Philipp Hauer's Blog

Engineering Management, Java Ecosystem, Kotlin, Sociology of Software Development

Best Practices for Unit Testing in Kotlin

Posted on Feb 12, 2018. Updated on Jun 12, 2022

Unit Testing in Kotlin is fun and tricky at the same time. We can benefit a lot from Kotlin’s powerful language features to write readable and concise unit tests. But in order to write idiomatic Kotlin test code in the first place, there is a certain test setup required. This post contains best practices and guidelines to write unit test code in Kotlin that is idiomatic, readable, concise and produces reasonable failure messages.

Best Practices for Unit Testing in Kotlin


  • Most of the best practices for unit testing in Java also apply for Kotlin tests.
  • Use JUnit5
  • Test Class Lifecycle
    • Use @TestInstance(Lifecycle.PER_CLASS) to avoid the need for static members, which are non-idiomatic and cumbersome in Kotlin.
    • Instead of annotating every class with @TestInstance() you can change the default lifecycle with the file.
  • Test Fixtures
    • Reuse one instance of the test class for every test methods (by using @TestInstance())
    • Initialize the required objects in the constructor (init) or in a field declaration (apply() is helpful). This way, the fields can be immutable (val) and non-nullable.
    • Don’t use @BeforeAll. It forces us to use lateinit or nullable types.
  • Put the test method names in backticks and use spaces.
  • Use @Nested inner classes to group the test methods.
  • Mocks
    • Use MockK to create mocks in a convenient and idiomatic way. It can also mock final classes by default.
    • Create mocks only once and reset them in a @BeforeEach. This way, you can use val fields and achieve better performance.
  • Test Libraries
    • There are so many test libraries. Check them out and make up your own mind.
    • When it comes to assertion libraries, I recommend Kotest’s matchers. They are powerful, idiomatic and concise.
  • Take advantage of data classes
    • Create a reference object and compare it directly with the actual object using an equality assertion.
    • Use data classes to carry the test data (input and expected output) in a @ParameterizedTest.
  • Helper Functions
    • Write helper functions with default arguments to easily create instances with a complex structure.
    • Helper extension functions are useful to concisely create frequently used values (like Instant, UUID)
    • Test-specific extension functions are useful to extend libraries in an idiomatic way to remove distractions in the test code.


Workshop: Deep Dive Into Testing With Kotlin

I offer a hands-on workshop about testing in Kotlin. It covers the full range of testing: mock-based unit tests, integration tests, property-based tests, and data-driven tests. In the end, you will be able to write clean and idiomatic tests and apply battle-proven best practices. Check out the details.

Recording of the KotlinConf Talk

At the KotlinConf 2018, I held a talk about this topic. You can watch the video here. Please note that the blog post below is more up to date than the recording. For instance, I now recommend the Kotest matchers instead of AssertJ.

Java Best Practices

Most of the best practices for unit testing in Java are still valid because they are language-agnostic.

Recap: What is Idiomatic Kotlin Code?

Let’s recap a few points about idiomatic Kotlin code:

  • Immutability. We should use immutable references with val instead of var.
  • Non-Nullability. We should favor non-nullable types (String) over nullable types (String?).
  • No static access. It impedes proper object-oriented design, because static access harms testability (can’t exchange objects easily) and obfuscates dependencies and side-effects. Kotlin strongly encourages us to avoid static access by simply not providing an easy way to create static members.

But how can we transfer these best practices to our test code?

Avoid Static and Reuse the Test Class Instance

In JUnit4, a new instance of the test class is created for every test method. So the initial setup code (that is used by all test methods) must be static. Otherwise, the setup code would be re-executed again and again for each test method. In JUnit4, the solution is to make those members static. That’s ok for Java as it has a static keyword. Kotlin doesn’t have this direct mean - for good reasons because static access is an anti-pattern in general.

