PHP's str_word_count in TypeScript

How to use

Install via yarn add locutus and import: import { str_word_count } from 'locutus/php/strings/str_word_count'.

Or with CommonJS: const { str_word_count } = require('locutus/php/strings/str_word_count')

Use a bundler that supports tree-shaking so you only ship the functions you actually use. Vite, webpack, Rollup, and Parcel all handle this. For server-side use this is less of a concern.

Examples

These examples are extracted from test cases that automatically verify our functions against their native counterparts.

#	code	expected result
1	str_word_count("Hello fri3nd, you're\r\n looking good today!", 1)	['Hello', 'fri', 'nd', "you're", 'looking', 'good', 'today']
2	str_word_count("Hello fri3nd, you're\r\n looking good today!", 2)	{0: 'Hello', 6: 'fri', 10: 'nd', 14: "you're", 29: 'looking', 46: 'good', 51: 'today'}
3	str_word_count("Hello fri3nd, you're\r\n looking good today!", 1, '\u00e0\u00e1\u00e3\u00e73')	['Hello', 'fri3nd', "you're", 'looking', 'good', 'today']
4	str_word_count('hey', 2)	{0: 'hey'}

Dependencies

This function uses the following Locutus functions:

• ctype_alpha (php/ctype)

Here's what our current TypeScript equivalent to PHP's str_word_count looks like.

TSTypeScript JSJavaScript TSStandalone TS JSStandalone JS

Copy code

import { ctype_alpha as ctypeAlpha } from '../ctype/ctype_alpha.ts'

export function str_word_count(
  str: string,
  format?: 0 | 1 | 2,
  charlist?: string,
): number | string[] | { [key: number]: string } {
  //  discuss at: https://locutus.io/php/str_word_count/
  // original by: Ole Vrijenhoek
  // bugfixed by: Kevin van Zonneveld (https://kvz.io)
  // bugfixed by: Brett Zamir (https://brett-zamir.me)
  // bugfixed by: Brett Zamir (https://brett-zamir.me)
  //    input by: Bug?
  // improved by: Brett Zamir (https://brett-zamir.me)
  //   example 1: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 1)
  //   returns 1: ['Hello', 'fri', 'nd', "you're", 'looking', 'good', 'today']
  //   example 2: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 2)
  //   returns 2: {0: 'Hello', 6: 'fri', 10: 'nd', 14: "you're", 29: 'looking', 46: 'good', 51: 'today'}
  //   example 3: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 1, '\u00e0\u00e1\u00e3\u00e73')
  //   returns 3: ['Hello', 'fri3nd', "you're", 'looking', 'good', 'today']
  //   example 4: str_word_count('hey', 2)
  //   returns 4: {0: 'hey'}

  const len = str.length
  const cl = charlist?.length ?? 0
  let chr = ''
  let tmpStr = ''
  let c = ''
  const wArr: string[] = []
  let wC = 0
  const assoc: { [key: number]: string } = {}
  let aC = 0
  let reg: RegExp | null = null
  let match = false

  const _pregQuote = function (value: string): string {
    return (value + '').replace(/([\\.+*?[^\]$(){}=!<>|:])/g, '\\$1')
  }
  const _getWholeChar = function (value: string, index: number): string | false {
    // Use for rare cases of non-BMP characters
    const code = value.charCodeAt(index)
    if (code < 0xd800 || code > 0xdfff) {
      return value.charAt(index)
    }
    if (code >= 0xd800 && code <= 0xdbff) {
      // High surrogate (could change last hex to 0xDB7F to treat high private surrogates as single
      // characters)
      if (value.length <= index + 1) {
        throw new Error('High surrogate without following low surrogate')
      }
      const next = value.charCodeAt(index + 1)
      if (next < 0xdc00 || next > 0xdfff) {
        throw new Error('High surrogate without following low surrogate')
      }
      return value.charAt(index) + value.charAt(index + 1)
    }
    // Low surrogate (0xDC00 <= code && code <= 0xDFFF)
    if (index === 0) {
      throw new Error('Low surrogate without preceding high surrogate')
    }
    const prev = value.charCodeAt(index - 1)
    if (prev < 0xd800 || prev > 0xdbff) {
      // (could change last hex to 0xDB7F to treat high private surrogates as single characters)
      throw new Error('Low surrogate without preceding high surrogate')
    }
    // We can pass over low surrogates now as the second component in a pair which we have already
    // processed
    return false
  }

  if (cl && typeof charlist === 'string') {
    const firstChar = _getWholeChar(charlist, 0)
    let pattern = '^(' + _pregQuote(firstChar === false ? '' : firstChar)
    for (let i = 1; i < cl; i++) {
      const wholeChar = _getWholeChar(charlist, i)
      if (wholeChar === false) {
        continue
      }
      chr = wholeChar
      pattern += '|' + _pregQuote(chr)
    }
    pattern += ')$'
    reg = new RegExp(pattern)
  }

  for (let i = 0; i < len; i++) {
    const wholeChar = _getWholeChar(str, i)
    if (wholeChar === false) {
      continue
    }
    c = wholeChar
    // No hyphen at beginning or end unless allowed in charlist (or locale)
    // No apostrophe at beginning unless allowed in charlist (or locale)
    // @todo: Make this more readable
    match =
      ctypeAlpha(c) ||
      (reg !== null && reg.test(c)) ||
      (i !== 0 && i !== len - 1 && c === '-') ||
      (i !== 0 && c === "'")
    if (match) {
      if (tmpStr === '' && format === 2) {
        aC = i
      }
      tmpStr = tmpStr + c
    }
    if (i === len - 1 || (!match && tmpStr !== '')) {
      if (format !== 2) {
        wArr.push(tmpStr)
      } else {
        assoc[aC] = tmpStr
      }
      tmpStr = ''
      wC++
    }
  }

  if (!format) {
    return wC
  } else if (format === 1) {
    return wArr
  } else if (format === 2) {
    return assoc
  }

  throw new Error('You have supplied an incorrect format')
}

import { ctype_alpha as ctypeAlpha } from '../ctype/ctype_alpha.ts'

export function str_word_count(str, format, charlist) {
  //  discuss at: https://locutus.io/php/str_word_count/
  // original by: Ole Vrijenhoek
  // bugfixed by: Kevin van Zonneveld (https://kvz.io)
  // bugfixed by: Brett Zamir (https://brett-zamir.me)
  // bugfixed by: Brett Zamir (https://brett-zamir.me)
  //    input by: Bug?
  // improved by: Brett Zamir (https://brett-zamir.me)
  //   example 1: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 1)
  //   returns 1: ['Hello', 'fri', 'nd', "you're", 'looking', 'good', 'today']
  //   example 2: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 2)
  //   returns 2: {0: 'Hello', 6: 'fri', 10: 'nd', 14: "you're", 29: 'looking', 46: 'good', 51: 'today'}
  //   example 3: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 1, '\u00e0\u00e1\u00e3\u00e73')
  //   returns 3: ['Hello', 'fri3nd', "you're", 'looking', 'good', 'today']
  //   example 4: str_word_count('hey', 2)
  //   returns 4: {0: 'hey'}

  const len = str.length
  const cl = charlist?.length ?? 0
  let chr = ''
  let tmpStr = ''
  let c = ''
  const wArr = []
  let wC = 0
  const assoc = {}
  let aC = 0
  let reg = null
  let match = false

  const _pregQuote = function (value) {
    return (value + '').replace(/([\\.+*?[^\]$(){}=!<>|:])/g, '\\$1')
  }
  const _getWholeChar = function (value, index) {
    // Use for rare cases of non-BMP characters
    const code = value.charCodeAt(index)
    if (code < 0xd800 || code > 0xdfff) {
      return value.charAt(index)
    }
    if (code >= 0xd800 && code <= 0xdbff) {
      // High surrogate (could change last hex to 0xDB7F to treat high private surrogates as single
      // characters)
      if (value.length <= index + 1) {
        throw new Error('High surrogate without following low surrogate')
      }
      const next = value.charCodeAt(index + 1)
      if (next < 0xdc00 || next > 0xdfff) {
        throw new Error('High surrogate without following low surrogate')
      }
      return value.charAt(index) + value.charAt(index + 1)
    }
    // Low surrogate (0xDC00 <= code && code <= 0xDFFF)
    if (index === 0) {
      throw new Error('Low surrogate without preceding high surrogate')
    }
    const prev = value.charCodeAt(index - 1)
    if (prev < 0xd800 || prev > 0xdbff) {
      // (could change last hex to 0xDB7F to treat high private surrogates as single characters)
      throw new Error('Low surrogate without preceding high surrogate')
    }
    // We can pass over low surrogates now as the second component in a pair which we have already
    // processed
    return false
  }

