1. Overview

In this tutorial, we’ll have a look at writing tests using the framework support in Spring Boot. We’ll cover unit tests that can run in isolation as well as integration tests that will bootstrap Spring context before executing tests.

If you are new to Spring Boot, check out our intro to Spring Boot.

2. Project Setup

The application we’re going to use in this article is an API that provides some basic operations on an Employee Resource. This is a typical tiered architecture — the API call is processed from the Controller to Service to the Persistence layer.

3. Maven Dependencies

Let’s first add our testing dependencies:

<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-test</artifactId>
    <scope>test</scope>
    <version>3.1.5</version>
</dependency>
<dependency>
    <groupId>com.h2database</groupId>
    <artifactId>h2</artifactId>
    <scope>test</scope>
</dependency>

The spring-boot-starter-test is the primary dependency that contains the majority of elements required for our tests.

The H2 DB is our in-memory database. It eliminates the need for configuring and starting an actual database for test purposes.

3.1. JUnit 4

Starting with Spring Boot 2.4, JUnit 5’s vintage engine has been removed from spring-boot-starter-test. If we still want to write tests using JUnit 4, we need to add the following Maven dependency:

<dependency>
    <groupId>org.junit.vintage</groupId>
    <artifactId>junit-vintage-engine</artifactId>
    <scope>test</scope>
    <exclusions>
        <exclusion>
            <groupId>org.hamcrest</groupId>
            <artifactId>hamcrest-core</artifactId>
        </exclusion>
    </exclusions>
</dependency>

4. Integration Testing With @SpringBootTest

As the name suggests, integration tests focus on integrating different layers of the application. That also means no mocking is involved.

Ideally, we should keep the integration tests separated from the unit tests and should not run along with the unit tests. We can do this by using a different profile to only run the integration tests. A couple of reasons for doing this could be that the integration tests are time-consuming and might need an actual database to execute.

However in this article, we won’t focus on that, and we’ll instead make use of the in-memory H2 persistence storage.

The integration tests need to start up a container to execute the test cases. Hence, some additional setup is required for this — all of this is easy in Spring Boot:

@RunWith(SpringRunner.class)
@SpringBootTest(
  webEnvironment = SpringBootTest.WebEnvironment.MOCK,
  classes = Application.class)
@AutoConfigureMockMvc
@TestPropertySource(
  locations = "classpath:application-integrationtest.properties")
public class EmployeeRestControllerIntegrationTest {

    @Autowired
    private MockMvc mvc;

    @Autowired
    private EmployeeRepository repository;

    // write test cases here
}

The @SpringBootTest annotation is useful when we need to bootstrap the entire container. The annotation works by creating the ApplicationContext that will be utilized in our tests.

We can use the webEnvironment attribute of @SpringBootTest to configure our runtime environment; we’re using WebEnvironment.MOCK here so that the container will operate in a mock servlet environment.

Next, the @TestPropertySource annotation helps configure the locations of properties files specific to our tests. Note that the property file loaded with @TestPropertySource will override the existing application.properties file.

The application-integrationtest.properties contains the details to configure the persistence storage:

spring.datasource.url = jdbc:h2:mem:test
spring.jpa.properties.hibernate.dialect = org.hibernate.dialect.H2Dialect

If we want to run our integration tests against MySQL, we can change the above values in the properties file.

The test cases for the integration tests might look similar to the Controller layer unit tests:

@Test
public void givenEmployees_whenGetEmployees_thenStatus200()
  throws Exception {

    createTestEmployee("bob");

    mvc.perform(get("/api/employees")
      .contentType(MediaType.APPLICATION_JSON))
      .andExpect(status().isOk())
      .andExpect(content()
      .contentTypeCompatibleWith(MediaType.APPLICATION_JSON))
      .andExpect(jsonPath("$[0].name", is("bob")));
}

The difference from the Controller layer unit tests is that here nothing is mocked and end-to-end scenarios will be executed.

