locale-ai 0.6.3 → 0.6.5

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -31,11 +31,11 @@ Download 577 MB? (Y/n)
31
31
  - **Layouts**: `locales/en/*.json` (nested folders too) → `locales/fr/...`, or `messages/en.json` → `messages/fr.json`. `.json`, `.yaml`, `.yml`, `.po` (gettext), `.arb` (Flutter) and `.strings` (Apple) are all supported, mixed freely in the same project. Common folders (`locales`, `src/locales`, `public/locales`, `src/i18n`, `messages`, `lang`, ...) are found automatically, or pass `--input`.
32
32
  - **Placeholders are never translated or lost**: `{name}`, `{{count}}`, `%s`, `%1$s`, `%(name)s`, `%{name}`, `${name}`, `:attribute`, `$t(key)`, `@:key`, `<b>…</b>`, `<0>…</0>`, `&nbsp;`, URLs, e-mails.
33
33
  - **ICU MessageFormat**: `{count, plural, one {# file} other {# files}}`, `select`, nesting. Only the text is translated.
34
- - **i18next plurals**: adds the forms the target language needs (`_few`, `_many` for Arabic, Russian, Polish...), so the app doesn't fall back to the English string — each added form copies the `_other` translation verbatim rather than inflecting it, so review and adjust the wording of `_zero`/`_two`/`_few`/`_many` by hand for languages where those forms use a different noun/adjective form.
34
+ - **i18next plurals**: adds the forms the target language needs (`_few`, `_many` for Arabic, Russian, Polish...), so the app doesn't fall back to the English string — each added form copies the `_other` translation verbatim rather than inflecting it, so review and adjust the wording of `_zero`/`_two`/`_few`/`_many` by hand for languages where those forms use a different noun/adjective form. The `_one` form is translated with its `{{count}}` temporarily replaced by a literal "1", since the model otherwise sometimes picks the target's plural noun form despite the English source being grammatically singular.
35
35
  - **Incremental**: only missing keys are translated, existing translations are kept, re-runs are no-ops. `--overwrite` re-translates everything.
36
36
  - **Structure**: key order, numbers, booleans, arrays, indentation and trailing newline are preserved.
37
37
 
38
- **Format notes**: for `.po`, `msgid` is the stable key (never changed) and `msgstr` is the translated value, as gettext expects — the English `.po` file's `msgstr` must hold the source text (most generators already fill it; an empty `msgstr` is treated as empty, not inherited from `msgid`), and `msgid_plural`/`msgstr[n]` plural entries pass through untranslated (translate those by hand, or use ICU plurals in `.arb`/JSON/YAML instead). For `.arb`, `@key` metadata and `@@locale` are never translated; `@@locale` is set to the target locale on output.
38
+ **Format notes**: for `.po`, `msgid` is the stable key (never changed) and `msgstr` is the translated value, as gettext expects — the English `.po` file's `msgstr` must hold the source text (most generators already fill it; an empty `msgstr` is treated as empty, not inherited from `msgid`). `msgid_plural`/`msgstr[n]` plural entries are translated too (the singular and plural source forms), then expanded to however many `msgstr[n]` the target language's plural rules need (6 for Arabic, 1 for Japanese, ...) — not full grammatical inflection for in-between categories (`few`/`many`/...), which get the plural-sourced translation verbatim, the same simplification `addPluralForms` makes for i18next. `Plural-Forms` in the header is computed for the target language, not copied from the source; an existing target file's header is otherwise left as you last edited it (a hand-fixed `Plural-Forms`, `Last-Translator`, ...) instead of being regenerated on every rerun. For `.arb`, `@key` metadata and `@@locale` are never translated; `@@locale` is set to the target locale on output.
