Re: exclusions

[Eric van der Vlist <vdv@dyomedea.com>]

James Clark wrote:
> 
> The <restriction> element in XML Schema is used for restricting both
> simple types and complex types.  It's use for restricting simple types
> depends on the facets defined in XML Schema Part 2.  The whole idea of
> restricting a simple type depends on having a system of datatypes that
> has a set of facets defined.  This kind of restriction is part of the
> datatype system, and TREX already supports this.  

The restriction in XML Schema is kind of weird since you can have a
larger scope of possibilities after a restriction than before since for
most of the facets the new constraints are replacing previous ones even
if they are less restrictive.

Do we want to keep this behaviour for TREX ?

.../...
> 
> I don't see a whole lot of connection between this sort of restriction
> and excludeElement.

There might be... instead of a restriction ala XML Schema, I would
rather have some kind of generic intersection, union and exclusion
mecanisms that would allow these basic operations on the possible
patterns and datatypes.

The union can be done by a "choice" (it's allowing a scope of possible
patterns that is the union of the scopes of its children elements).

Exclusion is the subject of your mail and can be seen as complementing
the scope of its children.

It would be nice to have a restriction that would work as an
intersection -even if it's redundant and A intersection B can be written
like (A union B) excluding ((A excluding B) union (B excluding A))-

(wish I add a mathml enabled mailer !)

Eric

-- 
Rendez-vous à Paris pour net2001.
                         http://www.mynet2001.net/pgmonline2001/it2.html
------------------------------------------------------------------------
Eric van der Vlist       Dyomedea                    http://dyomedea.com
http://xmlfr.org         http://4xt.org              http://ducotede.com
------------------------------------------------------------------------