Skip to content

Creates the inverse of transliterated string to a regex. What? Basically, diacritic insensitiveness

License

Notifications You must be signed in to change notification settings

pocesar/js-diacritic-regex

Repository files navigation

Build Status Coverage Status

NPM

Diacritic regex

Creates the inverse of transliterated string to a regex. What? Basically, a regex that is diacritic insensitive

Why?

Sometimes the user will search for blasé, but your database is dumb and doesn't understand collations and diacritic insensitiveness, but it can compare stuff using regex, so there ya go.

How?

Suppose you have the word résumé but written improperly in the database as resume. The user is clever, and types it correctly into the search box. Gets nothing. How to search for all the weird cases people mistype stuff when comes to accents?

import { toRegex, toString } from 'diacritic-regex';

toRegex()('résumé') // => /r[eEÉéÈèÊêëË]s[úùÚÙüÜuU]m[eEÉéÈèÊêëË]/i;
toRegex({flags: 'mu'})('résumé') // => /r[eEÉéÈèÊêëË]s[úùÚÙüÜuU]m[eEÉéÈèÊêëË]/mu;
toRegex({
  flags: '',
  mappings: {
    'e': 'eéÉ'
  }
})('résumé') // => /r[eéÉ]s[úùÚÙüÜuU]m[eéÉ]/;

toString({
    mappings: {
        '*': ['\\S+'] // literals, won't try to wrap in []'s,
        'u': ['u']
    }
})('résumé*') // => 'r[eEÉéÈèÊêëË]sum[eEÉéÈèÊêëË]\S+'

If you want to change the mappings for all instances:

import { mappings } from 'diacritic-regex'

mappings['*'] = ['[\\S\\s]+']

Caveats

Be aware of RegExp.prototype.exec with g flag being stateful

The i flag is appended to the RegExp flags if you don't pass any flags to toRegex

Compatibility

Work in node and the browser, but needs polyfills for Array.reduce, Array.map and Object.keys depending on how old your target browser is

License

MIT