W3C home > Mailing lists > Public > xmlschema-dev@w3.org > July 2003

RE: XML Schema compliance

From: Savas Parastatidis <Savas.Parastatidis@newcastle.ac.uk>
Date: Wed, 2 Jul 2003 18:52:29 +0100
Message-ID: <BC28A9E979C56C44BCBC2DED313A447001D75BE6@bond.ncl.ac.uk>
To: "Henry S. Thompson" <ht@cogsci.ed.ac.uk>
Cc: <xmlschema-dev@w3.org>, "Jim Webber" <jim.webber@arjuna.com>, "Paul Watson" <Paul.Watson@newcastle.ac.uk>

> > <xs:complexType name="myType">
> >    <xs:complexContent>
> >       <xs:restriction base="xs:element">
> >          ...
> >       </xs:restriction>
> >    </xs:complexContent>
> > </xs:complexType>
> >
> > With xmlns:xs="http://www.w3.org/2001/XMLSchema"
> There's nothing I'm aware of in the REC to stop you doing this.
> There is a type definition with that name in the schema for schemas.
> Like any other type definition that doesn't include
> final='restriction', you can derive new type definitions from it by
> restriction.

Please see bellow why I think why you can't do that.

> Purists would probably argue you should include
> <xs:import namespace="http://www.w3.org/2001/XMLSchema"/>

I believe that this is not necessary. When you define an XML Schema, the
above namespace is assumed to be already imported. I may be wrong in
this but including the above line in XmlSpy and VS.NET 2003 does not
help in validating the XML Schema I posted in my previous message.

> What made you think this wasn't allowed?  I didn't understand your
> comment "since the XML Schema specification is defined as an infoset,
> the XML Schema <element> element that is defined in the normative XML
> Schema schema (I hope this makes sense) should not be used in the
> above manner".

Here's why I believe that "xs:element" cannot be used as base in an XML
Schema declaration.

The XML Schema infoset defines a tree of information items with the
{namespace} property for all of them set to
"http://www.w3.org/2001/XMLSchema" (section 3.1.2).

So, when one uses

<xs:restriction base="xs:element">

What should a tool do? Should it refer to the "element" complex type
defined in http://www.w3.org/2001/XMLSchema by the normative XML Schema
or should it refer to the more general, representation-independent XML
Schema information item defined by the XML Schema infoset in the same

The two tools I tried seem to agree with me although I don't know
whether this because of a bad implementation of the XML Schema standard.

Received on Wednesday, 2 July 2003 13:58:29 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:56:02 UTC