n2words 2.0.0 → 3.1.0

package/lib/languages/sr-Latn.js

@@ -1,113 +1,221 @@
-import { SlavicLanguage } from '../classes/slavic-language.js'
-
 /**
- * Serbian Latin language converter.
+ * Serbian Latin language converter - Functional Implementation
+ *
+ * Self-contained module with its own input validation, ready for subpath exports.
  *
- * Supports:
+ * Key features:
  * - Three-form pluralization (one/few/many)
- * - Gender agreement (jedan/jedna, dva/dve)
- * - Latin script representation
+ * - Gender: thousands are feminine, millions+ are masculine
+ * - Irregular hundreds (dvesta, trista, etc.)
+ * - Long scale naming with -ard forms
+ * - Latin script
  */
-export class SerbianLatin extends SlavicLanguage {
-  negativeWord = 'minus'
-  decimalSeparatorWord = 'zapeta'
-  zeroWord = 'nula'
-
-  onesWords = {
-    1: 'jedan',
-    2: 'dva',
-    3: 'tri',
-    4: 'četiri',
-    5: 'pet',
-    6: 'šest',
-    7: 'sedam',
-    8: 'osam',
-    9: 'devet'
+
+import { parseNumericValue } from '../utils/parse-numeric.js'
+import { validateOptions } from '../utils/validate-options.js'
+
+// ============================================================================
+// Vocabulary
+// ============================================================================
+
+const ONES_MASC = ['', 'jedan', 'dva', 'tri', 'četiri', 'pet', 'šest', 'sedam', 'osam', 'devet']
+const ONES_FEM = ['', 'jedna', 'dve', 'tri', 'četiri', 'pet', 'šest', 'sedam', 'osam', 'devet']
+const TEENS = ['deset', 'jedanaest', 'dvanaest', 'trinaest', 'četrnaest', 'petnaest', 'šesnaest', 'sedamnaest', 'osamnaest', 'devetnaest']
+const TENS = ['', '', 'dvadeset', 'trideset', 'četrdeset', 'pedeset', 'šezdeset', 'sedamdeset', 'osamdeset', 'devedeset']
+const HUNDREDS = ['', 'sto', 'dvesta', 'trista', 'četiristo', 'petsto', 'šesto', 'sedamsto', 'osamsto', 'devetsto']
+
+const ZERO = 'nula'
+const NEGATIVE = 'minus'
+const DECIMAL_SEP = 'zapeta'
+
+// Scale words: [singular, few, many]
+const SCALE_FORMS = [
+  ['hiljada', 'hiljade', 'hiljada'],
+  ['milion', 'miliona', 'miliona'],
+  ['milijarda', 'milijarde', 'milijarda'],
+  ['bilion', 'biliona', 'biliona'],
+  ['bilijarda', 'bilijarde', 'bilijarda'],
+  ['trilion', 'triliona', 'triliona'],
+  ['trilijarda', 'trilijarde', 'trilijarda'],
+  ['kvadrilion', 'kvadriliona', 'kvadriliona'],
+  ['kvadrilijarda', 'kvadrilijarde', 'kvadrilijarda']
+]
+
+// ============================================================================
+// Segment Building
+// ============================================================================
+
+function pluralize (n, forms) {
+  const num = typeof n === 'bigint' ? Number(n) : n
+  const lastDigit = num % 10
+  const lastTwoDigits = num % 100
+
+  if (lastTwoDigits >= 11 && lastTwoDigits <= 19) {
+    return forms[2]
   }
 
-  onesFeminineWords = {
-    1: 'jedna',
-    2: 'dve',
-    3: 'tri',
-    4: 'četiri',
-    5: 'pet',
-    6: 'šest',
-    7: 'sedam',
-    8: 'osam',
-    9: 'devet'
+  if (lastDigit === 1) return forms[0]
+  if (lastDigit >= 2 && lastDigit <= 4) return forms[1]
+  return forms[2]
+}
+
+function buildSegmentMasc (n) {
+  if (n === 0) return ''
+
+  const onesDigit = n % 10
+  const tensDigit = Math.floor(n / 10) % 10
+  const hundredsDigit = Math.floor(n / 100)
+
+  const parts = []
+
+  if (hundredsDigit > 0) {
+    parts.push(HUNDREDS[hundredsDigit])
+  }
+
+  if (tensDigit > 1) {
+    parts.push(TENS[tensDigit])
   }
 
