-
Notifications
You must be signed in to change notification settings - Fork 64
CCPP Framework Meeting Minutes 2024 04 15
Courtney Peverley edited this page Apr 15, 2024
·
3 revisions
Attendees: Mike Kavulich,
CCPP Framework (issues, PRs, discussions)
-
PR 549: Constituent updates
- Has comments from Steve, still needs one more review
-
Issue 553: Rethink CCPP framework development cycle and branch structure
- Discussed above
-
PR 555: Cleanup, adding unit tests for ccpp_track_variables.py, activate "chunked data" CI test
- Has comments from Steve, still needs one more review
-
PR 556: Add missing MPI target in src/CMakeLists.txt
- Approved
Standard names (issues, PRs, discussions)
-
Issue 57: Feedback on the meaning of: sea_ice_area_fraction_of_sea_area_fraction
- I still need to update this with latest discoveries about inconsistent use
-
Issue 58: Variable name changing policy and procedures
- Who should we reach out to for adding code owners?
- Issue 61: Clarification regarding _at_interface suffix
-
PR 62: Update Standard Name Rules table for dimensionless units, change variables improperly using "1" to "fraction"
- Per discussion, "frac" was updated to "fraction"
- Issue 63: Dealing with unusual units
- Issue 64: Need to replace the use of "volumetric" with "volume_fraction" when referring to soil quantities
- PR 65: Add a rule to clarify vertical stagger suffixes
CCPP Framework
- Issue #553 - Michael K to add thoughts; other to chime in as desired
- PR #555
- Michael to respond to Steve’s comments
- Dustin to also review
- PR #556 - add missing MPI target in src/CMakeLists.txt
- Approved
- Dustin - was working in SCM because it was using wrapper compilers that already had MPI linked in
- Will be merged in when UFS testing is done
- Michael K/Grant to bring up at next bi-friday meeting to be combined with next UFS PR
Standard Names
- Issue #57: feedback on meaning of sea_ice_area_fracton_of_sea_area_fraction
- Michael K to update issue/get discussion up to date
- There’s a lot more in the issue - Michael K to take a look if something has been missed that needs to be addressed
- Issue #58: variable name changing policy and procedures
- Michael K to add to wider meeting
- Dom - start with what we have and solicit feedback in PR?
- Dom to open PR
- Issue #61: clarification regarding _at_interface suffix
- Addressed in PR#65
- Dom: Remove all; other than that we’re ok
- Other people should chime in to agree/disagree on removal of “all”
- Cheryl to chime in with agreement
- Michael K to prioritize his review
- Dom: Remove all; other than that we’re ok
- Addressed in PR#65
- PR #62: update standard name rules for dimensionless units
- “frac” -> “fraction”
- Re-review requested from Jesse
- Issue #63 - dealing with unusual units
- Units of 1 is sometimes used as a catch-all (e.g. ln(Pa))
- Dom - remembering discussion from a few years ago - what does ln(Pa) mean?
- Michael K - technically a correct unit, but definitely a weird one
- Cheryl - would vote for one-off weird units of 1
- Dustin - this would actually be a simple conversion
- Courtney - what if one scheme had it in ln(mb) and another had ln(Pa) - those aren’t equivalent and would cause wrong answers if we had units = 1 for both.
- Michael W - check other unit conversion libraries - how do they handle ln units?
- Would make things easier if we ever decide to externalize our unit conversion library
- Dom - how common are these types of units? Is this worth the effort for an edge case?
- Courtney - would vote for not implementing a converter, but updating the unit from 1 to “ln(Pa)”
- Capgen can’t handle ln(Pa), so we’d have to update the regex (or possibly skip the check entirely?)
- Michael K to look into this
- Cheryl - could update to ln_Pa (generally agreement on this)
- Issue #64: need to replace the use of “volumetric” with “volume_fraction” when referring to soil quantities
- Hopefully a straight-forward correction; waiting on PR from Jesse
Discussion
- Dustin: Should prioritize efforts
- Specifically, Dom has a few things going on, so want to use his time as best as possible before he moves to the Navy
- Dom had issues with deferred-length characters with older version of GNU compiler
- Will have conflicts between Dom’s work and Dustin’s work; Dustin working to get SCM stuff up to dates optional arguments updates