39
39
 
40
40
  ## Why locale-ai?
41
41
 
@@ -50,6 +50,7 @@ Download 577 MB? (Y/n)
50
50
 
51
51
  ```bash
52
52
  locale-ai translate --to fr,pt-BR [--from en] [--input dir] [--dry-run] [--overwrite] [--fast] [--report file] [-y]
53
+ locale-ai check [--from en] [--input dir] # CI: fails if a translated locale has missing keys or placeholder mismatches
53
54
  locale-ai models list [--from en] # supported language pairs
54
55
  locale-ai models installed
55
56
  locale-ai models install en-de # pre-download (e.g. in CI)
@@ -68,6 +69,7 @@ Models are cached once per machine in `~/.cache/locale-ai` (`LOCALE_AI_CACHE` to
68
69
  ```json
69
70
  {
70
71
  "doNotTranslate": ["Acme", "Acme Pro"],
72
+ "doNotTranslateKeys": ["brand.name", "config.statusEnum"],
71
73
  "glossary": {
72
74
  "fr": { "Workspace": "Espace de travail" },
73
75
  "de": { "Sign in": "Einloggen" }
@@ -77,6 +79,8 @@ Models are cached once per machine in `~/.cache/locale-ai` (`LOCALE_AI_CACHE` to
77
79
 
78
80
  `glossary` entries replace whole strings. `models` adds your own translation models (next section). A built-in glossary covers ~60 standard UI labels ("Sign in", "Settings", "Log out", ...) for fr, es, de, it, pt-BR, nl and ar, because one-to-three word labels are where sentence-level MT is weakest.
79
81
 
82
+ `doNotTranslate` protects a term everywhere it appears in any string. `doNotTranslateKeys` instead protects a key path (dot/`[i]` notation, e.g. `brand.name`, `items[0]`) — the whole value at that path, and everything nested under it if it's an object or array, is copied to every target verbatim. Use it for an enum array (`"statusEnum": ["ACTIVE", "PENDING"]`, which you'd otherwise have to list value-by-value in `doNotTranslate`) or a field a model might respell or inflect even when the word is listed in `doNotTranslate` elsewhere in the same string.
83
+
80
84
  ## Add a language, or a better model
81
85
 
82
86
  Any [MarianMT / OPUS-MT](https://huggingface.co/Helsinki-NLP) model on Hugging Face can be used. Convert it once (needs Python):
package/dist/cli.js CHANGED
@@ -9,14 +9,19 @@ import { CATALOG, customEntries, findById, reachableTargets, resolveRoute, } fro
9
9
  import { translateStrings } from "./engine.js";
10
10
  import { CONFIG_FILE, glossaryLookup, loadConfig, protectRegex } from "./glossary.js";
11
11
  import { intlCode, parseLocale } from "./lang.js";
12
- import { addPluralForms, buildTarget, detectLayout, pendingStrings, readLocaleDoc, sourceFiles, targetFile, } from "./locales.js";
12
+ import { addPluralForms, buildTarget, detectLayout, existingTargets, flattenStrings, keyPathProtect, pendingStrings, pluralOneStrings, readLocaleDoc, sourceFiles, targetFile, } from "./locales.js";
13
13
  import { cacheRoot, install, isInstalled, loadTranslator, localModelProblem, planInstall, readManifest, uninstall, } from "./models.js";
14
+ import { mask } from "./placeholders.js";
14
15
  /** Read from package.json so `npm version` is the single source of truth. */
15
16
  const VERSION = JSON.parse(fs.readFileSync(fileURLToPath(new URL("../package.json", import.meta.url)), "utf8")).version;
16
17
  const HELP = `locale-ai ${VERSION} — translate i18n locale files locally, on CPU, for free.
17
18
 
18
19
  Usage:
19
20
  locale-ai translate --to <locales> [options]
21
+ locale-ai check [--from <locale>]
22
+ Validate that already-translated locales have the same keys and
23
+ placeholders as the source, without translating. Exits non-zero
24
+ on a mismatch (CI-friendly).
