cats-effect: Unable to format file due to bug in scalafmt

[rossabaker]

This template is a guideline, not a strict requirement.

• Version: 2.4.1
• Integration: sbt-scalafmt
• Configuration:

version = 2.4.1
align = none
align.openParenCallSite = true
align.openParenDefnSite = true
maxColumn = 120
continuationIndent.defnSite = 2
assumeStandardLibraryStripMargin = true
danglingParentheses = true
rewrite.rules = [AvoidInfix, SortImports, RedundantBraces, RedundantParens, SortModifiers]
docstrings = JavaDoc
lineEndings=preserve

Steps

Given code like this:

  testAsync("Resource.make is equivalent to a partially applied bracket") { implicit ec =>
    check { (acquire: IO[String], release: String => IO[Unit], f: String => IO[String]) =>
      acquire.bracket(f)(release) <-> Resource.make(acquire)(release).use(f)
    }
  }

When I run scalafmt like this:

sbt lawsJVM/scalafmtAll

Problem

Scalafmt fails:

[error] /home/ross/src/cats-effect/laws/shared/src/test/scala/cats/effect/ResourceTests.scala:56: error: Unable to format file due to bug in scalafmt
[error]       acquire.bracket(f)(release) <-> Resource.make(acquire)(release).use(f)
[error]                                 ^: /home/ross/src/cats-effect/laws/shared/src/test/scala/cats/effect/ResourceTests.scala
[error] (lawsJVM / Test / scalafmt) /home/ross/src/cats-effect/laws/shared/src/test/scala/cats/effect/ResourceTests.scala:56: error: Unable to format file due to bug in scalafmt
[error]       acquire.bracket(f)(release) <-> Resource.make(acquire)(release).use(f)
[error]                                 ^: /home/ross/src/cats-effect/laws/shared/src/test/scala/cats/effect/ResourceTests.scala

Expectation

scalafmt succeeds and the output still compiles

Workaround

Not upgrading to 2.4.1.

Notes

Similar to #1599, but not the same as what @larsrh reported for cats-effect on 2.4.0.

[kitbellew]

Similar to #1599, but not the same as what @larsrh reported for cats-effect on 2.4.0.

what was the problem in 2.4.0? i don't think i have seen that report.

[larsrh]

@kitbellew See #1599 (comment)

[kitbellew]

@kitbellew See #1599 (comment)

@larsrh is there a code example which can be used to reproduce the problem?

[larsrh]

Yes, you can run `scalafmtAll` on the PR.

…

On 17 February 2020 09:35:40 CET, Albert Meltzer ***@***.***> wrote: > @kitbellew See [#1599 (comment)](#1599 (comment)) @larsrh is there a code example which can be used to reproduce the problem? -- You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: #1714 (comment)

[kitbellew]

Yes, you can run scalafmtAll on the PR.

of course.

[kitbellew]

@kitbellew See #1599 (comment)

@larsrh is there a code example which can be used to reproduce the problem?

@larsrh please see if your problem is fixed by #1719.

also, it would make it easier in the future if you could share your problems with the tool by submitting a new issue with the configuration and exact steps to reproduce.

[rossabaker]

I can test. Are snapshots published anywhere? I checked Sonatype, and don't see any past 2.3.1. Thanks!

[kitbellew]

I can test. Are snapshots published anywhere? I checked Sonatype, and don't see any past 2.3.1.

honestly, i don't know. i build locally and run this:

v=2.4.1
# build
sbt publishLocal
# turn into a binary
coursier bootstrap org.scalameta:scalafmt-cli_2.13:$v-SNAPSHOT \
  -f -r sonatype:snapshots \
  -o scalafmt.new --standalone --main org.scalafmt.cli.Cli
# run
./scalafmt.new --test