@ingglish/phonemes 0.1.0

package/dist/index.js

@@ -0,0 +1,649 @@
+// src/arpabet.ts
+var ARPABET_VOWELS = [
+  "AA",
+  // father, hot
+  "AE",
+  // cat, bat
+  "AH",
+  // but, cup (stressed) / schwa (unstressed)
+  "AO",
+  // thought, law
+  "AW",
+  // cow, how
+  "AY",
+  // my, time
+  "EH",
+  // bed, red
+  "ER",
+  // bird, her
+  "EY",
+  // say, day
+  "IH",
+  // bit, sit
+  "IY",
+  // bee, see
+  "OW",
+  // go, show
+  "OY",
+  // boy, toy
+  "UH",
+  // book, put
+  "UW"
+  // too, blue
+];
+var ARPABET_CONSONANTS = [
+  // Stops
+  "B",
+  "D",
+  "G",
+  "K",
+  "P",
+  "T",
+  // Fricatives
+  "DH",
+  // the, this
+  "F",
+  "HH",
+  // hat
+  "S",
+  "SH",
+  // ship
+  "TH",
+  // think
+  "V",
+  "Z",
+  "ZH",
+  // measure
+  // Affricates
+  "CH",
+  // chat
+  "JH",
+  // just
+  // Nasals
+  "M",
+  "N",
+  "NG",
+  // sing
+  // Liquids
+  "L",
+  "R",
+  // Glides
+  "W",
+  "Y"
+];
+var VOWELS_SET = new Set(ARPABET_VOWELS);
+var CONSONANTS_SET = new Set(ARPABET_CONSONANTS);
+var STRESS_MARKER_REGEX = /[012]$/;
+function getStress(phoneme) {
+  const lastChar = phoneme.codePointAt(phoneme.length - 1);
+  if (lastChar >= 48 && lastChar <= 50) {
+    return lastChar - 48;
+  }
+  return null;
+}
+function isVowel(phoneme) {
+  const base = stripStress(phoneme);
+  return VOWELS_SET.has(base);
+}
+function stripStress(phoneme) {
+  const lastChar = phoneme.codePointAt(phoneme.length - 1);
+  if (lastChar >= 48 && lastChar <= 50) {
+    return phoneme.slice(0, -1);
+  }
+  return phoneme;
+}
+
+// src/phonotactics.ts
+var VALID_ONSETS = /* @__PURE__ */ new Set([
+  // Single consonants (all are valid onsets except NG)
+  "B",
+  // Two-consonant clusters: consonant + liquid (L, R)
+  "B L",
+  "B R",
+  // Two-consonant clusters: consonant + glide (W, Y)
+  "B Y",
+  "CH",
+  "D",
+  "DH",
+  "D R",
+  "D W",
+  "D Y",
+  "F",
+  "F L",
+  "F R",
+  "F Y",
+  "G",
+  "G L",
+  "G R",
+  "G W",
+  "G Y",
+  "HH",
+  "HH W",
+  "HH Y",
+  "JH",
+  "K",
+  "K L",
+  "K R",
+  "K W",
+  "K Y",
+  "L",
+  "L Y",
+  "M",
+  "M Y",
+  "N",
+  "N Y",
+  "P",
+  "P L",
+  "P R",
+  "P Y",
+  "R",
+  "S",
+  "SH",
+  "SH R",
+  // Two-consonant clusters: s + consonant
+  "S K",
+  // Three-consonant clusters: s + stop + liquid/glide
+  "S K R",
+  "S K W",
+  "S K Y",
+  "S L",
+  "S M",
+  "S N",
+  "S P",
+  "S P L",
+  "S P R",
+  "S P Y",
+  "S T",
+  "S T R",
+  "S T Y",
+  "S W",
+  "S Y",
+  "T",
+  "TH",
+  "TH R",
+  "TH W",
+  "T R",
+  "T W",
+  "T Y",
+  "V",
+  "V Y",
+  "W",
+  "Y",
+  "Z",
+  "ZH"
+]);
+function findOnsetStart(consonants) {
+  if (consonants.length === 0) {
+    return 0;
+  }
+  for (let start = 0; start < consonants.length; start++) {
+    const candidate = consonants.slice(start);
+    if (isValidOnset(candidate)) {
+      return start;
+    }
+  }
+  return consonants.length - 1;
+}
+function isValidOnset(consonants) {
+  if (consonants.length === 0) {
+    return true;
+  }
+  const key = consonants.join(" ");
+  return VALID_ONSETS.has(key);
+}
+
+// src/format-registry.ts
+var registry = /* @__PURE__ */ new Map();
+var isLatinScriptCache = /* @__PURE__ */ new Map();
+var preservesCaseCache = /* @__PURE__ */ new Map();
+function getFormatHandler(name) {
+  return registry.get(name);
+}
+function getFormatIsLatinScript(name) {
+  return isLatinScriptCache.get(name) ?? true;
+}
+function getFormatJoinSeparator(name) {
