如何利用Hibernate4的annotation配置一对多（OneToMany）关系？

首先，来看看部门（department）和雇员（employee）的ER关系图：

这个关系该的理解为：一个雇员拥有一个部门，一个部门对应多个雇员。从关系上理解，雇员为拥有方（the owning side），部门为被拥有方（the owned side）。当你用Hibenate4的annotation去配置映射关系的时候，有个需要注意的地方：@JoinColumn和@OneToMany中的mappedBy属性是不能同时存在的。也即，在拥有方（the owning side）你只能这么写：

@ManyToOne(cascade = {CascadeType.PERSIST, CascadeType.MERGE})
@JoinColumn(name = "department_id")
private Department department;

在被拥有方（the owned side）你只能这么写：

@OneToMany(mappedBy = "department", cascade = CascadeType.ALL)
@Fetch(FetchMode.SUBSELECT)
@OrderBy("id")
private Set<Employee> employees = new LinkedHashSet<Employee>();

你可以用一下测试代码：

public class HibernateTest {

	private static SessionFactory sf;
	
	@BeforeClass
	public static void setUpBeforeClass() throws Exception {
		Configuration cfg = new Configuration();
		cfg.configure();
		ServiceRegistry sr = new ServiceRegistryBuilder().applySettings(cfg.getProperties()).buildServiceRegistry();
		try {
			sf = cfg.configure()
					.addAnnotatedClass(Employee.class)
					.buildSessionFactory(sr);
		} catch (Exception e) {
			e.printStackTrace();
		}
	}
	
	@Test
	public void test() {
		Session s = sf.openSession();
		Transaction tx = s.beginTransaction();
		
		Department d = new Department();
		d.setName("development");
		
		Employee e = new Employee();
		e.setAge(33);
		e.setSex(1);
		e.setCellphone("86186xxxxxxx8");
		e.setLastname("Tian");
		e.setFirstname("Max");
		e.setEmail("max.tian@xxxx.com");
		e.setDepartment(d);
		
		s.persist(e);
		tx.commit();
		s.close();
		
	}
}

关于mappedBy的理解：

1. 只有OneToOne,OneToMany,ManyToMany上才有mappedBy属性，ManyToOne不存在该属性；
   
2. mappedBy标签一定是定义在被拥有方（the owned side），他指向拥有方（the owning side）；
   
3. mappedBy的含义，应该理解为，拥有方能够自动维护跟被拥有方的关系；
   
4. mappedBy跟JoinColumn/JoinTable总是处于互斥的。
   

数据库表结构：

-- Table: department

-- DROP TABLE department;

CREATE TABLE department
(
  id integer NOT NULL,
  name character varying(60),
  CONSTRAINT department_pk PRIMARY KEY (id)
)
WITH (
  OIDS=FALSE
);
ALTER TABLE department
  OWNER TO postgres;

-- Table: employee

-- DROP TABLE employee;

CREATE TABLE employee
(
  id integer NOT NULL,
  first_name character varying(60),
  last_name character varying(60),
  full_name character varying(120),
  email character varying(255),
  cell_phone character varying(15),
  age smallint,
  sex smallint,
  department_id integer,
  CONSTRAINT employee_pk PRIMARY KEY (id)
)
WITH (
  OIDS=FALSE
);
ALTER TABLE employee
  OWNER TO postgres;