org.eclipse.persistence.descriptors.SerializedObjectPolicy Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of eclipselink Show documentation
EclipseLink build based upon Git transaction f2b9fc5
There is a newer version: 5.0.0-B03
/*
 * Copyright (c) 2013, 2018 Oracle and/or its affiliates. All rights reserved.
 *
 * This program and the accompanying materials are made available under the
 * terms of the Eclipse Public License v. 2.0 which is available at
 * http://www.eclipse.org/legal/epl-2.0,
 * or the Eclipse Distribution License v. 1.0 which is available at
 * http://www.eclipse.org/org/documents/edl-v10.php.
 *
 * SPDX-License-Identifier: EPL-2.0 OR BSD-3-Clause
 */

// Contributors:
//     24 April 2013-2.5.1 ailitche
//       SerializedObjectPolicy initial API and implementation
package org.eclipse.persistence.descriptors;

import java.io.Serializable;
import java.util.List;

import org.eclipse.persistence.internal.helper.DatabaseField;
import org.eclipse.persistence.internal.sessions.AbstractRecord;
import org.eclipse.persistence.internal.sessions.AbstractSession;
import org.eclipse.persistence.queries.ObjectLevelReadQuery;

/**
 * SerializedObjectPolicy (SOP) could be set on a non-aggregate descriptor.
 *
 * If SerializedObjectPolicy is specified Eclipselink writes out the whole entity object with its
 * privately owned (and nested privately owned) entities and element collections into an additional
 * (likely BLOB) field in the database.
 *
 * If SerializedObjectPolicy is set on an entity then SerializedObjectPolicies with the same field are set
 * on all inheriting entities.
 *
 * The goal is to make reads from the database faster.
 * The draw back is slower writes into the database.
 * So SerializedObjectPolicy may make sense for read-only / read-mostly application
 * for Entity, which always loads all its dependent entities and / or ElementCollections.
 *
 * To use SerializedObjectPolicy, ObjectLevelReadQuery should set a boolean flag
 * @see ObjectLevelReadQuery#setShouldUseSerializedObjectPolicy
 *
 * In case the serialized object column contains null or obsolete version of the object
 * the query using SerializedObjectPolicy would either throw exception or - if all other fields have been read, too -
 * would build the object using these fields (exactly as in case SerializedObjectPolicy is not used).
 *
 * Note that currently no default implementation of SerializedObjectPolicy is available
 * and this class should be provided by the user.
 *
 * @author ailitche
 * @since EclipseLink 2.5.1
 *
 * @see org.eclipse.persistence.annotations.SerializedObject
 */
public interface SerializedObjectPolicy extends Cloneable, Serializable {
    /** get owning descriptor */
    ClassDescriptor getDescriptor();
    /** set owning descriptor */
    void setDescriptor(ClassDescriptor descriptor);
    /** get the field that stores sopObject in the database (sopField) */
    DatabaseField getField();
    /** set the field that stores sopObject in the database (sopField) */
    void setField(DatabaseField field);

    SerializedObjectPolicy clone();
    /** instantiate policy for child descriptor */
    SerializedObjectPolicy instantiateChild();

    /** unless the field is already initialized by parent descriptor, add the field to the owning descriptor and set field's type */
    void initializeField(AbstractSession session);
    /** initialize the policy */
    void initialize(AbstractSession session);
    /** postinitialize the policy: determine which mappings are included, which fields should be selected */
    void postInitialize(AbstractSession session);

    /**
     * Lists the database fields that should be read by the query using the policy.
     * To allow recovery in case of null or invalid sopObject, then this method should return all the fields define by descriptor
     * (descriptor.getFields()).
     */
    List getSelectionFields();
    /**
     * Lists the database fields that should be read by the query using the policy, in case all inherited objects are read using outer joining.
     * To allow recovery in case of null or invalid sopObject, then this method should return all the fields define by descriptor
     * (descriptor.getAllFields()).
     */
    List getAllSelectionFields();

    /** Serialize the object and put the result into the row as a value corresponding to the policy field */
    void putObjectIntoRow(AbstractRecord databaseRow, Object object, AbstractSession session);
    /**
     * Deserialize the object from the value corresponding to the policy field, nullify that value, set the object into the row using setSopObject, also return it.
     * If the object is null or invalid, behaviour depend on whether the policy allows the query to recover or not (see comments to getFieldsToSelect and getAllFieldsToSelect methods):
     * if recovery is possible then the method should return null, otherwise throw QueryException (query is a parameter of this method only because it's required by QueryException).
     */
    Object getObjectFromRow(AbstractRecord databaseRow, AbstractSession session, ObjectLevelReadQuery query);
}