+  return registry.get(name)?.joinSeparator ?? "";
+}
+function getFormatLabel(name) {
+  return registry.get(name)?.label ?? name;
+}
+function getFormatNativeLabel(name) {
+  const handler = registry.get(name);
+  return handler?.nativeLabel ?? handler?.label ?? name;
+}
+function getFormatPreservesCase(name) {
+  return preservesCaseCache.get(name) ?? true;
+}
+function registerFormat(name, handler) {
+  const existing = registry.get(name);
+  const merged = { ...existing, ...handler };
+  registry.set(name, merged);
+  isLatinScriptCache.set(name, merged.isLatinScript ?? true);
+  preservesCaseCache.set(name, merged.preservesCase ?? merged.isLatinScript ?? true);
+}
+
+// src/ingglish-maps.ts
+var INGGLISH_VOWEL_MAP = {
+  // Monophthongs
+  AA: "o",
+  // father, hot, rock (but AA+R → 'ar' in star, car)
+  AE: "a",
+  // cat, bat, had (but AE+R → 'arr' in arrow, barrow)
+  AH: "uh",
+  // but, cup, son (stressed /ʌ/; unstressed /ə/ AH0 → 'a' in conversion)
+  AO: "aw",
+  // thought, caught, law (but AO+R → 'or' in store, more)
+  // Diphthongs
+  AW: "ou",
+  // cow, how, out
+  AY: "ai",
+  // my, eye, time
+  EH: "e",
+  // bed, red, said (but EH+R → 'air' in air, care, there)
+  ER: "er",
+  // bird, her, nurse
+  EY: "ay",
+  // say, day, make
+  IH: "i",
+  // bit, sit, gym
+  IY: "ee",
+  // bee, see, machine
+  OW: "oh",
+  // go, show, coat
+  OY: "oi",
+  // boy, toy, coin
+  UH: "u",
+  // book, put, could
+  UW: "oo"
+  // too, blue, food
+};
+var INGGLISH_CONSONANT_MAP = {
+  // Stops (plosives)
+  B: "b",
+  // bat, cab
+  // Affricates
+  CH: "ch",
+  // chat, batch
+  D: "d",
+  // dog, bed
+  // Fricatives
+  DH: "dh",
+  // the, this (voiced) - distinguishes from TH
+  F: "f",
+  // fat, laugh
+  G: "g",
+  // go, big
+  // Aspirate
+  HH: "h",
+  // hat, ahead
+  JH: "j",
+  // just, edge
+  K: "k",
+  // cat, back
+  // Liquids
+  L: "l",
+  // let, well
+  // Nasals
+  M: "m",
+  // man, come
+  N: "n",
+  // no, pen
+  NG: "ng",
+  // sing, thing
+  P: "p",
+  // pat, cup
+  R: "r",
+  // run, car
+  S: "s",
+  // sat, miss
+  SH: "sh",
+  // she, push
+  T: "t",
+  // top, cat
+  TH: "th",
+  // think, bath (voiceless)
+  V: "v",
+  // van, love
+  // Semivowels (glides)
+  W: "w",
+  // wet, away
+  Y: "y",
+  // yes, you
+  Z: "z",
+  // zoo, is
+  ZH: "zh"
+  // measure, beige
+};
+var ARPABET_TO_INGGLISH_MAP = {
+  ...INGGLISH_VOWEL_MAP,
+  ...INGGLISH_CONSONANT_MAP
+};
+var INGGLISH_TO_ARPABET_MAP = Object.fromEntries(
+  Object.entries(ARPABET_TO_INGGLISH_MAP).map(([arpabet, ingglish]) => [ingglish, arpabet])
+);
+var R_COLORED_VOWELS = [
+  { arpabet: "AA", prefix: "a" },
+  // star, car, far → 'ar'
+  { arpabet: "AO", prefix: "o" },
+  // store, more, for → 'or'
+  { arpabet: "EH", prefix: "ai" },
+  // air, care, there → 'air'
+  { arpabet: "AE", prefix: "ar" },
+  // arrow, barrow, carrot → 'arr'
+  { arpabet: "IH", prefix: "ee" },
+  // beer, beard, fear → 'eer'
+  { arpabet: "UH", prefix: "u" },
+  // tour, cure, pure → 'ur' (CURE vowel, experimentable)
+  { arpabet: "AH", prefix: "uh" }
+  // curry, burroughs → 'uhr' (AH=uh, prevents AH0+R collision with 'ar')
+];
+var R_COLORED_FORWARD = new Map(
+  R_COLORED_VOWELS.map(({ arpabet, prefix }) => [arpabet, prefix])
+);
+var R_COLORED_REVERSE_3CHAR = Object.fromEntries(
+  R_COLORED_VOWELS.filter(({ prefix }) => prefix.length === 2).map(({ arpabet, prefix }) => [
+    prefix + "r",
+    [arpabet, "R"]
+  ])
