n2words 2.0.0 → 3.1.0

package/lib/languages/gu.js

@@ -1,132 +1,150 @@
-import { SouthAsianLanguage } from '../classes/south-asian-language.js'
-
 /**
- * Gujarati language converter.
+ * Gujarati language converter - Functional Implementation
+ *
+ * Self-contained module with its own input validation, ready for subpath exports.
  *
- * Supports:
+ * Key features:
  * - Indian numbering system (હજાર, લાખ, કરોડ)
  * - Gujarati script
+ * - 3-2-2 grouping pattern (last 3 digits, then groups of 2)
  * - Complete word forms for 0-99
+ * - Per-digit decimal reading
+ */
+
+import { parseNumericValue } from '../utils/parse-numeric.js'
+
+// ============================================================================
+// Vocabulary
+// ============================================================================
+
+const ZERO = 'શૂન્ય'
+const NEGATIVE = 'ઋણ'
+const DECIMAL_SEP = 'દશાંશ'
+const HUNDRED = 'સો'
+
+const BELOW_HUNDRED = [
+  'શૂન્ય', 'એક', 'બે', 'ત્રણ', 'ચાર', 'પાંચ', 'છ', 'સાત', 'આઠ', 'નવ',
+  'દસ', 'અગિયાર', 'બાર', 'તેર', 'ચૌદ', 'પંદર', 'સોળ', 'સત્તર', 'અઢાર', 'ઓગણીસ',
+  'વીસ', 'એકવીસ', 'બાવીસ', 'ત્રેવીસ', 'ચોવીસ', 'પચીસ', 'છવ્વીસ', 'સત્તાવીસ', 'અઠ્ઠાવીસ', 'ઓગણત્રીસ',
+  'ત્રીસ', 'એકત્રીસ', 'બત્રીસ', 'તેત્રીસ', 'ચોત્રીસ', 'પાંત્રીસ', 'છત્રીસ', 'સાડત્રીસ', 'અડત્રીસ', 'ઓગણચાલીસ',
+  'ચાલીસ', 'એકતાલીસ', 'બેતાળીસ', 'ત્રેતાળીસ', 'ચુંમાલીસ', 'પિસ્તાલીસ', 'છેતાળીસ', 'સુડતાળીસ', 'અડતાળીસ', 'ઓગણપચાસ',
+  'પચાસ', 'એકાવન', 'બાવન', 'ત્રેપન', 'ચોપન', 'પંચાવન', 'છપ્પન', 'સત્તાવન', 'અઠ્ઠાવન', 'ઓગણસાઠ',
+  'સાઠ', 'એકસઠ', 'બાસઠ', 'ત્રેસઠ', 'ચોસઠ', 'પાંસઠ', 'છાસઠ', 'સડસઠ', 'અડસઠ', 'અગણોસિત્તેર',
+  'સિત્તેર', 'એકોતેર', 'બોતેર', 'તોતેર', 'ચુમોતેર', 'પંચોતેર', 'છોતેર', 'સિત્યોતેર', 'ઇઠ્યોતેર', 'ઓગણાએંસી',
+  'એંસી', 'એક્યાસી', 'બ્યાસી', 'ત્યાસી', 'ચોર્યાસી', 'પંચાસી', 'છ્યાસી', 'સિત્યાસી', 'અઠ્યાસી', 'નેવ્યાસી',
+  'નેવું', 'એકાણું', 'બાણું', 'ત્રાણું', 'ચોરાણું', 'પંચાણું', 'છન્નું', 'સત્તાણું', 'અઠ્ઠાણું', 'નવ્વાણું'
+]
+
+// Scale words: index 0 = units (empty), 1 = thousand, 2 = lakh, 3 = crore, etc.
+const SCALE_WORDS = ['', 'હજાર', 'લાખ', 'કરોડ', 'અબજ', 'ખરબ', 'નીલ', 'પદ્મ', 'શંખ']
+
+// ============================================================================
+// Segment Building
+// ============================================================================
+
+/**
+ * Builds words for a 0-999 segment.
