Java Programming

Hi,

I'm working with java.net.URL to code a simple program that read a file and get the archive described in this file. Actualy, the file is an HTML document done by a third party, and the program just have to read this file and download all the other document referenced via an tag. The problem is any valid HTML should be accepted, so the anchor may contain a relative URL. I just can't figure out how to get the absolute URL.

I was first surprised that no method does this in the URL class, then I realized it's the server's job to do so. There are some thricky case, like this intern URL we realy have at the office :
http://wiki/index.php/Accueil
The base URL is http://wiki/ and /Accueil is actually a parameter of index.php (I hate PHP).

I couldn't find the base URL in the HTTP header either. But the server has to give it somehow, or else my FireFox would not be able to render links or images on our wiki.

Is there anyway in Java to ask for the base URL of a specific document, or compute it ?