[TOC]

SpringData JPA是spring基于ORM框架、JPA規(guī)范的基礎(chǔ)上封裝的一套JPA應(yīng)用框架诈乒，可以使開發(fā)者使用極簡(jiǎn)的代碼實(shí)現(xiàn)對(duì)數(shù)據(jù)庫(kù)的訪問(wèn)和操作淑趾。它提供了包括增刪改查等在內(nèi)的基本功能，且易于擴(kuò)展岖圈。

springdata jpa鹦牛、jpa和hibernate三者關(guān)系

通俗來(lái)講springdata jpa是對(duì)jpa規(guī)范的一層封裝搞糕，hibernate實(shí)現(xiàn)了jpa規(guī)范。

java代碼----->springdata jpa ------>jpa規(guī)范------>hibernate------>jdbc ----->mysql數(shù)據(jù)庫(kù)

graph LR
A[java代碼] -->B(spring data jpa)
B --> |jpa規(guī)范| C(hibernate)
C -->|jdbc| D(mysql數(shù)據(jù)庫(kù))

我們使用java代碼調(diào)用springdata jpa的api曼追，springdata jpa封裝了jpa規(guī)范窍仰，并且內(nèi)部使用的是hibernate實(shí)現(xiàn)，hibernate封裝了jdbc進(jìn)行數(shù)據(jù)庫(kù)操作礼殊。

入門案例

1驹吮、創(chuàng)建工程，導(dǎo)入依賴

    compile group: 'org.hibernate', name: 'hibernate-core', version: '5.4.3.Final'
    compile group: 'org.hibernate', name: 'hibernate-c3p0', version: '5.4.3.Final'
    compile group: 'mysql', name: 'mysql-connector-java', version: '8.0.16'
    compile group: 'org.springframework.data', name: 'spring-data-jpa', version: '2.1.9.RELEASE'
    testCompile group: 'org.springframework', name: 'spring-test', version: '5.1.8.RELEASE'

2晶伦、編寫spring配置文件

• 配置spring相關(guān)

• 數(shù)據(jù)源信息

• jpa的實(shí)現(xiàn)方式

• 配置要用到的實(shí)體類

• 配置jpa實(shí)現(xiàn)方的配置信息

• 配置事務(wù)管理器

• 聲明式事務(wù)

<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:jpa="http://www.springframework.org/schema/data/jpa"
       xmlns:context="http://www.springframework.org/schema/context"
       xsi:schemaLocation="http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans.xsd http://www.springframework.org/schema/data/jpa http://www.springframework.org/schema/data/jpa/spring-jpa.xsd http://www.springframework.org/schema/context https://www.springframework.org/schema/context/spring-context.xsd">

    <!--spring-->
    <!--配置spring的注解掃描-->
    <context:component-scan base-package="com.lxf"/>


    <!--spring data jpa-->
    <!--整合spring data jpa-->
    <jpa:repositories base-package="com.lxf.dao" entity-manager-factory-ref="entityManagerFactory" transaction-manager-ref="transactionManager" />

    <!--創(chuàng)建實(shí)體管理器工廠,交給spring管理-->
    <bean id="entityManagerFactory" class="org.springframework.orm.jpa.LocalContainerEntityManagerFactoryBean">
        <!--配置數(shù)據(jù)源-->
        <property name="dataSource" ref="dataSource"/>
        <!--配置要掃描的包碟狞，實(shí)體所在包-->
        <property name="packagesToScan" value="com.lxf.entity"/>
        <!--配置jpa的實(shí)現(xiàn)方-->
        <property name="persistenceProvider">
            <bean class="org.hibernate.jpa.HibernatePersistenceProvider"/>
        </property>

        <!--jpa的實(shí)現(xiàn)方的配置-->
        <property name="jpaVendorAdapter">
            <bean class="org.springframework.orm.jpa.vendor.HibernateJpaVendorAdapter">
                <!--數(shù)據(jù)庫(kù)類型-->
                <property name="database" value="MYSQL"/>
                <!--控制臺(tái)顯示sql語(yǔ)句-->
                <property name="showSql" value="true"/>
                <!--是否自動(dòng)創(chuàng)建數(shù)據(jù)庫(kù)表-->
                <property name="generateDdl" value="true"/>
                <!--數(shù)據(jù)庫(kù)方言-->
                <property name="databasePlatform" value="org.hibernate.dialect.MySQLDialect"/>
            </bean>
        </property>

