JPQL介绍
JPQL语言,即 Java Persistence Query Language 的简称。JPQL 是一种和 SQL 非常类似的中间性和对象化查询语言,它最终会被编译成针对不同底层数据库的 SQL 查询,从而屏蔽不同数据库的差异。如果会hibernate的hql查询,那么jpql会上手的很快。
JPQL语言的语句可以是 select 语句、update 语句或delete语句,它们都通过 Query 接口封装执行。
Query接口封装了执行数据库查询的相关方法。调用 EntityManager 的 createQuery、create NamedQuery 及 createNativeQuery 方法可以获得查询对象,进而可调用 Query 接口的相关方法来执行查询操作。
实体类User和Order如下
@Entity
@Table(name = "t_user")
public class User {
@Id//必须指定主键
@GeneratedValue(strategy=GenerationType.IDENTITY)
private Long id;
@Column(name="user_name")
private String userName;
private String password;
private String telephone;
private String email;
@Column(name="create_time")
@Temporal(TemporalType.DATE)
private Date createTime;
@JoinColumn(name="user_id")
@OneToMany(cascade=CascadeType.REMOVE)
private Set orders = new HashSet<>();
//映射必须定义个空构造器
public User() {
}
public User(String userName, String password, String telephone, String email, Date createTime) {
this.userName = userName;
this.password = password;
this.telephone = telephone;
this.email = email;
this.createTime = createTime;
}
public String getUserInfo(){
return "username:"+this.userName+",email:"+this.email;
}
get、set、toString方法
}
@Entity
@Table(name = "t_order")
public class Order {
@Id
@GeneratedValue
private Long id;
//付款金额
private Integer payment;
//支付方式 0:支付宝 1:微信
private Integer channel;
//支付状态
private Integer status=0;
//创建时间
@Column(name="create_time")
@Temporal(TemporalType.DATE)
private Date createTime;
//单向多对1,很智能,数据库中会出现关联字段user_id,或者自己指定外键列名
@ManyToOne(targetEntity=User.class,fetch=FetchType.LAZY)
private User user;
public Order() {
}
public Order(Integer payment, Integer channel, Integer status, Date createTime) {
this.payment = payment;
this.channel = channel;
this.status = status;
this.createTime = createTime;
}
get、set、toString方法
}
public class JpqlTest {
private EntityManager entityManager;
private EntityManagerFactory entityManagerFactory;
private EntityTransaction transaction;
@Before
public void init(){
entityManagerFactory = Persistence.createEntityManagerFactory("jpa-2");
entityManager = entityManagerFactory.createEntityManager();
transaction = entityManager.getTransaction();
transaction.begin();
}
@After
public void destory(){
transaction.commit();
entityManager.close();
entityManagerFactory.close();
}
@Test
/*
* 创建查询对象三种方式:1、createQuery();
* 获取单一(实体类)查询结果:query.getSingleResult()
*/
public void testJpql(){
//1、创建jpql语句(SELECT u可以省去)
String jpql = "SELECT u FROM User u WHERE u.id=?";
//2、创建查询对象
Query query = entityManager.createQuery(jpql);
//3、设置查询参数,位置从1开始
query.setParameter(1, 18l);
//4、执行查询(返回单一实体查询结果)
User user = (User)query.getSingleResult();
System.out.println(user.getUserInfo());
System.out.println(user.getOrders().size());
}
}
控制台输出:
Hibernate:
select
user0_.id as id1_2_,
user0_.create_time as create_t2_2_,
user0_.email as email3_2_,
user0_.password as password4_2_,
user0_.telephone as telephon5_2_,
user0_.user_name as user_nam6_2_
from
t_user user0_
where
user0_.id=?
username:Hession,email:[email protected]
Hibernate:
select
orders0_.user_id as user_id6_2_1_,
orders0_.id as id1_1_1_,
orders0_.id as id1_1_0_,
orders0_.channel as channel2_1_0_,
orders0_.create_time as create_t3_1_0_,
orders0_.payment as payment4_1_0_,
orders0_.status as status5_1_0_,
orders0_.user_id as user_id6_1_0_
from
t_order orders0_
where
orders0_.user_id=?
2
一对多默认就是延迟加载,所以会发送2条sql语句分别查询User信息和Order信息。
api:
query.getSingleResult():执行查询,返回单一实体
注:jpql的编写基本跟hql一样,都是面向对象查询
@Test
/*
* 获取查询实体集合:query.getResultList()
*/
public void testJpql2(){
//1、创建jpql语句(SELECT u可以省去)
String jpql = "SELECT u FROM User u WHERE u.id >?";
//2、创建查询语句
Query query = entityManager.createQuery(jpql);
//3、设置查询参数,位置从1开始
query.setParameter(1, 10l);
//4、执行查询(返回查询的实体集合)
List list = query.getResultList();
for (User user : list) {
System.out.println(user.getUserName());
}
}
控制台输出:
Hibernate:
select
略...
