Adding more contravariant instances

[DavidGregory084]

Some more work on #579.

[DavidGregory084]

I'm not 100% sure where instances for types from Algebra belong - happy to move them somewhere a bit less ad-hoc

[codecov-io]

Current coverage is 89.84%

Merging #929 into master will increase coverage by 0.95%

1. 6 files (not in diff) in ...main/scala/cats/free were created. more
2. 1 files (not in diff) in .../src/main/scala/cats were created. more
3. 2 files (not in diff) in ...cats/laws/discipline were modified. more
   • Hits -5
4. 5 files (not in diff) in ...main/scala/cats/laws were modified. more
   • Hits -2
5. 7 files (not in diff) in ...in/scala/cats/syntax were modified. more
   • Misses +13
   • Hits -18
6. 9 files (not in diff) in .../main/scala/cats/std were modified. more
   • Misses +15
   • Hits +35
7. 3 files (not in diff) in ...n/scala/cats/functor were modified. more
   • Misses +10
   • Hits -11
8. 8 files (not in diff) in ...main/scala/cats/data were modified. more
   • Misses -2
   • Hits -40
9. 14 files (not in diff) in .../src/main/scala/cats were modified. more
   • Misses +27
   • Hits -19
10. 2 files in ...cats/laws/discipline were modified. more
    • Hits +3

@@             master    #929   diff @@
=======================================
  Files           226     180     -46   
  Lines          2916    2177    -739   
  Methods        2865       0   -2865   
  Branches         48      42      -6   
=======================================
- Hits           2592    1956    -636   
+ Misses          324     221    -103   
  Partials          0       0           

Powered by Codecov. Last updated by 00839fb...ad4ab8b

[DavidGregory084]

I was able (I think) to write an implementation for StateT as:

def contramap[B](f: B => A)(implicit F: Contravariant[F], M: Monad[F]) =
  StateT(s1 =>
    F.contramap(run(s1)) {
      case (s2,  a) => (s2,  f(a)) 
    }
  ) 

However, this would typically mean resolving implicits for e.g Contravariant[Cokleisli[Option, ?, Int]] and Monad[Cokleisli[Option, Int, ?]] at the same time - I'm not really sure how to arrange that.

[non]

Hi @DavidGregory084, sorry for the long wait!

These instances look great! Would you be willing to update this PR?

As far as instances for the kernel (e.g. contravariant functors for Eq, PartialOrder, and Order) I would actually recommend putting those in the Contravariant companion object (in core). I would also recommend adding a note that the resulting instances are only law-abiding if the functions provided are one-to-one (injections).

[DavidGregory084]

No problem, this should be up to date now 😸

[ceedubs]

Sorry - we've been merging some sweeping PRs recently that have caused a lot of merge conflicts :(. Would you mind updating again?

[DavidGregory084]

@ceedubs Yeah, that's fine, should be up to date again :)

[ceedubs]

👍 thanks a bunch, @DavidGregory084! @non would you be willing to do a review before this goes stale again?

[non]

Looks great, thanks! 👍