is there a way to introduce a delay that can be skipped in a test? I tried using

runBlocking {
    delay(10000)
}

and then testing it with

runBlockingTest

, but since it’s in another

runBlocking

block the delay is not being skipped and my test is, in fact, hanging for 10 secs. I’d like to test that the delay is indeed executed but without really having to wait for it to expire

Is there no literals for

Short

? In other words, we aren't able to make compile time constants for Shorts?

Hey guys! I'm facing some issues with the no-arg plugin + lazy initialization:

@No-Arg
data class MyClass(val foo: String) {
    val bar by lazy { foo[0] }
}

// Create Foo through reflection, for example, getting it from a database. It will use the zero-arg constructor
val myFoo = getFoo()

myFoo.bar -> NullPointerException

When inspecting the bytecode, the bar$delegate (or something like this) is not initialized in the no-arg constructor, and thus is null in runtime. Do anybody have an idea on how to solve this?

Not sure if there's a better channel for this. I had a program throw an exception into main. I actually figured out what the issue was, but when I run my program in Intellij, I can't find even a half decent stack trace or the name of the exception, it's literally just a mountain of gradle garbage. How do I find my actual stack trace? The exception ends with (124 more) but I can't seem to expand it any further

I've just stumbled into a binary incompatibility issue.

java.lang.NoSuchMethodError: No direct method <init>(Lkotlin/coroutines/CoroutineContext;I)V in class Lkotlinx/coroutines/flow/internal/ChannelFlow; or its super classes (declaration of 'kotlinx.coroutines.flow.internal.ChannelFlow' appears in /data/app/~~ccBt4dlqHFv3AkedliSnSw==/redacted.package.name.debug-QqdNYvkOKVfFKcKUHZwhVg==/base.apk!classes18.dex)
        at kotlinx.coroutines.reactive.PublisherAsFlow.<init>(ReactiveFlow.kt:51)
        at kotlinx.coroutines.reactive.PublisherAsFlow.<init>(ReactiveFlow.kt:50)
        at kotlinx.coroutines.reactive.ReactiveFlowKt.asFlow(ReactiveFlow.kt:30)

I had a library compiled with

org.jetbrains.kotlinx:kotlinx-coroutines-core:1.3.9

My project used a more recent version

org.jetbrains.kotlinx:kotlinx-coroutines-core:1.4.2

this apparently caused a binary incompatibility.... It is troubling to me because

asFlow

wasn't marked experimental. should I open a ticket / tweet to someone about it? I shouldn't need to release a new version of the library to make it binary compatible with greater versions. WHY this is DEEPLY troubling: If the language doesn't guarantee binary backward compatibility with a previous minor version (1.3.x -> 1.4.x) it means that developers will have to constantly update libraries and eventually maintain multiple versions compiled against multiple version of kotlin for projects that still use a previous minor. THIS IS A NO GO for any language. Libraries compiled for Java 6 run just fine on Java 13. I'm not kotlin need to be that conservative, but at least not break backward compatibility on stable api between minors? Deprecate them for a couple of minors is ok. This is not.

I've got a standalone helper method which does some work and I want to be able to log information within it in the context of the calling methods, but I find it a bit awkward/annoying to have to pass a logger to it, has anyone come up with a good strategy for things like that? I thought about: 1. Define a

HasLogger

interface (which defines a

logger

property) and make the helper method an extension on that, so that way it could access the logger, but that requires all classes make access to their logger instance public, which seems wrong. 2. Make the helper function an extension function on

Logger

directly. This feels like it would be awkward at the call site, as the function itself has nothing to do with logging but would look like

logger.myHelperFunction(...)

. Even wrapping the call in a

with(logger)

feels a bit weird because it feels like the fact that it logs shoudn't be so explicit. 3. Just suck it up and pass a logger instance. This works but I find it distracting. I have an auto-close-style helper

withResource

and calling

withResource(myResource) { ... }

feels a lot better than

withResource(logger, myResource) { ... }

which seems like it makes it confusing as to which resource is being managed. 4. Have the method use its own logger. This works but our logger instances are part of a hierarchy and include contextual information from the parents, which it would be nice to have when the helper is called in the scope of some class' logger. Has anyone done anything that worked well for a use case like this?

Is there a stdlib function for something like fold or reduce with a stopping condition?

var current: Segment? = null
data
  .segments
  .find { segment ->
    shouldSegmentBeShown()
      .let { (meetsDependency, crossesFrequencyThreshold) ->
        if(meetsDependency && !crossesFrequencyThreshold) {
          current = segment
        }
        meetsDependency && crossesFrequencyThreshold
      }
}

When enabling granular metadata in gradle:

kotlin.mpp.enableGranularSourceSetsMetadata=true
kotlin.native.enableDependencyPropagation=false

I'm getting this, and can't build

java.lang.IllegalStateException: e: Failed to resolve Kotlin library: D:\Projects\KotlinProjects\keyboard-mouse-kt\keyboard\build\kotlinSourceSetMetadata\commonMain\org.jetbrains.kotlinx-atomicfu-common\org.jetbrains.kotlinx-atomicfu-common-nativeInterop.klib

It seems the lib is generated and present.

