I think a better (or more moddern) approach would be to use an HTML5 parser (like html5lib), as it has a standar way to parse INVALID documents too, and then re

I think a better (or more moddern) approach would be to use an HTML5 parser (like html5lib), as it has a standar way to parse INVALID documents too, and then recreate the HTML with the DOM. BTW: Paul Irish has this ticket asking for help/someone to update lazyweb to use an HTML5 parser: https://github.com/paulirish/lazyweb-requests/issues#issue/20 https://github.com/paulirish/lazyweb-requests/issues#issue/2...