-  teensWords = {
-    0: 'deset',
-    1: 'jedanaest',
-    2: 'dvanaest',
-    3: 'trinaest',
-    4: 'četrnaest',
-    5: 'petnaest',
-    6: 'šesnaest',
-    7: 'sedamnaest',
-    8: 'osamnaest',
-    9: 'devetnaest'
+  if (tensDigit === 1) {
+    parts.push(TEENS[onesDigit])
+  } else if (onesDigit > 0) {
+    parts.push(ONES_MASC[onesDigit])
   }
 
-  twentiesWords = {
-    2: 'dvadeset',
-    3: 'trideset',
-    4: 'četrdeset',
-    5: 'pedeset',
-    6: 'šezdeset',
-    7: 'sedamdeset',
-    8: 'osamdeset',
-    9: 'devedeset'
+  return parts.join(' ')
+}
+
+function buildSegmentFem (n) {
+  if (n === 0) return ''
+
+  const onesDigit = n % 10
+  const tensDigit = Math.floor(n / 10) % 10
+  const hundredsDigit = Math.floor(n / 100)
+
+  const parts = []
+
+  if (hundredsDigit > 0) {
+    parts.push(HUNDREDS[hundredsDigit])
   }
 
-  hundredsWords = {
-    1: 'sto',
-    2: 'dvesta',
-    3: 'trista',
-    4: 'četiristo',
-    5: 'petsto',
-    6: 'šesto',
-    7: 'sedamsto',
-    8: 'osamsto',
-    9: 'devetsto'
+  if (tensDigit > 1) {
+    parts.push(TENS[tensDigit])
   }
 
-  pluralForms = {
-    1: ['hiljada', 'hiljade', 'hiljada'], // 10 ^ 3
-    2: ['milion', 'miliona', 'miliona'], // 10 ^ 6
-    3: ['milijarda', 'milijarde', 'milijarda'], // 10 ^ 9
-    4: ['bilion', 'biliona', 'biliona'], // 10 ^ 12
-    5: ['bilijarda', 'bilijarde', 'bilijarda'], // 10 ^ 15
-    6: ['trilion', 'triliona', 'triliona'], // 10 ^ 18
-    7: ['trilijarda', 'trilijarde', 'trilijarda'], // 10 ^ 21
-    8: ['kvadrilion', 'kvadriliona', 'kvadriliona'], // 10 ^ 24
-    9: ['kvadrilijarda', 'kvadrilijarde', 'kvadrilijarda'], // 10 ^ 27
-    10: ['kvintilion', 'kvintiliona', 'kvintiliona'] // 10 ^ 30
+  if (tensDigit === 1) {
+    parts.push(TEENS[onesDigit])
+  } else if (onesDigit > 0) {
+    parts.push(ONES_FEM[onesDigit])
   }
 
-  /**
-   * Maps segment indices to whether they are grammatically feminine.
-   * In Serbian, thousands (index 1) are feminine, others are masculine.
-   * @type {Object.<number, boolean>}
-   */
-  scaleGenders = {
-    1: true // thousands are feminine (others default to false)
+  return parts.join(' ')
+}
+
+// ============================================================================
+// Conversion Functions
+// ============================================================================
+
+function integerToWords (n, options = {}) {
+  if (n === 0n) return ZERO
+
+  if (n < 1000n) {
+    return options.gender === 'feminine' ? buildSegmentFem(Number(n)) : buildSegmentMasc(Number(n))
   }
 
