PHP's metaphone in JavaScript

How to use

You you can install via yarn add locutus and require this function via const metaphone = require('locutus/php/strings/metaphone').

It is important to use a bundler that supports tree-shaking so that you only ship the functions that you actually use to your browser, instead of all of Locutus, which is massive. Examples are: Parcel, webpack, or rollup.js. For server-side use this is typically less of a concern.

Examples

Please note that these examples are distilled from test cases that automatically verify our functions still work correctly. This could explain some quirky ones.

#codeexpected result
1metaphone('Gnu')'N'
2metaphone('bigger')'BKR'
3metaphone('accuracy')'AKKRS'
4metaphone('batch batcher')'BXBXR'

Here’s what our current JavaScript equivalent to PHP's metaphone looks like.

module.exports = function metaphone(word, maxPhonemes) {
// discuss at: https://locutus.io/php/metaphone/
// original by: Greg Frazier
// improved by: Brett Zamir (https://brett-zamir.me)
// improved by: Rafał Kukawski (https://blog.kukawski.pl)
// example 1: metaphone('Gnu')
// returns 1: 'N'
// example 2: metaphone('bigger')
// returns 2: 'BKR'
// example 3: metaphone('accuracy')
// returns 3: 'AKKRS'
// example 4: metaphone('batch batcher')
// returns 4: 'BXBXR'

const type = typeof word

if (type === 'undefined' || (type === 'object' && word !== null)) {
// weird!
return null
}

// infinity and NaN values are treated as strings
if (type === 'number') {
if (isNaN(word)) {
word = 'NAN'
} else if (!isFinite(word)) {
word = 'INF'
}
}

if (maxPhonemes < 0) {
return false
}

maxPhonemes = Math.floor(+maxPhonemes) || 0

// alpha depends on locale, so this var might need an update
// or should be turned into a regex
// for now assuming pure a-z
const alpha = 'ABCDEFGHIJKLMNOPQRSTUVWXYZ'
const vowel = 'AEIOU'
const soft = 'EIY'
const leadingNonAlpha = new RegExp('^[^' + alpha + ']+')

word = typeof word === 'string' ? word : ''
word = word.toUpperCase().replace(leadingNonAlpha, '')

if (!word) {
return ''
}

const is = function (p, c) {
return c !== '' && p.indexOf(c) !== -1
}

let i = 0
let cc = word.charAt(0) // current char. Short name because it's used all over the function
let nc = word.charAt(1) // next char
let nnc // after next char
let pc // previous char
const l = word.length
let meta = ''
// traditional is an internal param that could be exposed for now let it be a local var
const traditional = true

switch (cc) {
case 'A':
meta += nc === 'E' ? nc : cc
i += 1
break
case 'G':
case 'K':
case 'P':
if (nc === 'N') {
meta += nc
i += 2
}
break
case 'W':
if (nc === 'R') {
meta += nc
i += 2
} else if (nc === 'H' || is(vowel, nc)) {
meta += 'W'
i += 2
}
break
case 'X':
meta += 'S'
i += 1
break
case 'E':
case 'I':
case 'O':
case 'U':
meta += cc
i++
break
}

// eslint-disable-next-line no-unmodified-loop-condition,max-len
for (; i < l && (maxPhonemes === 0 || meta.length < maxPhonemes); i += 1) {
cc = word.charAt(i)
nc = word.charAt(i + 1)
pc = word.charAt(i - 1)
nnc = word.charAt(i + 2)

if (cc === pc && cc !== 'C') {
continue
}

switch (cc) {
case 'B':
if (pc !== 'M') {
meta += cc
}
break
case 'C':
if (is(soft, nc)) {
if (nc === 'I' && nnc === 'A') {
meta += 'X'
} else if (pc !== 'S') {
meta += 'S'
}
} else if (nc === 'H') {
meta += !traditional && (nnc === 'R' || pc === 'S') ? 'K' : 'X'
i += 1
} else {
meta += 'K'
}
break
case 'D':
if (nc === 'G' && is(soft, nnc)) {
meta += 'J'
i += 1
} else {
meta += 'T'
}
break
case 'G':
if (nc === 'H') {
if (!(is('BDH', word.charAt(i - 3)) || word.charAt(i - 4) === 'H')) {
meta += 'F'
i += 1
}
} else if (nc === 'N') {
if (is(alpha, nnc) && word.substr(i + 1, 3) !== 'NED') {
meta += 'K'
}
} else if (is(soft, nc) && pc !== 'G') {
meta += 'J'
} else {
meta += 'K'
}
break
case 'H':
if (is(vowel, nc) && !is('CGPST', pc)) {
meta += cc
}
break
case 'K':
if (pc !== 'C') {
meta += 'K'
}
break
case 'P':
meta += nc === 'H' ? 'F' : cc
break
case 'Q':
meta += 'K'
break
case 'S':
if (nc === 'I' && is('AO', nnc)) {
meta += 'X'
} else if (nc === 'H') {
meta += 'X'
i += 1
} else if (!traditional && word.substr(i + 1, 3) === 'CHW') {
meta += 'X'
i += 2
} else {
meta += 'S'
}
break
case 'T':
if (nc === 'I' && is('AO', nnc)) {
meta += 'X'
} else if (nc === 'H') {
meta += '0'
i += 1
} else if (word.substr(i + 1, 2) !== 'CH') {
meta += 'T'
}
break
case 'V':
meta += 'F'
break
case 'W':
case 'Y':
if (is(vowel, nc)) {
meta += cc
}
break
case 'X':
meta += 'KS'
break
case 'Z':
meta += 'S'
break
case 'F':
case 'J':
case 'L':
case 'M':
case 'N':
case 'R':
meta += cc
break
}
}

return meta
}

A community effort

Not unlike Wikipedia, Locutus is an ongoing community effort. Our philosophy follows The McDonald’s Theory. This means that we assimilate first iterations with imperfections, hoping for others to take issue with-and improve them. This unorthodox approach has worked very well to foster fun and fruitful collaboration, but please be reminded to use our creations at your own risk. THE SOFTWARE IS PROVIDED "AS IS" has never been more true than for Locutus.

Now go and: [ View on GitHub | Edit on GitHub | View Raw ]


« More PHP strings functions


Star