Working with Databases

Using JPA

Adding a dependency on ws.ament.hammock:jpa-core will add basic JPA integration to your project.

Default EntityManager & DataSource

If you do nothing else with Hammock, out of the box a single EntityManagerFactory will be created with a default DataSource defined. You will still need to declare a dependency on your JDBC driver, as well as define the following properties:

Property Name	Default Value	Description
hammock.jpa.__default.datasource	__default	The name to use for the default EntityManager
hammock.datasource.__default.url	N/A	The JDBC URL to use for your DataSource. Either this or datasource have to be specified
hammock.datasource.__default.user	N/A	The JDBC User to use for your DataSource. It is not required, since basic connectors like H2 and HSQL don’t require it.
hammock.datasource.__default.password	N/A	The JDBC Password to use for your DataSource. It is not required, since basic connectors like H2 and HSQL don’t require it.
hammock.datasource.__default.datasource	N/A	The name of a previously defined DataSource to use. You should only use this property if you are not using the url/user/password combination.

Property Name

Default Value

Description

hammock.jpa.__default.datasource

__default

The name to use for the default EntityManager

hammock.datasource.__default.url

N/A

The JDBC URL to use for your DataSource. Either this or datasource have to be specified

hammock.datasource.__default.user

N/A

The JDBC User to use for your DataSource. It is not required, since basic connectors like H2 and HSQL don’t require it.

hammock.datasource.__default.password

N/A

The JDBC Password to use for your DataSource. It is not required, since basic connectors like H2 and HSQL don’t require it.

hammock.datasource.__default.datasource

N/A

The name of a previously defined DataSource to use. You should only use this property if you are not using the url/user/password combination.

Vendor Specific Modules

There are three vendor specific modules for JPA support:

ws.ament.hammock:jpa-hibernate (JPA 2.1)
ws.ament.hammock:jpa-eclipselink (JPA 2.2)
ws.ament.hammock:jpa-openjpa (JPA 2.0)

Each of these dependencies bring in the JPA Core module from Hammock plus the actual runtime dependencies to make these framework work. There is no production code in any of these modules, so you can just as easily declare a dependency on jpa-core as well as your preferred framework.

Defining DataSources

Hammock allows you to define DataSource instances using a few methods:

Annotate any bean with @javax.annotation.sql.DataSourceDefinition with necessary properties to connect to your database
Define a producer @Produces @Database(name) ws.ament.hammock.jpa.DataSourceDefinition
Create additional properties that reference your datasource name, e.g. hammock.datasource.foo.url will create a DataSource named foo in your application.

If you are using the producer approach, you can inject a builder that helps create the datasource, like this:

   @Produces
   @Database("test2")
   public DataSourceDefinition createDataSource(DataSourceDefinitionBuilder builder) {
      return builder.url("jdbc:h2:mem:test_mem").build();
   }

Which will create a DataSource named test2 connected to the JDBC URL jdbc:h2:mem:test_mem.

Under the Hood

Each DataSource created is actually an instance of HikariDataSource

Managing Database Migrations

If you declare a dependency on ws.ament.hammock:util-flyway your application will have support for migrating databases. As the name implies, it’s based on FlywayDB. Defining any flyway property in any of your register ConfigSource`s will cause that property to be passed to Flyway, effectively the same properties that would go in to `flyway.conf. Hammock uses one special property, flyway.execute which controls what operations to execute.