hibernate基于连接表的多对多双向关联

sql脚本:

--创建表
create table Person4 ( personId number(10) not null primary key ,personName NVARCHAR2(100));  
create table PersonAddress4 ( personId4 number(10) not null, addressId4 number(10) not null,primary key
(personId4, addressId4));
create table Address4 ( addressId number(10) not null primary key,addressName NVARCHAR2(100));

person.hbm.xml:

<class name="com.chenjun.eshop.staffManage.domain.Person"
		table="Person4">

		<id name="id" column="personId">
			<generator class="assigned"><!-- 由程序分配主键 -->

			</generator>
		</id>

		<property name="personName"></property>

		<!--  personId4,addressId4都是连接表PersonAddress4里的字段-->
		<set name="addresses" table="PersonAddress4" cascade="all">
			<key column="personId4" />
			<!-- unique="true" 表示当前实体是一方，不是多方,也就是person是一 -->
			<many-to-many column="addressId4" 	class="com.chenjun.eshop.staffManage.domain.Address" />
		</set>

	</class>

address.hbm.xml:

<class name="com.chenjun.eshop.staffManage.domain.Address"
		table="Address4">

		<id name="id" column="addressId"><!-- addressid引用的是person表的主键personid，也就是两表共用主键 -->

			<generator class="assigned"><!-- 由程序分配主键 -->

			</generator>
		</id>

		<property name="addressName"></property>
		
		<set name="persones" inverse="true" table="PersonAddress4" cascade="all">
			<key column="addressId4" />
			<many-to-many column="personId4" class="com.chenjun.eshop.staffManage.domain.Person" />
		</set>
	</class>

测试代码:

private void addPerson()
    {

        Person person1 = new Person();
        person1.setId(1);
        person1.setPersonName("王五");

        Person person2 = new Person();
        person2.setId(2);
        person2.setPersonName("赵六");

        Address address1 = new Address();
        address1.setId(2001);
        address1.setAddressName("上海");

        Address address2 = new Address();
        address2.setId(2002);
        address2.setAddressName("北京");

        person1.getAddresses().add(address1);
        person1.getAddresses().add(address2);
        person2.getAddresses().add(address1);
        person2.getAddresses().add(address2);

        address1.getPersones().add(person1);//添加关联表数据
        address1.getPersones().add(person2);
        address2.getPersones().add(person1);
        address2.getPersones().add(person2);

        Session session = this.getHibernateTemplate().getSessionFactory().openSession();
        session.beginTransaction();

        session.save(person1);
        session.save(person2);
     
         //session.save(address1); //这种保存方法也可以
         //session.save(address2);
 session.beginTransaction().commit();
        session.close();
    }

sql日志:

 insert into Person4 (personName, personId) values (?, ?)

insert into Address4 (addressName, addressId) values (?, ?)
insert into Address4 (addressName, addressId) values (?, ?)

 insert into Person4 (personName, personId) values (?, ?)

insert into PersonAddress4 (personId4, addressId4) values (?, ?)
insert into PersonAddress4 (personId4, addressId4) values (?, ?)
insert into PersonAddress4 (personId4, addressId4) values (?, ?)
insert into PersonAddress4 (personId4, addressId4) values (?, ?)

上面的执行顺序为什么不是像下面这样的执行顺序,先保存persn,再保存address,再保存person,再保存address？谁能告诉我下

以上面注释方式来保存的sql日志:

 insert into Address4 (addressName, addressId) values (?, ?)
 insert into Person4 (personName, personId) values (?, ?)

 insert into Address4 (addressName, addressId) values (?, ?)
 insert into Person4 (personName, personId) values (?, ?)

 insert into PersonAddress4 (personId4, addressId4) values (?, ?)
 insert into PersonAddress4 (personId4, addressId4) values (?, ?)
 insert into PersonAddress4 (personId4, addressId4) values (?, ?)
 insert into PersonAddress4 (personId4, addressId4) values (?, ?)