Original Message From: "Florent Georges">

Hi Florent,

>> Personally I think that, subject to occurrence constraints, the
>> particle that is currently gobbling up input, should have
>> priority (i.e. they're greedy).
>
>  So the following, from the current example:
>
>    <xs:sequence maxOccurs="unbound">
>       <xs:element ref="elem"/>
>       <xs:any namespace="##targetNamespace" maxOccurs="unbound"/>
>    </xs:sequence>
>
> would be valid, while a bit of an obfuscation because the maxOccurs on
> the sequence won't be taken into account.  Right?

That's what I'd propose. If you wanted to allow the repeating sequence to 
take effect, then you'd change it to:

    <xs:sequence maxOccurs="unbound">
       <xs:element ref="elem"/>
       <xs:any namespace="##targetNamespace" notQName="elem" 
maxOccurs="unbound"/>
    </xs:sequence>

It's not quite the same as the current XSD 1.1 rules because as I understand 
it XSD 1.1 would use the following association:

<elem/> -> xs:element elem
<elem/> -> xs:any
<elem/> -> xs:element elem
<elem/> -> xs:any
etc.

Whereas the revised schema form would mean that the above input would be 
invalid due to the absence of a minOccurs in the xs:any.

However, the above particle assignment just doesn't seem particularly 
sensible to me, and I'd rather have simple rules than accommodate such odd 
instances.


>  Is the above quote what tells XML Schema 1.1?  If yes, where it is in
> the WD?

Not yet!

Regards,

Pete Cordell
Codalogic
For XML C++ data binding visit http://www.codalogic.com/lmx/



From petexmldev@c... Wed Jun 25 13:19:40 2008
Received: from maggie.w3.org ([193.51.208.68])