Black formatting of existing code and inclusion of formatting check in workflow

[JuliaKukulies]

No description provided.

[w-k-jones]

Is this the autoformatting making these changes? Might have to restrict it from these files (i.e. non-python files) as I think this would break them

[JuliaKukulies]

You are right, I fixed this!

[freemansw1]

I think we should merge this into dev rather than main. Also, I'm confused as to why the docker check hasn't run yet...

[w-k-jones]

Just changed the base to dev, thought that that might run the docker checks but it doesn't seem to

[freemansw1]

My guess is that the issue is that docker-image.yml specifies the branches to run it on and still calls main master. I've gone ahead and just removed the branch list, so it should work once @JuliaKukulies pulls the changes into her repo (it should rerun the actions).

[JuliaKukulies]

Great, seems to work now @freemansw1 !

[freemansw1]

Now that it's passing the CI checks, I'm happy for this to be merged, but I'll let @w-k-jones as it's his review.

[fsenf]

Hi, I was just about to merge the PR, but then did the black reformatting by myself to be really on the safe side. Interestingly, I don't get exactly the same results. In my case, some more spaces have been added, e.g.

> git diff fabians-PRcheck..juliasPR 
...

 def calculate_cog_untracked(mass, mask):
-    """ caluclate centre of gravity and mass for untracked parts of domain
+    """caluclate centre of gravity and mass for untracked parts of domain
     Input:
-    mass:       iris.cube.Cube 
+    mass:       iris.cube.Cube
...

Any idea what happend?

[fsenf]

And yes, I was running black via:

> for py in `find -iname '*.py'`; do echo $py; black $py;done

using

> conda list black
# packages in environment at /home/senf/.miniconda3:
#
# Name                    Version                   Build  Channel
black                     19.10b0                  py37_0    conda-forge

on my linux OS

[fsenf]

@sean , @w-k-jones & @JuliaKukulies : If you don't think that these super-small differences matter, please press the merge button and step forward!

[JuliaKukulies]

Hmm..that is interesting. I used black 22.1.0. I tried to see if the same thing happens, when I use the same version as you:

> conda list black 
# packages in environment at /home/juli/anaconda3/envs/blacktest:
#
# Name                    Version                   Build  Channel
black                     19.10b0                  py37_0    conda-forge

but all files remain unchanged.

These are really small differences, but on the other hand the idea with black is to have everything exactly consistent. If you go down the rabbit hole of formatting (section "empty lines"), I think it is against PEP-8 to have empty lines directly after the function definition, since you already have an indentation. Do you pass the formatting check when you push your changes?

[w-k-jones]

Hi, I was just about to merge the PR, but then did the black reformatting by myself to be really on the safe side. Interestingly, I don't get exactly the same results. In my case, some more spaces have been added, e.g.

> git diff fabians-PRcheck..juliasPR 
...

 def calculate_cog_untracked(mass, mask):
-    """ caluclate centre of gravity and mass for untracked parts of domain
+    """caluclate centre of gravity and mass for untracked parts of domain
     Input:
-    mass:       iris.cube.Cube 
+    mass:       iris.cube.Cube
...

Any idea what happend?

Not sure why your black formatting has included the extra spaces, however I believe that the PR is formatting correctly (without spaces) so I'm happy to merge if there are no other exceptions

[fsenf]

OK, as a last check: My black version on my OS leaves all files in @JuliaKukulies PR unchanged, so we won't have any circular PRs in future ... Let's go forward ...