//JUnit4. Don't:
class MongoDAOTestJUnit4 {

    companion object {
        private lateinit var mongo: GenericContainer
        private lateinit var mongoDAO: MongoDAO

        fun initialize() {
            mongo = startMongoContainer()
            mongoDAO = MongoDAO(, mongo.port)

    fun foo() {
        // test mongoDAO

Fortunately, JUnit5 provides the @TestInstance(Lifecycle.PER_CLASS) annotation. This way, a single instance of the test class is used for every method. Consequently, we can initialize the required objects once and assign them to normal fields of the test class. This happens only once because there is only one instance of the test class.

class MongoDAOTestJUnit5 {
    private val mongo = startMongoContainer().apply {
    private val mongoDAO = MongoDAO(, mongo.port)

    fun foo() {
        // test mongoDAO

First, this approach is more concise, because we don’t have to wrap everything into a companion object containing @JvmStatic annotated fields. Second, it’s idiomatic Kotlin code as we are using immutable non-nullable val references and can get rid of the nasty lateinit. Please note, that Kotlin’s apply() is really handy here. It allows object initialization and configuration without a constructor. But using a constructor or an initializer block (init { }) is sometimes more appropriate if the initialization code is getting more complex.

class MongoDAOTestJUnit5Constructor {
    private val mongo: KGenericContainer
    private val mongoDAO: MongoDAO

    init {
        mongo = startMongoContainer().apply {
        mongoDAO = MongoDAO(, mongo.port)

In fact, we don’t need JUnit5’s @BeforeAll (the equivalent of JUnit4’s @BeforeClass) anymore because we can utilize the means of object-oriented programming to initialize the test fixtures.

Side note: For me, the re-creation of a test class for each test method was a questionable approach anyway. It should avoid dependencies and side-effects between test methods. But it’s not a big deal to ensure independent test methods if the developer pays attention. For instance, we should not forget to reset or reinitialize fields in a @BeforeEach block and don’t (re-)assigned fields in general - which is not possible when we use val fields. ;-)

Change the Lifecycle Default for Every Test Class

Writing @TestInstance(TestInstance.Lifecycle.PER_CLASS) on every test class explicitly is cumbersome and easy to forget. Fortunately, we can set the default lifecycle for the whole project by creating the file src/test/resources/ with the content:

junit.jupiter.testinstance.lifecycle.default = per_class

Use Backticks and @Nested Inner Classes

  • Put the test method name in backticks. This allows spaces in the method name which highly improves readability. This way, we don’t need an additional @DisplayName annotation.
  • JUnit5’s @Nested is useful to group the test methods. Reasonable groups can be certain types of tests (like InputIsXY, ErrorCases) or one group for each method under test (GetDesign and UpdateDesign).
class DesignControllerTest {
    inner class GetDesigns {
        fun `all fields are included`() {}
        fun `limit parameter`() {}
        fun `filter parameter`() {}
    inner class DeleteDesign {
        fun `design is removed from db`() {}
        fun `return 404 on invalid id parameter`() {}
        fun `return 401 if not authorized`() {}
Readable and Grouped Tests Results in IntelliJ IDEA

Readable and Grouped Tests Results in IntelliJ IDEA

Kotlin Test Libraries

Due to the variety of available Kotlin test libraries we are spoilt for choice. Here is an incomplete overview of some Kotlin-native and Java libraries for testing, mocking and assertions (note that some libraries fit into multiple categories):

Test Frameworks Mocking Assertions
Kotlin Kotest, Spek Mockito-Kotlin, MockK Kotest, Kluent, Strikt, Atrium, HamKrest, Expekt, AssertK
Java JUnit5 AssertJ

For me, it’s a matter of taste. Check them out and make up your own mind. We ended up using plain JUnit5, MockK, and Kotest’s matchers.

Kotest for Assertions

For a long time, we still used the powerful AssertJ in our Kotlin projects and it’s still a good choice. But we moved to Kotest’s matchers. Kotest has the advantage of being written in Kotlin. So it can leverage Kotlin’s language features which allows you to write more idiomatic and concise assertions.

But Kotest is not only an assertion library; it’s a whole test framework. However, I still prefer JUnit5 (for now) due to its simplicity and the better tooling in the IDE. Fortunately, you can use Kotest’s matchers in isolation as they are located in a dedicated dependency.

Mock Handling

Final By Default

Classes and therefore methods are final by default in Kotlin. Unfortunately, some libraries like Mockito are relying on subclassing which fails in these cases. What are the solutions to this?

