kirilloid
Feb 8th

  String.fromCharCode accepts several arguments.
Replacing lines 34-36 with

    enc = String.fromCharCode((c1 >> 6) | 192, (c1 & 63) | 128);
} else {
    enc = String.fromCharCode((c1 >> 12) | 224, ((c1 >> 6) & 63) | 128, (c1 & 63) | 128);

may reduce execution time from 20x to 12x on mostly non-ascii strings (e.g. cyrillic text).

Brett Zamir
2 May '11

  @Gajus: JavaScript uses Unicode internally, even if your document is encoded in ISO-8859-1. This function should only be needed if you have a string already using that encoding (or otherwise you are double-encoding). You could provide the codepoints you have, and what you expect it to return.

Gajus
1 May '11

  Doesn't work with the following string:
€,´,€,´,水,Д,Є

Soulcyon
7 Apr '11

function utf8_encode(){
    var str = arguments[0] + "",
        len = str.length - 1,
        i = -1,
        result = "";
        
    while( !!(i++ - len) ){
        var c = str.charCodeAt(i),
            ops = [
                c,
                (c >> 6 | 192) + (c & 63 | 128),
                (c >> 12 | 224) + (c >> 6 & 63 | 128) + (c & 63 | 128)
            ],
            i =  c < 128 ? 0 : c < 2048 ? 1 : 2;
        result += String.fromCharCode(ops[i]);
    }
    return result;
}

Brett Zamir
15 Jan '11

  @Manu: What are you trying to do? Do you have some sample code?

Manu
10 Jan '11

  this function doesnt work for me
Do i have to declare the Javascript version?

Anthon Pang
9 Jan '11

  Patch using Eli's suggestion (though, he posted the equivalent for utf8_decode):

--- utf8_encode.js.old	2011-01-09 12:23:22.000000000 -0500
+++ utf8_encode.js	2011-01-09 12:23:49.000000000 -0500
@@ -11,6 +11,10 @@
     // *     example 1: utf8_encode('Kevin van Zonneveld');
     // *     returns 1: 'Kevin van Zonneveld'
 
+    if (typepof window.encodeURIComponent !== 'undefined') {
+        return unescape( window.encodeURIComponent( argString ));
+    }
+
     var string = (argString+''); // .replace(/\r\n/g, "\n").replace(/\r/g, "\n");
 
     var utftext = "";

Anthon Pang
8 Jan '11

  Eli. That's not cross browser portable, plus it won't work with some input, e.g., "malformed URI sequence" errors on FF.

Eli Grey
19 Sep '10

  This entire function can be replaced with the following.

function utf8_encode (argString) {
    return decodeURIComponent(escape(argString));
}

Brett Zamir
10 Apr '10

  @Cristián: I think what happened is that you must have copied the text directly from this page, but the commenting code often messes up the new lines. To get a pristine version (and the latest version), use the link "raw js source"...

Cristián Pérez
7 Apr '10

  Hey, you have an error.
You are using into the code the object string.*, but the argument name is "argString" instead of "string".

Using argString instead of string it works correctly

Keith
1 Dec '09

  This function will throw an exception if passed an empty string.

I think it needs to include "

try {} catch(e) {} return'';

" around its contents and the following line at the start:

 if (argString == '') return '';

Ben Pettit
5 May '09

  I made a fix so this function ran correctly in adobe javascript.

function utf8_encode ( string ) {
    // Encodes an ISO-8859-1 string to UTF-8  
    // 
    // version: 812.316
    // discuss at: http://phpjs.org/functions/utf8_encode
    // +   original by: Webtoolkit.info (http://www.webtoolkit.info/)
    // +   improved by: Kevin van Zonneveld (http://kevin.vanzonneveld.net)
    // +   improved by: sowberry
    // +    tweaked by: Jack
    // +   bugfixed by: Onno Marsman
    // +   improved by: Yves Sucaet
    // +   bugfixed by: Onno Marsman
    // +   adobe js by: Ben Pettit
    // *     example 1: utf8_encode('Kevin van Zonneveld');
    // *     returns 1: 'Kevin van Zonneveld'
    string	=	string.valueOf(); //    <-bp:  I added this line.
    
    string = (string+'').replace(/\r\n/g, "\n").replace(/\r/g, "\n");

    var utftext = "";
    var start, end;
    var stringl = 0;

Kevin van Zonneveld
14 Nov '08

  @ Onno Marsman: Sjeesh, it has been a long day... But that long.. Thx Onno.

Onno Marsman
14 Nov '08

  This is just weird. Of course the extra (string+'') is not necessary. The following would do exactly the same:

[CODE="Javascript"]
string = (string+'').replace(/\r\n/g, "\n").replace(/\r/g, "\n");
[/CODE]

or even something like (not tested):
[CODE="Javascript"]
string = (string+'').replace(/\r\n?/g, "\n");
[/CODE]

Kevin van Zonneveld
13 Nov '08

  @ Yves Sucaet: I don't see the harm in that :) thank you Yves!

Yves Sucaet
12 Nov '08

  I think it makes sense to replace
[CODE="Javascript"]
string = (string+'').replace(/\r\n/g,"\n");
[/CODE]

with

[CODE="Javascript"]
string = (string+'').replace(/\r\n/g,"\n");
string = (string+'').replace(/\r/g,"\n");
[/CODE]

Kevin van Zonneveld
27 Aug '08

  @ sowberry: Thank you for your improvement!

sowberry
8 Aug '08

  While looking for a javascript crc script, I found the version on webtoolkit.info as well as your subsequent modification.

Testing with a chunk of text a couple hundred characters long, with just a couple non-ascii values, I saw no significant improvement with your approach of using an array as a pseudo-StringBuilder. The issue is the use of String.fromCharCode for even ascii values, which forces too many string creations. The code below is about 3 times faster in my tests:

[CODE="Javascript"]
function utf8_encode(string) {
string = string.replace(/\r\n/g,"\n");
var utftext = "";
var start, end;

start = end = 0;
for (var n = 0; n < string.length; n++) {

var c = string.charCodeAt(n);
var enc = null;

if (c < 128) {
end++;
}
else if((c > 127) && (c < 2048)) {
enc = String.fromCharCode((c >> 6) | 192) + String.fromCharCode((c & 63) | 128);
}
else {
enc = String.fromCharCode((c >> 12) | 224) + String.fromCharCode(((c >> 6) & 63) | 128) + String.fromCharCode((c & 63) | 128);
}
if (enc != null)
{
if (end > start)
{
utftext += string.substring(start, end);
}
utftext += enc;
start = end = n+1;
}

}
if (end > start)
{
utftext += string.substring(start, string.length);
}

return utftext;
}
[/CODE]

Please feel free to post this to the various script repositories, as I am not especially active on the web. Thanks.