+);
+var R_COLORED_REVERSE_2CHAR = Object.fromEntries(
+  R_COLORED_VOWELS.filter(({ prefix }) => prefix.length === 1).map(({ arpabet, prefix }) => [
+    prefix + "r",
+    [arpabet, "R"]
+  ])
+);
+
+// src/to-ingglish.ts
+function arpabetPhonemeToIngglish(phoneme) {
+  if (phoneme === "AH0") {
+    return "a";
+  }
+  const base = stripStress(phoneme);
+  return ARPABET_TO_INGGLISH_MAP[base] ?? phoneme.toLowerCase();
+}
+function convertArpabet(arpabet, phonemeMap, rColoredMap, stressOverrides) {
+  let result = "";
+  const len = arpabet.length;
+  for (let i = 0; i < len; i++) {
+    const phoneme = arpabet[i];
+    const base = stripStress(phoneme);
+    if (i + 1 < len && arpabet[i + 1] === "R") {
+      const rPrefix = rColoredMap.get(base);
+      if (rPrefix !== void 0) {
+        result += rPrefix;
+        continue;
+      }
+    }
+    const stressOverride = stressOverrides.get(phoneme);
+    if (stressOverride !== void 0) {
+      result += stressOverride;
+      continue;
+    }
+    result += phonemeMap[base] ?? phoneme.toLowerCase();
+  }
+  return result;
+}
+var INGGLISH_FULL_MAP = {};
+for (const [base, spelling] of Object.entries(ARPABET_TO_INGGLISH_MAP)) {
+  INGGLISH_FULL_MAP[base] = spelling;
+  INGGLISH_FULL_MAP[base + "0"] = spelling;
+  INGGLISH_FULL_MAP[base + "1"] = spelling;
+  INGGLISH_FULL_MAP[base + "2"] = spelling;
+}
+INGGLISH_FULL_MAP.AH0 = "a";
+function arpabetToIngglish(arpabet) {
+  let result = "";
+  const len = arpabet.length;
+  for (let i = 0; i < len; i++) {
+    const phoneme = arpabet[i];
+    if (i + 1 < len && arpabet[i + 1] === "R") {
+      const base = stripStress(phoneme);
+      const rPrefix = R_COLORED_FORWARD.get(base);
+      if (rPrefix !== void 0) {
+        result += rPrefix;
+        continue;
+      }
+    }
+    result += INGGLISH_FULL_MAP[phoneme] ?? phoneme.toLowerCase();
+  }
+  return result;
+}
+registerFormat("ingglish", {
+  forward: arpabetToIngglish,
+  isLatinScript: true,
+  label: "Ingglish",
+  preservesCase: true
+});
+var EMPTY_R_COLORED = /* @__PURE__ */ new Map();
+var INGGLISH_STRESS_OVERRIDES = /* @__PURE__ */ new Map([["AH0", "a"]]);
+function arpabetToFormat(arpabet, format = "ingglish", options) {
+  if (format === "ingglish") {
+    if (options?.disableRColoring === true) {
+      return convertArpabet(
+        arpabet,
+        ARPABET_TO_INGGLISH_MAP,
+        EMPTY_R_COLORED,
+        INGGLISH_STRESS_OVERRIDES
+      );
+    }
+    return arpabetToIngglish(arpabet);
+  }
+  const handler = getFormatHandler(format);
+  if (handler?.forward) {
+    return handler.forward(arpabet, options);
+  }
+  return arpabetToIngglish(arpabet);
+}
+
+// src/to-pronunciation.ts
+var GUIDE_MAP = {};
+for (const [base, spelling] of Object.entries(ARPABET_TO_INGGLISH_MAP)) {
+  GUIDE_MAP[base] = spelling;
+  GUIDE_MAP[base + "0"] = spelling;
+  GUIDE_MAP[base + "1"] = spelling;
+  GUIDE_MAP[base + "2"] = spelling;
+}
+GUIDE_MAP.AH0 = "a";
+function arpabetToPronunciation(arpabet) {
+  const syllables = syllabify(arpabet);
+  return syllables.map(({ phonemes, stress }) => {
+    const spelling = syllableToSpelling(phonemes);
+    return stress >= 1 ? spelling.toUpperCase() : spelling;
+  }).join("-");
+}
+function registerPronunciation() {
+  registerFormat("pronunciation", {
+    forward: arpabetToPronunciation,
+    isLatinScript: true,
+    label: "Guide",
+    preservesCase: false
+  });
+}
+function syllabify(arpabet) {
