How do I use @Mock and @InjectMocks with JUnit?

To use @Mock and @InjectMocks with JUnit, you typically use them with Mockito.

  • @Mock creates a fake/mock dependency.
  • @InjectMocks creates the class under test and injects the mocks into it.
  • With JUnit 5, you enable Mockito using @ExtendWith(MockitoExtension.class).

1. Add Mockito dependencies

Maven

<dependencies>
    <dependency>
        <groupId>org.junit.jupiter</groupId>
        <artifactId>junit-jupiter</artifactId>
        <version>5.11.4</version>
        <scope>test</scope>
    </dependency>

    <dependency>
        <groupId>org.mockito</groupId>
        <artifactId>mockito-junit-jupiter</artifactId>
        <version>5.14.2</version>
        <scope>test</scope>
    </dependency>
</dependencies>

Gradle

dependencies {
    testImplementation 'org.junit.jupiter:junit-jupiter:5.11.4'
    testImplementation 'org.mockito:mockito-junit-jupiter:5.14.2'
}

test {
    useJUnitPlatform()
}

2. Example class to test

Suppose you have a service that depends on a repository:

public class UserService {
    private final UserRepository userRepository;

    public UserService(UserRepository userRepository) {
        this.userRepository = userRepository;
    }

    public String getUsernameById(Long id) {
        User user = userRepository.findById(id);

        if (user == null) {
            return "Unknown";
        }

        return user.getName();
    }
}

Repository:

public interface UserRepository {
    User findById(Long id);
}

Model:

public class User {
    private final Long id;
    private final String name;

    public User(Long id, String name) {
        this.id = id;
        this.name = name;
    }

    public Long getId() {
        return id;
    }

    public String getName() {
        return name;
    }
}

3. Use @Mock and @InjectMocks

import org.junit.jupiter.api.Test;
import org.junit.jupiter.api.extension.ExtendWith;
import org.mockito.InjectMocks;
import org.mockito.Mock;
import org.mockito.junit.jupiter.MockitoExtension;

import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.mockito.Mockito.when;

@ExtendWith(MockitoExtension.class)
class UserServiceTest {

    @Mock
    private UserRepository userRepository;

    @InjectMocks
    private UserService userService;

    @Test
    void getUsernameByIdReturnsUserNameWhenUserExists() {
        when(userRepository.findById(1L))
                .thenReturn(new User(1L, "Alice"));

        String result = userService.getUsernameById(1L);

        assertEquals("Alice", result);
    }

    @Test
    void getUsernameByIdReturnsUnknownWhenUserDoesNotExist() {
        when(userRepository.findById(99L))
                .thenReturn(null);

        String result = userService.getUsernameById(99L);

        assertEquals("Unknown", result);
    }
}

How it works

@Mock
private UserRepository userRepository;

This tells Mockito to create a mock implementation of UserRepository.

@InjectMocks
private UserService userService;

This tells Mockito to create a UserService instance and inject the mocked UserRepository into it.

Mockito tries injection in this order:

  1. Constructor injection
  2. Setter injection
  3. Field injection

Constructor injection is usually the best option because it makes dependencies explicit and easier to test.

Verifying mock interactions

You can also verify that a dependency method was called:

import static org.mockito.Mockito.verify;

@Test
void getUsernameByIdCallsRepository() {
    when(userRepository.findById(1L))
            .thenReturn(new User(1L, "Alice"));

    userService.getUsernameById(1L);

    verify(userRepository).findById(1L);
}

Common mistake: forgetting Mockito extension

If you forget this:

@ExtendWith(MockitoExtension.class)

then your @Mock fields may remain null, causing a NullPointerException.

