Search code examples
phpdomdomdocumenthtml-content-extraction

Extracting description out of an html page

I am trying to extract title and description out of web pages, using DOMdocument(), I am successful in extracting title like this

$d=new DOMDocument();
$d->loadHTML($html);
$title=$d->getElementsByTagName("title")->item(0)->textContent;

I can extract the description by looping through all meta tags and checking for the name="desctiption"attribute but looping makes the process slow so wanted to know if there can be a direct method for extracting content using some attribute selector in php DOMdocument??

Solution

  • I don't think this can be done with DOMDocument alone, but it is possible in combination with with DOMXPath:

    $html = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>Dom - Xpath test</title>
<meta name="description" content="The first description meta tag" />
<meta name="keywords" content="none, no-keywords" />
<meta name="description" content="the second description tag" />
</head>
<body>
<p>This is the test HTML</p>
</body>
</html>
';

$dom = new DOMDocument();
$dom->loadHTML($html);
$domx = new DOMXPath($dom);
$desc = $domx->query("//meta[@name='description']");

$i = 0;
while ($item = $desc->item($i++)) {
    echo '<p>'.$item->getAttribute('content').'</p>';
}