I don't see the source ...

To parse specific tags out of the returned HTMP content, I'd recommend using something like HTML::Parser. There are plenty of examples in the docs on how to pull content from specific tags.

Can't comment on the easier way without the code, but I don't get the "don't use LWP" comment. LWP::UserAgent is about the simplest, yet most complete package for automating web access I've ever used. What are they suggesting you use instead?