Hibernate.orgCommunity Documentation

Chapter 8. Association Mappings

8.1. Introduction
8.2. Unidirectional associations
8.2.1. Many-to-one
8.2.2. One-to-one
8.2.3. One-to-many
8.3. Unidirectional associations with join tables
8.3.1. One-to-many
8.3.2. Many-to-one
8.3.3. One-to-one
8.3.4. Many-to-many
8.4. Bidirectional associations
8.4.1. one-to-many / many-to-one
8.4.2. One-to-one
8.5. Bidirectional associations with join tables
8.5.1. one-to-many / many-to-one
8.5.2. one to one
8.5.3. Many-to-many
8.6. More complex association mappings

Association mappings are often the most difficult thing to implement correctly. In this section we examine some canonical cases one by one, starting with unidirectional mappings and then bidirectional cases. We will use Person and Address in all the examples.

Associations will be classified by multiplicity and whether or not they map to an intervening join table.

Nullable foreign keys are not considered to be good practice in traditional data modelling, so our examples do not use nullable foreign keys. This is not a requirement of Hibernate, and the mappings will work if you drop the nullability constraints.

A bidirectional many-to-one association is the most common kind of association. The following example illustrates the standard parent/child relationship.


<class name="Person">
    <id name="id" column="personId">
        <generator class="native"/>
    </id>
    <many-to-one name="address" 
        column="addressId"
        not-null="true"/>
</class>

<class name="Address">
    <id name="id" column="addressId">
        <generator class="native"/>
    </id>
    <set name="people" inverse="true">
        <key column="addressId"/>
        <one-to-many class="Person"/>
    </set>
</class>
create table Person ( personId bigint not null primary key, addressId bigint not null )
create table Address ( addressId bigint not null primary key )
        

If you use a List, or other indexed collection, set the key column of the foreign key to not null. Hibernate will manage the association from the collections side to maintain the index of each element, making the other side virtually inverse by setting update="false" and insert="false":


<class name="Person">
   <id name="id"/>
   ...
   <many-to-one name="address"
      column="addressId"
      not-null="true"
      insert="false"
      update="false"/>
</class>

<class name="Address">
   <id name="id"/>
   ...
   <list name="people">
      <key column="addressId" not-null="true"/>
      <list-index column="peopleIdx"/>
      <one-to-many class="Person"/>
   </list>
</class>

If the underlying foreign key column is NOT NULL, it is important that you define not-null="true" on the <key> element of the collection mapping. Do not only declare not-null="true" on a possible nested <column> element, but on the <key> element.

More complex association joins are extremely rare. Hibernate handles more complex situations by using SQL fragments embedded in the mapping document. For example, if a table with historical account information data defines accountNumber, effectiveEndDate and effectiveStartDatecolumns, it would be mapped as follows:


<properties name="currentAccountKey">
    <property name="accountNumber" type="string" not-null="true"/>
    <property name="currentAccount" type="boolean">
        <formula>case when effectiveEndDate is null then 1 else 0 end</formula>
    </property>
</properties>
<property name="effectiveEndDate" type="date"/>
<property name="effectiveStateDate" type="date" not-null="true"/>

You can then map an association to the current instance, the one with null effectiveEndDate, by using:


<many-to-one name="currentAccountInfo"
        property-ref="currentAccountKey"
        class="AccountInfo">
    <column name="accountNumber"/>
    <formula>'1'</formula>
</many-to-one>

In a more complex example, imagine that the association between Employee and Organization is maintained in an Employment table full of historical employment data. An association to the employee's most recent employer, the one with the most recent startDate, could be mapped in the following way:


<join>
    <key column="employeeId"/>
    <subselect>
        select employeeId, orgId 
        from Employments 
        group by orgId 
        having startDate = max(startDate)
    </subselect>
    <many-to-one name="mostRecentEmployer" 
            class="Organization" 
            column="orgId"/>
</join>

This functionality allows a degree of creativity and flexibility, but it is more practical to handle these kinds of cases using HQL or a criteria query.