JUnit 4 version

If you are using JUnit 4, use @RunWith(MockitoJUnitRunner.class) instead:

import org.junit.Test;
import org.junit.runner.RunWith;
import org.mockito.InjectMocks;
import org.mockito.Mock;
import org.mockito.junit.MockitoJUnitRunner;

import static org.junit.Assert.assertEquals;
import static org.mockito.Mockito.when;

@RunWith(MockitoJUnitRunner.class)
public class UserServiceTest {

    @Mock
    private UserRepository userRepository;

    @InjectMocks
    private UserService userService;

    @Test
    public void getUsernameByIdReturnsUserNameWhenUserExists() {
        when(userRepository.findById(1L))
                .thenReturn(new User(1L, "Alice"));

        String result = userService.getUsernameById(1L);

        assertEquals("Alice", result);
    }
}

Quick summary

Use this pattern for JUnit 5:

@ExtendWith(MockitoExtension.class)
class MyServiceTest {

    @Mock
    private MyRepository repository;

    @InjectMocks
    private MyService service;

    @Test
    void testSomething() {
        when(repository.findSomething()).thenReturn("value");

        String result = service.doSomething();

        assertEquals("value", result);
    }
}

Use @Mock for dependencies and @InjectMocks for the class you are testing.

How do I mock dependencies in unit tests?

To mock dependencies in unit tests, you usually use a mocking framework such as Mockito. Mocking lets you test one class in isolation without running the real logic of its collaborators.

Basic Mockito Example

Suppose you have a service that depends on another class:

@Service
public class MyService {

    private final MyDependency dependency;

    public MyService(MyDependency dependency) {
        this.dependency = dependency;
    }

    public void process() {
        dependency.doSomething();
    }
}

You can mock MyDependency in a unit test like this:

import org.junit.jupiter.api.Test;
import org.junit.jupiter.api.extension.ExtendWith;
import org.mockito.InjectMocks;
import org.mockito.Mock;
import org.mockito.junit.jupiter.MockitoExtension;

import static org.mockito.Mockito.verify;

@ExtendWith(MockitoExtension.class)
class MyServiceTest {

    @Mock
    private MyDependency dependency;

    @InjectMocks
    private MyService myService;

    @Test
    void processCallsDependency() {
        myService.process();

        verify(dependency).doSomething();
    }
}

What the annotations mean

  • @Mock creates a mock object.
  • @InjectMocks creates the class under test and injects the mocks into it.
  • @ExtendWith(MockitoExtension.class) enables Mockito support in JUnit 5.
  • verify(...) checks that a method was called.

Mocking return values

If the dependency returns a value, use when(...).thenReturn(...):

import static org.mockito.Mockito.when;

when(repository.findNameById(1L)).thenReturn("Alice");

Example:

@Test
void returnsMockedValue() {
    when(userRepository.findNameById(1L)).thenReturn("Alice");

    String result = userService.getUserName(1L);

    assertEquals("Alice", result);
}

Mocking exceptions

You can also make a mock throw an exception:

import static org.mockito.Mockito.when;

when(repository.findById(99L))
        .thenThrow(new EntityNotFoundException("User not found"));

For void methods, use doThrow(...):

import static org.mockito.Mockito.doThrow;

doThrow(new RuntimeException("Failure"))
        .when(dependency)
        .doSomething();

Verifying interactions

You can verify how your class interacted with its dependencies:

verify(dependency).doSomething();
verify(dependency, times(1)).doSomething();
verify(dependency, never()).doSomething();

Spring Boot unit test example

For a pure unit test, prefer Mockito without starting the Spring context:

@ExtendWith(MockitoExtension.class)
class MyServiceTest {

    @Mock
    private MyDependency dependency;

    @InjectMocks
    private MyService service;

    @Test
    void processCallsDependency() {
        service.process();

        verify(dependency).doSomething();
    }
}

Spring integration-style test

If you need the Spring context, use Spring’s test support and replace a bean with a mock:

@SpringBootTest
class MyServiceSpringTest {

    @MockitoBean
    private MyDependency dependency;

    @Autowired
    private MyService service;

    @Test
    void processCallsDependency() {
        service.process();

        verify(dependency).doSomething();
    }
}

Use this style when you want to test Spring wiring, configuration, transactions, security, or other framework behavior.

Rule of thumb

  • Use Mockito @Mock + @InjectMocks for fast unit tests.
  • Use Spring test annotations only when you need the Spring application context.
  • Mock external systems such as databases, APIs, message queues, and file systems.
  • Avoid mocking simple value objects or the class you are actually testing.

How do I use Mockito with JUnit?

To use Mockito with JUnit, you add Mockito to your test dependencies, enable Mockito in your JUnit test class, then create mocks and define their behavior.

Below is a simple JUnit 5 + Mockito example.

