Hibernate/JPA中的继承映射 实体extends的关系(每个子类独立一个表)

Hibernate/JPA中的继承映射 实体extends的关系(每个子类独立一个表)

http://blog.csdn.net/skytalemcc/archive/2010/01/15/5196286.aspx

 

 

Hibernate的继承映射包含了三种不同的策略：

每簇类使用一个表；
每个子类一个表；
每个具体内一个表（有限制）。
假设我们有四个类Animal，Dog，Cat，其代码如下：
文件名：Animal.java

class Animal {
    private String identifier;
    private String name;
    private String category;
    // setter and getter
}

文件名：Dog.java

class Dog extends Animal {
    private String
    // setter and getter
}

文件名：Cat.java

class Cat extends Animal {
    private String
    // setter and getter
}
每簇类使用一个表
       使用每簇类使用一个表的策略时，有一个限制就时子类不能有NOT NULL，映射文件为：
       文件名：Animal.hbm.xml

       <class name="Animal" table="TB_ANIMAL">
          <id name="identifier" type="string" column="IDENTIFIER">
             <generator class="uuid.hex"/>
          </id>
          <discriminator column="ANIMAL_TYPE" type="string"/>
          <property name="name" column="NAME" type="string"/>
         
          <subclass name="Dog" discriminator-value="DOG">
            
          </subclass>
          <subclass name="Cat" discriminator-value="CAT">
            
          </subclass>
       </class>
每个子类一个表
       使用每个子类一个表的策略时，可以使用一个映射文件实现，也可以分成多个映射文件来实现。每个子类一个映射文件的情况：
       文件名：Animal.hbm.xml

       <class name="Animal" table="ANIMAL">
          <id name="identifier" column="IDENTIFIER" type="string">
             <generator class="uuid.hex"/>
          </id>
          <property >
       </class>
        文件名：Dog.hbm.xml
       <joined-subclass name="Dog" table="DOG" extends="Animal">
          <key column="DOG_ID"/>
         
       </joined-subclass>
        文件名：Cat.hbm.xml
       <joined-subclass name="Cat" table="CAT" extends="Cat">
          <key column="CAT_ID"/>
         
       </joined-subclass>

       每个子类一个表的策略实际上一种one-to-one的映射。

每个具体内一个表（有限制）
       使用每个具体内一个表（有限制）策略时，每一个子类的映射文件将要包含所有父类中的属性，映射文件：
       文件名：Dog.hbm.xml

       <class name="Dog" table="DOG">
          <id name="identifier" column="IDENTIFIER" type="string">
             <generator class="uuid.hex"/>
          </id>
          <property name="name" column="NAME" type="string"/>
         
       </class>
        文件名：Cat.hbm.xml
       <class name="Cat" table="CAT">
          <id name="identifier" column="IDENTIFIER" type="string">
             <generator class="uuid.hex"/>
          </id>
          <property name="name" column="NAME" type="string"/>
         
       </class>

JPA中的实体层次设计
这部分的内容基本与Hibernate一致.JPA同样支持3种类型的继承形式:

1.Single Table Strategy ,单表策略,一张表包含基类与子类的所有数据,很多情况下都是采用这样的冗余设计,通过一个discriminator来区分

2.Table Per Class Strategy ,每个子类对应一张表,每张表都拥有基类的属性

3.Join Strategy ,仍然是每个子类对应一张表，但此表中不包含基类的属性,仅仅是此子类的扩展属性,共享基类的属性

以一个例子来说明3种情况:

一.单表策略

比如Pet作为基类,Cat和Dog继承此类并拥有自己的扩展属性,如:

package com.denny_blue.ejb3.inheritance;

import java.io.Serializable;

import javax.persistence.DiscriminatorColumn;
import javax.persistence.DiscriminatorType;
import javax.persistence.Entity;
import javax.persistence.GeneratedValue;
import javax.persistence.GenerationType;
import javax.persistence.Id;
import javax.persistence.Inheritance;
import javax.persistence.InheritanceType;

@Entity
@Inheritance(strategy = InheritanceType.SINGLE_TABLE)
@DiscriminatorColumn(name = "animal_type", discriminatorType = DiscriminatorType.STRING)
public class Pet implements Serializable {
private int id;

private String name;

private double weight;

public Pet() {
}

@Id
@GeneratedValue(strategy = GenerationType.AUTO)
public int getId() {
   return id;
}

public void setId(int id) {
   this.id = id;
}

public String getName() {
   return name;
}

public void setName(String name) {
   this.name = name;
}

public double getWeight() {
   return weight;
}

public void setWeight(double weight) {
   this.weight = weight;
}

}

Pet类值的注意的就是通过@Inheritance(strategy = InheritanceType.SINGLE_TABLE)确定采用单表策略,通过@DiscriminatorColumn确定了标志值的字段和类型，我想熟悉hibernate的朋友对这些都应该很熟悉.然后是两个子类:

//Cat.java

package com.denny_blue.ejb3.inheritance;

import javax.persistence.DiscriminatorColumn;
import javax.persistence.DiscriminatorType;
import javax.persistence.DiscriminatorValue;
import javax.persistence.Entity;
import javax.persistence.Inheritance;
import javax.persistence.InheritanceType;

@Entity
@Inheritance(strategy = InheritanceType.SINGLE_TABLE)
@DiscriminatorColumn(discriminatorType = DiscriminatorType.STRING)
@DiscriminatorValue("cat")
public class Cat extends Pet {
private String HairBall;

public String getHairBall() {
   return HairBall;
}

public void setHairBall(String hairBall) {
   HairBall = hairBall;
}

}

//Dog.java

package com.denny_blue.ejb3.inheritance;

import javax.persistence.DiscriminatorColumn;
import javax.persistence.DiscriminatorType;
import javax.persistence.DiscriminatorValue;
import javax.persistence.Entity;
import javax.persistence.Inheritance;
import javax.persistence.InheritanceType;

@Entity
@Inheritance(strategy=InheritanceType.SINGLE_TABLE)
@DiscriminatorColumn(discriminatorType=DiscriminatorType.STRING)
@DiscriminatorValue("dog")
public class Dog extends Pet {
private String trick;

public String getTrick() {
   return trick;
}

public void setTrick(String trick) {
   this.trick = trick;
}

}

两个子类最值的关注的就是@DiscriminatorValue注释,比如Cat的此值为cat,意味着当Cat类型的Entity存入数据库时,JPA将自动把cat的值赋给animal_type字段,Dog的值则为dog,由此就可以在同一张表中区分开两个不同的子类.

二.Table per Class

采用Table Per Class策略的话，每个子类都将单独建表，并且都独立拥有基类中的所有属性,互相之间不共享,在我们的例子中所要进行的修改很小,像这样:

//基类

@Entity
@Inheritance(strategy = InheritanceType.TABLE_PER_CLASS)
public class Pet implements Serializable {
private int id;

private String name;

private double weight;

........

//子类:不需要任何设置

@Entity
public class Dog extends Pet {
private String trick;

.......

.......

}

例：

@Entity
@Table(name = "TEST_A")
@Inheritance(strategy = InheritanceType.TABLE_PER_CLASS)
public class TestA implements java.io.Serializable {

// Fields

/**
*
*/
private static final long serialVersionUID = 5931010626625178698L;
private Long id;
private String name;
private String code;

// Constructors
public TestA() {

}

/** minimal constructor */
public TestA(Long id) {
   this.id = id;
}

/** full constructor */
public TestA(Long id, String name, String code) {
   this.id = id;
   this.name = name;
   this.code = code;
}

// Property accessors
@Id
@Column(name = "ID")
@GeneratedValue(strategy = GenerationType.AUTO, generator = "SEQ")
@SequenceGenerator(name="SEQ", sequenceName = "SEQ_TEMP_VOICE_FILE")
public Long getId() {
   return this.id;
}

public void setId(Long id) {
   this.id = id;
}

@Column(name = "NAME", unique = false, nullable = true, insertable = true, updatable = true, length = 200)
public String getName() {
   return this.name;
}

public void setName(String name) {
   this.name = name;
}

@Column(name = "CODE", unique = false, nullable = true, insertable = true, updatable = true, length = 200)
public String getCode() {
   return this.code;
}

public void setCode(String code) {
   this.code = code;
}

}

@Entity
@Table(name = "TEST_B")
public class TestB extends TestA {

// Fields

/**
*
*/
private static final long serialVersionUID = 4942564072409132522L;

/** minimal constructor */
public TestB() {

}
public TestB(Long id) {
   this.setId(id);
}

@Column(name = "FLAG", unique = false, nullable = true, insertable = true, updatable = true, precision = 1, scale = 0)
public Long getFlag() {
   return this.flag;
}

public void setFlag(Long flag) {
   this.flag = flag;
}

}

三.Join策略

每个子类同样独立建表,基类也独立建表，只不过所有的子类的表中只有扩展属性,他们共享基类的表,在我们的例子中修改下即可:

//基类

@Entity
@Inheritance(strategy = InheritanceType.JOINED)
public class Pet implements Serializable {
private int id;

private String name;

private double weight;

........

//子类

@Entity

@Inheritance(strategy = InheritanceType.JOINED)
public class Dog extends Pet {
private String trick;

 

本文来自CSDN博客，转载请标明出处：http://blog.csdn.net/skytalemcc/archive/2010/01/15/5196286.aspx