@delali/narsil 0.1.0

package/dist/languages/romanian.mjs

@@ -0,0 +1,634 @@
+// src/languages/romanian.ts
+var VOWELS = "a\u0103\xE2ei\xEEou";
+var RE_UPPER_I = /I/g;
+var RE_UPPER_U = /U/g;
+var STEP0_SUFFIXES = [
+  "iilor",
+  "ilor",
+  "elor",
+  "ului",
+  "iile",
+  "atei",
+  "a\u021Bie",
+  "a\u021Bia",
+  "ele",
+  "ile",
+  "iei",
+  "ul",
+  "ea",
+  "ii",
+  "ua",
+  "iua",
+  "aua",
+  "a"
+];
+var STEP1_SUFFIX_GROUPS = [
+  [["abilitate", "abilitati", "abilit\u0103\u021Bi", "abilita\u021Bi"].sort((a, b) => b.length - a.length), 1, "abil"],
+  [["ibilitate", "ibilitati", "ibilit\u0103\u021Bi", "ibilita\u021Bi"].sort((a, b) => b.length - a.length), 2, "ibil"],
+  [["ivitate", "ivitati", "ivit\u0103\u021Bi", "ivita\u021Bi"].sort((a, b) => b.length - a.length), 3, "iv"],
+  [
+    [
+      "ical\u0103",
+      "icale",
+      "icali",
+      "ical",
+      "ic\u0103",
+      "ice",
+      "ici",
+      "ic",
+      "icitate",
+      "icitati",
+      "icita\u021Bi",
+      "icit\u0103\u021Bi",
+      "iciva",
+      "icive",
+      "icivi",
+      "iciv",
+      "iciv\u0103",
+      "icator",
+      "icatori",
+      "icatoare",
+      "ic\u0103toare",
+      "ic\u0103tori"
+    ].sort((a, b) => b.length - a.length),
+    4,
+    "ic"
+  ],
+  [
+    ["ativ\u0103", "ative", "ativi", "ativ", "ativa", "a\u021Bie", "a\u021Biune", "ator", "atori", "atoare", "\u0103toare", "\u0103tori"].sort(
+      (a, b) => b.length - a.length
+    ),
+    5,
+    "at"
+  ],
+  [
+    ["itiv\u0103", "itive", "itivi", "itiv", "itiva", "i\u021Bie", "i\u021Biune", "itor", "itori", "itoare"].sort(
+      (a, b) => b.length - a.length
+    ),
+    6,
+    "it"
+  ]
+];
+var STEP1_STANDARD_SUFFIXES = [
+  [["abil\u0103", "abile", "abili", "abil", "abila"].sort((a, b) => b.length - a.length), null],
+  [["ibil\u0103", "ibile", "ibili", "ibil", "ibila"].sort((a, b) => b.length - a.length), null],
+  [["oas\u0103", "oase", "oasa", "os", "o\u0219i"].sort((a, b) => b.length - a.length), null],
+  [
+    [
+      "at\u0103",
+      "ate",
+      "ati",
+      "at",
+      "ata",
+      "it\u0103",
+      "ite",
+      "iti",
+      "it",
+      "ita",
+      "ant\u0103",
+      "ante",
+      "anti",
+      "ant",
+      "anta",
+      "ut\u0103",
+      "ute",
+      "uti",
+      "ut",
+      "uta",
+      "iv\u0103",
+      "ive",
+      "ivi",
+      "iv",
+      "iva",
+      "iune",
+      "iuni",
+      "itate",
+      "itati",
+      "ita\u021Bi",
+      "it\u0103\u021Bi"
+    ].sort((a, b) => b.length - a.length),
+    null
+  ],
+  [["ist\u0103", "iste", "isti", "ist", "ista", "ism", "isme", "i\u0219ti"].sort((a, b) => b.length - a.length), "ist"]
+];
+var STEP2_SUFFIXES = [
+  "seser\u0103\u021Bi",
+  "aser\u0103\u021Bi",
+  "iser\u0103\u021Bi",
+  "user\u0103\u021Bi",
+  "\xE2ser\u0103\u021Bi",
+  "ser\u0103\u021Bi",
+  "seser\u0103m",
+  "aser\u0103m",
+  "iser\u0103m",
+  "user\u0103m",
+  "\xE2ser\u0103m",
+  "ser\u0103m",
+  "sese\u0219ti",
+  "ase\u0219ti",
+  "ise\u0219ti",
+  "use\u0219ti",
+  "\xE2se\u0219ti",
+  "se\u0219ti",
+  "ar\u0103\u021Bi",
+  "ir\u0103\u021Bi",
+  "ur\u0103\u021Bi",
+  "\xE2r\u0103\u021Bi",
+  "ar\u0103m",
+  "ir\u0103m",
+  "ur\u0103m",
