Wendell Piez wrote:
The bottom line seems to be that at least one popular
processor (released by W3C, even) does not respect every
rule restricting colons in namespace-conformant documents.
Well, I would rather say that the input HTML document does not conform to the
namespaces rec. And that's "legitimate" as HTML is not XML. The parser here
has to make a choice: 1/ be strict and generate an error or 2/ be lax and
produce well-formed SAX events sequence, even if the node names do not conform
to NS rec.
As Tidy HTML is not directly targeted as producing an XDM instance, I'd say
this behavior is consistent. But well, I guess that's the kind of use cases
where the developer is supposed to know what he makes, and how to configure his
tools the right way to get the result he wants.
When you write such a parser, you have the freedom to define the mapping
between input and output as you want, but when you use such a parser, you have
to hope there is already something that does what you want, maybe by pushing
the right switches, or you have to write it yourself.
But as you said, I don't think this is still related to the XSLT rec. This
is more related to the design choices of a particular tool. You can try to
disable the namespace support of the parser you use to feed a transform, but
then, it is normal it will not handle correctly namespaces ;-)
Regards,
--
Florent Georges
http://www.fgeorges.org/
--~------------------------------------------------------------------
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
To unsubscribe, go to: http://lists.mulberrytech.com/xsl-list/
or e-mail: <mailto:xsl-list-unsubscribe(_at_)lists(_dot_)mulberrytech(_dot_)com>
--~--