Add PowerShell formatter linter

[bdovaz]

@nvuillam I put it in draft because I have the doubt of how to create a test that fails if this command never fails... I pass it an input that is not even a valid powershell code and it works, that is, it does nothing and returns the same input as output....

https://learn.microsoft.com/en-us/powershell/module/psscriptanalyzer/invoke-formatter?view=ps-modules

Formatter code that does not handle any errors:

https://github.com/PowerShell/PSScriptAnalyzer/blob/1.21.0/Engine/Formatter.cs

Unlike ScriptAnalyzer.cs which does handle errors (used by original powershell linter):

https://github.com/PowerShell/PSScriptAnalyzer/blob/1.21.0/Engine/ScriptAnalyzer.cs

[codecov-commenter]

Codecov Report

Merging #2176 (84db0ea) into main (c9e2d37) will increase coverage by 0.45%.
The diff coverage is 50.00%.

@@            Coverage Diff             @@
##             main    #2176      +/-   ##
==========================================
+ Coverage   82.36%   82.81%   +0.45%     
==========================================
  Files         167      167              
  Lines        4439     4445       +6     
==========================================
+ Hits         3656     3681      +25     
+ Misses        783      764      -19     

Impacted Files	Coverage Δ
megalinter/linters/PowershellLinter.py	81.48% <50.00%> (-13.76%)	⬇️
megalinter/reporters/UpdatedSourcesReporter.py	89.74% <0.00%> (+2.56%)	⬆️
...alinter/tests/test_megalinter/helpers/utilstest.py	89.01% <0.00%> (+8.05%)	⬆️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[nvuillam]

Don't forget to build with bash build.sh, it will generate the test class :)

[bdovaz]

@nvuillam done.

Can you read and answer me to the first message of this PR related to the tests?

I see that there are special cases where you exclude this type of test (linter_failure):

https://github.com/oxsecurity/megalinter/blob/v6.16.0/megalinter/tests/test_megalinter/helpers/utilstest.py#L214

If we don't find a solution I don't know if this would be the solution.

[nvuillam]

@bdovaz do powershell formatter have a --check mode, to just check if the file is already formatted or not ?

[bdovaz]

@nvuillam as you can see, no:

https://learn.microsoft.com/en-us/powershell/module/psscriptanalyzer/invoke-formatter?view=ps-modules

[nvuillam]

@bdovaz it seems that Microsoft does not like to build their apps like all others ^^

In that case you can add an empty file named no_test_failure (without extension) in the test folder

[nvuillam]

plz check comment :)

[nvuillam]

❌ Linted [REPOSITORY] files with [git_diff]: Found 1 error(s) - (0.03s)
- Using [git_diff v2.36.3] https://megalinter.io/2644adfa201ad6bd9bec5f21e8040cd699416064/descriptors/repository_git_diff
- MegaLinter key: [REPOSITORY_GIT_DIFF]
- Rules config: identified by [git_diff]
--Error detail:
.automation/test/powershell_formatter/good/powershell_good_1.ps1:2: new blank line at EOF.
.automation/test/powershell_formatter/good/powershell_good_1.psd1:4: new blank line at EOF.
.automation/test/powershell_formatter/good/powershell_good_1.psm1:2: new blank line at EOF.

That's really strange, I never saw this one :D
Maybe some error while mananing conflicts ?

[bdovaz]

❌ Linted [REPOSITORY] files with [git_diff]: Found 1 error(s) - (0.03s)
- Using [git_diff v2.36.3] https://megalinter.io/2644adfa201ad6bd9bec5f21e8040cd699416064/descriptors/repository_git_diff
- MegaLinter key: [REPOSITORY_GIT_DIFF]
- Rules config: identified by [git_diff]
--Error detail:
.automation/test/powershell_formatter/good/powershell_good_1.ps1:2: new blank line at EOF.
.automation/test/powershell_formatter/good/powershell_good_1.psd1:4: new blank line at EOF.
.automation/test/powershell_formatter/good/powershell_good_1.psm1:2: new blank line at EOF.

That's really strange, I never saw this one :D Maybe some error while mananing conflicts ?

@nvuillam Well I don't know what to do to fix this because if you look at my PR I simply copied the 3 files from the powershell folder as is.

[nvuillam]

@bdovaz what if you add a blank line in the files so they errors can disappear ?

[bdovaz]

@nvuillam I just did it but that is the solution? I just don't understand the error either, what it refers to and how to fix it.

Let's wait for the result of the build.

[nvuillam]

That's really really strange... as the rest seems legit, I force the PR, we'll see if it solves the issue ^^

[echoix]

This reminds me of when git changes the end of lines when checking-out, and changing them back when committing. It is normally controlled by a .gitattributes file (like https://github.com/oxsecurity/megalinter/blob/main/.gitattributes ). Are you working on Windows?

[bdovaz]

This reminds me of when git changes the end of lines when checking-out, and changing them back when committing. It is normally controlled by a .gitattributes file (like https://github.com/oxsecurity/megalinter/blob/main/.gitattributes ). Are you working on Windows?

@nvuillam @echoix yes, I'm on Windows.

[nvuillam]

I'm on windows too but never had the issue ^^

If it does not happen on this PR #2189 , let's consider the problem solved ^^

[nvuillam]

I learned something today :)

https://stackoverflow.com/questions/27059239/git-new-blank-line-at-eof

[echoix]

I'm on windows too but never had the issue ^^

Some of the .gitattributes won't get applied to an already cloned repo without any actions, so it doesn't exclude the problem.

https://docs.github.com/en/get-started/getting-started-with-git/configuring-git-to-handle-line-endings

https://git-scm.com/docs/gitattributes

[nvuillam]

Finally solved... i just remove the line at the end of the files >_<