unable to convert unicode character in javascript

Question

I have a function as follows in javascript

function myfun()
{
  var a = "'\u0c39";
  alert(a);

}

When i called this function i am able to see correct letter(This letter is telugu language letter).

My problem is my string is coming from java which is like "'\\u0c39". I used as follows

function myfun()
{
  var a = "'\\u0c39";
  alert(a.replace("\\\\","\\").toString());

}

But in alert '\u0c39 is coming.What might be the mistake here.Please help me.Thanks in advance...

You can't get `"\u0c39"` from `"\\u0c39"` by removing backslashes. Hint: there are no backslashes in `"\u0c39"` — John Dvorak, Jun 13 '13 at 12:48
@PSR `"\u0c39"` is a one-character string. `"\\u0c39"` is its encoded form, a six-character string. — John Dvorak, Jun 13 '13 at 12:49
@JanDvorak see here i created a link http://jsfiddle.net/FersM/1/ — PSR, Jun 13 '13 at 12:51
Have a look at http://stackoverflow.com/q/7885096 and see if it helps. — Felix Kling, Jun 13 '13 at 12:52
@PSR not sure how to do such unescaping except manually. On hack might be JSON-encode -> do the replacement -> JSON-decode, but I can see problems with that. — John Dvorak, Jun 13 '13 at 12:52
@FelixKling this is not working in IE.But working in firefox — PSR, Jun 13 '13 at 13:03

Brett Zamir · Accepted Answer · 2013-06-13T14:49:58.787

5

Working in IE 10 (including back to IE7 mode): http://jsfiddle.net/FersM/2/

Try this:

function myfun()
{
  var a = "'\\u0c39";
  a = a.replace(/\\u([a-f0-9]{4})/gi, function (n, hex) {
      return String.fromCharCode(parseInt(hex, 16));
  });

  alert(a);

}

If you want to understand it, the regular expression is looking for all instances of a backslash (double-escaping needed within a regex literal) followed by exactly 4 hexadecimal digits (of whatever case), and then the function will replace the contents of each sequence thus found with the return value. The return value uses fromCharCode to convert a Unicode "code point" value into the actual character, but it requires the code point as a number (decimal), so we must first convert the 4 hex characters (the first parenthetical match, and thus the 2nd argument of the function, with the first argument being the whole sequence match which we don't need) into a decimal using the parseInt function with the base argument as "16".

edited Jun 13 '13 at 14:49

answered Jun 13 '13 at 13:09

Brett Zamir

14,034
6
54
77

it is working for only single character – PSR Jun 13 '13 at 13:17
if i have var a = "'\\u0c2a\\u0c4d"; it is not working – PSR Jun 13 '13 at 13:17
I forgot the "g" flag... Added now... – Brett Zamir Jun 13 '13 at 13:26
FYI, the eval() solution is not a safe solution unless you are absolutely sure of the contents. – Brett Zamir Jun 14 '13 at 22:16

score 1 · Answer 2 · answered Jun 13 '13 at 14:05

1

You can use this eval code for that issue

http://jsfiddle.net/6Hmyk/

function myfun1()
{
  var a = "'\\u0c39";

   alert(eval('"'+a+'"'));   

}

myfun1();

In that code , \\ will change to \ and eval code run this as unicode character

answered Jun 13 '13 at 14:05

Taleh Ibrahimli

750
4
13
29

1

eval() may work, but it is not a safe solution (unless one happens to know the exact input and know that, for example, the code doesn't add its own quotes to escape out of your string and do something else). – Brett Zamir Jun 14 '13 at 08:51

unable to convert unicode character in javascript

2 Answers2