How to remove entire div with preg_replace

Question

Ok, as it is WordPress problem and it sadly goes a little deeper, I need to remove each representation of parent div and its inside:

<div class="sometestclass">
   <img ....>
   <div>.....</div>
   any other html tags
</div><!-- END: .sometestclass -->

The only idea I have is to match everything that starts with:

<div class="sometestclass">

and ends with:

<!-- END: .sometestclass -->

with all that is between (I can tag the end of parent div anyway I want, this is just a sample). Anybody have an idea how to do it with:

<?php $content = preg_replace('?????','',$content); ?>

[Don't parse HTML with regex, nanananana.](http://stackoverflow.com/a/1732454/333786) — Jakub Hampl, May 18 '12 at 19:42

Sampson · Accepted Answer · 2012-05-18T20:18:30.613

9

I wouldn't use a regular expression. Instead, I would use the DOMDocument class. Just find all of the div elements with that class, and remove them from their parent(s):

$html = "<p>Hello World</p>
         <div class='sometestclass'>
           <img src='foo.png'/>
           <div>Bar</div>
         </div>";

$dom = new DOMDocument;
$dom->loadHTML( $html );

$xpath = new DOMXPath( $dom );
$pDivs = $xpath->query(".//div[@class='sometestclass']");

foreach ( $pDivs as $div ) {
  $div->parentNode->removeChild( $div );
}

echo preg_replace( "/.*<body>(.*)<\/body>.*/s", "$1", $dom->saveHTML() );

Which results in:

<p>Hello World</p>

edited May 18 '12 at 20:18

answered May 18 '12 at 19:42

Sampson

265,109
74
539
565

Sadly I have no idea about it :(. Could You explain with some kind of sample? – Marcin Bobowski May 18 '12 at 19:45
Instead of cycling over all divs, he could use an XPath query, to directly find divs with the right class. – Pascal MARTIN May 18 '12 at 19:45
@JonathanSampson this example gives me an idea about how to do it, but it depends on a parent (not named) div. But what, if I would not have that parent div, and needed to finish with only
Hello World
? – Marcin Bobowski May 18 '12 at 20:01
so the input html is:
Hello World

Bar

– Marcin Bobowski May 18 '12 at 20:02
Ok, I get it, after few tries it almost works. Only problem is with encoding - it does something strange with utf-8 encoded local language characters :( – Marcin Bobowski May 18 '12 at 20:10
@MarcinBobowski You can set encoding in the constructor: `$dom = new DOMDocument('1.0', 'iso-8859-1');` – Sampson May 18 '12 at 20:14
@JonathanSampson - already tryied, setting utf-8, iso-8859-1, iso-8859-2 doesnt help. My country special chars like ąęśćź changes to iÄ™ Å‚ Å¼ etc. – Marcin Bobowski May 18 '12 at 20:21
@JonathanSampson. Thanks for this good answer. What is the `$1`? Id there another way of saving back, without using preg_replace? – Peyman Mohamadpour Jun 01 '14 at 15:10
1

@Arash The `$1` refers to the `(.*)` in the regular expression. – Sampson Jun 01 '14 at 21:27
@JonathanSampson Thanks for response. could you please translate the preg_replace line in english? – Peyman Mohamadpour Jun 02 '14 at 12:40

Blake · Answer 2 · 2012-06-14T00:35:58.810

6

<?php $content = preg_replace('/<div class="sometestclass">.*?<\/div><!-- END: .sometestclass -->/s','',$content); ?>

My RegEx is a bit rusty, but I think this should work. Do note that, as others have said, RegEx is not properly equipped to handle some of the complexities of HTML.

In addition, this pattern won't find embedded div elements with the class sometestclass. You would need recursion for that.

edited Jun 14 '12 at 00:35

answered May 18 '12 at 20:33

Blake

1,691
2
15
23

This might work, but already I have used DOMDocument method. But thanks for RegEx, its more easy to learn on examples :) – Marcin Bobowski May 18 '12 at 20:59

score 0 · Answer 3 · answered May 18 '12 at 19:44

0

How about just some CSS .sometestclass{display: none;} ?

answered May 18 '12 at 19:44

Ryan

650
6
14

1

Hiding html element is not good idea as per SEO Experts, many prefer removing elements to save html size. – Nisha Ankit Mar 27 '22 at 16:48

score 0 · Answer 4 · answered Jul 31 '13 at 12:40

For the UTF-8 issue, I found a hack at the PHP-manual

So my functions looks as follows:

function rem_fi_cat() {
/* This function removes images from _within_ the article.
 * If these images are enclosed in a "wp-caption" div-tag.
 * If the articles are post formatted as "image".
 * Only on home-page, front-page an in category/archive-pages.
 */
if ( (is_home() || is_front_page() || is_category()) && has_post_format( 'image' ) ) {
    $document = new DOMDocument();
    $content = get_the_content( '', true );
    if( '' != $content ) {
        /* incl. UTF-8 "hack" as described at 
         * http://www.php.net/manual/en/domdocument.loadhtml.php#95251
         */
        $document->loadHTML( '<?xml encoding="UTF-8">' . $content );
        foreach ($doc->childNodes as $item) {
            if ($item->nodeType == XML_PI_NODE) {
                $doc->removeChild($item); // remove hack
                $doc->encoding = 'UTF-8'; // insert proper
            }
        }
        $xpath = new DOMXPath( $document );
        $pDivs = $xpath->query(".//div[@class='wp-caption']");

        foreach ( $pDivs as $div ) {
            $div->parentNode->removeChild( $div );
        }

        echo preg_replace( "/.*<div class=\"entry-container\">(.*)<\/div>.*/s", "$1", $document->saveHTML() );

    }
}

}

How to remove entire div with preg_replace

4 Answers4

Linked