I'm trying to test a LiveData using the

getOrAwaitValue

helper function from the Google samples. But this LiveData is coming from a

MutableStateFlow

using

asLiveData

. There seems to be a race condition and I am not getting a value in my test. I've tried passing the

StandardTestDispatcher

to

asLiveData

but it didn't change anything.

Hi, I am trying to see if we can use the Intellij coverage agent with kover to replace jacoco. We currently use Jenkins and sonarqube to display the coverage results. The Jenkins jacoco plugin is looking for a *.exec file but when I point it to the *.ic file it doesn’t seem to be compatible. Is this correct?

Hello guys, we have a little problem with mocking in out unit tests. It looks a bit like an error per design in kotlin. but actually, i cant believe this 😄 we tried mockk and mockito regarding this and both result in the same problem. imagine we have an instance of the following class:

data class AppConfig(
    var smtp = SmtpConfig(
            enabled = false
            host = "test",
            port = 12345,
            useStartTLS = true,
            fromName = "test",
            from = "<mailto:test@test.de|test@test.de>"
            user = ""
            password = ""
    )
)

val classInstance = AppConfig()

what I want to do in one of our tests is to mock the value of smtp. What I did - here for example with mockito - was the following:

val spy = Mockito.spy(classInstance)
Mockito.`when`(spy.smtp).thenReturn(SmtpConfig(true)). //true is the enabled flag in this case, so the rest should be null

What Mockito and Mockk are doing in this case is to mock the get method behind that property. This can be verified by starting the debugger right in the line after the Mockito.\`when\` expression. The field is exactly like initialized above and when I click the getter in the debugger I get the result from spy.smtp, as you can see in the screenshot. I can understand why this happens, but kotlin simply does not use the getters of those properties why the tests wont work. So long text, simple question: Any idea how to fix this? Everything I found was a github issue from mockk with exactly the same problem and that has been closed because to support this a major rebuild would be necessary.

Is it possible to benchmark some code in a way that's not bound to a specific machine? I'd like to monitor performance regressions in CI but cannot make any assumption about what type of VM runs the code. Is something like this possible?

Hi, I am writing some unit tests for an

object

class. Is there a way to “reset” the instance? It keeps its state between tests.

Hi, I updated the Kotiln version and migrating to

runTest

alternate from

runBlocking

and I'm facing an issue in testing a specific use case. Here is a simplified version of the test case I want to run.

@Test(expected = NullPointerException::class)
    fun test() = runTest {

        //In my actual code, I'm testing a ViewModel which uses viewModelScope to launch a coroutine
        //and it will throw an exception for a specific case.
        val scope = CoroutineScope(SupervisorJob() + Dispatchers.Main.immediate)
        scope.launch {
            throw NullPointerException("error")
        }
    }

Hello I hope I am on the right channel… We have some issues with

Sonar

coverage. and

value

classes.. and

init

function. it looks like the coverage ist not right on sonar. we use the

jacoco

maven plugin version

0.8.7

I'm migrating a program to Kotlin 1.6, and from build.gradle to build.gradle.kts. I'm using

spek

and

mocck

for testing. I'm getting a gradle error and wondered if someone could help me understand what it means?

Execution failed for task ':compileTestKotlin'.
> Could not resolve all files for configuration ':testCompileClasspath'.
   > Could not resolve org.jetbrains.kotlin:kotlin-test-junit5:1.6.21.
     Required by:
         project :
      > Module 'org.jetbrains.kotlin:kotlin-test-junit5' has been rejected:
           Cannot select module with conflict on capability 'org.jetbrains.kotlin:kotlin-test-framework-impl:1.6.21' also provided by [org.jetbrains.kotlin:kotlin-test-junit:1.6.21(junitApi)]
   > Could not resolve org.jetbrains.kotlin:kotlin-test-junit:1.6.21.
     Required by:
         project : > io.insert-koin:koin-test:3.2.0 > io.insert-koin:koin-test-jvm:3.2.0
      > Module 'org.jetbrains.kotlin:kotlin-test-junit' has been rejected:
           Cannot select module with conflict on capability 'org.jetbrains.kotlin:kotlin-test-framework-impl:1.6.21' also provided by [org.jetbrains.kotlin:kotlin-test-junit5:1.6.21(junit5Api)]

From my build.gradle.kts file, I have the following test configuration:

dependencies {
// ...
// testing
	testImplementation("io.insert-koin:koin-test:$koin_version")
//	testImplementation("org.jetbrains.kotlin:kotlin-test:$kotlin_version")
	testImplementation("org.jetbrains.kotlin:kotlin-test-junit5:$kotlin_version")
	testImplementation("org.spekframework.spek2:spek-dsl-jvm:$spek_version")
	testRuntimeOnly("org.spekframework.spek2:spek-runner-junit5:$spek_version")
	testImplementation("io.mockk:mockk:$mockk_version")
}

tasks.test {
	useJUnitPlatform {
		includeEngines("spek2")
	}
}

tasks.withType<KotlinCompile> {
	kotlinOptions.jvmTarget = "11"
}

Just trying to make sense of the error really.

does

kover

support

src/androidTest

?