-  /** Selects Serbian plural form: 1 = singular, 2-4 = few, else = many. */
-  pluralize (n, forms) {
-    const lastDigit = n % 10n
-    const lastTwoDigits = n % 100n
+  return buildLargeNumberWords(n, options)
+}
 
-    if ((lastTwoDigits < 10n || lastTwoDigits > 20n) && lastDigit === 1n) {
-      return forms[0]
-    }
+function buildLargeNumberWords (n, options) {
+  const numStr = n.toString()
+  const len = numStr.length
 
-    if ((lastTwoDigits < 10n || lastTwoDigits > 20n) && lastDigit > 1n && lastDigit < 5n) {
-      return forms[1]
+  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
+  }
+
+  const parts = []
+  let scaleIndex = segments.length - 1
+
+  for (let i = 0; i < segments.length; i++) {
+    const segment = segments[i]
+
+    if (segment !== 0) {
+      if (scaleIndex === 0) {
+        parts.push(options.gender === 'feminine' ? buildSegmentFem(segment) : buildSegmentMasc(segment))
+      } else {
+        const scaleForms = SCALE_FORMS[scaleIndex - 1]
+        const scaleWord = pluralize(segment, scaleForms)
+        // Thousands (scaleIndex=1) are feminine, others masculine
+        const isFeminine = scaleIndex === 1
+        const segmentWord = isFeminine ? buildSegmentFem(segment) : buildSegmentMasc(segment)
+        parts.push(segmentWord + ' ' + scaleWord)
+      }
     }
 
-    return forms[2]
+    scaleIndex--
   }
+
+  return parts.join(' ')
+}
+
+function decimalPartToWords (decimalPart, options) {
+  let result = ''
+  let i = 0
+
+  while (i < decimalPart.length && decimalPart[i] === '0') {
+    if (result) result += ' '
+    result += ZERO
+    i++
+  }
+
+  const remainder = decimalPart.slice(i)
+  if (remainder) {
+    if (result) result += ' '
+    result += integerToWords(BigInt(remainder), options)
+  }
+
+  return result
 }
+
+/**
+ * Converts a numeric value to Serbian (Latin) words.
+ *
+ * @param {number | string | bigint} value - The numeric value to convert
+ * @param {Object} [options] - Optional configuration
+ * @param {('masculine'|'feminine')} [options.gender='masculine'] - Grammatical gender
+ * @returns {string} The number in Serbian Latin words
+ */
+function toWords (value, options) {
+  options = validateOptions(options)
+  const { isNegative, integerPart, decimalPart } = parseNumericValue(value)
+
+  let result = ''
+
+  if (isNegative) {
+    result = NEGATIVE + ' '
+  }
+
+  result += integerToWords(integerPart, options)
+
+  if (decimalPart) {
+    result += ' ' + DECIMAL_SEP + ' ' + decimalPartToWords(decimalPart, options)
+  }
+
+  return result
+}
+
+// ============================================================================
+// Exports
+// ============================================================================
+
+export { toWords }

package/lib/languages/sv.d.ts

@@ -1,14 +1,14 @@
 /**
- * Swedish language converter.
+ * Converts a numeric value to Swedish words.
  *
- * Supports:
- * - Hyphenation for compound tens (tjugo-tre)
- * - "och" (and) after hundreds
- * - Space-separated larger composites
+ * @param {number | string | bigint} value - The numeric value to convert
+ * @returns {string} The number in Swedish words
+ * @throws {TypeError} If value is not a valid numeric type
+ * @throws {Error} If value is not a valid number format
+ *
+ * @example
+ * toWords(42)       // 'fyrtio-två'
+ * toWords(101)      // 'hundra och ett'
+ * toWords(1000000)  // 'en miljon'
  */
-export class Swedish extends GreedyScaleLanguage {
-    scaleWords: (string | bigint)[][];
-    /** Combines two word-sets according to Swedish grammar rules. */
-    combineWordSets(preceding: any, following: any): any;
-}
-import { GreedyScaleLanguage } from '../classes/greedy-scale-language.js';
+export function toWords(value: number | string | bigint): string;