kotlinlang #arrow

Join the conversation Join Slack

Channels

100daysofcode

100daysofkotlin

androidgithubprojects

cincinnati-user-group

competitive-programming

event21-community-content

java-to-kotlin-refactoring

jvm-ir-backend-feedback

kotlin-multiplatform-contest

kotlinultimatechallenge

myndocs-oauth2-server

naming

navigation-architecture-component

practical-functional-programming

proguard

prozis-android-backup

refactoring-to-kotlin

russian-kotlinasfirst

skia-wasm-interop-temp

stayhungrystayfoolish

swagger-gradle-codegen

tricity-kotlin-user-group

arrow

prat

09/02/2017, 7:45 PM

i'm writing a simple test

testLaws(SemigroupKLaws.laws(SequenceKW.semigroupK(), applicative, Eq.any()))

and it failed

java.lang.AssertionError: Property failed for
SequenceKW(sequence=kotlin.collections.ArraysKt___ArraysKt$asSequence$$inlined$Sequence$1@5d5abcb9)
SequenceKW(sequence=kotlin.collections.ArraysKt___ArraysKt$asSequence$$inlined$Sequence$1@2973d20a)
SequenceKW(sequence=kotlin.collections.ArraysKt___ArraysKt$asSequence$$inlined$Sequence$1@1ec9a588))
	at io.kotlintest.properties.PropertyTestingKt.forAll(PropertyTesting.kt:43)
	at kategory.data.SequenceKWTest$$special$$inlined$laws$1.invoke(SemigroupKlaws.kt:17)
	at kategory.data.SequenceKWTest$$special$$inlined$laws$1.invoke(SemigroupKlaws.kt:6)
	at io.kotlintest.Spec$runTest$callable$1$1.invoke(Spec.kt:124)
	at io.kotlintest.Spec$runTest$callable$1$1.invoke(Spec.kt:15)
	at io.kotlintest.Spec$runTest$initialInterceptor$1$1.invoke(Spec.kt:116)
	at io.kotlintest.Spec$runTest$initialInterceptor$1$1.invoke(Spec.kt:15)
	at io.kotlintest.Spec.interceptTestCase(Spec.kt:78)
	at io.kotlintest.Spec$runTest$initialInterceptor$1.invoke(Spec.kt:116)
	at io.kotlintest.Spec$runTest$initialInterceptor$1.invoke(Spec.kt:15)
	at io.kotlintest.Spec$runTest$callable$1.call(Spec.kt:124)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
	at java.lang.Thread.run(Thread.java:745)

any thought?

raulraja

09/05/2017, 4:35 PM

Now if you have a project that depends on kategory and use any of the functions in that file the produced output links in the bytecode directly to that. And if then you exclude Kategory from the list of transitive dependencies manually the compile code can't resolve user provided artifacts because those files are also excluded so it fails with a NoClassDefFound exception

raulraja

09/14/2017, 10:30 PM

there is a

isSubtype

for

TypeMirror

comparisons and it does not work so I'm recursing all interfaces as I did before and comparing Strings

raulraja

09/14/2017, 10:31 PM

metaprogramming in Kotlin is very painful, I'm impressed how far you've got, I'd have given up right away xD

raulraja

09/17/2017, 8:35 PM

@Eugenio If an annotation processor performs codegen that should be expanded with other annotation processors, does it need to do anything special beside adding the annotations to codegen?

raulraja

10/04/2017, 7:55 AM

@ersin_ertan expanding on what @simon.vergauwen said that is a polymorphic function that can compute over any higher kind for which a

MonadError<F, NukeException>

exist. In the case of

Option

it can't provide such an instance because there is no way to store the

NukeException

, since

None

which is the unbiased case can't store that error. Same for

Try

since it's

Failure

it's fixed to

Throwable

and

NukeException

is not in the

Throwable

hierarchy. The only choice here is

Either

which is parametric in it's

Left

case.

attack

may be invoked in two different ways:https://gist.github.com/raulraja/40ac0d0f03f4a1a85f6d249464f30a95

raulraja

10/05/2017, 8:53 PM

@ersin_ertan what does

.p()

do?

simon.vergauwen

10/10/2017, 11:56 AM

I am not sure if I understand correctly what you mean by syntax for functional types

raulraja

10/30/2017, 2:27 PM

object test {

    sealed class BizError  {
        object Error1 : BizError()
        object Error2 : BizError()
        object Error3: BizError()
    }

    val ops: List<ValidatedNel<BizError, Int>> = listOf(
            1.validNel(),
            1.validNel(),
            1.validNel(),
            BizError.Error1.invalidNel(),
            BizError.Error2.invalidNel(),
            BizError.Error3.invalidNel()
    )

    val result: ValidatedNel<BizError, Int> = ops.reduce { a, b ->
        a.combine(b, NonEmptyList.semigroup())
    }
}

raulraja

10/31/2017, 12:09 PM

@simon.vergauwen I'm currently swamped with work, if you wanna do the honors go for it or I'll ask someone else if they can. I think @pablisco fixed it so that it just needs a tag