1. Add Dependencies

Maven

<dependencies>
    <!-- JUnit 5 -->
    <dependency>
        <groupId>org.junit.jupiter</groupId>
        <artifactId>junit-jupiter</artifactId>
        <version>5.11.4</version>
        <scope>test</scope>
    </dependency>

    <!-- Mockito Core -->
    <dependency>
        <groupId>org.mockito</groupId>
        <artifactId>mockito-core</artifactId>
        <version>5.14.2</version>
        <scope>test</scope>
    </dependency>

    <!-- Mockito integration for JUnit 5 -->
    <dependency>
        <groupId>org.mockito</groupId>
        <artifactId>mockito-junit-jupiter</artifactId>
        <version>5.14.2</version>
        <scope>test</scope>
    </dependency>
</dependencies>

Make sure your Maven Surefire plugin supports JUnit 5:

<build>
    <plugins>
        <plugin>
            <groupId>org.apache.maven.plugins</groupId>
            <artifactId>maven-surefire-plugin</artifactId>
            <version>3.5.2</version>
        </plugin>
    </plugins>
</build>

Gradle

dependencies {
    testImplementation 'org.junit.jupiter:junit-jupiter:5.11.4'
    testImplementation 'org.mockito:mockito-core:5.14.2'
    testImplementation 'org.mockito:mockito-junit-jupiter:5.14.2'
}

test {
    useJUnitPlatform()
}

2. Example Class to Test

Suppose you have a service that depends on a repository.

public class User {
    private final Long id;
    private final String name;

    public User(Long id, String name) {
        this.id = id;
        this.name = name;
    }

    public Long getId() {
        return id;
    }

    public String getName() {
        return name;
    }
}
import java.util.Optional;

public interface UserRepository {
    Optional<User> findById(Long id);
}
public class UserService {
    private final UserRepository userRepository;

    public UserService(UserRepository userRepository) {
        this.userRepository = userRepository;
    }

    public String getUserName(Long id) {
        return userRepository.findById(id)
                .map(User::getName)
                .orElse("Unknown User");
    }
}

3. Write a Mockito Test with JUnit 5

Use @ExtendWith(MockitoExtension.class) to enable Mockito support.

import org.junit.jupiter.api.Test;
import org.junit.jupiter.api.extension.ExtendWith;
import org.mockito.InjectMocks;
import org.mockito.Mock;
import org.mockito.junit.jupiter.MockitoExtension;

import java.util.Optional;

import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.mockito.Mockito.verify;
import static org.mockito.Mockito.when;

@ExtendWith(MockitoExtension.class)
class UserServiceTest {

    @Mock
    private UserRepository userRepository;

    @InjectMocks
    private UserService userService;

    @Test
    void shouldReturnUserNameWhenUserExists() {
        User user = new User(1L, "Alice");

        when(userRepository.findById(1L))
                .thenReturn(Optional.of(user));

        String result = userService.getUserName(1L);

        assertEquals("Alice", result);
        verify(userRepository).findById(1L);
    }

    @Test
    void shouldReturnUnknownUserWhenUserDoesNotExist() {
        when(userRepository.findById(99L))
                .thenReturn(Optional.empty());

        String result = userService.getUserName(99L);

        assertEquals("Unknown User", result);
        verify(userRepository).findById(99L);
    }
}

4. What the Mockito Annotations Mean

Annotation Purpose
@Mock Creates a mock object
@InjectMocks Creates the class under test and injects mocks into it
@ExtendWith(MockitoExtension.class) Enables Mockito support in JUnit 5

5. Common Mockito Methods

when(...).thenReturn(...)

Used to define mock behavior.

when(userRepository.findById(1L))
        .thenReturn(Optional.of(new User(1L, "Alice")));

verify(...)

Used to check whether a method was called.

verify(userRepository).findById(1L);

verify(..., times(...))

