Script4rss takes a plain text file which holds a description for how the
particular site should be converted and creates a perl script which is able to
do that in the most fast and efficient way (well, someday). Users don't have to
know how to program but they need to know regular expressions (although there
probably aren't a lot of these people).

At the moment, script4rss is in its early development, which translates to "it
can be used but you have to figure out how yourself" and "if you screw up, the
script does so as well". Options include:
  * Detect multiple catagories within an HTML page.
  * Extract information over multiple lines.
  * Pre-and append text in output.
  * Attempt to circumvent "variable" HTML.