Extended fix for cl-like variables of CESM2 models

[schlunma]

This PR extends the fixes for cl, cli and clw for the CESM2 models (the original fix only flips the vertical coords, but not the data itself).

I also had to move the function var_name_constraint to a new module (iris_helpers) because of circular dependencies in esmvaltool.cmor._fixes.shared and esmvaltool.preprocessor.shared. I'm happy to rename that module if anybody knows a better name.

Tasks

• Create an issue to discuss what you are going to do, if you haven't done so already (and add the link at the bottom)
• This pull request has a descriptive title that can be used in a changelog
• Add unit tests
• Public functions should have a numpy-style docstring so they appear properly in the API documentation. For all other functions a one line docstring is sufficient.
• If writing a new/modified preprocessor function, please update the documentation
• Circle/CI tests pass. Status can be seen below your pull request. If the tests are failing, click the link to find out why.
• Codacy code quality checks pass. Status can be seen below your pull request. If there is an error, click the link to find out why. If you suspect Codacy may be wrong, please ask by commenting.
• Please use yamllint to check that your YAML files do not contain mistakes
• If you make backward incompatible changes to the recipe format, make a new pull request in the ESMValTool repository and add the link below

If you need help with any of the tasks above, please do not hesitate to ask by commenting in the issue or pull request.

---

Related to #538 and #539.

[axel-lauer]

Just tested these fixes for the CESM2 model family with variables "cli" and "clw". It works! Nice!

[schlunma]

Are the failing tests related to #189? @bouweandela @valeriupredoi

[valeriupredoi]

can you merge the latest master in this one pls @schlunma 🍺

[schlunma]

I did (one hour ago).

[valeriupredoi]

pytest-flake8=1.0.5 is being superflaky

[schlunma]

Ready to merge from my side

[bouweandela]

@mattiarighi Could you please test?

@schlunma Is there no other way to fix the formula terms than changing the file? Because this is really inefficient and I was hoping to get rid of that function at some point..

[axel-lauer]

@mattiarighi I'll rerun the tests and merge this PR in case the tests are successful.

[schlunma]

@schlunma Is there no other way to fix the formula terms than changing the file? Because this is really inefficient and I was hoping to get rid of that function at some point..

I invested quite some time in this and came to the following conclusion: It is possible without modifying the file, but the code necessary for this is considerably longer (it involves adding all coordinates, their bounds and manually adding the coordinate factory). Since the attribute formula_term is simply wrong, fixing this also feels like the "natural" way to fix this file.

[bouweandela]

it involves adding all coordinates, their bounds and manually adding the coordinate factory

Maybe there is some code in iris that you could copy/paste? It would be a lot faster, since file fixes copy the entire file before starting to do anything. If you don't have the time right now, maybe keep it as an idea for the future?

[schlunma]

Yes, I agree that this is slower. I will keep it in mind, but as I already said, it's way more complicated and also doesn't feel like the right fix.

[axel-lauer]

Tests were successful, PR is read to be merged...