EntityManager The diagram above presents an overview of the EntityManager interface. For a complete treatment of the EntityManager API, see the Javadoc documentation. Methods whose parameter signatures consist of an ellipsis (...) are overloaded to take multiple parameter types. OpenJPA extends the standard EntityManager interface with the org.apache.openjpa.persistence.OpenJPAEntityManager interface to provide additional functionality. The EntityManager is the primary interface used by application developers to interact with the JPA runtime. The methods of the EntityManager can be divided into the following functional categories: Transaction association. Entity lifecycle management. Entity identity management. Cache management. Query factory. Closing.

EntityManager obtaining the Transaction transactions Transaction obtaining from EntityManager public EntityTransaction getTransaction (); Every EntityManager has a one-to-one relation with an EntityTransaction instance. In fact, many vendors use a single class to implement both the EntityManager and EntityTransaction interfaces. If your application requires multiple concurrent transactions, you will use multiple EntityManagers. You can retrieve the EntityTransaction associated with an EntityManager through the getTransaction method. Note that most JPA implementations can integrate with an application server's managed transactions. If you take advantage of this feature, you will control transactions by declarative demarcation or through the Java Transaction API (JTA) rather than through the EntityTransaction.

EntityManager lifecycle operations EntityManagers perform several actions that affect the lifecycle state of entity instances. public void persist(Object entity); EntityManager persist persist EntityManager persistent objects persisting Transitions new instances to managed. On the next flush or commit, the newly persisted instances will be inserted into the datastore. For a given entity A, the persist method behaves as follows: If A is a new entity, it becomes managed. If A is an existing managed entity, it is ignored. However, the persist operation cascades as defined below. If A is a removed entity, it becomes managed. If A is a detached entity, an IllegalArgumentException is thrown. The persist operation recurses on all relation fields of A whose cascades include CascadeType.PERSIST. This action can only be used in the context of an active transaction. public void remove(Object entity); EntityManager remove remove EntityManager persistent objects deleting Transitions managed instances to removed. The instances will be deleted from the datastore on the next flush or commit. Accessing a removed entity has undefined results. For a given entity A, the remove method behaves as follows: If A is a new entity, it is ignored. However, the remove operation cascades as defined below. If A is an existing managed entity, it becomes removed. If A is a removed entity, it is ignored. If A is a detached entity, an IllegalArgumentException is thrown. The remove operation recurses on all relation fields of A whose cascades include CascadeType.REMOVE. This action can only be used in the context of an active transaction. public void refresh(Object entity); EntityManager refresh refresh EntityManager persistent objects refreshing state transactions optimistic Use the refresh action to make sure the persistent state of an instance is synchronized with the values in the datastore. refresh is intended for long-running optimistic transactions in which there is a danger of seeing stale data. For a given entity A, the refresh method behaves as follows: If A is a new entity, it is ignored. However, the refresh operation cascades as defined below. If A is an existing managed entity, its state is refreshed from the datastore. If A is a removed entity, it is ignored. If A is a detached entity, an IllegalArgumentException is thrown. The refresh operation recurses on all relation fields of A whose cascades include CascadeType.REFRESH. public Object merge(Object entity); EntityManager merge detachment detachment JPA merge detachment data transfer object value object A common use case for an application running in a servlet or application server is to "detach" objects from all server resources, modify them, and then "attach" them again. For example, a servlet might store persistent data in a user session between a modification based on a series of web forms. Between each form request, the web container might decide to serialize the session, requiring that the stored persistent state be disassociated from any other resources. Similarly, a client/server application might transfer persistent objects to a client via serialization, allow the client to modify their state, and then have the client return the modified data in order to be saved. This is sometimes referred to as the data transfer object or value object pattern, and it allows fine-grained manipulation of data objects without incurring the overhead of multiple remote method invocations. JPA provides support for this pattern by automatically detaching entities when they are serialized or when a persistence context ends (see for an exploration of persistence contexts). The JPA merge API re-attaches detached entities. This allows you to detach a persistent instance, modify the detached instance offline, and merge the instance back into an EntityManager (either the same one that detached the instance, or a new one). The changes will then be applied to the existing instance from the datastore. A detached entity maintains its persistent identity, but cannot load additional state from the datastore. Accessing any persistent field or property that was not loaded at the time of detachment has undefined results. Also, be sure not to alter the version or identity fields of detached instances if you plan on merging them later. The merge method returns a managed copy of the given detached entity. Changes made to the persistent state of the detached entity are applied to this managed instance. Because merging involves changing persistent state, you can only merge within a transaction. If you attempt to merge an instance whose representation has changed in the datastore since detachment, the merge operation will throw an exception, or the transaction in which you perform the merge will fail on commit, just as if a normal optimistic conflict were detected. OpenJPA offers enhancements to JPA detachment functionality, including additional options to control which fields are detached. See in the Reference Guide for details. For a given entity A, the merge method behaves as follows: If A is a detached entity, its state is copied into existing managed instance A' of the same entity identity, or a new managed copy of A is created. If A is a new entity, a new managed entity A' is created and the state of A is copied into A'. If A is an existing managed entity, it is ignored. However, the merge operation still cascades as defined below. If A is a removed entity, an IllegalArgumentException is thrown. The merge operation recurses on all relation fields of A whose cascades include CascadeType.MERGE. public void lock (Object entity, LockModeType mode); EntityManager lock locking EntityManager This method locks the given entity using the named mode. The javax.persistence.LockModeType enum defines two modes: READ: Other transactions may concurrently read the object, but cannot concurrently update it. WRITE: Other transactions cannot concurrently read or write the object. When a transaction is committed that holds WRITE locks on any entities, those entities will have their version incremented even if the entities themselves did not change in the transaction. OpenJPA has additional APIs for controlling object locking. See in the Reference Guide for details. The following diagram illustrates the lifecycle of an entity with respect to the APIs presented in this section.

