fix(fls): archive path folding measured against Swift
41 path pairs replayed in the Swift core decide which spellings name the same file. Two corrections: full Unicode case folding (ß, ligatures, final sigma, micro, long s) instead of toLowerCase, and width folding of the whole Halfwidth and Fullwidth Forms block. Swift treats composed and decomposed spellings as one dictionary key; here the writer refuses both. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Claude Opus 5.5
+23
-5
@@ -112,21 +112,39 @@ export function validateEntryPath(path: string): void {
|
||||
}
|
||||
|
||||
/**
|
||||
* Folds full-width and half-width forms (U+FF01…U+FFEF) to their ordinary
|
||||
* counterparts — the part of NFKC that Foundation's `.widthInsensitive` does.
|
||||
* Foundation's `.widthInsensitive`: the Halfwidth and Fullwidth Forms block
|
||||
* (U+FF01…U+FFEE) folds to its NFKC form — full-width ASCII, half-width
|
||||
* katakana, half-width hangul (to conjoining jamo) and the symbols. Measured
|
||||
* against the Swift core (packages/fls/test/path-folding.*.json).
|
||||
*/
|
||||
function foldWidth(text: string): string {
|
||||
return text.replace(/[！-]/g, (character) => character.normalize('NFKC'));
|
||||
return text.replace(/[\uFF01-\uFFEE]/g, (character) => character.normalize('NFKC'));
|
||||
}
|
||||
|
||||
/**
|
||||
* Unicode full case folding, which JavaScript lacks: lower, upper, lower per
|
||||
* character gives it (`ß` → `ss`, `fi` → `fi`, `K` (Kelvin) → `k`, `ς` → `σ`,
|
||||
* `İ` → `i̇`). The one exception is `ı`, which default case folding leaves
|
||||
* alone and the round trip would turn into `i`.
|
||||
*/
|
||||
function foldCase(text: string): string {
|
||||
let folded = '';
|
||||
for (const character of text) {
|
||||
folded += character === '\u0131' ? character : character.toLowerCase().toUpperCase().toLowerCase();
|
||||
}
|
||||
return folded;
|
||||
}
|
||||
|
||||
/**
|
||||
* The key two paths share when they would name the same file on a case- and
|
||||
* width-insensitive file system: NFC, width-folded, case-folded, per component.
|
||||
* width-insensitive file system: NFC, then width and case folding, per
|
||||
* component — `precomposedStringWithCanonicalMapping.folding([.caseInsensitive,
|
||||
* .widthInsensitive])` in Swift.
|
||||
*/
|
||||
export function canonicalEntryKey(path: string): string {
|
||||
return path
|
||||
.split('/')
|
||||
.map((component) => foldWidth(component.normalize('NFC')).toLowerCase())
|
||||
.map((component) => foldCase(foldWidth(component.normalize('NFC'))))
|
||||
.join('/');
|
||||
}
|
||||
|
||||
|
||||
@@ -0,0 +1,207 @@
|
||||
[
|
||||
{
|
||||
"name": "ascii case",
|
||||
"first": "projekt/x/a.txt",
|
||||
"second": "projekt/x/A.TXT"
|
||||
},
|
||||
{
|
||||
"name": "fullwidth letter",
|
||||
"first": "projekt/x/Ａ.txt",
|
||||
"second": "projekt/x/A.txt"
|
||||
},
|
||||
{
|
||||
"name": "fullwidth digit",
|
||||
"first": "projekt/x/１.txt",
|
||||
"second": "projekt/x/1.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth katakana",
|
||||
"first": "projekt/x/ｱ.txt",
|
||||
"second": "projekt/x/ア.txt"
|
||||
},
|
||||
{
|
||||
"name": "sharp s vs SS",
|
||||
"first": "projekt/x/straße.txt",
|
||||
"second": "projekt/x/STRASSE.txt"
|
||||
},
|
||||
{
|
||||
"name": "sharp s vs ss",
|
||||
"first": "projekt/x/straße.txt",
|
||||
"second": "projekt/x/strasse.txt"
|
||||
},
|
||||
{
|
||||
"name": "capital sharp s",
|
||||
"first": "projekt/x/ẞ.txt",
|
||||
"second": "projekt/x/ß.txt"
|
||||
},
|
||||
{
|
||||
"name": "kelvin sign",
|
||||
"first": "projekt/x/K.txt",
|
||||
"second": "projekt/x/k.txt"
|
||||
},
|
||||
{
|
||||
"name": "angstrom sign",
|
||||
"first": "projekt/x/Å.txt",
|
||||
"second": "projekt/x/å.txt"
|
||||
},
|
||||
{
|
||||
"name": "ohm sign",
|
||||
"first": "projekt/x/Ω.txt",
|
||||
"second": "projekt/x/ω.txt"
|
||||
},
|
||||
{
|
||||
"name": "fi ligature",
|
||||
"first": "projekt/x/fi.txt",
|
||||
"second": "projekt/x/fi.txt"
|
||||
},
|
||||
{
|
||||
"name": "composed vs decomposed",
|
||||
"first": "projekt/x/é.txt",
|
||||
"second": "projekt/x/é.txt"
|
||||
},
|
||||
{
|
||||
"name": "dotted capital I",
|
||||
"first": "projekt/x/İ.txt",
|
||||
"second": "projekt/x/i.txt"
|
||||
},
|
||||
{
|
||||
"name": "dotless i",
|
||||
"first": "projekt/x/ı.txt",
|
||||
"second": "projekt/x/I.txt"
|
||||
},
|
||||