I know that kotlin accepts the following syntax

fun <T> f(): Foo where T : Bar, T : Baz { ... }

but can I do the same with for example a map?

val map = mapOf<String, T (where T:bar,  T:Baz)>

Hi! I was thinking about replacing spring completely with ktor and other tools. As most things seem to be more or less easily replaceable i wasa wondering how to manage transactions in such a case. How do other people address this issue? While there are plenty examples of how to do programmatic transactions with r2dbc or exposed they all are very simplistic and only show how to use it in the data access layer. but of course we actually need to handle it on service layer so the transaction spans over multiple db queries. What tools are there to abstract the transaction handling?

Would be nice if there was some compact way to claim to a when statement that something is exhausive when the compiler cannot see that it is:

return when(Math.floorMod(value / 90 * rot.dir, 4)) {
        0 -> vector
        1 -> Point(-vector.j, vector.i)
        2 -> Point(-vector.i, -vector.j)
        3 -> Point(vector.j,-vector.i)
        else -> throw Exception("impossible")
    }

THe compiler here forces me to have the else. And I can't even have an assert False in the else, because that doesn't type check, I have to throw an exception which forces me to make up a name. Maybe something like !else at the end of the when or something like that. Or maybe some kind of special function with a contract that lets you do modulus where the compiler knows that only 0 to N-1 are possible outputs

Just curious, why is it impossible for overload resolution to distinguish these:

operator fun<R> R.rem(transform: (R) -> R) = transform(this)
operator fun<R> R.rem(transform: R.() -> R) = this.transform()

I'm not even allowed to declare both

apologies if this has been brought up before, but there's no Kotlin 1.4.20 (or .21) plugin released for IntelliJ 2020.3. anyone know why? https://plugins.jetbrains.com/plugin/6954-kotlin/versions

Hi! I have:

interface Attributes<T>
enum class StringAttributes : Attributes<T> { ... }

val attributes: Map<Attributes<Any>, Any>

val others: Map<StringAttributes, String> = ...

However, if I write:

attributes = others

I get a

Type Mismatch

. I guess the solution has something to do with properly using in and out, however I don't understand what the correct way to use those is.

Hi guys, How can the second sentence produce a NullPointerException?

lateinit var aLiveData: LiveData<List<..>>
val list = aLiveData.value

The crashlytics said there is a NPE in the second line. It occurs occasionally. I dont really get why, I thought it could be about the lateinit so I change it to a nullable variable (despite the fact there is no "lateinit exception"). Is that correct?

@KayCee Hi, you don’t define aLiveData

lateinit var aLiveData: LiveData<List<..>>
aLiveData = // define live data
val list = aLiveData.value

What's equivalent for kotlin gradle script (build.gradle.kts) of the following configurations in this article/guide. Script is not recognizing

model { platforms {

and things like that... Gradle docs seem to have no docs for kts as well in this topic 😞

in

object Blah {
    val context = context()
}

how can i get the name of the object (“Blah”) from the context method?

Hi there, how do I get to make

10_000_000_000.0.toString()

to display as

10000000000.0

insted of

1.0E10

?

hey! is there a way to create an enum class with 'normal' strings?

hello all! I’ve been trying to update my projecto to 1.4.21 (from 1.3.72) and I’m having a couple of errors with kotlin-reflect

w: Consider providing an explicit dependency on kotlin-reflect 1.4 to prevent strange errors
w: Some runtime JAR files in the classpath have an incompatible version. Consider removing them from the classpath

It seems kotlin-reflect is being used by android lint which is causing this error. I’ve already tried to: • declare the right version of kotlin-reflect • resolve it 1.4.21 via

resolutionStrategy

But the error persists. Any suggestion on how I can fix it? Thank you 🙂

when are kotlin objects intantiated? in

object MyObj { val blah=getBlah }

, when is getBlah called? when the containing file is imported? or when a method references

MyObj.blah

?

is it possible to throw from inside a map function without having to change the return type of the function to List<Any>?

How would I suppress the "this declaration is experimental" for unsigned types?

Should smart casting work normally when using nested

sealed

classes? I see a lot of issues like this https://youtrack.jetbrains.com/issue/KT-13495 marked as fixed a long time ago yet I experience this problem in 1.4.21

How do an object inside a class with a companion object keyword and an object inside the same class with just the object keyword differ? Would the one withour "companion" would be created upon every instance of the class?

Is it possible to something similar to be done in Kotlin https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Object/assign

Let's say I'm developing a chat implementation. A Chat object (one to one communication with polling) create its own internal CoroutineScope with a new thread pool CoroutineContext. This avoid exposing suspend functions, and instead each is wrapped in scope.launch. Is this a good approach?

is there a way to do a computation inside a

when

block?

return if (string == null) null
        else {
            val id = Json.parse(string)?.asObject()?.get("id")
            when {
                id == null -> null
                id.isString -> Util.entry(id.asString(), string)
                else -> Util.entry(id.asObject(), string)
            }
        }

here, i first check a

if

then compute the

id

then execute my

when

using the computed

id

it feels not as readable as it could

i’d like to put everything in the

when

but then, how do i compute the

id

so i can use it in the later evaluations?