Trying to find the links on a page.
my regex is:
/<a\s[^>]*href=(\"\'??)([^\"\' >]*?)[^>]*>(.*)<\/a>/
but seems to fail at
<a title="this" href="that">what?</a>
How would I change my regex to deal with href not placed first in the a tag?
116
votes
10 Answers
213
votes
Reliable Regex for HTML are difficult. Here is how to do it with DOM:
$dom = new DOMDocument;
$dom->loadHTML($html);
foreach ($dom->getElementsByTagName('a') as $node) {
echo $dom->saveHtml($node), PHP_EOL;
}
The above would find and output the "outerHTML" of all A elements in the $html string.
To get all the text values of the node, you do
echo $node->nodeValue;
To check if the href attribute exists you can do
echo $node->hasAttribute( 'href' );
To get the href attribute you'd do
echo $node->getAttribute( 'href' );
To change the href attribute you'd do
$node->setAttribute('href', 'something else');
To remove the href attribute you'd do
$node->removeAttribute('href');
You can also query for the href attribute directly with XPath
$dom = new DOMDocument;
$dom->loadHTML($html);
$xpath = new DOMXPath($dom);
$nodes = $xpath->query('//a/@href');
foreach($nodes as $href) {
echo $href->nodeValue; // echo current attribute value
$href->nodeValue = 'new value'; // set new attribute value
$href->parentNode->removeAttribute('href'); // remove attribute
}
Also see:
On a sidenote: I am sure this is a duplicate and you can find the answer somewhere in here
19
votes
I agree with Gordon, you MUST use an HTML parser to parse HTML. But if you really want a regex you can try this one :
/^<a.*?href=(["\'])(.*?)\1.*$/
This matches <a at the begining of the string, followed by any number of any char (non greedy) .*? then href= followed by the link surrounded by either " or '
$str = '<a title="this" href="that">what?</a>';
preg_match('/^<a.*?href=(["\'])(.*?)\1.*$/', $str, $m);
var_dump($m);
Output:
array(3) {
[0]=>
string(37) "<a title="this" href="that">what?</a>"
[1]=>
string(1) """
[2]=>
string(4) "that"
}
4
votes
3
votes
why don't you just match
"<a.*?href\s*=\s*['"](.*?)['"]"
<?php
$str = '<a title="this" href="that">what?</a>';
$res = array();
preg_match_all("/<a.*?href\s*=\s*['\"](.*?)['\"]/", $str, $res);
var_dump($res);
?>
then
$ php test.php
array(2) {
[0]=>
array(1) {
[0]=>
string(27) "<a title="this" href="that""
}
[1]=>
array(1) {
[0]=>
string(4) "that"
}
}
which works. I've just removed the first capture braces.
3
votes
2
votes
Quick test: <a\s+[^>]*href=(\"\'??)([^\1]+)(?:\1)>(.*)<\/a> seems to do the trick, with the 1st match being " or ', the second the 'href' value 'that', and the third the 'what?'.
The reason I left the first match of "/' in there is that you can use it to backreference it later for the closing "/' so it's the same.
See live example on: http://www.rubular.com/r/jsKyK2b6do
2
votes
I'm not sure what you're trying to do here, but if you're trying to validate the link then look at PHP's filter_var()
If you really need to use a regular expression then check out this tool, it may help: http://regex.larsolavtorvik.com/
2
votes
Using your regex, I modified it a bit to suit your need.
<a.*?href=("|')(.*?)("|').*?>(.*)<\/a>
I personally suggest you use a HTML Parser
EDIT: Tested
0
votes
The following is working for me and returns both href and value of the anchor tag.
preg_match_all("'\<a.*?href=\"(.*?)\".*?\>(.*?)\<\/a\>'si", $html, $match);
if($match) {
foreach($match[0] as $k => $e) {
$urls[] = array(
'anchor' => $e,
'href' => $match[1][$k],
'value' => $match[2][$k]
);
}
}
The multidimensional array called $urls contains now associative sub-arrays that are easy to use.
