上篇文章中已经通过一个简单的HelloWorld程序讲解了Spring boot的基本原理和使用。本文主要讲解如何通过spring boot来访问数据库,本文会演示三种方式来访问数据库,第一种是JdbcTemplate,第二种是JPA,第三种是Mybatis。之前已经提到过,本系列会以一个博客系统作为讲解的基础,所以本文会讲解文章的存储和访问(但不包括文章的详情),因为最终的实现是通过MyBatis来完成的,所以,对于JdbcTemplate和JPA只做简单演示,MyBatis部分会完整实现对文章的增删改查。
一、准备工作
在演示这几种方式之前,需要先准备一些东西。第一个就是数据库,本系统是采用MySQL实现的,我们需要先创建一个tb_article的表:
DROP TABLE IF EXISTS `tb_article`; CREATE TABLE `tb_article` ( `id` bigint(20) NOT NULL AUTO_INCREMENT, `title` varchar(255) NOT NULL DEFAULT ‘‘, `summary` varchar(1024) NOT NULL DEFAULT ‘‘, `status` int(11) NOT NULL DEFAULT ‘0‘, `type` int(11) NOT NULL, `user_id` bigint(20) NOT NULL DEFAULT ‘0‘, `create_time` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP, `update_time` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP, `public_time` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP, PRIMARY KEY (`id`) ) ENGINE=InnoDB DEFAULT CHARSET=utf8;
后续的演示会对这个表进行增删改查,大家应该会看到这个表里面并没有文章的详情,原因是文章的详情比较长,如果放在这个表里面容易影响查询文章列表的效率,所以文章的详情会单独存在另外的表里面。此外我们需要配置数据库连接池,这里我们使用druid连接池,另外配置文件使用yaml配置,即application.yml(你也可以使用application.properties配置文件,没什么太大的区别,如果对ymal不熟悉,有兴趣也可以查一下,比较简单)。连接池的配置如下:
spring: datasource: url: jdbc:mysql://127.0.0.1:3306/blog?useUnicode=true&characterEncoding=UTF-8&useSSL=false driverClassName: com.mysql.jdbc.Driver username: root password: 123456 type: com.alibaba.druid.pool.DruidDataSource
最后,我们还需要建立与数据库对应的POJO类,代码如下:
public class Article { private Long id; private String title; private String summary; private Date createTime; private Date publicTime; private Date updateTime; private Long userId; private Integer status;
private Integer type;
}
好了,需要准备的工作就这些,现在开始实现数据库的操作。
二、与JdbcTemplate集成
首先,我们先通过JdbcTemplate来访问数据库,这里只演示数据的插入,上一篇文章中我们已经提到过,Spring boot提供了许多的starter来支撑不同的功能,要支持JdbcTemplate我们只需要引入下面的starter就可以了:
<dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-jdbc</artifactId> </dependency>
现在我们就可以通过JdbcTemplate来实现数据的插入了:
public interface ArticleDao { Long insertArticle(Article article); } @Repository public class ArticleDaoJdbcTemplateImpl implements ArticleDao { @Autowired private NamedParameterJdbcTemplate jdbcTemplate; @Override public Long insertArticle(Article article) { String sql = "insert into tb_article(title,summary,user_id,create_time,public_time,update_time,status) " + "values(:title,:summary,:userId,:createTime,:publicTime,:updateTime,:status)"; Map<String, Object> param = new HashMap<>(); param.put("title", article.getTitle()); param.put("summary", article.getSummary()); param.put("userId", article.getUserId()); param.put("status", article.getStatus()); param.put("createTime", article.getCreateTime()); param.put("publicTime", article.getPublicTime()); param.put("updateTime", article.getUpdateTime()); return (long) jdbcTemplate.update(sql, param); } }
我们通过JUnit来测试上面的代码:
@RunWith(SpringJUnit4ClassRunner.class) @SpringBootTest(classes = Application.class) public class ArticleDaoTest { @Autowired private ArticleDao articleDao; @Test public void testInsert() { Article article = new Article(); article.setTitle("测试标题"); article.setSummary("测试摘要"); article.setUserId(1L); article.setStatus(1); article.setCreateTime(new Date()); article.setUpdateTime(new Date()); article.setPublicTime(new Date()); articleDao.insertArticle(article); } }
要支持上面的测试程序,也需要引入一个starter:
<dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-test</artifactId> <scope>test</scope> </dependency>
