Linter CI part 1

[yaswant]

Description

Summary

Code linting via Super-Linter GitHub Action

Changes

Added

• Linter workflow file workflow/lint.yml
• Some manual language-specific overrides in linters/
• CODEOWNERS, currently @ssdteam but we can revisit and include individuals for each component in a separate PR
• A PR template

Dependency

None.

Impact

Automatic checks to assist code review

Issues addressed

Resolves

#51

Coordinated merge

NA

Checklist

• I have performed a self-review of my own changes

[yaswant]

some minor changes in ruff.toml

[yaswant]

• @yaswant to update shell scripts (Fixes Fix UMDP3 checker linting issues #53), switch off ruff linting for now..
• @james-bruten-mo to blackify Python scripts

[yaswant]

@james-bruten-mo Don't think there is anyway to override black defaults for SuperLinter. I see now SciTools/iris switched from black to ruff. So there must be a way AVD use the tool. I take the message that if you prefer black, then just don't try to override defaults via command-line.

[yaswant]

This is now ready for review and merge. @Pierre-siddall you can step away from the PR for now, but FYI we have disabled ruff in the workflow which will be revisited in later.

Requesting reviews from:

• @james-bruten-mo to check implementation changes
• @r-sharp to check POSIX compliance in the script we discussed in teams chat
• @t00sa workflow implantation
  Thanks

[james-bruten-mo]

I've just used this branch for the kgo install I've been doing today. This highlighted a couple of issues:

• The xc40's require scitools in order to have python3 available - this causes f-string issues when running kgo_update.py - I think we can just add module load scitools to the command in the new function as there's no problem with having it loaded on other platforms.
• The new printf syntax doesn't colour the strings correctly - instead it just prints out the colour code string. I tested Sams fix from Fix UMDP3 checker linting issues #53 which is to use echo -ne instead and that seems to work so suggest we go with that? I highlighted the locations in meto_run_kgo_script.sh but not in meto_update_kgo.sh.

Cheers

[Pierre-siddall]

This is now ready for review and merge. @Pierre-siddall you can step away from the PR for now, but FYI we have disabled ruff in the workflow which will be revisited in later.

Requesting reviews from:

• @james-bruten-mo to check implementation changes
• @r-sharp to check POSIX compliance in the script we discussed in teams chat
• @t00sa workflow implantation
  Thanks

@yaswant gotcha I'll step back from checking this out and focus on other stuff.

[james-bruten-mo]

I've had another go at installing kgo with this branch and hit issues again.
Given those changes are becoming reasonably large, can I suggest that we split a refactor of the the kgo shell script off into a separate PR, and just enforce shellcheck compliance (probably with some exemptions) in this ticket? Good development practice would probably suggest this ticket should only be for achieving linter compliance anyway.
We can then test the install script changes more thoroughly in slower time.

[yaswant]

I've had another go at installing kgo with this branch and hit issues again. Given those changes are becoming reasonably large, can I suggest that we split a refactor of the the kgo shell script off into a separate PR, and just enforce shellcheck compliance (probably with some exemptions) in this ticket? Good development practice would probably suggest this ticket should only be for achieving linter compliance anyway. We can then test the install script changes more thoroughly in slower time.

I've reverted the script back to the way it was before with fixes against shellcheck. Please give it another go.

[james-bruten-mo]

I've tested again and the printf statements with the colour codes at the end don't colour the line correctly. I've reverted these and added the relevant shellcheck ignores if that's an acceptable way to go?