XAP

JPA API

The Java Persistency API (JPA) is a Java-programming language framework that manages relational data in applications using the Java platform. GigaSpaces's JPA API allows you to use the JPA's functionality and annotations, and to execute JPQL queries on a SpaceClosed Where GigaSpaces data is stored. It is the logical cache that holds data objects in memory and might also hold them in layered in tiering. Data is hosted from multiple SoRs, consolidated as a unified data model.. GigaSpaces's JPA implementation is based on OpenJPA.

We recommend that you familiarize yourself with JPA before reading this section. We also suggest that you review the GigaSpaces PetClinic JPA Tutorial, which describes how a standard JPA application (the Spring PetClinic) can be adapted to GigaSpaces JPA and deployed to the GigaSpaces runtime environment.

GigaSpaces JPA Configuration

OpenJPA

OpenJPA's JAR file is included with the GigaSpaces distribution (provided under $GS_HOME/lib/optional/jpa), and the GigaSpaces-specific JPA implementation classes are part of the OpenSpaces JAR (located under $GS_HOME/lib/required/xap-openspaces.jar). Maven users should define the following dependency in their pom.xml file:

<dependencies>
  <dependency>
    <groupId>org.apache.openjpa</groupId>
    <artifactId>openjpa</artifactId>
    <version>3.2.2</version>
  </dependency>
</dependencies>

The persistence.xml File

To enable the GigaSpaces JPA implementation, you should specify the following three mandatory properties in your persistence.xml file:

  • BrokerFactory should be set to abstractstore, which tells OpenJPA that an alternate StoreManager (the layer responsible for interacting with the underlying database) is going to be used.
  • abstractstore.AbstractStoreManager should be set to org.openspaces.jpa.StoreManager, which tells OpenJPA to use the OpenSpaces StoreManager.
  • LockManager should be set to none because OpenJPA's default lock manager is set to version (Optimistic locking), which is currently unsupported.

Your persistence.xml file should be placed in any **/META-INF folder in your classpath.

BrokerFactory Property

There is no need to set the abstractstore.AbstractStoreManager property. Instead, make sure to set the BrokerFactory property to org.openspaces.jpa.BrokerFactory, as shown in the example below.

The following is an example of a GigaSpaces JPA persistence.xml configuration file:

<persistence-unit name="gigaspaces" transaction-type="RESOURCE_LOCAL">
    <provider>org.apache.openjpa.persistence.PersistenceProviderImpl</provider>
    <properties>
            <property name="BrokerFactory" value="abstractstore"/>
            <property name="abstractstore.AbstractStoreManager" value="org.openspaces.jpa.StoreManager"/>
            <property name="LockManager" value="none"/>
    </properties>
</persistence-unit>
<persistence-unit name="gigaspaces" transaction-type="RESOURCE_LOCAL">
    <provider>org.apache.openjpa.persistence.PersistenceProviderImpl</provider>
    <properties>
            <property name="BrokerFactory" value="org.openspaces.jpa.BrokerFactory"/>
            <property name="LockManager" value="none"/>
    </properties>
</persistence-unit>

Transaction Read Lock Level

The default GigaSpaces JPA read lock level is set to "read", which is equivalent to GigaSpaces's ReadModifiers.REPEATABLE_READ. In order to use ReadModifiers.EXCLUSIVE_READLOCK, the ReadLockLevel property should be set to "write":

  <property name="ReadLockLevel" value="write"/>

Space Connection Injection

Specifying a Space connection URL or a Space instance can be done in one of the ways described below.

Referencing an Existing Space Instance through Factory Properties

You can specify a Space instance when creating an EntityManagerFactory as follows:

GigaSpace gigaspace = ...
Properties properties = new Properties();
properties.put("ConnectionFactory", gigaspace.getSpace());
EntityManagerFactory emf = Persistence.createEntityManagerFactory("gigaspaces", properties);

Injection using Spring

You can inject either an EntityManager or EntityManagerFactory using Spring. Before reading this, we recommend that you familiarize yourself with Spring's JPA support.

