W3C home > Mailing lists > Public > xmlschema-dev@w3.org > May 2002

Re: Circular types in XML Schemas

From: Henry S. Thompson <ht@cogsci.ed.ac.uk>
Date: 01 May 2002 20:26:43 +0100
To: Jeni Tennison <jeni@jenitennison.com>
Cc: "Morris Matsa" <mmatsa@us.ibm.com>, Yuri de Wit <yuri.dewit@metaserver.com>, "'xmlschema-dev@w3.org'" <xmlschema-dev@w3.org>
Message-ID: <f5bsn5boej0.fsf@cogsci.ed.ac.uk>
Jeni Tennison <jeni@jenitennison.com> writes:

> Hi Morris,
> 
> > I believe that necessarily infinitely recursive unsatisfiable types
> > are forbidden in Schemas. This question was answered about a year
> > ago on this list [1], and the part of the schema spec quoted then
> > was [2]. Please tell me if I'm wrong.
> 
> Hmm... I don't think that the definition of an emptiable group
> (http://www.w3.org/TR/xmlschema-1/#cos-group-emptiable) relates to the
> question of infinitely recursive unsatisfiable types. As far as I can
> tell, the only places in which this Schema Component Constraint is
> used is when assessing whether it's OK to derive one type from another
> type, and whether an element's default/fixed value is OK.
> 
> There might be something else specifying a constraint like this
> somewhere, of course, but I don't think that's it. If there were such
> a constraint, I guess it would be a constraint on model groups that
> stated that a model group whose minimum effective total range is more
> than 0 cannot contain an element particle at any level whose minimum
> effective total range is more than 0 and whose type's content involves
> that model group at any level. But I can't see such a constraint
> anywhere.

There isn't one.  It would be very expensive to enforce, and not in at 
least my judgement worth it.

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]
Received on Wednesday, 1 May 2002 15:26:57 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:55:56 UTC