  */
-export class Gujarati extends SouthAsianLanguage {
-  negativeWord = 'ઋણ'
-  decimalSeparatorWord = 'દશાંશ'
-  zeroWord = 'શૂન્ય'
-  hundredWord = 'સો'
-  usePerDigitDecimals = true
-
-  belowHundredWords = [
-    'શૂન્ય',
-    'એક',
-    'બે',
-    'ત્રણ',
-    'ચાર',
-    'પાંચ',
-    'છ',
-    'સાત',
-    'આઠ',
-    'નવ',
-    'દસ',
-    'અગિયાર',
-    'બાર',
-    'તેર',
-    'ચૌદ',
-    'પંદર',
-    'સોળ',
-    'સત્તર',
-    'અઢાર',
-    'ઓગણીસ',
-    'વીસ',
-    'એકવીસ',
-    'બાવીસ',
-    'ત્રેવીસ',
-    'ચોવીસ',
-    'પચીસ',
-    'છવ્વીસ',
-    'સત્તાવીસ',
-    'અઠ્ઠાવીસ',
-    'ઓગણત્રીસ',
-    'ત્રીસ',
-    'એકત્રીસ',
-    'બત્રીસ',
-    'તેત્રીસ',
-    'ચોત્રીસ',
-    'પાંત્રીસ',
-    'છત્રીસ',
-    'સાડત્રીસ',
-    'અડત્રીસ',
-    'ઓગણચાલીસ',
-    'ચાલીસ',
-    'એકતાલીસ',
-    'બેતાળીસ',
-    'ત્રેતાળીસ',
-    'ચુંમાલીસ',
-    'પિસ્તાલીસ',
-    'છેતાળીસ',
-    'સુડતાળીસ',
-    'અડતાળીસ',
-    'ઓગણપચાસ',
-    'પચાસ',
-    'એકાવન',
-    'બાવન',
-    'ત્રેપન',
-    'ચોપન',
-    'પંચાવન',
-    'છપ્પન',
-    'સત્તાવન',
-    'અઠ્ઠાવન',
-    'ઓગણસાઠ',
-    'સાઠ',
-    'એકસઠ',
-    'બાસઠ',
-    'ત્રેસઠ',
-    'ચોસઠ',
-    'પાંસઠ',
-    'છાસઠ',
-    'સડસઠ',
-    'અડસઠ',
-    'અગણોસિત્તેર',
-    'સિત્તેર',
-    'એકોતેર',
-    'બોતેર',
-    'તોતેર',
-    'ચુમોતેર',
-    'પંચોતેર',
-    'છોતેર',
-    'સિત્યોતેર',
-    'ઇઠ્યોતેર',
-    'ઓગણાએંસી',
-    'એંસી',
-    'એક્યાસી',
-    'બ્યાસી',
-    'ત્યાસી',
-    'ચોર્યાસી',
-    'પંચાસી',
-    'છ્યાસી',
-    'સિત્યાસી',
-    'અઠ્યાસી',
-    'નેવ્યાસી',
-    'નેવું',
-    'એકાણું',
-    'બાણું',
-    'ત્રાણું',
-    'ચોરાણું',
-    'પંચાણું',
-    'છન્નું',
-    'સત્તાણું',
-    'અઠ્ઠાણું',
-    'નવ્વાણું'
-  ]
-
-  scaleWords = [
-    '',
-    'હજાર',
-    'લાખ',
-    'કરોડ',
-    'અબજ',
-    'ખરબ',
-    'નીલ',
-    'પદ્મ',
-    'શંખ'
-  ]
+function buildSegment (n) {
+  if (n === 0) return ''
+  if (n < 100) return BELOW_HUNDRED[n]
+
+  const hundreds = Math.trunc(n / 100)
+  const remainder = n % 100
+
+  if (remainder === 0) {
+    return BELOW_HUNDRED[hundreds] + ' ' + HUNDRED
+  }
+  return BELOW_HUNDRED[hundreds] + ' ' + HUNDRED + ' ' + BELOW_HUNDRED[remainder]
 }
+
+// ============================================================================
+// Conversion Functions
+// ============================================================================
+
+/**
+ * Converts a non-negative integer to Gujarati words.