Used to check how many times a method was called.

verify(userRepository, times(1)).findById(1L);

You need this static import:

import static org.mockito.Mockito.times;

6. Mockito Without Annotations

You can also create mocks manually.

import org.junit.jupiter.api.Test;

import java.util.Optional;

import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.mockito.Mockito.mock;
import static org.mockito.Mockito.when;

class UserServiceManualMockTest {

    @Test
    void shouldReturnUserName() {
        UserRepository userRepository = mock(UserRepository.class);
        UserService userService = new UserService(userRepository);

        when(userRepository.findById(1L))
                .thenReturn(Optional.of(new User(1L, "Alice")));

        String result = userService.getUserName(1L);

        assertEquals("Alice", result);
    }
}

7. Typical Mockito Test Structure

A clean Mockito test usually follows the Arrange, Act, Assert pattern:

@Test
void shouldReturnUserName() {
    // Arrange
    User user = new User(1L, "Alice");
    when(userRepository.findById(1L)).thenReturn(Optional.of(user));

    // Act
    String result = userService.getUserName(1L);

    // Assert
    assertEquals("Alice", result);
    verify(userRepository).findById(1L);
}

Summary

To use Mockito with JUnit 5:

  1. Add mockito-core and mockito-junit-jupiter.
  2. Add @ExtendWith(MockitoExtension.class) to your test class.
  3. Use @Mock for dependencies.
  4. Use @InjectMocks for the class being tested.
  5. Use when(...).thenReturn(...) to define behavior.
  6. Use verify(...) to check interactions.

How do I write tests for Java records?

Java records are compact classes designed to hold immutable data. Because records automatically provide a constructor, accessor methods, equals(), hashCode(), and toString(), testing them is usually simpler than testing ordinary classes.

In most cases, you do not need to test Java’s generated record behavior directly. Instead, test:

  • custom validation in the compact constructor
  • custom methods you add to the record
  • behavior that depends on equality or immutability
  • serialization/deserialization if the record is used with JSON or persistence frameworks

Example Record

Suppose you have this Java record:

public record User(String username, String email, int age) {

    public User {
        if (username == null || username.isBlank()) {
            throw new IllegalArgumentException("Username must not be blank");
        }

        if (email == null || !email.contains("@")) {
            throw new IllegalArgumentException("Email must be valid");
        }

        if (age < 0) {
            throw new IllegalArgumentException("Age must not be negative");
        }
    }

    public boolean isAdult() {
        return age >= 18;
    }
}

This record has:

  • three components: username, email, and age
  • validation in the compact constructor
  • a custom method named isAdult()

Basic JUnit 5 Test Class

Here is a simple JUnit 5 test class:

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.*;

class UserTest {

    @Test
    void shouldCreateUserWithValidData() {
        User user = new User("alice", "[email protected]", 25);

        assertEquals("alice", user.username());
        assertEquals("[email protected]", user.email());
        assertEquals(25, user.age());
    }

    @Test
    void shouldReturnTrueWhenUserIsAdult() {
        User user = new User("bob", "[email protected]", 20);

        assertTrue(user.isAdult());
    }

    @Test
    void shouldReturnFalseWhenUserIsNotAdult() {
        User user = new User("charlie", "[email protected]", 15);

        assertFalse(user.isAdult());
    }

    @Test
    void shouldRejectBlankUsername() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> new User("", "[email protected]", 25)
        );

        assertEquals("Username must not be blank", exception.getMessage());
    }

    @Test
    void shouldRejectInvalidEmail() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> new User("alice", "invalid-email", 25)
        );

        assertEquals("Email must be valid", exception.getMessage());
    }

    @Test
    void shouldRejectNegativeAge() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> new User("alice", "[email protected]", -1)
        );

        assertEquals("Age must not be negative", exception.getMessage());
    }
}

Testing Generated Accessor Methods

Record accessors use the component name directly. For example, if your record is:

public record Product(String name, double price) {
}

The accessors are:

product.name();
product.price();

not:

product.getName();
product.getPrice();

A basic test looks like this:

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.assertEquals;

class ProductTest {

    @Test
    void shouldExposeRecordComponents() {
        Product product = new Product("Keyboard", 49.99);

        assertEquals("Keyboard", product.name());
        assertEquals(49.99, product.price());
    }
}

However, for plain records with no validation or custom behavior, these tests often provide little value because they only verify Java-generated code.


Testing equals() and hashCode()

Records automatically generate equals() and hashCode() based on all record components.

