8. Field Mapping

8. Field Mapping
Prev	Chapter 13. Mapping Metadata	Next

8.1. Basic Mapping

8.1.1. LOBs
8.1.2. Enumerated
8.1.3. Temporal Types
8.1.4. The Updated Mappings

A basic field mapping stores the field value directly into a database column. The following field metadata types use basic mapping. These types were defined in Section 2, “ Field and Property Metadata ”.

Id fields.
Version fields.
Basic fields.

In fact, you have already seen examples of basic field mappings in this chapter - the mapping of all identity fields in Example 13.3, “ Identity Mapping ”. As you saw in that section, to write a basic field mapping you use the Column annotation to describe the column the field value is stored in. We discussed the Column annotation in Section 3, “ Column ”. Recall that the name of the column defaults to the field name, and the type of the column defaults to an appropriate type for the field type. These defaults allow you to sometimes omit the annotation altogether.

8.1.1. LOBs

Adding the Lob marker annotation to a basic field signals that the data is to be stored as a LOB (Large OBject). If the field holds string or character data, it will map to a CLOB (Character Large OBject) database column. If the field holds any other data type, it will be stored as binary data in a BLOB (Binary Large OBject) column. The implementation will serialize the Java value if needed.

The equivalent XML element is lob, which has no children or attributes.

Note

OpenJPA also supports LOB streaming. See Section 7.11, “ LOB Streaming ” in the Reference Guide for details.

8.1.2. Enumerated

You can apply the Enumerated annotation to your Enum fields to control how they map to the database. The Enumerated annotation's value one of the following constants from the EnumType enum:

EnumType.ORDINAL: The default. The persistence implementation places the ordinal value of the enum in a numeric column. This is an efficient mapping, but may break if you rearrange the Java enum declaration.
EnumType.STRING: Store the name of the enum value rather than the ordinal. This mapping uses a VARCHAR column rather than a numeric one.

The Enumerated annotation is optional. Any un-annotated enumeration field defaults to ORDINAL mapping.

The corresponding XML element is enumerated. Its embedded text must be one of STRING or ORIDINAL.

8.1.3. Temporal Types

The Temporal annotation determines how the implementation handles your basic java.util.Date and java.util.Calendar fields at the JDBC level. The Temporal annotation's value is a constant from the TemporalType enum. Available values are:

TemporalType.TIMESTAMP: The default. Use JDBC's timestamp APIs to manipulate the column data.
TemporalType.DATE: Use JDBC's SQL date APIs to manipulate the column data.
TemporalType.TIME: Use JDBC's time APIs to manipulate the column data.

If the Temporal annotation is omitted, the implementation will treat the data as a timestamp.

The corresponding XML element is temporal, whose text value must be one of: TIME, DATE, or TIMESTAMP.

8.1.4. The Updated Mappings

Below we present an updated diagram of our model and its associated database schema, followed by the corresponding mapping metadata. Note that the mapping metadata relies on defaults where possible. Also note that as a mapped superclass, Document can define mappings that will automatically transfer to its subclass' tables. In Section 8.3, “ Embedded Mapping ”, you will see how a subclass can override its mapped superclass' mappings.

Example 13.10. Basic Field Mapping

package org.mag;

@Entity
@IdClass(Magazine.MagazineId.class)
@Table(name="MAG")
@DiscriminatorValue("Mag")
public class Magazine {

    @Column(length=9)
    @Id private String isbn;
    @Id private String title;

    @Column(name="VERS")
    @Version private int version;
    
    private String name;
    private double price;

    @Column(name="COPIES")
    private int copiesSold;

    ...

    public static class MagazineId {
        ...
    }
}

@Entity
@Table(name="ART", uniqueConstraints=@Unique(columnNames="TITLE"))
@SequenceGenerator(name="ArticleSeq", sequenceName="ART_SEQ")
public class Article {

    @Id
    @GeneratedValue(strategy=GenerationType.SEQUENCE, generator="ArticleSeq") 
    private long id;

    @Column(name="VERS")
    @Version private int version;

    private String title;
    private byte[] content;