20
25
  locale-ai models list [--from <locale>]
21
26
  locale-ai models installed
22
27
  locale-ai models install <from-to | model-id> [--yes]
@@ -153,8 +158,10 @@ async function cmdTranslate(argv) {
153
158
  const config = loadConfig(process.cwd(), values.config);
154
159
  const catalog = catalogFor(config);
155
160
  const protect = protectRegex(config);
161
+ const isProtectedKey = keyPathProtect(config.doNotTranslateKeys);
156
162
  const layout = detectLayout(process.cwd(), from.tag, values.input);
157
163
  const sources = sourceFiles(layout, from.tag).map((f) => ({ file: f, doc: readLocaleDoc(f, from.tag) }));
164
+ const singularHints = new Set(sources.flatMap(({ doc }) => [...pluralOneStrings(doc.data)]));
158
165
  log(`locale-ai ${VERSION}`);
159
166
  log(`Found ${sources.length} ${from.tag} file${sources.length === 1 ? "" : "s"} in ${rel(layout.root)}/` +
160
167
  (layout.kind === "dir" ? `${from.tag}/` : ""));
@@ -172,7 +179,13 @@ async function cmdTranslate(argv) {
172
179
  const files = sources.map(({ file, doc }) => {
173
180
  const target = targetFile(layout, from.tag, locale.tag, file);
174
181
  const existing = fs.existsSync(target) ? readLocaleDoc(target, locale.tag).data : undefined;
175
- return { source: file, target, doc, existing, pending: pendingStrings(doc.data, existing, values.overwrite) };
182
+ return {
183
+ source: file,
184
+ target,
185
+ doc,
186
+ existing,
187
+ pending: pendingStrings(doc.data, existing, values.overwrite, isProtectedKey),
188
+ };
176
189
  });
177
190
  const unique = new Set(files.flatMap((f) => f.pending)).size;
178
191
  jobs.push({ locale, route, files, unique });
@@ -214,6 +227,7 @@ async function cmdTranslate(argv) {
214
227
  batchSize,
215
228
  protect,
216
229
  lookup: glossaryLookup(from, job.locale, config),
230
+ singularHints,
217
231
  onProgress: makeProgress((d, t) => `${job.locale.tag}: translating ${bar(d, t)} ${d}/${t}`),
218
232
  });
219
233
  endStatus();
@@ -230,7 +244,7 @@ async function cmdTranslate(argv) {
230
244
  let written = 0;
231
245
  let pluralsAdded = 0;
232
246
  for (const f of job.files) {
233
- const built = buildTarget(f.doc.data, f.existing, (s) => translations.get(s) ?? s, values.overwrite);
247
+ const built = buildTarget(f.doc.data, f.existing, (s) => translations.get(s) ?? s, values.overwrite, isProtectedKey);
234
248
  const { doc, added } = addPluralForms(built, intlCode(job.locale));
235
249
  pluralsAdded += added;
236
250
  const before = fs.existsSync(f.target) ? fs.readFileSync(f.target, "utf8") : undefined;
@@ -370,11 +384,70 @@ async function cmdModels(argv) {
370
384
  return sub ? 1 : 0;
371
385
  }
372
386
  }
387
+ /**
388
+ * Validates already-translated locales against the source without translating:
389
+ * same key structure, and the same placeholders in each matching string. Meant for
390
+ * CI, so a mismatch (a key a translator dropped, a {name} a translation lost) fails
391
+ * the build instead of shipping silently.
392
+ */
393
+ async function cmdCheck(argv) {
394
+ const { values } = parseArgs({
395
+ args: argv,
396
+ options: {
397
+ from: { type: "string", default: "en" },
398
+ input: { type: "string" },
399
+ },
400
+ });
401
+ const from = parseLocale(values.from);
402
+ const layout = detectLayout(process.cwd(), from.tag, values.input);
403
+ const sources = sourceFiles(layout, from.tag).map((f) => ({ file: f, doc: readLocaleDoc(f, from.tag) }));
404
+ const targets = existingTargets(layout, from.tag);
405
+ if (!targets.length) {
406
+ log(`No translated locales found next to ${from.tag}.`);
407
+ return 0;
408
+ }
409
+ let problems = 0;
410
+ for (const to of targets) {