The following example shows how to inject a Space-based EntityManagerFactory. The Spring XML configuration file declares a Space, an EntityManagerFactory, a transaction manager, and a JPA service beanClosed An application component that interacts with the space (using the read, write, take and notify operations), and implements a certain functionality. (this is our DAO):

<?xml version="1.0" encoding="UTF-8"?>

<beans xmlns="http://www.springframework.org/schema/beans"
       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
       xmlns:os-core="http://www.openspaces.org/schema/core"
       xmlns:tx="http://www.springframework.org/schema/tx"
       xmlns:context="http://www.springframework.org/schema/context"
       xsi:schemaLocation="http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans.xsd
       http://www.openspaces.org/schema/core http://www.openspaces.org/schema/17.0.0/core/openspaces-core.xsd
       http://www.springframework.org/schema/context http://www.springframework.org/schema/context/spring-context.xsd
       http://www.springframework.org/schema/tx http://www.springframework.org/schema/tx/spring-tx.xsd">

    <!-- space definition -->
    <os-core:embedded-space  id="space" space-name="jpaSpace" lookup-groups="test"/>

    <!-- gigaspace definition -->
    <os-core:giga-space id="gigaSpace" space="space"/>

    <!-- JPA entity manager factory definition -->
    <bean id="entityManagerFactory" class="org.springframework.orm.jpa.LocalContainerEntityManagerFactoryBean">
                <!-- this relies on the fact that our persistence.xml file defines a persistence unit named "gigaspaces" -->
        <property name="persistenceUnitName" value="gigaspaces"/>
        <property name="jpaVendorAdapter">
            <bean class="org.openspaces.jpa.OpenSpacesJpaVendorAdapter">
                <property name="space" value="#{gigaSpace.space}"/>
            </bean>
        </property>
    </bean>

    <!-- JPA transaction manager definition -->
    <bean id="transactionManager" class="org.springframework.orm.jpa.JpaTransactionManager">
        <property name="entityManagerFactory" ref="entityManagerFactory" />
    </bean>

    <!-- support annotations -->
    <bean class="org.springframework.orm.jpa.support.PersistenceAnnotationBeanPostProcessor" />
    <context:annotation-config/>
    <tx:annotation-driven transaction-manager="transactionManager"/>

    <!-- JPA example service definition -->
    <bean id="jpaService" class="org.openspaces.jpa.JpaService" />
</beans>

In our DAO, Spring injects the EntityManager using the @PersistentContext annotation. Spring creates an EntityManager using the EntityManagerFactory defined at the beginning of every transaction.

@Repository
@Transactional
public class JpaService {
    @PersistenceContext
    private EntityManager em;

    public JpaService() {
    }

    @Transactional
    public void persistObject() {
        em.persist(...);
    }
}

For detailed information regarding the persistence.xml file, refer to the OpenJPA Manual.

Listing Your Persistent Classes

When working with persistent classes, there are several ways to make the JPA layer aware of them.

When using Spring, this is done automatically unless otherwise specified (see below). Spring will scan your classpath, looking for classes that have the @Entity annotation.

When not using Spring, there are two options:

  • Point to an orm.xml file your persistence.xml file:

      <persistence-unit name="gigaspaces" transaction-type="RESOURCE_LOCAL">
        <mapping-file>META-INF/orm.xml</mapping-file>
        <exclude-unlisted-classes/>
      </persistence-unit>
    
  • Use the <class> tag in your pesistence.xml file. For example, if you're going to use the classes Trade, Book & Author, list them in your persistence.xml file as follows:

    <persistence-unit name="gigaspaces" transaction-type="RESOURCE_LOCAL">
        <provider>org.apache.openjpa.persistence.PersistenceProviderImpl</provider>
            <class>org.openspaces.objects.Trade</class>
            <class>org.openspaces.objects.Book</class>
            <class>org.openspaces.objects.Author</class>
        <properties>
            <property name="BrokerFactory" value="abstractstore"/>
            <property name="abstractstore.AbstractStoreManager" value="org.openspaces.jpa.StoreManager"/>
            <property name="LockManager" value="none"/>
        </properties>
    </persistence-unit>
    

Enhancing Your Classes

JPA classes are monitored at runtime for automatic dirty detection. There must be bytecode enhancement in order for this to be transparent to the user. OpenJPA offers 3 options for this:

  • Enhance at build time using an Ant or a Maven script.
  • Enhance at runtime using OpenJPA's Java agent enhancer.
  • When using Spring, you can specify a LoadTimeWeaver that will enhance the classes at load time.