        <!--jpa方言:高級(jí)特性-->

    </bean>

    <!--數(shù)據(jù)源-->
    <bean id="dataSource" class="com.mchange.v2.c3p0.ComboPooledDataSource">
        <property name="jdbcUrl" value="jdbc:mysql://localhost:3306/study?serverTimezone=GMT"/>
        <property name="user" value="root"/>
        <property name="password" value="crystal1024"/>
        <property name="driverClass" value="com.mysql.cj.jdbc.Driver"/>
    </bean>

    <!--配置事務(wù)管理器-->
    <bean id="transactionManager" class="org.springframework.orm.jpa.JpaTransactionManager">
        <property name="entityManagerFactory" ref="entityManagerFactory"/>
    </bean>

    <!--聲明式事務(wù)-->
    
</beans>

3、創(chuàng)建實(shí)體類婚陪，編寫實(shí)體類和數(shù)據(jù)庫(kù)表關(guān)系映射

參考JPA規(guī)范族沃。

4、編寫dao層接口

• 需要繼承兩個(gè)接口

  • JpaRepository：封裝了增刪改查分頁(yè)排序等基本操作近忙，具體可以看JpaRepository的父類

  graph TB
  A[Repository] -->B(CrudRepository)
  B --> C(PagingAndSortingRepository)
  C -->D(JpaRepository)
  

  • JpaSpecificationExecutor：封裝了標(biāo)準(zhǔn)查詢

• 提供相應(yīng)的泛型

  • JpaRepository
    • 操作的實(shí)體類型
    • 實(shí)體中主鍵類型
  • JpaSpecificationExecutor
    • 操作的實(shí)體類型

public interface UserDao extends JpaRepository<User,Integer>, JpaSpecificationExecutor<User> {
}

• 會(huì)通過(guò)動(dòng)態(tài)代理自動(dòng)生成相應(yīng)方法

5竭业、測(cè)試

@RunWith(SpringJUnit4ClassRunner.class)
@ContextConfiguration(locations = "classpath:spring_data_jpa.xml")
public class Test {
    @Autowired
    private UserDao userDao;

    @org.junit.Test
    public void textSave(){
        User user = new User();
        user.setName("小紅");
        user.setAge(22);
        User userResult = userDao.save(user);
        System.out.println(userResult);
    }
}

User{id=4, name='小紅', age=22, sex=null, address='null', phone='null'}

操作數(shù)據(jù)庫(kù)

調(diào)用spring data jpa的api

插入/更新

• save方法：傳入的實(shí)體對(duì)象有主鍵則更新智润，沒有主鍵則插入及舍。

    @org.junit.Test
    public void testSave(){
        User user = new User();
        user.setName("小紅");
        user.setAge(22);
        User userResult = userDao.save(user);
        System.out.println(userResult);
    }

刪除

• delete系列方法

    @org.junit.Test
    public void testDelete(){
        userDao.deleteById(2);
    }

查詢

• count：統(tǒng)計(jì)

• exists系列方法：數(shù)據(jù)庫(kù)中是否存在

• find系列方法：立即加載

• getOne：延遲加載，返回的是一個(gè)動(dòng)態(tài)代理對(duì)象

    @org.junit.Test
    public void testFindOne(){
//        Optional<User> user = userDao.findById(2);
//        System.out.println(user.get());
        User user = userDao.getOne(2);
        System.out.println(user);
    }

    @org.junit.Test
    public void testApi(){
        long count = userDao.count();
        boolean b = userDao.existsById(2);
    }

語(yǔ)句操作

除了調(diào)用spring data jpa內(nèi)置的api窟绷，我們也可以在dao接口中定義我們自己的方法锯玛，通過(guò)@Query聲明jpql或sql語(yǔ)句。