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.junit.jupiter.api.Assertions.assertNotEquals;

class ProductTest {

    @Test
    void shouldCompareRecordsByComponentValues() {
        Product first = new Product("Keyboard", 49.99);
        Product second = new Product("Keyboard", 49.99);
        Product third = new Product("Mouse", 19.99);

        assertEquals(first, second);
        assertEquals(first.hashCode(), second.hashCode());
        assertNotEquals(first, third);
    }
}

Again, you usually do not need this test unless your application depends heavily on equality behavior, such as using records as keys in a Map or elements in a Set.


Testing toString()

Records also generate a readable toString() method:

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.assertEquals;

class ProductTest {

    @Test
    void shouldGenerateReadableToString() {
        Product product = new Product("Keyboard", 49.99);

        assertEquals("Product[name=Keyboard, price=49.99]", product.toString());
    }
}

Be careful with this kind of test. It can be brittle because it depends on the exact string format.

Test toString() mainly when:

  • you override it
  • logs or messages depend on its output
  • the string representation is part of your expected behavior

Testing Constructor Validation

Records are commonly used with compact constructors for validation.

public record EmailAddress(String value) {

    public EmailAddress {
        if (value == null || value.isBlank()) {
            throw new IllegalArgumentException("Email must not be blank");
        }

        if (!value.contains("@")) {
            throw new IllegalArgumentException("Email must contain @");
        }
    }
}

Test both valid and invalid cases:

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.*;

class EmailAddressTest {

    @Test
    void shouldCreateEmailAddressWhenValueIsValid() {
        EmailAddress email = new EmailAddress("[email protected]");

        assertEquals("[email protected]", email.value());
    }

    @Test
    void shouldRejectNullEmail() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> new EmailAddress(null)
        );

        assertEquals("Email must not be blank", exception.getMessage());
    }

    @Test
    void shouldRejectBlankEmail() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> new EmailAddress(" ")
        );

        assertEquals("Email must not be blank", exception.getMessage());
    }

    @Test
    void shouldRejectEmailWithoutAtSign() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> new EmailAddress("invalid-email")
        );

        assertEquals("Email must contain @", exception.getMessage());
    }
}

Using Parameterized Tests for Records

Parameterized tests are useful when a record has multiple invalid input values.

import org.junit.jupiter.params.ParameterizedTest;
import org.junit.jupiter.params.provider.ValueSource;

import static org.junit.jupiter.api.Assertions.assertThrows;

class EmailAddressTest {

    @ParameterizedTest
    @ValueSource(strings = {"", " ", "invalid-email", "user.example.com"})
    void shouldRejectInvalidEmailValues(String value) {
        assertThrows(
                IllegalArgumentException.class,
                () -> new EmailAddress(value)
        );
    }
}

For more complex data, use @CsvSource:

import org.junit.jupiter.params.ParameterizedTest;
import org.junit.jupiter.params.provider.CsvSource;

import static org.junit.jupiter.api.Assertions.assertThrows;

class UserTest {

    @ParameterizedTest
    @CsvSource({
            "'', [email protected], 25",
            "' ', [email protected], 25",
            "alice, invalid-email, 25",
            "alice, [email protected], -1"
    })
    void shouldRejectInvalidUserData(String username, String email, int age) {
        assertThrows(
                IllegalArgumentException.class,
                () -> new User(username, email, age)
        );
    }
}

Testing Custom Methods in Records

If your record contains business logic, test that logic directly.

public record Money(String currency, int amount) {

    public boolean isPositive() {
        return amount > 0;
    }

    public Money add(Money other) {
        if (!currency.equals(other.currency())) {
            throw new IllegalArgumentException("Currencies must match");
        }

        return new Money(currency, amount + other.amount());
    }
}

Tests:

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.*;

class MoneyTest {

    @Test
    void shouldReturnTrueForPositiveAmount() {
        Money money = new Money("USD", 100);

        assertTrue(money.isPositive());
    }

    @Test
    void shouldAddMoneyWithSameCurrency() {
        Money first = new Money("USD", 100);
        Money second = new Money("USD", 50);

        Money result = first.add(second);

        assertEquals(new Money("USD", 150), result);
    }

    @Test
    void shouldRejectAddingDifferentCurrencies() {
        Money first = new Money("USD", 100);
        Money second = new Money("EUR", 50);

        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> first.add(second)
        );

        assertEquals("Currencies must match", exception.getMessage());
    }
}

Testing Immutability

Records are shallowly immutable. This means record components cannot be reassigned, but if a component refers to a mutable object, that object can still be changed.

