Thursday, April 28, 2011

Extract everything between <object></object>

I am using CURL to download a page. Now I want to extract this from the page:

<object classid="clsid:67DABFBF-D0AB-41fa-9C46-CC0F21721616" width="640"
        height="303.33333333333"
        codebase="http://go.divx.com/plugin/DivXBrowserPlugin.cab"
        id="object701207571">
    <param name="autoPlay" value="false" />
    <param name="custommode" value="Stage6" />
    <param name="src" value="" />
    <param name="movieTitle" value="Titanic" />
    <param name="bannerEnabled" value="false" />
    <param name="previewImage" 
           value="http://stagevu.com/img/thumbnail/oripmqeqzrccbig.jpg" />
    <embed type="video/divx" src="" width="640" height="303.33333333333"
           autoPlay="false" custommode="Stage6" movieTitle="Titanic"
           bannerEnabled="false"
           previewImage="http://stagevu.com/img/thumbnail/oripmqeqzrccbig.jpg"
           pluginspage="http://go.divx.com/plugin/download/"
           id="embed701207571">
    </embed>
</object>

Please help!

From stackoverflow

  • See Can you provide some examples of why it is hard to parse XML and HTML with a regex? for why this is probably the wrong thing to do.

    That said you might be able to get away with something like /(<object>.*?<\/object>)/s. This matches the string "<object>" followed by any number of characters up to the string "</object>". The s on the end tells . to match newlines (it normally doesn't).

    strager : +1 for the first paragraph.

  • This is partially in response to Owens (because I can't put code in a comment very well). That regex might not work for the object tag, basically because the opening <object> tag has attributes in it. Try this one instead:

    /(<object[^>]*>)(.*?)(<\/object>)/si

    It's case insensitive and broken into the three groupings for easy reference. It's not 100% perfect, but should help.

    strager : > is legal in an attribute value, IIRC.
    strager : Also, this does not handle nesting.
    St. John Johnson : Which is why it is hard to parse HTML with a Regex. But this will work for his attempt.
    Chas. Owens : Yeah, these are the dangers of trying to use a regex, which is why I used a half-hearted match-what-he-showed approach. Any time spent attempting to bullet proof the regex is time that should have been spent learning how to use a parser.

  • this regex will match all the line breaks between the opening and closing tags and capture the entire thing in one group

    /(<object[^>]*?>(?:[\s\S]*?)<\/object>)/gi
    porneL : This will fail if objects are nested.
    Scott Evernden : right .. but i don't think i've ever seen objects nested inside objects
    strager : It's completely legal. I've seen it. You can have an image object inside a video object inside a flash object, for example.

  • Using SimpleXML:

    $sxe = new SimpleXMLElement($xml);
$objects = $sxe->xpath('//object[@id="object701207571"]');
$object = $objects[0];

$params = $object->xpath('param');

foreach($params as $param)
{
    $attrs = $param->attributes();
    echo $attrs['name'] . ' = ' . $attrs['value'] . "\n";
}

// Get plain XML:
echo $object->asXML();

  • $doc = DOMDocument::loadHTML($html);
foreach($node->getElementsByTagName('object') as $object)
{
   echo $doc->saveXML($object);
}

  • @St. John Johnson Good sir its working ... thanx

    • Posted by Ku XI at 8:01 PM

    0 comments:

    Post a Comment

    Subscribe to: Post Comments (Atom)