Adds force-exlucde flag to ling/autocorrect commands

[ashfurrow]

This adds a --force-exclude option to the lint and autocorrect commands which will force SwiftLint to exclude a file if it matches the excluded config options, even if that file is explicitly passed with the --path option.

I could not test locally because I got the following exception:

Fixed this, ran into #1466.

➜  Moya git:(master) ✗ swiftlint
Loading configuration from '.swiftlint.yml'
Fatal error: Loading sourcekitd.framework/Versions/A/sourcekitd failed: file /Users/ashfurrow/SwiftLint/.build/checkouts/SourceKitten.git-6484296299232452758/Source/SourceKittenFramework/library_wrapper.swift, line 61
[1]    71720 illegal hardware instruction  swiftlint

Tested this locally:

➜  Moya git:(master) ✗ swiftlint lint --path Sources/Moya/AnyEncodable.swift --force-exclude --config .swiftlint.yml
Loading configuration from '.swiftlint.yml'
Linting Swift files at path Sources/Moya/AnyEncodable.swift
No lintable files found at path 'Sources/Moya/AnyEncodable.swift'

It fails because it "can't find" any files, but I think that makes sense.

Fixes #2051.

[SwiftLintBot]

	1 Error
🚫	Please rebase to get rid of the merge commits in this PR

	12 Messages
📖	Linting Aerial with this PR took 0.46s vs 0.41s on master (12% slower)
📖	Linting Alamofire with this PR took 3.65s vs 3.59s on master (1% slower)
📖	Linting Firefox with this PR took 14.0s vs 13.86s on master (1% slower)
📖	Linting Kickstarter with this PR took 21.85s vs 21.32s on master (2% slower)
📖	Linting Moya with this PR took 2.09s vs 2.04s on master (2% slower)
📖	Linting Nimble with this PR took 2.07s vs 2.02s on master (2% slower)
📖	Linting Quick with this PR took 0.6s vs 0.58s on master (3% slower)
📖	Linting Realm with this PR took 3.84s vs 3.79s on master (1% slower)
📖	Linting SourceKitten with this PR took 1.16s vs 1.17s on master (0% faster)
📖	Linting Sourcery with this PR took 5.26s vs 5.22s on master (0% slower)
📖	Linting Swift with this PR took 15.03s vs 14.77s on master (1% slower)
📖	Linting WordPress with this PR took 16.69s vs 16.61s on master (0% slower)

Generated by 🚫 Danger

[codecov-io]

Codecov Report

Merging #2056 into master will decrease coverage by 0.03%.
The diff coverage is 62%.

@@            Coverage Diff             @@
##           master    #2056      +/-   ##
==========================================
- Coverage   89.69%   89.66%   -0.04%     
==========================================
  Files         256      256              
  Lines       14907    14937      +30     
  Branches      973      973              
==========================================
+ Hits        13371    13393      +22     
- Misses       1519     1526       +7     
- Partials       17       18       +1

Impacted Files	Coverage Δ
Source/swiftlint/Commands/AutoCorrectCommand.swift	0% <0%> (ø)	⬆️
...iftlint/Extensions/Configuration+CommandLine.swift	0% <0%> (ø)	⬆️
Source/swiftlint/Commands/LintCommand.swift	0% <0%> (ø)	⬆️
...sts/SwiftLintFrameworkTests/IntegrationTests.swift	37.14% <100%> (ø)	⬆️
...s/SwiftLintFrameworkTests/ConfigurationTests.swift	94.57% <100%> (+0.55%)	⬆️
...ework/Extensions/Configuration+LintableFiles.swift	89.47% <75%> (ø)	⬆️
Source/SwiftLintFramework/Models/Command.swift	95.65% <0%> (-2.18%)	⬇️
...SwiftLintFramework/Extensions/File+SwiftLint.swift	95.09% <0%> (+0.35%)	⬆️

---

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 dddb0f6...37c413e. Read the comment docs.

[ashfurrow]

Okay, got it to run but it's not behaving as I'd expect. Will mark the PR as a WIP.

[ashfurrow]

Coooooool, had to brew uninstall swiftlint first. Ready for review 👍

[marcelofabri]

What should happen if you pass force-exclude and --path ADirectory?

[ashfurrow]

That's a good question. I would expect that the other files in the directory would be linted, but any files in excluded would not be. In practice, I just tried this and the file was included. Hmm. I'll do some debugging and let you know how things go.

[marcelofabri]

@ashfurrow you probably need to filter in https://github.com/realm/SwiftLint/pull/2056/files#diff-27e9905d3d17e50e5263bb7cae0fd09aR24 too.

I'd also make sure that we have tests for what happens if we have force-exclude and:

• --path aFileThatIsInExcluded
• --path aFileThatIsNotInExcluded
• --path aDirectoryThatIsInExcluded
• --path aDirectoryThatIsNotInExcludedButHasChildrenThatAre

[ashfurrow]

Okay so it turns out I didn't quite understand the semantics of how SwiftLint resolves config files. I re-tested your earlier question:

What should happen if you pass force-exclude and --path ADirectory?

This does work as I'd expect: SwiftLint picks up the other files but not the ones in excluded. I tested this with Moya:

swiftlint lint --path .  --force-exclude --config .swiftlint.yml

And this lints everything but does exclude the files in excluded. So it works as-expected. I added the tests you specified too, let me know if there's anything else I can do to wrap this up. And thanks again!

[ashfurrow]

Ack! Just saw the failures, will fix.

[marcelofabri]

Can you add two trailing whitespaces after .?

[ashfurrow]

Sure can. Would you like a separate PR that checks for these missing spaces using Danger? Could save you some time chasing after contributors 😄

[marcelofabri]

@ashfurrow Sorry for the delay. This looks good! ✅

Just added a small comment, but other than that and fixing the conflict, this is ready to be merged.

[ashfurrow]

Any update on this? I don’t understand the nature of the CI failure.

[marcelofabri]

@ashfurrow I just manually merged this. I had to rebase to get rid of the merge commit and also I had to run Sourcery to make sure the new tests were being run on Linux (f231e6d).

Thanks and sorry for the delay!

[ashfurrow]

Cool, thanks a lot!