@diEcho and @Dve are correct, you should learn to use something like the native DOMdocument class rather than using regex. Your code will be easier to read and maintain, and will handle malformed HTML much better.
Here is some sample code which may or may not do what you want:
$contents = '';
$doc = new DOMDocument();
$doc->load($page_url);
$nodes = $doc->getElementsByTagName('div');
foreach ($nodes as $node)
{
if($node->hasAttributes()){
$attributes = $element->attributes;
if(!is_null($attributes)){
foreach ($attributes as $index=>$attr){
if($attr->name == 'class' && $attr->value == 'rgz'){
$contents .= $node->nodeValue;
}
}
}
}
}