Skip to content

test(search): fix flaky authorsSort property generators #1803

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ClemRz merged 2 commits into from
Sep 4, 2026
Merged

test(search): fix flaky authorsSort property generators #1803

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
df80ec2
test(search): stop generating lone surrogates in authorsSort names
ClemRz Sep 4, 2026
c0cb970
test(search): derive authorsSort survival from the helper, not trim()
ClemRz Sep 4, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 19 additions & 4 deletions test/integration/2_utils/computeDocumentAuthorsSort.property.test.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,8 +24,14 @@ const nameArb = fc.oneof(
maxLength: 15,
}),
fc.string({
// Iterate by code point, not by `.split('')`: the astral-plane samples
// below (U+2000B, U+2A600) are two UTF-16 code units each, so splitting by
// code unit would offer their surrogate halves as standalone units and
// generate names holding lone surrogates. Those have no UTF-8 encoding —
// every one of them serializes to U+FFFD — which collapses names this
// helper orders as distinct and makes Property 4 fail on ~9% of seeds.
unit: fc.constantFrom(
...'ЯрославльΩμέγα山田太郎홍길동محمدמשהสมชาย𠀋𪘀'.split(''),
...'ЯрославльΩμέγα山田太郎홍길동محمدמשהสมชาย𠀋𪘀',
'\u{10FFFD}',
'�',
'~'
Expand All @@ -41,9 +47,18 @@ const namesArb = fc.array(fc.oneof(nameArb, fc.constantFrom(null, undefined)), {
maxLength: 5,
});

/** Arbitrary: a list holding at least one name that survives normalization. */
const nonEmptyNamesArb = namesArb.filter((names) =>
names.some((n) => typeof n === 'string' && n.trim().length > 0)
/**
* Arbitrary: a list holding at least one name that survives normalization.
*
* Survival is decided by asking the helper itself, not by `.trim()`: `trim`
* only strips whitespace, while normalizeName strips combining marks first, so
* a name made purely of diacritics is non-blank to `trim` yet normalizes away.
* U+1FDD, say, decomposes under NFKD to a space plus two combining marks and
* leaves nothing behind — a `trim`-based precondition would hand such a name to
* the properties below as "authored" and they would rightly disagree.
*/
const nonEmptyNamesArb = namesArb.filter(
(names) => computeDocumentAuthorsSort(names) !== EMPTY_AUTHORS_SORT_KEY
);

/** Strip the ordering bucket prefix to recover the normalized name. */
Expand Down