> sure if there is a relation (link) between them 
> xsl is the look, and the xml the meaning, but theres 
> some variants to research.
> I don't know the parsing algorihtm that the program 
> use, but it managed to do something, but is hard to 
> understand (i only put a couple of minutes in this)
> But go ahead!!! impressive results!

Hum, I fear that the xsl/xml files might be linked 
together, and therefore I'll have to crawl a little 
into some XML references (I stil didn't read the 
boring book "XML use guide"..)

For the current HTML/Javascript parser, the engine is 
quite efficient, but a **little** bit dirty as you may 
have noticed :)
The problem is that HTML becomes more and more 
complex, and if HTML 3.2 was really easy to parse (you 
can parse a standard 3.2 file in 10 lines of Perl!), 
HTML4+javascript is really a nuisance to analyze. 
HTTrack can handle complex things like document.write
('<a href=image.gif'); but not yet analyzing the code 
itself.

Maybe XML will simplify this.. or not