Hello all! I'm not sure if this question has been asked before, but I am really interested in your opinion on how logging should be performed in a purely functional way. Thank you!

it's not in Kotlin, but I found haskellbook.com very useful

I’m quite sure the inference will be happy with that

Hello guys, I'm trying to build intuition on finding fp patterns in real life applications, so can you correct me if I worng? Can I look at JSON as a functor? It seems to me that fmap can be written something like that:

deserialize().let(::f).serialize()

. But is it applicative? Or I am following the wrong direction and there are more suitable abstractions to reason about JSON? Also Parcelable from Android and JSON both are looking pretty similar to me. Are there any chances to prove this similarity in fp terms?

For those of you attending Kotlin Conf and interested in FP and Arrow https://twitter.com/raulraja/status/1046700333630722050

@simon.vergauwen I think that could be possible

when you say a separate idea instance, I should copy and paste the idea folder and choose a different directory for the config?

I'm running

assemble

in hope I'll end up with correct jar. Not sure how long that'll take. runIde took 17 hours on my machine.

@rameez this is a tutorial that people frequently use to get started and exposes you to exception free code with

Option

,

Try

,

Either

,

ApplicativeError

and

MonadError

https://arrow-kt.io/docs/patterns/error_handling/

now, take a look at this article and see what’s the next step: https://arrow-kt.io/docs/patterns/glossary/#type-constructors

However KotlinTest also has some Arrow specific matchers https://github.com/kotlintest/kotlintest/blob/master/doc/reference.md#arrow

Okay, I watched the talk! I'm glad you started with Optics, as that's the part that interests me most right now. And "Put

@optics

on your data classes and then you get methods you can use to copy and modify" is a much more approachable message to me than some introductions to Arrow I've seen.

outsider to FP here : most of the issues I have with FP documentation is that it uses words I don't know to explain other words I don't know, leaving me with 75 open tabs and still no idea what I'm reading (seriously, imagine reading this https://wiki.haskell.org/Monad without an advanced mathematical/FP understanding, it still gives me a headache). Just having a "it's like other common programming thing but with *differences*" section would really help newcomers.

It just explains interface but not the purpose

This message was deleted.

my two cents on the above discussion about naming conventions, also because I'm currently helping to percolate FP concepts to people coming from more imperative background (eg, some android developers more familiar with traditional java style/OOP). Although i can partially agree that names like Monad, Applicative etc might sound scary for those who come from a different background, i also think that choosing unconventional names like chanable or container is just a palliative to increase the appeal but at the same time it is extremely misleading. IMO we should focus (as programmers in general) to focus more on the concepts behind and try to understand if/when/how/WHY such abstractions and their properties can be helpful when writing code. Although chainability and containers refer to some behaviour we can qualitatively observe when using these concepts, I don't think this captures the essence of these concepts and the power of these abstractions. Although composability is at the heart of lot of concepts borrowed from category theory, terms like Containers or dotchains are quite generic, not precisely defined, they refer more to the way we expect to use things rather than their fundating properties, they shift our focus on a particular aspect of those things instead of driving us to an appreciation of how to reason in terms of these abstractions. It's actually a nice exercise to compare how either the monadic or applicative properties actually constrains us , what they allow us to do and achieve and what not.

@Mike why not use Left(....) and Right(...) instead?

so maybe we just need to write one?

Has anyone watched this series? https://www.youtube.com/playlist?list=PLbgaMIhjbmEnaH_LTkxLI7FMa2HsnawM_

Hi all! I'm a newcomer of the functional programming in general; after a 3-days applied FP course (in scala) I've learned the basics of functional programming; after the course I switched to Kotlin+Arrow since I already knew the language quite well. Currently I'm facing some problems when approaching to a further steps in my understanding; I'm trying to better understand the monad transformers topic, and I've found online this little example: https://underscore.io/blog/posts/2014/07/27/readerwriterstate.html I'm struggling to write it in arrow since I can't understand how to use the ReaderT/WriterT transformers in order to achieve the same result described in the post. Do you have any similar example but done with Kotlin+Arrow?

First when: Just do a check for null with a elvis operator or, if use are using Option, use a fold

Also, don't use

when

with

Either

, there's the

.fold

to solve this problem

so my guess is a behavior regression

Hi, I'm moving here the thread started before as it's changing a bit the topic and becoming a bit more generic. As a Kotlin developer, that has been introduced in the functional world recently (but with a Scala-based course), what do you think is a valid learning path? As the Arrow documentation is a bit incomplete at the moment, and it is very difficult to find also examples on the web on specific topics, I wonder if Kotlin+Arrow is a good path in order to learn the principles of FP... could it be better to start with some other language (es. Scala), learn the FP concepts (as there is a ton of material on the web) and then move back to Kotlin+Arrow? I'd like to stay on the Kotlin path but I'm struggling in finding out how to use Arrow to make the same things I found on the web with Scala (as in the previous ReaderT/WriterT question). What do you think about it?

Yeah, i’ll have a look today and tomorrow then) The problem is i off for one week starting Thursday, so if i won’t have time to finish, i’ll let you know here tomorrow evening)