Investigate generation of functions which have outputs that depend on their inputs

[LightAndLight]

There were a couple of questions in the reddit thread about this kind of idea.

I think it's cool but I don't know if it's possible. The main example of monotonicity requires knowledge of the entire domain and codomain, which isn't captured by a -> Gen b anyway. At the very least we need to be aware of the history of inputs to the function.

The memo-trie approach works by lazily enumerating all the possible inputs to the function, so if we wanted to generate a monotonic function then somehow each input would have to have a knot that knows about all the other inputs. I don't know that it works.

[tscholak]

lol, I just ran into this problem. any idea how this may be accomplished?

[LightAndLight]

The monotonic example, or something else? I haven't thought about it for a while. What are you trying to do?

[tscholak]

I wanted to generate a function isomorphic to a -> Either a b.

[tscholak]

In the meantime, I circumvented the issue by generating Either () (Fn a b) and mapping this to a -> Either a b down the line. In the Left case, it's just the identity.

[LightAndLight]

Do you mean you generated an a -> Either a b for some specific a and b (say, Int and Bool)? Or you want to generate a parametric version (forall a b. a -> Either a b - there's only one function of this type- Left)?

[tscholak]

no, no, you are right, it would be for some unknown, but specific a and b.

implementing the following is not possible with what Hedgehog.Function provides today:

mkGen
  :: forall a b
   . (Hedgehog.Function.Arg a, Hedgehog.Function.Vary a)
  => Hedgehog.Gen b
  -> Hedgehog.Gen (Hedgehog.Function.Fn a (Either a b))