Characters with ASCII > 128 are not correctly read in Javascript

Question

I have a HTML that includes a Javascript file. This script contains a special character, ASCII 152. When I try to display the charCodeAt, I get different results, but never the right one. Could you please advice? Thanks

TEST.HTML

<script type="text/javascript" charset=SEE BELOW src="test.js">
</script>

TEST.JS file with ANSI encoding

function d(a)
{
a=(a+"").split("");
alert(a[1].charCodeAt(0));
};
d("i˜g"); // Note that ˜ is 152 in ASCII

TEST.HTML with x-user-defined charset: alert shows 63384. With %63232 works, as every char >128 is displayed as 63232+char.
TEST.HTML with utf-8 charset: alert shows 65533. All chars > 128 are displayed as 65533.
TEST.HTML with Windows-1252 charset: alert shows 752. I cannot find a relation between ASCII and what is displayed.

TEST.JS file with UTF-8 encoding

function d(a)
{
a=(a+"").split("");
alert(a[1].charCodeAt(0));
};
d("i[x98]g"); // Note that x98 is 152

TEST.HTML with x-user-defined charset: alert shows 65533. All chars > 128 are displayed as 65533.
TEST.HTML with utf-8 charset: alert shows 65533. All chars > 128 are displayed as 65533.
TEST.HTML with Windows-1252 charset: alert shows 65533. All chars > 128 are displayed as 65533.

There is no characters in range 127-255 for utf8, and ASCII ends at 127... Also the character at position `1` in `"i[x98]g"` is a `"["` — Esailija, Apr 09 '12 at 08:38

Esailija · Accepted Answer · 2012-04-09T08:51:34.340

6

There is no characters in range 128-255 for utf8, and ASCII ends completely at 127... Also the character at position 1 in "i[x98]g" is a "[", the "[x98]" is meaningless.

Your function can be replaced with str.charCodeAt(1).

The character ˜ is Unicode Character 'SMALL TILDE' (U+02DC and can be written as "\u02DC", or String.fromCharCode(732)

edited Apr 09 '12 at 08:51

answered Apr 09 '12 at 08:45

Esailija

138,174
23
272
326

1

Additional info: [charCodeAt()](https://developer.mozilla.org/en/JavaScript/Reference/Global_Objects/String/charCodeAt) works with Unicode and does not really care about ASCII. – Álvaro González Apr 09 '12 at 09:01
Sorry, I meant ASCII EXTENDED. As per below, 152 is '˜' http://www.ascii-code.com/ [x98] is one character. It is the way Notepad++ represents that 152 char in UTF-8. 98 in hexadecimal is 152. – Apr 09 '12 at 10:12
Thank you all. Then I rephrase my question. I want to write a character representing integer 152 in TEST.JS as parameter of d(), so that when I run this function alert displays 152. I am open to use any encoding. – Apr 09 '12 at 10:16
1

@Arturo in that case you can to write `"\u0098"`, or this invisible character: `""`, or `String.fromCharCode(152)` – Esailija Apr 09 '12 at 10:23
@Arturo, that’s a completely new question and should be asked separately. I cannot make sense of that question either, as there is no character representing integer 152. If you wish to map characters to integers according to a scheme of your own, or applying someone else’s mapping, use table for that. But it is difficult to guess what might be the real question here. – Jukka K. Korpela Apr 09 '12 at 11:34
I believe there is some kind of problem between chars 152 (\u0098) and 732 (\u02dc), as both are displayed with the same symbol. How can I force charCodeAt() to use the correct one (152)? http://www.fileformat.info/info/unicode/char/2dc/index.htm http://www.fileformat.info/info/unicode/char/098/index.htm – Apr 09 '12 at 21:31
Thank you all. As suggested, I have opened a new question: http://stackoverflow.com/questions/10080605/special-character-u0098-read-as-u02dc-using-charcodeat – Apr 09 '12 at 22:20

score 0 · Answer 2 · answered Apr 09 '12 at 08:41

0

ASCII only has 127 characters. char 152 is non-existant

answered Apr 09 '12 at 08:41

hofnarwillie

3,563
10
49
73

1

Sorry, I meant ASCII EXTENDED. As per below, 152 is '˜' http://www.ascii-code.com/ – Apr 09 '12 at 10:08

Characters with ASCII > 128 are not correctly read in Javascript

2 Answers2

Linked