Re: [xml] xml Digest, Vol 89, Issue 16

[Alan Hogan <alanhogan gmail com>]

On Sep 24, 2011, at 5:00 AM, xml-request gnome org wrote:
> > "should be able to parse "real world" HTML, even if severely broken from a
> > specification point of view"
>
> It's documented, therefore it can be called a feature, not a bug :)
>
> Csaba

Sounds like a cop-out.

1. What is the point of parsing HTML if it is treated like dumb XML without an intelligent spec-aware layer 
on top of that (at least optionally)?

2. What alternative do people have to libXML if they want to generate a “real” DOM from HTML?

3. Why document an “HTML” mode if it isn’t really parsing HTML? Per HTML, there really ought to be a TBODY 
node, no ifs, ands, or buts.

ajh