CCPP Framework Meeting 10‐03‐2024

Agenda

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

---

Attendees: Mike Kavulich,

GitHub issues/PRs

CCPP Framework (issues, PRs, discussions)

• Add diagnostic object to CCPP-framework Issue#579

  • Further discussion later

• Add constituent tendency capability PR#584

  • Several people have offered comments, but still needs a full review or two

• Add constituent minimum values array interface Issue#585

• Return full error message from xmllint PR#586

  • Needs an approval from the NOAA/UFS side.

• Constituents object using wrong standard name Issue#587

• capgen's Fortran parser does not understand array declarations that have lower bounds specified Issue#588

  • Dustin will provide an example

• capgen host model DDT problem Issue#589

  • Problems handling ccpp_error_message and ccpp_error_code

Drafts

• DRAFT TO DISCUSS: Add support and testing for equivalent units PR#571

• DRAFT TO DISCUSS - add CCPP register phase PR#582

Standard names (issues, PRs, discussions)

• Update StandardNamesRules.rst: changesurface prefix to _at_surface suffix PR#72
  • Dom added: Can we come to a decision for https://github.com/ESCOMP/CCPPStandardNames/pull/72 - JCSDA is blocked by this

New items for discussion

Topic of the week: Add diagnostic object to CCPP-framework #579

Updates from last time

Previous notes

Other business

Meeting notes

CCPP Framework

• Constituent tendencies (#584)
  • Michael K to do a review
  • Steve G also reviewing
• XMLlint error message (#586)
  • Michael K to do a review
• Capgen’s fortran parser does not understand array declarations that have lower bounds specified (#588)
  • Dustin will provide example
  • In SCM: Community lake model and NOAA MP both have this lower bound difference
• Capgen host model DDT problem (#589)
  • Host model pass errcode and errmsg to cap, but should be able to pass DDT (which includes error message and error code)
• No updates on draft PRs

Standard Names

• Update rules - surface_* vs *_at_surface (#72)
  • Disagreement on which path forward
  • JEDI folks prefer “surface_*” to not have to change things on their side
  • Will move forward with those in the meeting approving the PR
    • Michael K to add a comment and approve
    • Grant added his comment & approval
• We don’t have people from JCSDA or the UK Met office at our meetings, so they can’t provide their input on standard names
  • Another meeting may be necessary to keep them included
  • They may not know this meeting is where we talk about standard names
    • Dom - could make it the second half of this meeting - they have a meeting from 9-10
    • Also includes wider JEDI audience (but Dom can represent for now)
    • Future attendees, possibly: MPAS, NOAA DA

Discussion

• Dustin - SCM is close to working with capgen
  • Opened physics PR
  • Got an issue with parsing
    • Grant - could be related to F77 leading ampersand
    • It’s definitely a bug we want to fix
    • Michael W volunteered to take a look at regular expression
    • Michael K - might want to stop supporting F77
    • Michael W - might want to consider porting to an external parser
      • Dom - would need to get 3rd party library through extensive approval for operational
      • Courtney - we’re not doing anything wild with our parser (just validating vs metadata)
      • Michael K - if we start parsing more complicated things, perhaps that’s what triggers a move to an external library
• Diagnostic object
  • None of us have really thought about it
  • Tabling it since CAM-SIMA folks don’t have time to implement it right now
• Dustin leading the next meeting 10/3
• In-person meeting 10/10
  • Location TBD based on SIMA homecoming event - NCAR folks to notify when they know!