ID: 34947
Updated by: [EMAIL PROTECTED]
Reported By: cold_candor at hotmail dot com
-Status: Open
+Status: Bogus
Bug Type: PCRE related
Operating System: windows XP Pro
PHP Version: 5.0.5
New Comment:
* stands for "0 or more", so "\s*(.*?)\s*" matches "" (empty string)
too.
Not PHP problem.
Previous Comments:
------------------------------------------------------------------------
[2005-10-21 16:41:36] cold_candor at hotmail dot com
Description:
------------
I can't imagine any reason why this should happen, and all the
documentation I've read says it shouldn't, so here you go:
If, when using the preg_replace function, I attempt to match against an
entire line to remove leading and ending whitespace (using ^ and $ but
no modifiers, like D or m), the function does what I'd expect.
However, if I neglect to use the leading ^, the function still matches
what I'd expect it to, but the replacement string is entered twice!
Even if I don't use the backreference in the replacement, all
non-reference related characters are still repeated!
If I split the replacement to deal with leading and ending whitespace
seperately, it always repeats the replacement string!
I did not configure anything special when I downloaded PHP, I simply
grabbed the windows zip file provided on the website (www.php.net),
opened it, added the path to my environment variables, and started
using it.
I know of nothing special about my setup.
I have made no changes to the PHP.ini file
If it turns out that this is not a bug, don't tell me to use the damn
support page, everything on there I either can't do or have tried with
no result or have no entries relating to my problem or (what I really
want) have no way to actually ask a question. Since you have to figure
it out anyway, just tell me what went wrong!
Reproduce code:
---------------
<?php
$text = " a b c ";
$newText = preg_replace("/\s*(.*?)\s*$/", "$1\n", $text);
for($i = 0; $i < strlen($newText); $i++) {
echo ord($newText[$i]), '~';
} // End for loop
echo "\n$newText";
$newText = preg_replace("/\s*(.*?)\s*$/", " $1\n gh ", $text);
for($i = 0; $i < strlen($newText); $i++) {
echo ord($newText[$i]), '~';
} // End for loop
echo "\n$newText";
$newText = preg_replace("/^\s*/", "", $text);
$newText = preg_replace("/\s*$/", "\n", $newText);
for($i = 0; $i < strlen($newText); $i++) {
echo ord($newText[$i]), '~';
} // End for loop
echo "\n$newText";
?>
Expected result:
----------------
The following three outputs should be produced (labels added for
readability):
ASCII values: 97~32~98~32~99~10~
Viewed output: "a b c
"
ASCII values: 32~32~97~32~98~32~99~10~32~103~104~32
Viewed output: " a b c
gh "
ASCII values: 97~32~98~32~99~10~
Viewed output: "a b c
"
Actual result:
--------------
The following three outputs are what was actually produced (labels
added for readability):
ASCII values: 97~32~98~32~99~10~10~
Viewed output: "a b c
"
ASCII values:
32~32~97~32~98~32~99~10~32~103~104~32~32~32~10~32~103~104~32~
Viewed output: " a b c
gh
gh"
ASCII values: 97~32~98~32~99~10~10~
Viewed output: "a b c
"
------------------------------------------------------------------------
--
Edit this bug report at http://bugs.php.net/?id=34947&edit=1
