> // TODO which we need to have access to the i...

# arrow

[pakoito]

// TODO which we need to have access to the implementation of map in Option

yes! it's also a fun exercise not to call fix() and see a stackoverflow exception because the function calls itself

[Jannis]

I've learned to really appreaciate the recursion icon intellij shows in these cases as sometimes even with .fix() this can happen (if the ext. function is also on to Kind<*, *>) ^^ Probably saved me many minutes of pointless staring at code

[Bruno]

another question:

fun <F> parseInt(value: Kind<F, Int>, functor: Functor<F>) =
    functor.run {
        value.map { it.toInt() }
    }

I have trouble with this run method...

If we had a way of saying Kind<F extends Functor, Int>
would it be possible to just say:

fun <F> parseInt(value: Kind<F extends Functor, Int>) = value.map { it.toInt() }

[pakoito]

that’s the thing, the fact that you have to pass a Functor<F> proves that one exists

[pakoito]

now, typeclasses, or extension interfaces, or traits, or whatever the name your lang/framework uses

[pakoito]

is the mechanism to inject that parameter you want

[pakoito]

F: Functor[_] is in fact one way Scala does it

[pakoito]

the current state of the art is

[pakoito]

fun <F> Functor<F>.parseInt(value: Kind<F extends Functor, Int>) =

[pakoito]

and the next generation is what the compiler plugin will do

[pakoito]

what we call sometimes KEEP87

[Bruno]

awesome thanks for the response!