+  const vowelIndices = [];
+  for (const [i, phoneme] of arpabet.entries()) {
+    if (isVowel(phoneme)) {
+      vowelIndices.push(i);
+    }
+  }
+  if (vowelIndices.length === 0) {
+    return [{ phonemes: arpabet, stress: 0 }];
+  }
+  const syllables = [];
+  let start = 0;
+  for (let vi = 0; vi < vowelIndices.length; vi++) {
+    const vowelIdx = vowelIndices[vi];
+    const stress = getStress(arpabet[vowelIdx]) ?? 0;
+    if (vi < vowelIndices.length - 1) {
+      const nextVowelIdx = vowelIndices[vi + 1];
+      const consonantStart = vowelIdx + 1;
+      const consonants = [];
+      for (let j = consonantStart; j < nextVowelIdx; j++) {
+        consonants.push(stripStress(arpabet[j]));
+      }
+      if (consonants.length === 0) {
+        syllables.push({ phonemes: arpabet.slice(start, consonantStart), stress });
+        start = consonantStart;
+      } else {
+        const onsetIdx = findOnsetStart(consonants);
+        const boundary = consonantStart + onsetIdx;
+        syllables.push({ phonemes: arpabet.slice(start, boundary), stress });
+        start = boundary;
+      }
+    } else {
+      syllables.push({ phonemes: arpabet.slice(start), stress });
+    }
+  }
+  return syllables;
+}
+function syllableToSpelling(phonemes) {
+  let result = "";
+  for (let i = 0; i < phonemes.length; i++) {
+    const phoneme = phonemes[i];
+    if (i + 1 < phonemes.length && phonemes[i + 1] === "R") {
+      const base = stripStress(phoneme);
+      const rPrefix = R_COLORED_FORWARD.get(base);
+      if (rPrefix !== void 0) {
+        result += rPrefix;
+        continue;
+      }
+    }
+    result += GUIDE_MAP[phoneme] ?? phoneme.toLowerCase();
+  }
+  return result;
+}
+
+// src/from-ingglish.ts
+var ARPABET_ALTERNATIVES = {
+  AE: [["AH"]],
+  // "a" could be AE (cat) or AH (schwa: about, the)
+  ER: [["EH", "R"]],
+  SH: [["S", "HH"]]
+  // "sh" could be SH (ship) or S+HH (exhume)
+};
+var ARPABET_ALTERNATIVES_ENTRIES = Object.entries(ARPABET_ALTERNATIVES);
+function expandArpabetAlternatives(arpabet) {
+  const results = [arpabet];
+  for (let i = 0; i < arpabet.length; i++) {
+    const alternatives = ARPABET_ALTERNATIVES[arpabet[i]];
+    if (alternatives !== void 0) {
+      for (const alt of alternatives) {
+        const expanded = [...arpabet.slice(0, i), ...alt, ...arpabet.slice(i + 1)];
+        results.push(expanded);
+      }
+    }
+  }
+  for (const [phoneme, alts] of ARPABET_ALTERNATIVES_ENTRIES) {
+    for (const alt of alts) {
+      if (alt.length === 1) {
+        let count = 0;
+        for (const p of arpabet) {
+          if (p === phoneme) {
+            count++;
+          }
+        }
+        if (count >= 2) {
+          results.push(arpabet.map((p) => p === phoneme ? alt[0] : p));
+        }
+      }
+    }
+  }
+  return results;
+}
+var TWO_CHAR_SPELLINGS = new Set(
+  Object.keys(INGGLISH_TO_ARPABET_MAP).filter((s) => s.length === 2)
+);
+var ONE_CHAR_SPELLINGS = new Set(
+  Object.keys(INGGLISH_TO_ARPABET_MAP).filter((s) => s.length === 1)
+);
+function ingglishToArpabet(ingglish) {
+  const result = [];
+  const str = ingglish.toLowerCase();
+  const len = str.length;
+  let pos = 0;
+  while (pos < len) {
+    if (pos + 3 <= len) {
+      const threeChar = str.slice(pos, pos + 3);
+      if (threeChar in R_COLORED_REVERSE_3CHAR) {
+        result.push(...R_COLORED_REVERSE_3CHAR[threeChar]);
+        pos += 3;
+        continue;
