PHP Dom没有检索元素

$code = '
<h1>Galeria </h1>

<div class="galeria">
    <ul id="galeria_list">
        <li>
          <img src="img.jpg" width="350" height="350" />
          <br />
          Teste
        </li>
    </ul>
</div>';


$dom = new DOMDocument;
$dom->validateOnParse = true;

$dom->loadHTML($code);

var_dump($dom->getElementById('galeria_list'));

The var_dump always returns NULL. Anyone know why? I can clearly see the element with the id galeria_list in $code. Why is this not getting the element?

And also, does anyone know how to prevent the domdocument from adding the <html> and <body> tags on the saveHTML method?

Thanks

It seems that the DOMDocument will not play nice with HTML fragments. You may want to either consider the DOMDocumentFragment (as dnagirl suggests) or consider extending the DOMDocument.

After a little research, I've put together a simple extension that will achieve what you are asking:

class MyDOMDocument extends DOMDocument {

    function getElementById($id) {

        //thanks to: http://www.php.net/manual/en/domdocument.getelementbyid.php#96500
        $xpath = new DOMXPath($this);
        return $xpath->query("//*[@id='$id']")->item(0);
    }

    function output() {

        // thanks to: http://www.php.net/manual/en/domdocument.savehtml.php#85165
        $output = preg_replace('/^<!DOCTYPE.+?>/', '',
                str_replace( array('<html>', '</html>', '<body>', '</body>'),
                        array('', '', '', ''), $this->saveHTML()));

        return trim($output);

    }

}

Usage

$dom = new MyDOMDocument();
$dom->loadHTML($code);

var_dump($dom->getElementById("galeria_list"));

echo $dom->output();

You might consider DOMDocumentFragment rather than DOMDocument if you don't want headers.

As for the id problem, this is from the manual:

<?php

$doc = new DomDocument;

// We need to validate our document before refering to the id
$doc->validateOnParse = true;
$doc->Load('book.xml');

echo "The element whose id is books is: " . $doc->getElementById('books')->tagName . "
";

?>

validateOnParse is likely the issue.

Someone worked around this problem in the PHP manual by using XPath: http://us3.php.net/manual/en/domdocument.getelementbyid.php#96500

It seems that loadhtml() does not "attach" the html dtd that defines id as an id-attribute to the DOM. But if the html document contains a DOCTYPE declaration it works as intended. (But my guess is you don't want to add a doctype and html skeleton, anyway:).

$code = '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
<html><head><title>...</title></head>
<body>
  <h1>Galeria </h1>
  <div class="galeria">
    <ul id="galeria_list">
      <li>
        <img src="img.jpg" width="350" height="350" />
        <br />
        Teste
      </li>
    </ul>
  </div>
</body></html>';

$dom = new DOMDocument;
$dom->loadhtml($code);
var_dump($dom->getElementById('galeria_list'));