Base64, TypedArrays, and UTF-8 / Unicode conversions in ES5.1 JavaScript (Browser and Node.js)
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Adrian Etter 8f89e4d637 allow es6 import of unibabel.hex 3 months ago
examples directory cleanup 3 years ago
.gitignore Initial commit 4 years ago
LICENSE update urls 1 year ago v2.1.7 1 year ago
bower.json v2.1.7 1 year ago
index.js v2.1.7 1 year ago
node.js publish to npm for browsers as per #7 3 years ago
package.json allow es6 import of unibabel.hex 3 months ago
unibabel.base32.js polyfill Mobile Safari's missing slice 3 years ago
unibabel.hex.js allow es6 import of unibabel.hex 3 months ago


Minimalistic Base64, TypedArrays, and UTF-8 / Unicode conversions in Browser (and Node) JavaScript. Optional add-on support for hex and base32.


See also

Are you in the right place?

Dear Node.js Users:

You SHOULD NOT use this module. You already have Buffer and thirty-two:

var buf = new Buffer('I ½ ♥ 💩', 'utf8');
buf.toString('binary'); // deprecated, do not use


You just include the index.js in a lovely script tag.

bower install --save unibabel
<script src="/bower_components/unibabel/index.js"></script>

Dear WebPack / Broccoli, Gulp / Grunt Users:

I don’t know how your build systems work these days, but Unibabel is exported as Unibabel and you can access it as window.Unibabel.


// TypedArray <--> UTF8
var uint8Array = Unibabel.strToUtf8Arr(str);
var str = Unibabel.utf8ArrToStr(uint8Array);

// TypedArray <--> Base64
var base64 = Unibabel.arrToBase64(uint8Array)
var uint8Array = Unibabel.base64ToArr(base64)

Normal APIs


  • utf8ToBuffer(utf8str) => array
  • bufferToUtf8(array) => string
  • utf8ToBase64(utf8str) => base64
  • base64ToUtf8(base64) => string
  • bufferToBase64(array) => base64
  • base64ToBuffer(base64) => array

Hex APIs


  • hexToBuffer(hexstr) => array
  • bufferToHex(array) => hexstr

Base32 APIs


  • base32ToBuffer(b32str) => array
  • bufferToBase32(array) => b32str

Helper APIs

  • utf8ToBinaryString(utf8str) => binstr
  • binaryStringToUtf8(binstr) => utf8str
  • bufferToBinaryString(buffer) => binstr
  • binaryStringToBuffer(binstr) => array


// Base64

var myArray = Unibabel.base64ToArr("QmFzZSA2NCDigJQgTW96aWxsYSBEZXZlbG9wZXIgTmV0d29yaw=="); // "Base 64 \u2014 Mozilla Developer Network"
var myBuffer = Unibabel.base64ToArr("QmFzZSA2NCDigJQgTW96aWxsYSBEZXZlbG9wZXIgTmV0d29yaw==").buffer; // "Base 64 \u2014 Mozilla Developer Network"


// Crazy Unicode

var sMyInput = "I'm a ☢ ☃ that plays 𝄢 guitar and spea̧͈͖ks Ar̽̾̈́͒͑ ̶̧̨̱̹̭̯ͧ̾ͬC̷̙̲̝͖ͭ̏ͥͮ͟Oͮ͏̮̪̝͍M̲̖͊̒ͪͩͬ̚̚͜!";
var aMyUTF8Input = Unibabel.strToUtf8Arr(sMyInput);
var sMyBase64 = Unibabel.arrToBase64(aMyUTF8Input);


var aMyUTF8Output = Unibabel.base64ToArr(sMyBase64);
var sMyOutput = Unibabel.utf8ArrToStr(aMyUTF8Output);



  • index.js and unibabel.hex.js are dual-licensed as Apache 2.0 and MIT.
  • unibabel.base32.js is a modified version of thirty-two and is therefore licensed MIT.

Some parts of the code were taken from MDN, which Mozilla has licensed in the Public Domain, which means that I am at liberty to re-license my copy under the Apache 2 and MIT licenses.




Added unibabel.base32.js


The new implementation is binary compatible with node.js, TextEncoder, and other more-common UTF-8 encodings.

It is also based on DOM APIs which result in much less code and are still backwards compatible all the way back to IE6 (not on purpose, just that it happens to work).



This version was based on the work by good folks at the MDN, however, the UTF-8 conversion was not byte-compatible with other UTF-8 conversions (such as node.js and TextEncoder), so don’t use it. See