RE: Is xdt:anyAtomicType itself atomic?

XQuery's type system and the subtype hierarchy are based on named types.
A named union type in that system is defined as being a subtype of
xs:simpleType and not a subtype of xs:anyAtomicType. So in that respect
the type itself is not atomic. 

However, an instance of that type will be of either xs:integer or
xs:string (and not of type T) and will be an atomic typed value.

So in general, union types (and list types) are not considered atomic
types. They are simple types and can only be used to designate the type
of a node (attribute or element). The typed value of such an attribute
or element will always be one of the member types of the union or list
and will be of an atomic type. 

Best regards
Michael 

> -----Original Message-----
> From: Paul J. Lucas [mailto:plucas@bea.com]
> Sent: Tuesday, November 09, 2004 6:43 PM
> To: Michael Rys
> Cc: public-qt-comments@w3.org
> Subject: RE: Is xdt:anyAtomicType itself atomic?
> 
> On Tue, 9 Nov 2004, Michael Rys wrote:
> 
> > Xdt:anyAtomicType is the abstract atomic type and thus should be
> > considered atomic in a general case (function signatures and implied
> > atomization), but is special (no cast) since it is abstract.
> 
> 	OK, that begs the question: can it be generalized?  E.g., given:
> 
> 		T = xs:integer | xs:string
> 
> 	Is T atomic?  More general still: is every type T, composed
> 	only by alternation of only atomic types, atomic?
> 
> 	- Paul
> 
> 
> > > -----Original Message-----
> > > From: public-qt-comments-request@w3.org
[mailto:public-qt-comments-
> > > request@w3.org] On Behalf Of Paul J. Lucas
> > > Sent: Tuesday, November 09, 2004 3:32 PM
> > > To: public-qt-comments@w3.org
> > > Subject: Is xdt:anyAtomicType itself atomic?
> > >
> > >
> > > 	My understanding is that xdt:anyAtomicType type is defined as
> > > 	the union of all atomic types and that atomic types are only
> > > 	those such as xs:integer, xs:string, etc.
> > >
> > > 	It would therefore seem to be the case that xdt:anyAtomicType
> > > 	itself is not "atomic."
> > >
> > > 	This makes perfect sense to me, but it is correct?  Or should
> > > 	xdt:anyAtomicType itslef also be considered atomic?
> > >
> > > 	- Paul
> > >
> >
> >

Received on Wednesday, 10 November 2004 06:42:33 UTC