This sounds like just another example of conditional type assignment:

type="if (@enum_extension) then xs:string else my:enumerated-type"

Michael Kay
http://www.saxonica.com/


> -----Original Message-----
> From: xmlschema-dev-request@w... 
> [mailto:xmlschema-dev-request@w...] On Behalf Of Bryce Nesbitt
> Sent: 02 May 2007 20:08
> To: xmlschema-dev@w...
> Cc: Pete Cordell; Chuck Herrick
> Subject: Re: Schema 1.1: xs:anyEnumeration considered?
> 
> 
> The original poster (that's me) was drawing an analogy 
> between Java and XML, perhaps a little unfair.  In the Java 
> world packages outside the core get an "x" in their name (for 
> "experimental" or "extended") until they are eventually 
> folded into the core.
> 
> For XML enums it might look like this.  Imagine an 
> enumeration containing "schwinn", "huffy" and "specialized", then:
> 
>     <vendor>specialized</vendor>
>     <vendor>heffy</vendor>
>     <vendor>Schwinn</vendor>
>     <vendor enum_extension="true">breezer</vendor>
> 
> when validated "heffy" would be invalid, but "breezer" is 
> clearly marked as existing outside the predefined list of 
> bicycle manufacturers.
> 
> 
> 
> Pete Cordell wrote:
> >
> > Gee, thanks.  Believe it or not I had already tried that, but it 
> > didn't actually give me any additional insight into what 
> the OP was suggesting.
> > About as useful as suggesting:
> >
> > 
> http://www.google.com/search?hl=en&q=computer+science&btnG=Google+Sear
> > ch
> 
> 


From petexmldev@t... Thu May 03 15:10:59 2007
Received: from aji.w3.org ([13