Blocking URL hacking attemps

I'm using a script called BlackHole that uses PHP and a "hidden" directory to catch bad bots and block them by their IP address from the site [http://perishablepress.com/blackhole-bad-bots/]. It's working very well, but I'd like to expand on the idea. When looking at our access logs, I see someone "adding code" to our URL's in hopes to, apparently, hack into our site. If our URL is like this: http://www.example.com/pages/bolts.php they are adding to it like this: http://www.example.com/pages/bolts.php/wp-content/themes/functions/thumb.php.cache/external_e19f4bb51bc2262e07d23b79b916c12c.php or http://www.example.com/pages/bolts.php/cart.php or http://www.example.com/pages/bolts.php/wp-content/themes/functions/thumb.phptimthumb.php or http://www.example.com/pages/bolts.php/bratislava/stare-mesto Is there a way that I can trap the "extra" info in the URL's and pass it to the BlackHole script to ban these attempts as well? I'm thinking I'll need to compare the entered URL with $_SERVER["PHP_SELF"] and if there's extra info, like another "*.php", after the page name, then ban the IP. I realize that's a rather simplistic explanation and there may be a better way to accomplish this. If anyone has any examples they can share, I'd really appreciate. Thanks, Marc