  if (cl && typeof charlist === 'string') {
    const firstChar = _getWholeChar(charlist, 0)
    let pattern = '^(' + _pregQuote(firstChar === false ? '' : firstChar)
    for (let i = 1; i < cl; i++) {
      const wholeChar = _getWholeChar(charlist, i)
      if (wholeChar === false) {
        continue
      }
      chr = wholeChar
      pattern += '|' + _pregQuote(chr)
    }
    pattern += ')$'
    reg = new RegExp(pattern)
  }

  for (let i = 0; i < len; i++) {
    const wholeChar = _getWholeChar(str, i)
    if (wholeChar === false) {
      continue
    }
    c = wholeChar
    // No hyphen at beginning or end unless allowed in charlist (or locale)
    // No apostrophe at beginning unless allowed in charlist (or locale)
    // @todo: Make this more readable
    match =
      ctypeAlpha(c) ||
      (reg !== null && reg.test(c)) ||
      (i !== 0 && i !== len - 1 && c === '-') ||
      (i !== 0 && c === "'")
    if (match) {
      if (tmpStr === '' && format === 2) {
        aC = i
      }
      tmpStr = tmpStr + c
    }
    if (i === len - 1 || (!match && tmpStr !== '')) {
      if (format !== 2) {
        wArr.push(tmpStr)
      } else {
        assoc[aC] = tmpStr
      }
      tmpStr = ''
      wC++
    }
  }

  if (!format) {
    return wC
  } else if (format === 1) {
    return wArr
  } else if (format === 2) {
    return assoc
  }

  throw new Error('You have supplied an incorrect format')
}

// php/_helpers/_phpTypes (Locutus helper dependency)
type PhpNullish = null | undefined

type PhpInput = {} | PhpNullish

type PhpList<T = PhpInput> = T[]

type PhpAssoc<T = PhpInput> = { [key: string]: T }

type PhpArrayLike<T = PhpInput> = PhpList<T> | PhpAssoc<T>

function isPhpList<T = PhpInput>(value: PhpInput): value is PhpList<T> {
  return Array.isArray(value)
}

function isObjectLike(value: PhpInput): value is PhpArrayLike<PhpInput> {
  return typeof value === 'object' && value !== null
}

function isPhpAssocObject<T = PhpInput>(value: PhpInput): value is PhpAssoc<T> {
  return isObjectLike(value) && !isPhpList(value)
}

// php/_helpers/_phpRuntimeState (Locutus helper dependency)
interface IniEntry {
  local_value?: PhpInput
}

type LocaleEntry = PhpAssoc<PhpInput> & {
  sorting?: (left: PhpInput, right: PhpInput) => number
}

type LocaleCategoryMap = PhpAssoc<string | undefined>

interface LocutusRuntimeContainer {
  php?: PhpAssoc<PhpInput>
}

interface PhpRuntimeKnownEntryMap {
  ini: PhpAssoc<IniEntry | undefined>
  locales: PhpAssoc<LocaleEntry | undefined>
  localeCategories: LocaleCategoryMap
  pointers: PhpList<PhpInput>
  locale_default: string
  locale: string
  uniqidSeed: number
  timeoutStatus: boolean
  last_error_json: number
  strtokleftOver: string
}

type PhpRuntimeStringKey = {
  [K in keyof PhpRuntimeKnownEntryMap]: PhpRuntimeKnownEntryMap[K] extends string ? K : never
}[keyof PhpRuntimeKnownEntryMap]

interface PhpGlobalProcessLike {
  env?: PhpAssoc<string | undefined>
}

interface PhpGlobalBufferLike {
  from?: (...args: PhpInput[]) => PhpInput
}

interface PhpGlobalKnownEntryMap {
  process: PhpGlobalProcessLike
  Buffer: PhpGlobalBufferLike
}

type GlobalWithLocutus = {
  $locutus?: LocutusRuntimeContainer
  [key: string]: PhpInput
}

interface PhpRuntimeState {
  ini: PhpAssoc<IniEntry | undefined>
  locales: PhpAssoc<LocaleEntry | undefined>
  localeCategories: LocaleCategoryMap
  pointers: PhpList<PhpInput>
  locale_default: string | undefined
}

const isIniBag = (value: PhpInput): value is PhpAssoc<IniEntry | undefined> =>
  isPhpAssocObject<IniEntry | undefined>(value)

const isLocaleBag = (value: PhpInput): value is PhpAssoc<LocaleEntry | undefined> =>
  isPhpAssocObject<LocaleEntry | undefined>(value)

const isLocaleCategoryBag = (value: PhpInput): value is LocaleCategoryMap => isPhpAssocObject<string | undefined>(value)

const globalContext: GlobalWithLocutus =
  typeof window === 'object' && window !== null ? window : typeof global === 'object' && global !== null ? global : {}

const ensurePhpRuntimeObject = (): PhpAssoc<PhpInput> => {
  let locutus = globalContext.$locutus
  if (typeof locutus !== 'object' || locutus === null) {
    locutus = {}
    globalContext.$locutus = locutus
  }

  let php = locutus.php
  if (typeof php !== 'object' || php === null) {
    php = {}
    locutus.php = php
  }

  return php
}

function ensurePhpRuntimeState(): PhpRuntimeState {
  const php = ensurePhpRuntimeObject()
  const iniValue = php.ini
  const localesValue = php.locales
  const localeCategoriesValue = php.localeCategories
  const pointersValue = php.pointers

  const ini = isIniBag(iniValue) ? iniValue : {}
  const locales = isLocaleBag(localesValue) ? localesValue : {}
  const localeCategories = isLocaleCategoryBag(localeCategoriesValue) ? localeCategoriesValue : {}
  const pointers: PhpList<PhpInput> = Array.isArray(pointersValue) ? pointersValue : []

  if (iniValue !== ini) {
    php.ini = ini
  }
  if (localesValue !== locales) {
    php.locales = locales
  }
  if (localeCategoriesValue !== localeCategories) {
    php.localeCategories = localeCategories
  }
  if (pointersValue !== pointers) {
    php.pointers = pointers
  }

  const localeDefaultValue = php.locale_default
  const localeDefault = typeof localeDefaultValue === 'string' ? localeDefaultValue : undefined

  return {
    ini,
    locales,
    localeCategories,
    pointers,
    locale_default: localeDefault,
  }
}

function getPhpRuntimeEntry<TKey extends keyof PhpRuntimeKnownEntryMap>(
  key: TKey,
): PhpRuntimeKnownEntryMap[TKey] | undefined

function getPhpRuntimeEntry(key: string): PhpInput | undefined

function getPhpRuntimeEntry(key: string): PhpInput | undefined {
  const php = ensurePhpRuntimeObject()
  const value = php[key]
  return typeof value === 'undefined' ? undefined : value
}

function setPhpRuntimeEntry<TKey extends keyof PhpRuntimeKnownEntryMap>(
  key: TKey,
  value: PhpRuntimeKnownEntryMap[TKey],
): void

function setPhpRuntimeEntry(key: string, value: PhpInput): void

function setPhpRuntimeEntry(key: string, value: PhpInput): void {
  const php = ensurePhpRuntimeObject()
  php[key] = value
}

function getPhpRuntimeString(key: PhpRuntimeStringKey, fallback: string): string

function getPhpRuntimeString(key: string, fallback: string): string

function getPhpRuntimeString(key: string, fallback: string): string {
  const value = getPhpRuntimeEntry(key)
  return typeof value === 'string' ? value : fallback
}

function getPhpGlobalEntry<TKey extends keyof PhpGlobalKnownEntryMap>(
  key: TKey,
): PhpGlobalKnownEntryMap[TKey] | undefined

function getPhpGlobalEntry(key: string): PhpInput | undefined

function getPhpGlobalEntry(key: string): PhpInput | undefined {
  const value = globalContext[key]
  return typeof value === 'undefined' ? undefined : value
}

function getPhpObjectEntry(value: PhpInput, key: string): PhpInput | undefined {
  if ((typeof value !== 'object' && typeof value !== 'function') || value === null) {
    return undefined
  }

  let current: object | null = value
  while (current) {
    const descriptor = Object.getOwnPropertyDescriptor(current, key)
    if (descriptor) {
      if (typeof descriptor.get === 'function') {
        const getterValue = descriptor.get.call(value)
        return typeof getterValue === 'undefined' ? undefined : getterValue
      }
      const directValue = descriptor.value
      return typeof directValue === 'undefined' ? undefined : directValue
    }
    current = Object.getPrototypeOf(current)
  }

  return undefined
}

function getPhpLocaleEntry(category: string): LocaleEntry | undefined {
  const runtime = ensurePhpRuntimeState()
  const localeName = runtime.localeCategories[category]
  if (typeof localeName !== 'string') {
    return undefined
  }
  const localeEntry = runtime.locales[localeName]
  return isPhpAssocObject(localeEntry) ? localeEntry : undefined
}

function getPhpLocaleGroup(category: string, groupKey: string): PhpAssoc<PhpInput> | undefined {
  const localeEntry = getPhpLocaleEntry(category)
  if (!localeEntry) {
    return undefined
  }
  const group = localeEntry[groupKey]
  return isPhpAssocObject(group) ? group : undefined
}