+  "\xE2r\u0103m",
+  "ea\u021Bi",
+  "ia\u021Bi",
+  "a\u021Bi",
+  "eam",
+  "iam",
+  "am",
+  "eaz\u0103",
+  "e\u021Bi",
+  "i\u021Bi",
+  "\xE2\u021Bi",
+  "ar\u0103",
+  "ir\u0103",
+  "ur\u0103",
+  "\xE2r\u0103",
+  "ser\u0103",
+  "aser\u0103",
+  "iser\u0103",
+  "user\u0103",
+  "\xE2ser\u0103",
+  "seser\u0103",
+  "easc",
+  "esc",
+  "\u0103sc",
+  "e\u0219te",
+  "\u0103\u0219te",
+  "e\u0219ti",
+  "\u0103\u0219ti",
+  "are",
+  "ere",
+  "ire",
+  "\xE2re",
+  "ase",
+  "ise",
+  "use",
+  "\xE2se",
+  "sese",
+  "se",
+  "ind",
+  "\xE2nd",
+  "indu",
+  "\xE2ndu",
+  "eze",
+  "ezi",
+  "ez",
+  "eai",
+  "iai",
+  "ai",
+  "\xE2i",
+  "ui",
+  "sei",
+  "asem",
+  "isem",
+  "usem",
+  "\xE2sem",
+  "sesem",
+  "em",
+  "im",
+  "\xE2m",
+  "\u0103m",
+  "eau",
+  "iau",
+  "au",
+  "a\u0219i",
+  "i\u0219i",
+  "u\u0219i",
+  "\xE2\u0219i"
+].sort((a, b) => b.length - a.length);
+var STEP2_VERB_GROUP_SESE = /* @__PURE__ */ new Set([
+  "sese",
+  "sesem",
+  "sese\u0219ti",
+  "seser\u0103m",
+  "seser\u0103\u021Bi",
+  "se",
+  "sei",
+  "se\u0219ti",
+  "ser\u0103",
+  "seser\u0103",
+  "ser\u0103m",
+  "ser\u0103\u021Bi"
+]);
+var STEP3_SUFFIXES = ["ie", "\u0103", "e", "i", "a"];
+function isVowel(ch) {
+  return VOWELS.includes(ch);
+}
+function isLowerVowel(ch) {
+  return VOWELS.includes(ch) || ch === "I" || ch === "U";
+}
+function markVowelAsConsonant(word) {
+  const chars = word.split("");
+  for (let i = 1; i < chars.length - 1; i++) {
+    if (isVowel(chars[i - 1]) && isVowel(chars[i + 1])) {
+      if (chars[i] === "u") chars[i] = "U";
+      if (chars[i] === "i") chars[i] = "I";
+    }
+  }
+  return chars.join("");
+}
+function computeRegions(word) {
+  let rv = word.length;
+  let r1 = word.length;
+  let r2 = word.length;
+  if (word.length >= 2) {
+    if (isLowerVowel(word[0])) {
+      if (isLowerVowel(word[1])) {
+        let i2 = 2;
+        while (i2 < word.length && isLowerVowel(word[i2])) i2++;
+        if (i2 < word.length) {
+          i2++;
+          rv = i2;
+        }
+      } else {
+        let i2 = 2;
+        while (i2 < word.length && !isLowerVowel(word[i2])) i2++;
+        if (i2 < word.length) {
+          rv = i2 + 1;
+        }
+      }
+    } else if (isLowerVowel(word[1])) {
+      let i2 = 2;
+      while (i2 < word.length && !isLowerVowel(word[i2])) i2++;
+      if (i2 < word.length) {
+        rv = i2 + 1;
+      }
+    } else {
+      let i2 = 2;
+      while (i2 < word.length) {
+        if (isLowerVowel(word[i2])) {
+          rv = i2 + 1;
+          break;
+        }
+        i2++;
+      }
+    }
+  }
+  let i = 1;
+  while (i < word.length) {
+    if (!isLowerVowel(word[i]) && isLowerVowel(word[i - 1])) {
+      r1 = i + 1;
+      break;
+    }
+    i++;
+  }
+  i = r1 + 1;
+  while (i < word.length) {
+    if (!isLowerVowel(word[i]) && isLowerVowel(word[i - 1])) {
+      r2 = i + 1;
+      break;
+    }
+    i++;
+  }
+  return { rv, r1, r2 };
+}
+function inRegion(word, regionStart, suffixLen) {
+  return word.length - suffixLen >= regionStart;
+}
+function trySuffix(word, suffixes) {
+  for (const s of suffixes) {
+    if (word.endsWith(s)) return s;
+  }
+  return null;
+}
