W3C home > Mailing lists > Public > xmlschema-dev@w3.org > August 2004

Re: Extension and all-group

From: Daniel Sitton <dsitton@panorama.com>
Date: Wed, 4 Aug 2004 14:18:21 +0300
Message-ID: <21F319296175054CA3A74B66159C6E5960F4F2@pan1.pansw.com>
To: <xmlschema-dev@w3.org>
Cc: <ht@cogsci.ed.ac.uk>
> Hello!
> This question was raised at news:microsoft.public.dotnet.xml newsgroup
> -
> is the following schema valid? .NET gives errors, while XML Spy and
> W3C validator say okay.
The schema is not valid.  The derived type violates the requirement
that the only place an 'all' group can appear is all alone as the only
group in a type.  XSV should check for this, but doesn't [1]:
 "<all> groups are [not] properly constrained as to how they appear in
I understand how this results from the definitions, but IMHO there is a
problem here.
Consider a case in which object A inherits from object B and both have
attributes that should be serialized. When building the matching schema,
it's either forgetting about the inheritance or using a 'sequence' and
not 'all' in the definition of   both the base type and the extending
type. The latter however has the poor outcome of forcing order on the
object's attributes. 
Am I missing something here?

Received on Wednesday, 4 August 2004 07:28:47 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:56:05 UTC