    ...
}


package org.mag.pub;

@Entity
@Table(name="COMP")
public class Company {

    @Column(name="CID")
    @Id private long id;

    @Column(name="VERS")
    @Version private int version;

    private String name;

    @Column(name="REV")
    private double revenue;

    ...
}

@Entity
@Table(name="AUTH")
public class Author {

    @Id
    @GeneratedValue(strategy=GenerationType.TABLE, generator="AuthorGen")
    @TableGenerator(name="AuthorGen", table="AUTH_GEN", pkColumnName="PK",
        valueColumnName="AID")
    @Column(name="AID", columnDefinition="INTEGER64")
    private long id;

    @Column(name="VERS")
    @Version private int version;

    @Column(name="FNAME")
    private String firstName;

    @Column(name="LNAME")
    private String lastName;

    ...
}

@Embeddable
public class Address {
    ...
}

package org.mag.subscribe;

@MappedSuperclass
public abstract class Document {

    @Id
    @GeneratedValue(strategy=GenerationType.IDENTITY)
    private long id;

    @Column(name="VERS")
    @Version private int version;

    ...
}

@Entity
@Table(schema="CNTRCT")
@Inheritance(strategy=InheritanceType.JOINED)
@DiscriminatorColumn(name="CTYPE")
public class Contract
    extends Document {

    @Lob
    private String terms;

    ...
}

@Entity
@Table(name="SUB", schema="CNTRCT")
@DiscriminatorColumn(name="KIND", discriminatorType=DiscriminatorType.INTEGER)
@DiscriminatorValue("1")
public class Subscription {

    @Id
    @GeneratedValue(strategy=GenerationType.IDENTITY)
    private long id;

    @Column(name="VERS")
    @Version private int version;

    @Column(name="START")
    private Date startDate;

    @Column(name="PAY")
    private double payment;

    ...

    @Entity
    @Table(name="LINE_ITEM", schema="CNTRCT")
    public static class LineItem
        extends Contract {

        @Column(name="COMM")
        private String comments;

        private double price;
        private long num;
        ...
    }
}

@Entity(name="Lifetime")
@DiscriminatorValue("2")
public class LifetimeSubscription
    extends Subscription {

    @Basic(fetch=FetchType.LAZY)
    @Column(name="ELITE")
    private boolean getEliteClub() { ... }
    public void setEliteClub(boolean elite) { ... }

    ...
}

@Entity(name="Trial")
@DiscriminatorValue("3")
public class TrialSubscription
    extends Subscription {

    @Column(name="END")
    public Date getEndDate() { ... }
    public void setEndDate(Date end) { ... }

    ...
}

The same metadata expressed in XML:

<entity-mappings xmlns="http://java.sun.com/xml/ns/persistence/orm" 
    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" 
    xsi:schemaLocation="http://java.sun.com/xml/ns/persistence/orm orm_1_0.xsd"
    version="1.0">
    <mapped-superclass class="org.mag.subscribe.Document">
        <attributes>
            <id name="id">
                <generated-value strategy="IDENTITY"/>
            </id>
            <version name="version">
                <column name="VERS"/>
            </version>
            ...
        </attributes>
    </mapped-superclass>
    <entity class="org.mag.Magazine">
        <table name="MAG"/>
        <id-class="org.mag.Magazine.MagazineId"/>
        <discriminator-value>Mag</discriminator-value>
        <attributes>
            <id name="isbn">
                <column length="9"/>
            </id>
            <id name="title"/>
            <basic name="name"/>
            <basic name="price"/>
            <basic name="copiesSold">
                <column name="COPIES"/>
            </basic>
            <version name="version">
                <column name="VERS"/>
            </version>
            ...
        </attributes>
    </entity>
    <entity class="org.mag.Article">
        <table name="ART">
            <unique-constraint>
               <column-name>TITLE</column-name>
            </unique-constraint>
        </table>
        <sequence-generator name="ArticleSeq", sequenceName="ART_SEQ"/>
        <attributes>
            <id name="id">
                <generated-value strategy="SEQUENCE" generator="ArticleSeq"/>
            </id>
            <basic name="title"/>
            <basic name="content"/>
            <version name="version">
                <column name="VERS"/>
            </version>
            ...
        </attributes>
    </entity>
    <entity class="org.mag.pub.Company">
        <table name="COMP"/>
        <attributes>
            <id name="id">
                <column name="CID"/>
            </id>
            <basic name="name"/>
            <basic name="revenue">
                <column name="REV"/>
            </basic>
        </attributes>
    </entity>
    <entity class="org.mag.pub.Author">
        <table name="AUTH"/>
        <attributes>
            <id name="id">
                <column name="AID" column-definition="INTEGER64"/>
                <generated-value strategy="TABLE" generator="AuthorGen"/>
                <table-generator name="AuthorGen" table="AUTH_GEN" 
                    pk-column-name="PK" value-column-name="AID"/>
            </id>
            <basic name="firstName">
                <column name="FNAME"/>
            </basic>
            <basic name="lastName">
                <column name="LNAME"/>
            </basic>
            <version name="version">
                <column name="VERS"/>
            </version>
            ...
        </attributes>
    </entity>
    <entity class="org.mag.subcribe.Contract">
        <table schema="CNTRCT"/>
        <inheritance strategy="JOINED"/>
        <discriminator-column name="CTYPE"/>
        <attributes>
            <basic name="terms">
                <lob/>
            </basic>
            ...
        </attributes>
    </entity>
    <entity class="org.mag.subcribe.Subscription">
        <table name="SUB" schema="CNTRCT"/>
        <inheritance strategy="SINGLE_TABLE"/>
        <discriminator-value>1</discriminator-value>
        <discriminator-column name="KIND" discriminator-type="INTEGER"/>
        <attributes>
            <id name="id">
                <generated-value strategy="IDENTITY"/>
            </id>
            <basic name="payment">
                <column name="PAY"/>
            </basic>
            <basic name="startDate">
                <column name="START"/>
            </basic>
            <version name="version">
                <column name="VERS"/>
            </version>
            ...
        </attributes>
    </entity>
    <entity class="org.mag.subscribe.Subscription.LineItem">
        <table name="LINE_ITEM" schema="CNTRCT"/>
        <primary-key-join-column name="ID" referenced-column-name="PK"/>
        <attributes>
            <basic name="comments">
                <column name="COMM"/>
            </basic>
            <basic name="price"/>
            <basic name="num"/>
            ...
        </attributes>
    </entity>
    <entity class="org.mag.subscribe.LifetimeSubscription" name="Lifetime">
        <discriminator-value>2</discriminator-value>
        <attributes>
            <basic name="eliteClub" fetch="LAZY">
                <column name="ELITE"/>
            </basic>
            ...
        </attributes>
    </entity>
    <entity class="org.mag.subscribe.TrialSubscription" name="Trial">
        <discriminator-value>3</discriminator-value>
        <attributes>
            <basic name="endDate">
                <column name="END"/>
            </basic>
            ...
        </attributes>
    </entity>
</entity-mappings>

8.2. Secondary Tables

Sometimes a logical record is spread over multiple database tables. JPA calls a class' declared table the primary table, and calls other tables that make up a logical record secondary tables. You can map any persistent field to a secondary table. Just write the standard field mapping, then perform these two additional steps:

Set the table attribute of each of the field's columns or join columns to the name of the secondary table.
Define the secondary table on the entity class declaration.

You define secondary tables with the SecondaryTable annotation. This annotation has all the properties of the Table annotation covered in Section 1, “ Table ” , plus a pkJoinColumns property.

The pkJoinColumns property is an array of PrimaryKeyJoinColumns dictating how to join secondary table records to their owning primary table records. Each PrimaryKeyJoinColumn joins a secondary table column to a primary key column in the primary table. See Section 6.2, “ Joined ” above for coverage of PrimaryKeyJoinColumn's properties.

The corresponding XML element is secondary-table. This element has all the attributes of the table element, but also accepts nested primary-key-join-column elements.