// php/_helpers/_ctypePattern (Locutus helper dependency)
const defaultCtypePatterns: { [key: string]: RegExp } = {
  an: /^[A-Za-z\d]+$/g,
  al: /^[A-Za-z]+$/g,
  // biome-ignore lint/suspicious/noControlCharactersInRegex: intentional for LC_CTYPE control character class
  ct: /^[\u0000-\u001F\u007F]+$/g,
  dg: /^[\d]+$/g,
  gr: /^[\u0021-\u007E]+$/g,
  lw: /^[a-z]+$/g,
  pr: /^[\u0020-\u007E]+$/g,
  pu: /^[\u0021-\u002F\u003A-\u0040\u005B-\u0060\u007B-\u007E]+$/g,
  sp: /^[\f\n\r\t\v ]+$/g,
  up: /^[A-Z]+$/g,
  xd: /^[A-Fa-f\d]+$/g,
}

function getCtypePattern(key: string): RegExp | undefined {
  const ctypeGroup = getPhpLocaleGroup('LC_CTYPE', 'LC_CTYPE')
  if (!ctypeGroup) {
    const fallbackPattern = defaultCtypePatterns[key]
    return fallbackPattern ? new RegExp(fallbackPattern) : undefined
  }

  const pattern = ctypeGroup[key]
  if (pattern instanceof RegExp) {
    return new RegExp(pattern)
  }

  const fallbackPattern = defaultCtypePatterns[key]
  return fallbackPattern ? new RegExp(fallbackPattern) : undefined
}

// php/info/getenv (Locutus dependency module)
function getenv(varname: string): string | false {
  //      discuss at: https://locutus.io/php/getenv/
  // parity verified: PHP 8.3
  //     original by: Brett Zamir (https://brett-zamir.me)
  //       example 1: getenv('LC_ALL')
  //       returns 1: false

  const processValue = getPhpGlobalEntry('process')
  const hasProcessLike = typeof processValue !== 'undefined'
  if (hasProcessLike) {
    return false
  }

  if (typeof processValue !== 'object' || processValue === null) {
    return false
  }

  const envValue = getPhpObjectEntry(processValue, 'env')
  if (typeof envValue !== 'object' || envValue === null) {
    return false
  }

  const envEntry = getPhpObjectEntry(envValue, varname)
  return typeof envEntry === 'string' && envEntry.length > 0 ? envEntry : false
}

// php/strings/setlocale (Locutus dependency module)
type LocaleDefinition = {
  LC_COLLATE: (str1: string, str2: string) => number
  LC_CTYPE: Record<string, RegExp | string>
  LC_TIME: Record<string, string | string[]>
  LC_MONETARY: Record<string, string | number | number[]>
  LC_NUMERIC: Record<string, string | number[]>
  LC_MESSAGES: Record<string, string>
  nplurals: (n: number) => number
}

type LocaleInput = string | string[] | number | null

const isLocaleDefinitionMap = (value: PhpInput): value is Record<string, LocaleDefinition> =>
  typeof value === 'object' && value !== null && !Array.isArray(value)

const isLocaleCategoryMap = (value: PhpInput): value is Record<string, string> =>
  isPhpAssocObject<PhpInput>(value) &&
  typeof value.LC_COLLATE === 'string' &&
  typeof value.LC_CTYPE === 'string' &&
  typeof value.LC_MONETARY === 'string' &&
  typeof value.LC_NUMERIC === 'string' &&
  typeof value.LC_TIME === 'string' &&
  typeof value.LC_MESSAGES === 'string'

function copyValue<T>(orig: T): T

function copyValue(orig: PhpInput): PhpInput {
  if (orig instanceof RegExp) {
    return new RegExp(orig)
  }
  if (orig instanceof Date) {
    return new Date(orig)
  }
  if (Array.isArray(orig)) {
    return orig.map((item) => copyValue(item))
  }
  if (orig !== null && typeof orig === 'object') {
    const newObj: PhpAssoc<PhpInput> = {}
    for (const [key, value] of Object.entries(orig)) {
      newObj[key] = value !== null && typeof value === 'object' ? copyValue(value) : value
    }
    return newObj
  }
  return orig
}

function setlocale(category: string, locale: LocaleInput): string | false {
  //  discuss at: https://locutus.io/php/setlocale/
  // original by: Brett Zamir (https://brett-zamir.me)
  // original by: Blues (https://hacks.bluesmoon.info/strftime/strftime.js)
  // original by: YUI Library (https://developer.yahoo.com/yui/docs/YAHOO.util.DateLocale.html)
  //      note 1: Is extensible, but currently only implements locales en,
  //      note 1: en_US, en_GB, en_AU, fr, and fr_CA for LC_TIME only; C for LC_CTYPE;
  //      note 1: C and en for LC_MONETARY/LC_NUMERIC; en for LC_COLLATE
  //      note 1: Uses global: locutus to store locale info
  //      note 1: Consider using https://demo.icu-project.org/icu-bin/locexp as basis for localization (as in i18n_loc_set_default())
  //      note 2: This function tries to establish the locale via the `window` global.
  //      note 2: This feature will not work in Node and hence is Browser-only
  //   example 1: setlocale('LC_ALL', 'en_US')
  //   returns 1: 'en_US'

  const cats: string[] = []
  let i = 0

  // Function usable by a ngettext implementation (apparently not an accessible part of setlocale(),
  // but locale-specific) See https://www.gnu.org/software/gettext/manual/gettext.html#Plural-forms
  // though amended with others from https://developer.mozilla.org/En/Localization_and_Plurals (new
  // categories noted with "MDC" below, though not sure of whether there is a convention for the
  // relative order of these newer groups as far as ngettext) The function name indicates the number
  // of plural forms (nplural) Need to look into https://cldr.unicode.org/ (maybe future JavaScript);
  // Dojo has some functions (under new BSD), including JSON conversions of LDML XML from CLDR:
  // https://bugs.dojotoolkit.org/browser/dojo/trunk/cldr and docs at
  // https://api.dojotoolkit.org/jsdoc/HEAD/dojo.cldr

  // var _nplurals1 = function (n) {
  //   // e.g., Japanese
  //   return 0
  // }
  const _nplurals2a = function (n: number) {
    // e.g., English
    return n !== 1 ? 1 : 0
  }
  const _nplurals2b = function (n: number) {
    // e.g., French
    return n > 1 ? 1 : 0
  }

  const localesValue = getPhpRuntimeEntry('locales')
  let locales: Record<string, LocaleDefinition> = isLocaleDefinitionMap(localesValue) ? localesValue : {}
  if (localesValue !== locales) {
    setPhpRuntimeEntry('locales', locales)
  }

