- From: m.taschuk at newcastle.ac.uk (Morgan Taschuk)
- Subject: [cellml-dev] RDF Representation
- Date: Thu, 28 May 2009 12:25:19 +0100
Hi,
>
I'm not sure how you are generating these stubs... do you mean that you
>
generated CORBA stubs? The native API binding does not require CORBA
>
stubs to be generated. It uses multiple inheritance of interfaces
>
(allowed in Java) with a single implementing class for each top-level
>
object.
I'm generating the stubs with a program bundled with the Java compiler called
IDL2Java for the sole purpose of generating documentation for the CDA. I took
the IDL files in the interfaces directory of the CDA. I like to be able to
see all of the available methods in a library in Javadoc style. As a
by-product, the program generated CORBA stubs that have the queryInterface
method, so I initially thought I may be able to do something tricky with
those stubs and a few inheritance tricks. However, without multiple
inheritance, I don't see a way to subclass both the IObject necessary for
queryInterface and the RDFRepresentation class created for the Java bindings.
>
Please feel free to implement this and submit a patch if you
>
feel you can.
Unfortunately, I don't speak C++ at all, so my implementing it would probably
take longer.
>
Things work as they stand, so it's not like we cannot
>
work, unlike Morgan, from what I understand.
Yes, this oversight effectively stops all of my work with the CDA, since my
primary motivation is to semi-automate the annotation of CellML models with
RDF. I would really appreciate any work on this problem. I'll be following
the tracker item with bated breath and high anticipation. :)
Thanks,
Morgan
_______________________________________________
cellml-tools-developers mailing list
cellml-tools-developers at cellml.org
http://www.cellml.org/mailman/listinfo/cellml-tools-developers
Archive powered by MHonArc 2.6.18.