How to replace text over multiple lines using preg_replace

Question

Hi have the following content within an html page that stretches multiple lines

                

        
                      

              相关标签:
       
      

Answer 1

you can also use [\s\S] instead of . combined with the DOTALL flag s for matching everyting because [\s\S] means exactly the same: match everything; \s matches all space-characters (including newline) and \S machtes everything that is not a space-character (i.e. everything else). in some cases/implementations of regular expressions, this works better than enabling DOTALL

caution: .* with the flag for DOTALL as well as [\s\S] are both "hungry" and won't stop reading the string. if you want them to stop at a certain position, (e.g. the first </div>), use the non-greedy operator ? behind your quantifier, e.g. .*?

Answer 2

It is possible to use regex to strip out chunks of html data, but you need to wrap the html with custom html tags which get ignored by browsers. For example:

<?php
$html='
<div>This will be shown</div>
<custom650 rel="nofollow">
  <p class="subformedit">
    <a href="#" class="mylink">Link</a>
    <div class="morestuff">
      ... more html in here ...
    </div>
  </p>
</custom650>
<div>This will also be shown</div>
';

To strip the tags with the rel="nofollow" attributes, you can use the following regex:

$newhtml = preg_replace('/<([^\s]+)[^>]*rel="nofollow"[^>]*>.*?<\/\1>/si', '', $html);

From experience, start the custom tags on a new line. Undoubtedly a hack, but might help someone.

Answer 3

it is the "s" flag, it enables . to capture newlines

Answer 4

If this weren't HTML, I'd tell you to use the DOTALL modifier to change the meaning of . from 'match everything except new line' to 'match everything':

preg_replace('/(.*)<\/div>/s','abc',$body);

But this is HTML, so use an HTML parser instead.