  // Reconcile Windows vs. *nix locale names?
  // Allow different priority orders of languages, esp. if implement gettext as in
  // LANGUAGE env. var.? (e.g., show German if French is not available)
  if (!locales.fr_CA?.LC_TIME?.x) {
    // Can add to the locales
    locales = {}
    setPhpRuntimeEntry('locales', locales)

    locales.en = {
      LC_COLLATE: function (str1, str2) {
        // @todo: This one taken from strcmp, but need for other locales; we don't use localeCompare
        // since its locale is not settable
        return str1 === str2 ? 0 : str1 > str2 ? 1 : -1
      },
      LC_CTYPE: {
        // Need to change any of these for English as opposed to C?
        an: /^[A-Za-z\d]+$/g,
        al: /^[A-Za-z]+$/g,
        // biome-ignore lint/suspicious/noControlCharactersInRegex: intentional for LC_CTYPE control character class
        ct: /^[\u0000-\u001F\u007F]+$/g,
        dg: /^[\d]+$/g,
        gr: /^[\u0021-\u007E]+$/g,
        lw: /^[a-z]+$/g,
        pr: /^[\u0020-\u007E]+$/g,
        pu: /^[\u0021-\u002F\u003A-\u0040\u005B-\u0060\u007B-\u007E]+$/g,
        sp: /^[\f\n\r\t\v ]+$/g,
        up: /^[A-Z]+$/g,
        xd: /^[A-Fa-f\d]+$/g,
        CODESET: 'UTF-8',
        // Used by sql_regcase
        lower: 'abcdefghijklmnopqrstuvwxyz',
        upper: 'ABCDEFGHIJKLMNOPQRSTUVWXYZ',
      },
      LC_TIME: {
        // Comments include nl_langinfo() constant equivalents and any
        // changes from Blues' implementation
        a: ['Sun', 'Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat'],
        // ABDAY_
        A: ['Sunday', 'Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday'],
        // DAY_
        b: ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'],
        // ABMON_
        B: [
          'January',
          'February',
          'March',
          'April',
          'May',
          'June',
          'July',
          'August',
          'September',
          'October',
          'November',
          'December',
        ],
        // MON_
        c: '%a %d %b %Y %r %Z',
        // D_T_FMT // changed %T to %r per results
        p: ['AM', 'PM'],
        // AM_STR/PM_STR
        P: ['am', 'pm'],
        // Not available in nl_langinfo()
        r: '%I:%M:%S %p',
        // T_FMT_AMPM (Fixed for all locales)
        x: '%m/%d/%Y',
        // D_FMT // switched order of %m and %d; changed %y to %Y (C uses %y)
        X: '%r',
        // T_FMT // changed from %T to %r  (%T is default for C, not English US)
        // Following are from nl_langinfo() or https://www.cptec.inpe.br/sx4/sx4man2/g1ab02e/strftime.4.html
        alt_digits: '',
        // e.g., ordinal
        ERA: '',
        ERA_YEAR: '',
        ERA_D_T_FMT: '',
        ERA_D_FMT: '',
        ERA_T_FMT: '',
      },
      // Assuming distinction between numeric and monetary is thus:
      // See below for C locale
      LC_MONETARY: {
        // based on Windows "english" (English_United States.1252) locale
        int_curr_symbol: 'USD',
        currency_symbol: '$',
        mon_decimal_point: '.',
        mon_thousands_sep: ',',
        mon_grouping: [3],
        // use mon_thousands_sep; "" for no grouping; additional array members
        // indicate successive group lengths after first group
        // (e.g., if to be 1,23,456, could be [3, 2])
        positive_sign: '',
        negative_sign: '-',
        int_frac_digits: 2,
        // Fractional digits only for money defaults?
        frac_digits: 2,
        p_cs_precedes: 1,
        // positive currency symbol follows value = 0; precedes value = 1
        p_sep_by_space: 0,
        // 0: no space between curr. symbol and value; 1: space sep. them unless symb.
        // and sign are adjacent then space sep. them from value; 2: space sep. sign
        // and value unless symb. and sign are adjacent then space separates
        n_cs_precedes: 1,
        // see p_cs_precedes
        n_sep_by_space: 0,
        // see p_sep_by_space
        p_sign_posn: 3,
        // 0: parentheses surround quantity and curr. symbol; 1: sign precedes them;
        // 2: sign follows them; 3: sign immed. precedes curr. symbol; 4: sign immed.
        // succeeds curr. symbol
        n_sign_posn: 0, // see p_sign_posn
      },
      LC_NUMERIC: {
        // based on Windows "english" (English_United States.1252) locale
        decimal_point: '.',
        thousands_sep: ',',
        grouping: [3], // see mon_grouping, but for non-monetary values (use thousands_sep)
      },
      LC_MESSAGES: {
        YESEXPR: '^[yY].*',
        NOEXPR: '^[nN].*',
        YESSTR: '',
        NOSTR: '',
      },
      nplurals: _nplurals2a,
    }
    locales.en_US = copyValue(locales.en)
    locales.en_US.LC_TIME.c = '%a %d %b %Y %r %Z'
    locales.en_US.LC_TIME.x = '%D'
    locales.en_US.LC_TIME.X = '%r'
    // The following are based on *nix settings
    locales.en_US.LC_MONETARY.int_curr_symbol = 'USD '
    locales.en_US.LC_MONETARY.p_sign_posn = 1
    locales.en_US.LC_MONETARY.n_sign_posn = 1
    locales.en_US.LC_MONETARY.mon_grouping = [3, 3]
    locales.en_US.LC_NUMERIC.thousands_sep = ''
    locales.en_US.LC_NUMERIC.grouping = []

    locales.en_GB = copyValue(locales.en)
    locales.en_GB.LC_TIME.r = '%l:%M:%S %P %Z'

    locales.en_AU = copyValue(locales.en_GB)
    // Assume C locale is like English (?) (We need C locale for LC_CTYPE)
    locales.C = copyValue(locales.en)
    locales.C.LC_CTYPE.CODESET = 'ANSI_X3.4-1968'
    locales.C.LC_MONETARY = {
      int_curr_symbol: '',
      currency_symbol: '',
      mon_decimal_point: '',
      mon_thousands_sep: '',
      mon_grouping: [],
      p_cs_precedes: 127,
      p_sep_by_space: 127,
      n_cs_precedes: 127,
      n_sep_by_space: 127,
      p_sign_posn: 127,
      n_sign_posn: 127,
      positive_sign: '',
      negative_sign: '',
      int_frac_digits: 127,
      frac_digits: 127,
    }
    locales.C.LC_NUMERIC = {
      decimal_point: '.',
      thousands_sep: '',
      grouping: [],
    }
    // D_T_FMT
    locales.C.LC_TIME.c = '%a %b %e %H:%M:%S %Y'
    // D_FMT
    locales.C.LC_TIME.x = '%m/%d/%y'
    // T_FMT
    locales.C.LC_TIME.X = '%H:%M:%S'
    locales.C.LC_MESSAGES.YESEXPR = '^[yY]'
    locales.C.LC_MESSAGES.NOEXPR = '^[nN]'

    locales.fr = copyValue(locales.en)
    locales.fr.nplurals = _nplurals2b
    locales.fr.LC_TIME.a = ['dim', 'lun', 'mar', 'mer', 'jeu', 'ven', 'sam']
    locales.fr.LC_TIME.A = ['dimanche', 'lundi', 'mardi', 'mercredi', 'jeudi', 'vendredi', 'samedi']
    locales.fr.LC_TIME.b = [
      'jan',
      'f\u00E9v',
      'mar',
      'avr',
      'mai',
      'jun',
      'jui',
      'ao\u00FB',
      'sep',
      'oct',
      'nov',
      'd\u00E9c',
    ]
    locales.fr.LC_TIME.B = [
      'janvier',
      'f\u00E9vrier',
      'mars',
      'avril',
      'mai',
      'juin',
      'juillet',
      'ao\u00FBt',
      'septembre',
      'octobre',
      'novembre',
      'd\u00E9cembre',
    ]
    locales.fr.LC_TIME.c = '%a %d %b %Y %T %Z'
    locales.fr.LC_TIME.p = ['', '']
    locales.fr.LC_TIME.P = ['', '']
    locales.fr.LC_TIME.x = '%d.%m.%Y'
    locales.fr.LC_TIME.X = '%T'

    locales.fr_CA = copyValue(locales.fr)
    locales.fr_CA.LC_TIME.x = '%Y-%m-%d'
  }
  let currentLocale = getPhpRuntimeString('locale', '')
  if (!currentLocale) {
    currentLocale = 'en_US'
    // Try to establish the locale via the `window` global
    if (typeof window !== 'undefined' && window.document) {
      const d = window.document
      const NS_XHTML = 'https://www.w3.org/1999/xhtml'
      const NS_XML = 'https://www.w3.org/XML/1998/namespace'
      const htmlNsElement = d.getElementsByTagNameNS ? d.getElementsByTagNameNS(NS_XHTML, 'html')[0] : undefined
      if (htmlNsElement) {
        const xmlLang = htmlNsElement.getAttributeNS(NS_XML, 'lang')
        if (xmlLang) {
          currentLocale = xmlLang
        } else {
          const htmlLang = htmlNsElement.getAttribute('lang')
          if (htmlLang) {
            currentLocale = htmlLang
          }
        }
      } else {
        const htmlElement = d.getElementsByTagName('html')[0]
        const htmlLang = htmlElement?.getAttribute('lang')
        if (htmlLang) {
          currentLocale = htmlLang
        }
      }
    }
  }
  // PHP-style
  currentLocale = currentLocale.replace('-', '_')
  // @todo: locale if declared locale hasn't been defined
  if (!(currentLocale in locales)) {
    const languageLocale = currentLocale.replace(/_[a-zA-Z]+$/, '')
    if (languageLocale in locales) {
      currentLocale = languageLocale
    }
  }
  setPhpRuntimeEntry('locale', currentLocale)