Lucy
jackson
Hession
api:
query.getResultList():返回查询到的实体集合
@Test
/*
* 创建查询对象三种方式:2、createNamedQuery();
* 前提:在查询实体上添加注解,并设置好name及查询语句
* 这里需要在User实体类上添加
* @NamedQuery(name="jpqlNamedQuery",query="SELECT u FROM User u WHERE u.id=1l")
*/
public void testJpql3(){
//传入的字符串即为实体类上定义的name属性
Query query = entityManager.createNamedQuery("jpqlNamedQuery");
User user = (User)query.getSingleResult();
System.out.println(user.getUserInfo());
}
api:
entityManager.createNamedQuery(“jpqlNamedQuery”):查询语句写在实体类上,设置好name,从而创建查询语句,这种方式在hibernate中也有,不过hibernate是把这些写在了配置文件中。
@Test
/*
* 创建查询对象三种方式:3、createNativeQuery();
*/
public void testJpqlNative(){
String sql = "SELECT u.user_name FROM t_user u WHERE u.id=?";
Query query = entityManager.createNativeQuery(sql);
query.setParameter(1, 1l);
Object result = query.getSingleResult();
System.out.println(result);
}
api:
entityManager.createNativeQuery(sql):传入的是sql语句,得到的是Object对象或者泛型为Object的list。
@Test
/*
* 查询部分属性然后封装为实体类
* 默认情况下,若只查询部分属性,则将返回Object[]类型的结果.或者Object[]类型的list
* 也可以在实体类中创建对应的构造器,然后再JPQL语句中利用对应的构造器返回实体类的对象
*/
public void testJpql4(){
String jpql = "SELECT u.userName,u.telephone FROM User u";
Query query = entityManager.createQuery(jpql);
List
输出结果分别为:
Torra
15922222226
Torra
13922222226
Lucy
15522222226
jackson
13322222226
Hession
15822222226
Torra
Torra
Lucy
jackson
Hession
1、分页查询
@Test
public void testJpql6(){
String jpql = "SELECT u FROM User u WHERE u.id>?";
Query query = entityManager.createQuery(jpql);
query.setParameter(1, 1l);
//分页查询,下标1开始,查询2条
query.setFirstResult(1);
query.setMaxResults(2);
List list = query.getResultList();
for (User user : list) {
System.out.println(user);
}
}
api:
query.setFirstResult(pageNum):设置分页查询下标开始位置;
query.setMaxResults(pageSize):设置分页查询记录条数;
控制台输出:
Hibernate:
select
user0_.id as id1_2_,
user0_.create_time as create_t2_2_,
user0_.email as email3_2_,
user0_.password as password4_2_,
user0_.telephone as telephon5_2_,
user0_.user_name as user_nam6_2_
from
t_user user0_
where
user0_.id>? limit ?, ?
User [id=13, userName=Lucy, password=123456, telephone=15522222226, [email protected]]
User [id=14, userName=jackson, password=123456, telephone=13322222226, email=14]
2、排序
按照id降序排列
@Test//排序
public void testJpql7(){
String jpql = "SELECT u FROM User u WHERE u.id>? ORDER BY u.id DESC";
Query query = entityManager.createQuery(jpql);
query.setParameter(1, 1l);
List list = query.getResultList();
for (User user : list) {
System.out.println(user);
}
}
排序语句跟sql语句一样,直接使用ORDER BY 字段 排序方式。
另外Group By,having,函数,都是跟写sql语句一样。
3、DML语句
修改id为14的用户的用户名和邮箱
@Test
public void testJpql8(){
String jpql = "UPDATE User u SET u.userName=? , u.email=? WHERE u.id=?";
Query query = entityManager.createQuery(jpql);
query.setParameter(1, "LuoB");
query.setParameter(2, "[email protected]");
query.setParameter(3,14l);
query.executeUpdate();
}
控制台输出:
Hibernate:
update
t_user
set
user_name=?,
email=?
where
id=?
说到这里,感觉还是hibernate的hql和qbc查询好用,而jpa也只是hibernate的一个子集,即使jpa与spring整合了(确实可以整合)似乎也没hibernate好用,但是jpa的好用之处不在这,在于它和spring data 的结合使用,而spring data与关系型数据库结合中并没有hibernate只有jpa(有mybatis),所以要会spring data + jpa得先知道jpa,用了spring data之后就能体会到快速开发的感觉了。