The examples below demonstrate how to use the lifecycle methods presented in the previous section. The examples are appropriate for out-of-container use. Within a container, EntityManagers are usually injected, and transactions are usually managed. You would therefore omit the createEntityManager and close calls, as well as all transaction demarcation code. persistent objects persisting example // create some objects Magazine mag = new Magazine("1B78-YU9L", "JavaWorld"); Company pub = new Company("Weston House"); pub.setRevenue(1750000D); mag.setPublisher(pub); pub.addMagazine(mag); Article art = new Article("JPA Rules!", "Transparent Object Persistence"); art.addAuthor(new Author("Fred", "Hoyle")); mag.addArticle(art); // persist EntityManager em = emf.createEntityManager(); em.getTransaction().begin(); em.persist(mag); em.persist(pub); em.persist(art); em.getTransaction().commit(); // or we could continue using the EntityManager... em.close(); persistent objects updating example Magazine.MagazineId mi = new Magazine.MagazineId(); mi.isbn = "1B78-YU9L"; mi.title = "JavaWorld"; // updates should always be made within transactions; note that // there is no code explicitly linking the magazine or company // with the transaction; JPA automatically tracks all changes EntityManager em = emf.createEntityManager(); em.getTransaction().begin(); Magazine mag = em.find(Magazine.class, mi); mag.setPrice(5.99); Company pub = mag.getPublisher(); pub.setRevenue(1750000D); em.getTransaction().commit(); // or we could continue using the EntityManager... em.close(); persistent objects deleting example // assume we have an object id for the company whose subscriptions // we want to delete Object oid = ...; // deletes should always be made within transactions EntityManager em = emf.createEntityManager(); em.getTransaction().begin(); Company pub = (Company) em.find(Company.class, oid); for (Subscription sub : pub.getSubscriptions()) em.remove(sub); pub.getSubscriptions().clear(); em.getTransaction().commit(); // or we could continue using the EntityManager... em.close(); This example demonstrates a common client/server scenario. The client requests objects and makes changes to them, while the server handles the object lookups and transactions. // CLIENT: // requests an object with a given oid Record detached = (Record) getFromServer(oid); ... // SERVER: // send object to client; object detaches on EM close Object oid = processClientRequest(); EntityManager em = emf.createEntityManager(); Record record = em.find(Record.class, oid); em.close(); sendToClient(record); ... // CLIENT: // makes some modifications and sends back to server detached.setSomeField("bar"); sendToServer(detached); ... // SERVER: // merges the instance and commit the changes Record modified = (Record) processClientRequest(); EntityManager em = emf.createEntityManager(); em.getTransaction().begin(); Record merged = (Record) em.merge(modified); merged.setLastModified(System.currentTimeMillis()); merged.setModifier(getClientIdentityCode()); em.getTransaction().commit(); em.close();