5. Test Configuration With @TestConfiguration

As we’ve seen in the previous section, a test annotated with @SpringBootTest will bootstrap the full application context, which means we can @Autowire any bean that’s picked up by component scanning into our test:

@RunWith(SpringRunner.class)
@SpringBootTest
public class EmployeeServiceImplIntegrationTest {

    @Autowired
    private EmployeeService employeeService;

    // class code ...
}

However, we might want to avoid bootstrapping the real application context but use a special test configuration. We can achieve this with the @TestConfiguration annotation. There are two ways of using the annotation. Either on a static inner class in the same test class where we want to @Autowire the bean:

@RunWith(SpringRunner.class)
public class EmployeeServiceImplIntegrationTest {

    @TestConfiguration
    static class EmployeeServiceImplTestContextConfiguration {
        @Bean
        public EmployeeService employeeService() {
            return new EmployeeService() {
                // implement methods
            };
        }
    }

    @Autowired
    private EmployeeService employeeService;
}

Alternatively, we can create a separate test configuration class:

@TestConfiguration
public class EmployeeServiceImplTestContextConfiguration {
    
    @Bean
    public EmployeeService employeeService() {
        return new EmployeeService() { 
            // implement methods 
        };
    }
}

Configuration classes annotated with @TestConfiguration are excluded from component scanning, therefore we need to import it explicitly in every test where we want to @Autowire it. We can do that with the @Import annotation:

@RunWith(SpringRunner.class)
@Import(EmployeeServiceImplTestContextConfiguration.class)
public class EmployeeServiceImplIntegrationTest {

    @Autowired
    private EmployeeService employeeService;

    // remaining class code
}

6. Mocking With @MockBean

Our Service layer code is dependent on our Repository:

@Service
public class EmployeeServiceImpl implements EmployeeService {

    @Autowired
    private EmployeeRepository employeeRepository;

    @Override
    public Employee getEmployeeByName(String name) {
        return employeeRepository.findByName(name);
    }
}

However, to test the Service layer, we don’t need to know or care about how the persistence layer is implemented. Ideally, we should be able to write and test our Service layer code without wiring in our full persistence layer.

To achieve this, we can use the mocking support provided by Spring Boot Test.

Let’s have a look at the test class skeleton first:

@RunWith(SpringRunner.class)
public class EmployeeServiceImplIntegrationTest {

    @TestConfiguration
    static class EmployeeServiceImplTestContextConfiguration {
 
        @Bean
        public EmployeeService employeeService() {
            return new EmployeeServiceImpl();
        }
    }

    @Autowired
    private EmployeeService employeeService;

    @MockBean
    private EmployeeRepository employeeRepository;

    // write test cases here
}

To check the Service class, we need to have an instance of the Service class created and available as a @Bean so that we can @Autowire it in our test class. We can achieve this configuration using the @TestConfiguration annotation.

Another interesting thing here is the use of @MockBean. It creates a Mock for the EmployeeRepository, which can be used to bypass the call to the actual EmployeeRepository:

@Before
public void setUp() {
    Employee alex = new Employee("alex");

    Mockito.when(employeeRepository.findByName(alex.getName()))
      .thenReturn(alex);
}

Since the setup is done, the test case will be simpler:

@Test
public void whenValidName_thenEmployeeShouldBeFound() {
    String name = "alex";
    Employee found = employeeService.getEmployeeByName(name);
 
     assertThat(found.getName())
      .isEqualTo(name);
 }

7. Integration Testing With @DataJpaTest

We’re going to work with an entity named Employee, which has an id and a name as its properties:

@Entity
@Table(name = "person")
public class Employee {

    @Id
    @GeneratedValue(strategy = GenerationType.AUTO)
    private Long id;

    @Size(min = 3, max = 20)
    private String name;

    // standard getters and setters, constructors
}

And here’s our repository using Spring Data JPA:

@Repository
public interface EmployeeRepository extends JpaRepository<Employee, Long> {

    public Employee findByName(String name);

}

That’s it for the persistence layer code. Now let’s head toward writing our test class.

First, let’s create the skeleton of our test class:

@RunWith(SpringRunner.class)
@DataJpaTest
public class EmployeeRepositoryIntegrationTest {

    @Autowired
    private TestEntityManager entityManager;

    @Autowired
    private EmployeeRepository employeeRepository;

    // write test cases here

}

@RunWith(SpringRunner.class) provides a bridge between Spring Boot test features and JUnit. Whenever we are using any Spring Boot testing features in our JUnit tests, this annotation will be required.

@DataJpaTest provides some standard setup needed for testing the persistence layer:

  • configuring H2, an in-memory database
  • setting Hibernate, Spring Data, and the DataSource
  • performing an @EntityScan
  • turning on SQL logging

To carry out DB operations, we need some records already in our database. To setup this data, we can use TestEntityManager.

The Spring Boot TestEntityManager is an alternative to the standard JPA EntityManager that provides methods commonly used when writing tests.

EmployeeRepository is the component that we are going to test.

