JDBI Dropwizard Integration Testing
Java
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
src
.gitignore
.travis.yml
LICENSE
README.md
pom.xml
setup_mysql_database.sql
setup_postgresql_database.sql

README.md

JDIT

Build Status Maven Central

Overview

JDIT is a library for simplifying of integration testing of JDBI3 data access objects

What it does

The library does the following things:

  • Supports HSQLDB, PostgreSQL, MySQL, H2 databases;
  • Optionally migrates a user-defined sql schema;
  • The database and the DBI instance are shared between the tests, so they are performed quickly;
  • Provides a JUnit runner for running DBI-related tests;
  • Supports injecting a DBI DAO or a SQL object to the current test by annotating a tested instance;
  • Supports injecting a DBI instance or an Handle instance to the current test for performing SQL requests against the database;
  • Sweeps data from the database and reset sequences after every test. All data changes performed in tests are discarded (but the schema remains), so the database is in the clean state before every test. It affords to tests to be independent and don't impact to each other.
  • Supports executing of an arbitrary SQL script before every test (or set of tests) by specifying an annotation on a test method or test class.

Getting started

Define a simple SQL Object to test

public interface PlayerDao {

    @GetGeneratedKeys
    @SqlUpdate("insert into players(first_name, last_name, birth_date, weight, height)"
            + " values (:first_name, :last_name, :birth_date, :weight, :height)")
    Long createPlayer(@Bind("first_name") String firstName,
                      @Bind("last_name") String lastName,
                      @Bind("birth_date") Date birthDate,
                      @Bind("height") int height, @Bind("weight") int weight);

    @SqlQuery("select last_name from players order by last_name")
    List<String> getPlayerLastNames();

    @SqlQuery("select count(*) from players where year(birth_date) = :year")
    int getAmountPlayersBornInYear(@Bind("year") int year);

    @SqlQuery("select * from players where first_name=:first_name and " +
              "last_name=:last_name")	      
    Optional<Player> findPlayer(@Bind("first_name") String firstName,
                                @Bind("last_name") String lastName);
}

Add Maven dependency

<dependency>
    <groupId>com.github.arteam</groupId>
    <artifactId>jdit</artifactId>
    <version>0.8</version>
    <scope>test</scope>
</dependency>

Setup a database driver

If your test database is different from the production one, add a test database driver via Maven:

<dependency>
	<groupId>org.hsqldb</groupId>
	<artifactId>hsqldb</artifactId>
	<version>2.3.3</version>
	<scope>test</scope>
</dependency>

Create a test resources directory

You need to create a test resource directory to host resources. By default it's 'src/test/resources'.

If you want to use another directory, don't forget to set it in Maven as a test resources directory in the build section:

<build>
    <testResources>
        <testResource>
           <directory>src/custom_test_resources</directory>
        </testResource>
    </testResources>
</build>

Define a database schema

Add a file with a database schema to your test resources directory. By default it should have name schema.sql.

create table players(
    id  identity,
    first_name varchar(128) not null,
    last_name varchar(128) not null,
    birth_date date not null,
    weight int not null,
    height int not null
);

Write a test

@RunWith(DBIRunner.class)
public class PlayerDaoTest {

    @TestedSqlObject
    PlayerDao playerDao;

    @DBIHandle
    Handle handle;

    @Test
    public void testCreatePlayer() {
        Long playerId = playerDao.createPlayer("Vladimir", "Tarasenko", date("1991-12-13"), 184, 90);
        List<Map<String,Object>> rows = handle.select("select * from players where id=?", playerId);
        assertFalse(rows.isEmpty());

        Map<String, Object> row = rows.get(0);
        assertEquals(0, row.get("id"));
        assertEquals("Vladimir", row.get("first_name"));
        assertEquals("Tarasenko", row.get("last_name"));
        assertEquals(date("1991-12-13"), row.get("birth_date"));
        assertEquals(184, row.get("height"));
        assertEquals(90, row.get("weight"));
    }

    private static Date date(String textDate) {
        return ISODateTimeFormat.date().parseDateTime(textDate).toDate();
    }
}

You should see something like this in output:

