This is a parser for HTML documents. Unlike for XML documents, it must include a certain amount of error-correction to account for HTML features like self-terminating tags, unterminated tags, and incorrect nesting. The input is tokenised by the XML lexer (a separate lexer is not required for HTML).