[xml] A possible problem with libxml2

[Steve Underwood <steveu coppice org>]

Hi,

The attached HTML fragment is the start of an HTML document generated by
MS Word. The HTML parser in libxml2 2.3.9 chokes on this, as it parses
the XML document description stuff. As far as I can see, the markers
before and after the XML should cause the XML to be ignored as an HTML
comment, but it isn't. Am I missing something, or is libxml really doing
something wrong here?

Regards,
Steve