这是今晚尝试让JPA / Hibernate在新项目中工作的第43个障碍。
当尝试创建然后持久保存我的Staffer
类时,我得到:
SEVERE: Column 'Person_ID' cannot be null
SEVERE: Could not synchronize database state with session
Staffer
包含一个Person
,一个Office
和一个Location
,其中Person
是主键。我正在手动创建这些实体中的每一个,将它们持久化,附加它们,验证它们不为null,其id不为null,然后持久化我的Staffer
。我无法弄清楚为什么JPA / Hibernate在尝试坚持时为何抱怨一个(显然)空人员字段:
EntityManager em = EMF.get().createEntityManager();
em.getTransaction().begin();
Location location = new Location();
location.setCity("my city");
location = em.merge(location);
assertNotNull(location.getId());
Office office = new Office();
office.setName("my office");
office.setLocation(location);
office = em.merge(office);
assertNotNull(office.getId());
Person person = new Person();
person.setFirstName("first");
person.setLastName("last");
person = em.merge(person);
assertNotNull(person.getId());
Staffer staffer = new Staffer();
staffer.setPerson(person);
staffer.setCellPhone("555-555-5555");
staffer.setOffice(office);
staffer.setHomeLocation(location);
staffer.setPersonalEmail("[email protected]");
assertNotNull(staffer.getPerson());
assertNotNull(staffer.getPerson().getId());
staffer = em.merge(staffer); // CODE FAILS HERE
em.getTransaction().commit();
这是我的模型代码:
@Entity
@Table(name = "Staffer")
public class Staffer implements Serializable
{
@Id
@OneToOne(cascade = CascadeType.PERSIST)
@JoinColumn(name = "Person_ID")
private Person person;
@ManyToOne(cascade = CascadeType.PERSIST)
@JoinColumn(name = "Office_ID")
private Office office;
@ManyToOne(cascade = CascadeType.PERSIST)
@JoinColumn(name = "Home_Address_Location_ID")
private Location homeLocation;
@Column(name = "Home_Phone")
private String homePhone;
@Column(name = "Cell_Phone")
private String cellPhone;
@Column(name = "Personal_Email")
private String personalEmail;
@Override
public boolean equals(Object o)
{
if (this == o) return true;
if (o == null || getClass() != o.getClass()) return false;
Staffer staffer = (Staffer) o;
if (homeLocation != null ? !homeLocation.equals(staffer.homeLocation) : staffer.homeLocation != null)
return false;
if (office != null ? !office.equals(staffer.office) : staffer.office != null) return false;
if (person != null ? !person.equals(staffer.person) : staffer.person != null) return false;
return true;
}
@Override
public int hashCode()
{
int result = person != null ? person.hashCode() : 0;
result = 31 * result + (office != null ? office.hashCode() : 0);
result = 31 * result + (homeLocation != null ? homeLocation.hashCode() : 0);
return result;
}
// getters & setters, etc.
}
@Entity
@Table(name = "Person")
public class Person
{
@Id
@GeneratedValue(strategy = GenerationType.AUTO)
@Column(name = "Person_ID")
private Long id;
@Column(name = "First_Name")
private String firstName;
@Column(name = "Last_Name")
private String lastName;
// getters & setters, etc.
}
@Entity
@Table(name = "Office")
public class Office
{
@Id
@GeneratedValue(strategy= GenerationType.AUTO)
@Column(name = "Office_ID")
private Long id;
@Column(name = "Name")
private String name;
@ManyToOne(cascade = CascadeType.PERSIST)
@JoinColumn(name = "Location_ID")
private Location location;
// getters & setters, etc.
}
@Entity
@Table(name = "Location")
public class Location
{
@Id
@GeneratedValue(strategy= GenerationType.AUTO)
@Column(name = "Location_ID")
private Long id;
@Column(name ="City")
private String city;
// getters & setters, etc.
}
另外,我不确定为什么当Hibernate创建表时,它会创建Home_Address_Location_ID和Office字段作为Staffer键的一部分。我的意图是仅使用Person作为主键。
+--------------------------+--------------+------+-----+---------+-------+
| Field | Type | Null | Key | Default | Extra |
+--------------------------+--------------+------+-----+---------+-------+
| Cell_Phone | varchar(255) | YES | | NULL | |
| Home_Phone | varchar(255) | YES | | NULL | |
| Personal_Email | varchar(255) | YES | | NULL | |
| Person_ID | bigint(20) | NO | PRI | NULL | |
| Home_Address_Location_ID | bigint(20) | YES | MUL | NULL | |
| Office_ID | bigint(20) | YES | MUL | NULL | |
+--------------------------+--------------+------+-----+---------+-------+
最佳答案
在我看来Staffer
确实是Person
的子类。我将以这种方式建模(Hibernate会更快乐)。可能使用每个子类的表设置。
其他选项包括Staffer
(ugh)上的代理ID或某种嵌入式ID伏都教(没想到最后一个到)。