Now let’s write our first test case:

@Test
public void whenFindByName_thenReturnEmployee() {
    // given
    Employee alex = new Employee("alex");
    entityManager.persist(alex);
    entityManager.flush();

    // when
    Employee found = employeeRepository.findByName(alex.getName());

    // then
    assertThat(found.getName())
      .isEqualTo(alex.getName());
}

In the above test, we’re using the TestEntityManager to insert an Employee in the DB and reading it via the find by name API.

The assertThat(…) part comes from the Assertj library, which comes bundled with Spring Boot.

8. Unit Testing With @WebMvcTest

Our Controller depends on the Service layer; let’s only include a single method for simplicity:

@RestController
@RequestMapping("/api")
public class EmployeeRestController {

    @Autowired
    private EmployeeService employeeService;

    @GetMapping("/employees")
    public List<Employee> getAllEmployees() {
        return employeeService.getAllEmployees();
    }
}

Since we’re only focused on the Controller code, it’s natural to mock the Service layer code for our unit tests:

@RunWith(SpringRunner.class)
@WebMvcTest(EmployeeRestController.class)
public class EmployeeRestControllerIntegrationTest {

    @Autowired
    private MockMvc mvc;

    @MockBean
    private EmployeeService service;

    // write test cases here
}

To test the Controllers, we can use @WebMvcTest. It will auto-configure the Spring MVC infrastructure for our unit tests.

In most cases, @WebMvcTest will be limited to bootstrap a single controller. We can also use it along with @MockBean to provide mock implementations for any required dependencies.

@WebMvcTest also auto-configures MockMvc, which offers a powerful way of easy testing MVC controllers without starting a full HTTP server.

Having said that, let’s write our test case:

@Test
public void givenEmployees_whenGetEmployees_thenReturnJsonArray()
  throws Exception {
    
    Employee alex = new Employee("alex");

    List<Employee> allEmployees = Arrays.asList(alex);

    given(service.getAllEmployees()).willReturn(allEmployees);

    mvc.perform(get("/api/employees")
      .contentType(MediaType.APPLICATION_JSON))
      .andExpect(status().isOk())
      .andExpect(jsonPath("$", hasSize(1)))
      .andExpect(jsonPath("$[0].name", is(alex.getName())));
}

The get(…) method call can be replaced by other methods corresponding to HTTP verbs like put(), post(), etc. Please note that we are also setting the content type in the request.

MockMvc is flexible, and we can create any request using it.

9. Auto-Configured Tests

One of the amazing features of Spring Boot’s auto-configured annotations is that it helps to load parts of the complete application and test-specific layers of the codebase.

In addition to the above-mentioned annotations, here’s a list of a few widely used annotations:

  • @WebF**luxTest: We can use the @WebFluxTest annotation to test Spring WebFlux controllers. It’s often used along with @MockBean to provide mock implementations for required dependencies.

  • @JdbcTest: We can use the @JdbcTest annotation to test JPA applications, but it’s for tests that only require a DataSource. The annotation configures an in-memory embedded database and a JdbcTemplate.

  • @JooqTest: To test jOOQ-related tests, we can use @JooqTest annotation, which configures a DSLContext.

  • @DataMongoTest: To test MongoDB applications, @DataMongoTest is a useful annotation. By default, it configures an in-memory embedded MongoDB if the driver is available through dependencies, configures a MongoTemplate, scans for @Document classes, and configures Spring Data MongoDB repositories.

  • @DataRedisTestmakes it easier to test Redis applications. It scans for @RedisHash classes and configures Spring Data Redis repositories by default.

  • @DataLdapTest configures an in-memory embedded LDAP (if available), configures a LdapTemplate, scans for @Entry classes, and configures Spring Data LDAP repositories by default.

  • @RestClientTest: We generally use the @RestClientTest annotation to test REST clients. It auto-configures different dependencies such as Jackson, GSON, and Jsonb support; configures a RestTemplateBuilder; and adds support for MockRestServiceServer by default.

  • @JsonTest: Initializes the Spring application context only with those beans needed to test JSON serialization.

You can read more about these annotations and how to further optimize integrations tests in our article Optimizing Spring Integration Tests.

10. Conclusion

In this article, we took a deep dive into the testing support in Spring Boot and showed how to write unit tests efficiently.

The complete source code of this article can be found over on GitHub. The source code contains many more examples and various test cases.

And if you want to keep learning about testing, we have separate articles related to integration tests, optimizing Spring integration tests, and unit tests in JUnit 5.


« 上一篇: Java Stripe API教程
» 下一篇: JAX-WS入门介绍