{
|
||||
"name": "final sigma",
|
||||
"first": "projekt/x/ς.txt",
|
||||
"second": "projekt/x/σ.txt"
|
||||
},
|
||||
{
|
||||
"name": "micro sign",
|
||||
"first": "projekt/x/µ.txt",
|
||||
"second": "projekt/x/μ.txt"
|
||||
},
|
||||
{
|
||||
"name": "circled letter",
|
||||
"first": "projekt/x/Ⓐ.txt",
|
||||
"second": "projekt/x/ⓐ.txt"
|
||||
},
|
||||
{
|
||||
"name": "superscript two",
|
||||
"first": "projekt/x/².txt",
|
||||
"second": "projekt/x/2.txt"
|
||||
},
|
||||
{
|
||||
"name": "ideographic space",
|
||||
"first": "projekt/x/a b.txt",
|
||||
"second": "projekt/x/a b.txt"
|
||||
},
|
||||
{
|
||||
"name": "umlaut case",
|
||||
"first": "projekt/x/Ä.txt",
|
||||
"second": "projekt/x/ä.txt"
|
||||
},
|
||||
{
|
||||
"name": "fullwidth tilde",
|
||||
"first": "projekt/x/～.txt",
|
||||
"second": "projekt/x/~.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth hangul",
|
||||
"first": "projekt/x/ﾡ.txt",
|
||||
"second": "projekt/x/ㄱ.txt"
|
||||
},
|
||||
{
|
||||
"name": "fullwidth won sign",
|
||||
"first": "projekt/x/￦.txt",
|
||||
"second": "projekt/x/₩.txt"
|
||||
},
|
||||
{
|
||||
"name": "different names",
|
||||
"first": "projekt/x/a.txt",
|
||||
"second": "projekt/x/b.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth left arrow",
|
||||
"first": "projekt/x/￩.txt",
|
||||
"second": "projekt/x/←.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth forms light vertical",
|
||||
"first": "projekt/x/￨.txt",
|
||||
"second": "projekt/x/│.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth katakana middle dot",
|
||||
"first": "projekt/x/･.txt",
|
||||
"second": "projekt/x/・.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth hangul filler",
|
||||
"first": "projekt/x/ﾠ.txt",
|
||||
"second": "projekt/x/ㅤ.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth hangul last",
|
||||
"first": "projekt/x/ￜ.txt",
|
||||
"second": "projekt/x/ㅣ.txt"
|
||||
},
|
||||
{
|
||||
"name": "fullwidth cent sign",
|
||||
"first": "projekt/x/￠.txt",
|
||||
"second": "projekt/x/¢.txt"
|
||||
},
|
||||
{
|
||||
"name": "long s",
|
||||
"first": "projekt/x/ſ.txt",
|
||||
"second": "projekt/x/s.txt"
|
||||
},
|
||||
{
|
||||
"name": "capital sharp s vs ss",
|
||||
"first": "projekt/x/ẞ.txt",
|
||||
"second": "projekt/x/ss.txt"
|
||||
},
|
||||
{
|
||||
"name": "ff ligature",
|
||||
"first": "projekt/x/ff.txt",
|
||||
"second": "projekt/x/ff.txt"
|
||||
},
|
||||
{
|
||||
"name": "dotless i vs i",
|
||||
"first": "projekt/x/ı.txt",
|
||||
"second": "projekt/x/i.txt"
|
||||
},
|
||||
{
|
||||
"name": "dotted capital I vs i dot",
|
||||
"first": "projekt/x/İ.txt",
|
||||
"second": "projekt/x/i̇.txt"
|
||||
},
|
||||
{
|
||||
"name": "greek iota with dialytika and tonos",
|
||||
"first": "projekt/x/ΐ.txt",
|
||||
"second": "projekt/x/ΐ.txt"
|
||||
},
|
||||
{
|
||||
"name": "cherokee small letter",
|
||||
"first": "projekt/x/ꭰ.txt",
|
||||
"second": "projekt/x/Ꭰ.txt"
|
||||
},
|
||||
{
|
||||
"name": "fullwidth lower vs ascii upper",
|
||||
"first": "projekt/x/ａ.txt",
|
||||
"second": "projekt/x/A.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth hangul vs its NFKC form",
|
||||
"first": "projekt/x/ﾡ.txt",
|
||||
"second": "projekt/x/ᄀ.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth hangul filler vs its NFKC form",
|
||||
"first": "projekt/x/ﾠ.txt",
|
||||
"second": "projekt/x/ᅠ.txt"
|
||||
},
|
||||
{
|
||||
"name": "halfwidth hangul last vs its NFKC form",
|
||||
"first": "projekt/x/ￜ.txt",
|
||||
"second": "projekt/x/ᅵ.txt"
|
||||
}
|
||||
]
|
||||
@@ -0,0 +1,166 @@
|
||||
[
|
||||
{
|
||||
"name" : "ascii case",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "fullwidth letter",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "fullwidth digit",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth katakana",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "sharp s vs SS",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "sharp s vs ss",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "capital sharp s",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "kelvin sign",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "angstrom sign",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "ohm sign",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "fi ligature",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "composed vs decomposed",
|
||||
"outcome" : "sameKey"
