@delali/narsil 0.1.0

package/dist/languages/turkish.mjs

@@ -0,0 +1,675 @@
+// src/languages/turkish.ts
+var VOWELS = "ae\u0131io\xF6u\xFC";
+var BACK_UNROUNDED_VOWELS = "a\u0131";
+var FRONT_UNROUNDED_VOWELS = "ei";
+var BACK_ROUNDED_VOWELS = "ou";
+var FRONT_ROUNDED_VOWELS = "\xF6\xFC";
+function isVowel(ch) {
+  return VOWELS.includes(ch);
+}
+function vowelCount(word) {
+  let count = 0;
+  for (const ch of word) {
+    if (isVowel(ch)) count++;
+  }
+  return count;
+}
+function lastVowel(word) {
+  for (let i = word.length - 1; i >= 0; i--) {
+    if (isVowel(word[i])) return word[i];
+  }
+  return null;
+}
+function checkVowelHarmony(stem2, suffix) {
+  const lv = lastVowel(stem2);
+  if (lv === null) return false;
+  const firstSuffixVowel = suffix.split("").find(isVowel);
+  if (!firstSuffixVowel) return true;
+  if (BACK_UNROUNDED_VOWELS.includes(lv)) {
+    return BACK_UNROUNDED_VOWELS.includes(firstSuffixVowel);
+  }
+  if (FRONT_UNROUNDED_VOWELS.includes(lv)) {
+    return FRONT_UNROUNDED_VOWELS.includes(firstSuffixVowel);
+  }
+  if (BACK_ROUNDED_VOWELS.includes(lv)) {
+    return BACK_ROUNDED_VOWELS.includes(firstSuffixVowel) || firstSuffixVowel === "u";
+  }
+  if (FRONT_ROUNDED_VOWELS.includes(lv)) {
+    return FRONT_ROUNDED_VOWELS.includes(firstSuffixVowel) || firstSuffixVowel === "\xFC";
+  }
+  return false;
+}
+function stripSuffix(word, suffix) {
+  return word.slice(0, word.length - suffix.length);
+}
+function checkPreconditionN(word) {
+  if (word.endsWith("n") && word.length > 1) {
+    return isVowel(word[word.length - 2]);
+  }
+  return word.length > 0 && !word.endsWith("n") && isVowel(word[word.length - 1]);
+}
+function checkPreconditionY(word) {
+  if (word.endsWith("y") && word.length > 1) {
+    return isVowel(word[word.length - 2]);
+  }
+  return word.length > 0 && !word.endsWith("y") && isVowel(word[word.length - 1]);
+}
+function tryStripWithHarmony(word, suffixes, minVowels) {
+  for (const suffix of suffixes) {
+    if (word.endsWith(suffix)) {
+      const candidate = stripSuffix(word, suffix);
+      if (vowelCount(candidate) >= minVowels && checkVowelHarmony(candidate, suffix)) {
+        return candidate;
+      }
+    }
+  }
+  return word;
+}
+function tryStripWithHarmonyAndPrecondN(word, suffixes, minVowels) {
+  for (const suffix of suffixes) {
+    if (word.endsWith(suffix)) {
+      const candidate = stripSuffix(word, suffix);
+      if (vowelCount(candidate) >= minVowels && checkVowelHarmony(candidate, suffix) && checkPreconditionN(candidate)) {
+        return candidate;
+      }
+    }
+  }
+  return word;
+}
+function tryStripWithHarmonyAndPrecondY(word, suffixes, minVowels) {
+  for (const suffix of suffixes) {
+    if (word.endsWith(suffix)) {
+      const candidate = stripSuffix(word, suffix);
+      if (vowelCount(candidate) >= minVowels && checkVowelHarmony(candidate, suffix) && checkPreconditionY(candidate)) {
+        return candidate;
+      }
+    }
+  }
+  return word;
+}
+function tryStripMinVowels(word, suffixes, minVowels) {
+  for (const suffix of suffixes) {
+    if (word.endsWith(suffix)) {
+      const candidate = stripSuffix(word, suffix);
+      if (vowelCount(candidate) >= minVowels) return candidate;
+    }
+  }
+  return word;
+}
+function tryRemovePossessiveLari(word) {
