- From: Shane McCarron <shane@aptest.com>
- Date: Tue, 03 Apr 2007 14:33:34 -0500
- To: Ben Adida <ben@adida.net>
- CC: Karl Dubost <karl@w3.org>, Olivier Théreaux <ot@w3.org>, QA Dev <public-qa-dev@w3.org>, RDFa force <public-rdf-in-xhtml-tf@w3.org>, HTML WG <w3c-html-wg@w3.org>
Ben Adida wrote: > I'm confused: this requires re-declaring all the XML namespaces in the > DTD declaration? That doesn't seem like a viable direction for > validating pages in the wild. > Well - there are other ways to do it. I was just demonstrating that it would work. I didn't mean to imply it was a good idea. You and some others seemed to feel it was impossible. > Is this a DTD vs. XML Schema validation issue? > Yes. Sort of anyway. The w3c validator, and indeed MOST validators out there, rely upon the sgmls/nsgmls/onsgmls tool that is now part of OpenSP. This tool is DTD based, and has no provision I know of for permitting random attributes to just show up. We addressed this in XHTML M12N many years ago by providing this hook. You could achieve the same thing by teaching the validator abouts lots of well know namespaces and prefixes... and then having it automagically shoehorn them into the internal subset before processing. You could also have the validator pre-scan the document, extract the namespace declarations, and do the same thing. It has not really been an issue to date, but those are a couple of pretty straightforward solutions that would work more or less immediately with XHTML Family markup languages. I would be happy to implement either one if anyone is interested in letting me. -- Shane P. McCarron Phone: +1 763 786-8160 x120 Managing Director Fax: +1 763 786-8180 ApTest Minnesota Inet: shane@aptest.com
Received on Tuesday, 3 April 2007 19:34:04 UTC