In the following example, we move the Article.content field we mapped in Section 8.1, “ Basic Mapping ” into a joined secondary table, like so:

Example 13.11. Secondary Table Field Mapping

package org.mag;

@Entity
@Table(name="ART")
@SecondaryTable(name="ART_DATA", 
    pkJoinColumns=@PrimaryKeyJoinColumn(name="ART_ID", referencedColumnName="ID"))
public class Article {

    @Id private long id;

    @Column(table="ART_DATA")
    private byte[] content;

    ...
}

And in XML:

<entity class="org.mag.Article">
    <table name="ART"/>
    <secondary-table name="ART_DATA">
        <primary-key-join-column name="ART_ID" referenced-column-name="ID"/>
    </secondary-table>
    <attributes>
        <id name="id"/>
        <basic name="content">
            <column table="ART_DATA"/>
        </basic>
        ...
    </attributes>
</entity>

8.3. Embedded Mapping

Chapter 5, Metadata describes JPA's concept of embeddable objects. The field values of embedded objects are stored as part of the owning record, rather than as a separate database record. Thus, instead of mapping a relation to an embeddable object as a foreign key, you map all the fields of the embeddable instance to columns in the owning field's table.

JPA defaults the embedded column names and descriptions to those of the embeddable class' field mappings. The AttributeOverride annotation overrides a basic embedded mapping. This annotation has the following properties:

String name: The name of the embedded class' field being mapped to this class' table.
Column column: The column defining the mapping of the embedded class' field to this class' table.

The corresponding XML element is attribute-override. It has a single name attribute to name the field being overridden, and a single column child element.

To declare multiple overrides, use the AttributeOverrides annotation, whose value is an array of AttributeOverride s. In XML, simply list multiple attribute-override elements in succession.

To override a many to one or one to one relationship, use the AssociationOverride annotation in place of AttributeOverride. AssociationOverride has the following properties:

String name: The name of the embedded class' field being mapped to this class' table.
JoinColumn[] joinColumns: The foreign key columns joining to the related record.

The corresponding XML element is association-override. It has a single name attribute to name the field being overridden, and one or more join-column child elements.

To declare multiple relation overrides, use the AssociationOverrides annotation, whose value is an array of AssociationOverride s. In XML, simply list multiple association-override elements in succession.

Example 13.12. Embedded Field Mapping

In this example, Company overrides the default mapping of Address.street and Address.city. All other embedded mappings are taken from the Address embeddable class.

package org.mag.pub;

@Entity
@Table(name="COMP")
public class Company {

    @Embedded
    @AttributeOverrides({
        @AttributeOverride(name="street", column=@Column(name="STRT")),
        @AttributeOverride(name="city", column=@Column(name="ACITY"))
    })
    private Address address;
    
    ...
}

@Entity
@Table(name="AUTH")
public class Author {

    // use all defaults from Address class mappings
    private Address address;
    
    ...
}

@Embeddable
public class Address {

    private String street;
    private String city;
    @Column(columnDefinition="CHAR(2)")
    private String state;
    private String zip;
}

The same metadata expressed in XML:

<entity class="org.mag.pub.Company">
    <table name="COMP"/>
    <attributes>
        ...
        <embedded name="address">
            <attribute-override name="street">
               <column name="STRT"/>
            </attribute-override>
            <attribute-override name="city">
               <column name="ACITY"/>
            </attribute-override>
        </embedded>
    </attributes>
</entity>
<entity class="org.mag.pub.Author">
    <table name="AUTH"/>
    <attributes>
        <embedded name="address">
            <!-- use all defaults from Address -->
        </embedded>
    </attributes>
</entity>
<embeddable class="org.mag.pub.Address">
    <attributes>
        <basic name="street"/>
        <basic name="city"/>
        <basic name="state">
            <column column-definition="CHAR(2)"/>
        </basic>
        <basic name="zip"/>
    </attributes>
</embeddable>

You can also use attribute overrides on an entity class to override mappings defined by its mapped superclass or table-per-class superclass. The example below re-maps the Document.version field to the Contract table's CVERSION column.