  • Use interfaces
  • open the class and methods explicitly for subclassing
  • Enable the incubating feature of Mockito to mock final classes. For this, create a file with the name org.mockito.plugins.MockMaker in test/resources/mockito-extensions with the content mock-maker-inline.
  • Use MockK instead of Mockito/Mockito-Kotlin. It supports mocking final classes by default.

Use MockK

I highly recommend using MockK. First, you don’t have to worry any longer about final classes or additional interfaces. Second, MockK provides a convenient and idiomatic API for writing mocks and verifications.

val clientMock: UserClient = mockk()
val user = User(id = 1, name = "Ben")
every { clientMock.getUser(any()) } returns user
val daoMock: UserDAO = mockk(relaxed = true)

val scheduler = UserScheduler(clientMock, daoMock)

verifySequence {

Note the usage of lambdas in verifySequence { } to nicely group verifications. Moreover, MockK provides reasonable error messages containing all tracked calls if verification fails. I also like that MockK fails with an exception if an unspecified method is called on a mock (strict mocking by default). So you don’t run into mysterious NullPointerExceptions known from Mockito.

Besides, MockK’s relaxed mocks are useful if the class under test uses a certain object, but you don’t want to define the behavior of this mock because it’s not relevant for the test. A relaxed mock returns dummy objects containing empty values.

val clientMock: UserClient = mockk(relaxed = true)
// we don't have to mock clientMock.getUser()
println(clientMock.getUser(1).age) // 0

But mind, that relaxed mocks can also lead to tricky errors, when you forget to mock a required method. I recommend using strict mocks by default and relaxed ones only if you really need it.

Create Mocks Once

Recreating mocks before every test is slow and requires the usage of lateinit var. So the variable can be reassigned which can harm the independence of each test.

class DesignControllerTest_RecreatingMocks {

    private lateinit var dao: DesignDAO
    private lateinit var mapper: DesignMapper
    private lateinit var controller: DesignController

    fun init() {
        dao = mockk()
        mapper = mockk()
        controller = DesignController(dao, mapper)

    // takes 2 s!
    fun foo() {

Instead, create the mock instance once and reset them before or after each test. It’s significantly faster (2 s vs 250 ms in the example) and allows using immutable fields with val.

// Do:
class DesignControllerTest {

    private val dao: DesignDAO = mockk()
    private val mapper: DesignMapper = mockk()
    private val controller = DesignController(dao, mapper)

    fun init() {

    // takes 250 ms
    fun foo() {

Handle Classes with State

The presented create-once-approach for the test fixture and the classes under test only works if they don’t have any state or can be reset easily (like mocks). In other cases, re-creation before each test is inevitable.

class DesignViewTest {

    private val dao: DesignDAO = mockk()
    private lateinit var view: DesignView // the class under test has state

    fun init() {
        view = DesignView(dao) // re-creation is required

    fun changeButton() {
        view.button.caption shouldBe "Hi"
        view.button.caption shouldBe "Hallo"

Utilize Data Classes

Data Classes for Assertions

Single Objects

If possible don’t compare each property for your object with a dedicated assertion. This bloats your code and - even more important - leads to an unclear failure message.

// Don't
val actualDesign = client.requestDesign(id = 1) shouldBe 2 // ComparisonFailure
actualDesign.userId shouldBe 9 shouldBe "Cat"

This leads to poor failure messages:

org.opentest4j.AssertionFailedError: expected:<2> but was:<1>
Expected :2
Actual   :1

Expected: 2. Actual: 1? What is the semantics of the numbers? Design id or User id? What is the context/the containing class? Hard to say.

Instead, create an instance of the data classes with the expected values and use it directly in a single equality assertion.

