In that case the following example might be worth a look ( HTML::TreeBuilder::LibXML vs HTML::TreeBuilder::XPath ):
https://gist.github.com/hernan604/8466937 But remember HTML::TreeBuilder::LibXML will accept more html5 tags than HTML::TreeBuilder::XPath =) On Tue, Mar 4, 2014 at 7:35 PM, James Laver <james.la...@gmail.com> wrote: > > On 4 Mar 2014, at 22:10, DAVID HODGKINSON <daveh...@gmail.com> wrote: > > > For what I'm thinking, a way of relating named divs (and lists of) on > > a page to the hash elements needed for poking into DBIx::Class. > > > > As for Web::Scraper, it's Miyagawa-ware, so definitely worth looking > > at. > > Sounds like what you actually want is a handful of app-specific lines of > code around HTML::TreeBuilder. You can fetch with LWP (maybe LWP::Simple if > your needs are small) or WWW::Mechanize for more complex stuff, or whatever > else. > > FWIW, last time I got involved in web scraping, this approach worked quite > well and while it's not immediately reusable, it's pretty straightforward. > > James >