Example 13.13. Mapping Mapped Superclass Field

@MappedSuperclass
public abstract class Document {

    @Column(name="VERS")
    @Version private int version;

    ...
}

@Entity
@Table(schema="CNTRCT")
@Inheritance(strategy=InheritanceType.JOINED)
@DiscriminatorColumn(name="CTYPE")
@AttributeOverride(name="version", column=@Column(name="CVERSION"))
public class Contract
    extends Document {
    ...
}

The same metadata expressed in XML form:

<mapped-superclass class="org.mag.subcribe.Document">
    <attributes>
        <version name="version">
            <column name="VERS">
        </version>
        ...
    </attributes>
</mapped-superclass>
<entity class="org.mag.subcribe.Contract">
    <table schema="CNTRCT"/>
    <inheritance strategy="JOINED"/>
    <discriminator-column name="CTYPE"/>
    <attribute-override name="version">
        <column name="CVERSION"/>
    </attribute-override>
    <attributes>
        ...
    </attributes>
</entity>

8.4. Direct Relations

A direct relation is a non-embedded persistent field that holds a reference to another entity. many to one and one to one metadata field types are mapped as direct relations. Our model has three direct relations: Magazine's publisher field is a direct relation to a Company, Magazine's coverArticle field is a direct relation to Article, and the LineItem.magazine field is a direct relation to a Magazine. Direct relations are represented in the database by foreign key columns:

You typically map a direct relation with JoinColumn annotations describing how the local foreign key columns join to the primary key columns of the related record. The JoinColumn annotation exposes the following properties:

String name: The name of the foreign key column. Defaults to the relation field name, plus an underscore, plus the name of the referenced primary key column.
String referencedColumnName: The name of the primary key column being joined to. If there is only one identity field in the related entity class, the join column name defaults to the name of the identity field's column.
boolean unique: Whether this column is guaranteed to hold unique values for all rows. Defaults to false.

JoinColumn also has the same nullable , insertable, updatable, columnDefinition, and table properties as the Column annotation. See Section 3, “ Column ” for details on these properties.

The join-column element represents a join column in XML. Its attributes mirror the above annotation's properties:

name
referenced-column-name
unique
nullable
insertable
updatable
column-definition
table

When there are multiple columns involved in the join, as when a LineItem references a Magazine in our model, the JoinColumns annotation allows you to specify an array of JoinColumn values. In XML, simply list multiple join-column elements.

Note

OpenJPA supports many non-standard joins. See Section 6, “ Non-Standard Joins ” in the Reference Guide for details.

Example 13.14. Direct Relation Field Mapping

package org.mag;

@Table(name="AUTH")
public class Magazine {

    @Column(length=9)
    @Id private String isbn;
    @Id private String title;

    @OneToOne
    @JoinColumn(name="COVER_ID" referencedColumnName="ID")
    private Article coverArticle;

    @ManyToOne
    @JoinColumn(name="PUB_ID" referencedColumnName="CID")
    private Company publisher;

    ...
}

@Table(name="ART")
public class Article {

    @Id private long id;

    ...
}


package org.mag.pub;

@Table(name="COMP")
public class Company {

    @Column(name="CID")
    @Id private long id;

    ...
}


package org.mag.subscribe;

public class Subscription {
    ...

    @Table(name="LINE_ITEM", schema="CNTRCT")
    public static class LineItem
        extends Contract {

        @ManyToOne
        @JoinColumns({
            @JoinColumn(name="MAG_ISBN" referencedColumnName="ISBN"),
            @JoinColumn(name="MAG_TITLE" referencedColumnName="TITLE")
        })
        private Magazine magazine;

        ...
    }
}

The same metadata expressed in XML form:

<entity class="org.mag.Magazine">
    <table name="MAG"/>
    <id-class="org.mag.Magazine.MagazineId"/>
    <attributes>
        <id name="isbn">
            <column length="9"/>
        </id>
        <id name="title"/>
        <one-to-one name="coverArticle">
            <join-column name="COVER_ID" referenced-column-name="ID"/>
        </one-to-one>
        <many-to-one name="publisher">
            <join-column name="PUB_IC" referenced-column-name="CID"/>
        </many-to-one>
        ...
    </attributes>
</entity>
<entity class="org.mag.Article">
    <table name="ART"/>
    <attributes>
        <id name="id"/>
        ...
    </attributes>
</entity>
<entity class="org.mag.pub.Company">
    <table name="COMP"/>
    <attributes>
        <id name="id">
            <column name="CID"/>
        </id>
        ...
    </attributes>
</entity>
<entity class="org.mag.subscribe.Subscription.LineItem">
    <table name="LINE_ITEM" schema="CNTRCT"/>
    <primary-key-join-column name="ID" referenced-column-name="PK"/>
    <attributes>
        <many-to-one name="magazine">
            <join-column name="MAG_ISBN" referenced-column-name="ISBN"/>
            <join-column name="MAG_TITLE" referenced-column-name="TITLE"/>
        </many-to-one>
        ...
    </attributes>
</entity>

When the entities in a one to one relation join on shared primary key values rather than separate foreign key columns, use the PrimaryKeyJoinColumn(s) annotation or primary-key-join-column elements in place of JoinColumn(s) / join-column elements.

8.5. Join Table

A join table consists of two foreign keys. Each row of a join table associates two objects together. JPA uses join tables to represent collections of entity objects: one foreign key refers back to the collection's owner, and the other refers to a collection element.

one to many and many to many metadata field types can map to join tables. Several fields in our model use join table mappings, including Magazine.articles and Article.authors.

You define join tables with the JoinTable annotation. This annotation has the following properties:

String name: Table name. If not given, the name of the table defaults to the name of the owning entity's table, plus an underscore, plus the name of the related entity's table.
String catalog: Table catalog.
String schema: Table schema.
JoinColumn[] joinColumns: Array of JoinColumn showing how to associate join table records with the owning row in the primary table. This property mirrors the pkJoinColumns property of the SecondaryTable annotation in functionality. See Section 8.2, “ Secondary Tables ” to refresh your memory on secondary tables.

If this is a bidirectional relation (see Section 2.10.1, “ Bidirectional Relations ” ), the name of a join column defaults to the inverse field name, plus an underscore, plus the referenced primary key column name. Otherwise, the join column name defaults to the field's owning entity name, plus an underscore, plus the referenced primary key column name.
JoinColumn[] inverseJoinColumns: Array of JoinColumns showing how to associate join table records with the records that form the elements of the collection. These join columns are used just like the join columns for direct relations, and they have the same naming defaults. Read Section 8.4, “ Direct Relations ” for a review of direct relation mapping.

join-table is the corresponding XML element. It has the same attributes as the table element, but includes the ability to nest join-column and inverse-join-column elements as children. We have seen join-column elements already; inverse-join-column elements have the same attributes.

Here are the join table mappings for the diagram above.

Example 13.15. Join Table Mapping

package org.mag;

@Entity
@Table(name="MAG")
public class Magazine {

    @Column(length=9)
    @Id private String isbn;
    @Id private String title;

    @OneToMany(...)
    @OrderBy
    @JoinTable(name="MAG_ARTS",
        joinColumns={
            @JoinColumn(name="MAG_ISBN", referencedColumnName="ISBN"),
            @JoinColumn(name="MAG_TITLE", referencedColumnName="TITLE")
        },
        inverseJoinColumns=@JoinColumn(name="ART_ID", referencedColumnName="ID"))
    private Collection<Article> articles;

    ...
}

@Entity
@Table(name="ART")
public class Article {

    @Id private long id;

    @ManyToMany(cascade=CascadeType.PERSIST)
    @OrderBy("lastName, firstName")
    @JoinTable(name="ART_AUTHS",
        joinColumns=@JoinColumn(name="ART_ID", referencedColumnName="ID"),
        inverseJoinColumns=@JoinColumn(name="AUTH_ID", referencedColumnName="AID"))
    private Collection<Author> authors;