411
+ let missing = 0;
412
+ let mismatched = 0;
413
+ for (const { file, doc } of sources) {
414
+ const target = targetFile(layout, from.tag, to, file);
415
+ if (!fs.existsSync(target)) {
416
+ missing += flattenStrings(doc.data).length;
417
+ log(`✗ ${to}: ${rel(target)} is missing entirely`);
418
+ continue;
419
+ }
420
+ const srcFlat = flattenStrings(doc.data);
421
+ const tgtByPath = new Map(flattenStrings(readLocaleDoc(target, to).data).map((e) => [e.path, e.value]));
422
+ for (const { path: p, value: srcVal } of srcFlat) {
423
+ const tgtVal = tgtByPath.get(p);
424
+ if (tgtVal === undefined) {
425
+ missing++;
426
+ log(`✗ ${to}: ${rel(target)} missing key "${p}"`);
427
+ continue;
428
+ }
429
+ const srcTokens = mask(srcVal, "natural").tokens.toSorted();
430
+ const tgtTokens = mask(tgtVal, "natural").tokens.toSorted();
431
+ if (srcTokens.join("\u0001") !== tgtTokens.join("\u0001")) {
432
+ mismatched++;
433
+ log(`✗ ${to}: ${rel(target)} "${p}" placeholders differ: source [${srcTokens.join(", ")}], target [${tgtTokens.join(", ")}]`);
434
+ }
435
+ }
436
+ }
437
+ if (missing || mismatched)
438
+ problems++;
439
+ else
440
+ log(`✓ ${to}`);
441
+ }
442
+ return problems ? 1 : 0;
443
+ }
373
444
  async function main() {
374
445
  const [cmd, ...rest] = process.argv.slice(2);
375
446
  switch (cmd) {
376
447
  case "translate":
377
448
  return cmdTranslate(rest);
449
+ case "check":
450
+ return cmdCheck(rest);
378
451
  case "models":
379
452
  return cmdModels(rest);
380
453
  case "-v":
package/dist/engine.js CHANGED
@@ -208,6 +208,36 @@ async function splitFallback(units, hops, opts, progress, issues) {
208
208
  issues.push({ source: unit.owner, kind: "placeholder-fallback", output: unit.result });
209
209
  }
210
210
  }
211
+ const COUNT_PLACEHOLDER = /\{\{\s*count\s*\}\}/;
212
+ /**
213
+ * A "_one" key's source is grammatically singular ("You have {{count}} file"), but the
214
+ * masked marker that stands in for {{count}} carries no number of its own, so the model
215
+ * sometimes still produces the target's PLURAL noun form (confirmed: en→fr "fileCount_one"
216
+ * came back "Vous avez {{count}} fichiers"). Retranslating with a literal "1" in place of
217
+ * {{count}} gives the model the real grammatical cue; only applied when {{count}} is the
218
+ * string's only placeholder (anything else needs the normal masking pipeline to survive),
219
+ * and only kept when the "1" survives translation intact enough to restore {{count}}.
220
+ */
221
+ async function fixSingularGrammar(translations, singularHints, remaining, hops, opts, progress) {
222
+ for (const s of singularHints) {
223
+ if (!remaining.has(s) || translations.get(s) === undefined)
224
+ continue;
225
+ const tokens = mask(s, "natural").tokens;
226
+ if (tokens.length !== 1 || !COUNT_PLACEHOLDER.test(tokens[0]))
227
+ continue;
228
+ const m = COUNT_PLACEHOLDER.exec(s);
229
+ const pinned = s.slice(0, m.index) + "1" + s.slice(m.index + m[0].length);
230
+ const prepared = prepare(pinned);
231
+ const out = (await runModels(new Set([prepared]), hops, opts, progress)).get(prepared) ?? "";
232
+ if (isDegenerate(pinned, out))
233
+ continue;
234
+ const post = postprocess(pinned, out, opts.targetLang);
235
+ const restored = /(?<![\d.])1(?![\d.])/.exec(post);
236
+ if (!restored)
237
+ continue; // can't confirm the "1" survived: keep the original translation
238
+ translations.set(s, post.slice(0, restored.index) + m[0] + post.slice(restored.index + 1));
239
+ }
240
+ }
211
241
  export async function translateStrings(strings, hops, opts) {
212
242
  const issues = [];
213
243
  const progress = { done: 0, total: 0 };
