This repository was archived by the owner on Nov 4, 2026. It is now read-only.
Repository navigation
Commit dac022c
test(scripts): cite the bare-ValidationRule negatives by file, not by line; mark the two retired names synthetic (objectui#10781) (#10791)
Fixes #10781
Clause-②: no. It is a test-fixture citation, and nothing published
moves.
## What changed
One file, `scripts/__tests__/check-doc-component-types.test.ts`,
comments only. With comments masked, the test's code is byte-identical
to `origin/main` (measured below).
- The negative fixtures of the bare-`ValidationRule` matcher (test "the
bare-`ValidationRule` match discriminates every near-spelling that
really exists") no longer carry line addresses.
- `AdvancedValidationRule` and `ValidationRuleType`: objectui#10719 (PR
objectui#10759) retired both names. No live spelling puts a suffixed
name on the `type … =` branch, so there is nothing truthful to re-cite
`ValidationRuleType` to, and route 1 of the card has no live pair. Both
stay as **SYNTHETIC** negatives (route 2). The comment above them says
so, gives the reason each shape is kept, and points at the retirement
pin `phase35-validation-types-retired-10719.test.ts` in
`packages/types`.
- `ObjectValidationRule`, `DesignerValidationRule`,
`FieldValidationRules`, `ValidationRuleDraft` and
`BaseValidationRuleShape` are re-cited by the file they live in, with no
line. AGENTS.md #11 and the citation gate's own header both say an
address is repaired by turning it into a content anchor, not by moving
the number.
- The block comment no longer says every negative is a spelling this
repository writes. That is only true of the negatives with a file beside
them; the rest are shapes. It also states that nothing re-derives either
kind of citation (AGENTS.md #9).
- The same file has one other stale cross-file address, in the
objectui#5106 key-table block: the `plugin-grid.mdx` specimen of a "none
— `skipFallback: true`" row. It now points at the `view:grid` row by
content. Acceptance notes say why it is in this PR.
## Why this is wider than the card's Finding
The card's Finding names two addresses. Its Acceptance reads: "No
fixture in that file cites a `packages/types/src/**` address that does
not hold the cited spelling at head." Measured at `origin/main`
`bf14b6460`, **none** of the six addresses into `packages/types/src`
held its spelling:
- objectui#10759 removed the two lines the Finding names.
- The other four were wrong before that. They cite `data-protocol.ts`
(twice, including the `BaseValidationRuleShape` quote), `designer.ts`
and `form.ts`, and each already pointed at an unrelated line on
`97b6c21^`, the tree before objectui#10759.
Changing only two would leave the Acceptance false. The seventh fixture
address, app-shell `ObjectValidationsPanel.tsx`, still held; it is
converted anyway so the list uses one citation form.
## Evidence (head `a90af71878`)
| Check | Reading |
|---|---|
| `pnpm exec vitest run
scripts/__tests__/check-doc-component-types.test.ts` (under
`os-verify-lock`) | `Tests 125 passed (125)`, `VERDICT command-exit 0` |
| Masked-code equality against `bf14b6460`: `stripComments` from
`scripts/js-comment-mask.mjs`, with blank lines and trailing spaces
dropped | equal. Positive control: dropping one negative fixture from
the base reads unequal |
| `pnpm check:new-line-citations` | `VERDICT
new-cross-file-line-citations: 0 new citation(s), enforcement
report-only -> exit 0` |
| `node scripts/cross-file-line-citation-census.mjs --json`, rows whose
citing file is this test | 7 before, 0 after. 1 parent bff63cc commit dac022c
1 file changed
Lines changed: 21 additions & 12 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1625 | 1625 | | |
1626 | 1626 | | |
1627 | 1627 | | |
1628 | | - | |
| 1628 | + | |
| 1629 | + | |
1629 | 1630 | | |
1630 | 1631 | | |
1631 | 1632 | | |
| |||
2497 | 2498 | | |
2498 | 2499 | | |
2499 | 2500 | | |
2500 | | - | |
2501 | | - | |
2502 | | - | |
2503 | | - | |
| 2501 | + | |
| 2502 | + | |
| 2503 | + | |
| 2504 | + | |
| 2505 | + | |
| 2506 | + | |
| 2507 | + | |
2504 | 2508 | | |
2505 | 2509 | | |
2506 | 2510 | | |
| |||
2515 | 2519 | | |
2516 | 2520 | | |
2517 | 2521 | | |
2518 | | - | |
2519 | | - | |
2520 | | - | |
2521 | | - | |
2522 | | - | |
2523 | | - | |
2524 | | - | |
| 2522 | + | |
| 2523 | + | |
| 2524 | + | |
| 2525 | + | |
| 2526 | + | |
| 2527 | + | |
| 2528 | + | |
| 2529 | + | |
| 2530 | + | |
| 2531 | + | |
| 2532 | + | |
| 2533 | + | |
2525 | 2534 | | |
2526 | 2535 | | |
2527 | 2536 | | |
| |||
0 commit comments