regex to replace all trems but what is wraped in a no replace phase

@Ray you got it backward  so.. from  

<ul>
<!--glossover-start-->  <li><a href="#pci">What is PCI?</a></li>  <li><a href="#dss">What is the 'PCI DSS'?</a></li><!--glossover-end-->
  <li><a href="#apply">To whom does PCI apply?</a></li>
  <li><a href="#refuse">What if a Merchant refuses to cooperate?</a></li>
  <li><a href="#okay">What does a Merchant have to do in order to satisfy the PCI requirements?</a></li>
  <li><a href="#saq1">What is the 'Self-Assessment Questionnaire'?</a></li>
  <li><a href="#saq2">What is the 'SAQ'?</a></li>
  <li><a href="#compliance">What is meant by 'Compliance'?</a></li>
  <li><a href="#validation">What is meant by 'Validation'?</a></li>
  <li><a href="#levels">How are the different Merchant Levels defined?</a></li>
  <li><a href="#remediation">What is meant by 'Remediation'?</a></li>
  <li><a href="#government">Is PCI a government program? Is it a law?</a></li>
  <li><a href="#panoptic">Are Merchants required to use the tools provided through this web-site to fix any compliance problems?</a></li>
  <li><a href="#Internet">If a merchant does not have Internet access what options do they have to complete the SAQ forms?</a></li>
  <li><a href="#printRemed">How do I print my Fix-It Plan?</a></li>
  <li><a href="#glossary">Glossary of Technical Terms</a></li>
  <li><a href="#submitQuestion">Submit a Question</a></li>
</ul>

Select allOpen in new window

where I look to replace PCI  with

<span class="gloss"><strong>$term_link</strong></span>

which would give me

<ul>
<!--glossover-start-->  <li><a href="#pci">What is PCI?</a></li>  <li><a href="#dss">What is the 'PCI DSS'?</a></li><!--glossover-end-->
  <li><a href="#apply">To whom does <span class="gloss"><strong>PCI</strong></span> apply?</a></li>
  <li><a href="#refuse">What if a Merchant refuses to cooperate?</a></li>
  <li><a href="#okay">What does a Merchant have to do in order to satisfy the <span class="gloss"><strong>PCI</strong></span> requirements?</a></li>
  <li><a href="#saq1">What is the 'Self-Assessment Questionnaire'?</a></li>
  <li><a href="#saq2">What is the 'SAQ'?</a></li>
  <li><a href="#compliance">What is meant by 'Compliance'?</a></li>
  <li><a href="#validation">What is meant by 'Validation'?</a></li>
  <li><a href="#levels">How are the different Merchant Levels defined?</a></li>
  <li><a href="#remediation">What is meant by 'Remediation'?</a></li>
  <li><a href="#government">Is <span class="gloss"><strong>PCI</strong></span> a government program? Is it a law?</a></li>
  <li><a href="#panoptic">Are Merchants required to use the tools provided through this web-site to fix any compliance problems?</a></li>
  <li><a href="#Internet">If a merchant does not have Internet access what options do they have to complete the SAQ forms?</a></li>
  <li><a href="#printRemed">How do I print my Fix-It Plan?</a></li>
  <li><a href="#glossary">Glossary of Technical Terms</a></li>
  <li><a href="#submitQuestion">Submit a Question</a></li>
</ul>

Select allOpen in new window

So that anyything in the <!--glossover-start--> <!--glossover-end--> area doesn't get replaced.


