Blah... the parser is quite complex even when "http://" is required, because there are many possible types of URLs...
http://localhost/
http://localhost:80/
http://127.0.0.1/
ftp://msgplus.net/
http://www.msgplus.net/whatever/blah%20spam.jpg
http://whatpulse.bounceme.net/
I think adding a parser for www.something would be a waste of time... people should use the [url] tags like so:
www.mess.be
/ 
