#37794 [Bgs]: preg_split doesn't work as it should be with \W on utf-8 string
| From: | jdespatis at yahoo dot fr | Date: | Wed, 14 Jun 2006 08:44:04 +0000 |
| Subject: | #37794 [Bgs]: preg_split doesn't work as it should be with \W on utf-8 string | ||
| References: | 1 | Groups: | php.bugs |
| Request: | Send a blank email to php-bugs+get-98339@lists.php.net to get a copy of this message | ||
ID: 37794
User updated by: jdespatis at yahoo dot fr
Reported By: jdespatis at yahoo dot fr
Status: Bogus
Bug Type: PCRE related
Operating System: Linux 2.6.15 Debian Testing
PHP Version: 5.1.4
New Comment:
Ok.
However i've read again the documentation
http://fr.php.net/manual/en/reference.pcre.pattern.syntax.php
And i don't see it's explicitely said "in utf-8 mode don't use \w"
i can only see: "Since PHP 4.4.0 and 5.1.0, three additional escape
sequences to match generic character types are available when UTF-8
mode is selected. "
So, a reader understand this as: \w works AND in utf8 i have also \p{}
Would it be possible to update the documentation ? (for example, now, i
have a doubt on \d, is it working on utf8 ?, i dunno...)
One thing more: i've found that ucwords() and ucfirst() are not utf8
aware, the documentation should be updated i think
Thanks
Previous Comments:
------------------------------------------------------------------------
[2006-06-13 21:08:46] nlopess@php.net
sorry, my last comment is incorrect. in utf mode you should use the
property escapes (\p{..}), instead of non utf8-aware escapes, like \W.
------------------------------------------------------------------------
[2006-06-13 18:35:27] nlopess@php.net
/\W/ means match any non-whitespace. you probably want to use \w (lower
case)
------------------------------------------------------------------------
[2006-06-13 11:53:50] jdespatis at yahoo dot fr
Description:
------------
preg_split("/\W/u", $utf8_string) cuts the words !
Reproduce code:
---------------
print_r(preg_split("/(\W)/u", "этот",
-1,
PREG_SPLIT_DELIM_CAPTURE));
(watch out, i've put an utf8 string (you need to translate the html
code into utf8), it's a russian string, (when you see the characters,
you can see etot, with e being an epsilon inverted)
For now, i succeed in making my code work by using:
\P{L} instead of \W
Expected result:
----------------
Array
(
[0] => этот
)
Actual result:
--------------
Array
(
[0] =>
[1] => э
[2] =>
[3] => т
[4] =>
[5] => о
[6] =>
[7] => т
[8] =>
)
------------------------------------------------------------------------
--
Edit this bug report at http://bugs.php.net/?id=37794&edit=1