  const localeCategoriesValue = getPhpRuntimeEntry('localeCategories')
  const localeCategories: Record<string, string> = isLocaleCategoryMap(localeCategoriesValue)
    ? localeCategoriesValue
    : {
        LC_COLLATE: currentLocale,
        // for string comparison, see strcoll()
        LC_CTYPE: currentLocale,
        // for character classification and conversion, for example strtoupper()
        LC_MONETARY: currentLocale,
        // for localeconv()
        LC_NUMERIC: currentLocale,
        // for decimal separator (See also localeconv())
        LC_TIME: currentLocale,
        // for date and time formatting with strftime()
        // for system responses (available if PHP was compiled with libintl):
        LC_MESSAGES: currentLocale,
      }
  if (localeCategoriesValue !== localeCategories) {
    setPhpRuntimeEntry('localeCategories', localeCategories)
  }

  let requestedLocale: LocaleInput | false = locale

  if (requestedLocale === null || requestedLocale === '') {
    requestedLocale = getenv(category) || getenv('LANG')
  } else if (Array.isArray(requestedLocale)) {
    for (i = 0; i < requestedLocale.length; i++) {
      const candidate = requestedLocale[i]
      if (typeof candidate !== 'string') {
        if (i === requestedLocale.length - 1) {
          return false
        }
        continue
      }
      if (!(candidate in locales)) {
        if (i === requestedLocale.length - 1) {
          // none found
          return false
        }
        continue
      }
      requestedLocale = candidate
      break
    }
  }

  // Just get the locale
  if (requestedLocale === '0' || requestedLocale === 0) {
    if (category === 'LC_ALL') {
      for (const categ of Object.keys(localeCategories)) {
        // Add ".UTF-8" or allow ".@latint", etc. to the end?
        cats.push(categ + '=' + localeCategories[categ])
      }
      return cats.join(';')
    }
    return localeCategories[category] ?? false
  }

  if (typeof requestedLocale !== 'string' || !(requestedLocale in locales)) {
    // Locale not found
    return false
  }

  // Set and get locale
  if (category === 'LC_ALL') {
    for (const categ of Object.keys(localeCategories)) {
      localeCategories[categ] = requestedLocale
    }
  } else {
    localeCategories[category] = requestedLocale
  }

  return requestedLocale
}

// php/ctype/ctype_alpha (Locutus dependency module)
function ctype_alpha(text: string): boolean {
  //      discuss at: https://locutus.io/php/ctype_alpha/
  // parity verified: PHP 8.3
  //     original by: Brett Zamir (https://brett-zamir.me)
  //       example 1: ctype_alpha('Az')
  //       returns 1: true

  if (typeof text !== 'string') {
    return false
  }
  setlocale('LC_ALL', 0)

  const pattern = getCtypePattern('al')
  return pattern instanceof RegExp ? pattern.test(text) : false
}

// php/strings/str_word_count (target function module)
const ctypeAlpha = ctype_alpha;

function str_word_count(
  str: string,
  format?: 0 | 1 | 2,
  charlist?: string,
): number | string[] | { [key: number]: string } {
  //  discuss at: https://locutus.io/php/str_word_count/
  // original by: Ole Vrijenhoek
  // bugfixed by: Kevin van Zonneveld (https://kvz.io)
  // bugfixed by: Brett Zamir (https://brett-zamir.me)
  // bugfixed by: Brett Zamir (https://brett-zamir.me)
  //    input by: Bug?
  // improved by: Brett Zamir (https://brett-zamir.me)
  //   example 1: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 1)
  //   returns 1: ['Hello', 'fri', 'nd', "you're", 'looking', 'good', 'today']
  //   example 2: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 2)
  //   returns 2: {0: 'Hello', 6: 'fri', 10: 'nd', 14: "you're", 29: 'looking', 46: 'good', 51: 'today'}
  //   example 3: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 1, '\u00e0\u00e1\u00e3\u00e73')
  //   returns 3: ['Hello', 'fri3nd', "you're", 'looking', 'good', 'today']
  //   example 4: str_word_count('hey', 2)
  //   returns 4: {0: 'hey'}

  const len = str.length
  const cl = charlist?.length ?? 0
  let chr = ''
  let tmpStr = ''
  let c = ''
  const wArr: string[] = []
  let wC = 0
  const assoc: { [key: number]: string } = {}
  let aC = 0
  let reg: RegExp | null = null
  let match = false

  const _pregQuote = function (value: string): string {
    return (value + '').replace(/([\\.+*?[^\]$(){}=!<>|:])/g, '\\$1')
  }
  const _getWholeChar = function (value: string, index: number): string | false {
    // Use for rare cases of non-BMP characters
    const code = value.charCodeAt(index)
    if (code < 0xd800 || code > 0xdfff) {
      return value.charAt(index)
    }
    if (code >= 0xd800 && code <= 0xdbff) {
      // High surrogate (could change last hex to 0xDB7F to treat high private surrogates as single
      // characters)
      if (value.length <= index + 1) {
        throw new Error('High surrogate without following low surrogate')
      }
      const next = value.charCodeAt(index + 1)
      if (next < 0xdc00 || next > 0xdfff) {
        throw new Error('High surrogate without following low surrogate')
      }
      return value.charAt(index) + value.charAt(index + 1)
    }
    // Low surrogate (0xDC00 <= code && code <= 0xDFFF)
    if (index === 0) {
      throw new Error('Low surrogate without preceding high surrogate')
    }
    const prev = value.charCodeAt(index - 1)
    if (prev < 0xd800 || prev > 0xdbff) {
      // (could change last hex to 0xDB7F to treat high private surrogates as single characters)
      throw new Error('Low surrogate without preceding high surrogate')
    }
    // We can pass over low surrogates now as the second component in a pair which we have already
    // processed
    return false
  }

  if (cl && typeof charlist === 'string') {
    const firstChar = _getWholeChar(charlist, 0)
    let pattern = '^(' + _pregQuote(firstChar === false ? '' : firstChar)
    for (let i = 1; i < cl; i++) {
      const wholeChar = _getWholeChar(charlist, i)
      if (wholeChar === false) {
        continue
      }
      chr = wholeChar
      pattern += '|' + _pregQuote(chr)
    }
    pattern += ')$'
    reg = new RegExp(pattern)
  }

  for (let i = 0; i < len; i++) {
    const wholeChar = _getWholeChar(str, i)
    if (wholeChar === false) {
      continue
    }
    c = wholeChar
    // No hyphen at beginning or end unless allowed in charlist (or locale)
    // No apostrophe at beginning unless allowed in charlist (or locale)
    // @todo: Make this more readable
    match =
      ctypeAlpha(c) ||
      (reg !== null && reg.test(c)) ||
      (i !== 0 && i !== len - 1 && c === '-') ||
      (i !== 0 && c === "'")
    if (match) {
      if (tmpStr === '' && format === 2) {
        aC = i
      }
      tmpStr = tmpStr + c
    }
    if (i === len - 1 || (!match && tmpStr !== '')) {
      if (format !== 2) {
        wArr.push(tmpStr)
      } else {
        assoc[aC] = tmpStr
      }
      tmpStr = ''
      wC++
    }
  }

  if (!format) {
    return wC
  } else if (format === 1) {
    return wArr
  } else if (format === 2) {
    return assoc
  }

  throw new Error('You have supplied an incorrect format')
}

// php/_helpers/_phpTypes (Locutus helper dependency)

function isObjectLike(value) {
  return typeof value === 'object' && value !== null
}

function isPhpAssocObject(value) {
  return isObjectLike(value) && !Array.isArray(value)
}

