It depends *very*much* on what you're trying to extract.
I once had my motd come from
<?php
$x = `curl -s http://www.care2.com/`;
$t = strpos($x, 'DAILY QUACK UP');
$y = substr($x, $t, 300);
$t = strpos($y, '</td>');
$z = substr($y, 0, $t);
$z = str_replace('</b></font></a><br>', '', $z);
$z = str_replace('<BR>', '', $z);
echo $z;
?>
================
You may want to try these classes:
Class: HTMLparser
http://www.phpclasses.org/browse.html/package/244.html
Class: HTMLSaxhttp://bytes.com/groups/php/2515-parsing-html
http://www.phpclasses.org/htmlsax
From: http://bytes.com/groups/php/2515-parsing-html