Add OrderTests for SortedSet #2915
Conversation
Codecov Report
@@ Coverage Diff @@
## master #2915 +/- ##
=========================================
Coverage ? 94.26%
=========================================
Files ? 364
Lines ? 6952
Branches ? 188
=========================================
Hits ? 6553
Misses ? 399
Partials ? 0Continue to review full report at Codecov.
|
| @@ -19,6 +19,8 @@ class SortedSetSuite extends CatsSuite { | |||
| checkAll("Semigroupal[SortedSet]", SerializableTests.serializable(Semigroupal[SortedSet])) | |||
|
|
|||
| checkAll("SortedSet[Int]", FoldableTests[SortedSet].foldable[Int, Int]) | |||
| checkAll("Order[SortedSet[Int]]", OrderTests[SortedSet[Int]].order) | |||
| checkAll("Order.reverse(Order[SortedSet[Int]])", OrderTests(Order.reverse(Order[SortedSet[Int]])).order) | |||
| checkAll("PartialOrder[SortedSet[Int]]", PartialOrderTests[SortedSet[Int]].partialOrder) | |||
There was a problem hiding this comment.
@kailuowang Do we still need these?
There was a problem hiding this comment.
It is redundant, but we are not very consistent when it comes testing laws already covered by the type class laws lower in the hierarchy. That's why I didn't raise an objection here. There is benefit in redundant tests, also sometimes it's hard to track which tests are already covered and remove them. Reducing redundancy to minimum (i.e. always test only the laws lowest in the hierarchy) might have its own problem as well. So right now we are doing it rather arbitrarily. We should probably come up with a principle and add it to the guideline. I created an issue for it #2919
I found we have PartialOrderTests for SortedSet, but not OrderTests.