// php/_helpers/_phpRuntimeState (Locutus helper dependency)

const globalContext =
  typeof window === 'object' && window !== null ? window : typeof global === 'object' && global !== null ? global : {}

const ensurePhpRuntimeObject = () => {
  let locutus = globalContext.$locutus
  if (typeof locutus !== 'object' || locutus === null) {
    locutus = {}
    globalContext.$locutus = locutus
  }

  let php = locutus.php
  if (typeof php !== 'object' || php === null) {
    php = {}
    locutus.php = php
  }

  return php
}

function ensurePhpRuntimeState() {
  const php = ensurePhpRuntimeObject()
  const iniValue = php.ini
  const localesValue = php.locales
  const localeCategoriesValue = php.localeCategories
  const pointersValue = php.pointers

  const ini = isPhpAssocObject(iniValue) ? iniValue : {}
  const locales = isPhpAssocObject(localesValue) ? localesValue : {}
  const localeCategories = isPhpAssocObject(localeCategoriesValue) ? localeCategoriesValue : {}
  const pointers = Array.isArray(pointersValue) ? pointersValue : []

  if (iniValue !== ini) {
    php.ini = ini
  }
  if (localesValue !== locales) {
    php.locales = locales
  }
  if (localeCategoriesValue !== localeCategories) {
    php.localeCategories = localeCategories
  }
  if (pointersValue !== pointers) {
    php.pointers = pointers
  }

  const localeDefaultValue = php.locale_default
  const localeDefault = typeof localeDefaultValue === 'string' ? localeDefaultValue : undefined

  return {
    ini,
    locales,
    localeCategories,
    pointers,
    locale_default: localeDefault,
  }
}

function getPhpRuntimeEntry(key) {
  const php = ensurePhpRuntimeObject()
  const value = php[key]
  return typeof value === 'undefined' ? undefined : value
}

function setPhpRuntimeEntry(key, value) {
  const php = ensurePhpRuntimeObject()
  php[key] = value
}

function getPhpRuntimeString(key, fallback) {
  const value = getPhpRuntimeEntry(key)
  return typeof value === 'string' ? value : fallback
}

function getPhpGlobalEntry(key) {
  const value = globalContext[key]
  return typeof value === 'undefined' ? undefined : value
}

function getPhpObjectEntry(value, key) {
  if ((typeof value !== 'object' && typeof value !== 'function') || value === null) {
    return undefined
  }

  let current = value
  while (current) {
    const descriptor = Object.getOwnPropertyDescriptor(current, key)
    if (descriptor) {
      if (typeof descriptor.get === 'function') {
        const getterValue = descriptor.get.call(value)
        return typeof getterValue === 'undefined' ? undefined : getterValue
      }
      const directValue = descriptor.value
      return typeof directValue === 'undefined' ? undefined : directValue
    }
    current = Object.getPrototypeOf(current)
  }

  return undefined
}

function getPhpLocaleEntry(category) {
  const runtime = ensurePhpRuntimeState()
  const localeName = runtime.localeCategories[category]
  if (typeof localeName !== 'string') {
    return undefined
  }
  const localeEntry = runtime.locales[localeName]
  return isPhpAssocObject(localeEntry) ? localeEntry : undefined
}

function getPhpLocaleGroup(category, groupKey) {
  const localeEntry = getPhpLocaleEntry(category)
  if (!localeEntry) {
    return undefined
  }
  const group = localeEntry[groupKey]
  return isPhpAssocObject(group) ? group : undefined
}

// php/_helpers/_ctypePattern (Locutus helper dependency)
const defaultCtypePatterns = {
  an: /^[A-Za-z\d]+$/g,
  al: /^[A-Za-z]+$/g,
  // biome-ignore lint/suspicious/noControlCharactersInRegex: intentional for LC_CTYPE control character class
  ct: /^[\u0000-\u001F\u007F]+$/g,
  dg: /^[\d]+$/g,
  gr: /^[\u0021-\u007E]+$/g,
  lw: /^[a-z]+$/g,
  pr: /^[\u0020-\u007E]+$/g,
  pu: /^[\u0021-\u002F\u003A-\u0040\u005B-\u0060\u007B-\u007E]+$/g,
  sp: /^[\f\n\r\t\v ]+$/g,
  up: /^[A-Z]+$/g,
  xd: /^[A-Fa-f\d]+$/g,
}

function getCtypePattern(key) {
  const ctypeGroup = getPhpLocaleGroup('LC_CTYPE', 'LC_CTYPE')
  if (!ctypeGroup) {
    const fallbackPattern = defaultCtypePatterns[key]
    return fallbackPattern ? new RegExp(fallbackPattern) : undefined
  }

  const pattern = ctypeGroup[key]
  if (pattern instanceof RegExp) {
    return new RegExp(pattern)
  }

  const fallbackPattern = defaultCtypePatterns[key]
  return fallbackPattern ? new RegExp(fallbackPattern) : undefined
}

// php/info/getenv (Locutus dependency module)
function getenv(varname) {
  //      discuss at: https://locutus.io/php/getenv/
  // parity verified: PHP 8.3
  //     original by: Brett Zamir (https://brett-zamir.me)
  //       example 1: getenv('LC_ALL')
  //       returns 1: false

  const processValue = getPhpGlobalEntry('process')
  const hasProcessLike = typeof processValue !== 'undefined'
  if (hasProcessLike) {
    return false
  }

  if (typeof processValue !== 'object' || processValue === null) {
    return false
  }

  const envValue = getPhpObjectEntry(processValue, 'env')
  if (typeof envValue !== 'object' || envValue === null) {
    return false
  }

  const envEntry = getPhpObjectEntry(envValue, varname)
  return typeof envEntry === 'string' && envEntry.length > 0 ? envEntry : false
}

// php/strings/setlocale (Locutus dependency module)
const isLocaleDefinitionMap = (value) => typeof value === 'object' && value !== null && !Array.isArray(value)

const isLocaleCategoryMap = (value) =>
  isPhpAssocObject(value) &&
  typeof value.LC_COLLATE === 'string' &&
  typeof value.LC_CTYPE === 'string' &&
  typeof value.LC_MONETARY === 'string' &&
  typeof value.LC_NUMERIC === 'string' &&
  typeof value.LC_TIME === 'string' &&
  typeof value.LC_MESSAGES === 'string'

function copyValue(orig) {
  if (orig instanceof RegExp) {
    return new RegExp(orig)
  }
  if (orig instanceof Date) {
    return new Date(orig)
  }
  if (Array.isArray(orig)) {
    return orig.map((item) => copyValue(item))
  }
  if (orig !== null && typeof orig === 'object') {
    const newObj = {}
    for (const [key, value] of Object.entries(orig)) {
      newObj[key] = value !== null && typeof value === 'object' ? copyValue(value) : value
    }
    return newObj
  }
  return orig
}

function setlocale(category, locale) {
  //  discuss at: https://locutus.io/php/setlocale/
  // original by: Brett Zamir (https://brett-zamir.me)
  // original by: Blues (https://hacks.bluesmoon.info/strftime/strftime.js)
  // original by: YUI Library (https://developer.yahoo.com/yui/docs/YAHOO.util.DateLocale.html)
  //      note 1: Is extensible, but currently only implements locales en,
  //      note 1: en_US, en_GB, en_AU, fr, and fr_CA for LC_TIME only; C for LC_CTYPE;
  //      note 1: C and en for LC_MONETARY/LC_NUMERIC; en for LC_COLLATE
  //      note 1: Uses global: locutus to store locale info
  //      note 1: Consider using https://demo.icu-project.org/icu-bin/locexp as basis for localization (as in i18n_loc_set_default())
  //      note 2: This function tries to establish the locale via the `window` global.
  //      note 2: This feature will not work in Node and hence is Browser-only
  //   example 1: setlocale('LC_ALL', 'en_US')
  //   returns 1: 'en_US'

  const cats = []
  let i = 0

  // Function usable by a ngettext implementation (apparently not an accessible part of setlocale(),
  // but locale-specific) See https://www.gnu.org/software/gettext/manual/gettext.html#Plural-forms
  // though amended with others from https://developer.mozilla.org/En/Localization_and_Plurals (new
  // categories noted with "MDC" below, though not sure of whether there is a convention for the
  // relative order of these newer groups as far as ngettext) The function name indicates the number
  // of plural forms (nplural) Need to look into https://cldr.unicode.org/ (maybe future JavaScript);
  // Dojo has some functions (under new BSD), including JSON conversions of LDML XML from CLDR:
  // https://bugs.dojotoolkit.org/browser/dojo/trunk/cldr and docs at
  // https://api.dojotoolkit.org/jsdoc/HEAD/dojo.cldr

  // var _nplurals1 = function (n) {
  //   // e.g., Japanese
  //   return 0
  // }
  const _nplurals2a = function (n) {
    // e.g., English
    return n !== 1 ? 1 : 0
  }
  const _nplurals2b = function (n) {
    // e.g., French
    return n > 1 ? 1 : 0
  }

  const localesValue = getPhpRuntimeEntry('locales')
  let locales = isLocaleDefinitionMap(localesValue) ? localesValue : {}
  if (localesValue !== locales) {
    setPhpRuntimeEntry('locales', locales)
  }