Each EntityManager is responsible for managing the persistent identities of the managed objects in the persistence context. The following methods allow you to interact with the management of persistent identities. The behavior of these methods is deeply affected by the persistence context type of the EntityManager; see for an explanation of persistence contexts. public <T> T find(Class<T> cls, Object oid); EntityManager find identity find EntityManager identity retrieving objects by identity This method returns the persistent instance of the given type with the given persistent identity. If the instance is already present in the current persistence context, the cached version will be returned. Otherwise, a new instance will be constructed and loaded with state from the datastore. If no entity with the given type and identity exists in the datastore, this method returns null. public <T> T getReference(Class<T> cls, Object oid); EntityManager getReference identity getReference EntityManager identity retrieving objects by identity EntityNotFoundException This method is similar to find, but does not necessarily go to the database when the entity is not found in cache. The implementation may construct a hollow entity and return it to you instead. Hollow entities do not have any state loaded. The state only gets loaded when you attempt to access a persistent field. At that time, the implementation may throw an EntityNotFoundException if it discovers that the entity does not exist in the datastore. The implementation may also throw an EntityNotFoundException from the getReference method itself. Unlike find, getReference does not return null. public boolean contains(Object entity); EntityManager contains contains EntityManager Returns true if the given entity is part of the current persistence context, and false otherwise. Removed entities are not considered part of the current persistence context.

EntityManager cache public void flush(); EntityManager flush flush EntityManager transactions flushing changes before commit The flush method writes any changes that have been made in the current transaction to the datastore. If the EntityManager does not already have a connection to the datastore, it obtains one for the flush and retains it for the duration of the transaction. Any exceptions during flush cause the transaction to be marked for rollback. See . Flushing requires an active transaction. If there isn't a transaction in progress, the flush method throws a TransactionRequiredException. public FlushModeType getFlushMode(); public void setFlushMode(FlushModeType flushMode); EntityManager FlushMode FlushMode The EntityManager's FlushMode property controls whether to flush transactional changes before executing queries. This allows the query results to take into account changes you have made during the current transaction. Available javax.persistence.FlushModeType constants are: COMMIT: Only flush when committing, or when told to do so through the flush method. Query results may not take into account changes made in the current transaction. AUTO: The implementation is permitted to flush before queries to ensure that the results reflect the most recent object state. You can also set the flush mode on individual Query instances. OpenJPA only flushes before a query if the query might be affected by data changed in the current transaction. Additionally, OpenJPA allows fine-grained control over flushing behavior. See the Reference Guide's . public void clear(); EntityManager clear clear EntityManager Clearing the EntityManager effectively ends the persistence context. All entities managed by the EntityManager become detached.

EntityManager as Query factory Query Query creating public Query createQuery(String query); Query objects are used to find entities matching certain criteria. The createQuery method creates a query using the given Java Persistence Query Language (JPQL) string. See for details. public Query createNamedQuery(String name); This method retrieves a query defined in metadata by name. The returned Query instance is initialized with the information declared in metadata. For more information on named queries, read . public Query createNativeQuery(String sql); public Query createNativeQuery(String sql, Class resultCls); public Query createNativeQuery(String sql, String resultMapping); Native queries are queries in the datastore's native language. For relational databases, this the Structured Query Language (SQL). elaborates on JPA's native query support.

EntityManager properties information &properties_info.xml;

EntityManager closing public boolean isOpen(); public void close(); When an EntityManager is no longer needed, you should call its close method. Closing an EntityManager releases any resources it is using. The persistence context ends, and the entities managed by the EntityManager become detached. Any Query instances the EntityManager created become invalid. Calling any method other than isOpen on a closed EntityManager results in an IllegalStateException. You cannot close a EntityManager that is in the middle of a transaction. If you are in a managed environment using injected entity managers, you should not close them.