+  return tryStripMinVowels(word, ["lar\u0131", "leri"], 2);
+}
+function tryRemovePlural(word) {
+  return tryStripWithHarmony(word, ["lar", "ler"], 2);
+}
+function tryRemovePossessiveM(word) {
+  return tryStripWithHarmony(word, ["\u0131m", "im", "um", "\xFCm", "m"], 2);
+}
+function tryRemovePossessiveN(word) {
+  return tryStripWithHarmony(word, ["\u0131n", "in", "un", "\xFCn", "n"], 2);
+}
+function tryRemovePossessiveMiz(word) {
+  return tryStripWithHarmony(word, ["m\u0131z", "miz", "muz", "m\xFCz"], 2);
+}
+function tryRemovePossessiveNiz(word) {
+  return tryStripWithHarmony(word, ["n\u0131z", "niz", "nuz", "n\xFCz"], 2);
+}
+function tryRemoveCaseDa(word) {
+  return tryStripWithHarmony(word, ["da", "de", "ta", "te"], 2);
+}
+function tryRemoveCaseDan(word) {
+  return tryStripWithHarmony(word, ["dan", "den", "tan", "ten"], 2);
+}
+function tryRemoveCaseNa(word) {
+  return tryStripWithHarmony(word, ["na", "ne"], 2);
+}
+function tryRemoveCaseA(word) {
+  return tryStripWithHarmony(word, ["a", "e"], 2);
+}
+function tryRemoveGenitive(word) {
+  return tryStripWithHarmonyAndPrecondN(word, ["\u0131n", "in", "un", "\xFCn"], 2);
+}
+function tryRemoveAccusative(word) {
+  const long = tryStripWithHarmony(word, ["n\u0131", "ni", "nu", "n\xFC"], 2);
+  if (long !== word) return long;
+  return tryStripWithHarmonyAndPrecondY(word, ["\u0131", "i", "u", "\xFC"], 2);
+}
+function tryRemoveCopulaDi(word) {
+  return tryStripWithHarmonyAndPrecondY(
+    word,
+    [
+      "dik",
+      "tik",
+      "duk",
+      "tuk",
+      "d\xFCk",
+      "t\xFCk",
+      "d\u0131k",
+      "t\u0131k",
+      "dim",
+      "tim",
+      "dum",
+      "tum",
+      "d\xFCm",
+      "t\xFCm",
+      "d\u0131m",
+      "t\u0131m",
+      "din",
+      "tin",
+      "dun",
+      "tun",
+      "d\xFCn",
+      "t\xFCn",
+      "d\u0131n",
+      "t\u0131n",
+      "di",
+      "ti",
+      "du",
+      "tu",
+      "d\xFC",
+      "t\xFC",
+      "d\u0131",
+      "t\u0131"
+    ],
+    2
+  );
+}
+function tryRemoveCopulaSa(word) {
+  return tryStripWithHarmonyAndPrecondY(word, ["sak", "sek", "sam", "sem", "san", "sen", "sa", "se"], 2);
+}
+function tryRemoveCopulaMis(word) {
+  return tryStripWithHarmonyAndPrecondY(word, ["mi\u015F", "mu\u015F", "m\xFC\u015F", "m\u0131\u015F"], 2);
+}
+function tryRemoveCopulaSiniz(word) {
+  return tryStripWithHarmony(word, ["siniz", "sunuz", "s\xFCn\xFCz", "s\u0131n\u0131z"], 2);
+}
+function tryRemoveCopulaLar(word) {
+  return tryStripWithHarmony(word, ["lar", "ler"], 2);
+}
+function tryRemoveCopulaCasina(word) {
+  return tryStripMinVowels(word, ["cas\u0131na", "cesine"], 2);
+}
+function tryRemoveCopulaDir(word) {
+  return tryStripWithHarmony(word, ["dir", "tir", "dur", "tur", "d\xFCr", "t\xFCr", "d\u0131r", "t\u0131r"], 2);
+}
+function tryRemoveWithLa(word) {
+  return tryStripWithHarmonyAndPrecondY(word, ["la", "le"], 2);
+}
+function tryRemoveWithCa(word) {
+  return tryStripWithHarmonyAndPrecondN(word, ["ca", "ce"], 2);
+}
+function tryRemoveKen(word) {
+  if (word.endsWith("ken")) {
+    const candidate = stripSuffix(word, "ken");
+    if (vowelCount(candidate) >= 2 && checkPreconditionY(candidate)) {
+      return candidate;
+    }
+  }
+  return word;
+}
+function changed(original, current) {