Any good examples of unit test setup for a multiplatform module, including jvm, android, ios, watchos, macos and optionally js? Thanks!

New Version of KMock is out: https://github.com/bitPogo/kmock/releases

And KFixture https://github.com/bitPogo/kfixture

hi folks, I'm looking for an equivalent of PITest for mutation testing, for Kotlin. PITest which is optimized for Java doesn't give great results for Kotlin, and this plugin was deprecated. The author recommends a commercial tool called arcmutate. Any thoughts, experience to share?

Hey Assuming we need to advance coroutine time using

advanceUntilIdle()

method. I wonder where this method belongs in BDD (

When

or

Then

block?)

@Test
    fun `test something`() = runTest {
        // given
        setup code

        // when
        subject.init()
        advanceUntilIdle()

        // then
        verify something
    }

or

@Test
    fun `test something`() = runTest {
        // given
        setup code

        // when
        subject.init()

        // then
        advanceUntilIdle()
        verify something
    }

CanI have the JS Promise handling of

runTest {}

but not the delay-skipping behaviour? Delays are a part of the behaviour I want to test

Stop Using JUnit BeforeEach!

How to Test Shared Behaviour with JUnit Contracts

A New Way to Manage JUnit Test State in Kotlin

Is it good to test all error case in one test func? for example:

@Test
fun `my error test`() {
    every { apiManager.foo() }.returnsMany(
        Observable.just(Response.error("Error A".toResponseBody())),
        Observable.just(Response.error("Error B".toResponseBody())),
        Observable.just(Response.error("Error C".toResponseBody())),
    )

    repeat(3) { presenter.bar() }

    verifyOrder {
        view.showErrorA()
        view.showErrorB()
        view.showErrorC()
    }
}

Has anyone managed to get the TestParameterInjector constructor injection working with kotlin value classes? It looks like Kotlin generates an additional constructor when there’s a value class in the constructor parameter list, which confuses the library only expecting a single constructor.

Hi, are there any plans to add

@DisplayName

and

@Nested

to the kotlin.test core annotations list? I find these annotations useful when writing JUnit tests in Kotlin/JVM.

@DisplayName

-- Because Kotlin/JS does not support spaces in backtick function names, and Kotlin/native does not allow parenthesis having a

@DisplayName

annotation allows for a more readable test name (and the ability to specify the class description).

@Nested

-- I often find it useful to use this in my tests, especially to group different partitions or use cases for the thing under test. For example, testing a "to list" or "list of" method with zero, one, and many items.

I’m implemented the Paging 3 library and I’d like to test my repository. I’ve seen the Paging 3 testing webpage but there’s only theory about the repository tests. My repository looks like this:

class RepositoriesRepositoryImpl @Inject constructor(
    @IoCoroutineDispatcher private val ioCoroutineDispatcher: CoroutineDispatcher,
    private val service: GithubService,
    private val database: RepositoriesDatabase
) : RepositoriesRepository {
    @OptIn(ExperimentalPagingApi::class)
    override suspend fun fetchRepositories() = withContext(ioCoroutineDispatcher) {
        return@withContext Pager(config = PagingConfig(
            pageSize = NETWORK_PAGE_SIZE, enablePlaceholders = false
        ),
            remoteMediator = GithubRemoteMediator(ioCoroutineDispatcher, service, database),
            pagingSourceFactory = { database.repositoriesDao().allRepositories() }).flow
    }

    companion object {
        const val NETWORK_PAGE_SIZE = 10
    }
}

And this is my test class:

@ExperimentalCoroutinesApi
class RepositoriesRepositoryTest {
    private lateinit var sut: RepositoriesRepository
    private val testDispatcher = UnconfinedTestDispatcher()
    private val fakeGithubService: GithubService = mock {
        onBlocking { fetchRepos(any(), any()) } doSuspendableAnswer {
            Response.success((listOf()))
        }
    }
    private val inMemoryDatabase = Room.inMemoryDatabaseBuilder(
        ApplicationProvider.getApplicationContext(), RepositoriesDatabase::class.java
    ).build()

    @Before
    fun setup() {
        sut = RepositoriesRepositoryImpl(testDispatcher, fakeGithubService, inMemoryDatabase)
    }

    @Test
    fun test1() = runTest {
        sut.fetchRepositories().test {
            val item: PagingData<Repository> = awaitItem()
            assertNull(item)
        }
    }
}

What I’m trying is to assert the content of the

PagingData

, just to see that it contains whatever number repository I’m passing in the test expectation. How could I do this? Thanks!

