I’m reading https://adventures92.medium.com/kotlin-arrow-kt-eval-monad-a-beginners-guide-bff50e49d07c and I’m struggling to understand what’s the point of

Eval

. •

Eval.now

- the author says that “This is the equivalent of just calling a function normally in Kotlin.” •

Eval.later

- how is it different from

lazy

delegated property? •

Eval.always

- how is it different from just calling a function each time? I presume that the examples in the article didn’t fully utilize the power of this monad, like chain it in a creative way. Is it the case? If not, I’d be grateful for any explanation on the above ☝️

https://github.com/arrow-kt/arrow/pull/2921 😅

ok, that’s a bit of a bummer from the article’s perspective 😄 could you reveal why you decided to remove them?

I’m struggling to understand what’s the point of

Eval

.

On the JVM we have the issue of stack-safety, and when working over

F

, or wrappers, you cannot use

tailrec

. So we there is a need to resort to special tricks to make wrappers stack-safe. Originally the point of

Eval

was to create stack-safe implementations of

Traverse

, but we have since then overcome this need and found simpler solutions. I.e. we can use

tailrec

with

suspend

, and you can use

bind

with

Iterable#map

etc. So the need for

Eval

has kind-of disappeared within the use-case of

wrappers

. The Kotlin Std has since then also added

DeepRecursiveFunction

which when we originally had

Eval

did not exist.

I guess this article need an update and a bit more explanation. Thanks you @Piotr Krzemiński for picking it out.

@Anand Verma if you want someone to proof read feel free to share it here or in #arrow-contributors ☺️ There is plenty of people happy to help!

Thanks @simon.vergauwen thanks a ton for your valuable feedback.