- From: David Carlisle <davidc@nag.co.uk>
- Date: Fri, 29 Aug 2008 09:47:07 +0100
- To: public-html@w3.org
If you want a DTD that just does the html5 (that is, essentially html+mathml) entity definitions and not any validation you could use http://www.w3.org/2003/entities/2007/w3centities-f.ent (which is linked from section 2 of http://www.w3.org/TR/xml-entity-names/ ) It's made from the same unicode.xml file from which the entity definitions for the HTML serialisation are being extracted so it should match. (Actually that file probably includes a couple of the ISO sets that are not in html or mathml, but if they need to be omitted then an extra predicate in the xpath statement controlling the generation would be all that is required.) If you want a DTD for validation as well, then (assuming someone makes a relax schema) it probably makes most sense to use trang or some other convertor to convert the relax ng to dtd and then add a reference to the entity definitions. (That's what I'm planning on doing for mathml3, although currently the draft mathml3 schema uses too many relax features so trang won't convert it, but I expect that in the end it will be easy enough to devise a pre-processing step that simplifies the relax grammar before passing it to trang. This just addresses the possibility of generating html5 related dtd; I don't want to comment on more political/policy issues about whether such a dtd should be hosted by some organisation or referenced from a spec etc. David ________________________________________________________________________ The Numerical Algorithms Group Ltd is a company registered in England and Wales with company number 1249803. The registered office is: Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom. This e-mail has been scanned for all viruses by Star. The service is powered by MessageLabs. ________________________________________________________________________
Received on Friday, 29 August 2008 08:47:42 UTC