Hello everybody, I have a model like

sealed interface AffectsA
sealed interface AffectsB

sealed interface Action {

  object Act1 : Action, AffectsA
  object Act2 : Action, AffectsA, AffectsB
  object Act3 : Action, AffectsB
}

and a class like

class Reducer(
  private val subReducer1: X,
  private val subReducer2: Y
) {

  fun reduce(state: State, action: Action): State =
    when (action) {
      is Act1 -> state.copy(
        a = subReducer1.reduce(state.a, action)
      )
      Act2 -> state.copy(
        a = subReducer1.reduce(state.a, action),
        b = subReducer2.reduce(state.b, action)
      )
      Act3 -> state.copy(
        b = subReducer2.reduce(state.b, action)
      )
    }
}

I wanna test that for every action, the correct sub-reducer is called. Something like

reducer.reduce(someState, Act2)
verify {
  subReducer1.reduce(someState.a, Act2)
  subReducer2.reduce(someState.b, Act2)
}

/// 

reducer.reduce(someState, Act3)
verify {
  subReducer1 wasNot Called
  subReducer2.reduce(someState.b, Act2)
}

I usually like to use

org.junit.runners.Parameterized

for these cases, but I wanna ensure that all the cases are covered, a.k.a. every sub-type of Action. Got any idea?

Does anyone have problems with mocking with Kotlin 1.7.20 https://github.com/mockito/mockito/issues/2777 ?

Is the kotlin team planning on an analog to Junit timeout rules? I currently have a mpp test suite where the failure mode is a synchronous infinite loop, which I would like to fail after like 100ms.

I’m working on a class that handles performing a suspending task (logging the user in) as well as maintaining a SharedFlow so other components can be alerted when the task is performed:

class ExampleClass(
    private val loginApi: LoginApi,
    externalScope: CoroutineScope,
) {
    private val _loginFlow = MutableSharedFlow<String>()

    val loginFlow = _loginFlow.shareIn(externalScope, SharingStarted.Lazily)

    suspend fun performLogin(): String {
        val result = loginApi.performLogin()
        _loginFlow.emit(result)
        return result
    }
}

interface LoginApi {
    suspend fun performLogin(): String
}

I’ve written a test case for this:

class ExampleClassTest {

    private val mockApi = mockk<LoginApi> {
        coEvery { performLogin() } returns "hello"
    }

    private val testCoroutineScheduler = UnconfinedTestDispatcher()
    private val testScope = TestScope(testCoroutineScheduler)

    @Test
    fun `this test passes`() = runTest(testCoroutineScheduler) {
        val exampleClass = ExampleClass(mockApi, testScope)
        expect("hello") {
            exampleClass.performLogin()
        }
    }
}

Can anyone help me understand why calling runTest directly on my TestScope causes this test case to fail with an eventual UncompletedCoroutinesError? The error message says the test coroutine is not completing, and also that there’s an active child job.

@Test
    fun `this test times out`() = testScope.runTest {
        val exampleClass = ExampleClass(mockApi, testScope)
        expect("hello") {
            exampleClass.performLogin()
        }
    }

I’ve just been informed for cases like this, you’re supposed to use

TestScope.backgroundScope

. Perhaps this example is more correct:

class ExampleClassTest {
    private val mockApi = mockk<LoginApi> { coEvery { performLogin() } returns "hello" }
    private val testCoroutineScheduler = UnconfinedTestDispatcher()
    private val testScope = TestScope(testCoroutineScheduler)

    @Test
    fun `this test passes`() = testScope.runTest {
        val exampleClass = ExampleClass(mockApi, testScope.backgroundScope)
        expect("hello") {
            exampleClass.performLogin()
        }
    }
}

Hi @Gabriel Feo Did you find a solution for this?

I find nice that

kotlin.test

besides convenient functions like assertFailsWith, it has inspections to guide devs to use the proper one, which will give more useful error messages