Re: Importing XMLSchema.xsd

Henry:  I wonder whether we shouldn't open an issue covering both the lack 
of clarity regarding what processors should do, as well as the need for a 
mechanism that reliably supports what's requested here.  Presumably the 
situation is somewhat complicated by the fact that the Schema for Schemas 
is not quite a legal schema, insofar as one of its purposes is to 
introduce or at least provide models for the very simple types that are 
built in and that we normally prohibit users from creating, I.e. types 
that derive directly from anySimpleType.

I wonder whether in a future version of schema we might somehow factor the 
Schema for Schema into one piece that would indeed be a legal schema 
document covering the syntax for <xsd:element>, etc., and perhaps even the 
definitions of the built in derived types.  The other piece providing the 
"magical" derivations of the builtin primitive types.  Presumably, with a 
bit of care, we could arrange things so that the former could be imported 
for use in other vocabularies, as is being requested in this thread? 

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------








ht@cogsci.ed.ac.uk (Henry S. Thompson)
Sent by: xmlschema-dev-request@w3.org
11/13/2003 03:34 AM

 
        To:     "Simon Cox" <Simon.Cox@csiro.au>
        cc:     <jddahl@micron.com>, <xmlschema-dev@w3.org>, (bcc: Noah 
Mendelsohn/Cambridge/IBM)
        Subject:        Re: Importing XMLSchema.xsd



"Simon Cox" <Simon.Cox@csiro.au> writes:

> Re: Importing XMLSchema.xsdYes - WXS components are all available in
> a schema without having to <import>.
>
> And as you point out, a simpleType can appear in an instance, as the
> value of an xsi:type.
>
> But the goal is to be able to have type *definitions* (in
> particular, restictions of the simple types) appear in an *instance*
> document.
>
> For example, I'd like to be able to indicate to the processor at run
> time, **in the instance**, that (for example) we have a numeric
> range.  So rather than inventing a new mini-schema language, I'd
> like to use WXS components, for example:
>
> <xs:simpleType ... > 
>   <xs:restriction base="xs:double">
>     <xs:minInclusive value="45.7e9"/>
>     <xs:maxInclusive value="467.2e9"/>
>   </xs:restriction>
> </xs:simpleType> 
>
> I thought in this case it needed us to <import> the S4S into the
> schema that validates the instance.

I think I understand your scenario -- it would appear to be the case
that different implementations treat the sForS specially in different
ways, which is unfortunate but not I guess surprising.

What happens if you include references to e.g. xsd:simpleType in the
schema document for _your_ namespace _without_ an import?

ht
-- 
  Henry S. Thompson, HCRC Language Technology Group, University of 
Edinburgh
                      Half-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 Thursday, 13 November 2003 17:29:26 UTC