RE: building an object model of a XML schema

[Jeff Lowery <jlowery@scenicsoft.com>]

> Remember that UML as it stands now has some degree of flexibility, you
can
> use some of your own symbols and apply your own semantics to them.  An
> example is OPEN's use of more clearly specified notions of composition
and
> containment.  So there might be some scope for adapting UML for your
needs.

Well, that's hopeful. 

...
> I don't really want to model objects or XML, what I'm looking for is
an
> abstract mapping of types.

.The trouble with this is, I think, that xml "types" generally include
.aspects of the xml syntax - that is, markup issues - but you wouldn't
.want
.this things getting into your abstract "object" model, I imagine.

Yeah, I think I said something about removing the XML-ness from XML Schema
to derive this new schema language, then add in some OO archetypes (types
with well-understood implicit behaviors BUT NO INTERFACE AND NO
IMPLEMENTATION!!! -- thats up to the code generator).

.ASN.1 (yes, that again!) would seem to get quite close to what you have
.said, but it's info set (equivalent) naturally doesn't have many of the
xml
.infoset constructs.  So when you want to automatically map asn.1 to xml
or
.vice-versa, you have to restrict the xml structures that you use.  If
you're
.willing to do that, you would definitely have some hope of a mapping
to/from
"types.

Tom, I've warned you once and I'm not going to tell you again...

Alright, maybe I have to look into more closely. Bleccch, another language;
it never ends.

Cheers,

Tom P



------------------------------------------------------------------
The xml-dev list is sponsored by XML.org, an initiative of OASIS
<http://www.oasis-open.org>

The list archives are at http://lists.xml.org/archives/xml-dev/

To unsubscribe from this elist send a message with the single word
"unsubscribe" in the body to: xml-dev-request@lists.xml.org