从上面的代码可以看出,其实除了引入jdbc的start之外,基本没有配置,这都是spring boot的自动帮我们完成了配置的过程。上面的代码需要注意的Application类的位置,该类必须位于Dao类的父级的包中,比如这里Dao都位于com.pandy.blog.dao这个包下,现在我们把Application.java这个类从com.pandy.blog这个包移动到com.pandy.blog.app这个包中,则会出现如下错误:
Caused by: org.springframework.beans.factory.NoSuchBeanDefinitionException: No qualifying bean of type ‘com.pandy.blog.dao.ArticleDao‘ available: expected at least 1 bean which qualifies as autowire candidate. Dependency annotations: {@org.springframework.beans.factory.annotation.Autowired(required=true)} at org.springframework.beans.factory.support.DefaultListableBeanFactory.raiseNoMatchingBeanFound(DefaultListableBeanFactory.java:1493) at org.springframework.beans.factory.support.DefaultListableBeanFactory.doResolveDependency(DefaultListableBeanFactory.java:1104) at org.springframework.beans.factory.support.DefaultListableBeanFactory.resolveDependency(DefaultListableBeanFactory.java:1066) at org.springframework.beans.factory.annotation.AutowiredAnnotationBeanPostProcessor$AutowiredFieldElement.inject(AutowiredAnnotationBeanPostProcessor.java:585) ... 28 more
也就是说,找不到ArticleDao的实现,这是什么原因呢?上一篇博文中我们已经看到@SpringBootApplication这个注解继承了@ComponentScan,其默认情况下只会扫描Application类所在的包及子包。因此,对于上面的错误,除了保持Application类在Dao的父包这种方式外,也可以指定扫描的包来解决:
@SpringBootApplication @ComponentScan({"com.pandy.blog"}) public class Application { public static void main(String[] args) throws Exception { SpringApplication.run(Application.class, args); } }
三、与JPA集成
现在我们开始讲解如何通过JPA的方式来实现数据库的操作。还是跟JdbcTemplate类似,首先,我们需要引入对应的starter:
<dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-data-jpa</artifactId> </dependency>
然后我们需要对POJO类增加Entity的注解,并指定表名(如果不指定,默认的表名为article),然后需要指定ID的及其生成策略,这些都是JPA的知识,与Spring boot无关,如果不熟悉的话可以看下JPA的知识点:
@Entity(name = "tb_article") public class Article { @Id @GeneratedValue private Long id; private String title; private String summary; private Date createTime; private Date publicTime; private Date updateTime; private Long userId; private Integer status; }
最后,我们需要继承JpaRepository这个类,这里我们实现了两个查询方法,第一个是符合JPA命名规范的查询,JPA会自动帮我们完成查询语句的生成,另一种方式是我们自己实现JPQL(JPA支持的一种类SQL的查询)。
public interface ArticleRepository extends JpaRepository<Article, Long> { public List<Article> findByUserId(Long userId); @Query("select art from com.pandy.blog.po.Article art where title=:title") public List<Article> queryByTitle(@Param("title") String title); }
好了,我们可以再测试一下上面的代码:
@RunWith(SpringJUnit4ClassRunner.class) @SpringBootTest(classes = Application.class) public class ArticleRepositoryTest { @Autowired private ArticleRepository articleRepository; @Test public void testQuery(){ List<Article> articleList = articleRepository.queryByTitle("测试标题"); assertTrue(articleList.size()>0); } }
注意,这里还是存在跟JdbcTemplate类似的问题,需要将Application这个启动类未于Respository和Entity类的父级包中,否则会出现如下错误:
Caused by: org.springframework.beans.factory.NoSuchBeanDefinitionException: No qualifying bean of type ‘com.pandy.blog.dao.ArticleRepository‘ available: expected at least 1 bean which qualifies as autowire candidate. Dependency annotations: {@org.springframework.beans.factory.annotation.Autowired(required=true)} at org.springframework.beans.factory.support.DefaultListableBeanFactory.raiseNoMatchingBeanFound(DefaultListableBeanFactory.java:1493) at org.springframework.beans.factory.support.DefaultListableBeanFactory.doResolveDependency(DefaultListableBeanFactory.java:1104) at org.springframework.beans.factory.support.DefaultListableBeanFactory.resolveDependency(DefaultListableBeanFactory.java:1066) at org.springframework.beans.factory.annotation.AutowiredAnnotationBeanPostProcessor$AutowiredFieldElement.inject(AutowiredAnnotationBeanPostProcessor.java:585) ... 28 more
