In this guide, we will move from basic JUnit setup to advanced property-based testing and coroutine simulation. Forget the old @Test annotations that feel clunky. Kotlin allows us to write tests that read like plain English. The Setup (Gradle) // build.gradle.kts dependencies { testImplementation("org.junit.jupiter:junit-jupiter:5.10.0") testImplementation("org.jetbrains.kotlin:kotlin-test-junit5") } The First Test: Clean Syntax Notice how we avoid assertThat(actual).isEqualTo(expected) (Hamcrest) or Assert.assertEquals() (JUnit). Instead, we use Kotlin's infix functions via the kotlin.test library.
src/ test/kotlin/ # Unit tests (run fast, no Android/Server) integrationTest/ # Integration tests (use real DB) testFixtures/ # Shared test data (factories, builders) curso de testing kotlin
import kotlin.test.Test import kotlin.test.assertEquals import kotlin.test.assertNotNull class CalculatorTest { In this guide, we will move from basic
Use TestDispatcher and advanceUntilIdle() to control time precisely. Module 4: Mocking with MockK (Not Mockito) If you come from Java, you know Mockito. For Kotlin, you need MockK . Why? Because Mockito fails when dealing with final classes (Kotlin classes are final by default) and suspend functions. Mocking a suspend function import io.mockk.coEvery import io.mockk.coVerify import io.mockk.mockk class RepositoryTest { The Setup (Gradle) // build
@Test fun `verify API is called only once`() = runTest { // 1. Create mock val api = mockk<MyApi>() // 2. Stub a suspend function (coEvery) coEvery { api.getData() } returns "Mocked Response" val repo = Repository(api) // 3. Execute val result = repo.refreshData() // 4. Verify (coVerify) coVerify(exactly = 1) { api.getData() } result shouldBe "Mocked Response" } } Traditional testing (Example-based) says: "Give input 2+2, check output 4." Property-based testing says: "For ALL integers, addition should be commutative."
It’s time to change that. Welcome to your conceptual .
