Create a gist now

Instantly share code, notes, and snippets.

What would you like to do?
Javascript Base64 UTF8 for the Browser / Server. Base64 UTF-8 Encoding and Decoding Libraries / Modules for AMD, CommonJS, Nodejs and Browsers. Cross-browser compatible.
// UTF8 Module
//
// Cleaner and modularized utf-8 encoding and decoding library for javascript.
//
// copyright: MIT
// author: Nijiko Yonskai, @nijikokun, nijikokun@gmail.com
(function (name, definition, context, dependencies) {
if (typeof context['module'] !== 'undefined' && context['module']['exports']) { if (dependencies && context['require']) { for (var i = 0; i < dependencies.length; i++) context[dependencies[i]] = context['require'](dependencies[i]); } context['module']['exports'] = definition.apply(context); }
else if (typeof context['define'] !== 'undefined' && context['define'] === 'function' && context['define']['amd']) { define(name, (dependencies || []), definition); }
else { context[name] = definition.apply(context); }
})('utf8', function () {
return {
encode: function (string) {
if (typeof string !== 'string') return string;
else string = string.replace(/\r\n/g, "\n");
var output = "", i = 0, charCode;
for (i; i < string.length; i++) {
charCode = string.charCodeAt(i);
if (charCode < 128)
output += String.fromCharCode(charCode);
else if ((charCode > 127) && (charCode < 2048))
output += String.fromCharCode((charCode >> 6) | 192),
output += String.fromCharCode((charCode & 63) | 128);
else
output += String.fromCharCode((charCode >> 12) | 224),
output += String.fromCharCode(((charCode >> 6) & 63) | 128),
output += String.fromCharCode((charCode & 63) | 128);
}
return output;
},
decode: function (string) {
if (typeof string !== 'string') return string;
var output = "", i = 0, charCode = 0;
while (i < string.length) {
charCode = string.charCodeAt(i);
if (charCode < 128)
output += String.fromCharCode(charCode),
i++;
else if ((charCode > 191) && (charCode < 224))
output += String.fromCharCode(((charCode & 31) << 6) | (string.charCodeAt(i + 1) & 63)),
i += 2;
else
output += String.fromCharCode(((charCode & 15) << 12) | ((string.charCodeAt(i + 1) & 63) << 6) | (string.charCodeAt(i + 2) & 63)),
i += 3;
}
return output;
}
};
}, this);
// Base64 Module
//
// Cleaner, modularized and properly scoped base64 encoding and decoding module for strings.
//
// copyright: MIT
// author: Nijiko Yonskai, @nijikokun, nijikokun@gmail.com
(function (name, definition, context, dependencies) {
if (typeof context['module'] !== 'undefined' && context['module']['exports']) { if (dependencies && context['require']) { for (var i = 0; i < dependencies.length; i++) context[dependencies[i]] = context['require'](dependencies[i]); } context['module']['exports'] = definition.apply(context); }
else if (typeof context['define'] !== 'undefined' && context['define'] === 'function' && context['define']['amd']) { define(name, (dependencies || []), definition); }
else { context[name] = definition(); }
})('base64', function (utf8) {
var $this = this;
var $utf8 = utf8 || this.utf8;
var map = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/=";
return {
encode: function (input) {
if (typeof $utf8 === 'undefined') throw { error: "MissingMethod", message: "UTF8 Module is missing." };
if (typeof input !== 'string') return input;
else input = $utf8.encode(input);
var output = "", a, b, c, d, e, f, g, i = 0;
while (i < input.length) {
a = input.charCodeAt(i++);
b = input.charCodeAt(i++);
c = input.charCodeAt(i++);
d = a >> 2;
e = ((a & 3) << 4) | (b >> 4);
f = ((b & 15) << 2) | (c >> 6);
g = c & 63;
if (isNaN(b)) f = g = 64;
else if (isNaN(c)) g = 64;
output += map.charAt(d) + map.charAt(e) + map.charAt(f) + map.charAt(g);
}
return output;
},
decode: function (input) {
if (typeof $utf8 === 'undefined') throw { error: "MissingMethod", message: "UTF8 Module is missing." };
if (typeof input !== 'string') return input;
else input = input.replace(/[^A-Za-z0-9\+\/\=]/g, "");
var output = "", a, b, c, d, e, f, g, i = 0;
while (i < input.length) {
d = map.indexOf(input.charAt(i++));
e = map.indexOf(input.charAt(i++));
f = map.indexOf(input.charAt(i++));
g = map.indexOf(input.charAt(i++));
a = (d << 2) | (e >> 4);
b = ((e & 15) << 4) | (f >> 2);
c = ((f & 3) << 6) | g;
output += String.fromCharCode(a);
