preg_replace replace all instances

Spud_Nic

I'm almost positive I've seen this before on the forum, but haven't been able to find an answer with the search, so apologies if it has been asked before!

I have some code - see below - which I'm using to match a string pattern, then add some characters, and then replace the original in the text body.

However, the code I'm using only replaces the first instance (I have three in my text) of the pattern.

My code is:

		$WorklogEntryTitleExtract = preg_match("/<div class=\"wt\">.*<\/div wt>/s", $aExistingWorklogData[0], $aExistingTitle);

	if ($WorklogEntryTitleExtract !== 0)  {
		$aNewTitle[0] = $aExistingTitle[0]."\n----------------------------------------";
	}//end if

	$aExistingWorklogData[0] = preg_replace("/<div class=\"wt\">.*<\/div wt>/s", $aNewTitle[0], $aExistingWorklogData[0]);

Can anyone help. Cheers, and apologies again if I'm missing a obvious solution!

tsinka

Hi,

use preg_match_all. $aExistingTitle will contain all matches. Loop through the array and replace the values. The manual contains a good description of how the array will look like.

Alternatively use print_r($aExistingTitle) to display it.

Edit:

This one might to the trick, too:

$aExistingWorklogData[0] = preg_replace("/(<div class=\"\w\t\">.*?</div \w\t>)/","\1\n----------------------------------------"),$aExistingWorklogData[0]);

You might need to adjust the wt part, seems like there are some backslashes in your pattern but you have to type them twice if you want them to be displayed in your post.

Edit: PHPBuilder messes up the code ... put two backslashes in front of the 1 and remove the whitespace in front of ).

Thomas

Spud_Nic

thanks for your reply

I've tried to add preg_match_all into my code, but the array that it puts the results into only contains 1, when I use the count() function. When I print it to the screen, it contains all the text which is it meant to be parsing! I've tried with a simple regex, but the same thing happens.

I also tried that line of code which you wrote in your response, but didn't get me anywhere!!! :-(

I've only had limited experience working with preg_match before, so I maybe missing something really obvious. My whole code is:

$PagePath =  "index.php";
$OpenPage = @fopen($PagePath, "r"); 
$ContentsBuffer = fread($OpenPage, filesize($PagePath));	
if ($EntriesExtract == 0)  {
	$aExistingWorklogData[0] = "Error.";
} else {	
	$WorklogEntryTitleExtract = preg_match_all("/<div class=\"wt\">.*<\/div wt>/s", $aExistingWorklogData[0], $aExistingTitle);

print count($aExistingTitle);	
// where the loop would go!
//if ($WorklogEntryTitleExtract !== 0)  {
//$aNewTitle[0] = $aExistingTitle[0]."\n----------------------------------------";
//}//end if
//$aExistingWorklogData[0] = preg_replace("/<div class=\"wt\">.*<\/div wt>/s", $aNewTitle[0], $aExistingWorklogData[0]);

} //end if

fclose($OpenPage);	

print "<pre>".$aExistingWorklogData[0]."</pre>";
exit;

It's not particularly long, but there must be a reason why preg_match_all isn't working as it should. I've also had a look at the manual and can't see anything I'm missing from there!

So fruststrating! grateful for any help!

tsinka

Hi,

I'm not sure what exactly you're trying to do. You fopen the index.php file but nothing is done with the contents of index.php. $ContentsBuffer is never used.

Do you want to replace all occurences of
<div class="wt">text</div wt>
by
<div class="wt">text</div wt>\n------------------------
or
<div class="wt">text\n------------------------</div wt> ?

Where do $EntriesExtract and $aExistingWorklogData[0] come from ?

<div wt> doesn't look like a valid div closing tag.
Thomas

Spud_Nic

thanks for your reply

I've got it working now, after many hours of trying! It appears my original code was fine, except the regex wasn't making it behave properly. Still don't know why, but instead I've changed the regex I'm using, and its fine!

Yes, </div wt> is not a valid tag, so wouldn't pass an xhtml validator, but I'm using it as a unique identifier, so PHP knows where to start and stop looking! Thats all.

Anyway, thanks for your help!