feat: add --config cli argument

[reiddraper]

This is useful for projects that want to provide their own proselintrc
which is stored in the repository for the project itself. Further, it's
useful for installations where putting a config-file outside of a
sandboxed directory is not possible.

---

Resolves #356. Resolves #1091.

[reiddraper]

Looks like there's a few build failures related to different APIs on different Python versions. Before I fix each of them, is this a contribution you'd be interested in?

[carlsmedstad]

Apart from needing a rebase on main and some clean up to pass the CI, this looks good to me.

Should @reiddraper continue this work? @Nytelife26 @suchow

[Nytelife26]

Thanks for bringing this to my attention, @carlsmedstad. I'll get to work on bringing it up to date and merging now, since it is a useful feature. We could also put automatically detecting proselintrc files up to the filesystem boundary as part of this PR, too, but that's another thing.

[carlsmedstad]

Great, detecting project specific config files seems like something that would be useful after this has been merged. This would resolve #1112 and #702.

I'm a proponent of the pattern of having three levels of configuration: system, user and project scope.

[Nytelife26]

I'm a proponent of the pattern of having three levels of configuration: system, user and project scope.

Me too ultimately. The user configuration should check `~/.config` too, though.

[codecov]

Codecov Report

Merging #1081 (905ead6) into main (50ac6b9) will increase coverage by 0.98%.
The diff coverage is 95.65%.

❗ Current head 905ead6 differs from pull request most recent head b825a57. Consider uploading reports for the commit b825a57 to get more accurate results

@@            Coverage Diff             @@
##             main    #1081      +/-   ##
==========================================
+ Coverage   89.20%   90.19%   +0.98%     
==========================================
  Files          83       83              
  Lines        1214     1203      -11     
==========================================
+ Hits         1083     1085       +2     
+ Misses        131      118      -13     

Flag	Coverage Δ
macos-latest	90.19% <95.65%> (+0.98%)	⬆️
py3.6	89.22% <95.65%> (+1.06%)	⬆️
py3.7	89.22% <95.65%> (+1.06%)	⬆️
py3.8	90.19% <95.65%> (+0.98%)	⬆️
py3.9	90.19% <95.65%> (+0.98%)	⬆️
pypypy3	89.22% <95.65%> (+1.06%)	⬆️
ubuntu-latest	90.19% <95.65%> (+0.98%)	⬆️
windows-latest	90.19% <95.65%> (+0.98%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
proselint/command_line.py	44.23% <66.66%> (+0.54%)	⬆️
proselint/tools.py	77.56% <100.00%> (+3.74%)	⬆️

---

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 50ac6b9...b825a57. Read the comment docs.

[carlsmedstad]

Definitely, I could make a PR tomorrow if you'd like?

[suchow]

@Nytelife26 The CI run is failing here for one substantive reason (coverage) and one transition-to-Github-actions reason. In particular, something is breaking during the Danger action.

[Nytelife26]

@Nytelife26 The CI run is failing here for one substantive reason (coverage) and one transition-to-Github-actions reason. In particular, something is breaking during the Danger action.

I am aware of the coverage issue. The test suite is not currently very comprehensive, it seems. As for the Danger action, I believe that is a result of the force push I performed following the rebase, given that Danger specifies that as a potential cause in the error log.

[Nytelife26]

cc: @suchow

[suchow]

The new test looks good, just need to import subprocess up top and I think it should be good to go!

[benknoble]

This doesn't resolve #1091, which is about individual checks versus categories.