓 Docs Drift Check

Nothing in this diff resolved to a documentable surface (no symbol, route or SDK anchor derived from 0 changed package(s)), so this run has no opinion about the docs.

What this run could not see

• 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 — 0 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 75ddcd1b41e74823b7bb9f3fe9e159a189fa2a91 → packageMentionDocs.