|
||||
},
|
||||
{
|
||||
"name" : "dotted capital I",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "dotless i",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "final sigma",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "micro sign",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "circled letter",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "superscript two",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "ideographic space",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "umlaut case",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "fullwidth tilde",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth hangul",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "fullwidth won sign",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "different names",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth left arrow",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth forms light vertical",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth katakana middle dot",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth hangul filler",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth hangul last",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "fullwidth cent sign",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "long s",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "capital sharp s vs ss",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "ff ligature",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "dotless i vs i",
|
||||
"outcome" : "distinct"
|
||||
},
|
||||
{
|
||||
"name" : "dotted capital I vs i dot",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "greek iota with dialytika and tonos",
|
||||
"outcome" : "sameKey"
|
||||
},
|
||||
{
|
||||
"name" : "cherokee small letter",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "fullwidth lower vs ascii upper",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth hangul vs its NFKC form",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth hangul filler vs its NFKC form",
|
||||
"outcome" : "collide"
|
||||
},
|
||||
{
|
||||
"name" : "halfwidth hangul last vs its NFKC form",
|
||||
"outcome" : "collide"
|
||||
}
|
||||
]
|
||||
@@ -0,0 +1,52 @@
|
||||
// Which archive paths name the same file: the TypeScript folding against what
|
||||
// the Swift core decides for the same pairs (path-folding.swift.json, written
|
||||
// by tools/swift-oracle/run.sh; the Swift folding is Linux Foundation's).
|
||||
|
||||
import { readFileSync } from 'node:fs';
|
||||
import { dirname, join } from 'node:path';
|
||||
import { fileURLToPath } from 'node:url';
|
||||
import { describe, expect, it } from 'vitest';
|
||||
import { FlsArchiveError, writeZipEntries } from '../src/index.js';
|
||||
|
||||
const here = dirname(fileURLToPath(import.meta.url));
|
||||
const cases = JSON.parse(readFileSync(join(here, 'path-folding.cases.json'), 'utf8')) as {
|
||||
name: string;
|
||||
first: string;
|
||||
second: string;
|
||||
}[];
|
||||
const swift = new Map(
|
||||
(JSON.parse(readFileSync(join(here, 'path-folding.swift.json'), 'utf8')) as { name: string; outcome: string }[]).map(
|
||||
(item) => [item.name, item.outcome],
|
||||
),
|
||||
);
|
||||
|
||||
function outcome(first: string, second: string): string {
|
||||
try {
|
||||
writeZipEntries(
|
||||
new Map([
|
||||
[first, Uint8Array.of(1)],
|
||||
[second, Uint8Array.of(2)],
|
||||
]),
|
||||
);
|
||||
return 'distinct';
|
||||
} catch (error) {
|
||||
if (error instanceof FlsArchiveError) return error.code === 'duplicateEntry' ? 'collide' : error.code;
|
||||
throw error;
|
||||
}
|
||||
}
|
||||
|
||||
describe('archive path folding matches Swift', () => {
|
||||
it('has a Swift result for every case', () => {
|
||||
expect(cases.every((item) => swift.has(item.name))).toBe(true);
|
||||
});
|
||||
|
||||
for (const item of cases) {
|
||||
it(item.name, () => {
|
||||
// Swift strings compare by canonical equivalence, so a composed and a
|
||||
// decomposed spelling are one dictionary key there ("sameKey"). A Map
|
||||
// here keeps both — and the writer then refuses them as one file.
|
||||
const expected = swift.get(item.name) === 'sameKey' ? 'collide' : swift.get(item.name);
|
||||
expect(outcome(item.first, item.second)).toBe(expected);
|
||||
});
|
||||
}
|
||||
});
|
||||
Reference in New Issue
Block a user