// Do
val actualDesign = client.requestDesign(id = 1)

val expectedDesign = Design(id = 2, userId = 9, name = "Cat")
actualDesign shouldBe expectedDesign

This way, we get a nice and descriptive failure message:

data class diff for com.phauer.Design
└ id: expected:<2> but was:<1>

expected:<Design(id=2, userId=9, name=Cat)> but was:<Design(id=1, userId=9, name=Cat)>
Expected :Design(id=2, userId=9, name=Cat)
Actual   :Design(id=1, userId=9, name=Cat)

We take advantage of Kotlin’s data classes. They implement equals() and toString() out of the box. So the equals check works and we get a really nice failure message. Plus, Kotest detects data classes and outputs the property that is different. Moreover, by using named arguments, the code for creating the expected object becomes very readable.


We can take this approach even further and apply it to lists. Here, Kotest’s powerful list assertions are shining.

// Do
val actualDesigns = client.getAllDesigns()

    Design(id = 1, userId = 9, name = "Cat"),
    Design(id = 2, userId = 4, name = "Dog")
java.lang.AssertionError: Expecting: [
    Design(id=1, userId=9, name=Cat),
    Design(id=2, userId=4, name=Dog)
] but was: [
    Design(id=1, userId=9, name=Cat),
    Design(id=2, userId=4, name=Dogggg)
Some elements were missing: [
    Design(id=2, userId=4, name=Dog)
] and some elements were unexpected: [
    Design(id=2, userId=4, name=Dogggg)

How cool is that?

Other Useful Kotests Assertions

Usually, comparing all properties of a data class is what you need in the test. But from time to time, it’s useful to ignore some properties or to compare only some properties.

actualDesign.shouldBeEqualToIgnoringFields(expectedDesign, Design::id)
actualDesign.shouldBeEqualToUsingFields(expectedDesign, Design::userId, Design::name)

Group Assertions With Kotest’s asClue

If you really want to assert only a few properties of a data class, consider grouping the assertions using Kotest’s asClue{} function. First, it nicely groups the code. Second, failure messages will also print the object that asClue was invoked on. That provides very useful context.

actualDesign.asClue { shouldBe 2 shouldBe "Cat"
org.opentest4j.AssertionFailedError: Design(id=1, userId=9, name=Cat) expected:<2> but was:<1>
Expected :2
Actual   :1

Data Classes for Parameterized Tests

Data classes can also be used for parameterized tests. Due to the automatic toString() implementation, we get a readable test result output in IDEA and the build.

fun `parse valid tokens`(data: TestData) {
    parse(data.input) shouldBe data.expected

private fun validTokenProvider() = Stream.of(
    TestData(input = "1511443755_2", expected = Token(1511443755, "2")),
    TestData(input = "151175_13521", expected = Token(151175, "13521")),
    TestData(input = "151144375_id", expected = Token(151144375, "id")),
    TestData(input = "15114437599_1", expected = Token(15114437599, "1")),
    TestData(input = null, expected = null)

data class TestData(
    val input: String?,
    val expected: Token?
If we use data classes as test parameters we get readable test results

If we use data classes as test parameters we get readable test results

Helper Functions

Use Helper Functions with Default Arguments to Ease Object Creation

In reality, data structures are complex and nested. Creating those objects again and again in the tests can be cumbersome. In those cases, it’s handy to write a utility function that simplifies the creation of the data objects. Kotlin’s default arguments are really nice here as they allow every test to set only the relevant properties and don’t have to care about the remaining ones.

fun createDesign(
    id: Int = 1,
    name: String = "Cat",
    date: Instant = Instant.ofEpochSecond(1518278198),
    tags: Map<Locale, List<Tag>> = mapOf(
        Locale.US to listOf(Tag(value = "$name in English")),
        Locale.GERMANY to listOf(Tag(value = "$name in German"))
) = Design(
    id = id,
    userId = 9,
    name = name,
    fileName = name,
    dateCreated = date,
    dateModified = date,
    tags = tags

// Usage
// only set the properties that are relevant for the current test
val testDesign = createDesign()
val testDesign2 = createDesign(id = 1, name = "Fox") 
val testDesign3 = createDesign(id = 1, name = "Fox", tags = mapOf())

This leads to concise and readable object creation code.

  • Don’t add default arguments to the data classes in the production code just to make your tests easier. If they are used only for the tests, they should be located in the test folder. So use helper functions like the one above and set the default arguments there.
  • Don’t use copy() just to ease object creation. Extensive copy() usage is hard to read; especially with nested structures. Prefer the helper functions.
  • Locate all helper functions in a single file like CreationUtils.kt. This way, we can reuse the functions like lego bricks for each test.