+function step0(word, r1) {
+  const matched = trySuffix(word, STEP0_SUFFIXES);
+  if (matched && inRegion(word, r1, matched.length)) {
+    switch (matched) {
+      case "ul":
+      case "ului":
+        return word.slice(0, -matched.length);
+      case "aua":
+        return `${word.slice(0, -matched.length)}a`;
+      case "ea":
+      case "ele":
+      case "elor":
+        return `${word.slice(0, -matched.length)}e`;
+      case "iua":
+      case "iei":
+      case "ile":
+      case "iile":
+      case "ilor":
+      case "iilor":
+      case "ii":
+        return `${word.slice(0, -matched.length)}i`;
+      case "a\u021Bia":
+      case "a\u021Bie":
+      case "atei":
+        return `${word.slice(0, -matched.length)}a\u021Bi`;
+      case "ua":
+        return word.slice(0, -matched.length);
+      case "a":
+        return word.slice(0, -1);
+      default:
+        return word.slice(0, -matched.length);
+    }
+  }
+  return word;
+}
+function step1(word, _r1, r2) {
+  for (const [suffixes, _id, replacement] of STEP1_SUFFIX_GROUPS) {
+    const matched = trySuffix(word, suffixes);
+    if (matched && inRegion(word, r2, matched.length)) {
+      return { word: `${word.slice(0, -matched.length)}${replacement}`, changed: true };
+    }
+  }
+  for (const [suffixes, replacement] of STEP1_STANDARD_SUFFIXES) {
+    const matched = trySuffix(word, suffixes);
+    if (matched && inRegion(word, r2, matched.length)) {
+      if (replacement === "ist") {
+        return { word: `${word.slice(0, -matched.length)}ist`, changed: true };
+      }
+      const stripped = word.slice(0, -matched.length);
+      if (matched === "iune" || matched === "iuni") {
+        if (stripped.endsWith("\u021B")) {
+          return { word: `${stripped.slice(0, -1)}t`, changed: true };
+        }
+      }
+      return { word: stripped, changed: true };
+    }
+  }
+  return { word, changed: false };
+}
+function step2(word, rv) {
+  const matched = trySuffix(word, STEP2_SUFFIXES);
+  if (matched && inRegion(word, rv, matched.length)) {
+    const stripped = word.slice(0, -matched.length);
+    if (STEP2_VERB_GROUP_SESE.has(matched)) {
+      if (!isVowel(stripped[stripped.length - 1]) && stripped[stripped.length - 1] !== "u") {
+        return { word, changed: false };
+      }
+    }
+    return { word: stripped, changed: true };
+  }
+  return { word, changed: false };
+}
+function step3(word, rv) {
+  const matched = trySuffix(word, STEP3_SUFFIXES);
+  if (matched && inRegion(word, rv, matched.length)) {
+    return word.slice(0, -matched.length);
+  }
+  return word;
+}
+function restoreMarked(word) {
+  return word.replace(RE_UPPER_I, "i").replace(RE_UPPER_U, "u");
+}
+function stem(word) {
+  if (word.length < 3) return word;
+  word = word.toLowerCase();
+  word = markVowelAsConsonant(word);
+  const { rv, r1, r2 } = computeRegions(word);
+  word = step0(word, r1);
+  const result1 = step1(word, r1, r2);
+  word = result1.word;
+  const step1Changed = result1.changed;
+  if (!step1Changed) {
+    const result2 = step2(word, rv);
+    word = result2.word;
+  }
+  word = step3(word, rv);
+  word = restoreMarked(word);
+  return word;
+}
+var stopWords = /* @__PURE__ */ new Set([