+ *
+ * Uses BigInt modulo for segment extraction (faster than string slicing).
+ * South Asian 3-2-2 grouping: first 3 digits, then groups of 2.
+ *
+ * @param {bigint} n - Non-negative integer to convert
+ * @returns {string} Gujarati words
+ */
+function integerToWords (n) {
+  if (n === 0n) return ZERO
+
+  // Fast path: numbers < 1000 (direct lookup)
+  if (n < 1000n) {
+    return buildSegment(Number(n))
+  }
+
+  // Extract segments using BigInt modulo
+  const segments = []
+  segments.push(Number(n % 1000n))
+  let temp = n / 1000n
+
+  while (temp > 0n) {
+    segments.push(Number(temp % 100n))
+    temp = temp / 100n
+  }
+
+  // Build result string (process from most-significant to least)
+  const words = []
+  for (let i = segments.length - 1; i >= 0; i--) {
+    const segment = segments[i]
+    if (segment === 0) continue
+
+    if (i === 0) {
+      words.push(buildSegment(segment))
+    } else {
+      words.push(BELOW_HUNDRED[segment])
+    }
+
+    if (i > 0 && SCALE_WORDS[i]) {
+      words.push(SCALE_WORDS[i])
+    }
+  }
+
+  return words.join(' ')
+}
+
+function decimalPartToWords (decimalPart) {
+  // Per-digit decimal reading
+  const digits = []
+  for (const char of decimalPart) {
+    const d = parseInt(char, 10)
+    digits.push(d === 0 ? ZERO : BELOW_HUNDRED[d])
+  }
+  return digits.join(' ')
+}
+
+/**
+ * Converts a numeric value to Gujarati words.
+ *
+ * @param {number | string | bigint} value - The numeric value to convert
+ * @returns {string} The number in Gujarati words
+ */
+function toWords (value) {
+  const { isNegative, integerPart, decimalPart } = parseNumericValue(value)
+
+  let result = ''
+
+  if (isNegative) {
+    result = NEGATIVE + ' '
+  }
+
+  result += integerToWords(integerPart)
+
+  if (decimalPart) {
+    result += ' ' + DECIMAL_SEP + ' ' + decimalPartToWords(decimalPart)
+  }
+
+  return result
+}
+
+// ============================================================================
+// Exports
+// ============================================================================
+
+export { toWords }

package/lib/languages/ha.d.ts

@@ -0,0 +1,7 @@
+/**
+ * Converts a numeric value to Hausa words.
+ *
+ * @param {number | string | bigint} value - The numeric value to convert
+ * @returns {string} The number in Hausa words
+ */
+export function toWords(value: number | string | bigint): string;

package/lib/languages/ha.js

@@ -0,0 +1,225 @@
+/**
+ * Hausa language converter - Functional Implementation
+ *
+ * Self-contained module with its own input validation, ready for subpath exports.
+ *
+ * Key features:
+ * - Authentic Boko orthography with ɗ (hooked d) and ' (glottal stop)
+ * - Teens with "sha" prefix (sha ɗaya = 11)
+ * - Compound numbers with "da" connector (ashirin da ɗaya = 21)
+ * - Arabic loanwords for tens (ashirin, talatin, arba'in, etc.)
+ * - Reversed multiplier order: "biyu ɗari" (200), "biyu dubu" (2000)
+ * - Implicit one before ɗari and dubu
+ * - Per-digit decimal reading
+ */
+
+import { parseNumericValue } from '../utils/parse-numeric.js'
+
+// ============================================================================
+// Vocabulary
+// ============================================================================
+
+const ONES = ['', 'ɗaya', 'biyu', 'uku', 'huɗu', 'biyar', 'shida', 'bakwai', 'takwas', 'tara']
+const TEENS = ['goma', 'sha ɗaya', 'sha biyu', 'sha uku', 'sha huɗu', 'sha biyar', 'sha shida', 'sha bakwai', 'sha takwas', 'sha tara']
+// Arabic loanwords for tens
+const TENS = ['', '', 'ashirin', 'talatin', "arba'in", 'hamsin', 'sittin', "saba'in", 'tamanin', "tis'in"]
+
+const HUNDRED = 'ɗari'
+const THOUSAND = 'dubu'
+
+const ZERO = 'sifiri'
+const NEGATIVE = 'babu'
+const DECIMAL_SEP = 'digo'
+
+// Short scale
+const SCALE_WORDS = ['', THOUSAND, 'miliyan', 'biliyan']
+
+// ============================================================================
+// Precomputed Lookup Table
+// ============================================================================
+
+/**
+ * Build segment for 0-999 with Hausa patterns.