The first option (enhance at build time) is the best in terms of performance and suitability for the GigaSpaces runtime environment.For detailed information about how to enhance your entities, refer to the OpenJPA Entity Enhancement page.

GigaSpaces JPA Entities

An entity class must meet the following requirements:

  • The class must be annotated with the javax.persistence.Entity annotation.
  • The class must have a public no-argument constructor (the class may have other constructors).
  • GigaSpaces and JPA annotations can only be declared on Getters, NOT on fields.

Annotations

GigaSpaces JPA Entities must have both JPA and GigaSpaces annotations for the following annotations:

GigaSpaces JPA
@SpaceId @Id/@EmbeddedId
@SpaceExclude `@Transient

As with GigaSpaces POJOs, you can use the @SpaceIndex & @SpaceRouting annotations with GigaSpaces JPA entities.

Indexes should only be declared in the owning entity of a relationship. Examples are available on the JPA Relationships page.

THe following is an example of a basic JPA Entity:

@Entity
public class Trade {
  private Long id;
  private Double quantity;
  private List<Double> rates;
  private boolean state;

  // Public no-argument constructor
  public Trade() {
  }

  // Both SpaceId and Id should be declared on the id property
  @Id
  @SpaceId
  public Long getId() {
    return this.id;
  }

  // Persistent property, no additional GigaSpaces annotations needs to be used.
  public Double getQuantity() {
    return this.quantity;
  }

  // A persistent collection property. In this case we'll use a GigaSpaces annotation
  // for indexing its values.
  @ElementCollection
  @SpaceIndex(path = "[*]")
  public List<Double> getRates() {
    return this.rates;
  }

  // A transient property. In this case we'll use both GigaSpaces and JPA annotations
  @Transient
  @SpaceExclude
  public boolean getState() {
    return this.state;
  }

  /* Additional Getters & Setters... */

}

For auto-generated ID declaration and complex object ID declaration, refer to JPA Entity ID.

The following is an example of a JPA Owner entity with a one-to-many relationship:

@Entity
public class Owner {
    //
    private Integer id;
    private String name;
    private List<Pet> pets;
    //
    public Owner() {
    }
    public Owner(Integer id, String name, List<Pet&gt; pets) {
        super();
        this.id = id;
        this.name = name;
        this.pets = pets;
    }
    //
    @Id
    @SpaceId
    public Integer getId() {
        return id;
    }
    public void setId(Integer id) {
        this.id = id;
    }

    @SpaceRouting
    public String getName() {
        return name;
    }
    public void setName(String name) {
        this.name = name;
    }
    @OneToMany(cascade = CascadeType.ALL)
    public List<Pet> getPets() {
        return pets;
    }
    public void setPets(List<Pet> pets) {
        this.pets = pets;
    }
}

Non-Indexed Fields

Non-indexed fields that are not used for queries should be placed within a user-defined class (payload object) and have their getter and setter placed within the payload class. This improves read/write performance because these fields are not introduced to the Space class model.

JPA Query Language (JPQL)

GigaSpaces JPA supports a subset of JPQL. Here are a few examples of the supported queries:

Querying on Properties of Nested Objects

EntityManagerFactory emf = Persistence.createEntityManagerFactory("gigaspaces");
EntityManager em = emf.createEntityManager();
Query query = em.createQuery("SELECT c from org.openspaces.objects.Customer c WHERE c.address.country = 'United States'");
List<Customer> customers = (List<Customer>) query.getResultList();
em.close();
emf.close();

JOIN support for one to many relationship (Owner –> List<Pet>)

EntityManagerFactory emf = Persistence.createEntityManagerFactory("gigaspaces");
EntityManager em = emf.createEntityManager();
Query query = em.createQuery("SELECT o FROM org.openspaces.objects.Owner o JOIN o.pets p WHERE p.name = :name");
query.setParameter("name", "Whiskey");
Owner owner = (Owner) query.getSingleResult();
em.close();
emf.close();

When specifying entity names in the GigaSpaces JPQL, the full class qualified name should be used as shown in the above examples.