+  return original !== current;
+}
+var CONSONANT_SOFTENING = {
+  b: "p",
+  c: "\xE7",
+  d: "t",
+  \u011F: "k"
+};
+function handleConsonantSoftening(word) {
+  if (word.length > 0) {
+    const last = word[word.length - 1];
+    if (last in CONSONANT_SOFTENING) {
+      return word.slice(0, -1) + CONSONANT_SOFTENING[last];
+    }
+  }
+  return word;
+}
+function handleVowelInsertion(word) {
+  if (word.length < 2) return word;
+  const last = word[word.length - 1];
+  if (last !== "d" && last !== "g") return word;
+  const beforeLast = word.slice(0, -1);
+  const lv = lastVowel(beforeLast);
+  if (lv === null) return word;
+  let insertedVowel;
+  if (BACK_UNROUNDED_VOWELS.includes(lv)) insertedVowel = "\u0131";
+  else if (FRONT_UNROUNDED_VOWELS.includes(lv)) insertedVowel = "i";
+  else if (BACK_ROUNDED_VOWELS.includes(lv)) insertedVowel = "u";
+  else if (FRONT_ROUNDED_VOWELS.includes(lv)) insertedVowel = "\xFC";
+  else insertedVowel = "\u0131";
+  return beforeLast + insertedVowel + last;
+}
+function nominalVerbSuffixes(word) {
+  const original = word;
+  let removed = false;
+  let result;
+  result = tryRemoveCopulaMis(word);
+  if (changed(word, result)) {
+    word = result;
+    removed = true;
+  }
+  if (!removed) {
+    result = tryRemoveCopulaDi(word);
+    if (changed(word, result)) {
+      word = result;
+      removed = true;
+    }
+  }
+  if (!removed) {
+    result = tryRemoveCopulaSa(word);
+    if (changed(word, result)) {
+      word = result;
+      removed = true;
+    }
+  }
+  if (!removed) {
+    result = tryRemoveKen(word);
+    if (changed(word, result)) {
+      word = result;
+      removed = true;
+    }
+  }
+  if (!removed) {
+    result = tryRemoveCopulaCasina(word);
+    if (changed(word, result)) {
+      word = result;
+      result = tryRemoveCopulaSiniz(word);
+      if (changed(word, result)) {
+        word = result;
+      } else {
+        result = tryRemoveCopulaLar(word);
+        if (changed(word, result)) {
+          word = result;
+        } else {
+          result = tryRemoveCopulaDir(word);
+          if (changed(word, result)) {
+            word = result;
+          } else {
+            result = tryRemoveCopulaMis(word);
+            if (changed(word, result)) word = result;
+          }
+        }
+      }
+      return { word, removed: true };
+    }
+  }
+  if (removed) {
+    result = tryRemovePossessiveM(word);
+    if (changed(word, result)) {
+      word = result;
+    } else {
+      result = tryRemovePossessiveN(word);
+      if (changed(word, result)) {
+        word = result;
+      } else {
+        result = tryRemovePossessiveMiz(word);
+        if (changed(word, result)) {
+          word = result;
+        } else {
+          result = tryRemovePossessiveNiz(word);
+          if (changed(word, result)) {
+            word = result;
+          } else {
+            result = tryRemoveCopulaSiniz(word);
+            if (changed(word, result)) {
+              word = result;
+            } else {
+              result = tryRemoveCopulaLar(word);
+              if (changed(word, result)) word = result;
+            }
+          }
+        }
+      }
+    }
+    result = tryRemoveCopulaMis(word);
+    if (changed(word, result)) word = result;
+  }
