question regarding php function preg_replace

Question

Hi, I want to dynamically remove specific tags and their content from an html file and thought of using preg_replace but can't get the syntax right. Basically it should, for example, do something like : Replace everything between (and including) "" by nothing.

Could anybody help me out on this please ?

Answer 1

I would suggest not trying to do this with a regular expression. A safer approach would be to use something like

Simple HTML DOM

Here is the link to the API Reference: Simple HTML DOM API Reference

Another option would be to use DOMDocument

The idea here is to use a real HTML parser to parse the data and then you can move/traverse through the tree and remove whichever elements/attributes/text you need to. This is a much cleaner approach than trying to use a regular expression to replace data within the HTML.

<?php
    $doc = new DOMDocument;
    $doc->loadHTMLFile('blah.html');

    $content       = $doc->documentElement;
    $table         = $content->getElementsByTagName('table')->item(0);
    $delfirstTable = $content->removeChild($table);

    echo $doc->saveHTML();
?>

Answer 2

If you are trying to sanitize your data, it is often recommended that you use a whitelist as opposed to blacklisting certain terms and tags. This is easier to sanitize and prevent XSS attacks. There's a well known library called HTML Purifier that, although large and somewhat slow, has amazing results regarding purifying your data.

Answer 3

PSEUDO CODE

function replaceMe($html_you_want_to_replace,$html_dom) {
   return preg_replace(/^$html_you_want_to_replace/, '', $html_dom);
}

HTML Before

<div>I'm Here</div><div>I'm next</div>

<?php
$html_dom = "<div>I'm Here</div><div>I'm next</div>";
$get_rid_of = "<div>I'm Here</div>";
replaceMe($get_rid_of);
?>

HTML After

<div>I'm next</div>

I know it's a hack job

Answer 4

If you don't know what is between the tags, Phill's response won't work.

This will work if there's no other tags in between, and is definitely the easier case. You can replace the div with whatever tag you need, obviously.

preg_replace('#<div>[^<]+</div>#','',$html);

If there could be other tags in the middle, this should work, but could cause problems. You're probably better off going with the DOM solution above, if so

preg_replace('#<div>.+</div>#','',$html);

These aren't tested

Answer 5

Easy dude.

To have a Ungreedy regexpr, use the U modifier And to make it multiline, use the s modifier. Knowing that, to remove all paragraphes use this pattern :

#<p[^>]*>(.*)?</p>#sU

Explain :

• I use # delimiter to not have to protect my \ characters (to have a more readable pattern)
• <p[^>]*> : part detecting an opening paragraph (with a hypothetic style, such as )
• (.*)? : Everything (in "Ungreedy mode")
• </p> : Obviously, the closing paragraph

Hope that help !