On 30 Aug, 12:26, Martin Honnen <mahotr...@yahoo.de> wrote:
> RolfK wrote:
> > Case one:
> > Combination which may exists are A1,AA2,A3 or A1,A4,A5.
> > <E A1=3D"" A2=3D"" A3=3D"">
> > <E A1=3D"" =A0A4"" A5=3D"">
>
> > Case two:
> > The existance of an attribute depnds on a the VALUE of another
> > attribute
> > A1=3Da =3D=3D> A1,A2,A3
> > A1=3Db =3D=3D> A1,A4,A5
> > A1=3Dc =3D=3D> A1,A6,A7
>
> I don't think such restrictions can be modelled with the W3C XML schema
> language.
>
> --
>
> =A0 =A0 =A0 =A0 Martin Honnen
> =A0 =A0 =A0 =A0http://JavaScript.FAQTs.com/

I agree Martin.  People recommend using things like Schematron or
Relax-NG for this type of thing.

FWIW XSD 1.1 looks like it will include functionality for this sort of
thing in the way of schematron like assertions (xs:assert).  For those
interested last call for the current draft of XSD 1.1 ends this Friday
12th Sept!  How long it will take before you can assume the additional
functionality is in your tools is anyones guess!

HTH,

Pete Cordell
Codalogic Ltd
Interface XML to C++ the easy way using XML C++
data binding to convert XSD schemas to C++ classes.
Visit http://www.codalogic.com/lmx/ for more info