Question

$text = "<p>keep me</p> <p>strip me &nbsp;</p>
$pattern = "/<[^\/>]*>(\&nbsp\;)*<\/[^>]*>/"; 
$text =  preg_replace($pattern, '', $text);

Hi, I need to strip “quasi-empty” p tags from a html string. there’s always only a   as a trigger in the p element. how can I strip it with regex?

Answer 1

The following pattern will match all <p> </p> blocks that include   along with any accompanying text, as per your example.

$text = "<p>keep me</p> <p>strip me &nbsp;</p>";
$pattern = "/<p>[^<]*&nbsp\;[^<]*<\/p>/"; 
$output =  preg_replace($pattern, '', $text);

If you actually want it to only strip out <p> </p> blocks with   and spaces, use the following pattern instead:

$pattern = "/<p>(\s*&nbsp\;\s*)+<\/p>/"; 

If you want to only strip out <p> </p> blocks that have an   and up to a certain number of characters, use the following (setting the $maxChars variables as you see fit):

$maxCharsBefore = 10;
$maxCharsAfter = 10;
$pattern = "/<p>[^<]{0,".$maxCharsBefore."}&nbsp\;[^<]{0,".$maxCharsAfter."}<\/p>/";