@@ -236,5 +266,7 @@ export async function translateStrings(strings, hops, opts) {
236
266
  const parts = p.units.map((u) => u.result ?? u.text);
237
267
  translations.set(p.source, p.split ? p.split.rebuild(parts) : parts[0]);
238
268
  }
269
+ if (opts.singularHints?.size)
270
+ await fixSingularGrammar(translations, opts.singularHints, new Set(remaining), hops, opts, progress);
239
271
  return { translations, issues };
240
272
  }
package/dist/glossary.js CHANGED
@@ -84,6 +84,8 @@ export function loadConfig(cwd, file) {
84
84
  const cfg = JSON.parse(fs.readFileSync(p, "utf8"));
85
85
  if (cfg.doNotTranslate && !Array.isArray(cfg.doNotTranslate))
86
86
  throw new Error('"doNotTranslate" must be an array');
87
+ if (cfg.doNotTranslateKeys && !Array.isArray(cfg.doNotTranslateKeys))
88
+ throw new Error('"doNotTranslateKeys" must be an array');
87
89
  if (cfg.models && !Array.isArray(cfg.models))
88
90
  throw new Error('"models" must be an array');
89
91
  return { ...cfg, dir: path.dirname(p) };
package/dist/locales.js CHANGED
@@ -82,6 +82,23 @@ export function targetFile(layout, from, to, src) {
82
82
  return path.join(layout.root, `${to}${layout.ext}`);
83
83
  return path.join(layout.root, to, path.relative(path.join(layout.root, from), src));
84
84
  }
85
+ /** Locale tags already translated alongside `from` in this layout (for `locale-ai check`). */
86
+ export function existingTargets(layout, from) {
87
+ if (layout.kind === "file") {
88
+ const ext = layout.ext;
89
+ return fs
90
+ .readdirSync(layout.root)
91
+ .filter((f) => f.endsWith(ext) && f.slice(0, -ext.length) !== from)
92
+ .map((f) => f.slice(0, -ext.length))
93
+ .sort();
94
+ }
95
+ return fs
96
+ .readdirSync(layout.root, { withFileTypes: true })
97
+ .filter((e) => e.isDirectory() && !e.name.startsWith(".") && e.name !== "node_modules" && e.name !== from)
98
+ .map((e) => e.name)
99
+ .filter((name) => listLocaleFiles(path.join(layout.root, name)).length > 0)
100
+ .sort();
101
+ }
85
102
  const finalNewline = (text) => text === "" || text.endsWith("\n");
86
103
  const withNewline = (text, keep) => text.replace(/\n*$/, keep ? "\n" : "");
87
104
  function readJsonDoc(file, text) {
@@ -242,6 +259,10 @@ function pluralFormsHeader(lang) {
242
259
  const branches = Array.from({ length: n - 1 }, (_, i) => `n==${i + 1} ? ${i} :`).join(" ");
243
260
  return `nplurals=${n}; plural=(${branches} ${n - 1});`;
244
261
  }
262
+ /** The `nplurals=N` count from pluralFormsHeader, for expanding msgstr[] to the right length. */
263
+ function pluralCount(lang) {
264
+ return Number(/nplurals=(\d+)/.exec(pluralFormsHeader(lang))[1]);
265
+ }
245
266
  function parsePo(text) {
246
267
  const entries = [];
247
268
  let leading = [];
@@ -318,12 +339,26 @@ function parsePo(text) {
318
339
  flushEntry();
319
340
  return { entries, leading };
320
341
  }
342
+ /** Synthetic data keys for a plural entry's two translatable forms. */
343
+ const pluralOneKey = (e) => poKey(e) + "\u0005one";
344
+ const pluralOtherKey = (e) => poKey(e) + "\u0005other";
321
345
  function readPoDoc(file, text) {
322
346
  const { entries, leading } = parsePo(text);
323
347
  const data = newObj();
324
- for (const e of entries)
325
- if (!e.isHeader && e.msgidPlural === undefined)
348
+ for (const e of entries) {
349
+ if (e.isHeader)
350
+ continue;
351
+ if (e.msgidPlural === undefined)
326
352
  data[poKey(e)] = e.msgstr;
353
+ else {
354
+ // Like msgstr for a singular entry, msgstrPlural holds this FILE's own text: the
355
+ // English source form in the source .po, the translated forms in a target .po
356
+ // (msgid/msgid_plural are the stable keys, identical across every locale).
