[Versa] Issue 100A proposal, extension functions
Chimezie Ogbuji
chimezie at gmail.com
Sun Sep 4 08:59:05 MDT 2005
>
> I like this, plus it solves the issue of adding all of the functions
> like "rdf:type()" etc.
Yup.. don't need class (covered by rdf:type(member)) or value (covered
by rdf:value(res)). Though I think rdf-list and rdf-container are
still needed to cast RDF sets/lists to versa sets and lists,
especially since we have no literal representation for sets/list, we
mostly use the list/set functions. These two functions probably need
a section describing their use as constructors, they aren't listed in
the function section.
>
> Should we have processor produce a warning?
I'm not sure, I guess it depends on if the processor thinks using
functions this way has an ill-effect in some way or just that the user
should be aware of it. I can imagine using them much more often than
the forward traversal operator equivalents, so would I expect a
warning raised everytime I do? I'm thinking maybe it should be left up
to the implementation to decide.
More information about the Versa
mailing list