+  return { word, removed: changed(original, word) };
+}
+function nounSuffixes(word) {
+  let result;
+  result = tryRemovePossessiveLari(word);
+  if (changed(word, result)) return result;
+  result = tryRemovePlural(word);
+  if (changed(word, result)) {
+    word = result;
+    result = tryRemoveAccusative(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveLari(word);
+    if (changed(word, result)) return result;
+    return word;
+  }
+  result = tryRemoveCaseDa(word);
+  if (changed(word, result)) {
+    word = result;
+    result = tryRemovePossessiveLari(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveM(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveN(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveMiz(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveNiz(word);
+    if (changed(word, result)) return result;
+    return word;
+  }
+  result = tryRemoveCaseDan(word);
+  if (changed(word, result)) {
+    word = result;
+    result = tryRemovePossessiveLari(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveM(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveN(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveMiz(word);
+    if (changed(word, result)) return result;
+    result = tryRemovePossessiveNiz(word);
+    if (changed(word, result)) return result;
+    return word;
+  }
+  result = tryRemoveCaseNa(word);
+  if (changed(word, result)) {
+    word = result;
+    result = tryRemoveGenitive(word);
+    if (changed(word, result)) return result;
+    result = tryRemoveAccusative(word);
+    if (changed(word, result)) return result;
+    return word;
+  }
+  result = tryRemoveGenitive(word);
+  if (changed(word, result)) {
+    word = result;
+    result = tryRemovePossessiveLari(word);
+    if (changed(word, result)) return result;
+    result = tryRemoveCopulaMis(word);
+    if (changed(word, result)) return result;
+    return word;
+  }
+  result = tryRemoveAccusative(word);
+  if (changed(word, result)) {
+    word = result;
+    result = tryRemoveCopulaMis(word);
+    if (changed(word, result)) return result;
+    return word;
+  }
+  result = tryRemoveCaseA(word);
+  if (changed(word, result)) return result;
+  result = tryRemoveWithCa(word);
+  if (changed(word, result)) {
+    word = result;
+    result = tryRemoveGenitive(word);
+    if (changed(word, result)) return result;
+    result = tryRemoveAccusative(word);
+    if (changed(word, result)) return result;
+    return word;
+  }
+  result = tryRemoveWithLa(word);
+  if (changed(word, result)) return result;
+  return word;
+}
+function stem(word) {
+  if (vowelCount(word) < 2) return word;
+  const { word: afterNominal, removed: nominalRemoved } = nominalVerbSuffixes(word);
+  word = afterNominal;
+  if (!nominalRemoved) return word;
+  word = nounSuffixes(word);
+  if (word.endsWith("soyad") || word === "ad") return word;
+  word = handleVowelInsertion(word);
+  word = handleConsonantSoftening(word);
+  return word;
+}
+var stopWords = /* @__PURE__ */ new Set([
+  "acaba",
+  "acep",
+  "adeta",
+  "altm\u0131\u015F",
+  "alt\u0131",
+  "ama",
+  "ancak",
+  "arada",
+  "asl\u0131nda",
+  "aynen",
+  "ayr\u0131ca",
+  "az",
+  "bana",
+  "bari",
+  "bazen",
+  "baz\u0131",
+  "belki",
+  "ben",
+  "benden",
+  "beni",
+  "benim",
