Re: [xml-dev] Subtyping in XML

[ht@cogsci.ed.ac.uk (Henry S. Thompson)]

Paul Prescod <paulp@prescod.net> writes:

> On Wed, 4 Sep 2002, Jeff Lowery wrote:
> 
> > What you're saying is that base members come before extended members? That's
> > fine, and it's not possible to extend (A, B, C[3]) with (A, B, C[3], C*) in
> > XML Schema, although it is possible to extend it to (A, B, C[3], D, C*).
> > It's that last case that I consider Not So Bad, but Paul seems to find
> > objectionable.
> 
> Base type: TRUE|FALSE
> 
> Extended type: ((TRUE|FALSE), TRUE)

Sorry, just to be clear, those are DTD-notations for content models
with element names, nothing to do with simple types or booleans, right?

> This will defacto extend the enumeration and worse, allow two conflicting
> values at once. 

Using the word enumeration here is potentially misleading -- it's
usually used in DTD/schema context to discuss enumerated values for
attributes/simply-typed elements.  If that's how you're using it, the
above is _not_ possible in W3C XML Schema.

> Most SAX filters will not guard against these kinds of things.

The semantics you attribute to your element names etc. is surely your
business, and I wouldn't expect SAX or any other syntax-level tool to
interfere.

ht
-- 
  Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
          W3C Fellow 1999--2002, part-time member of W3C Team
     2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
	    Fax: (44) 131 650-4587, e-mail: ht@cogsci.ed.ac.uk
		     URL: http://www.ltg.ed.ac.uk/~ht/
 [mail really from me _always_ has this .sig -- mail without it is forged spam]