    ...
}


package org.mag.pub;

@Entity
@Table(name="AUTH")
public class Author {

    @Column(name="AID", columnDefinition="INTEGER64")
    @Id private long id;

    ...
}

The same metadata expressed in XML:

<entity class="org.mag.Magazine">
    <table name="MAG"/>
    <attributes>
        <id name="isbn">
            <column length="9"/>
        </id>
        <id name="title"/>
        <one-to-many name="articles">
            <order-by/>
            <join-table name="MAG_ARTS">
                <join-column name="MAG_ISBN" referenced-column-name="ISBN"/>
                <join-column name="MAG_TITLE" referenced-column-name="TITLE"/>
                <inverse-join-column name="ART_ID" referenced-column-name="ID"/>
            </join-table>
        </one-to-many>
        ...
    </attributes>
</entity>
<entity class="org.mag.Article">
    <table name="ART"/>
        <attributes>
            <id name="id"/>
            <many-to-many name="authors">
                <order-by>lastName, firstName</order-by>
                <join-table name="ART_AUTHS">
                    <join-column name="ART_ID" referenced-column-name="ID"/>
                    <inverse-join-column name="AUTH_ID" referenced-column-name="AID"/>
                </join-table>
            <cascade>
                <cascade-persist/>
            </cascade>
            </many-to-many>
            ...
       </attributes>
</entity>
<entity class="org.mag.pub.Author">
    <table name="AUTH"/>
    <attributes>
        <id name="id">
            <column name="AID" column-definition="INTEGER64"/>
        </id>
        ...
    </attributes>
</entity>

8.6. Bidirectional Mapping

Section 2.10.1, “ Bidirectional Relations ” introduced bidirectional relations. To map a bidirectional relation, you map one field normally using the annotations we have covered throughout this chapter. Then you use the mappedBy property of the other field's metadata annotation or the corresponding mapped-by XML attribute to refer to the mapped field. Look for this pattern in these bidirectional relations as you peruse the complete mappings below:

Magazine.publisher and Company.mags.
Article.authors and Author.articles.

8.7. Map Mapping

All map fields in JPA are modeled on either one to many or many to many associations. The map key is always derived from an associated entity's field. Thus map fields use the same mappings as any one to many or many to many fields, namely dedicated join tables or bidirectional relations. The only additions are the MapKey annotation and map-key element to declare the key field. We covered these additions in Section 2.14, “ Map Key ”.

The example below maps Subscription's map of LineItems to the SUB_ITEMS join table. The key for each map entry is the LineItem's num field value.

Example 13.16. Join Table Map Mapping

package org.mag.subscribe;

@Entity
@Table(name="SUB", schema="CNTRCT")
public class Subscription {

    @OneToMany(cascade={CascadeType.PERSIST,CascadeType.REMOVE})
    @MapKey(name="num")
    @JoinTable(name="SUB_ITEMS", schema="CNTRCT",
        joinColumns=@JoinColumn(name="SUB_ID"),
        inverseJoinColumns=@JoinColumn(name="ITEM_ID"))
    private Map<Long,LineItem> items;

    ...

    @Entity
    @Table(name="LINE_ITEM", schema="CNTRCT")
    public static class LineItem
        extends Contract {

        private long num;

        ...
    }
}

The same metadata expressed in XML:

<entity class="org.mag.subscribe.Subscription">
    <table name="SUB" schema="CNTRCT"/>
    <attributes>
        ...
        <one-to-many name="items">
            <map-key name="num">
            <join-table name="SUB_ITEMS" schema="CNTRCT">
                <join-column name="SUB_ID"/>
                <inverse-join-column name="ITEM_ID"/>
            </join-table>
            <cascade>
                <cascade-persist/>
                <cascade-remove/>
            </cascade>
        </one-to-many>
        ...
    </attributes>
</entity>
<entity class="org.mag.subscribe.Subscription.LineItem">
    <table name="LINE_ITEM" schema="CNTRCT"/>
    <attributes>
        ...
        <basic name="num"/>
        ...
    </attributes>
</entity>