Skip to main content

Notice: This Wiki is now read only and edits are no longer possible. Please see: for the plan.

Jump to: navigation, search

Introduction to Java Persistence API (ELUG)

Elug draft icon.png For the latest EclipseLink documentation, please see

This section introduces concepts of Java Persistence API and provides general information on it.

What Is Java Persistence API?

The Java Persistence API (JPA) is a lightweight framework for Java persistence (see Persisting Objects) based on Plain Old Java Object (POJO). JPA is a part of EJB 3.0 and 3.1 specifications. JPA provides an object relational mapping approach that lets you declaratively define how to map Java objects to relational database tables in a standard, portable way. You can use this API to create, remove and query across lightweight Java objects within both an EJB 3.0/3.1-compliant container and a standard Java SE 5/6 environment.

For more information, see the following:

What Do You Need to Develop with JPA

To start developing with JPA, you need the following:

Relational Database

To develop your applications with JPA, you can use any relational database.

Domain Model Classes

Your domain model should consist of classes representing entities–lightweight persistent domain objects. The easiest way to define an entity class is by using the @Entity annotation (see Using Metadata Annotations), as the following example shows:

 public class Employee implements Serializable {
    private static final long serialVersionUID = -8865917134914502125L;

For more information on entities, see the following:

persistence.xml File

Use the persistence.xml file to package your entities.

For more information and examples, see the following:

Object Relational Mapping Metadata

Object relational mapping metadata specifies the mapping of your domain model classes to the database.

You can express this metadata in the form of annotations and/or XML.

Metadata Annotations and ORM.xml File

A metadata annotation represents a Java language feature that lets you attach structured and typed metadata to the source code. Annotations alone are sufficient for the metadata specification–you do not need to use XML. Annotations for object relational mapping are in the javax.persistence package. For more information and examples, see Chapter 8 "Metadata Annotations" of the JPA Specification

An object relational mapping XML file is optional. If you choose to provide one, then it should contain mapping information for the classes listed in it. The persistence provider loads an orm.xml file (or other mapping file) as a resource. If you provide a mapping file, the classes and mapping information specified in the mapping file will be used. For more information and examples, see the following:

Overriding Annotations with XML

XML mapping metadata may combine with and override annotation metadata. For more information and examples, see the following:

Advantages and Disadvantages of Using Annotations

Metadata annotations are relatively simple to use and understand. They provide in-line metadata located with the code that this metadata is describing–you do not need to replicate the source code context of where the metadata applies.

On the other hand, annotations unnecessarily couple the metadata to the code. Thus, changes to metadata require changing the source code.

Advantages and Disadvantages of Using XML

The following are advantages of using XML:

  • no coupling between the metadata and the source code;
  • compliance with the existing, pre-EJB 3.0 development process;
  • support in IDEs and source control systems;

The main disadvantages of mapping with XML are the complexity and the need for replication of the code context.

Persistence Provider

The persistence provider supplies the implementation of the JPA specification.

The persistence provider handles the object relational mapping of the relationships, including their loading and storing to the database (as specified in the metadata of the entity class), and the referential integrity of the relationships (as specified in the database).

For example, the EclipseLink persistence provider ensures that relational descriptors are created for annotated objects, as well as mappings are created based on annotations.

Persistence Application Code

To manage entities (see Domain Model Classes) in your persistence application, you need to obtain an entity manager from an EntityManagerFactory. How you get the entity manager and its factory largely depends on the Java environment in which you are developing your application.

Container-Managed Entity Manager

In the Java EE environment, you acquire an entity manager by injecting it using the @PersistenceContext annotation (dependency injection), as the Obtaining an Entity Manager Through Dependency Injection example shows, or using a direct lookup of the entity manager in the JNDI namespace, as the Performing JNDI Lookup of an Entity Manager example shows.

Example 17-1

Obtaining an Entity Manager Through Dependency Injection

 public EntityManager em;

Note: You can only use the @PersistenceContext annotation injection on session beans, servlets and JSP.

Example 17-2

Performing JNDI Lookup of an Entity Manager

 @PersistenceContext(name="ProjectEM", unitName="Project")
 public class ProjectSessionBean implements Project {
     SessionContext ctx;
     public void makeCurrent() {
         try {
            EntityManager em = (EntityManager)ctx.lookup("ProjectEM");

See the following EclipseLink post for concurrency issues when using an EntityManager directly on a servlet outside of a function local variable instead of a Statelss session bean.

The container would manage the life cycle of this entity manager–your application does not have to create it or close it.

For more information and examples, see the following sections of the JPA Specification:

  • Section 3.1 "EntityManager"
  • Section 5.2.1 "Obtaining an Entity Manager in the Java EE Environment"
  • Section 5.3.1 "Obtaining an Entity Manager Factory in a Java EE Container"

Application-Managed Entity Manager

In the Java SE environment, not the container but the application manages the life cycle of an entity manager. You would create this entity manager using the EntityManagerFactory's method createEntityManager. You have to use the javax.persistence.Persistence class to bootstrap an EntityManagerFactory instance, as this example shows:

Application-Managed Entity Manager in the Java SE Environment

 public class Employee {
     public static void main(String[] args) {
         EntityManagerFactory emf =
         EntityManager em = emf.createEntityManager();

Notice that you need to explicitly close the entity manager and the factory.

In the Java EE environment, you can use the application-managed entity managers as well. You would create it using the @PersistenceUnit annotation to declare a reference to the EntityManagerFactory for a persistence unit, as the following example shows:

 EntityManagerFactory emf;

Note: You can only use the @PersistenceContext annotation injection on session beans, servlets and JSP.

For more information and examples, see the following sections of the JPA Specification:

  • Section 5.2.2 "Obtaining an Application-managed Entity Manager"
  • Section 5.3.2 "Obtaining an Entity Manager Factory in a Java SE Environment"

Transaction Management

Transactions define when new, changed or removed entities are synchronized to the database.

JPA supports the following two types of transaction management:

Container-managed entity managers always use JTA transactions. Application-managed entity managers may use JTA or resource-local transactions. The default transaction type for Java EE application is JTA.

You define the transaction type for a persistence unit and configure it using the persistence.xml file (see persistence.xml File).

For more information, see Section 5.5 "Controlling Transactions" of the JPA Specification.

JTA Transaction Management

JTA transactions are the transactions of the Java EE server.

As section 5.5.1 "JTA Entity Managers" of the JPA Specification defines, "An entity manager whose transactions are controlled through JTA is a JTA entity manager. A JTA entity manager participates in the current JTA transaction, which is begun and committed external to the entity manager and propagated to the underlying resource manager."

Resource-Local Transactions

Resource-local transactions are the native transactions of the JDBC drivers that are referenced by a persistence unit. Your application explicitly controls these transactions. Your application interacts with the resource-local transactions by acquiring an implementation of the EntityTransaction interface from the entity manager.

For more information and examples, see the following sections of the JPA Specification.

  • Section 5.5.2 "Resource-local Entity Managers"
  • Section "The EntityTransaction Interface"

Copyright Statement

Back to the top