I am getting

Engine exception: Kotlin reflection implementation is not found at runtime. Make sure you have kotlin-reflect.jar in the classpath

when running kotest tests from Android Studio with the help of kotest idea plugin. The same running from command line using gradle works just fine.

Kotest 5.5.5 released https://kotest.io/docs/next/changelog.html

Is Kotest broken on JS? I can't figure out a way to execute JS tests. Even the demo doesn't seem to run the JS tests:

$ ./gradlew jvmTest jsBrowserTest
> Task :jvmTest
io.kotest.examples.mpp.UUIDJvmTest[jvm] > uuids should be in be type 4 format[jvm] PASSED
io.kotest.examples.mpp.data.DataDrivenTest[jvm] > PythagTriple(a=3, b=4, c=5)[jvm] PASSED
io.kotest.examples.mpp.data.DataDrivenTest[jvm] > PythagTriple(a=6, b=8, c=10)[jvm] PASSED
io.kotest.examples.mpp.uuid.UUIDTestCommon[jvm] > uuids should be somewhat unique![jvm] PASSED
Deprecated Gradle features were used in this build, making it incompatible with Gradle 8.0.

When I run Kotest tests by clicking the green arrow on the test class, it uses Gradle to run them, but when I run individual tests, or run from the Kotest plugin window, it runs them differently (directly with JDK?). How can I make the plugin use Gradle to run the tests?

Hello I am trying Json support in Kotest from https://kotest.io/docs/assertions/json/json-overview.html Is there support for ignoring some values like https://github.com/lukas-krecan/JsonUnit#ignoring-values ?

I've started getting with the last couple (5.5.4 and 5.5.5) versions of kotest.

[4.128s][warning][gc,alloc] ClassGraph-worker-16: Retried waiting for GCLocker too often allocating 1026 words

UnknownClass.Kotest > UnknownClass.initializationError FAILED
    org.junit.platform.commons.JUnitException at EngineExecutionOrchestrator.java:113
        Caused by: io.kotest.engine.config.ConfigurationException at ConfigManager.kt:29
            Caused by: io.github.classgraph.ClassGraphException at ClassGraph.java:1606
                Caused by: java.lang.OutOfMemoryError at NestedJarHandler.java:654

UnknownClass.Kotest FAILED

If I keep re-running it eventually works. Is there a kotest leak? Or am I missing some kotest setting or is there something else I should be modifying? thx

In preparing PR #3384 for review, I tried to build, but keep getting errors that seem unrelated to the work I did. Advise, please. Thanks! Here's an example:

Failed to execute all tests:
  :kotest-assertions:kotest-assertions-json:iosX64Test: java.lang.IllegalStateException: command '/usr/bin/xcrun' exited with errors (exit code: 148):kotest-assertions:kotest-assertions-json:watchosX64Test: java.lang.IllegalStateException: command '/usr/bin/xcrun' exited with errors (exit code: 148)

I’m wondering if kotest could get snapshot tests support. See https://github.com/kotest/kotest/issues/777

I just updated the kotest libraries to v5.5.5 and it seems there are some files missing the the Maven Central repository. There is an issue open in the kotest github's repository: https://github.com/kotest/kotest/issues/3388 Is there a way to get around this issue?

FAILURE: Build failed with an exception.

* What went wrong:
Execution failed for task ':common-core:compileTestKotlinIosX64'.
> Could not resolve all files for configuration ':common-core:iosX64TestCompileKlibraries'.
   > Could not find kotest-common.klib (io.kotest:kotest-common-iosx64:5.5.5).
     Searched in the following locations:
         <https://repo.maven.apache.org/maven2/io/kotest/kotest-common-iosx64/5.5.5/kotest-common-iosx64-5.5.5.klib>

Where may I find snapshot version numbers for Kotest?

Hello! What's the best way if I want to have an exhaustive generator + well known edge cases?

And just test the exhaustive list + edge-cases

hi folks, i have a question about the usage of the kotest-bom. I was trying to harmonize the kotest version numbers in a maven multimodule setup and put the kotest-bom into the parent of that project. I have noticed that the bom does not contain all kotest artifacts (e.g. kotest-assertions-core-jvm is not there). In effect i had to define the mentioned dependency with the version number. Is it on purpose that the bom does not contain all kotest artifacts?

Hello folks! I have a question with regards to clues - certain other frameworks, most notably Spock and Hamcrest, when an assertion fails they show the data under test that failed, by default. Is there any way to do this in Kotest without having to wrap everything in clues?

Hello people. I’m struggling to find the right way to use `CoroutinesScope`s in Kotest. I’m refactoring a test where I need a scope for my SUT, and before I had this:

private val appScope = TestScope()
private val sut = Sut(appScope)

@Test
fun `some test`() = appScope.runTest { ... }

Now, with Kotest, I did this:

class MyTest : BehaviorSpec({

    Given("abc") {
        val sut = Sut(testScope)

It worked, till I needed to call

advanceUntilIdle()

(after this code) so, to achieve that, I had to add

coroutineTestScope = true

and it became

class MyTest : BehaviorSpec({
    coroutineTestScope = true

    Given("abc") {
        val sut = Sut(testScope)

But this doesn’t seem to work correctly, as the Flow internal to my SUT never emits. I’ll drop my SUT in the 🧵

What's the latest on when

5.6.0

might be released?

