[SOLVED] Regular expression help needed

[TeNDoLLA]

I have some predefined HTML where I want to match <tr></tr> patterns and stuff between them and replace it with newline depending on the value of the $tag. Anyone can show me the proper pattern for it? Here is some example html and the php code for the pattern that I have tried.

 

<tr>
<td>[:some_tag:]</td>
<td> some text </td>
</tr>
<tr>
<td> [:another_tag:] </td>
<td> another text </td>
</tr>

 

<?php
$pregSearch[] = '/(\r?\n)<tr>\r?\n<td>.*\[:'. $tag .':\].*<\/td>\r?\n<td>.*<\/td>\r?\n<\/tr>\r?\n/';
$pregReplace[] = '$1'; 

[nrg_alpha]

One possible solution:

 

Example:

$html = <<<EOD
<tr>
<td>[:some_tag:]</td>
<td> some text </td>
</tr>
<tr>
<td> [:another_tag:] </td>
<td> another text </td>
</tr>
EOD;

$tagName = 'some_tag';
$html = preg_replace('#<tr>\R*<td>\[:'. $tagName .':\]</td>.*?</tr>#is', "\n", $html);
echo $html;

 

This of course makes some rigid assumptions (going by the sample you have listed). If there are tr tags with attributes, simply change <tr> to <tr[^>]*> in the pattern.. same goes for td if need be. This also assumes there is only terminating whitespace characters between <tr> and <td>.. if not, simply change \R* to .*?

[TeNDoLLA]

No go with this either. And don't really know why. There is newlines after tr and some td tags. And I think the 's' modifier ignores these and removes just everything (should remove) matching this pattern but there will be problem with using the 's' modifier because I don't want to remove all these patterns. Only the ones that has NULL as value for $tagName.

[nrg_alpha]

There is newlines after tr and some td tags.

 

Would changing \R* to [\r\n]* help?

 

And I think the 's' modifier ignores these and removes just everything (should remove) matching this pattern but there will be problem with using the 's' modifier because I don't want to remove all these patterns. Only the ones that has NULL as value for $tagName.

 

I think you misunderstand the s modifier. This only applies to the dot in .*? for example. Without that modifier, the dot will stop matching once it hits a newline (as by default, this is the one thing it doesn't match). And this will break the pattern before it can completely finish what needs to be matched.

[TeNDoLLA]

Thanks for your replies and help! I got the problem solved.. was kind of more like stupid user error I had the matching stuff in wrong order compared to the actual html code.

[nrg_alpha]

Glad to hear... Please don't forget to flag this thread as solved.