在hibernate注解(三)中,我提高过一对一(@OneToOne)懒加载失效的问题。虽然给出了解决方法,但并没有给出完整的解决方案。今天我专门针对该问题进行讨论。至于懒加载失效的原因,在之前的文章中已经我已经叙述过了,就不再重复了,不明白的可以去看看。
数据库:myqsl
代码:主:Student,从:Card
表:
DROP TABLE IF EXISTS `student`; CREATE TABLE `student` ( `ID` int(11) NOT NULL, `NAME` varchar(50) NOT NULL, `CARD_ID` int(11) DEFAULT NULL, PRIMARY KEY (`ID`), KEY `PK_CARD_ID` (`CARD_ID`), CONSTRAINT `PK_CARD_ID` FOREIGN KEY (`CARD_ID`) REFERENCES `card` (`ID`) ON DELETE NO ACTION ON UPDATE NO ACTION ) ENGINE=InnoDB DEFAULT CHARSET=utf8; DROP TABLE IF EXISTS `card`; CREATE TABLE `card` ( `ID` int(11) NOT NULL, `CODE` varchar(32) NOT NULL, PRIMARY KEY (`ID`) ) ENGINE=InnoDB DEFAULT CHARSET=utf8;
代码:
package com.po; import javax.persistence.CascadeType; import javax.persistence.Column; import javax.persistence.Entity; import javax.persistence.FetchType; import javax.persistence.GeneratedValue; import javax.persistence.GenerationType; import javax.persistence.Id; import javax.persistence.JoinColumn; import javax.persistence.OneToOne; import javax.persistence.Table; @Entity @Table(name = "Student") public class Student { private int id; private String name; private Card card; @Id @GeneratedValue(strategy = GenerationType.IDENTITY) @Column(name = "ID", unique = true, nullable = false) public int getId() { return id; } public void setId(int id) { this.id = id; } @Column(name = "NAME", nullable = false, length = 50) public String getName() { return name; } public void setName(String name) { this.name = name; } @OneToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL) @JoinColumn(name = "CARD_ID") public Card getCard() { return card; } public void setCard(Card card) { this.card = card; } }
package com.po; import javax.persistence.CascadeType; import javax.persistence.Column; import javax.persistence.Entity; import javax.persistence.FetchType; import javax.persistence.GeneratedValue; import javax.persistence.GenerationType; import javax.persistence.Id; import javax.persistence.OneToOne; import javax.persistence.Table; @Entity @Table(name = "card") public class Card { private int id; private String code; private Student student; @Id @GeneratedValue(strategy = GenerationType.IDENTITY) @Column(name = "ID", unique = true, nullable = false) public int getId() { return id; } public void setId(int id) { this.id = id; } @Column(name = "CODE", length = 32, nullable = false) public String getCode() { return code; } public void setCode(String code) { this.code = code; } @OneToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL, mappedBy = "card") public Student getStudent() { return student; } public void setStudent(Student student) { this.student = student; } }
在card表增加一个student表的外键字段STUDENT_ID,并在Card类的@OneToOne下增加@JoinColumn(name = "STUDENT_ID"),去掉mappedBy = "card",即
DROP TABLE IF EXISTS `card`; CREATE TABLE `card` ( `ID` int(11) NOT NULL, `CODE` varchar(32) NOT NULL, `STUDENT_ID` int(11) DEFAULT NULL, PRIMARY KEY (`ID`), KEY `PK_STUDENT_ID` (`STUDENT_ID`), CONSTRAINT `PK_STUDENT_ID` FOREIGN KEY (`STUDENT_ID`) REFERENCES `student` (`ID`) ON DELETE NO ACTION ON UPDATE NO ACTION ) ENGINE=InnoDB DEFAULT CHARSET=utf8;
public class Card { // ... 略 @OneToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL) @JoinColumn(name = "STUDENT_ID") public Student getStudent() { return student; } // ... 略 }
优点:不改变Student与Card在代码中的对应关系(一对一)
缺点:需要同时维护Student和Card的两个外键。
改为主键关联。
DROP TABLE IF EXISTS `student`; CREATE TABLE `student` ( `ID` int(11) NOT NULL, `NAME` varchar(50) NOT NULL, PRIMARY KEY (`ID`), CONSTRAINT `PK_CARD_ID` FOREIGN KEY (`ID`) REFERENCES `card` (`ID`) ON DELETE NO ACTION ON UPDATE NO ACTION ) ENGINE=InnoDB DEFAULT CHARSET=utf8;
public class Student { // ... 略 @Id @GenericGenerator(name = "PK_Card", strategy = "foreign", parameters = @Parameter(name = "property", value = "card")) @GeneratedValue(generator = "PK_Card") @Column(name = "ID", unique = true, nullable = false) public int getId() { return id; } // ... 略 @OneToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL, optional = false) @PrimaryKeyJoinColumn public Card getCard() { return card; } // ... 略 }
public class Card { // ... 略 @OneToOne(fetch = FetchType.LAZY, cascade = CascadeType.ALL, mappedBy = "card", optional = false) public Student getStudent() { return student; } // ... 略 }
除了改变student表的主键、外键结构外,Student类和Card类也要做相应修改,尤其注意“optional”,要设置为false,否则无法实现懒加载。
优点:不改变Student与Card在代码中的对应关系(一对一)
缺点:改动较大,且使用主键关联具有局限性。
PS:主键关联的局限性
使用主键关联会影响数据存储结构,主键关联是一种强耦合,以上述为例:Card存在时,Student才能存在,Card消亡时,Student也随之消失。这是因为Student的主键依赖于Card主键,Student无法独立存在(就是说必须先有学生卡,才能有学生)。
将Card类中的OneToOne改为OneToMany(一对多)。
public class Card { private Set students; // ... 略 @OneToMany(fetch = FetchType.LAZY, cascade = CascadeType.ALL, mappedBy = "card") public Set<Student> getStudents() { return students; } public void setStudents(Set students) { this.students = students; } // ... 略 }
优点:数据库不用修改
缺点:需要修改Student与Card在代码中的对应关系
放弃用注解的方式,改为Xml方式来实现hibernate模型设计,并在Card Xml的OneToOne标签中添加constrained属性,靠注解解决的办法已经没有了(instrument增强就算了吧,很麻烦)。
最后,我们来评估下以上方案的可行性。
方案一:从可读性来讲,是最容易理解的,但需要维护两个外键,如果程序控制不好的话,容易出问题,即关联错误。
方案二:主键关联虽然有些约束,但也取决于业务需求,比如订单和订单详情,采用主键关联也挺合适的,只是不适合相对灵活的对象关系。
方案三:改动在我看来是最小的了,牺牲了一定的可读性(关系从Card角度看变为了一对多),我个人比较喜欢该种方案,因此推荐。
方案四:如果不采用注解,而采用Xml的话,我是很推荐这种方案的,注解虽然优点多,也趋于主流,但最传统的Xml,功能还是最强大的。但如果你仅为了解决该问题,而将注解和Xml混合使用的话,我建议你还是放弃吧。