why does this regular expression always return null?

Question

I need to check for classname in a string.

My string looks like this:

testing ui-li ui-li-static ui-btn-up-f ui-first-child ui-filter-hidequeue

And I'm looking for ui-filter-hidequeue anywhere (!) in the string.

This is what I'm trying. It always returns null, althought the string I'm trying to match is there:

var classPassed = 'ui-filter-hidequeue',
    classes = 'testing ui-li ui-li-static ui-btn-up-f ui-first-child ui-filter-hidequeue',
    c = new RegExp('(?:\s|^)' + classPassed + '(?:\s|$)');

console.log( classes.match(c) )   // returns null

Question:
Can someone tell me what I'm doing wrong in my regex?

Thanks!

Where do you get the string from? If these are the classes for an element, jQuery lets you test class membership directly, I think. — Thilo, Mar 25 '13 at 00:44
@AndrewWhitaker: yes. jQuery is easy, but I'm trying in pure Javascript. — frequent, Mar 25 '13 at 00:48

score 2 · Accepted Answer · answered Mar 25 '13 at 00:44

2

You need \\s. \s inside a string would escape the s, which does not need escaping, and evaluates to just a "s". I.e. when you're using literal regexps, backslash escapes the regexp characters; and strings also use backslash to escape string characters. When you build a regexp from a string, you need to think about both of these layers: \\ is string-escapey way of saying \, then \s is the regexp-escapey way of saying whitespace.

Also as other comments show, there are WAY better ways to test for class presence. :) But I just answered the direct error.

answered Mar 25 '13 at 00:44

Amadan

179,482
20
216
275

ok. Still curious on "way better". – frequent Mar 25 '13 at 00:47
@frequent: I think you got it, with Thilo's comment above. Regexp is an overkill, besides being unreliable for non-controlled input. – Amadan Mar 25 '13 at 00:53

score 0 · Answer 2 · answered Mar 25 '13 at 00:44

0

Try using this:

c = new RegExp('\\b' + classPassed + '\\b');

The \b metacharacter matches at the beginning/end of a word; it stands for boundary (as in, word boudnary).

answered Mar 25 '13 at 00:44

lxop

6,551
2
19
35

1

`\b` is not appropriate, as it will also match at the boundary between the minus sign and a word. – Amadan Mar 25 '13 at 00:46
ok. thanks for the info. and explanation on `b`! – frequent Mar 25 '13 at 00:46
@Amadan, ah, good point, I sometimes forget about the minus signs being part of a 'word' in HTML classes etc – lxop Mar 25 '13 at 00:47

why does this regular expression always return null?

2 Answers2