A quick way to preg_match a utf-8 string

Question

$input = "žąsis su šešiolika žąsyčių";
preg_match_all("/\b(žąs\S*)/iu", $input, $output_array);
print_r($output_array);

returns a one big nothing. I want it to return both "žąsis" and "žąsyčių". Seems to be a simple problem but I can't find a simple answer. Should I encode both the subject and the pattern somehow or?..

And by meaning "one big nothing" I mean an empty multidimensional array

Array ( [0] => Array ( ) [1] => Array ( ) )

Is it possible that UTF-8 support is not enabled on the PCRE library PHP is using? — cmbuckley, Nov 24 '13 at 19:50
possible duplicate of [preg\_match rule for utf-8](http://stackoverflow.com/questions/14511866/preg-match-rule-for-utf-8) — Fez Vrasta, Nov 24 '13 at 19:50
@FezVrasta - I read that, it did not help me a one bit, so it is hardly a duplicate. — August, Nov 24 '13 at 19:52
@August you should get a warning if UTF-8 support is not enabled; are you suppressing errors? — cmbuckley, Nov 24 '13 at 20:02
@cbuckley I've set error_reporting(E_ALL); at the beginning of the script - no errors. — August, Nov 24 '13 at 20:06
Try directly using the unicode properties. `preg_match_all('/\b(\p{L}{2}[a-z]\S+)/ui', $input, $output_array);` — hwnd, Nov 24 '13 at 20:39
@hwnd returns Array ( [0] => Array ( [0] => ešiolika ) [1] => Array ( [0] => ešiolika ) ) not sure why, because I do not understand regex that well — August, Nov 24 '13 at 20:48

jerdiggity · Accepted Answer · 2013-11-26T03:33:34.057

1

Try adding a UTF8 sequence to the beginning of the pattern:

$input = "žąsis su šešiolika žąsyčių";
preg_match_all("/(*UTF8)(žąs\S*)/iu", $input, $output_array);
print_r($output_array);

EDIT:

I tested this on PHP 5.2.17 and 5.3.20... I don't seem to have any problems while using 5.3.20 but I do get the same empty output while using 5.2.17. While I couldn't find any documentation that addressed why this happens, the problem seems to go away when removing the first \b (word boundary). Here's a screenshot with the output, PHP version, loaded extensions, and source code (if this doesn't help, make sure you're saving your documents in UTF8 instead of whatever Windows likes to save them as):

enter image description here

edited Nov 26 '13 at 03:33

answered Nov 24 '13 at 22:18

jerdiggity

3,655
1
29
41

Does not return anything at all. – August Nov 25 '13 at 21:14
I thought you said it did... No? Anyway, see latest revision for some (possible) debugging info if you want to compare your environment to mine. – jerdiggity Nov 26 '13 at 03:34
Nope, removing \b did the trick. Although I have no idea why :)) You are awesome jerediggity thanks! – August Nov 28 '13 at 15:34

A quick way to preg_match a utf-8 string

1 Answers1