Over the past couple of months I've been working on a new version of
Beautiful Soup that delegates the parsing chores to an external
parser. I'm now happy to announce the release of the alpha version of
Beautiful Soup 4. Instead of trying to maintain its own HTML parser,
BS4 lets you plug in a faster (lxml) or more accurate (html5lib) HTML
parser. I still have some work to do (notably, to add support for
Python 3), but you should be able to use 4.0.0a everywhere you would
use 3.2, and get better results.
Get 4.0.0a from the project homepage:
http://www.crummy.com/software/BeautifulSoup/
Or check out lp:beautifulsoup/bs4 from Launchpad to track my development.
File bugs here: https://bugs.launchpad.net/beautifulsoup
Leonard