Jorge Castillo

11/03/2017, 2:51 PM

Oh, don’t forget to attend @pakoito’s talk today, he is gonna talk about how kategory works behind the scenes. Both talks complement each other in a really good way

raulraja

11/05/2017, 9:36 PM

https://github.com/kategory/kategory/pull/441

snrostov

11/06/2017, 10:19 AM

Yet another use case for type classes in the UI:

// in existed module
data class User(val name: String)

// in other module
object UserRenderer: Renderer<User> {
    fun User.render() = Row {
        +Column { +"Name: " }
        +Column { +name }        
    }
}

fun someOtherComponent(user: User)
  = Row { +user }

👏 1

raulraja

11/07/2017, 2:57 PM

For example look at this way to do DI type checked by the compiler and eliminating the need for DI frameworks:

interface Context<A> {
  fun A.config(): Config
}

package prod

object ProdContext: Context<Service> {
  fun Service.config(): Config = ProdConfig
}

package test

object TestContext: Context<Service> {
  fun Service.config(): Config = TestConfig
}

package prod

service.config() // ProdConfig

package test

service.config() // TestConfig

pakoito

12/02/2017, 11:39 PM

@wiyarmir can answer that best, I’m a faithful customer of jitpack 😄

raulraja

12/02/2017, 11:39 PM

@Czar it may not be needed after some recent changes if you can resolve all modules with jcenter. Meaning that you just found a piece of outdated documentation 🙂

pakoito

12/20/2017, 7:02 PM

I personally oversee that it doesn’t happen, so it isn’t expected 😄

kevinherron

12/20/2017, 10:26 PM

@pakoito @raulraja re: earlier, removing the

crossinline

modifier from the parameters to

map

and

flatMap

allow me to call suspending functions. if you’d like, I’ll PR those changes as well. I suspect there’s more than just the

Try

data structure being affected by this.

👍 1

dh44t

12/22/2017, 9:49 AM

https://www.packtpub.com/application-development/functional-kotlin

👍 5

😱 3

🎉 8

alex.hart

01/04/2018, 12:20 PM

Also it’s super cool that Jesse Wilson is a Co-Organizer

pakoito

01/15/2018, 5:33 PM

that's just a thin wrapper over PCollections. We'd like our own built up from scratch. @glenkpeterson made Paguro for Java and was taking a look on his spare time 😄

🔥 2

dh44t

01/18/2018, 2:59 PM

I'm writing one section of my book about State, I had a few examples of it. The code examples are Apache licensed, as soon as I have it public I'll post it here

👍 3

yousuf.haque

01/18/2018, 3:30 PM

question, has anyone already written/found a retrofit converter factory for nullables into

Option

raulraja

01/19/2018, 12:04 AM

Scala has already changed to a right biased either since 2.11

✅ 3

rawtoast

01/24/2018, 1:48 PM

@cedric Interesting, from experience with Scala I would advise you don’t do that for Option. In Scala there is a difference between

Option(null)

(returns

None

) and

Some(null)

(returns an Option holding

null

). It’s understandable but can be confusing and lead to bugs as not everyone knows this edge case. So for safety you end up creating/assigning with

Option

and only using

Some

for pattern matching/test assertions. Not really tried that in Kotlin, as I tend to use Option.fromNullable which seems to work in a sane manner 🙂

aeruhxi

01/24/2018, 6:31 PM

semantically,

Some(null)

doesn't make sense at all 🤔

raulraja

02/04/2018, 1:56 AM

https://github.com/arrow-kt/arrow/pull/649

raulraja

02/08/2018, 12:47 PM

https://twitter.com/arrow_kt/status/961562757152862208

👏 6

👍 7

🎉 11

thanerian

02/08/2018, 8:31 PM

https://github.com/arrow-kt/arrow/blob/master/modules/docs/arrow-docs/docs/docs/quickstart/blogs/README.md you only need to modify It and send a PR

simon.vergauwen

02/12/2018, 10:05 AM

Isn't it really confusing we are releasing docs on every merge?

Title

simon.vergauwen

02/12/2018, 10:05 AM

Isn't it really confusing we are releasing docs on every merge?

pguardiola

02/12/2018, 10:35 AM

yeah i was wondering the same 🤔

btw after releasing

0.7.0

should be more stable

raulraja

02/12/2018, 10:42 AM

yes that is confusing and a larger issue is that docs need to be versioned and stored. We don't have a strategy yet to do that but generating the site in the current version number and serving all versions may be an option.

User's may want to consult the docs for 0.5.1 and those are no longer available since we are making destructive changes in the doc site.

👍 1

View count: 7