@kaufmed isn't [^glossover\-start] saying any characters except those listed?  ie: [^notWantedPhrase]



Please let me know if there is any thing else I can add to clear up the need.  thank you.. Cheers -Jeremy

So here is what I tried with

<?php
//if (!isset(cmsms())) exit;
global $CMS_ADMIN_PAGE;
//print($CMS_ADMIN_PAGE.' admin');

if($CMS_ADMIN_PAGE!=1){
	if (substr($_GET['page'], 0,8)!="glossary"){
		$content=$params["content"];
		global $gCms;
		$config = $gCms->GetConfig();
		$tmp_file=$config['root_path'].'/tmp//templates_c/gloss.tmp';
				if(!file_exists($tmp_file)){
					$db = $gCms->db;
					$query = 'SELECT * FROM '.cms_db_prefix().'module_glossary_term';
					$result=$db->Execute($query,array());
					$resultsarray=array();
					while ($result && !$result->EOF){
						$term=$result->fields['term'];
						$returnid='69';
						$id='asdga';
						$def = $result->fields['definition'];
						$prettyurl = 'glossary/'.$term.'/'.$returnid;
						$term_link = $this->CreateFrontendLink($id, 'show', $returnid, $term,array('tid'=>$result->fields['term_id']), '', false, false, 'rel="'.$def.'"', false,$prettyurl);
						$resultsarray[]=array('term'=>$term,'term_link'=>$term_link);
						$result->MoveNext();
					}
					$fn = $tmp_file;
					$fp = fopen($fn, "w");
					if (!$fp) {
						return 0;
					}
					fwrite($fp, serialize($resultsarray));
					fclose($fp);
					$data=file_get_contents($tmp_file);
				}else{
					$data=file_get_contents($tmp_file);
				}
				//debug_display('$data---->'.$data);
			$result=unserialize($data);
		
			foreach($result as $key => $item){
				$term=$item['term'];
				$term_link=$item['term_link'];
				//$pattern = '/()'.$term.'/';
				//$replacement='<span class="gloss"><strong>'.$term_link .'</strong></span>';
				//debug_display('$replacement---->'.$replacement);
				//$content = preg_replace($pattern, $replacement, $content); 
				
				
				
				
				
				/*$content = preg_replace("|(?<![^glossover\-start])(?!<[^<>]*?)(?<![?./&])\b$term\b(?!:)(?![^<>]*?>)(?![^glossover\-end])|imsU","<span class=\"gloss\"><strong>$term_link</strong></span>" , $content);*/
	$gloss = '#<!--glossover-start-->.*?<!--glossover-end-->#';		// Pattern for a glossy section
	$term_link = "<span class=\"gloss\"><strong>$term_link</strong></span>";
	
	if (preg_match_all($gloss, $content, $glossed))	{			// Find all glossy sections
		$content = preg_replace($gloss, $temp_replace, $content);	// Replace glossy sections with temporary placeholder
		$content = str_replace($term, $term_link, $content);		// Replace non-glossy PCI occurrences

		preg_match_all("/$temp_replace/", $content, $temps);		// Find all placeholders
		$content = str_replace($temps[0], $glossed[0], $content);	// Replace placeholders with original values.
	}

	//echo $content;

				  // Acronyms wrapped in dollar signs will just be unwrapped: 
				  // So: $AOL$ will become AOL, without the <acronym title="America Online">AOL</acronym> 
      			$content = preg_replace("|[$]<span class=\"gloss\"><strong>$term_link</strong></span>[$]|imsU" , "$term" , $content);
				
				
				
			}
			$params["content"]=trim($content);
	}
}

Select allOpen in new window

where $term included PCI with its it's link in $term_link .  


Also I thought preg_replace what fast then preg_match_all as there is not an array created and such, or am I off on that?

$content = preg_replace("|(?!<[^<>]*?)(?<![?./&])\b$term\b(?!:)(?![^<>]*?>)|imsU","<span class=\"gloss\"><strong>$term_link</strong></span>" , $content);

That does match the $term, so I thought I just needed to addjust the grouping and put the pattern be tween a Not followed by and Not following  group. .. or that was the thought :)

tk for the help.. Cheers -Jeremy

it's the working anwser build of the rest.  tk all for the help

Wow!  Haven't seen you in a while!  Welcome back, ~Ray