• @Query
  • value：數(shù)據(jù)庫(kù)操作語(yǔ)句
  • nativeQuery：是否是原生查詢，默認(rèn)false攘残，即默認(rèn)使用jpql查詢
• @Modifying：聲明當(dāng)前是一個(gè)更新操作拙友，需要修改數(shù)據(jù)庫(kù)數(shù)據(jù)。
  • 只能用于void或int/Integer的返回類型
  • 因?yàn)樾枰薷臄?shù)據(jù)庫(kù)數(shù)據(jù)歼郭，未防止修改失敗造成未知后果遗契，需要搭配事務(wù)管理來(lái)是使用
• @Transactional：添加事務(wù)管理支持
  • 一般需要設(shè)置rollbackFor或者noRollbackFor，來(lái)表示什么情況下進(jìn)行事務(wù)回滾
• @Rollback：是否可以回滾病曾，默認(rèn)true

jpql查詢

public interface UserDao extends JpaRepository<User, Integer>, JpaSpecificationExecutor<User> {

    @Query(value = "from User where name = :name and age = :age")
    public User findUserByName(@Param("name") String userName,@Param("age") int age);
}

jpql更新

@Query(value = "update User set name = :name where id = :id")
@Modifying
public Integer updateNameById(@Param("id") int id,@Param("name") String userName);


    @org.junit.Test
    @Transactional(rollbackFor = Exception.class)
    //@Rollback(value = false)//如果設(shè)置為fasle牍蜂，即使發(fā)生異常也不會(huì)回滾
    public void testJpql(){
        User user = userDao.findUserByName("lili",18);
        System.out.println(user);

        userDao.updateNameById(user.getId(),"lili_2");
    }

原生sql語(yǔ)句查詢

public interface UserDao extends JpaRepository<User, Integer>, JpaSpecificationExecutor<User> {
    @Query(value = "select * from user where name = :name and age = :age",nativeQuery = true)
    public User findUserByName(@Param("name") String userName,@Param("age") int age);
}

約定規(guī)則查詢

spring data jpa制定了一些約定，如果按照這些約定來(lái)定義方法名泰涂，則會(huì)自動(dòng)解析出sql語(yǔ)句鲫竞。

findBy + 屬性名 + 查詢方式 + （And|Or） + 屬性名 + 查詢方式...

簡(jiǎn)單挑幾個(gè)示例：

public interface UserDao extends JpaRepository<User, Integer>, JpaSpecificationExecutor<User> {
    public User findByName(String name);

    public User findByNameLike(String name);

    public User findByNameLikeAndAge(String name, int age);

    public List<User> findByIdBetween(int idMin, int idMax);
}



    @org.junit.Test
    public void testName(){
        User user1 = userDao.findByName("tom");
        System.out.println(user1);

        User user2 = userDao.findByNameLike("t%");
        System.out.println(user2);

        User user3 = userDao.findByNameLikeAndAge("tom",18);
        System.out.println(user3);

        List<User> users = userDao.findByIdBetween(1, 3);
        users.forEach(new Consumer<User>() {
            @Override
            public void accept(User user) {
                System.out.println(user);
            }
        });
    }

Hibernate: select user0_.id as id1_0_, user0_.address as address2_0_, user0_.age as age3_0_, user0_.name as name4_0_, user0_.phone as phone5_0_, user0_.sex as sex6_0_ from user user0_ where user0_.name=?
User{id=3, name='tom', age=18, sex=1, address='null', phone='null'}
Hibernate: select user0_.id as id1_0_, user0_.address as address2_0_, user0_.age as age3_0_, user0_.name as name4_0_, user0_.phone as phone5_0_, user0_.sex as sex6_0_ from user user0_ where user0_.name like ? escape ?
User{id=3, name='tom', age=18, sex=1, address='null', phone='null'}
Hibernate: select user0_.id as id1_0_, user0_.address as address2_0_, user0_.age as age3_0_, user0_.name as name4_0_, user0_.phone as phone5_0_, user0_.sex as sex6_0_ from user user0_ where (user0_.name like ? escape ?) and user0_.age=?
User{id=3, name='tom', age=18, sex=1, address='null', phone='null'}
Hibernate: select user0_.id as id1_0_, user0_.address as address2_0_, user0_.age as age3_0_, user0_.name as name4_0_, user0_.phone as phone5_0_, user0_.sex as sex6_0_ from user user0_ where user0_.id between ? and ?
User{id=2, name='lili2', age=18, sex=1, address='null', phone='null'}
User{id=3, name='tom', age=18, sex=1, address='null', phone='null'}

