Parsers

I want to write a wikitext parser, but I don't know how ENTERPRISE quality parsers work.
I hear a lot of shit about parsing html with regex, am I do assume it is also inadequate for wikitext?

create data structures that hold intermediary parses and fully formatted output.. dicks should work. also you use beautifulsoup or its equivalent.