Persisting Collection Properties

You can make a collection property persistent by using the @ElementCollection annotation. The following example shows an entity with a collection of Integers:

@Entity
public class Card {
  // ...

  private List<Integer> numbers;

  @ElementCollection
  @SpaceIndex(path = "[*]") // the list values will be indexed.
  public List<Integer> getNumbers() {
    return this.numbers;
  }

  public void setNumbers(List<Integer> numbers) {
    this.numbers = numbers;
  }

  // ...
}

In order to query the Card entity using a specific Integer in the numbers collection we use JPQL's MEMBER OF:

EntityManagerFactory emf = Persistence.createEntityManagerFactory("gigaspaces");
EntityManager em = emf.createEntityManager();
Query query = em.createQuery("SELECT c FROM org.openspaces.objects.Card c WHERE :number MEMBER OF c.numbers");
query.setParameter("number", "10");
Card card = (Card) query.getSingleResult();
em.close();
emf.close();

Persisting Enum Properties

JPA allows persisting Enum properties using the @Enumerated annotation, as shown below:

// A Vehicle entity which has an Enum property
@Entity
public class Vehicle {
  // Enum Declaration
  public enum VehicleType { CAR, TRUCK, BIKE };

  private Integer id;
  private String name;
  private VehicleType type;

  public Vehicle() {
  }

  @Id
  @SpaceId
  public Integer getId() {
    return this.id;
  }

  public String getName() {
    return this.name;
  }

  @Enumerated
  public VehicleType getType() {
    return this.type;
  }

  /* Additional Getters & Setters */

}

We used the @Enumerated annotation for persisting an Enum property. Specifying a value for the @Enumerated.value() attribute has no effect because Enums are saved in GigaSpaces as is.

Enums In JPQL

You can query according to an Enum property by setting an Enum parameter, or by using the Enum's value in the query string:

EntityManager em = emf.createEntityManager();

// Query using an Enum parameter
Query query1 = em.createQuery("SELECT vehicle FROM com.gigaspaces.objects.Vehicle vehicle WHERE vehicle.type = :type");
query1.setParameter("type", VehicleType.CAR);
Vehicle result1 = (Vehicle) query1.getSingleResult();

// Query using an Enum in query's string
Query query2 = em.createQuery("SELECT vehicle FROM com.gigaspaces.objects.Vehicle vehicle WHERE vehicle.type = 'BIKE'");
Vehicle result2 = (Vehicle) query2.getSingleResult();

Interoperability

The GigaSpaces JPA implementation is fully interoperable with the GigaSpaces native POJO API. For instance, you can persist a JPA entity and read it using the native POJOClosed Plain Old Java Object. A regular Java object with no special restrictions other than those forced by the Java Language Specification and does not require any classpath.-driven Space API:

@Entity
public class Author {
  private Integer id;
  private String name;
  private List<Book> books;

  public Author() {
  }

  @Id
  @SpaceId
  public Integer getId() {
    return this.id;
  }

  public String getName() {
    return this.name;
  }

  @OneToMany(cascade = CascadeType.ALL)
  @SpaceIndex(path = "[*].id")
  public List<Book> getBooks() {
    return this.books;
  }

  // Additional Getters & Setters...

}

@Entity
public class Book implements Serializable {
  private Integer id;
  private String name;

  public Book() {
  }

  @Id
  @SpaceId
  public Integer getId() {
    return this.id;
  }

  public String getName() {
    return this.name;
  }

  // Additional Getters & Setters...

}

GigaSpace gigaspace = ...
Book book1 = new Book(10, "Book Title 1");
Book book2 = new Book(20, "Book Title 2");
List<Book> books = new ArrayList<Book>();
books.add(book1);
books.add(book2);
Author author = new Author();
author.setId(1234);
author.setBooks(books);

// Persist using GigaSpaces JPA..
Properties properties = new Properties();
properties.put("ConnectionFactory", gigaspace.getSpace());
EntityManagerFactory emf = Persistence.createEntityManagerFactory("gigaspaces", properties);
EntityManager em = emf.createEntityManager();
em.getTransaction().begin();
em.persist(author);
em.getTransaction().commit();
em.close();

// Read using Space API..
Author result = gigaspace.readById(Author.class, 1234);

// Or even SQLQuery..
SQLQuery<Author> query = new SQLQuery<Author>(Author.class, "id = 1234");
result = gigaspace.read(query);

// Or by a certain book..
query = new SQLQuery<Author>(Author.class, "books[*].id = 10");
result = gigaspace.read(query);

Native Query Execution

The GigaSpaces JPA native query execution is a powerful feature used to execute the items discussed in the following subsections.

