We've finally succeeded in passing all the 112 parser tests found in the html5lib-tests suite's test1.dat file. There are MANY other files to go, with even more tests but this is actually a big milestone. The more tests we get to pass, the more other tests will automatically pass so things should speed up from this point on.

The parser at its current setup will never be the fastest one, but I'm more concerned on accuracy than speed. As long as we pass the tests, we can refactor the whole tokenizer/parser later on.

So for the next few weeks I probably will be focussing on doing more work on the parser tests which hopefully result in the GosubEngine to be able to parser a lot of real html pages (it currently doesn't).

Meanwhile, there is lots of progress done on the CSS3 tokenizer/parser and element attributes that will be important when generating the real DOM tree. I will try and do some research on implementing a javascript engine, but it is not the most important part at the moment. I also will think about a bit about the networking layer and how it all interconnects within the engine.

A lot of work, but amazing what we have accomplished in such a short time!