当然,同样也可以通过注解@EnableJpaRepositories指定扫描的JPA的包,但是还是不行,还会出现如下错误:
Caused by: java.lang.IllegalArgumentException: Not a managed type: class com.pandy.blog.po.Article at org.hibernate.jpa.internal.metamodel.MetamodelImpl.managedType(MetamodelImpl.java:210) at org.springframework.data.jpa.repository.support.JpaMetamodelEntityInformation.<init>(JpaMetamodelEntityInformation.java:70) at org.springframework.data.jpa.repository.support.JpaEntityInformationSupport.getEntityInformation(JpaEntityInformationSupport.java:68) at org.springframework.data.jpa.repository.support.JpaRepositoryFactory.getEntityInformation(JpaRepositoryFactory.java:153) at org.springframework.data.jpa.repository.support.JpaRepositoryFactory.getTargetRepository(JpaRepositoryFactory.java:100) at org.springframework.data.jpa.repository.support.JpaRepositoryFactory.getTargetRepository(JpaRepositoryFactory.java:82) at org.springframework.data.repository.core.support.RepositoryFactorySupport.getRepository(RepositoryFactorySupport.java:199) at org.springframework.data.repository.core.support.RepositoryFactoryBeanSupport.initAndReturn(RepositoryFactoryBeanSupport.java:277) at org.springframework.data.repository.core.support.RepositoryFactoryBeanSupport.afterPropertiesSet(RepositoryFactoryBeanSupport.java:263) at org.springframework.data.jpa.repository.support.JpaRepositoryFactoryBean.afterPropertiesSet(JpaRepositoryFactoryBean.java:101) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.invokeInitMethods(AbstractAutowireCapableBeanFactory.java:1687) at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.initializeBean(AbstractAutowireCapableBeanFactory.java:1624) ... 39 more
这个错误说明识别不了Entity,所以还需要通过注解@EntityScan来指定Entity的包,最终的配置如下:
@SpringBootApplication @ComponentScan({"com.pandy.blog"}) @EnableJpaRepositories(basePackages="com.pandy.blog") @EntityScan("com.pandy.blog") public class Application { public static void main(String[] args) throws Exception { SpringApplication.run(Application.class, args); } }
四、与MyBatis集成
最后,我们再看看如何通过MyBatis来实现数据库的访问。同样我们还是要引入starter:
<dependency> <groupId>org.mybatis.spring.boot</groupId> <artifactId>mybatis-spring-boot-starter</artifactId> <version>1.1.1</version> </dependency>
由于该starter不是spring boot官方提供的,所以版本号于Spring boot不一致,需要手动指定。
MyBatis一般可以通过XML或者注解的方式来指定操作数据库的SQL,个人比较偏向于XML,所以,本文中也只演示了通过XML的方式来访问数据库。首先,我们需要配置mapper的目录。我们在application.yml中进行配置:
mybatis: config-locations: mybatis/mybatis-config.xml mapper-locations: mybatis/mapper/*.xml type-aliases-package: com.pandy.blog.po
这里配置主要包括三个部分,一个是mybatis自身的一些配置,例如基本类型的别名。第二个是指定mapper文件的位置,第三个POJO类的别名。这个配置也可以通过 Java configuration来实现,由于篇幅的问题,我这里就不详述了,有兴趣的朋友可以自己实现一下。
配置完后,我们先编写mapper对应的接口:
public interface ArticleMapper { public Long insertArticle(Article article); public void updateArticle(Article article); public Article queryById(Long id); public List<Article> queryArticlesByPage(@Param("article") Article article, @Param("pageSize") int pageSize, @Param("offset") int offset); }
该接口暂时只定义了四个方法,即添加、更新,以及根据ID查询和分页查询。这是一个接口,并且和JPA类似,可以不用实现类。接下来我们编写XML文件:
<?xml version="1.0" encoding="UTF-8" ?> <!DOCTYPE mapper PUBLIC "-//mybatis.org//DTD Mapper 3.0//EN" "http://mybatis.org/dtd/mybatis-3-mapper.dtd" > <mapper namespace="com.pandy.blog.dao.ArticleMapper"> <resultMap id="articleMap" type="com.pandy.blog.po.Article"> <id column="id" property="id" jdbcType="INTEGER"/> <result column="title" property="title" jdbcType="VARCHAR"/> <result column="summary" property="summary" jdbcType="VARCHAR"/> <result column="user_id" property="userId" jdbcType="INTEGER"/> <result column="status" property="status" jdbcType="INTEGER"/> <result column="create_time" property="createTime" jdbcType="TIMESTAMP"/> <result column="update_time" property="updateTime" jdbcType="TIMESTAMP"/> <result column="public_time" property="publicTime" jdbcType="TIMESTAMP"/> </resultMap> <sql id="base_column"> title,summary,user_id,status,create_time,update_time,public_time </sql> <insert id="insertArticle" parameterType="Article"> INSERT INTO tb_article(<include refid="base_column"/>) VALUE (#{title},#{summary},#{userId},#{status},#{createTime},#{updateTime},#{publicTime}) </insert> <update id="updateArticle" parameterType="Article"> UPDATE tb_article <set> <if test="title != null"> title = #{title}, </if> <if test="summary != null"> summary = #{summary}, </if> <if test="status!=null"> status = #{status}, </if> <if test="publicTime !=null "> public_time = #{publicTime}, </if> <if test="updateTime !=null "> update_time = #{updateTime}, </if> </set> WHERE id = #{id} </update> <select id="queryById" parameterType="Long" resultMap="articleMap"> SELECT id,<include refid="base_column"></include> FROM tb_article WHERE id = #{id} </select> <select id="queryArticlesByPage" resultMap="articleMap"> SELECT id,<include refid="base_column"></include> FROM tb_article <where> <if test="article.title != null"> title like CONCAT(‘%‘,${article.title},‘%‘) </if> <if test="article.userId != null"> user_id = #{article.userId} </if> </where> limit #{offset},#{pageSize} </select> </mapper>
最后,我们需要手动指定mapper扫描的包:
@SpringBootApplication @MapperScan("com.pandy.blog.dao") public class Application { public static void main(String[] args) throws Exception { SpringApplication.run(Application.class, args); } }
好了,与MyBatis的集成也完成了,我们再测试一下:
@RunWith(SpringJUnit4ClassRunner.class) @SpringBootTest(classes = Application.class) public class ArticleMapperTest { @Autowired private ArticleMapper mapper; @Test public void testInsert() { Article article = new Article(); article.setTitle("测试标题2"); article.setSummary("测试摘要2"); article.setUserId(1L); article.setStatus(1); article.setCreateTime(new Date()); article.setUpdateTime(new Date()); article.setPublicTime(new Date()); mapper.insertArticle(article); } @Test public void testMybatisQuery() { Article article = mapper.queryById(1L); assertNotNull(article); } @Test public void testUpdate() { Article article = mapper.queryById(1L); article.setPublicTime(new Date()); article.setUpdateTime(new Date()); article.setStatus(2); mapper.updateArticle(article); } @Test public void testQueryByPage(){ Article article = new Article(); article.setUserId(1L); List<Article> list = mapper.queryArticlesByPage(article,10,0); assertTrue(list.size()>0); } }
五、总结
本文演示Spring boot与JdbcTemplate、JPA以及MyBatis的集成,整体上来说配置都比较简单,以前做过相关配置的同学应该感觉比较明显,Spring boot确实在这方面给我们提供了很大的帮助。后续的文章中我们只会使用MyBatis这一种方式来进行数据库的操作,这里还有一点需要说明一下的是,MyBatis的分页查询在这里是手写的,这个分页在正式开发中可以通过插件来完成,不过这个与Spring boot没什么关系,所以本文暂时通过这种手动的方式来进行分页的处理。