+  "beri",
+  "be\u015F",
+  "bile",
+  "bin",
+  "bir",
+  "biraz",
+  "biri",
+  "birka\xE7",
+  "birkez",
+  "bir\xE7ok",
+  "bir\u015Fey",
+  "bir\u015Feyi",
+  "biz",
+  "bizden",
+  "bize",
+  "bizi",
+  "bizim",
+  "bu",
+  "buna",
+  "bunda",
+  "bundan",
+  "bunlar",
+  "bunlar\u0131",
+  "bunlar\u0131n",
+  "bunu",
+  "bunun",
+  "burada",
+  "b\xF6yle",
+  "b\xF6ylece",
+  "b\xFCt\xFCn",
+  "da",
+  "daha",
+  "dahi",
+  "dahil",
+  "daima",
+  "dair",
+  "dayanarak",
+  "de",
+  "defa",
+  "de\u011Fil",
+  "diye",
+  "di\u011Fer",
+  "doksan",
+  "dokuz",
+  "dolay\u0131",
+  "dolay\u0131s\u0131yla",
+  "d\xF6rt",
+  "edecek",
+  "eden",
+  "ederek",
+  "edilecek",
+  "ediliyor",
+  "edilmesi",
+  "ediyor",
+  "elli",
+  "en",
+  "etmesi",
+  "etti",
+  "etti\u011Fi",
+  "etti\u011Fini",
+  "e\u011Fer",
+  "fakat",
+  "gibi",
+  "g\xF6re",
+  "halbuki",
+  "halen",
+  "hangi",
+  "hani",
+  "hari\xE7",
+  "hatta",
+  "hele",
+  "hem",
+  "hen\xFCz",
+  "hep",
+  "hepsi",
+  "her",
+  "herhangi",
+  "herkes",
+  "herkesin",
+  "hi\xE7",
+  "hi\xE7bir",
+  "iken",
+  "iki",
+  "ila",
+  "ile",
+  "ilgili",
+  "ilk",
+  "illa",
+  "ise",
+  "itibaren",
+  "itibariyle",
+  "iyi",
+  "iyice",
+  "i\xE7in",
+  "i\u015Fte",
+  "kadar",
+  "kar\u015F\u0131n",
+  "katrilyon",
+  "kendi",
+  "kendilerine",
+  "kendini",
+  "kendisi",
+  "kendisine",
+  "kendisini",
+  "kere",
+  "kez",
+  "ki",
+  "kim",
+  "kimden",
+  "kime",
+  "kimi",
+  "kimse",
+  "k\u0131rk",
+  "lakin",
+  "madem",
+  "milyar",
+  "milyon",
+  "mu",
+  "m\xFC",
+  "m\u0131",
+  "nas\u0131l",
+  "ne",
+  "neden",
+  "nedenle",
+  "nerde",
+  "nere",
+  "nerede",
+  "nereye",
+  "nitekim",
+  "niye",
+  "ni\xE7in",
+  "o",
+  "olan",
+  "olarak",
+  "oldu",
+  "olduklar\u0131n\u0131",
+  "oldu\u011Fu",
+  "oldu\u011Funu",
+  "olmad\u0131",
+  "olmad\u0131\u011F\u0131",
+  "olmak",
+  "olmas\u0131",
+  "olmayan",
+  "olmaz",
+  "olsa",
+  "olsun",
+  "olup",
+  "olur",
+  "olursa",
+  "oluyor",
+  "on",
+  "ona",
+  "ondan",
+  "onlar",
+  "onlardan",
+  "onlari",
+  "onlar\u0131",
+  "onlar\u0131n",
+  "onu",
+  "onun",
+  "otuz",
+  "oysa",
+  "pek",
+  "ra\u011Fmen",
+  "sadece",
+  "sanki",
+  "sekiz",
+  "seksen",
+  "sen",
+  "senden",
+  "seni",
+  "senin",
+  "siz",
+  "sizden",
+  "sizi",
+  "sizin",
+  "sonra",
+  "taraf\u0131ndan",
+  "trilyon",
+  "t\xFCm",
+  "var",
+  "vard\u0131",
+  "ve",
+  "veya",
+  "veyahut",
+  "ya",
+  "yahut",
+  "yani",
+  "yapacak",
+  "yapmak",
+  "yapt\u0131",
+  "yapt\u0131klar\u0131",
+  "yapt\u0131\u011F\u0131",
+  "yapt\u0131\u011F\u0131n\u0131",
+  "yap\u0131lan",
+  "yap\u0131lmas\u0131",
+  "yap\u0131yor",
+  "yedi",
+  "yerine",
+  "yetmi\u015F",
+  "yine",
+  "yirmi",
+  "yoksa",
+  "y\xFCz",
+  "zaten",
+  "\xE7ok",
+  "\xE7\xFCnk\xFC",
+  "\xF6yle",
+  "\xFCzere",
+  "\xFC\xE7",
+  "\u015Fey",
+  "\u015Feyden",
+  "\u015Feyi",
+  "\u015Feyler",
+  "\u015Fu",
+  "\u015Funa",
+  "\u015Funda",
+  "\u015Fundan",
+  "\u015Funlar\u0131",
+  "\u015Funu",
+  "\u015F\xF6yle"
+]);
+var turkish = {
+  name: "turkish",
+  stemmer: stem,
+  stopWords,
+  tokenizer: { splitPattern: /[^a-z0-9çğıöşü]+/gi }
+};
+
+export { turkish };

package/dist/languages/twi.d.ts

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