Skip to content

CCPP Framework Meeting Minutes 2024 04 22

Jump to bottom
Courtney Peverley edited this page Apr 23, 2024 · 3 revisions

Agenda

Attendees: Mike Kavulich,

GitHub issues/PRs

CCPP Framework (issues, PRs, discussions)

Standard names (issues, PRs, discussions)

New items for discussion

Updates from last time

Previous notes

Other business

Meeting notes

CCPP Framework

  • No new issues/PRs
  • Constituents
    • In progress
  • Issue #553 - rethink dev cycle
    • Michael K to try to get to it
    • Welcome feedback from others as well!
  • PR #555 - cleanup, unit tests for ccpp_track_variables
    • Fixes failure message so it comes through properly
    • Dom - every change must go through UFS testing
    • Could also be combined with another ccpp-physics PR - not answer-changing
    • Michael to merge #556 into #555
  • PR #556 - missing MPI target
    • Approved; waiting for testing
    • Grant - being combined with next ccpp-physics PR

Standard Names

  • Issue #58
    • Waiting on one more collaborator to accept invite and approve - may comment him out if he takes too long
  • Issue #63 - unusual units
    • Michael K - how do we feel about adding “other”?
      • Grant - “1” is the grab-bag
      • Michael K - that’s not explicitly stated in the definition - we should update the documentation to say that
      • Cheryl - “other” is easier to grep for
      • Michael K - could also use “none” but would be technically incorrect
      • Grant - would be nice to specify weird units in long name if we’re using the catch-all unit
    • Michael K to open a PR with the proposed change and others can chime in there
  • PR #65 - vertical stagger suffixes - merged
    • Resulted in 2 issues: #67, #68
    • Dom - the PR went in, but had last-minute disapproval - see below discussion

Discussion

  • Dom - optional arguments PR
    • Needs optional attribute in fortran code when fortran arguments are passed and associated
      • Easy to make mistakes - takes time
      • Help?
        • Dustin to help!
      • Also: Can we take capgen parser and have it do a consistency check for fortran vs metadata optional attribute?
        • Courtney to tackle
  • Handling conflicts within Standard Names repo
    • We should act faster in the future to avoid these lingering PRs that are asking for people’s opinions
    • Grant - the discussions can be very tiring, no one wants to make the final decision
    • Michael W - How do we deal with this long-term?
      • Michael K - unclear. We kind of got stuck with the standard names repo, but it’s not our primary thing
      • Dom - everyone who agrees to use CCPP framework agrees to using the standard names
      • Dom - Greg didn’t want to be code owner, so he gets overruled
        • Also have to give time limit for responding to PRs for code owners
      • Michael K - how do we specify the rules for standard names repo? Wiki entry
        • Grant - PR template?
        • Michael W - contributor’s guide / have to sign an agreement that you understand the rules?
          • Dom - ECMWF has that for when you open your first PR; could look at that language
  • Michael K - how do we keep the physics repo(s) in sync with standard name repo and how do we enforce this?
    • Courtney - Jesse’s tool in the standard names repo that checks vs standard names repo
      • Dom - could be incorporated into CI for ccpp-physics
    • Courtney - could maybe come at it from both sides and notify stakeholders when a standard name they use is changed
      • Michael K - good in theory, could be information overload/too many emails
    • Grant - agreed that we need something. Current system is not manageable.
    • Michael K to add this to his to-do list to do CI for ccpp-physics
      • Will likely be controversial/big lift at first to bring ccpp-physics up to date with standard names repo; less so after
  • we will meet next week

CCPP Framework wiki

Clone this wiki locally