if (f != 64) output += String.fromCharCode(b);
if (g != 64) output += String.fromCharCode(c);
}
return $utf8.decode(output);
}
}
}, this, [ "utf8" ]);
// Base64 Module
//
// Cleaner, modularized and properly scoped base64 encoding and decoding module for strings.
//
// copyright: MIT
// author: Nijiko Yonskai, @nijikokun, nijikokun@gmail.com
(function (name, definition, context, dependencies) {
if (typeof context['module'] !== 'undefined' && context['module']['exports']) { if (dependencies && context['require']) { for (var i = 0; i < dependencies.length; i++) context[dependencies[i]] = context['require'](dependencies[i]); } context['module']['exports'] = definition.apply(context); }
else if (typeof context['define'] !== 'undefined' && context['define'] === 'function' && context['define']['amd']) { define(name, (dependencies || []), definition); }
else { context[name] = definition(); }
})('base64', function (utf8) {
var $this = this;
var $utf8 = utf8 || this.utf8;
var map = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/=";
return {
encode: function (input) {
if (typeof $utf8 === 'undefined') throw { error: "MissingMethod", message: "UTF8 Module is missing." };
if (typeof input !== 'string') return input;
else input = $utf8.encode(input);
var output = "", a, b, c, d, e, f, g, i = 0;
while (i < input.length) {
a = input.charCodeAt(i++);
b = input.charCodeAt(i++);
c = input.charCodeAt(i++);
d = a >> 2;
e = ((a & 3) << 4) | (b >> 4);
f = ((b & 15) << 2) | (c >> 6);
g = c & 63;
if (isNaN(b)) f = g = 64;
else if (isNaN(c)) g = 64;
output += map.charAt(d) + map.charAt(e) + map.charAt(f) + map.charAt(g);
}
return output;
},
decode: function (input) {
if (typeof $utf8 === 'undefined') throw { error: "MissingMethod", message: "UTF8 Module is missing." };
if (typeof input !== 'string') return input;
else input = input.replace(/[^A-Za-z0-9\+\/\=]/g, "");
var output = "", a, b, c, d, e, f, g, i = 0;
while (i < input.length) {
d = map.indexOf(input.charAt(i++));
e = map.indexOf(input.charAt(i++));
f = map.indexOf(input.charAt(i++));
g = map.indexOf(input.charAt(i++));
a = (d << 2) | (e >> 4);
b = ((e & 15) << 4) | (f >> 2);
c = ((f & 3) << 6) | g;
output += String.fromCharCode(a);
if (f != 64) output += String.fromCharCode(b);
if (g != 64) output += String.fromCharCode(c);
}
return $utf8.decode(output);
}
}
}, this, [ "utf8" ]);
// UTF8 Module
//
// Cleaner and modularized utf-8 encoding and decoding library for javascript.
//
// copyright: MIT
// author: Nijiko Yonskai, @nijikokun, nijikokun@gmail.com
(function (name, definition, context, dependencies) {
if (typeof context['module'] !== 'undefined' && context['module']['exports']) { if (dependencies && context['require']) { for (var i = 0; i < dependencies.length; i++) context[dependencies[i]] = context['require'](dependencies[i]); } context['module']['exports'] = definition.apply(context); }
else if (typeof context['define'] !== 'undefined' && context['define'] === 'function' && context['define']['amd']) { define(name, (dependencies || []), definition); }
else { context[name] = definition.apply(context); }
})('utf8', function () {
return {
encode: function (string) {
if (typeof string !== 'string') return string;
else string = string.replace(/\r\n/g, "\n");
var output = "", i = 0, charCode;
for (i; i < string.length; i++) {
charCode = string.charCodeAt(i);
if (charCode < 128)
output += String.fromCharCode(charCode);
else if ((charCode > 127) && (charCode < 2048))
output += String.fromCharCode((charCode >> 6) | 192),
output += String.fromCharCode((charCode & 63) | 128);
else
output += String.fromCharCode((charCode >> 12) | 224),
output += String.fromCharCode(((charCode >> 6) & 63) | 128),
output += String.fromCharCode((charCode & 63) | 128);
}
return output;
},
decode: function (string) {
if (typeof string !== 'string') return string;
var output = "", i = 0, charCode = 0;
while (i < string.length) {
charCode = string.charCodeAt(i);
if (charCode < 128)
output += String.fromCharCode(charCode),
i++;
else if ((charCode > 191) && (charCode < 224))
output += String.fromCharCode(((charCode & 31) << 6) | (string.charCodeAt(i + 1) & 63)),
i += 2;
else
output += String.fromCharCode(((charCode & 15) << 12) | ((string.charCodeAt(i + 1) & 63) << 6) | (string.charCodeAt(i + 2) & 63)),
i += 3;
}
return output;
}
};
}, this);

