Skip to content
Integration tests made easy
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
docs
gradle/wrapper
testing
.gitignore
.travis.yml
build.gradle
gradlew
gradlew.bat
readme.adoc
settings.gradle

readme.adoc

Testegration: Integration Testing for Java

Maven Central Build Status docs

Powerful & flexible testing lifecycle + opinionated defaults.
Reference documentation

Integration tests made easy

No more:

  • manual DB installation, setup or cleanup for local development

  • single testing db shared between many devs

  • separate db migrations and hibernate configuration for production and in-memory db

  • untested stored procedures and committed transactions (due to always-rollback policy)

  • problems with testing other kinds of storage (NoSQL, LDAP, search engine, queue, distributed FS, you name it)

  • command line only tests that are painful to debug

How would you test it?

Let’s have a look at a typical repository that uses a little bit more complex sql than all hello-world examples. It’s a typical spring-boot application with postgres and flyway.

Application.java
@Repository class Repo {
    @Autowired JdbcTemplate jdbcTemplate;

    /** fails at first batch containing null */
    public void save_ints_in_batches_by_two(List<Integer> ints) {
        jdbcTemplate.batchUpdate("insert into some_table values (?)", ints, 2,
                                                            (ps, value) -> ps.setInt(1, value));
    }

    public int count() {return jdbcTemplate.queryForObject("select my_count()", Integer.class);}
}

@SpringBootApplication public class Application {
	public static void main(String[] args) {SpringApplication.run(Application.class, args);}
}
V1__first_table.sql
create table some_table (field integer not null);
V2__function_and_data.sql
-- this function just counts rows of some_table
CREATE FUNCTION my_count() RETURNS bigint AS $$ select count(1) from some_table; $$ LANGUAGE sql;

-- we often need some starting data. e.g. dictionaries
insert into some_table values (2);
  • You can’t test it using in-memory db - sql syntax won’t match. You need some real db to be started and prepared

  • You can’t test it using auto rollback as this code uses commits.

How would an ideal test look like?

Maybe we would like to write something like this:

@SpringBootTest @RunWith(SpringRunner.class)
public class RepoTest {
    @Autowired Repo repo;

    @Test public void should_insert_all_values() {
        repo.save_ints_in_batches_by_two(Arrays.asList(1,2,3,4,5));

        int count = repo.count();
        assertThat(count).isEqualTo(5);
    }

    @Test public void should_fail_on_batch_containing_null() {
        assertThatCode(()->
                    repo.save_ints_in_batches_by_two(Arrays.asList(1,2,3,null,5))
            ).hasCauseInstanceOf(Exception.class);

        int count = repo.count();
        assertThat(count).isEqualTo(2);
    }
}

Buuut what about installing and starting postgres? What about clean-up between tests? What about connection parameters? What about…​

Yep, you’ll need to solve all those problems. Using Testegration you’ll do it in 20 lines of code.

Let’s get started

gradle import
testCompile 'net.piotrturski.testegration:postgres:0.0.2-beta'
test base-class
import net.piotrturski.testegration.postgres.Postgres;
...
Postgres.docker("my-project", "9.6.1").run();

This line will run postgres docker image in version 9.6.1 (or connect to your local postgres instance if you have one) and prepare test schema (or other if you override the default).

One more thing is missing. Testegration tries to be framework agnostic so we have to connect it to spring-boot and JUnit. In this case we’ll do it by setting connection parameters as system properties and preparing db before each test and before building spring context:

CommonDbTest.java
public class CommonDbTest {

    @BeforeClass public static void beforeClass() { prepareDb(); }
    @Before public void before() { prepareDb(); }

    static void prepareDb() {
        Postgres.docker("my-project", "9.6.1")
                .with(exposeDbPropertiesForSpringBoot)
                .run();
    }

    static PartialDefinition<PostgresConf> exposeDbPropertiesForSpringBoot = def ->
                def.postConfigure(runtime ->
                        ImmutableMap.of(
                                "spring.datasource.url", runtime.getConfig().getJdbcUrl(),
                                "spring.datasource.username", runtime.getConfig().user,
                                "spring.datasource.password", runtime.getConfig().passwd)
                                .forEach(System::setProperty));
}

Make your db tests extend CommonDbTest and that’s it. Now you can run your test from IDE, console or CI server.

To fully understand what’s going on, check documentation. You will find out that similar strategy applies to other testing frameworks, non-spring environments or non-db storage types.

Free for small projects

It’s free for up to 20 calls per build. Even for commercial use. So you can:

  • quickly start a project that requires storage tests

  • easily evaluate if Testegration is a good fit for your existing project of any size

  • use it for your studies or pet projects

You can’t perform that action at this time.