+ * Hausa uses reversed order for hundreds: "biyu ɗari" (200)
+ * And "da" connector for ones: "ashirin da ɗaya" (21)
+ */
+function buildSegment (n) {
+  if (n === 0) return ''
+
+  const ones = n % 10
+  const tensDigit = Math.floor(n / 10) % 10
+  const hundredsDigit = Math.floor(n / 100)
+
+  const parts = []
+
+  // Hundreds: implicit one, or "biyu ɗari" (reversed order)
+  if (hundredsDigit > 0) {
+    if (hundredsDigit === 1) {
+      parts.push(HUNDRED)
+    } else {
+      // Reversed: multiplier + hundredWord
+      parts.push(ONES[hundredsDigit] + ' ' + HUNDRED)
+    }
+  }
+
+  // Tens and ones
+  const tensOnes = n % 100
+
+  if (tensOnes === 0) {
+    // Just hundreds
+  } else if (tensOnes < 10) {
+    // Single digit - with "da" connector if after hundreds
+    if (hundredsDigit > 0) {
+      parts.push('da ' + ONES[ones])
+    } else {
+      parts.push(ONES[ones])
+    }
+  } else if (tensOnes < 20) {
+    // Teens (10-19): "sha X"
+    parts.push(TEENS[ones])
+  } else if (ones === 0) {
+    // Even tens (20, 30, 40, etc.)
+    parts.push(TENS[tensDigit])
+  } else {
+    // Tens + ones with "da" connector
+    parts.push(TENS[tensDigit] + ' da ' + ONES[ones])
+  }
+
+  return parts.join(' ')
+}
+
+// ============================================================================
+// Conversion Functions
+// ============================================================================
+
+function integerToWords (n) {
+  if (n === 0n) return ZERO
+
+  if (n < 1000n) {
+    return buildSegment(Number(n))
+  }
+
+  return buildLargeNumberWords(n)
+}
+
+/**
+ * Checks if a word is a single digit (1-9).
+ */
+function isSingleDigit (word) {
+  return ONES.slice(1).includes(word)
+}
+
+function buildLargeNumberWords (n) {
+  const numStr = n.toString()
+  const len = numStr.length
+
+  const segments = []
+  const segmentSize = 3
+
+  const remainderLen = len % segmentSize
+  let pos = 0
+  if (remainderLen > 0) {
+    segments.push(Number(numStr.slice(0, remainderLen)))
+    pos = remainderLen
+  }
+  while (pos < len) {
+    segments.push(Number(numStr.slice(pos, pos + segmentSize)))
+    pos += segmentSize
+  }
+
+  // Build raw parts (segment words and scale words)
+  const rawParts = []
+  let scaleIndex = segments.length - 1
+
+  for (let i = 0; i < segments.length; i++) {
+    const segment = segments[i]
+
+    if (segment !== 0) {
+      const scaleWord = SCALE_WORDS[scaleIndex] || ''
+
+      if (scaleIndex === 0) {
+        rawParts.push(buildSegment(segment))
+      } else {
+        rawParts.push(buildSegment(segment))
+        rawParts.push(scaleWord)
+      }
+    }
+
+    scaleIndex--
+  }
+
+  // Filter out implicit "ɗaya" before ɗari or dubu
+  const filtered = []
+  for (let i = 0; i < rawParts.length; i++) {
+    const part = rawParts[i]
+    const nextPart = rawParts[i + 1]
+
+    // Skip "ɗaya" before ɗari or dubu (implicit one)
+    if (part === 'ɗaya' && nextPart && (nextPart === HUNDRED || nextPart === THOUSAND)) {
+      continue
+    }
+
+    filtered.push(part)
+  }
+
+  // Join with correct separators
+  const result = []
+  for (let i = 0; i < filtered.length; i++) {
+    const part = filtered[i]
+    const prevPart = i > 0 ? filtered[i - 1] : null
+
+    // Determine if we need "da" connector
+    // Use "da" when current is a single digit following a scale word
+    if (prevPart && isSingleDigit(part) &&
+        (prevPart === THOUSAND || prevPart === HUNDRED ||
+         SCALE_WORDS.includes(prevPart))) {
+      result.push(' da ')
+    } else if (i > 0) {
+      result.push(' ')
+    }
+
+    result.push(part)
+  }
+
+  return result.join('')
+}
+
+function decimalPartToWords (decimalPart) {
+  // Per-digit decimal reading
+  const digits = []
+  for (const char of decimalPart) {
+    const d = parseInt(char, 10)
+    digits.push(d === 0 ? ZERO : ONES[d])
+  }
+  return digits.join(' ')
+}
+
+/**
+ * Converts a numeric value to Hausa words.