Hello all, question about Arbs. I have a situation where my code is trying to generate an impossible-to-generate arb. To show a simple example:

data class Foo(val x: Boolean)

val fooArb: Arb<Foo> = arbitrary {
    Foo(x = false)
}

val stuck = fooArb.filter { it.x }.next()

Now, obviously this won't work. This is a simple example but in my case, the arb is much more complex (and transient) so the situation wasn't immediately obviously. What I would have hoped would happen is that the program would eventually time out or throw trying to generate the "next" arb, but what I am witnessing is the kotest test hangs seemingly indefinitely, making it very hard to debug. What I am wonder is if there is a way to better detect this situation? Can I set a timeout? Force an error? Etc... ** Not a contribution

Hi, We are using DescribeSpec. Currently I am transforming some TestNG Classes to kotest and run into the issue to change the amount of invocations I found this Pull Request, but when I look into config with `describe(“foo”).config() I can see only:

fun config(
      enabled: Boolean? = null,
      enabledIf: EnabledIf? = null,
      enabledOrReasonIf: EnabledOrReasonIf? = null,
      tags: Set<Tag>? = null,
      timeout: Duration? = null,
      failfast: Boolean? = null,
      blockingTest: Boolean? = null,
      coroutineTestScope: Boolean? = null,
      test: suspend T.() -> Unit
   )

When I try to do that at the root level overriding the

defaultTestCaseConfig

function as described here, I see that this method is now deprecated, but I do not understand how else to do that. Maybe someone can help me?

Good morning. I have a problem migrating my small integration test suite to Kotest. I have a custom

MockAppExtension

(previously

MockAppRule

) that sets up Koin with Ktor’s `MockEngine`s and in-memory SqlDelight database. I have 3 test classes with 17 tests in total. If I run a single class, all of them pass with no problem, but if I run tests for the whole module, 2 out of 17 fail with

io.ktor.serialization.JsonConvertException: Illegal input

on a network request I wonder what could make the difference 🤔 I tried with different `IsolationMode`s

The concurrency docs say:

We can configure this feature by setting the configuration field

concurrencyMode

or the system property

kotest.framework.concurrency.mode

.

Where is “the configuration field concurrencyMode”? Even searching the codebase, the only place I can find “concurrencyMode” is in the docs.

Hello everyone, I'm not sure which framework to blame, but I can't get test filtering working with Kotest, Maven Failsafe and Tags, given a simple example like this:

package com.example.demo

import io.kotest.core.spec.style.ShouldSpec
import org.junit.jupiter.api.Tag
import org.junit.jupiter.api.fail

@Tag("IntegrationTests")
class IntegrationIT : ShouldSpec({

    should("fail") {
        fail { "fail" }
    }
})

This test will always be run by failsafe, even if I specify the groups/tags to run to be something else

Hello, is there a way to group multiple tests that follow the

StringSpec

a bit like we an use InnerClass with JUnit5? What I have :

"myMethod does ABC when 123" { 

}

"myMethod does XYZ when 666" {

}

What I want :

"myMethod" { 
    "does ABC when 123" {

    }
    "does XYZ when 666" {

    }
}

Hey so I'm making use of https://kotest.io/docs/framework/datatesting/data-driven-testing.html What would be the best way to run individual tests? My list of tests and test names are stable. (See picture) when I right click a test I can't rerun a specific test. These tests are also from files, I would love to be able to double click the test name and for the file to open up. There's a

Rerun Failed Tests

button which is also useful, but I'm not sure how the failed test names are fed in so I can filter the tests being run. Thanks. Edit: I just realized I've been using a for loop over my input and just calling

test

multiple times. I tried

withData

and it's the same.

If using

InstancePerTest

or

InstancePerLeaf

, does

autoClose(x)

close x when the spec class is finished (ie: all of the tests in the spec have completed) or when the specific spec instance is finished?

hi folks, while working on a project that uses the http4k kotest matchers we found that its a bit hard to re-use kotests json matchers. So we have made some changes to the json matchers to make it easier to re-use them. I hope that change is acceptable and can me merged

hello. I am facing a "possibly" simple issue. How would i initialise the class under test before each test run in a ShouldSpec test? I have a test like the one below but i do not know how to correctly initialize the

engine

. I don't want to change the isolation mode as i would like to do some other data loading once upfront (to feed the engine).

class EngineTest : ShouldSpec({

    val engineInput = loadExpensiveInput()

    // does not work and lateinit is not available
    var engine: Engine
    
    beforeEach {
        engine = Engine(engineInput)
    }

    should("be ready") {
        engine.ready shouldBe true
    }
}

Something like the following would be nice.

val engine by beforeEach { Engine(engineInput) }

Hi everyone! I would like to know if anyone here used Kotest on an Android app with Hilt (DI framework) and what your experiences were. I don't see much content online about the two of them together. I don't have experience in either, besides creating the first few classes in separate projects. Did anyone have any compatibility issues?

My common tests don’t seem to be running at all when targeting ios specifically: •

jvmTest

runs

commonTest

targeting JVM and runs tests •

iosSimulatorArm64Test

doesn’t seem to compile/run

commonTest

targets iOS target 🤔 I have

framework-engine

dependency added to

commonTest

, anything I might be missing?