spliddit

This is a fork of essdot/spliddit

Spliddit — unicode-aware JS string splitting

Split a string into its constituent characters, without munging emoji and other non-BMP code points.

Installation

npm i @kiraind/spliddit

Why?

The native String#split implementation does not pay attention to surrogate pairs. When the code units of a surrogate pair are split apart, they are not intelligible on their own. Unless they are put back together in the correct order, individual code units will cause problems in code that handles strings.

Consider:

const emojiMessage = 'Hello 😤'

emojiMessage.split('').reverse().join('')
// => String with messed-up emoji

spliddit will correctly split the string into strings consisting of legible characters:

import spliddit from '@kiraind/spliddit'

const emojiMessage = 'Hello 😤'

spliddit(emojiMessage).reverse().join('')
// => '😤 olleH'

Also, since surrogate pairs take up two spaces in the Javascript string to represent a single character, spliddit can help you correctly count the number of code points (characters) in the string:

const emoji = '🍔'
const han = '𠬠典'

emoji.length
// => 2
han.length
// => 3

spliddit(emoji).length
// => 1
spliddit(han).length
// => 2

Alternatively, you can pass spliddit an array that potentially has broken-apart surrogate pairs, and spliddit will return an array that has them put back together:

const myCoolString = '😎 Fooool'

// Messed-up array beginning with a split-apart surrogate pair :(
const myBustedArray = myCoolString.split('')

// Aww yeah cool guy is back
const myCoolFixedArray = spliddit(myBustedArray)

Delimiter

You can also pass spliddit a second argument, a string or RegExp representing the delimiter to split by. The native String#split implementation does this correctly, so spliddit just passes through to String#split in this case.

spliddit('hi🍔hi', '🍔')
// => ['hi', 'hi']

spliddit('123a456', 'a')
// => ['123', '456']

Country Flags

Country flags like 🇦🇴 are composed of two regional indicator Unicode characters. Each regional indicator character is represented as a surrogate pair in JavaScript strings, so country flags take up 4 code units. The regional indicator symbols follow the alphabet, and the two regional indicators used follow the country's code.

(For example, 🇮🇹 , Italy's flag, is U+1F1EE 'REGIONAL INDICATOR SYMBOL LETTER I' followed by U+1F1F9 'REGIONAL INDICATOR SYMBOL LETTER T'.)

spliddit will split pairs of regional indicator characters (4 total code units) into one character even though they consist of two Unicode code points.

Skin tone emoji

Skin tone emojis (👩🏾) are composed of a color-neutral emoji that depicts humans (👩), followed by one of the 5 Unicode skin tone modifier characters (🏻, 🏼, 🏽, 🏾, 🏿). The emoji character and the skin tone modifier are each represented as a surrogate pair in JavaScript strings.

spliddit will split these sequences (4 total code units) into one character even though they consist of two Unicode code points.

Other functions

spliddit.hasPair(s)

Tells if s contains a surrogate pair.

import * as spliddit from '@kiraind/spliddit'

spliddit.hasPair('Look 👀 wow')
// => true
spliddit.hasPair('abcdef')
// => false

spliddit.isFirstOfSurrogatePair(c)

Tells if c[0] (the first item in c) is the first code unit of a surrogate pair. (Character codes 0xD800 through 0xDFFF)

const s = '👴'
const sFirst = s[0]
const sArr = s.split('')

spliddit.isFirstOfSurrogatePair(s)
// => true

spliddit.isFirstOfSurrogatePair(sFirst)
// => true

spliddit.isFirstOfSurrogatePair(sArr)
// => true

spliddit.isFirstOfSurrogatePair(sArr[0])
// => true

spliddit.isFirstOfSurrogatePair('a')
// => false