Example:

import java.util.List;

public record Order(List<String> items) {
}

This record is not deeply immutable:

import java.util.ArrayList;
import java.util.List;

Order order = new Order(new ArrayList<>(List.of("Book")));
order.items().add("Pen");

To make it safer, copy the list:

import java.util.List;

public record Order(List<String> items) {

    public Order {
        items = List.copyOf(items);
    }
}

Then test it:

import org.junit.jupiter.api.Test;

import java.util.ArrayList;
import java.util.List;

import static org.junit.jupiter.api.Assertions.*;

class OrderTest {

    @Test
    void shouldDefensivelyCopyItems() {
        List<String> items = new ArrayList<>();
        items.add("Book");

        Order order = new Order(items);

        items.add("Pen");

        assertEquals(List.of("Book"), order.items());
    }

    @Test
    void shouldExposeUnmodifiableItems() {
        Order order = new Order(new ArrayList<>(List.of("Book")));

        assertThrows(
                UnsupportedOperationException.class,
                () -> order.items().add("Pen")
        );
    }
}

This is a valuable test because it verifies your own defensive-copying behavior, not just Java-generated record behavior.


What Should You Actually Test?

For Java records, focus your tests on behavior you wrote yourself.

Good things to test:

Feature Should You Test It? Why
Accessor methods Usually no Generated by Java
equals() / hashCode() Sometimes Useful if equality is important in your domain
toString() Rarely Usually generated and brittle to assert
Compact constructor validation Yes This is your logic
Custom methods Yes This is your logic
Defensive copying Yes Important for immutability
Serialization/deserialization Yes, if used Important for APIs and persistence

Recommended Testing Style

Use clear test names:

@Test
void shouldRejectNegativeAge() {
    // test body
}

Follow the Arrange-Act-Assert pattern:

@Test
void shouldCreateUserWithValidData() {
    // Arrange
    String username = "alice";
    String email = "[email protected]";
    int age = 25;

    // Act
    User user = new User(username, email, age);

    // Assert
    assertEquals(username, user.username());
    assertEquals(email, user.email());
    assertEquals(age, user.age());
}

Summary

To test Java records:

  1. Do not over-test generated code.
  2. Test constructor validation.
  3. Test custom methods.
  4. Test defensive copying for mutable components.
  5. Test JSON or persistence integration only when records are used that way.
  6. Use JUnit 5 assertions such as assertEquals(), assertTrue(), assertFalse(), and assertThrows().

A plain record like this usually needs no dedicated unit test:

public record Point(int x, int y) {
}

But a record like this should be tested:

public record Age(int value) {

    public Age {
        if (value < 0) {
            throw new IllegalArgumentException("Age must not be negative");
        }
    }

    public boolean isAdult() {
        return value >= 18;
    }
}

Because it contains behavior that belongs to your application, not just Java’s generated record features.

How do I write tests for utility classes?

Utility classes are usually tested like any other unit: test their public behavior, especially calculations, transformations, validation rules, edge cases, and error handling.

The main difference is that utility classes often have static methods, so your tests usually call the method directly instead of creating an object.


1. Test Useful Behavior, Not the Fact That It Is a Utility Class

A utility class is worth testing when it contains logic such as:

  • formatting
  • parsing
  • validation
  • calculations
  • normalization
  • mapping
  • filtering
  • sorting
  • date/time handling
  • null/empty handling
  • error handling

Example utility class:

public final class StringUtils {

    private StringUtils() {
    }

    public static String normalizeEmail(String email) {
        if (email == null || email.isBlank()) {
            throw new IllegalArgumentException("Email must not be blank");
        }

        return email.trim().toLowerCase();
    }
}

A good test focuses on the behavior:

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.junit.jupiter.api.Assertions.assertThrows;

class StringUtilsTest {

    @Test
    void normalizeEmailTrimsAndLowercasesEmail() {
        String result = StringUtils.normalizeEmail("  [email protected]  ");

        assertEquals("[email protected]", result);
    }

    @Test
    void normalizeEmailThrowsExceptionWhenEmailIsNull() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> StringUtils.normalizeEmail(null)
        );

        assertEquals("Email must not be blank", exception.getMessage());
    }

    @Test
    void normalizeEmailThrowsExceptionWhenEmailIsBlank() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> StringUtils.normalizeEmail("   ")
        );

        assertEquals("Email must not be blank", exception.getMessage());
    }
}

2. Static Utility Methods Are Fine to Test Directly

If the method is static, call it directly:

@Test
void calculatePercentageReturnsExpectedValue() {
    int result = MathUtils.percentageOf(25, 200);

    assertEquals(50, result);
}

You do not need Mockito or Spring for this kind of test.

