Hibernate一对多双向关联及inverse的作用

                 Hibernate一对多双向关联及inverse的作用

       在测试Hibernate的一对多双向关联映射时，碰到很有趣的问题，跟inverse属性直接相关。

1、People.hbm.xml

< hibernate-mapping   default-lazy ="false" >   
    < class  name ="com.persistent.People"  table ="people" >      
      < id  name ="id"  column ="peopleId"  unsaved-value ="0" >     
         < generator  class ="increment" >         
         </ generator >     
      </ id >      
      < property  name ="name"  column ="name" ></ property >      
      < set  name ="addresses"  cascade ="save-update" >
         < key  column ="peopleId"  not-null ="true"   />
         < one-to-many  class ="com.persistent.Address" />
      </ set >      
    </ class >
</ hibernate-mapping >

2、Address.hbm.xml

< hibernate-mapping >
   < class  name ="com.persistent.Address"  table ="address" >   
     < id  name ="id"  column ="addressId"  unsaved-value ="0" >
         < generator  class ="increment" >
         </ generator >
     </ id >     
     < many-to-one  name ="people"  column ="peopleId"  insert ="false"  update ="false" ></ many-to-one >     
     < property  name ="addressName"  column ="addressName" ></ property >     
     < property  name ="codeNumber"  column ="codeNumber" ></ property >     
   </ class >   
</ hibernate-mapping >

3、People.java和Address.java

public   class  People  ... {    
  private long id;
  private String name;
  private Set addresses = new HashSet();
  ...
}

public   class  Address  ... {    
  private long id;
  private People people;
  private String addressName;
  private String codeNumber;
  ...
}         

4、数据库结构

  people表：{peopleId,name}

  address表：{addressId,peopleId,addressName,codeNumber}

5、测试代码

People people  =   new  People();
people.setName( "linda " );
Address address  =   new  Address();
address.setAddressName( "yunnan " );
address.setCodeNumber( " 564123 " );
address.setPeople(people);
people.getAddresses().add(address);
Session session  =  HibernateSessionFactory.getSession();
session.beginTransaction();
session.save(people);
session.getTransaction().commit();

6、运行结果

      上面测试代码运行起来正确：

Hibernate: select max(peopleId) from people
Hibernate: select max(addressId) from address
Hibernate: insert into people (name, peopleId) values (?, ?)
Hibernate: insert into address (addressName, codeNumber, peopleId, addressId) values (?, ?, ?, ?)
Hibernate: update address set peopleId=? where addressId=?

      如果将People.hbm.xml映射改写一下：

< set  name ="addresses"  cascade ="save-update"  inverse ="true" >
         < key  column ="peopleId"  not-null ="true"   />
         < one-to-many  class ="com.persistent.Address" />
</ set >

      不同之处在于添加了inverse="true"，结果：
Hibernate: select max(peopleId) from people
Hibernate: select max(addressId) from address
Hibernate: insert into people (name, peopleId) values (?, ?)
Hibernate: insert into address (addressName, codeNumber, addressId) values (?, ?, ?)

      可以看到，peopleId并没有写入到关联的address当中，数据库address表中相应记录的peopleId字段为空。

7、分析

      在Hibernate中，术语inverse是反转的意思，在关联关系中，inverse="false"为主控方，由主控方负责维护对象的关联关系。所以上面的映射文件改动之后，address为主控方，people为被控方，但是测试代码只进行了一个保存操作session.save(people)，这是针对people的，因此无法正确级联保存address。而原来的映射文件中（虽然没有明确指明，Hibernate默认inverse="false"），people为主控方，因此保存people时它会保证关联的address的正确保存。

      也就是说，Hibernate仅仅按照主控方对象的状态的变化来同步更新数据库。按照原来的映射文件，people.getAddresses().add(address)，即主控方对象的状态发生了改变，因此数据库会跟着对象状态的变化来同步更新数据库；而address.setPeople(people)，即被控方对象的状态发生了改变，它是不能触发对象和数据库的同步更新的。