学步园

Bag（结合了List与Set），可以重复且没有顺序的一种集合，是Hibernate提供的。HIbernate使用jdk的List模拟Bag。其配置与Hibernate映射List时基本相同。

以Team、Student为例：

持久化类Team（与映射List<Student>时相同）：

package bean;

import java.util.List;

public class Team {
	private long id;
	private String name;
	private List<Student> students;
        //省略get、set方法
}

持久化类Student（与映射List<Student>时相同）：

package bean;

public class Student {
	private long id;
	private long cardID;
	private String name;
	private int age;
	private Team team;
       //省略get、set方法
}

对象关系映射文件Team.hbm.xml:

<hibernate-mapping>
	<class name="bean.Team" table="team">
		<id name="id" column="id" type="long">
			<generator class="increment"></generator>
		</id>
		<property name="name" column="name" type="string"></property>
		<bag name="students" cascade="all" table="students">
			<key column="team_id"></key>
			 <one-to-many class="bean.Student"/> 
		</bag>
	</class>
</hibernate-mapping>

Student.hbm.xml(与映射List<Student>时相同)：

<hibernate-mapping>
	<class name="bean.Student" table="students">
		<id name="id" column="id" type="long">
			<generator class="increment"></generator>
		</id>
		<property name="name" column="name" type="string"></property>
		<property name="age" column="age" type="int"></property>
		<property name="cardID" column="cardID" type="long"></property>
        <many-to-one name="team" class="bean.Team" column="team_id" cascade="none"></many-to-one>
	</class>
</hibernate-mapping>

并将它们加入到主配置文件中：

    <mapping resource="Team.hbm.xml"/>
    <mapping resource="Student.hbm.xml"/> 

说明：

Hibernate依然将Team的List中的元素放置到另一张表中（此处即students表），与映射List不同的是<bag>中不需要指定index索引（因为Bag中的元素是可重复并且无续的），team表和students表结构如下：

级联保存Team对象：

              tx=session.beginTransaction();
              Team t1=new Team();
              t1.setName("team1");
              t1.setStudents(new ArrayList<Student>());
              t1.getStudents().add(new Student(1001L,23,"zhangsan"));
              t1.getStudents().add(new Student(1002L,23,"lisi"));
              t1.getStudents().add(new Student(1001L,23,"zhangsan"));           
              session.save(t1);
              tx.commit();

此时输出的SQL语句为：

Hibernate: select max(id) from team
Hibernate: select max(id) from students
Hibernate: insert into team (name, id) values (?, ?)
Hibernate: insert into students (name, age, cardID, team_id, id) values (?, ?, ?, ?, ?)
Hibernate: insert into students (name, age, cardID, team_id, id) values (?, ?, ?, ?, ?)
Hibernate: insert into students (name, age, cardID, team_id, id) values (?, ?, ?, ?, ?)
Hibernate: update students set team_id=? where id=?
Hibernate: update students set team_id=? where id=?
Hibernate: update students set team_id=? where id=?

可以看出执行流程是先取一个默认值（实际为NULL）给team_id向students表中插入一条记录，然后再执行update语句更新students表中的对应记录的team_id字段值。

若将Team.hbm.xml中<bag>元素的inverse属性值改为true：

		<bag name="students" cascade="save-update" table="students" inverse="true"><!--指定由students表维持两表之间的关联关系-->

然后再执行上述保存Team对象的语句，此时输出的SQL语句为：

Hibernate: select max(id) from team
Hibernate: select max(id) from students
Hibernate: insert into team (name, id) values (?, ?)
Hibernate: insert into students (name, age, cardID, team_id, id) values (?, ?, ?, ?, ?)
Hibernate: insert into students (name, age, cardID, team_id, id) values (?, ?, ?, ?, ?)
Hibernate: insert into students (name, age, cardID, team_id, id) values (?, ?, ?, ?, ?)

对比可知，并没有执行上述的update语句，原因就在于是students表维持两表之间的关联关系而不再是Team表，通过对比以下students表中的前三条记录和后三条记录就可以理解inverse的作用，并且Bag存放的元素是可以重复及无续的：

转载请注明出处：http://blog.csdn.net/jialinqiang/article/details/8688481