Avoid loading the Spring context just to test a utility class:

@SpringBootTest
class MathUtilsTest {
    // Usually unnecessary for a utility class
}

Prefer a plain JUnit test:

class MathUtilsTest {
    // Fast unit tests
}

3. Cover Normal Cases, Edge Cases, and Failure Cases

For utility classes, edge cases are often the most valuable tests.

For example, for a number utility:

public final class NumberUtils {

    private NumberUtils() {
    }

    public static boolean isEven(int number) {
        return number % 2 == 0;
    }
}

Tests:

import org.junit.jupiter.params.ParameterizedTest;
import org.junit.jupiter.params.provider.ValueSource;

import static org.junit.jupiter.api.Assertions.assertFalse;
import static org.junit.jupiter.api.Assertions.assertTrue;

class NumberUtilsTest {

    @ParameterizedTest
    @ValueSource(ints = {2, 4, 0, -6})
    void isEvenReturnsTrueForEvenNumbers(int number) {
        assertTrue(NumberUtils.isEven(number));
    }

    @ParameterizedTest
    @ValueSource(ints = {1, 3, -5})
    void isEvenReturnsFalseForOddNumbers(int number) {
        assertFalse(NumberUtils.isEven(number));
    }
}

Good things to test:

  • normal valid input
  • zero
  • negative numbers
  • empty strings or collections
  • null, if allowed or explicitly rejected
  • boundary values
  • invalid input
  • rounding behavior
  • duplicate values
  • case sensitivity
  • timezone/date boundaries

4. Use Parameterized Tests for Repeated Input/Output Cases

Utility methods often have many input/output examples. Parameterized tests keep them clean.

Example:

import org.junit.jupiter.params.ParameterizedTest;
import org.junit.jupiter.params.provider.CsvSource;

import static org.junit.jupiter.api.Assertions.assertEquals;

class SlugUtilsTest {

    @ParameterizedTest
    @CsvSource({
            "'Hello World', 'hello-world'",
            "' Java  Testing ', 'java-testing'",
            "'Spring Boot', 'spring-boot'",
            "'already-clean', 'already-clean'"
    })
    void toSlugReturnsNormalizedSlug(String input, String expected) {
        String result = SlugUtils.toSlug(input);

        assertEquals(expected, result);
    }
}

This is better than writing four nearly identical tests.


5. Test Exceptions Clearly

If the utility method rejects invalid input, test that explicitly.

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.junit.jupiter.api.Assertions.assertThrows;

class DateUtilsTest {

    @Test
    void parseDateThrowsExceptionWhenDateHasInvalidFormat() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> DateUtils.parseIsoDate("07/13/2026")
        );

        assertEquals("Date must use ISO format: yyyy-MM-dd", exception.getMessage());
    }
}

You do not always need to assert the exception message, but it can be useful when the message is part of the behavior you care about.


6. Avoid Testing Trivial Wrappers Around Libraries

This is usually not valuable:

public static String upper(String value) {
    return value.toUpperCase();
}

A test for this mostly tests Java itself:

assertEquals("ABC", StringUtils.upper("abc"));

That is probably not worth it unless your method adds meaningful behavior, such as locale handling, null handling, trimming, or business-specific normalization.

More useful:

public static String normalizeCode(String value) {
    if (value == null || value.isBlank()) {
        throw new IllegalArgumentException("Code must not be blank");
    }

    return value.trim().toUpperCase(Locale.ROOT);
}

This has behavior worth testing.


7. Usually Do Not Test the Private Constructor

Many Java utility classes have a private constructor:

public final class MoneyUtils {

    private MoneyUtils() {
    }

    public static BigDecimal roundToCents(BigDecimal amount) {
        return amount.setScale(2, RoundingMode.HALF_UP);
    }
}

In most cases, do not write a test just to call the private constructor for coverage. That test gives little confidence and only exists to satisfy a coverage number.