+ *
+ * @param {number | string | bigint} value - The numeric value to convert
+ * @returns {string} The number in Hausa words
+ */
+function toWords (value) {
+  const { isNegative, integerPart, decimalPart } = parseNumericValue(value)
+
+  let result = ''
+
+  if (isNegative) {
+    result = NEGATIVE + ' '
+  }
+
+  result += integerToWords(integerPart)
+
+  if (decimalPart) {
+    result += ' ' + DECIMAL_SEP + ' ' + decimalPartToWords(decimalPart)
+  }
+
+  return result
+}
+
+// ============================================================================
+// Exports
+// ============================================================================
+
+export { toWords }

package/lib/languages/hbo.d.ts

@@ -1,113 +1,13 @@
 /**
- * Biblical Hebrew language converter.
+ * Converts a numeric value to Biblical Hebrew words.
  *
- * Supports:
- * - Three-form pluralization (one/few/many)
- * - Gender agreement (masculine/feminine)
- * - Optional "ו" (and) conjunction control
+ * @param {number | string | bigint} value - The numeric value to convert
+ * @param {Object} [options] - Optional configuration
+ * @param {('masculine'|'feminine')} [options.gender='masculine'] - Grammatical gender
+ * @param {string} [options.andWord] - Custom conjunction word
+ * @returns {string} The number in Biblical Hebrew words
  */
-export class BiblicalHebrew extends SlavicLanguage {
-    constructor(options?: {});
-    onesWords: {
-        1: string;
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-        7: string;
-        8: string;
-        9: string;
-    };
-    teensWords: {
-        0: string;
-        1: string;
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-        7: string;
-        8: string;
-        9: string;
-    };
-    twentiesWords: {
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-        7: string;
-        8: string;
-        9: string;
-    };
-    hundredsWords: {
-        1: string;
-        2: string;
-        3: string;
-    };
-    pluralForms: {
-        1: string;
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-        7: string;
-        8: string;
-        9: string;
-    };
-    femininePluralForms: {
-        1: string;
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-        7: string;
-        8: string;
-        9: string;
-    };
-    scale: {
-        1: string;
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-    };
-    scalePlural: {
-        1: string;
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-    };
-    onesFeminineWords: {
-        1: string;
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-        7: string;
-        8: string;
-        9: string;
-    };
-    teensFeminineWords: {
-        0: string;
-        1: string;
-        2: string;
-        3: string;
-        4: string;
-        5: string;
-        6: string;
-        7: string;
-        8: string;
-        9: string;
-    };
-    /** Converts integer part with Biblical Hebrew three-form pluralization. */
-    integerToWords(integerPart: any): string;
-}
-import { SlavicLanguage } from '../classes/slavic-language.js';
+export function toWords(value: number | string | bigint, options?: {
+    gender?: "masculine" | "feminine" | undefined;
+    andWord?: string | undefined;
+}): string;