@remnic/core 9.45.2 → 9.45.4

package/dist/{chunk-VNA7W7QF.js → chunk-SVH57F5Z.js}

@@ -87,6 +87,14 @@ function isIntlWordBoundary(source, index) {
   if (index <= 0 || index >= source.length) return true;
   return UNICODE_WORD_SEGMENTER?.segment(source).containing(index)?.index === index;
 }
+function isStandaloneJapaneseLeadingParticle(particle, source, boundaryIndex) {
+  if (!JAPANESE_WORD_SEGMENTER || boundaryIndex <= 0 || !JAPANESE_PARTICLE_RE.test(particle)) {
+    return false;
+  }
+  const particleStart = boundaryIndex - particle.length;
+  const segment = JAPANESE_WORD_SEGMENTER.segment(source).containing(particleStart);
+  return segment?.index === particleStart && segment.index + segment.segment.length === boundaryIndex;
+}
 function isKoreanUnspacedQuestionBoundary(suffix) {
   return KOREAN_UNSPACED_QUESTION_PARTICLES.some(
     (particle) => suffix.startsWith(particle) && KOREAN_INTERROGATIVE_PREFIXES.some((prefix) => suffix.startsWith(prefix, particle.length))
@@ -102,8 +110,8 @@ function isKoreanParticleBoundary(suffix) {
   const nextCharacter = firstUnicodeCharacter(suffix.slice(particleRun.length));
   return KOREAN_PARTICLES.some((particle) => particle === particleRun) && (nextCharacter.length === 0 || !UNICODE_WORD_OR_NUMBER_RE.test(nextCharacter));
 }
-function isUnicodePhraseBoundary(character, suffix = "", source = "", boundaryIndex = -1, strictUnicode = false, allowLeadingParticle = false) {
-  return character.length === 0 || !UNICODE_WORD_OR_NUMBER_RE.test(character) || allowLeadingParticle && isJapaneseParticleBoundary(character) || isJapaneseParticleBoundary(suffix) || !strictUnicode && boundaryIndex >= 0 && !JAPANESE_KANA_RE.test(suffix) && isIntlWordBoundary(source, boundaryIndex) || allowLeadingParticle && isKoreanParticleBoundary(character) || isKoreanParticleBoundary(suffix);
+function isUnicodePhraseBoundary(character, suffix = "", source = "", boundaryIndex = -1, strictUnicode = false) {
+  return character.length === 0 || !UNICODE_WORD_OR_NUMBER_RE.test(character) || isJapaneseParticleBoundary(suffix) || !strictUnicode && boundaryIndex >= 0 && !JAPANESE_KANA_RE.test(suffix) && isIntlWordBoundary(source, boundaryIndex) || isKoreanParticleBoundary(suffix);
 }
 function containsPhrase(haystack, needle) {
   if (!needle) return false;
@@ -114,7 +122,7 @@ function containsPhrase(haystack, needle) {
   while (offset >= 0) {
     const before = lastUnicodeCharacter(searchHaystack.slice(0, offset));
     const after = searchHaystack.slice(offset + searchNeedle.length);
-    if (isUnicodePhraseBoundary(before, "", searchHaystack, offset, caseInsensitive, true) && isUnicodePhraseBoundary(
+    if ((isUnicodePhraseBoundary(before, "", searchHaystack, offset, caseInsensitive) || isStandaloneJapaneseLeadingParticle(before, searchHaystack, offset)) && isUnicodePhraseBoundary(
       firstUnicodeCharacter(after),
       after,
       searchHaystack,
@@ -131,4 +139,4 @@ function containsPhrase(haystack, needle) {
 export {
   containsPhrase
 };
-//# sourceMappingURL=chunk-VNA7W7QF.js.map
+//# sourceMappingURL=chunk-SVH57F5Z.js.map

