How to remove link with preg_replace();?

Question

I'm not sure how to explain this, so I'll show it on my code.

<a href="link.php">First</a> and 
<a href="link.php" class="delete">Second</a> and 
<a href="link.php">Third</a>

how can I delete opening <a href="link.php" class="delete"> and closing </a> but not the rest?

I'm asking for preg_replace(); and I'm not looking for DomDocument or others methods to do it. I just want to see example on preg_replace();

how is it achievable?

provided there is no more HTML inside the `..` element, then it's relatively simple. If there is (or if there could be) any sub elements, then it becomes much more difficult. — Spudley, Jul 21 '11 at 15:04
Okay, @Tomalak, show me example with ANY method to achive this — genesis, Jul 21 '11 at 15:06
Are you trying to disable links to the current page? If you explain what you *actually* want, you will get better solutions than asking how to do what you already decided was best. I'm sure there is some reason for this, and it can be achieved without a regular expression. — Wesley Murch, Jul 21 '11 at 15:07
I can understand using `preg_replace()` for such a trivial thing. But it is not the appropriate solution. Albeit unlikely, if there were nested anchor tag this would fail. That's one of the reasons Regular Expressions — Jason McCreary, Jul 21 '11 at 15:08
@Tomalak Geret'kal: Comment was targeted at the other Tomalak. I just edited the comment to clarify. — webbiedave, Jul 21 '11 at 15:11
@Tomalak: Can you tell me, what's wrong with preg_replacing html? Is there some real answer WHY is it bad idea? — genesis, Jul 21 '11 at 15:12
@webbiedave: Thank you for support. However I have to tell that this is very begginer question though. I have to think about it little bit more before asking new question. Sure I know solution for this, but someone had to remind me :/ — genesis, Jul 21 '11 at 15:13
@genesis: http://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags/1732454#1732454 — Lightness Races in Orbit, Jul 21 '11 at 15:15
@genesis: The answer is in my first comment, really. I'm not berating you, I'm just being direct, so no offense. *"How can I parse/modify HTML with regex and why is it such a bad idea?"* is a topic that has been discussed on SO *to no end* and I won't start repeating the arguments here. — Tomalak, Jul 21 '11 at 15:17
@Tomalak:Okay. I knew answer for my own question for sure, but I didn't about such an easy answer like hakre given. — genesis, Jul 21 '11 at 15:19

hakre · Accepted Answer · 2011-07-21T15:10:45.763

7

Only pick the groups you want to preserve:

$pattern = '~(<a href="[^"]*" class="delete">)([^<]*)(</a>)~';
//                   1                           2      3
$result = preg_replace($pattern, '$2', $subject);

You find more examples on the preg_replace manual page.

edited Jul 21 '11 at 15:10

answered Jul 21 '11 at 15:05

hakre

193,403
52
435
836

@genesis: You wanted to delete links with class `"delete"`? No? This code does the opposite. – Lightness Races in Orbit Jul 21 '11 at 15:09
@hakre: [Oh, really](http://www.ideone.com/oGoBM)? I suggest you try out your code before you post it. – Lightness Races in Orbit Jul 21 '11 at 15:11
@Tomalak: I've read the question that the text should be preserved, but the link removed. Naturally this only works on strings that match the regex, so no parser here. – hakre Jul 21 '11 at 15:12
@Tomalak: Was just using XMLReader for parsing [lately for an answer](http://stackoverflow.com/questions/6775785/how-can-i-edit-my-code-to-echo-the-data-of-childs-element-where-my-search-term-w/6777403#6777403). – hakre Jul 21 '11 at 15:17

Tomalak · Answer 2 · 2011-07-21T15:52:11.117

5

Since you asked me in the comments to show any method of doing this, here it is.

$html =<<<HTML
<a href="link.php">First</a> and 
<a href="link.php" class="delete">Second</a> and 
<a href="link.php">Third</a>
HTML;

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

$xpath = new DOMXPath($dom);

$elems = $xpath->query("//a[@class='delete']");
foreach ($elems as $elem) {
    $elem->parentNode->removeChild($elem);
}

echo $dom->saveHTML();

Note that saveHTML() saves a complete document even if you only parsed a fragment.

As of PHP 5.3.6 you can add a $node parameter to specify the fragment it should return - something like $xpath->query("/*/body")[0] would work.

edited Jul 21 '11 at 15:52

answered Jul 21 '11 at 15:43

Tomalak

332,285
67
532
628

Which 5.3.6 PHP you use for `$xpath->query("/*/body")[0]` ? Not 5.4? – hakre Jul 22 '11 at 01:50
@hakre: Not sure what you mean - what has the PHP version to do with an XPath expression? – Tomalak Jul 22 '11 at 08:20

score 4 · Answer 3 · answered Mar 23 '13 at 07:58

4

$pattern = '/<a (.*?)href=[\"\'](.*?)\/\/(.*?)[\"\'](.*?)>(.*?)<\/a>/i';
$new_content = preg_replace($pattern, '$5', $content);

answered Mar 23 '13 at 07:58

Devan

117
5

guido · Answer 4 · 2011-07-21T15:33:54.520

$pattern = '/<a[^<>]*?class="delete"[^<>]*?>(.*?)<\/a>/';

$test = '<a href="link.php">First</a> and <a href="url2.html" class="delete">Second</a> and <a href="link.php">Third</a>';
echo preg_replace($pattern, '$1', $test)."\n";

$test = '<a href="link.php">First</a> and <a href="url2.html"><b class="delete">seriously</b></a> and <a href="link.php">Third</a>';
echo preg_replace($pattern, '$1', $test)."\n";

$test = '<a href="link.php">First</a> and <a href="url2.html" class="delete"><b class="delete">seriously</b></a> and <a href="link.php">Third</a>';
echo preg_replace($pattern, '$1', $test)."\n";

$test = '<a href="link.php">First</a> and <a  class="delete" href="url2.html">Second</a> and <a href="link.php">Third</a>';
echo preg_replace($pattern, '$1', $test)."\n";

score 0 · Answer 5 · answered Jul 21 '11 at 15:09

preg_replace('@<a href="[^"]+" class="delete">(.+?)</a>@', '$1', $html_string);

It is important to understand this is not an ideal solution. First, it requires markup in this exact format. Second, if there were, say, a nested anchor tag (albeit unlikely) this would fail. These are some of the many reasons why Regular Expressions should not be used for parsing/manipulating HTML.

How to remove link with preg_replace();?

5 Answers5

Linked