Changed ddt_type metadata keyword to just type. #383
Conversation
Added infrastructure and a method to find the intrinsic elements of a compound array or DDT
gold2718
added
the
capgen-unification
| found in this variable (via finding a variable in <check_dict> with | ||
| the same standard name). | ||
| Currently, an array of DDTs is not processeed (return None) since | ||
| Fortran does not support a way to reference those elements. |
There was a problem hiding this comment.
I am not sure of the implication of this, hope we can clarify at the ccpp-framework meeting.
There was a problem hiding this comment.
I guess it depends on whether or not the NOAA models or the SCM will ever need this feature. We generate code to make sure that all input variables in the active physics suite have been initialized. To do this, we need to be able to look into arrays of fields or into DDTs to find fields that could be initialized (2D or 3D physical fields). Is this something you will likely need?
There was a problem hiding this comment.
I am not sure. Maybe we should look at some specific examples of how the UFS works at the moment in our next developer meeting (or in a separate meeting) to see if this is already an issue or not.
There was a problem hiding this comment.
Okay. You can prepare by answering the question:
At run time, does / will the UFS need to query the standard names used by a suite?
There was a problem hiding this comment.
I don't think so. We have never used the standard names at run time (after switching to the static build 2 years ago), only at build time.
Also, added infrastructure and a method to find the intrinsic elements of a compound array or DDT
closes #377