W3C home > Mailing lists > Public > www-ws-desc@w3.org > February 2004

RE: WSD WG requests to XML Schema WG

From: David Orchard <dorchard@bea.com>
Date: Fri, 27 Feb 2004 16:27:18 -0800
Message-ID: <32D5845A745BFB429CBDBADA57CD41AF05AA4B63@ussjex01.amer.bea.com>
To: "Glen Daniels" <gdaniels@sonicsoftware.com>, "Matthew Fuchs" <matt@westbridgetech.com>, <w3c-xml-schema-ig@w3.org>
Cc: <www-ws-desc@w3.org>

Glen,

I agree on the issue of self-describing type hierarchies, and I made this suggestion in my "list of things schema could do" document that apparently hasn't gotten much of a read...

It seems to me the relationship between the name's type and the xsi:type should be rigorously defined.  Is there a guarantee that the xsi:type is a compatible type wrt name?  If you don't know, then processing name without knowing the xsi:type could be erroneous.

Dave

> -----Original Message-----
> From: Glen Daniels [mailto:gdaniels@sonicsoftware.com]
> Sent: Sunday, February 22, 2004 11:49 AM
> To: David Orchard; Matthew Fuchs; w3c-xml-schema-ig@w3.org
> Cc: www-ws-desc@w3.org
> Subject: Re: WSD WG requests to XML Schema WG
> 
> 
> Hi folks!
> 
> David Orchard wrote:
> > If an older receiver gets the <name xsi:type="nameType2">, 
> how does it
> > know that nameType2 is an extension of nameType?  Is there 
> some way of
> > saying "because  the receiver knows about nameType1 and it gets an
> > element called name, it can ignore anything that is 
> different between
> > nameType1 and nameType2?
> 
> We had this same issue in a lot of the early SOAP encoding 
> conversations,
> and I remember talking through this in detail with Eric 
> Prud'Hommeaux back
> in 2000 in the context of object types and inheritance.  We 
> came up with a
> parentTypes="" (not sure what we called it tho) attribute 
> which would allow
> self-describing inheritance hierarchies for compatible 
> extension types,
> though it never went beyond that into any of the WG's.
> 
> If you know the schema for <name>, I think it's OK to simply 
> ignore the
> xsi:type - assuming what you're receiving is still 
> schema-valid with respect
> to the base type, it shouldn't matter if you don't understand 
> the new type.
> 
> I believe .NET used to (or maybe still does) *always* ignore 
> the xsi:type
> attribute....
> 
> --Glen
> 
> 
Received on Friday, 27 February 2004 19:27:28 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:15:02 UTC