Now that I'm filling more and more data states, i can actually parse more and more html5 code. To test all this, i've generated some testcases through ChatGPT, but I actually found a test suite for html5: https://github.com/html5lib/html5lib-tests

I can actually use these tests to see if the tokenizer works as expected, so I'm spending some time now to building a html5tester that will test the tokenizer against each test in this html5lib testsuite.

Rust is becoming clearer by the day and I'm starting to understand the things i'm doing. That is progress!