TODO.md

HTML

• Tokenizer
  • character buffering to reduce the number of token objects ?
  • Since we now buffer some character tokens, add tests for whitespace handling, i.e. for when the tree builder needs to ignore whitespace-only tokens
  • Check places where we could use chars_(while|until) instead of consuming one byte at a time. Typically, when the spec says « Ignore the character »
• TreeBuilder
  • parse errors
• Serializer
  • boolean attributes

CSS

• Syntax Module Level 3
• Selectors Module Level 4
  • Correctly handle namespaces
  • Correctly handle case sensitivity
  • Tests
    • https://www.w3.org/Style/CSS/Test/
    • http://test.csswg.org/harness/
    • http://test.csswg.org/suites/selectors-4_dev/nightly-unstable/html/toc.htm
    • http://test.csswg.org/suites/selectors-3_dev/nightly-unstable/xhtml1/toc.xht
    • https://github.com/dperini/nwsapi/tree/master/test
    • https://github.com/facelessuser/soupsieve/tree/master/tests