  // Reconcile Windows vs. *nix locale names?
  // Allow different priority orders of languages, esp. if implement gettext as in
  // LANGUAGE env. var.? (e.g., show German if French is not available)
  if (!locales.fr_CA?.LC_TIME?.x) {
    // Can add to the locales
    locales = {}
    setPhpRuntimeEntry('locales', locales)

    locales.en = {
      LC_COLLATE: function (str1, str2) {
        // @todo: This one taken from strcmp, but need for other locales; we don't use localeCompare
        // since its locale is not settable
        return str1 === str2 ? 0 : str1 > str2 ? 1 : -1
      },
      LC_CTYPE: {
        // Need to change any of these for English as opposed to C?
        an: /^[A-Za-z\d]+$/g,
        al: /^[A-Za-z]+$/g,
        // biome-ignore lint/suspicious/noControlCharactersInRegex: intentional for LC_CTYPE control character class
        ct: /^[\u0000-\u001F\u007F]+$/g,
        dg: /^[\d]+$/g,
        gr: /^[\u0021-\u007E]+$/g,
        lw: /^[a-z]+$/g,
        pr: /^[\u0020-\u007E]+$/g,
        pu: /^[\u0021-\u002F\u003A-\u0040\u005B-\u0060\u007B-\u007E]+$/g,
        sp: /^[\f\n\r\t\v ]+$/g,
        up: /^[A-Z]+$/g,
        xd: /^[A-Fa-f\d]+$/g,
        CODESET: 'UTF-8',
        // Used by sql_regcase
        lower: 'abcdefghijklmnopqrstuvwxyz',
        upper: 'ABCDEFGHIJKLMNOPQRSTUVWXYZ',
      },
      LC_TIME: {
        // Comments include nl_langinfo() constant equivalents and any
        // changes from Blues' implementation
        a: ['Sun', 'Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat'],
        // ABDAY_
        A: ['Sunday', 'Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday'],
        // DAY_
        b: ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'],
        // ABMON_
        B: [
          'January',
          'February',
          'March',
          'April',
          'May',
          'June',
          'July',
          'August',
          'September',
          'October',
          'November',
          'December',
        ],
        // MON_
        c: '%a %d %b %Y %r %Z',
        // D_T_FMT // changed %T to %r per results
        p: ['AM', 'PM'],
        // AM_STR/PM_STR
        P: ['am', 'pm'],
        // Not available in nl_langinfo()
        r: '%I:%M:%S %p',
        // T_FMT_AMPM (Fixed for all locales)
        x: '%m/%d/%Y',
        // D_FMT // switched order of %m and %d; changed %y to %Y (C uses %y)
        X: '%r',
        // T_FMT // changed from %T to %r  (%T is default for C, not English US)
        // Following are from nl_langinfo() or https://www.cptec.inpe.br/sx4/sx4man2/g1ab02e/strftime.4.html
        alt_digits: '',
        // e.g., ordinal
        ERA: '',
        ERA_YEAR: '',
        ERA_D_T_FMT: '',
        ERA_D_FMT: '',
        ERA_T_FMT: '',
      },
      // Assuming distinction between numeric and monetary is thus:
      // See below for C locale
      LC_MONETARY: {
        // based on Windows "english" (English_United States.1252) locale
        int_curr_symbol: 'USD',
        currency_symbol: '$',
        mon_decimal_point: '.',
        mon_thousands_sep: ',',
        mon_grouping: [3],
        // use mon_thousands_sep; "" for no grouping; additional array members
        // indicate successive group lengths after first group
        // (e.g., if to be 1,23,456, could be [3, 2])
        positive_sign: '',
        negative_sign: '-',
        int_frac_digits: 2,
        // Fractional digits only for money defaults?
        frac_digits: 2,
        p_cs_precedes: 1,
        // positive currency symbol follows value = 0; precedes value = 1
        p_sep_by_space: 0,
        // 0: no space between curr. symbol and value; 1: space sep. them unless symb.
        // and sign are adjacent then space sep. them from value; 2: space sep. sign
        // and value unless symb. and sign are adjacent then space separates
        n_cs_precedes: 1,
        // see p_cs_precedes
        n_sep_by_space: 0,
        // see p_sep_by_space
        p_sign_posn: 3,
        // 0: parentheses surround quantity and curr. symbol; 1: sign precedes them;
        // 2: sign follows them; 3: sign immed. precedes curr. symbol; 4: sign immed.
        // succeeds curr. symbol
        n_sign_posn: 0, // see p_sign_posn
      },
      LC_NUMERIC: {
        // based on Windows "english" (English_United States.1252) locale
        decimal_point: '.',
        thousands_sep: ',',
        grouping: [3], // see mon_grouping, but for non-monetary values (use thousands_sep)
      },
      LC_MESSAGES: {
        YESEXPR: '^[yY].*',
        NOEXPR: '^[nN].*',
        YESSTR: '',
        NOSTR: '',
      },
      nplurals: _nplurals2a,
    }
    locales.en_US = copyValue(locales.en)
    locales.en_US.LC_TIME.c = '%a %d %b %Y %r %Z'
    locales.en_US.LC_TIME.x = '%D'
    locales.en_US.LC_TIME.X = '%r'
    // The following are based on *nix settings
    locales.en_US.LC_MONETARY.int_curr_symbol = 'USD '
    locales.en_US.LC_MONETARY.p_sign_posn = 1
    locales.en_US.LC_MONETARY.n_sign_posn = 1
    locales.en_US.LC_MONETARY.mon_grouping = [3, 3]
    locales.en_US.LC_NUMERIC.thousands_sep = ''
    locales.en_US.LC_NUMERIC.grouping = []

    locales.en_GB = copyValue(locales.en)
    locales.en_GB.LC_TIME.r = '%l:%M:%S %P %Z'

    locales.en_AU = copyValue(locales.en_GB)
    // Assume C locale is like English (?) (We need C locale for LC_CTYPE)
    locales.C = copyValue(locales.en)
    locales.C.LC_CTYPE.CODESET = 'ANSI_X3.4-1968'
    locales.C.LC_MONETARY = {
      int_curr_symbol: '',
      currency_symbol: '',
      mon_decimal_point: '',
      mon_thousands_sep: '',
      mon_grouping: [],
      p_cs_precedes: 127,
      p_sep_by_space: 127,
      n_cs_precedes: 127,
      n_sep_by_space: 127,
      p_sign_posn: 127,
      n_sign_posn: 127,
      positive_sign: '',
      negative_sign: '',
      int_frac_digits: 127,
      frac_digits: 127,
    }
    locales.C.LC_NUMERIC = {
      decimal_point: '.',
      thousands_sep: '',
      grouping: [],
    }
    // D_T_FMT
    locales.C.LC_TIME.c = '%a %b %e %H:%M:%S %Y'
    // D_FMT
    locales.C.LC_TIME.x = '%m/%d/%y'
    // T_FMT
    locales.C.LC_TIME.X = '%H:%M:%S'
    locales.C.LC_MESSAGES.YESEXPR = '^[yY]'
    locales.C.LC_MESSAGES.NOEXPR = '^[nN]'

    locales.fr = copyValue(locales.en)
    locales.fr.nplurals = _nplurals2b
    locales.fr.LC_TIME.a = ['dim', 'lun', 'mar', 'mer', 'jeu', 'ven', 'sam']
    locales.fr.LC_TIME.A = ['dimanche', 'lundi', 'mardi', 'mercredi', 'jeudi', 'vendredi', 'samedi']
    locales.fr.LC_TIME.b = [
      'jan',
      'f\u00E9v',
      'mar',
      'avr',
      'mai',
      'jun',
      'jui',
      'ao\u00FB',
      'sep',
      'oct',
      'nov',
      'd\u00E9c',
    ]
    locales.fr.LC_TIME.B = [
      'janvier',
      'f\u00E9vrier',
      'mars',
      'avril',
      'mai',
      'juin',
      'juillet',
      'ao\u00FBt',
      'septembre',
      'octobre',
      'novembre',
      'd\u00E9cembre',
    ]
    locales.fr.LC_TIME.c = '%a %d %b %Y %T %Z'
    locales.fr.LC_TIME.p = ['', '']
    locales.fr.LC_TIME.P = ['', '']
    locales.fr.LC_TIME.x = '%d.%m.%Y'
    locales.fr.LC_TIME.X = '%T'