357
+ const forms = e.msgstrPlural ?? [];
358
+ data[pluralOneKey(e)] = forms[0] ?? "";
359
+ data[pluralOtherKey(e)] = forms[forms.length - 1] ?? forms[0] ?? "";
360
+ }
361
+ }
327
362
  const keepNewline = finalNewline(text);
328
363
  const serialize = (d, targetTag, existingText) => {
329
364
  const obj = isObj(d) ? d : newObj();
@@ -347,7 +382,19 @@ function readPoDoc(file, text) {
347
382
  }
348
383
  else if (e.msgidPlural !== undefined) {
349
384
  lines.push(`msgid "${escapeC(e.msgid)}"`, `msgid_plural "${escapeC(e.msgidPlural)}"`);
350
- (e.msgstrPlural ?? []).forEach((s, i) => lines.push(`msgstr[${i}] "${escapeC(s)}"`));
385
+ const oneKey = pluralOneKey(e);
386
+ const otherKey = pluralOtherKey(e);
387
+ seen.add(oneKey);
388
+ seen.add(otherKey);
389
+ const one = typeof obj[oneKey] === "string" ? obj[oneKey] : e.msgid;
390
+ const other = typeof obj[otherKey] === "string" ? obj[otherKey] : e.msgidPlural;
391
+ const n = pluralCount(targetTag);
392
+ // Not full grammatical inflection (same simplification as addPluralForms for
393
+ // i18next): index 0 gets the singular-sourced form, every other index the
394
+ // plural-sourced one. A language with only one category (ja, zh, ...) gets
395
+ // just the plural form, since it has no separate singular count.
396
+ for (let i = 0; i < n; i++)
397
+ lines.push(`msgstr[${i}] "${escapeC(n === 1 ? other : i === 0 ? one : other)}"`);
351
398
  }
352
399
  else {
353
400
  const key = poKey(e);
@@ -390,47 +437,104 @@ const isObj = (v) => typeof v === "object" && v !== null && !Array.isArray(v);
390
437
  const newObj = () => Object.create(null);
391
438
  /** A string needs translating when the target has no non-empty string at the same path. */
392
439
  const needsTranslation = (existing, overwrite) => overwrite || typeof existing !== "string" || existing === "";
393
- export function pendingStrings(source, existing, overwrite, out = []) {
394
- if (typeof source === "string") {
395
- if (needsTranslation(existing, overwrite))
396
- out.push(source);
397
- }
398
- else if (Array.isArray(source)) {
399
- source.forEach((v, i) => pendingStrings(v, Array.isArray(existing) ? existing[i] : undefined, overwrite, out));
400
- }
401
- else if (isObj(source)) {
402
- for (const k of Object.keys(source))
403
- pendingStrings(source[k], isObj(existing) ? existing[k] : undefined, overwrite, out);
404
- }
440
+ /** Flattens a locale tree into key-path -> string-value pairs (for `locale-ai check`). */
441
+ export function flattenStrings(data, prefix = "", out = []) {
442
+ if (typeof data === "string")
443
+ out.push({ path: prefix || "(root)", value: data });
444
+ else if (Array.isArray(data))
445
+ data.forEach((v, i) => flattenStrings(v, `${prefix}[${i}]`, out));
446
+ else if (isObj(data))
447
+ for (const k of Object.keys(data))
448
+ flattenStrings(data[k], prefix ? `${prefix}.${k}` : k, out);
449
+ return out;
450
+ }
451
+ /**
452
+ * True when `path` is, or is nested under, one of `doNotTranslateKeys` — the same
453
+ * dot/`[i]` notation flattenStrings produces (e.g. "brand.name", "config.statusEnum").
454
+ * Protecting a path that points at an object or array protects every string nested
455
+ * under it too, which is what makes an enum array ("statusEnum": ["ACTIVE", ...])
456
+ * protectable without enumerating each value as a doNotTranslate term.
457
+ */
458
+ export function keyPathProtect(doNotTranslateKeys) {
459
+ const keys = (doNotTranslateKeys ?? []).filter(Boolean);
460
+ if (!keys.length)
461
+ return undefined;
462
+ return (path) => keys.some((k) => path === k || path.startsWith(`${k}.`) || path.startsWith(`${k}[`));
463
+ }
464
+ export function pendingStrings(source, existing, overwrite, isProtected) {
465
+ const out = [];
466
+ const visit = (s, e, path) => {
467
+ if (typeof s === "string") {
468
+ if (!isProtected?.(path) && needsTranslation(e, overwrite))
469
+ out.push(s);
470
+ }
471
+ else if (Array.isArray(s)) {
472
+ s.forEach((v, i) => visit(v, Array.isArray(e) ? e[i] : undefined, `${path}[${i}]`));
473
+ }
474
+ else if (isObj(s)) {
475
+ for (const k of Object.keys(s))
476
+ visit(s[k], isObj(e) ? e[k] : undefined, path ? `${path}.${k}` : k);
477
+ }
478
+ };
479
+ visit(source, existing, "");
405
480
  return out;
406
481
  }
407
482
  /**
408
483
  * Builds the target document: source structure and key order, existing
409
484
  * translations kept (unless overwrite), new strings from `translate`, and keys
410
- * that only exist in the target preserved at the end.