標(biāo)準(zhǔn)查詢（Specification）

我們上面提到過(guò)，springdata jpa的dao層一般繼承2個(gè)接口JpaRepository和JpaSpecificationExecutor逼蒙。JpaRepository封裝了crud从绘、統(tǒng)計(jì)、排序是牢、分頁(yè)的常見操作僵井，而JpaSpecificationExecutor基于JPA的criteria查詢封裝了另一種查詢方式，我們之前一直在使用JpaRepositoru中的方法驳棱，下面來(lái)看下JpaSpecificationExecutor接口驹沿，它里面只提供了5個(gè)方法：

public interface JpaSpecificationExecutor<T> {
    //查詢一個(gè)
    Optional<T> findOne(@Nullable Specification<T> spec);
    //查詢?nèi)?    List<T> findAll(@Nullable Specification<T> spec);
    //查詢?nèi)? 提供分頁(yè)功能
    Page<T> findAll(@Nullable Specification<T> spec, Pageable pageable);
    //查詢?nèi)浚峁┡判蚬δ?    List<T> findAll(@Nullable Specification<T> spec, Sort sort);
    //統(tǒng)計(jì)
    long count(@Nullable Specification<T> spec);
}

可以看到蹈胡，這5個(gè)方法有個(gè)共同點(diǎn)渊季，接收一個(gè)Specification參數(shù)。

Specification

Specification是對(duì)JPA規(guī)范中Root罚渐、CriteriaQuery却汉、CriteriaBuilder的一層封裝，用于構(gòu)建過(guò)濾條件荷并。實(shí)例化Specification需要實(shí)現(xiàn)它的toPerdicate方法：

//參數(shù)含義在我的另一文JPA規(guī)范中有介紹合砂，簡(jiǎn)單說(shuō)來(lái)Root用于獲得查詢屬性，CriteriaBuilder用于構(gòu)建過(guò)濾條件源织，CriteriaQuery用于指定最終查詢語(yǔ)句翩伪，這里一般不會(huì)使用，默認(rèn)為where語(yǔ)句谈息。
Predicate toPredicate(Root<T> root, CriteriaQuery<?> query, CriteriaBuilder criteriaBuilder);

注意這里創(chuàng)建出來(lái)的是where查詢語(yǔ)句缘屹。

來(lái)個(gè)簡(jiǎn)單示例僻孝，查詢表中年齡大于等于18的所有河南人：

    @Test
    public void test(){
        Specification<User> specification = new Specification<User>() {
            @Override
            public Predicate toPredicate(Root<User> root, CriteriaQuery<?> query, CriteriaBuilder criteriaBuilder) {
                //分別構(gòu)造各個(gè)單屬性的過(guò)濾條件
                Predicate namePredicate = criteriaBuilder.like(root.get("address"), "河南%");
                Predicate agePredicate = criteriaBuilder.ge(root.get("age"), 18);//大于等于

                //組合成最終的過(guò)濾條件
                Predicate predicate = criteriaBuilder.and(namePredicate, agePredicate);
                return predicate;
            }
        };
        
        //查詢
        List<User> users = userDao.findAll(specification);
        users.forEach(new Consumer<User>() {
            @Override
            public void accept(User user) {
                System.out.println(user);
            }
        });
    }

如果要添加排序和分頁(yè)蒿囤，可以使用Sort和Pageable。

• Sort：排序

Sort sort = new Sort(Sort.Direction.DESC,"id");//排序?qū)傩钥梢栽O(shè)置多個(gè)
List<User> users = userDao.findAll(specification,sort);

• Pageable：分頁(yè)袱蜡，是一個(gè)接口，可以通過(guò)PageRequest構(gòu)建實(shí)例互亮。

Sort sort = new Sort(Sort.Direction.DESC,"id");
//Pageable pageable = PageRequest.of(0,10);//pageIndex犁享，pageSize
Pageable pageable = PageRequest.of(0,10,sort);
Page<User> users = userDao.findAll(specification, pageable);
users.forEach(new Consumer<User>() {
   @Override
   public void accept(User user) {
      System.out.println(user);
   }
});

spring boot中的springdata jpa配置

application.yaml

spring:
  datasource:
    driver-class-name: com.mysql.cj.jdbc.Driver
    url: jdbc:mysql://localhost:3306/study?serverTimezone=GMT
    username: root
    password: crystal1024
  jpa:
    show-sql: true
    hibernate:
      ddl-auto: update

Demo源碼地址

https://github.com/lunxinfeng/jpa/tree/master/springdatajpa