Helper Extension Functions for Frequently Used Values

I prefer to use fixed values instead of randomized or changing values. But writing code like Instant.ofEpochSecond(1525420010L) again and again is annoying and blows the code.

// Don't
val date1 = Instant.ofEpochSecond(1L)
val date2 = Instant.ofEpochSecond(2L)
val date3 = Instant.ofEpochSecond(3L)
val uuid1 = UUID.fromString("00000000-000-0000-0000-000000000001");
val uuid2 = UUID.fromString("00000000-000-0000-0000-000000000002");

Fortunately, we can write extension functions for frequently used objects like Instant, UUID or MongoDB’s ObjectId.

fun Int.toInstant(): Instant 
    = Instant.ofEpochSecond(this.toLong())

fun Int.toUUID(): UUID 
    = UUID.fromString("00000000-0000-0000-a000-${this.toString().padStart(11, '0')}")

fun String.toObjectId(): ObjectId 
    = ObjectId(this.padStart(24, '0'))

The code will become concise while keeping the different values visible:

val date1 = 1.toInstant()
val date2 = 2.toInstant()
val date3 = 3.toInstant()
val uuid1 = 1.toUUID()
val uuid2 = 2.toUUID()

The failure messages can be easily traced back to the test code:

Expected :00000000-0000-0000-a000-00000000001
Actual   :00000000-0000-0000-a000-00000000002

Test-Specific Extension Functions

Extension Functions can be useful to extend an existing library in a natural way. Here is an example: Kotest’s plusOrMinus() for floating comparison requires a tolerance parameter. This leads to a lot of code duplication, bloated code, and distraction while reading the tests.

// Don't
taxRate1 shouldBe 0.3f.plusOrMinus(0.001f)
taxRate2 shouldBe 0.2f.plusOrMinus(0.001f)
taxRate3 shouldBe 0.5f.plusOrMinus(0.001f)

Instead, define the extension function shouldBeCloseTo() on Float which delegates to a plusOrMinus() invocation with a fixed tolerance. This method is specific for the current test class and can be placed there.

private infix fun Float.shouldBeCloseTo(expected: Float) 
    = this shouldBe expected.plusOrMinus(0.001f)

The essential test logic becomes clearer and the code is still idiomatic as we stick to the pattern of Kotest.

// Usage:
taxRate1 shouldBeCloseTo 0.3f
taxRate2 shouldBeCloseTo 0.2f
taxRate3 shouldBeCloseTo 0.5f

The same technique can be used for all libraries with a fluent API.

Testcontainers: Reuse a Single Container

Testcontainers is a Java API to control containers within your test code. That’s great for executing your tests against a real database instead of an in-memory database. I highly recommend this approach. However, starting a new container for each test is usually a big waste of time. It’s better to create the container once and reuse it for every test. Kotlin’s object singletons and lazy initialized properties (by lazy { }) are very helpful here.

object MongoContainer {
    val instance by lazy { startMongoContainer() }

    private fun startMongoContainer() = KGenericContainer("mongo:4.0.2").apply {

// Usage:
class DesignDAOTest {
    private val container = MongoContainer.instance
    private val dao = DesignDAO(, container.port) // pseudo-code
    fun foo() { }
  • This way, we don’t need any test framework integration (like a JUnit5 extension for Testcontainers).
  • Don’t forget to clean up the database before each test method using a @BeforeEach method.
  • Be careful when you are running the tests in parallel. Consider using different “database names” (or “schema” in case of RDBs) for each test class to avoid side-effects.

Workshop: Deep Dive Into Testing With Kotlin

I offer a hands-on workshop about testing in Kotlin. It covers the full range of testing: mock-based unit tests, integration tests, property-based tests, and data-driven tests. In the end, you will be able to write clean and idiomatic tests and apply battle-proven best practices. Check out the details.

Further Reading

  • I highly recommend the book Kotlin in Action by Dmitry Jemerov and Svetlana Isakova