Codecov Report

Merging #1088 (0fb7166) into master (636e505) will decrease coverage by 0.04%.
The diff coverage is 100.00%.

❗ Current head 0fb7166 differs from pull request most recent head b0eac29. Consider uploading reports for the commit b0eac29 to get more accurate results

@@            Coverage Diff             @@
##           master    #1088      +/-   ##
==========================================
- Coverage   92.64%   92.61%   -0.04%     
==========================================
  Files          63       63              
  Lines         829      839      +10     
  Branches       26        8      -18     
==========================================
+ Hits          768      777       +9     
- Misses         61       62       +1     

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

Hi @fthomas , can you merge this MR please?

@FunFunFine Could you add tests for these instances to https://github.com/fthomas/refined/blob/master/modules/cats/shared/src/test/scala/eu/timepit/refined/cats/CommutativeSemigroupAndMonoidLawTests.scala ?

@fthomas it's been a while, but here it goes.

Thanks @FunFunFine!

Can we also add commutativeSemigroup / commutativeMonoid tests like this:

checkAll("CommutativeSemigroup[PosBigDecimal]", CommutativeSemigroupTests[PosBigDecimal].commutativeSemigroup)

Or do they also fail because those instances are only approximately associative, too (like PosDouble for example)?

The compilation fails with the following error when I try to add this test:

diverging implicit expansion for type org.scalacheck.Arbitrary[(scala.math.BigDecimal, eu.timepit.refined.numeric.Greater[shapeless._0])]
starting with method arbTuple2 in trait ArbitraryArities
  checkAll("CommutativeSemigroup[PosBigDecimal]", CommutativeSemigroupTests[PosBigDecimal].commutativeSemigroup)

I don't really understand what the problem is because it works for other types, but not for BigDecimal apparently.

Oh, that fails because we lack Arbitrary instances for refined BigDecimal types. I'm fine with merging this without these tests then.

Someone who has access to merging needs to merge it