XML : Creating a browser:no use of html class,read html as xml,parse it,in java

this question is about this thread making a browser without using JTextPane or any other class that reads HTML but I create a new one so that I give more detailed information and questions. We (the team) cannot make use of the html class or whatever reads/parses html. We have to do the parsing ourselves( no webkit etc) .\

How can we read html as xml in java?

the browser will be tested in "good" websites and the requirements are not high. We just have to find a way to depicture the site WITHOUT css or javascript.

any ideas/articles that could give information on how to implement?