Hi all, I am working on refactoring code that might be useful for code generation as well as a range of other applications into their own components. While nothing is stabilised yet, there are a number of already two components that are probably complete enough to be usable right now (note that the CCGS doesn't use the new services yet, this will probably be the last step of the refactoring process, if it is even done, which still needs to be discussed further). I also plan to use the new services to build a system for dealing with non-ODE models as part of my PhD.
Changes are being checked in to https://svn.physiomeproject.org/svn/physiome/CellML_DOM_API/branches/refactor_ccgs_branch/ The new services have to be explicitly turned on using newly added configure options. The following new modules have been implemented and have had tests written: AnnoTools: Provides a simplified interface for setting and getting pooled annotations on CellML objects (all annotations are removed at once when the pool is destroyed, and each pool places its unique identifier in the annotations, so different sets of annotations will not overlap). See the interface at: https://svn.physiomeproject.org/svn/physiome/CellML_DOM_API/branches/refactor_ccgs_branch/interfaces/AnnoTools.idl CUSES: Provides unit simplification and expansion services. It is not clear whether this needs to provide more attributes and operations, but having the object framework in place will make it easy to add these. See the interface at: https://svn.physiomeproject.org/svn/physiome/CellML_DOM_API/branches/refactor_ccgs_branch/interfaces/CUSES.idl I have also started writing an IDL for CeVAS (the variable association service, used to efficiently handle the problem of determining which variables are connected). This is available at https://svn.physiomeproject.org/svn/physiome/CellML_DOM_API/branches/refactor_ccgs_branch/interfaces/CeVAS.idl. I welcome any feedback about the work done so far, and in particular feedback about how useful the current interfaces are to people working with CellML, and any suggested improvements (or patches :) ). Best regards, Andrew Miller _______________________________________________ cellml-discussion mailing list [email protected] http://www.cellml.org/mailman/listinfo/cellml-discussion