23:57:30.091 [main] INFO  DBI - Handle [org.skife.jdbi.v2.BasicHandle@18cc8e9] obtained in 783 millis
23:57:30.157 [main] INFO  DBI - batch:[[create table players(     id  identity,     first_name varchar(128) not null,     last_name varchar(128) not null,     birth_date date not null,     weight int not null,     height int not null )]] took 3 millis
23:57:30.158 [main] INFO  DBI - Handle [org.skife.jdbi.v2.BasicHandle@18cc8e9] released
23:57:30.159 [main] INFO  DBI - Handle [org.skife.jdbi.v2.BasicHandle@3dacfa] obtained in 0 millis
23:57:30.639 [main] INFO  DBI - statement:[/* PlayerDao.createPlayer */ insert into players(first_name, last_name, birth_date, weight, height) values (?, ?, ?, ?, ?)] took 0 millis
23:57:30.664 [main] INFO  DBI - statement:[select * from players where id=?] took 0 millis
23:57:30.676 [main] INFO  DBI - statement:[TRUNCATE SCHEMA public RESTART IDENTITY AND COMMIT] took 0 millis
23:57:30.679 [main] INFO  DBI - Handle [org.skife.jdbi.v2.BasicHandle@3dacfa] released

Things to notice:

  • Annotation @RunWith is crucial. It makes the test aware of a DBI context. Without it nothing will work.
  • Annotation @TestedSqlObject is used for marking a tested SQL object.
  • Annotation @DBIHandle is used for obtaining a reference to a handle to the active database for performing queries.
  • During a first invocation the schema has been migrated to the database. It happens only once for all tests.
  • As you see from the logs, data was swept from the database after the completion of the test. But the schema wasn't removed.

Load data before a test

Write a SQL DML script that populates the DB with needed data for testing.

Give it a name, say, playerDao/players.sql and place it into the test resources directory.

insert into players(first_name, last_name, birth_date, weight, height)
values ('Vladimir','Tarasenko', '1991-12-13', 99, 184);
insert into players(first_name, last_name, birth_date, weight, height)
values ('Tyler','Seguin', '1992-01-30', 88, 185);
insert into players(first_name, last_name, birth_date, weight, height)
values ('Ryan','Ellis', '1991-01-03', 79, 176);
insert into players(first_name, last_name, birth_date, weight, height)
values ('John','Tavares', '1990-09-20', 93, 185);

Load this script before the test execution.

@DataSet("playerDao/players.sql")
@Test
public void testGetPlayerListNames(){
    List<String> playerLastNames = playerDao.getPlayerLastNames();
    assertEquals(playerLastNames, ImmutableList.of("Ellis", "Seguin", "Tarasenko", "Tavares"));
}

Annotation @DataSet is used for marking a script that should be loaded before a test.

If you find that you reuse the same data set for different tests, consider to place this annotation on a class level.

@RunWith(DBIRunner.class)
@DataSet("playerDao/players.sql")
public class PlayerDaoTest {

In this mode a script will be loaded for every method in the test. Nevertheless, this script can be overridden by a method level annotation.

Configuration

JDIT reads a configuration file of the following format:

db.url=jdbc:hsqldb:mem:jdbi-testing
db.username=sa
db.password=

schema.migration.enabled=true
schema.migration.location=schema.sql

dbi.factory=com.github.arteam.jdit.StandardDBIFactory
  • db.url - Database URL;
  • db.username - Database username;
  • db.password - Database password;
  • schema.migration.enabled - Whether schema migration is enabled;
  • schema.migration.location - Location of the database schema in resources; If it's a directory, then all .sql files in the directory are processed as the schema;
  • dbi.factory - Implementation of a factory for creating DBI instances.

If you need to override this configuration, you should place the jdit.properties file in your test resources directory with needed changes.

For example, for overriding the schema location you should create a file with following content:

schema.migration.location=db/migration

If you need to specify properties for a specific test you can do it with the @JditProperties annotation on the the test class level.

Examples

More examples available in a separate repository.

Dependencies

Compatability

Version 0.1 is compatible with Dropwizard 0.7.1 and 0.8.0

Version 0.2 is compatible with Dropwizard 0.8.1

Version 0.3 is compatible with Dropwizard 0.9.*

Version 0.4.* is compatible with Dropwizard 0.9.*

Version 0.5.* is compatible with Dropwizard 1.0.*

Version 0.6.* is compatible with Dropwizard 1.1.*

Version 0.7 and higher are not depended on Dropwizard

Availability

The lastest version is available in the Maven Central repository. Old artifacts are available in the [JCenter] (https://bintray.com/arteam/maven/jdit) repository

<repositories>
        <repository>
            <id>jcenter</id>
            <name>bintray</name>
            <url>http://jcenter.bintray.com</url>
        </repository>
</repositories>