+  "acea",
+  "aceasta",
+  "aceast\u0103",
+  "aceea",
+  "acei",
+  "aceia",
+  "acel",
+  "acela",
+  "acele",
+  "acelea",
+  "acest",
+  "acesta",
+  "aceste",
+  "acestea",
+  "ace\u015Fti",
+  "ace\u015Ftia",
+  "acolo",
+  "acum",
+  "ai",
+  "aici",
+  "al",
+  "ale",
+  "altceva",
+  "altcineva",
+  "am",
+  "ar",
+  "are",
+  "asemenea",
+  "asta",
+  "astea",
+  "ast\u0103zi",
+  "asupra",
+  "au",
+  "avea",
+  "avem",
+  "ave\u0163i",
+  "azi",
+  "a\u015F",
+  "a\u015Fadar",
+  "a\u0163i",
+  "bine",
+  "ca",
+  "care",
+  "ce",
+  "cel",
+  "ceva",
+  "chiar",
+  "cine",
+  "cineva",
+  "contra",
+  "cu",
+  "cum",
+  "cumva",
+  "c\xE2nd",
+  "c\xE2t",
+  "c\xE2te",
+  "c\xE2\u0163i",
+  "c\u0103",
+  "c\u0103ci",
+  "c\u0103rei",
+  "c\u0103ror",
+  "c\u0103rui",
+  "c\u0103tre",
+  "da",
+  "dac\u0103",
+  "dar",
+  "datorit\u0103",
+  "dat\u0103",
+  "de",
+  "deci",
+  "deja",
+  "deoarece",
+  "departe",
+  "de\u015Fi",
+  "din",
+  "dintre",
+  "doi",
+  "dou\u0103",
+  "drept",
+  "dup\u0103",
+  "ea",
+  "ei",
+  "el",
+  "ele",
+  "eram",
+  "este",
+  "eu",
+  "e\u015Fti",
+  "face",
+  "fi",
+  "fie",
+  "fiecare",
+  "fii",
+  "fim",
+  "fiu",
+  "fi\u0163i",
+  "f\u0103r\u0103",
+  "iar",
+  "ieri",
+  "la",
+  "le",
+  "li",
+  "lor",
+  "lui",
+  "l\xE2ng\u0103",
+  "mai",
+  "mea",
+  "mei",
+  "mele",
+  "mereu",
+  "meu",
+  "mi",
+  "mie",
+  "mine",
+  "mult",
+  "mult\u0103",
+  "mul\u0163i",
+  "m\xE2ine",
+  "m\u0103",
+  "ne",
+  "nevoie",
+  "nici",
+  "nic\u0103ieri",
+  "nimeni",
+  "nimic",
+  "ni\u015Fte",
+  "noastre",
+  "noastr\u0103",
+  "noi",
+  "nostru",
+  "nou\u0103",
+  "no\u015Ftri",
+  "nu",
+  "ori",
+  "oricare",
+  "orice",
+  "oricine",
+  "oricum",
+  "oric\xE2nd",
+  "oric\xE2t",
+  "oriunde",
+  "pe",
+  "pentru",
+  "peste",
+  "poate",
+  "pot",
+  "prea",
+  "prima",
+  "primul",
+  "prin",
+  "pu\u0163in",
+  "p\xE2n\u0103",
+  "sa",
+  "sale",
+  "sau",
+  "se",
+  "spre",
+  "sub",
+  "sunt",
+  "suntem",
+  "sunte\u0163i",
+  "s\u0103",
+  "s\u0103i",
+  "s\u0103u",
+  "ta",
+  "tale",
+  "te",
+  "timp",
+  "tine",
+  "toate",
+  "toat\u0103",
+  "tot",
+  "totu\u015Fi",
+  "to\u0163i",
+  "trei",
+  "tu",
+  "t\u0103i",
+  "t\u0103u",
+  "un",
+  "una",
+  "unde",
+  "undeva",
+  "unei",
+  "unele",
+  "unii",
+  "unor",
+  "unu",
+  "unui",
+  "unul",
+  "vi",
+  "voastre",
+  "voastr\u0103",
+  "voi",
+  "vostru",
+  "vou\u0103",
+  "vo\u015Ftri",
+  "vreme",
+  "vreo",
+  "vreun",
+  "v\u0103",
+  "zece",
+  "zi",
+  "\xEEi",
+  "\xEEl",
+  "\xEEmi",
+  "\xEEmpotriva",
+  "\xEEn",
+  "\xEEnainte",
+  "\xEEnaintea",
+  "\xEEncotro",
+  "\xEEnc\xE2t",
+  "\xEEntre",
+  "\xEEntruc\xE2t",
+  "\xEE\u0163i",
+  "\u0103la",
+  "\u0103lea",
+  "\u0103sta",
+  "\u0103stea",
+  "\u015Fapte",
+  "\u015Fase",
+  "\u015Fi",
+  "\u0163i",
+  "\u0163ie"
+]);
+var romanian = {
+  name: "romanian",
+  stemmer: stem,
+  stopWords,
+  tokenizer: { splitPattern: /[^a-z0-9ăâîșț]+/gi }
+};
+
+export { romanian };

package/dist/languages/russian.d.ts

@@ -0,0 +1,5 @@
+import { L as LanguageModule } from '../language-ClUGLuoC.js';
+
+declare const russian: LanguageModule;
+
+export { russian };