ccpp_prebuild.py mkstatic.py need for ccpp_loop_counter in metadata_request

[climbfuji]

Discussed in #389

^{Originally posted by grantfirl August 10, 2021}
During the IAP CCPP work, I came across an error that I haven't seen before. Specifically, I'm trying to run ccpp_prebuild.py on the new host and physics metadata. To do so, I have a simple SDF with one scheme for which there is host metadata for all variables.

When running ccpp_prebuild.py, I'm getting the following error:

Traceback (most recent call last):
File "./ccpp/framework/scripts/ccpp_prebuild.py", line 849, in
main()
File "./ccpp/framework/scripts/ccpp_prebuild.py", line 826, in main
arguments_request, config['caps_dir'])
File "./ccpp/framework/scripts/ccpp_prebuild.py", line 568, in generate_suite_and_group_caps
suite.write(metadata_request, metadata_define, arguments)
File "/Volumes/d1/grantf/code/IAP_cas-esm2/models/atm/iap/ccpp/framework/scripts/mkstatic.py", line 676, in write
group.write(metadata_request, metadata_define, arguments)
File "/Volumes/d1/grantf/code/IAP_cas-esm2/models/atm/iap/ccpp/framework/scripts/mkstatic.py", line 901, in write
ccpp_loop_counter_target_name = metadata_request[CCPP_LOOP_COUNTER][0].target
KeyError: 'ccpp_loop_counter'

However, from the console output, ccpp_loop_counter is being parsed on the host side from ccpp_types.F90.

It looks like this error crops up when ccpp_loop_counter is not being requested by any of the physics in a suite? Is this necessary or is there a workaround?

[climbfuji]

#436 addresses this issue.