package/dist/chunk-SVH57F5Z.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/entity-retrieval-boundaries.ts"],"sourcesContent":["const UNICODE_WORD_OR_NUMBER_RE = /[\\p{L}\\p{M}\\p{N}]/u;\n\nfunction firstUnicodeCharacter(value: string): string {\n  const codePoint = value.codePointAt(0);\n  return codePoint === undefined ? \"\" : String.fromCodePoint(codePoint);\n}\n\nfunction lastUnicodeCharacter(value: string): string {\n  const codePoint = [...value].at(-1);\n  return codePoint ?? \"\";\n}\n\nconst JAPANESE_PARTICLE_RE = /^[のはがをにへともやかでだ]$/u;\nconst JAPANESE_MULTI_CHARACTER_PARTICLES = [\n  \"について\",\n  \"にとって\",\n  \"によって\",\n  \"に関して\",\n  \"に対して\",\n  \"とは\",\n  \"では\",\n  \"には\",\n  \"にも\",\n  \"って\",\n] as const;\nconst KOREAN_PARTICLES = [\n  \"에서\", \"에게\", \"한테\", \"으로\", \"까지\", \"부터\", \"보다\", \"처럼\", \"같이\",\n  \"이나\", \"라도\", \"밖에\", \"마다\", \"조차\", \"이랑\", \"하고\", \"은\", \"는\",\n  \"이\", \"가\", \"을\", \"를\", \"의\", \"께\", \"와\", \"과\", \"도\", \"만\", \"든\",\n  \"에\", \"로\", \"나\", \"랑\",\n] as const;\nconst KOREAN_UNSPACED_QUESTION_PARTICLES = [\"은\", \"는\", \"이\", \"가\"] as const;\nconst KOREAN_INTERROGATIVE_PREFIXES = [\n  \"어디\", \"무엇\", \"뭐\", \"왜\", \"어떻게\", \"언제\", \"누구\", \"어느\", \"몇\", \"얼마\",\n] as const;\n\nconst HANGUL_RUN_RE = /^[\\p{Script=Hangul}]+/u;\n\nconst UNICODE_WORD_SEGMENTER =\n  typeof Intl.Segmenter === \"function\" ? new Intl.Segmenter(undefined, { granularity: \"word\" }) : null;\nconst KOREAN_WORD_SEGMENTER =\n  typeof Intl.Segmenter === \"function\" ? new Intl.Segmenter(\"ko\", { granularity: \"word\" }) : null;\n\nconst JAPANESE_WORD_SEGMENTER =\n  typeof Intl.Segmenter === \"function\" ? new Intl.Segmenter(\"ja\", { granularity: \"word\" }) : null;\nconst JAPANESE_KANA_RE = /[\\p{Script=Hiragana}\\p{Script=Katakana}]/u;\n\nfunction isJapaneseParticleBoundary(suffix: string): boolean {\n  const firstSegment = JAPANESE_WORD_SEGMENTER?.segment(suffix).containing(0)?.segment;\n  if (firstSegment !== undefined) {\n    if (JAPANESE_MULTI_CHARACTER_PARTICLES.some((particle) => particle === firstSegment)) return true;\n    return JAPANESE_PARTICLE_RE.test(firstSegment);\n  }\n  const firstCharacter = firstUnicodeCharacter(suffix);\n  return (\n    JAPANESE_MULTI_CHARACTER_PARTICLES.some((particle) => suffix.startsWith(particle)) ||\n    (JAPANESE_PARTICLE_RE.test(firstCharacter) &&\n      !UNICODE_WORD_OR_NUMBER_RE.test(firstUnicodeCharacter(suffix.slice(firstCharacter.length))))\n  );\n}\n\nfunction isIntlWordBoundary(source: string, index: number): boolean {\n  if (index <= 0 || index >= source.length) return true;\n  return UNICODE_WORD_SEGMENTER?.segment(source).containing(index)?.index === index;\n}\n\nfunction isStandaloneJapaneseLeadingParticle(\n  particle: string,\n  source: string,\n  boundaryIndex: number,\n): boolean {\n  if (!JAPANESE_WORD_SEGMENTER || boundaryIndex <= 0 || !JAPANESE_PARTICLE_RE.test(particle)) {\n    return false;\n  }\n  const particleStart = boundaryIndex - particle.length;\n  const segment = JAPANESE_WORD_SEGMENTER.segment(source).containing(particleStart);\n  