[
Lists Home |
Date Index |
Thread Index
]
That makes sense. So in effect, unless datatypes are generated at runtime
(I don't know how), the aggregate set of datatypes to be supported must
somehow be inspected and checked against the local framework/opsys library.
If they aren't there, there has to be a place to go download them from
pretty much the way we treat plugins. RELAX doesn't care; it just notifies
the system or user about the requirements. Is that it?
Wouldn't that reinforce pushing toward a single framework/platform given
the costs of attempting to provide support for any given one? I am thinking
of what happens to those of us who have Oracle, SQL Server, etc, backends
and are mapping before starting up ODBC. Hmm. I'm probably confused about
just what the bolt-in can achieve with RELAX. I'll have to study it more.
len
From: John Cowan [mailto:jcowan@reutershealth.com]
Bullard, Claude L (Len) scripsit:
> What happens in an instance if it contains productions in different
> namespaces where each has a RELAX NG schema but each schema references
> a different data type library? I assume the validator uses the different
> definitions according to the namespace and does not merge these.
Correct. RELAX NG does not know about relationships between datatypes
even in the same library, still less in different libraries.
Since each datatype has a (hopefully) unique {namespace name, local name}
pair identifying it, there can be no confusion between different types,
but there is no ability to merge them either.
|