Focus on the public methods instead.

If your team enforces 100% coverage, you may see reflection-based tests for private constructors, but they are usually low-value.


8. Keep Utility Tests Framework-Free When Possible

For a simple utility class, your test usually needs only JUnit:

import org.junit.jupiter.api.Test;

import static org.junit.jupiter.api.Assertions.assertEquals;

class MoneyUtilsTest {

    @Test
    void roundToCentsRoundsHalfUp() {
        BigDecimal result = MoneyUtils.roundToCents(new BigDecimal("10.235"));

        assertEquals(new BigDecimal("10.24"), result);
    }
}

Avoid unnecessary:

  • @SpringBootTest
  • database setup
  • mocks
  • dependency injection
  • web layer testing tools

A utility test should usually be fast and isolated.


9. Test Static Methods by Result, Not Implementation

Avoid tests that depend on how the utility method works internally.

Prefer:

@Test
void maskEmailHidesLocalPartExceptFirstCharacter() {
    String result = MaskingUtils.maskEmail("[email protected]");

    assertEquals("a****@example.com", result);
}

Avoid trying to verify internal helper calls or private methods. If you refactor the internals, the test should still pass as long as the behavior stays the same.


10. Example: Testing a Calculation Utility

Production code:

import java.math.BigDecimal;
import java.math.RoundingMode;

public final class TaxUtils {

    private TaxUtils() {
    }

    public static BigDecimal calculateTax(BigDecimal amount, BigDecimal taxRate) {
        if (amount == null) {
            throw new IllegalArgumentException("Amount must not be null");
        }

        if (taxRate == null) {
            throw new IllegalArgumentException("Tax rate must not be null");
        }

        if (amount.signum() < 0) {
            throw new IllegalArgumentException("Amount must not be negative");
        }

        return amount.multiply(taxRate)
                .setScale(2, RoundingMode.HALF_UP);
    }
}

Test code:

import org.junit.jupiter.api.Test;

import java.math.BigDecimal;

import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.junit.jupiter.api.Assertions.assertThrows;

class TaxUtilsTest {

    @Test
    void calculateTaxReturnsRoundedTaxAmount() {
        BigDecimal result = TaxUtils.calculateTax(
                new BigDecimal("19.99"),
                new BigDecimal("0.0825")
        );

        assertEquals(new BigDecimal("1.65"), result);
    }

    @Test
    void calculateTaxReturnsZeroWhenAmountIsZero() {
        BigDecimal result = TaxUtils.calculateTax(
                BigDecimal.ZERO,
                new BigDecimal("0.0825")
        );

        assertEquals(new BigDecimal("0.00"), result);
    }

    @Test
    void calculateTaxThrowsExceptionWhenAmountIsNull() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> TaxUtils.calculateTax(null, new BigDecimal("0.0825"))
        );

        assertEquals("Amount must not be null", exception.getMessage());
    }

    @Test
    void calculateTaxThrowsExceptionWhenTaxRateIsNull() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> TaxUtils.calculateTax(new BigDecimal("19.99"), null)
        );

        assertEquals("Tax rate must not be null", exception.getMessage());
    }

    @Test
    void calculateTaxThrowsExceptionWhenAmountIsNegative() {
        IllegalArgumentException exception = assertThrows(
                IllegalArgumentException.class,
                () -> TaxUtils.calculateTax(
                        new BigDecimal("-1.00"),
                        new BigDecimal("0.0825")
                )
        );

        assertEquals("Amount must not be negative", exception.getMessage());
    }
}

Quick Checklist

When testing utility classes, ask:

  • Does the method contain real logic?
  • What is the normal expected output?
  • What are the edge cases?
  • What invalid inputs should fail?
  • Should null be allowed or rejected?
  • Are there rounding, formatting, locale, or timezone concerns?
  • Can similar cases be written as a parameterized test?
  • Am I testing my behavior, not Java/Spring/a library?
  • Can this test run without Spring, a database, or mocks?

In short: test utility classes through their public methods, keep the tests small and fast, and focus on behavior that could realistically break.