Support cyclic-import message with --jobs

[jacobtylerwalls]

Type of Changes

	Type
✓	🐛 Bug fix

Description

Closes #4171

[codecov]

Codecov Report

Merging #8807 (61f551f) into main (d203de7) will increase coverage by 0.01%.
The diff coverage is 87.50%.

❗ Current head 61f551f differs from pull request most recent head fac0949. Consider uploading reports for the commit fac0949 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #8807      +/-   ##
==========================================
+ Coverage   95.87%   95.89%   +0.01%     
==========================================
  Files         173      173              
  Lines       18503    18510       +7     
==========================================
+ Hits        17740    17750      +10     
+ Misses        763      760       -3     

Impacted Files	Coverage Δ
pylint/checkers/imports.py	94.18% <87.50%> (-0.13%)	⬇️

... and 1 file with indirect coverage changes

[Pierre-Sassoulas]

LGTM, but I don't understand why we cover the reduce without covering the get_map_data.. Should we pragma: no cover it ?

[jacobtylerwalls]

I verified that the line is being hit, but it's hit during parallel processing, so there could be a bug in codecov. Let's not add a pragma for something that's a bug in codecov.

[Pierre-Sassoulas]

The [skip ci] in the terminal commit prevent the CI from running properly.

[jacobtylerwalls]

The [skip ci] in the terminal commit prevent the CI from running properly.

I did it on purpose to avoid wasting resources on linting a commit that removed lines from an RST file. I'll just override :-)

[jacobtylerwalls]

Also, GitHub was not showing additional commits this morning, I opened a bug report with them and they just said "it happens sometimes" and didn't show it on githubstatus or anything.

[jacobtylerwalls]

Also, the primer was failing on the first commit because the project cache was missing. It's possible that without frequent enough commits on main, the PRs will always fail because the needed cache gets evicted?

[Pierre-Sassoulas]

I was wondering about the primer too. Funny that we would have an unusual drop in commit frequency and hitting never seen bugs 😄 (I blame the high temperatures, and the attention astroid is getting lately).