对象映射关系分为单向关系和双向关系,单向关系只在一方对象上存在对方对象,双向关系是在双方对象上存在彼此对象。

一、单向关系

public class Department implements Serializable {
    @Id
    @GeneratedValue(strategy = GenerationType.IDENTITY)
    private Integer departmentId;

    private String departmentCode;
    
    @OneToMany(cascade = {CascadeType.ALL})
    @JoinColumn
    private List<Employee> employeeList;
}

在没有@JoinColumn时,将多增加一个中间关系表,由此表来维护两个对象关系,增加之后只有两个表,由Employee表维护关系。默认情况下jpa会使用主键来做关联,并在子表中增加外键约束。

二、单向关系使用code关联

在设计表结构时,主键一般会使用自增ID,但在做子表关联时由于分布式结构原因不想使用自增ID来做关系维护,则可自定义字段code来维护关系,如下:

public class Department implements Serializable {
    @Id
    @GeneratedValue(strategy = GenerationType.IDENTITY)
    private Integer departmentId;

    private String departmentCode;

    @OneToMany(cascade = {CascadeType.ALL})
    @JoinColumn(name = "departmentCode", referencedColumnName = "departmentCode")
    private List<Employee> employeeList;
}

在employee表中会增加一个department_code字段来维护关联关系。
注意:从add操作中的SQL可以看出,employee是先insert之后,再去update关系字段的,多一步update。

Hibernate: insert into department (department_code, department_name) values (?, ?)
Hibernate: insert into employee (employee_code, employee_name) values (?, ?)
Hibernate: insert into employee (employee_code, employee_name) values (?, ?)
Hibernate: update employee set department_code=? where employee_id=?
Hibernate: update employee set department_code=? where employee_id=?

三、双向关系

public class Department implements Serializable {
    @Id
    @GeneratedValue(strategy = GenerationType.IDENTITY)
    private Integer departmentId;

    private String departmentCode;

    @OneToMany(cascade = {CascadeType.ALL}, mappedBy = "department")
    private List<Employee> employeeList;
}

public class Employee {
    @Id
    @GeneratedValue(strategy = GenerationType.IDENTITY)
    private Integer employeeId;

    private String employeeCode;

    @ManyToOne
    private Department department;
}

使用mappedBy将去掉中间关系表,由employee维护department的关系。

注意事项
1.在add操作中子对象employee必须设置主对象,否则数据库层面是没有维护外键关系的,如:

@Test
public void add(){
    Department department = Department.builder()
            .departmentCode("D001")
            .departmentName("部门1")
            .build();

    Employee employee = Employee.builder()
            .employeeCode("E001")
            .employeeName("员工1")
            .build();
    // 如无此操作,将无关联关系        
    employee.setDepartment(department);
    
    List<Employee> employeeList = new ArrayList<>();
    employeeList.add(employee);

    department.setEmployeeList(employeeList);
    departmentRepository.save(department);
}

2.此种设置是不需要额外update关系的

Hibernate: insert into department (department_code, department_name) values (?, ?)
Hibernate: insert into employee (department_department_id, employee_code, employee_name) values (?, ?, ?)
Hibernate: insert into employee (department_department_id, employee_code, employee_name) values (?, ?, ?)

3.关闭子对象中的父对象toString,避免无限循环调用

四、双向关系使用code关联

同单向关系类似,且有额外的update操作
code关联,使用JoinColumn时不可同时使用mappedBy

五、级联操作设置

  • CascadeType.PERSIST:级联保存,在保存department的同时保存employee对象
  • CascadeType.MERGE:级联更新,将department和employee视为一个整体,任何一个对象有变化,都会更新
  • CascadeType.REMOVE:级联删除

    • 当没有设置时,delete主对象时,子对象只是去掉关系;remove子对象时也只是去掉关系,如果增加orphanRemoval = true则会删除remove的子对象
    • 当有设置时,delete主对象时,子对象同样会被删除
  • CascadeType.REFRESH:级联刷新(较少使用),在并发的场景下避免脏数据
  • CascadeType.DETACH:级联脱管(较少使用)
  • CascadeType.ALL:以上全部,需要根据实际情况谨慎设置,以免产生混乱

六、源代码

https://gitee.com/hypier/barr...

此处输入图片的描述


barry的异想世界
20 声望3 粉丝