    locales.fr_CA = copyValue(locales.fr)
    locales.fr_CA.LC_TIME.x = '%Y-%m-%d'
  }
  let currentLocale = getPhpRuntimeString('locale', '')
  if (!currentLocale) {
    currentLocale = 'en_US'
    // Try to establish the locale via the `window` global
    if (typeof window !== 'undefined' && window.document) {
      const d = window.document
      const NS_XHTML = 'https://www.w3.org/1999/xhtml'
      const NS_XML = 'https://www.w3.org/XML/1998/namespace'
      const htmlNsElement = d.getElementsByTagNameNS ? d.getElementsByTagNameNS(NS_XHTML, 'html')[0] : undefined
      if (htmlNsElement) {
        const xmlLang = htmlNsElement.getAttributeNS(NS_XML, 'lang')
        if (xmlLang) {
          currentLocale = xmlLang
        } else {
          const htmlLang = htmlNsElement.getAttribute('lang')
          if (htmlLang) {
            currentLocale = htmlLang
          }
        }
      } else {
        const htmlElement = d.getElementsByTagName('html')[0]
        const htmlLang = htmlElement?.getAttribute('lang')
        if (htmlLang) {
          currentLocale = htmlLang
        }
      }
    }
  }
  // PHP-style
  currentLocale = currentLocale.replace('-', '_')
  // @todo: locale if declared locale hasn't been defined
  if (!(currentLocale in locales)) {
    const languageLocale = currentLocale.replace(/_[a-zA-Z]+$/, '')
    if (languageLocale in locales) {
      currentLocale = languageLocale
    }
  }
  setPhpRuntimeEntry('locale', currentLocale)

  const localeCategoriesValue = getPhpRuntimeEntry('localeCategories')
  const localeCategories = isLocaleCategoryMap(localeCategoriesValue)
    ? localeCategoriesValue
    : {
        LC_COLLATE: currentLocale,
        // for string comparison, see strcoll()
        LC_CTYPE: currentLocale,
        // for character classification and conversion, for example strtoupper()
        LC_MONETARY: currentLocale,
        // for localeconv()
        LC_NUMERIC: currentLocale,
        // for decimal separator (See also localeconv())
        LC_TIME: currentLocale,
        // for date and time formatting with strftime()
        // for system responses (available if PHP was compiled with libintl):
        LC_MESSAGES: currentLocale,
      }
  if (localeCategoriesValue !== localeCategories) {
    setPhpRuntimeEntry('localeCategories', localeCategories)
  }

  let requestedLocale = locale

  if (requestedLocale === null || requestedLocale === '') {
    requestedLocale = getenv(category) || getenv('LANG')
  } else if (Array.isArray(requestedLocale)) {
    for (i = 0; i < requestedLocale.length; i++) {
      const candidate = requestedLocale[i]
      if (typeof candidate !== 'string') {
        if (i === requestedLocale.length - 1) {
          return false
        }
        continue
      }
      if (!(candidate in locales)) {
        if (i === requestedLocale.length - 1) {
          // none found
          return false
        }
        continue
      }
      requestedLocale = candidate
      break
    }
  }

  // Just get the locale
  if (requestedLocale === '0' || requestedLocale === 0) {
    if (category === 'LC_ALL') {
      for (const categ of Object.keys(localeCategories)) {
        // Add ".UTF-8" or allow ".@latint", etc. to the end?
        cats.push(categ + '=' + localeCategories[categ])
      }
      return cats.join(';')
    }
    return localeCategories[category] ?? false
  }

  if (typeof requestedLocale !== 'string' || !(requestedLocale in locales)) {
    // Locale not found
    return false
  }

  // Set and get locale
  if (category === 'LC_ALL') {
    for (const categ of Object.keys(localeCategories)) {
      localeCategories[categ] = requestedLocale
    }
  } else {
    localeCategories[category] = requestedLocale
  }

  return requestedLocale
}

// php/ctype/ctype_alpha (Locutus dependency module)
function ctype_alpha(text) {
  //      discuss at: https://locutus.io/php/ctype_alpha/
  // parity verified: PHP 8.3
  //     original by: Brett Zamir (https://brett-zamir.me)
  //       example 1: ctype_alpha('Az')
  //       returns 1: true

  if (typeof text !== 'string') {
    return false
  }
  setlocale('LC_ALL', 0)

  const pattern = getCtypePattern('al')
  return pattern instanceof RegExp ? pattern.test(text) : false
}

// php/strings/str_word_count (target function module)
const ctypeAlpha = ctype_alpha

function str_word_count(str, format, charlist) {
  //  discuss at: https://locutus.io/php/str_word_count/
  // original by: Ole Vrijenhoek
  // bugfixed by: Kevin van Zonneveld (https://kvz.io)
  // bugfixed by: Brett Zamir (https://brett-zamir.me)
  // bugfixed by: Brett Zamir (https://brett-zamir.me)
  //    input by: Bug?
  // improved by: Brett Zamir (https://brett-zamir.me)
  //   example 1: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 1)
  //   returns 1: ['Hello', 'fri', 'nd', "you're", 'looking', 'good', 'today']
  //   example 2: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 2)
  //   returns 2: {0: 'Hello', 6: 'fri', 10: 'nd', 14: "you're", 29: 'looking', 46: 'good', 51: 'today'}
  //   example 3: str_word_count("Hello fri3nd, you're\r\n       looking          good today!", 1, '\u00e0\u00e1\u00e3\u00e73')
  //   returns 3: ['Hello', 'fri3nd', "you're", 'looking', 'good', 'today']
  //   example 4: str_word_count('hey', 2)
  //   returns 4: {0: 'hey'}

  const len = str.length
  const cl = charlist?.length ?? 0
  let chr = ''
  let tmpStr = ''
  let c = ''
  const wArr = []
  let wC = 0
  const assoc = {}
  let aC = 0
  let reg = null
  let match = false

  const _pregQuote = function (value) {
    return (value + '').replace(/([\\.+*?[^\]$(){}=!<>|:])/g, '\\$1')
  }
  const _getWholeChar = function (value, index) {
    // Use for rare cases of non-BMP characters
    const code = value.charCodeAt(index)
    if (code < 0xd800 || code > 0xdfff) {
      return value.charAt(index)
    }
    if (code >= 0xd800 && code <= 0xdbff) {
      // High surrogate (could change last hex to 0xDB7F to treat high private surrogates as single
      // characters)
      if (value.length <= index + 1) {
        throw new Error('High surrogate without following low surrogate')
      }
      const next = value.charCodeAt(index + 1)
      if (next < 0xdc00 || next > 0xdfff) {
        throw new Error('High surrogate without following low surrogate')
      }
      return value.charAt(index) + value.charAt(index + 1)
    }
    // Low surrogate (0xDC00 <= code && code <= 0xDFFF)
    if (index === 0) {
      throw new Error('Low surrogate without preceding high surrogate')
    }
    const prev = value.charCodeAt(index - 1)
    if (prev < 0xd800 || prev > 0xdbff) {
      // (could change last hex to 0xDB7F to treat high private surrogates as single characters)
      throw new Error('Low surrogate without preceding high surrogate')
    }
    // We can pass over low surrogates now as the second component in a pair which we have already
    // processed
    return false
  }

  if (cl && typeof charlist === 'string') {
    const firstChar = _getWholeChar(charlist, 0)
    let pattern = '^(' + _pregQuote(firstChar === false ? '' : firstChar)
    for (let i = 1; i < cl; i++) {
      const wholeChar = _getWholeChar(charlist, i)
      if (wholeChar === false) {
        continue
      }
      chr = wholeChar
      pattern += '|' + _pregQuote(chr)
    }
    pattern += ')$'
    reg = new RegExp(pattern)
  }

  for (let i = 0; i < len; i++) {
    const wholeChar = _getWholeChar(str, i)
    if (wholeChar === false) {
      continue
    }
    c = wholeChar
    // No hyphen at beginning or end unless allowed in charlist (or locale)
    // No apostrophe at beginning unless allowed in charlist (or locale)
    // @todo: Make this more readable
    match =
      ctypeAlpha(c) ||
      (reg !== null && reg.test(c)) ||
      (i !== 0 && i !== len - 1 && c === '-') ||
      (i !== 0 && c === "'")
    if (match) {
      if (tmpStr === '' && format === 2) {
        aC = i
      }
      tmpStr = tmpStr + c
    }
    if (i === len - 1 || (!match && tmpStr !== '')) {
      if (format !== 2) {
        wArr.push(tmpStr)
      } else {
        assoc[aC] = tmpStr
      }
      tmpStr = ''
      wC++
    }
  }

  if (!format) {
    return wC
  } else if (format === 1) {
    return wArr
  } else if (format === 2) {
    return assoc
  }

  throw new Error('You have supplied an incorrect format')
}

Improve this function

Locutus is a community effort following The McDonald's Theory: we ship first iterations, hoping others will improve them. If you see something that could be better, we'd love your contribution.

View on GitHub · Edit on GitHub · View Raw

---

« More PHP strings functions

Star