For more information on the following topics:

SQLQuery Execution

SQLQuery execution using JPA native query API is relatively simple, and done as follows:

// SQLQuery execution
EntityManager em = emf.createEntityManagerFactory();
Query query = em.createNativeQuery("name = 'John Doe'", Author.class);
Author author = (Author) query.getSingleResult();

// SQLQuery execution with parameters
query = em.createNativeQuery("name = ?", Author.class);
query.setParameter(1, "John Doe");
author = (Author) query.getSingleResult();

For more details about the SQLQuery syntax, refer to the SQLQuery page.

Task Execution

Using GigaSpaces's JPA native query API, you can execute tasks over the Space as follows:

// Task definition
public class MyTask implements Task<Integer> {

  @TaskGigaSpace
  private transient GigaSpace gigaSpace;
  private Object routingClosed The mechanism that is in charge of routing the objects into and out of the corresponding partitions. The routing is based on a designated attribute inside the objects that are written to the Space, called the Routing Index.;

  public MyTask(Object routing) {
    this.routing = routing;
  }

  public Integer execute() throws Exception {
    return gigaSpace.count(new Author());
  }

  @SpaceRouting
  public Object getRouting() {
    return this.routing;
  }
}

// Task execution
Query query = em.createNativeQuery("execute ?");    // Special syntax for task execution
query.setParameter(1, new MyTask(1));               // We pass our task instance as a parameter to the query
Integer result = (Integer) query.getSingleResult(); // Task execution always returns a single result

Task execution using JPA's native query API is always synchronous.

Getting an EntityManagerFactory Instance in a Task

You can get an EntityManagerFactory instance (according to the bean definition in puClosed This is the unit of packaging and deployment in the GigaSpaces Data Grid, and is essentially the main GigaSpaces service. The Processing Unit (PU) itself is typically deployed onto the Service Grid. When a Processing Unit is deployed, a Processing Unit instance is the actual runtime entity..xml) by implementing the ApplicationContextAware interface. For example:

public class MyTask implements Task<Integer>, ApplicationContextAware {

  private transient EntityManagerFactory emf;

  public void setApplicationContext(ApplicationContext context) throws BeansException {
    // Get the entityManagerFactory bean
    emf = (EntityManagerFactory) context.getBean("entityManagerFactory");
  }

  public Integer execute() throws Exception {

    // Create an EntityManager..
    EntityManager em = emf.createEntityManager();

    // ...

    em.close();

    return 0;
  }

}

Another option (instead of using the ApplicationContextAware interface) is to annotate your task with the @AutowireTask annotation, and annotate the EntityManagerFactory property with a @Resource annotation.

For more information about GigaSpaces tasks, refer to Task Execution.

Dynamic Script Execution

In addition to task execution, the GigaSpaces JPA native query execution also offers the ability to execute dynamic scripts such as Groovy, JavaScript & JRuby over the Space. Dynamic script execution over the Space is based on task execution and remoting, so your Processing UnitClosed This is the unit of packaging and deployment in the GigaSpaces Data Grid, and is essentially the main GigaSpaces service. The Processing Unit (PU) itself is typically deployed onto the Service Grid. When a Processing Unit is deployed, a Processing Unit instance is the actual runtime entity. must have a remoting scripting executor service:

<!-- The service exporter exposing the scripting service -->
<os-remoting:service-exporter id="serviceExporter">
     <os-remoting:service ref="scriptingExecutor"/>
</os-remoting:service-exporter>

The next step is using the exposed scripting service on the client side, using JPA's native query API:

// Dynamic Script execution
Script script = new StaticScript("GroovyScript", "groovy", "println 'Dynamic Script Execution using JPA'; return 0");

Query query = em.createNativeQuery("execute ?");     // Special syntax for script execution (similar to task execution)
query.setParameter(1, script);                       // We pass our script as a parameter to the query
Integer result = (Integer) query.getSingleResult();  // Script execution always returns a single result

For more information about dynamic script execution, refer to Dynamic Language Tasks.

GigaSpaces JPA Limitations

For a list of unsupported JPA features and limitations, refer to GigaSpaces JPA Limitations.