JPA中的實體層次設計

這部分的內容基本與Hibernate一致.JPA同樣支援3種類型的繼承形式:

1.Single Table Strategy ,單表策略,一張表包含基類與子類的所有資料,很多情況下都是採用這樣的冗餘設計,通過一個discriminator來區分

2.Table Per Class Strategy ,每個子類對應一張表,每張表都擁有基類的屬性

3.Join Strategy ,仍然是每個子類對應一張表，但此表中不包含基類的屬性,僅僅是此子類的擴充屬性,共用基類的屬性

以一個例子來說明3種情況:

一.單表策略

比如Pet作為基類,Cat和Dog繼承此類並擁有自己的擴充屬性,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類值的注意的就是通過@Inheritance(strategy = InheritanceType.SINGLE_TABLE)確定採用單表策略,通過@DiscriminatorColumn確定了標誌值的欄位和類型，我想熟悉hibernate的朋友對這些都應該很熟悉.然後是兩個子類:

//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;
 }

}

兩個子類最值的關注的就是@DiscriminatorValue注釋,比如Cat的此值為cat,意味著當Cat類型的Entity存入資料庫時,JPA將自動把cat的值賦給animal_type欄位,Dog的值則為dog,由此就可以在同一張表中區分開兩個不同的子類.

二.Table per Class

採用Table Per Class策略的話，每個子類都將單獨建表，並且都獨立擁有基類中的所有屬性,互相之間不共用,在我們的例子中所要進行的修改很小,像這樣:

//基類

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

 private String name;

 private double weight;

........

//子類:不需要任何設定

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

  .......

  .......

三.Join策略

每個子類同樣獨立建表,基類也獨立建表，只不過所有的子類的表中只有擴充屬性,他們共用基類的表,在我們的例子中修改下即可:

//基類

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

 private String name;

 private double weight;

  ........

//子類

@Entity

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

  .......

  .......

這部分的內容實在沒什麼新意,與hibernate完全一致.JAVA EE5向spring和hibernate借鑒了太多東西.