CCPP Framework Meeting Minutes 2025 02 27

Agenda

• GitHub issues/PRs
• New items
• Updates
• Other business
• notes

---

Attendees: Michael Kavulich,

GitHub issues/PRs

CCPP Framework (issues, PRs, discussions)

• Add support and testing for equivalent units #571

  • On the back-burner

• Constituent index lookup #622

  • Approved, should this go in first?

• Allow for differing module/file names Issue#636

  • Capgen in SCM: Differing source and module name #637
    • Approved by Dom, still needs a SIMA reviewer

• [ccpp_fortran_to_metadata.py errors out when given a file that has a constituent properties object Issue#643])https://github.com/NCAR/ccpp-framework/issues/643)

• [https://github.com/NCAR/ccpp-framework/pull/644)

  • Approved, merge 622 then this one?

• Current intel compiler does not allow >255 subset arguments Issue#645

Drafts

• Testing refactor #612

  • Continued updates and discussion

• Capgen in SCM: DDT subfields into Group Caps #640

  • Ready for review?

Standard names (issues, PRs, discussions)

• add emissions names #86

  • Merged.

• Second rules update in v1 branch, update names, update write_standard_name_table.py to allow for subsections #87

  • Still need some discussion

New items for discussion

Updates from last time

Previous notes

Other business

Brian Dobbins has started testing the Intel compiler limit problem, and will be attending the beginning of the meeting, so it might be good to discuss this issue first if possible.

Updated branch rules:

• For main, Added "Dismiss stale pull request approvals when new commits are pushed" and increased number of required reviews to 3 (to match number of regtests)
• For develop, Added "Require linear history" (which prevents merge commits to that branch specifically). This allows me to re-enable merge commits as an option, but creates a potentially confusing message on PRs where merge commit is the default.

Offer suggestions/comments on development workflow rules:

• Latest updates were applied: https://github.com/NCAR/ccpp-framework/wiki/Development-Workflow
• New comments and suggestions here: https://docs.google.com/document/d/1iMh_iKsUrB_mOMF6lGkFggG85JV62NFkTjKJboqWvV8/edit?tab=t.0

Meeting notes

CCPP Framework

• #633 and #637 are ready for review!
• Michael W to open issue for bug in comparing fortran vs metadata local names for host variables

Standard Names

• Michael K to rename to ESMStandardNames TODAY!

Discussion

• 255 arguments issue with intel

  • It’s not a hard limit on arguments, but a hard limit on “specification expressions”
    • E.g. array bounds
  • Documentation here
  • Brian D wants more info on what’s causing the problem for Dustin, maybe an example
  • Soren: Dustin’s example wouldn’t work even after messing with flags a bit
    • ccpp_SCM_GFS_v17_p8_ugwpv1
  • Are we explicitly passing array bounds? If so, that’s likely the problem
  • Dom - need to subset chunks of array for things like radiation
    • Need to know if capgen is subsetting at the group level or scheme level
    • Courtney to look into how capgen handles this/get the example from Dustin/Soren
  • Brian’s analysis of physics meta data
• Next week’s meeting will be in-person in the Mesa Lab for those in the area