[4.6] Distribution.files is None in some packages

[nicoddemus]

Backport of #5389

Question: when backporting, should we squash commits? I vote yes.

[blueyed]

Question: when backporting, should we squash commits? I vote yes.

Yeah.
On / for the PR already, or when merging (i.e. squash-merging then)?

[nicoddemus]

On / for the PR already, or when merging (i.e. squash-merging then)?

I think it is up to the person creating the PR: they can open the PR without squashing, but squashing before merging.

I think it is easier to just open the PR with commits already squashed.

[codecov]

Codecov Report

Merging #5395 into 4.6-maintenance will decrease coverage by 4.5%.
The diff coverage is 100%.

@@                 Coverage Diff                 @@
##           4.6-maintenance    #5395      +/-   ##
===================================================
- Coverage            94.28%   89.78%   -4.51%     
===================================================
  Files                  115      115              
  Lines                26362    26377      +15     
  Branches              2603     2603              
===================================================
- Hits                 24855    23682    -1173     
- Misses                1178     2363    +1185     
- Partials               329      332       +3

Impacted Files	Coverage Δ
src/_pytest/config/__init__.py	93.79% <ø> (ø)	⬆️
testing/test_config.py	99.83% <100%> (+0.34%)	⬆️
testing/test_reports.py	10.49% <0%> (-89.51%)	⬇️
testing/logging/test_reporting.py	21.03% <0%> (-78.97%)	⬇️
testing/test_runner_xunit.py	29.82% <0%> (-70.18%)	⬇️
testing/examples/test_issue519.py	33.33% <0%> (-66.67%)	⬇️
testing/test_runner.py	34.7% <0%> (-61.61%)	⬇️
testing/test_resultlog.py	53.6% <0%> (-46.4%)	⬇️
testing/test_warnings.py	26.77% <0%> (-36.07%)	⬇️
testing/test_pytester.py	55.21% <0%> (-27.89%)	⬇️
... and 37 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 937f945...0ae2771. Read the comment docs.

[nicoddemus]

Let's see what @asottile says then. 👍

[asottile]

the command I was using was git cherry-pick -m1 <themergecommit>, this looks like 3 cherry picks?

I documented the procedure I used in #5275 -- it ~should end up as a single commit on the backport branch

[nicoddemus]

Closing in favor of #5401