I use the
Quiotix HTML Parser which is a good tool to walk through the HTML, look for stuff, change it, and regenerate HTML.
If you use something like that you can find any relative addresses and fix them up. Here's code I use for that in my Wiki ... the "nameSpace" variable is the absolute path to the page that contains the relative links. It doesn't support "." to indicate current directory, but otherwise I'm pretty sure it works
[ April 07, 2005: Message edited by: Stan James ]