tadaa9 commented May 18, 2013

There is an error in the "decode" function of "base64". You must use the "window.atob" function and not the "window.btoa" function.

Owner

nijikokun commented May 24, 2013

Thanks! I've updated it to use atob @tadaa9

Thanks for that code. In some Browsers (i.e. my Chrome28 and FF22) the native btoa function has problems with multi byte characters (throws InvalidCharacterError).

So, wouldn't it be better to utf8 de/encode those strings, too? Something like:

encode: function (input) {
      if (typeof $utf8 === 'undefined') throw { error: "MissingMethod", message: "UTF8 Module is missing." };
      if (typeof input !== 'string') return input;
      else input = $utf8.encode(input);
      if (typeof $this.btoa !== 'undefined') return $this.btoa(input); // moved
      var output = "", a, b, c, d, e, f, g, i = 0;
      // ...

and

decode: function (input) {
      if (typeof $utf8 === 'undefined') throw { error: "MissingMethod", message: "UTF8 Module is missing." };
      if (typeof input !== 'string') return input;
      else input.replace(/[^A-Za-z0-9\+\/\=]/g, "");
      if (typeof $this.atob !== 'undefined') return $utf8.decode($this.atob(input)); // moved and decoded
      // ...

With these changes i get no problems with lines like:

base64.decode(base64.encode('Hello World 秦始皇帝'))
Owner

nijikokun commented Sep 27, 2013

@delbertooo Actually this makes more sense. Updating to reflect changes

I'm having a lot of trouble getting this library to work because I can find no documentation or examples of how to integrate this library and I'm unsure if it's the content I'm attempting to decode or the way I'm attempting to use it. Could we get a link to a couple examples demonstrating it's use and any details concerning Content Delivery Network (CDN) access?

Owner

nijikokun commented Dec 9, 2013

@modusinternet simply copy the raw data from the first file on this gist (it includes both base64 and utf8)

then you simply base64.decode(base64.encode('Hello World 秦始皇帝'))

vote539 commented Jun 25, 2014

Where you have context['define'] === 'function' on line 9, I think you meant to put typeof context['define'] === 'function'.

vote539 commented Jun 25, 2014

Also, I'm not having much luck with the Gist. The following example seems to fail.

var data = "ZGlzcCgiSGVsbG8gV8OzcmxkIDMiKTsKJSBow6lsbMOzIHfDp3JsZAo=";
console.log(require("base64").decode(data));

/* * * Expected Value * * *\
disp("Hello Wórld 3");
% hélló wçrld
*/

/* * * Actual Value * * *\
disp("Hello Wórld 3");
% hélló wçrld
*/
Owner

nijikokun commented Sep 11, 2014

@vote539 I've removed atob and btoa as they are the culprit here, they do not do well with utf8 and I don't really like the workaround for it: https://developer.mozilla.org/en-US/docs/Web/API/WindowBase64/Base64_encoding_and_decoding#The_.22Unicode_Problem.22

Owner

nijikokun commented Sep 11, 2014

The code you provided with the updated gist works as expected now. Sorry for the late reply.

@nijikokun you should replace your "input.replace" (base64.module.js#L44) with following code. Strings are immutable in JavaScript.

https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/replace
http://stackoverflow.com/a/4717855

// @component base64.module.js
input = input.replace(/[^A-Za-z0-9\+\/\=]/g, "");
Owner

nijikokun commented Nov 11, 2016

@friedemannsommer you are correct, gist has been updated 👍 nice catch

shubo commented Feb 21, 2017 edited

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment