View source for EclipseLink/UserGuide/JPA/Basic JPA Development/Entities/SecondaryTable

{{EclipseLink_UserGuide
|info=y
|toc=y
|eclipselink=y
|eclipselinktype=JPA
|api=y
|apis=
* [http://www.eclipse.org/eclipselink/api/latest/javax/persistence/SecondaryTable.html @SecondaryTable]
* [http://www.eclipse.org/eclipselink/api/latest/javax/persistence/SecondaryTables.html @SecondaryTables]
* [http://www.eclipse.org/eclipselink/api/latest/javax/persistence/PrimaryKeyJoinColumn.html @PrimaryKeyJoinColumn]
* [http://www.eclipse.org/eclipselink/api/latest/javax/persistence/PrimaryKeyJoinColumns.html @PrimaryKeyJoinColumns]
|nativeapi=y
|nativeapis=
* [http://www.eclipse.org/eclipselink/api/latest/org/eclipse/persistence/descriptors/ClassDescriptor.html ClassDescriptor]
** [http://www.eclipse.org/eclipselink/api/latest/org/eclipse/persistence/descriptors/ClassDescriptor.html#addTableName(java.lang.String) addTable()]
** [http://www.eclipse.org/eclipselink/api/latest/org/eclipse/persistence/descriptors/ClassDescriptor.html#addForeignKeyFieldNameForMultipleTable(java.lang.String,%20java.lang.String) addForeignKeyFieldName ForMultipleTable()]
* [http://www.eclipse.org/eclipselink/api/latest/org/eclipse/persistence/descriptors/DescriptorQueryManager.html DescriptorQueryManager]
** [http://www.eclipse.org/eclipselink/api/latest/org/eclipse/persistence/descriptors/DescriptorQueryManager.html#setMultipleTableJoinExpression(org.eclipse.persistence.expressions.Expression) setMultipleTableJoinExpression()]
* [http://www.eclipse.org/eclipselink/api/latest/org/eclipse/persistence/expressions/Expression.html Expression]
}}

=@SecondaryTable=
You can use the <tt>@SecondaryTable</tt> annotation or <code><nowiki><secondary-table></nowiki></code> XML element to configure an entity to map to multiple tables.  When an entity instance is persisted, it will insert into each of the tables.  When an entity instance is read, the tables will be joined.  This allows for the entity to define mappings that make use of any of the columns in any of the tables.

Secondary tables should only be used when the entity logically has its data spread across multiple tables.  If the tables represent relationships, then this is better modeled using relationship mappings such as <code>OneToOne</code> and <code>OneToMany</code>.

By default the secondary table is joined to the primary table's id columns.  This assumes that the secondary table has the same id columns defined with the same name.
If the secondary table uses different names for the id columns, then a set or <tt>@PrimaryKeyJoinColumn</tt>s must be provided.  <tt>@PrimaryKeyJoinColumn</tt> can also be used for <tt>JOINED</tt> inheritance to specify how the child table is joined to the parent table.  <tt>@PrimaryKeyJoinColumn</tt> still requires that the secondary table have all of the id columns of the primary table, for more advanced multiple table joins see [[#Advanced Multiple Table Configuration|Advanced Multiple Table Configuration]].

{{EclipseLink_AttributeTable
|caption=@SecondaryTable Attributes
|content=<tr>
 <td>'''<tt>name</tt>'''</td>
 <td>The name of the table
 <td></td>
 <td>Yes</td>
</tr>
<tr>
 <td>'''<tt>catalog</tt>'''</td>
 <td>A <tt>String</tt> catalog name.
 <td>Default catalog for database</td>
 <td>No</td>
</tr>
<tr>
 <td>'''<tt>schema</tt>'''</td>
 <td>The <tt>String</tt> name of the schema.</td>
 <td>Default schema of the database</td>
 <td>No</td>
</tr>
<tr>
 <td>'''<tt>uniqueConstraints</tt>'''</td>
 <td>This is used only by DDL generation.  By default only a primary key and foreign key constraints are defined, if desired set the value of this attribute to an array of one or more <tt>UniqueConstraint</tt> instances. 
{{EclipseLink_Spec|section=Section 11.1.49 "UniqueConstraint Annotation"}}</td>
 <td>No additional constraints</td>
 <td>No</td>
</tr>
}}

{{EclipseLink_Spec|section=Section 11.1.42 "SecondaryTable Annotation"}}

{{EclipseLink_AttributeTable
|caption=@PrimaryKeyJoinColumn Attributes
|content=<tr>
 <td>'''<tt>name</tt>'''</td>
 <td>The name foreign key column in the secondary table that references the id column in the primary table
 <td>id column</td>
 <td>No</td>
</tr>
<tr>
 <td>'''<tt>referencedColumnName</tt>'''</td>
 <td>A name of the id column in the primary table that is being referenced.  This is only required for composite ids.
 <td>id column</td>
 <td>No</td>
</tr>
<tr>
 <td>'''<tt>columnDefinition</tt>'''</td>
 <td>Optional column description for use with DDL generation.</td>
 <td></td>
 <td>No</td>
</tr>
}}

{{EclipseLink_Spec|section=Section 11.1.40 "PrimaryKeyJoinColumn Annotation"}}

The following example shows how to use this annotation to specify the table for <code>Employee</code>.

<span id="Example: @Table"></span>
======''Example: Using @Table''======
<source lang="java">
@Entity
@Table(name="EMP")
@SecondaryTable(name="SALARY")
public class Employee implements Serializable {
    ...
    @Id
    public Long getId() {
        return id;
    }
    ...
}
</source>

======''Example: Using <code><nowiki><table-generator></nowiki></code>''======

<source lang="xml">
<entity class="Employee">
    <table name="EMP"/>
    <secondary-table name="SALARY"/>
    <attributes>
        <id name="id"/>
        ...
    </attributes>
</entity>
</source>
<br>

==Advanced Multiple Table Configuration==
JPA requires that secondary tables contain the id of the entity.  EclipseLink allows for the multiple table join to be based on foreign keys, or on specific criteria.

To define a secondary table that is joined through a foreign key, the <code>ClassDescriptor.addForeignKeyFieldNameForMultipleTable()</code> API is used.  This column name given to this API should be prefixed by its table name, this allows for foreign keys from any of the secondary tables to be used in the multiple table join.  Note that all of the column values still need to be mapped at least once.

If the multiple table join is more complex, an EclipseLink <code>Expression</code> can be used to define the join.  The <code>Expression</code> is set on the <code>ClassDescriptor</code>'s query manager using the <code>DescriptorQueryManager.setMultipleTableJoinExpression()</code> API.  It is even possible to define a join <code>Expression</code> that make use of an outer-join on certain databases.

In certain advanced cases multiple tables can be encapsulated using a database view.  This allows for any table join to be define in SQL and for the entity to be simply mapped to the view.  To map an entity to a view, just use the view name instead of the table name.  If the view is not updateable, your database may support instead of triggers to allow updates.

======''Example: Using a DescriptorCustomizer to configure a secondary table joined through a foreign key''======
<source lang="java">
@Entity
@Table(name="EMP")
@SecondaryTable(name="EMP_INFO")
@Customizer(EmployeeCustomizer.class)
public class Employee implements Serializable {
    @Id
    private Long id;
    @Basic
    @Column(name="ID", table="EMP_INFO")
    private String infoId;
    ...
}
</source>

<source lang="java">
public class EmployeeCustomizer implements DescriptorCustomizer {
    public void customize(ClassDescriptor descriptor) {
        descriptor.getAdditionalTablePrimaryKeyFields().clear();
        descriptor.getMultipleTableForeignKeys()().clear();
        descriptor.addForeignKeyFieldNameForMultipleTable("EMP.INFO_ID", "EMP_INFO.ID");
    }
}
</source>

======''Example: Using a DescriptorCustomizer to configure a secondary table joined through an Expression''======
<source lang="java">
@Entity
@Table(name="EMP")
@SecondaryTable(name="EMP_INFO")
@Customizer(EmployeeCustomizer.class)
public class Employee implements Serializable {
    @Id
    private Long id;
    @Basic
    @Column(name="ID", table="EMP_INFO")
    private String infoId;
    ...
}
</source>

<source lang="java">
public class EmployeeCustomizer implements DescriptorCustomizer {
    public void customize(ClassDescriptor descriptor) {
        ExpressionBuilder emp = new ExpressionBuilder();
        Expression join = emp.getField("EMP.INFO_ID").equal(emp.getField("EMP_INFO.ID").and(emp.getField("EMP_INFO.ARCHIVE").equal(false));
        descriptor.getQueryManager().setMultipleTableJoinExpression(join);
    }
}
</source>

{{EclipseLink_JPA
|previous=[[EclipseLink/UserGuide/JPA/Basic JPA Development/Entities/Table|@Table]]
|next=    [[EclipseLink/UserGuide/JPA/Basic JPA Development/Entities/Inheritance|@Inheritance]]
|up=      [[EclipseLink/UserGuide/JPA/Basic JPA Development/Entities/Creating_and_Configuring_Entities|Configuring Entities]]
|version=2.2.0 DRAFT}}