Hey all,

Before I start, I just want to say that below you will find a modifier of "U".  I know a lot of people say this should not be used, so if there is a way that my issue can be resolved where the U modifier won't be needed, great!

 Language: PHP (4)

Users submit content via a wysiwyg editor we provide.  This means we allow html markup.  I perform cleanup on what the provide (sanity type stuff) and then I must split their content into multiple pages every x number of words (I have this covered).

Anyways, I am using a regexp to capture all html blocks so that I get the content ("display" words) and its wrapped html markup.

To elaborate, here is what is desired:

Original:

 <p style='TEXT-ALIGN: center'><strong><span style='TEXT-DECORATION: underline'><span style='FONT-FAMILY: Times New Roman' size='3;'>Strange and Unexplained Happenings</span></span></strong></p> <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Author: ZeusFluff.</span></strong></p> <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Date Started: 7/14/06. Date Finished: 7/14/06.</span></strong></p>

[0] => <p style='TEXT-ALIGN: center'><strong><span style='TEXT-DECORATION: underline'><span style='FONT-FAMILY: Times New Roman' size='3;'>Strange and Unexplained Happenings</span></span></strong></p>
[1] => <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Author: ZeusFluff.</span></strong></p>
[2] => <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Date Started: 7/14/06. Date Finished: 7/14/06.</span></strong></p>

 <p style='TEXT-ALIGN: center'><strong><span style='TEXT-DECORATION: underline'><span style='FONT-FAMILY: Times New Roman' size='3;'>Strange and Unexplained Happenings</span></span></strong></p> <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Author: ZeusFluff.</span></strong></p> <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Date Started: 7/14/06. Date Finished: 7/14/06.</span></strong></p>

preg_match_all("/<[^\/]?(\w+)((\s+\w+(\s*=\s*(?:\".*?\"|'.*?'|[^'\">\s]+))?)+\s*|\s*)\/?>(.*)<\/\\1>/iusU" ,$html_markup,$arry,PREG_PATTERN_ORDER);

 <p style='TEXT-ALIGN: center'><strong><span style='TEXT-DECORATION: underline'><span style='FONT-FAMILY: Times New Roman' size='3;'>Strange and Unexplained Happenings</span></span></strong></p> <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Author: ZeusFluff.</span></strong></p> <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Date Started: 7/14/06. Date Finished: 7/14/06.</span></strong></p>

preg_match_all("/<[^\/](.+)>(.*)<\/(.+)>/iusU",$html_markup,$arry,PREG_PATTERN_ORDER);

How do you plan to handle nested tags?  Also, it would seem that your intended matches are just <p>...</p> blocks, is that the case or is it just the example?

Raw Match Pattern:
<p[^>]*>.*?</p>

PHP Code Example:
<?php
preg_match_all('~<p[^>]*>.*?</p>~s',$sourcestring,$matches);
echo print_r($matches,true);
?>
$matches array:
(
    [0] => Array
        (
            [0] => <p style='TEXT-ALIGN: center'><strong><span style='TEXT-DECORATION: underline'><span style='FONT-FAMILY: Times New Roman' size='3;'>Strange and Unexplained Happenings</span></span></strong></p>
            [1] => <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Author: ZeusFluff.</span></strong></p>
            [2] => <p><strong><span style='FONT-FAMILY: Times New Roman' size= '3;'>Date Started: 7/14/06. Date Finished: 7/14/06.</span></strong></p>
        )

)

Hi,

 Thank you for the reply.  This works almost 100%!  But, yes, I do need to support nested tags.  For instance, if someone has something like this:

<p style='MARGIN: 0in 1in 0pt; TEXT-ALIGN: justify'>“Ok I say I&#39;m not gonna hurt you and then you go and do something like that,” Noel said pulling the blood stained dagger out of his chest, “it&#39;s just not sane. Ok I&#39;m gonna put him down and maybe we can talk about this.” Noel said calmly. <strong><em>Slowly lessening his grip on Allan&#39;s throat he let him drop to the forest floor. Allan sweat drenched and fearing he might need a new change of under garments soon looked from Noel to the blood stained dagger lying not three feet away.</em></strong> <span style='FONT-WEIGHT: normal; COLOR: red'><em>He grabbed it and lunged a third time, this time though he hit nothing but the tree to the other side of Noel.</em></span></p>

Alright!  Very close here.

Here is my next challenge and for the life of me, I can't get this next thing to work right.

<p class="classname" style='somestyle' id='someid'>It felt like we were dancing around, trying to figure each other out, and she was looking to my reaction and decision to tell her <em>something here</em>. I had no idea what I was doing. I felt like I was in a foreign country, and I didn’t know the language. Why the hell did I care that I passed her test? But I did. I most definitely did.</p>

I'm trying to break this up using regexp so that I get a structure like:

[0]
     [0] => <p class="classname" style='somestyle' id='someid'>
     [1] => <em>
[1]
     [0]=> It felt like we were dancing around, trying to figure each other out, and she was looking to my reaction and decision to tell her <em>something here</em>. I had no idea what I was doing. I felt like I was in a foreign country, and I didn’t know the language. Why the hell did I care that I passed her test? But I did. I most definitely did.
     [1]=> something here
[2]
     [0]=></p>
     [1]=></em>

or something like that.

I fear given the simplicity of your original sample and my understanding of your goal from the original question, that's far from very close.

Please be specific about what exactly you want to accomplish, do you want to strip tags, find some split points then reassemble with tags?

Also, please please explain your latest sample matches further.  Thanks.

Hi ddrudik,

I apologize for that.  Let me elaborate further on my original post.

We have a site where we allow users to submit content.  This content is in html (we provide a wysiwyg) and many people copy and paste from MS Word.

So, I first go through the process of cleaning up what they submit so that what I have left is the "cleansed" html (I use tidy and some other logic to clean up the source).  By the time the cleaning process is complete, I'm left with a "workable" set of html.

I then need to split the content up into multiple pages.  We define a page as having N number of words.  So, this means that I have to be able to split apart the html, loop through the words until I hit N, end that page (by closing up the markup with the appropriate closing tags, etc), then start a new page and make sure that the starting tags are a continuation of the tags that the content was in when I split it.

Here is an example: http://fanfictiondev.myfandoms.com/index.php/fanfiction/show/317?site_id=13

But the problem with that example is prevalent even on the first page.  You'll notice that the paragraph, " Excellent. I’m so relieved to hear it wasn’t a date that I was suddenly myself and instead of trying to figure out what he hell was happening, I acted like myself. I kissed her good night. You know…straight girl kissing good night. Not that I ever thought of it that way before, but I guess that’s ....." should actually have only the word Excellent. as being italicized.

A further example of what I'm needing to do:

CONTENT:

While I understand your goal I don't readily see a quick solution to the issue.