+      }
+    }
+    if (pos + 2 <= len) {
+      const twoChar = str.slice(pos, pos + 2);
+      if (twoChar in R_COLORED_REVERSE_2CHAR) {
+        result.push(...R_COLORED_REVERSE_2CHAR[twoChar]);
+        pos += 2;
+        continue;
+      }
+      if (TWO_CHAR_SPELLINGS.has(twoChar)) {
+        result.push(INGGLISH_TO_ARPABET_MAP[twoChar]);
+        pos += 2;
+        continue;
+      }
+    }
+    const oneChar = str[pos];
+    if (ONE_CHAR_SPELLINGS.has(oneChar)) {
+      result.push(INGGLISH_TO_ARPABET_MAP[oneChar]);
+      pos += 1;
+      continue;
+    }
+    pos += 1;
+  }
+  return result.length > 0 ? result : null;
+}
+
+// src/custom-format.ts
+function createCustomConverter(config) {
+  const mergedMap = { ...ARPABET_TO_INGGLISH_MAP, ...config.phonemeMap };
+  const mergedRColored = new Map(R_COLORED_FORWARD);
+  for (const [vowel] of R_COLORED_FORWARD) {
+    if (vowel in config.phonemeMap && !(vowel in config.rColoredPrefixes)) {
+      mergedRColored.set(vowel, config.phonemeMap[vowel]);
+    }
+  }
+  for (const [key, value] of Object.entries(config.rColoredPrefixes)) {
+    mergedRColored.set(key, value);
+  }
+  const stressOverrides = /* @__PURE__ */ new Map([["AH0", "a"]]);
+  for (const [key, value] of Object.entries(config.phonemeMap)) {
+    const lastChar = key.codePointAt(key.length - 1);
+    if (lastChar >= 48 && lastChar <= 50) {
+      stressOverrides.set(key, value);
+    }
+  }
+  const emptyRColored = /* @__PURE__ */ new Map();
+  return (arpabet, options) => convertArpabet(
+    arpabet,
+    mergedMap,
+    options?.disableRColoring === true ? emptyRColored : mergedRColored,
+    stressOverrides
+  );
+}
+export {
+  ARPABET_CONSONANTS,
+  ARPABET_TO_INGGLISH_MAP,
+  ARPABET_VOWELS,
+  R_COLORED_FORWARD,
+  STRESS_MARKER_REGEX,
+  arpabetPhonemeToIngglish,
+  arpabetToFormat,
+  arpabetToIngglish,
+  createCustomConverter,
+  expandArpabetAlternatives,
+  findOnsetStart,
+  getFormatHandler,
+  getFormatIsLatinScript,
+  getFormatJoinSeparator,
+  getFormatLabel,
+  getFormatNativeLabel,
+  getFormatPreservesCase,
+  getStress,
+  ingglishToArpabet,
+  isVowel,
+  registerFormat,
+  registerPronunciation,
+  stripStress
+};

package/package.json

@@ -0,0 +1,48 @@
+{
+  "name": "@ingglish/phonemes",
+  "version": "0.1.0",
+  "description": "Phoneme definitions and ARPAbet/IPA/Ingglish conversion maps",
+  "type": "module",
+  "main": "./dist/index.js",
+  "module": "./dist/index.mjs",
+  "types": "./dist/index.d.ts",
+  "exports": {
+    ".": {
+      "source": "./src/index.ts",
+      "import": {
+        "types": "./dist/index.d.ts",
+        "default": "./dist/index.js"
+      },
+      "require": {
+        "types": "./dist/index.d.cts",
+        "default": "./dist/index.cjs"
+      }
+    }
+  },
+  "files": [
+    "dist"
+  ],
+  "sideEffects": false,
+  "engines": {
+    "node": ">=16"
+  },
+  "scripts": {
+    "build": "tsup",
+    "build:fast": "tsup src/index.ts --format esm",
+    "lint": "eslint --cache src",
+    "test": "vitest run --no-color",
+    "bench": "vitest bench --no-color",
+    "prepublishOnly": "npm run build"
+  },
+  "author": "Paul Tarjan",
+  "license": "MIT",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/ptarjan/ingglish.git",
+    "directory": "packages/phonemes"
+  },
+  "homepage": "https://github.com/ptarjan/ingglish#readme",
+  "bugs": {
+    "url": "https://github.com/ptarjan/ingglish/issues"
+  }
+}