-
Notifications
You must be signed in to change notification settings - Fork 64
CCPP Framework Meeting Minutes 2021 03 23
Dom Heinzeller edited this page Mar 23, 2021
·
3 revisions
- Process for creating a master cross-lab standard name resource
- How long do we have to support python 2?
- Have to jump through a lot of hoops to keep things running (e.g., only updating modified files).
-
capgen.pyprogress - Transition to
capgen.py - Other business
- We need a naming convention, the details are not as important as long as there is a naming convention
- Identify process/rules how to generate standard names
- Smaller group to meet for furter discussion (this is a time consuming exercise)
- Should NCAR have the list that Ligia and Grant have been working on? Add to repository as starting point?
- Yes, then Dave can communicate this list to the NCAR CCPP team
- The spreadsheet that Ligia and Grant created is a better starting point than the PDF that ccpp_prebuild.py generates each time it runs
- Update the standard names that have been defined until now in a feature branch
- Then create the pdf/html and share with NCAR (but this contains only the names )
- Parallel effort/target for this unification: CCPP standard names repository in ESCOMP organization
- Having the list of standard names in the ESCOMP organization means that developers have to add it there when they create a new name in the physics
- This could be automated in the future
- We could modify ccpp_prebuild.py (capgen.py) to output a format that an ESCOMP ingest/compare tool can understand
- This could be the master baselist from which to start from
- Some conflicts need to be resolved, for example: x/y winds versus eastward/northward winds
- No need for Python 2 from UFS, MMM neither; Steve will check
- On Cheyenne, need to load module python (3.7.9), ncarenv (1.3) and then run ncar_pylib
- Ongoing (from both ends), we have a PR from Julie that fixes some of the metadata/Fortran inconsistencies in the current ccpp-physics
- n/a