regression test system bug fixes, eliminate MOM6 warnings (#2197), add xr_cnvcld flag to FV3 (#2185)

[SamuelTrahanNOAA]

Commit Queue Requirements:

• Fill out all sections of this template.
• All sub component pull requests have been reviewed by their code managers.
• Run the full Intel+GNU RT suite (compared to current baselines) on either Hera/Derecho/Hercules
• Commit 'test_changes.list' from previous step

---

Description:

Part 1: atparse.bash

The tests/atparse.bash ignored the last line of the input if it lacked an end-of-line character (see #2199).

Now it will process such a line, and insert an end-of-line at the end if one was missing. (Some regression tests assume the output ends with an end-of-line.)

Also, I commented the file and updated the test program.

Part 2: Regression test system Rocoto fixes

A stray <partition></partition> on Hera rocoto_workflow.xml caused sporadic job failures on the new Hera Rocky nodes (#2206)

Inconsistencies in Rocoto logging have been corrected (see #2207)

And there has been some general clean-up in the tests/ directory, such as only loading rocoto or ecflow modules when needed and replacing tabs with eight spaces.

Parts 3 & 4: Combined two other pull requests to this one.

Pull requests #2185 and #2197 have been merged into this branch. Neither should change results. Preliminary testing on Hera and Hercules has shown no changes to baselines from this combined PR.

Commit Message:

* UFSWM - atparse.bash: correctly handle input that doesn't end with an end-of-line character. Fix some bugs in Rocoto support and clean up rt.sh.
  * FV3 - namelist flag xr_cnvcld to control if suspended grid-mean convective cloud condensate should be included in cloud fraction and optical depth calculation in radiation in the GFS suite
    * ccpp - physics-level changes to implement new namelist variable
  * MOM6 - update MOM6 code to eliminate all compiler warnings

Priority:

• Normal

Git Tracking

UFSWM:

• Fixes Bug in atparse script #2199
• Fixed MOM6 issue 128
• Fixes Empty partition tag causes some regression tests to fail on Hera Rocky using Rocoto #2206
• Fixes Inconsistent logging in rt.sh Rocoto mode #2207
• Fixes update MOM6 to eliminate compiling warning #2196
• Includes PR update MOM6 to remove compiling warning #2197
• Fixes Address convection/cloud/radiation interaction at coarse resolution  #2184
• Includes PR Introduce namelist flag xr_cnvcld to control if suspended grid-mean convective cloud condensate should be included in cloud fraction and optical depth calculation in radiation in the GFS suite.  #2185

Sub component Pull Requests:

UFS-WM PR 2185
FV3: sub-PR 799
ccpp-physics: sub-PR 184

UFS-WM PR 2197
MOM6 sub-PR 130

UFSWM Blocking Dependencies:

• None

---

Changes

Regression Test Changes (Please commit test_changes.list):

• No Baseline Changes.

Input data Changes:

• None.

Library Changes/Upgrades:

• No Updates

---

Testing Log:

• RDHPCS
  • Hera
  • Orion
  • Hercules
  • Jet
  • Gaea
  • Derecho
• WCOSS2
  • Dogwood/Cactus
  • Acorn
• CI
• opnReqTest (complete task if unnecessary)

[SamuelTrahanNOAA]

I'm running regression tests on Hercules now. Will report back when I have something interesting to say about them.

[BrianCurtis-NOAA]

@SamuelTrahanNOAA If you get some time, could you cleanup that script with: https://www.shellcheck.net/ i have work moving towards a GH action to check bash scripts, and am working towards updating all scripts in UFSWM to comply.

I ask because I am not sure I 100% understand this file, which might come in handy to comply with shellcheck, and it seems you might.

If not, that's fine, it will be worked on either way.

[SamuelTrahanNOAA]

I ask because I am not sure I 100% understand this file

Hopefully the comments I'm adding will help people understand it.

[SamuelTrahanNOAA]

I ran it through ShellCheck and found only one genuine error. ShellCheck's other complaints were limitations of its parser.

[BrianCurtis-NOAA]

I ran it through ShellCheck and found only one genuine error. ShellCheck's other complaints were limitations of its parser.

Thanks!

[SamuelTrahanNOAA]

Some of the regression tests assume atparse always outputs a trailing end-of-line character. I'll update it to add the missing end-of-line.

[SamuelTrahanNOAA]

And now it works!

All tests passed on Hercules with no changes to baselines.

[jkbk2004]

@SamuelTrahanNOAA I always wonder why we use @[] for parsing parm namlist files. I mean why not ${}. Parsing with environment might be a handy approach to parse with simple bash script. Any idea?

[SamuelTrahanNOAA]

I always wonder why we use @[] for parsing parm namlist files. I mean why not ${}. Parsing with environment might be a handy approach to parse with simple bash script. Any idea?

Using ${varname} prevents you from parsing a shell script with atparse. The syntax must be easily identifiable from any other language it may parse. No languages that I know of use @[varname], so that's what I went with.

[SamuelTrahanNOAA]

The full version of atparse is more capable. You can do things like @[varname:-default value], there are if/else-if/else/end-if equivalents, and you can compare variables. The bash version only has @[varname].

[BrianCurtis-NOAA]

Also curious, why are we using atparse? What was the need it met?

[jkbk2004]

It appears @lisa-bengtsson is on leave until Friday, so it is unlikely we'll hear back from her. The commit message has a sentence straight from her PR description, so the content should be correct. I think we can use it even if we don't hear back.

Yes, @lisa-bengtsson OK'd to merge once all tests run successfully. The commit messages look good.

[SamuelTrahanNOAA]

I see two problems left:

The FV3 PR has not been reviewed: NOAA-EMC/fv3atm#799. If there are problems, someone will need to edit the branch to fix them.

Once the CCPP PR is merged (ufs-community/ccpp-physics#184) someone will need to update .gitmodules in the branch for the FV3 PR (NOAA-EMC/fv3atm#799) before the FV3 PR can be merged.

[FernandoAndrade-NOAA]

FYI Gaea is currently very unresponsive today to the point that it is taking a long time to clone and does so incorrectly. I've opened up a support ticket with them to see what's going on.

[jkbk2004]

No baseline change! we can skip Gaea due to the system issue. We can follow up with Gaea in the next PR if there is issue.

[SamuelTrahanNOAA]

We are changing the rt.sh slightly, and it is possible that change will break GAEA. Unlikely, but possible. Consider running only a few of the regression tests on GAEA when it comes back to see if the regression test system works.

[BrianCurtis-NOAA]

Acorn needed an adjustment to the changes where the ecflow if statements were, i'll push them after I run testing. Give me time to get those tests running.

[BrianCurtis-NOAA]

There's still something not quite right with Acorn using ecflow. The problem i'm having happens with both the current dev branch and this branch, so it's not a UFSWM issue, yet, at least. It's not worth holding this PR up. Please skip Acorn.

[zach1221]

There's still something not quite right with Acorn using ecflow. The problem i'm having happens with both the current dev branch and this branch, so it's not a UFSWM issue, yet, at least. It's not worth holding this PR up. Please skip Acorn.

Ok, I think we're skipping Gaea as well, due to continued latency. So, we can begin the final review/merging process.

[zach1221]

@SamuelTrahanNOAA Fv3atm and MOM6 PRs are merged.
NOAA-EMC/fv3atm@1ba8410
NOAA-EMC/MOM6@ab7bd14