[mpas-developers] additions to MPAS registry and I/O

[Michael Duda]

Todd,

you're exactly right -- this would allow us to specify the
time-invariant equivalent of a variable like xtime in the
registry. The first use that we had in mind was to specify
vertical grid parameters such as model top pressure and reference
pressure.

Cheers,
Michael


On Wed, May 05, 2010 at 04:02:58PM -0600, Todd Ringler wrote:
> 
> Hi Michael,
> 
> The change looks good to me. This would seem to be the time-invariant  
> counterpart to variables like xTime.
> 
> Cheers,
> Todd
> 
> On May 5, 2010, at 3:44 PM, Michael Duda wrote:
> 
> >Hi, All.
> >
> >I apologize if this already went through to the mailing list
> >yesterday; I suspsect it might not have due to the size of the
> >attachments, which I've replaced with references to files in
> >a repository branch.
> >
> >Currently, it's not possible to define a time-invariant, real
> >scalar field in the registry; however, I've got a set of changes
> >to the registry and framework code that enable this, and I've
> >tested them in the context of the mpas_cam_coupling branch.
> >Essentially, the changes would make it possible to define
> >something like
> >
> >     var real    ptop ( ) iro ptop - -
> >
> >that would be accessible in the code as
> >
> >  block % mesh % ptop % scalar
> >
> >The files that I'm proposing to commit to the trunk can be found
> >in https://svn-mpas-model.cgd.ucar.edu/branches/mpas_cam_coupling
> >as:
> >
> >  src/framework/module_io_input.F
> >  src/framework/module_io_output.F
> >  src/registry/gen_inc.c
> >
> >The changes will have no effect on current code, but would just
> >expand the dimensionality of fields specifiable in the registry.
> >If anyone has any objections or suggestions for improvements,
> >please feel free to comment.
> >
> >Cheers,
> >Michael
> >_______________________________________________
> >mpas-developers mailing list
> >mpas-developers at mailman.ucar.edu
> >http://mailman.ucar.edu/mailman/listinfo/mpas-developers
>