485
+ * that only exist in the target preserved at the end. A string at a protected
486
+ * path (see keyPathProtect) is always copied from the source verbatim.
411
487
  */
412
- export function buildTarget(source, existing, translate, overwrite) {
413
- if (typeof source === "string") {
414
- return needsTranslation(existing, overwrite) ? translate(source) : existing;
415
- }
416
- if (Array.isArray(source)) {
417
- return source.map((v, i) => buildTarget(v, Array.isArray(existing) ? existing[i] : undefined, translate, overwrite));
418
- }
419
- if (isObj(source)) {
420
- const out = newObj();
421
- for (const k of Object.keys(source)) {
422
- out[k] = buildTarget(source[k], isObj(existing) ? existing[k] : undefined, translate, overwrite);
488
+ export function buildTarget(source, existing, translate, overwrite, isProtected) {
489
+ const visit = (s, e, path) => {
490
+ if (typeof s === "string") {
491
+ if (isProtected?.(path))
492
+ return s;
493
+ return needsTranslation(e, overwrite) ? translate(s) : e;
423
494
  }
424
- if (isObj(existing))
425
- for (const k of Object.keys(existing))
426
- if (!(k in out))
427
- out[k] = existing[k];
428
- return out;
429
- }
430
- return source;
495
+ if (Array.isArray(s))
496
+ return s.map((v, i) => visit(v, Array.isArray(e) ? e[i] : undefined, `${path}[${i}]`));
497
+ if (isObj(s)) {
498
+ const out = newObj();
499
+ for (const k of Object.keys(s))
500
+ out[k] = visit(s[k], isObj(e) ? e[k] : undefined, path ? `${path}.${k}` : k);
501
+ if (isObj(e))
502
+ for (const k of Object.keys(e))
503
+ if (!(k in out))
504
+ out[k] = e[k];
505
+ return out;
506
+ }
507
+ return s;
508
+ };
509
+ return visit(source, existing, "");
431
510
  }
432
511
  const CATEGORIES = ["zero", "one", "two", "few", "many", "other"];
433
512
  const PLURAL_KEY = /^(.+?)(_ordinal)?_(zero|one|two|few|many|other)$/;
513
+ /**
514
+ * Source strings at a cardinal i18next "_one" plural key ("itemCount_one": "{{count}}
515
+ * item"). These are grammatically singular in English, but the model sometimes infers
516
+ * the TARGET's plural form anyway because {{count}} reads as an opaque marker rather
517
+ * than a concrete number — the engine uses this set to retranslate with a real "1" in
518
+ * place of {{count}} and recover the correct grammatical number (see engine.ts).
519
+ */
520
+ export function pluralOneStrings(data, out = new Set()) {
521
+ const visit = (v) => {
522
+ if (Array.isArray(v)) {
523
+ v.forEach(visit);
524
+ return;
525
+ }
526
+ if (!isObj(v))
527
+ return;
528
+ for (const k of Object.keys(v)) {
529
+ const m = PLURAL_KEY.exec(k);
530
+ if (m && !m[2] && m[3] === "one" && typeof v[k] === "string")
531
+ out.add(v[k]);
532
+ visit(v[k]);
533
+ }
534
+ };
535
+ visit(data);
536
+ return out;
537
+ }
434
538
  /**
435
539
  * i18next stores plurals as `key_one` / `key_other`. Languages such as Arabic,
436
540
  * Russian or Polish need more forms (`_few`, `_many` ...) and i18next does not fall
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "locale-ai",
3
- "version": "0.6.3",
3
+ "version": "0.6.5",
4
4
  "description": "Translate your i18n locale files locally on CPU. Free, offline, no API keys.",
5
5
  "author": "Taha KHABOUSS",
6
6
  "repository": {