[Automated PR] Bump KTLint and/or Google Java Formatter

[macisamuele]

Bump KTLint and/or Google Java Formatter

Auto-generated by bump-releases

[codecov]

Codecov Report

Merging #67 (1aaa0a4) into master (dff1c5e) will not change coverage.
The diff coverage is 100.00%.

@@            Coverage Diff            @@
##            master       #67   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           10        10           
  Lines          318       321    +3     
=========================================
+ Hits           318       321    +3     

Impacted Files	Coverage Δ
...ormatters_pre_commit_hooks/pretty_format_kotlin.py	100.00% <100.00%> (ø)
..._formatters_pre_commit_hooks/pretty_format_yaml.py	100.00% <100.00%> (ø)

---

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 dff1c5e...1aaa0a4. Read the comment docs.

[cortinico]

For some reason tests are failing on windows. Do you happen to know why @macisamuele ?

[macisamuele]

I'll be looking into the windows failures.
The bright side is that tests did got executed for the automated PR (so at least #62 worked)

[macisamuele]

@cortinico from a very superficial (possibly still incorrect) analysis I see that windows tests are failing because:

• ktlint requires relative paths, but it does look for paths with the forward-slash (/) and in windows the path separator is \ (example in here).
  I have to check how in real pre-commit executions the path would be passed in to understand if this is testing issue or this would break during real execution as well
• even assuming that we can replace \ with / (here), tests would continue to fail because seems that ktlint does not write the fixed file even if --autofix is passed in.

Considering the above findings, I'm going to look more into this hoping to identify what is the root cause of this and hoping that it is testing setup rather than a big regression in ktlint.

Edit: After a little debugging I've found out what is happening in here

1. ktlint does not accept \ (on Windows) in the input paramters
2. ktlint uses \ (on Windows) paths
3. the tool uses the ktlint output to decide on which files to run the formatter (using then the \ as provided by ktling, which does not work :( )

[macisamuele]

Manually rebased PR on top of #68 .
Please let's merge the 2 PRs separately.