docs(i18n): correct the stale NOT empty clause in the i18n baseline note

[yinlianghui-tw]

Fixes#6164

Verified at d76de60c9.

scripts/i18n-call-site-key-baseline.json's note claimed missingMembers was NOT empty. #6023 / PR #6163 took that list to zero, so the clause became false. One line changes; nothing else in the file, and nothing else in the repo.

- "missingMembers is a THIRD list (objectui#4964), NOT part of that pair and NOT empty -- so",
+ "missingMembers is a THIRD list (objectui#4964), NOT part of that pair and empty since objectui#6023 -- so",

Premise re-verified, not inherited

An earlier round on this card stopped because missingMembers still held its 2 entries. That stop was correct then, so the premise was re-measured rather than taken from the dispatch:

• PR fix(i18n): define home.recentApps.itemType.report and .metadata in all ten packs #6163 reports "merged":true, merged_at 2026-08-24T22:25:09Z.
• It was squash-merged, so git branch -r --contains cd8d6594d still prints only its own feature branch — the containment check has to run on the squash commit. git merge-base --is-ancestor b08b7eba9 origin/main exits 0, and b08b7eba9 fix(i18n): define home.recentApps.itemType.report and .metadata in all ten packs (#6163) is in git log origin/main.
• The decisive check, on origin/main at e5263ed33: missingKeys len 0, missingPrefixes len 0, missingMembers len 0. The clause is false; the edit is owed.

The note as it stood on main, before editing

 7 'Of the three lists in this file, missingKeys and missingPrefixes are the PAIR that'
8 'objectui#3546 ratcheted to zero -- BOTH LISTS ARE NOW EMPTY. The 258-key stock it opened'
9 'with was paid off across seven slices, the last of them two template families. Keep both'
10 'of that pair: empty is their terminal, load-bearing state. Any NEW unresolved call-site'
11 'key is `unexpected` against an empty baseline and fails the build, which is the point.'
12 'missingMembers is a THIRD list (objectui#4964), NOT part of that pair and NOT empty -- so'
13 'the sentence above counts missingKeys and missingPrefixes only, never all three. It holds'
14 'debt the exact-member check MEASURED on its first run, not debt this repo took on.'

Only element 12 changes. The pair sentinel (element 8), both in-note occurrences of fails the build (elements 1 and 11) and the third-list scoping — NOT part of that pair plus element 13's counts missingKeys and missingPrefixes only, never all three — are kept byte-for-byte. The element count is unchanged at 15, so no prose was reflowed across an element boundary; that matters because the guard asserts on the joined array and is sensitive to the join spaces.

Both sentinels are load-bearing, and both were proved so

residue-namespaces-3546.test.tsx:792-793 asserts two literal substrings, not one. One red proves one string is checked, so each was ablated separately, each with the other left intact as a control. Every mutation was confirmed on disk by anchored grep -c on the target text before the result was read, and each leg restored under a trap ... EXIT INT TERM running git checkout HEAD -- followed by the path (the change was committed first, so the restore had a commit to return to). git status --porcelain was empty after both.

Leg A — mutate the sentinel to the exact tempting rewrite the card warns about (BOTH LISTS ARE NOW EMPTY to ALL THREE LISTS ARE NOW EMPTY; fails the build count held at 3 as a control). Guard exit 1:

AssertionError: expected 'Keys a t() call site references that …' to contain 'BOTH LISTS ARE NOW EMPTY'
Tests 1 failed | 42 passed (43)

Leg B — mutate only the two in-note occurrences of fails the build to breaks the build, leaving the sentinel present and leaving the third occurrence (in //missingMembers) untouched as a control. That control is the interesting half: grep -c "fails the build" still returns 1 for the file, yet the guard reddens — so the assertion reads note specifically, not the file. Guard exit 1:

AssertionError: expected 'Keys a t() call site references that …' to contain 'fails the build'
Tests 1 failed | 42 passed (43)

Restored, then green again — exit 0, the same test named:

✓ |dom| … the ratchet is empty — this is the terminal state, not a partial one 1ms
Test Files 1 passed (1) / Tests 43 passed (43)

Gates

Every exit code captured before any pipe (redirect to a file, then read it), and each line quoted is the gate's own verdict, not a bare $?. Root vitest only, per objectui#3378.

• pnpm check:i18n-keys — exit 0: Every in-scope call-site key resolves against the en pack (2822 keys), … and every dynamic key family either checks its members against a declared vocabulary or says in writing why it has none.
• pnpm vitest run packages/i18n/src/__tests__ scripts/__tests__/check-i18n-call-site-keys.test.ts — exit 0, Test Files 55 passed (55) / Tests 993 passed (993).
• node scripts/check-changeset-presence.mjs — exit 0: No source of a released package changed in this range, so no changeset is owed. No changeset, and no skip-changeset label — that label does not exist in this repo.
• node scripts/check-control-bytes.mjs — exit 0, OK (scanned 5101 tracked text file(s); skipped 85 binary). Also scanned the edited file directly with a grep -naP control-character class — no match.
• JSON parse and key stability, asserted rather than eyeballed: the file parses; top-level key order is identical; missingKeys, //, missingPrefixes, //missingMembers and missingMembers all compare equal to HEAD; note differs at exactly one index, 12.
• Lint narrowing is a measurement, not a skip: eslint's own config puts this file outside its universe — File ignored because no matching configuration was supplied — so the diff's lintable set is empty. CI runs the full farm regardless.

Reported, not fixed

The same merge left two more stale statements in this file, both outside this card's one-clause scope: note's closing It holds debt … (present tense about an empty list) and //missingMembers's Both entries below … (there are now zero entries below). Widening the diff would have spent this card's safety margin on prose that is not load-bearing, so both are filed as #6179 with the ablation findings carried across. The PM decides whether the tense of a comment is above the line.

---

Generated by Claude Code