Add MaybeK wrapper for RxJava Maybe

[JorgeCastilloPrz]

Fixes #860

• Added MaybeK
• Added instances
• Added some tests including law tests for the instances.

@pakoito I've added fold(), foldLeft, foldRight etc to it since I saw those where also defined in other "equivalent" data types like Option and maybe also represents code branching. Also a Foldable instance. But to be honest I'm not 100% sure the code on those methods / instance is correct, but FoldableLaws are passing, which seems suspicious. Could you take a look?

I've not added Traverse instance for it yet since traverse uses to rely on fold and I'm still not confident with my implementation.

[pakoito]

Dooocs

[JorgeCastilloPrz]

Yeees I'll add docs this weekend, but need your help on those folds :D

[pakoito]

IIRC Maybe has the same semantics as Observable, except it's max 1 return.

https://github.com/arrow-kt/arrow/blob/master/modules/effects/arrow-effects-rx2/src/main/kotlin/arrow/effects/ObservableK.kt#L33-L42

[JorgeCastilloPrz]

Given that Single does not provide reduce operation (bc max 1 return), the equivalent to that foldLeft()implementation from ObservableK, but in terms of Single, would be the one I have now (I've redefined it in terms of fold()).

On foldRight() you usually need laziness (for being able to evaluate infinite streams), but here I'm gonna have a single value at most, so I'm a bit concerned about whether that ObservableK impl based on Eval.defer() applies here. Still, I've kept is as deferred eval even for 1 element and redefined when in terms of fold(). It should be equivalent to what you have there.

[JorgeCastilloPrz]

There you got the docs also, back to you @pakoito

[codecov]

Codecov Report

Merging #862 into master will increase coverage by 0.29%.
The diff coverage is 85.18%.

@@             Coverage Diff              @@
##             master     #862      +/-   ##
============================================
+ Coverage     44.85%   45.15%   +0.29%     
- Complexity      628      640      +12     
============================================
  Files           300      302       +2     
  Lines          7709     7763      +54     
  Branches        834      837       +3     
============================================
+ Hits           3458     3505      +47     
- Misses         3947     3954       +7     
  Partials        304      304

Impacted Files	Coverage Δ	Complexity Δ
...2/src/main/kotlin/arrow/effects/MaybeKInstances.kt	81.81% <81.81%> (ø)	0 <0> (?)
...ffects-rx2/src/main/kotlin/arrow/effects/MaybeK.kt	87.5% <87.5%> (ø)	12 <12> (?)
...ics/src/main/kotlin/arrow/optics/instances/mapk.kt	91.66% <0%> (+4.16%)	0% <0%> (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 5ed0c8d...5ebf7ea. Read the comment docs.

[pakoito]

According to RxJava's docs, this.maybe.blockingGet() returns null for no value and A for a value. Given that RxJava2 doesn't allow nulls in Maybe, you can rewrite this to be simpler:

maybe.blockingGet().let {
  if (it == null) ifEmpty() else ifSome(if)
}

[pakoito]

I would suggest adding a new test for it too.

[JorgeCastilloPrz]

I'll address both comments tomorrow. Out of home today, Champions League finals tonight 💥

If you're hurry for a release, feel free to push those changes. Otherwise I'll do.

[JorgeCastilloPrz]

Done and done @pakoito 👍

[pakoito]

Let's mergeee!

[JorgeCastilloPrz]

👍