return segment?.index === particleStart\n    && segment.index + segment.segment.length === boundaryIndex;\n}\nfunction isKoreanUnspacedQuestionBoundary(suffix: string): boolean {\n  return KOREAN_UNSPACED_QUESTION_PARTICLES.some((particle) =>\n    suffix.startsWith(particle) &&\n    KOREAN_INTERROGATIVE_PREFIXES.some((prefix) => suffix.startsWith(prefix, particle.length)),\n  );\n}\n\nfunction isKoreanParticleBoundary(suffix: string): boolean {\n  if (isKoreanUnspacedQuestionBoundary(suffix)) return true;\n  const firstSegment = KOREAN_WORD_SEGMENTER?.segment(suffix).containing(0)?.segment;\n  if (firstSegment !== undefined) {\n    return KOREAN_PARTICLES.some((particle) => particle === firstSegment);\n  }\n  const particleRun = suffix.match(HANGUL_RUN_RE)?.[0] ?? \"\";\n  const nextCharacter = firstUnicodeCharacter(suffix.slice(particleRun.length));\n  return (\n    KOREAN_PARTICLES.some((particle) => particle === particleRun) &&\n    (nextCharacter.length === 0 || !UNICODE_WORD_OR_NUMBER_RE.test(nextCharacter))\n  );\n}\n\nfunction isUnicodePhraseBoundary(\n  character: string,\n  suffix: string = \"\",\n  source: string = \"\",\n  boundaryIndex: number = -1,\n  strictUnicode = false,\n): boolean {\n  return (\n    character.length === 0\n    || !UNICODE_WORD_OR_NUMBER_RE.test(character)\n    || isJapaneseParticleBoundary(suffix)\n    || (!strictUnicode &&\n      boundaryIndex >= 0 &&\n      !JAPANESE_KANA_RE.test(suffix) &&\n      isIntlWordBoundary(source, boundaryIndex))\n    || isKoreanParticleBoundary(suffix)\n  );\n}\n\nexport function containsPhrase(haystack: string, needle: string): boolean {\n  if (!needle) return false;\n  const caseInsensitive = /^[a-z0-9 ]+$/i.test(needle);\n  const searchHaystack = caseInsensitive ? haystack.toLowerCase() : haystack;\n  const searchNeedle = caseInsensitive ? needle.toLowerCase() : needle;\n  let offset = searchHaystack.indexOf(searchNeedle);\n  while (offset >= 0) {\n    const before = lastUnicodeCharacter(searchHaystack.slice(0, offset));\n    const after = searchHaystack.slice(offset + searchNeedle.length);\n    if (\n      (\n        isUnicodePhraseBoundary(before, \"\", searchHaystack, offset, caseInsensitive)\n        || isStandaloneJapaneseLeadingParticle(before, searchHaystack, offset)\n      )\n      && isUnicodePhraseBoundary(\n        firstUnicodeCharacter(after),\n        after,\n        searchHaystack,\n        offset + searchNeedle.length,\n        caseInsensitive,\n      )\n    ) {\n      return true;\n    }\n    offset = searchHaystack.indexOf(searchNeedle, offset + searchNeedle.length);\n  }\n  return false;\n}\n"],"mappings":";AAAA,IAAM,4BAA4B;AAElC,SAAS,sBAAsB,OAAuB;AACpD,QAAM,YAAY,MAAM,YAAY,CAAC;AACrC,SAAO,cAAc,SAAY,KAAK,OAAO,cAAc,SAAS;AACtE;AAEA,SAAS,qBAAqB,OAAuB;AACnD,QAAM,YAAY,CAAC,GAAG,KAAK,EAAE,GAAG,EAAE;AAClC,SAAO,aAAa;AACtB;AAEA,IAAM,uBAAuB;AAC7B,IAAM,qCAAqC;AAAA,EACzC;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AACF;AACA,IAAM,mBAAmB;AAAA,EACvB;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAChD;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAK;AAAA,EAC/C;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AAAA,EAClD;AAAA,EAAK;AAAA,EAAK;AAAA,EAAK;AACjB;AACA,IAAM,qCAAqC,CAAC,UAAK,UAAK,UAAK,QAAG;AAC9D,IAAM,gCAAgC;AAAA,EACpC;AAAA,EAAM;AAAA,EAAM;AAAA,EAAK;AAAA,EAAK;AAAA,EAAO;AAAA,EAAM;AAAA,EAAM;AAAA,EAAM;AAAA,EAAK;AACtD;AAEA,IAAM,gBAAgB;AAEtB,IAAM,yBACJ,OAAO,KAAK,cAAc,aAAa,IAAI,KAAK,UAAU,QAAW,EAAE,aAAa,OAAO,CAAC,IAAI;AAClG,IAAM,wBACJ,OAAO,KAAK,cAAc,aAAa,IAAI,KAAK,UAAU,MAAM,EAAE,aAAa,OAAO,CAAC,IAAI;AAE7F,IAAM,0BACJ,OAAO,KAAK,cAAc,aAAa,IAAI,KAAK,UAAU,MAAM,EAAE,aAAa,OAAO,CAAC,IAAI;AAC7F,IAAM,mBAAmB;AAEzB,SAAS,2BAA2B,QAAyB;AAC3D,QAAM,eAAe,yBAAyB,QAAQ,MAAM,EAAE,WAAW,CAAC,GAAG;AAC7E,MAAI,iBAAiB,QAAW;AAC9B,QAAI,mCAAmC,KAAK,CAAC,aAAa,aAAa,YAAY,EAAG,QAAO;AAC7F,WAAO,qBAAqB,KAAK,YAAY;AAAA,EAC/C;AACA,QAAM,iBAAiB,sBAAsB,MAAM;AACnD,SACE,mCAAmC,KAAK,CAAC,aAAa,OAAO,WAAW,QAAQ,CAAC,KAChF,qBAAqB,KAAK,cAAc,KACvC,CAAC,0BAA0B,KAAK,sBAAsB,OAAO,MAAM,eAAe,MAAM,CAAC,CAAC;AAEhG;AAEA,SAAS,mBAAmB,QAAgB,OAAwB;AAClE,MAAI,SAAS,KAAK,SAAS,OAAO,OAAQ,QAAO;AACjD,SAAO,wBAAwB,QAAQ,MAAM,EAAE,WAAW,KAAK,GAAG,UAAU;AAC9E;AAEA,SAAS,oCACP,UACA,QACA,eACS;AACT,MAAI,CAAC,2BAA2B,iBAAiB,KAAK,CAAC,qBAAqB,KAAK,QAAQ,GAAG;AAC1F,WAAO;AAAA,EACT;AACA,QAAM,gBAAgB,gBAAgB,SAAS;AAC/C,QAAM,UAAU,wBAAwB,QAAQ,MAAM,EAAE,WAAW,aAAa;AAChF,SAAO,SAAS,UAAU,iBACrB,QAAQ,QAAQ,QAAQ,QAAQ,WAAW;AAClD;AACA,SAAS,iCAAiC,QAAyB;AACjE,SAAO,mCAAmC;AAAA,IAAK,CAAC,aAC9C,OAAO,WAAW,QAAQ,KAC1B,8BAA8B,KAAK,CAAC,WAAW,OAAO,WAAW,QAAQ,SAAS,MAAM,CAAC;AAAA,EAC3F;AACF;AAEA,SAAS,yBAAyB,QAAyB;AACzD,MAAI,iCAAiC,MAAM,EAAG,QAAO;AACrD,QAAM,eAAe,uBAAuB,QAAQ,MAAM,EAAE,WAAW,CAAC,GAAG;AAC3E,MAAI,iBAAiB,QAAW;AAC9B,WAAO,iBAAiB,KAAK,CAAC,aAAa,aAAa,YAAY;AAAA,EACtE;AACA,QAAM,cAAc,OAAO,MAAM,aAAa,IAAI,CAAC,KAAK;AACxD,QAAM,gBAAgB,sBAAsB,OAAO,MAAM,YAAY,MAAM,CAAC;AAC5E,SACE,iBAAiB,KAAK,CAAC,aAAa,aAAa,WAAW,MAC3D,cAAc,WAAW,KAAK,CAAC,0BAA0B,KAAK,aAAa;AAEhF;AAEA,SAAS,wBACP,WACA,SAAiB,IACjB,SAAiB,IACjB,gBAAwB,IACxB,gBAAgB,OACP;AACT,SACE,UAAU,WAAW,KAClB,CAAC,0BAA0B,KAAK,SAAS,KACzC,2BAA2B,MAAM,KAChC,CAAC,iBACH,iBAAiB,KACjB,CAAC,iBAAiB,KAAK,MAAM,KAC7B,mBAAmB,QAAQ,aAAa,KACvC,yBAAyB,MAAM;AAEtC;AAEO,SAAS,eAAe,UAAkB,QAAyB;AACxE,MAAI,CAAC,OAAQ,QAAO;AACpB,QAAM,kBAAkB,gBAAgB,KAAK,MAAM;AACnD,QAAM,iBAAiB,kBAAkB,SAAS,YAAY,IAAI;AAClE,QAAM,eAAe,kBAAkB,OAAO,YAAY,IAAI;AAC9D,MAAI,SAAS,eAAe,QAAQ,YAAY;AAChD,SAAO,UAAU,GAAG;AAClB,UAAM,SAAS,qBAAqB,eAAe,MAAM,GAAG,MAAM,CAAC;AACnE,UAAM,QAAQ,eAAe,MAAM,SAAS,aAAa,MAAM;AAC/D,SAEI,wBAAwB,QAAQ,IAAI,gBAAgB,QAAQ,eAAe,KACxE,oCAAoC,QAAQ,gBAAgB,MAAM,MAEpE;AAAA,MACD,sBAAsB,KAAK;AAAA,MAC3B;AAAA,MACA;AAAA,MACA,SAAS,aAAa;AAAA,MACtB;AAAA,IACF,GACA;AACA,aAAO;AAAA,IACT;AACA,aAAS,eAAe,QAAQ,cAAc,SAAS,aAAa,MAAM;AAAA,EAC5E;AACA,SAAO;AACT;","names":[]}

package/dist/{chunk-WIPEXG6R.js → chunk-SXIMQLGV.js}

@@ -19,7 +19,7 @@ import {
 import {
   decryptMemoryDirToPlaintext,
   migrateMemoryDirToEncrypted
-} from "./chunk-5TFP4CUL.js";
+} from "./chunk-XQ3HXAQH.js";
 import {
   generateSalt
 } from "./chunk-A6XUJE5D.js";
@@ -499,4 +499,4 @@ export {
   renderStatusReport,
   createPassphraseReader
 };
-//# sourceMappingURL=chunk-WIPEXG6R.js.map
+//# sourceMappingURL=chunk-SXIMQLGV.js.map