wtf-8 is a well-tested WTF-8 encoder/decoder written in JavaScript. WTF-8 is a superset of UTF-8: it can encode/decode any given Unicode code point, including those of (unpaired) surrogates. Here’s an online demo.

Feel free to fork if you see possible improvements!

Via npm:

npm install wtf-8

Via Bower:

bower install wtf-8

Via Component:

component install mathiasbynens/wtf-8

In a browser:

<script src="wtf-8.js"></script>

In Narwhal, Node.js, and RingoJS ≥ v0.8.0:

var wtf8 = require('wtf-8');

In Rhino:

load('wtf-8.js');

Using an AMD loader like RequireJS:

require(
  {
    'paths': {
      'wtf-8': 'path/to/wtf-8'
    }
  },
  ['wtf-8'],
  function(wtf8) {
    console.log(wtf8);
  }
);

Encodes any given JavaScript string (string) as WTF-8, and returns the WTF-8-encoded version of the string.

// U+00A9 COPYRIGHT SIGN; see https://codepoints.net/U+00A9
wtf8.encode('\xA9');
// → '\xC2\xA9'
// U+10001 LINEAR B SYLLABLE B038 E; see https://codepoints.net/U+10001
wtf8.encode('\uD800\uDC01');
// → '\xF0\x90\x80\x81'

Decodes any given WTF-8-encoded string (byteString) as WTF-8, and returns the WTF-8-decoded version of the string. It throws an error when malformed WTF-8 is detected.

wtf8.decode('\xC2\xA9');
// → '\xA9'
wtf8.decode('\xF0\x90\x80\x81');
// → '\uD800\uDC01'
// → U+10001 LINEAR B SYLLABLE B038 E

A string representing the semantic version number.

wtf-8 has been tested in (at least) the latest versions of Chrome, Opera, Firefox, Safari, IE, Node.js, Narwhal, RingoJS, PhantomJS, and Rhino.

Mathias Bynens

wtf-8 is available under the MIT license.