SpringBoot整合MyBatis实战
本文从本人博客搬运，原文格式更加美观，可以移步原文阅读：SpringBoot整合MyBatis实战
基本增删改查
1.准备工作
给数据库创建表tb_user
CREATE TABLE `tb_user` (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键ID',
`user_name` varchar(30) NOT NULL DEFAULT '' COMMENT '姓名',
`age` int(11) unsigned NOT NULL DEFAULT '0' COMMENT '年龄',
`email` varchar(50) NOT NULL DEFAULT '' COMMENT '邮箱',
`create_time` datetime DEFAULT NULL COMMENT '创建时间',
`update_time` datetime DEFAULT NULL COMMENT '修改时间',
`version` int(11) unsigned NOT NULL DEFAULT '1' COMMENT '乐观锁字段',
`status` varchar(50) NOT NULL DEFAULT '' COMMENT '状态:启用ENABLE，禁用DISABLE',
PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=12 DEFAULT CHARSET=utf8 COMMENT='用户表';
编写表对应的实体类
@Data
@AllArgsConstructor
@NoArgsConstructor
public class UserPO {
private Long id;
private String userName;
private Integer age;
private String email;
private Date createTime;
private Date updateTime;
private Integer version;
private UserStatusEnum status;
}
编写用户状态的枚举类
public enum UserStatusEnum {
/**
* 启用
*/
ENABLE,
/**
* 禁用
*/
DISABLE;
}
添加maven依赖
<dependency>
<groupId>org.mybatis.spring.bootgroupId>
<artifactId>mybatis-spring-boot-starterartifactId>
<version>2.1.0version>
dependency>
MyBatis-Spring-Boot-Starter依赖于MyBatis-Spring，它们与SpringBoot的版本关系如下
| MyBatis-Spring-Boot-Starter | MyBatis-Spring | Spring Boot | Java |
|---|---|---|---|
| 2.2 | 2.0 (need 2.0.6+ for enable all features) | 2.5 or higher | 8 or higher |
| 2.1 | 2.0 (need 2.0.6+ for enable all features) | 2.1 - 2.4 | 8 or higher |
| 1.3 | 1.3 | 1.5 | 6 or higher |
yaml中配置数据库基本信息
spring:
datasource:
driver-class-name: com.mysql.cj.jdbc.Driver
url: jdbc:mysql://192.168.157.130:3306/mybatis_plus?serverTimezone=GMT%2B8
username: root
password: root
2.mapper.xml文件版
编写UserMapper.java，添加增删改查方法，并标注@Mapper注解
@Mapper
public interface UserMapper {
void insert(UserPO userPO);
void update(UserPO userPO);
void deleteById(Long userId);
List<UserPO> getAll();
UserPO getById(Long userId);
}
编写UserMapper.xml
<?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.baobao.mybatis.mapper.UserMapper">
<insert id="insert" useGeneratedKeys="true" keyProperty="id">
insert into tb_user(
<if test="userName != null and userName != ''">user_name,</if>
<if test="age != null">age,</if>
<if test="email != null and email != ''">email,</if>
create_time, update_time
) values (
<if test="userName != null and userName != ''">#{userName},</if>
<if test="age != null">#{age},</if>
<if test="email != null and email != ''">#{email},</if>
sysdate(), sysdate()
)
</insert>
<update id="update">
update tb_user
<set>
<if test="userName != null and userName != ''">user_name = #{userName},</if>
<if test="age != null">age = #{age},</if>
<if test="email != null and email != ''">email = #{email},</if>
<if test="status != null">status = #{status},</if>
update_time = sysdate(), version = version + 1
</set>
where id = #{id}
</update>
<delete id="deleteById">
delete from tb_user where id = #{userId}
</delete>
<select id="getAll" resultType="com.baobao.mybatis.entity.po.UserPO">
select * from tb_user
</select>
<select id="getById" resultType="com.baobao.mybatis.entity.po.UserPO">
select * from tb_user where id = #{userId}
</select>
</mapper>
在yaml中配置mapper.xml文件的位置
编写测试类，测试插入
@Test
void testInsert() {
UserPO userPO = new UserPO();
userPO.setUserName("baobao");
userPO.setAge(18);
userMapper.insert(userPO);
}
测试更新
@Test
void testUpdate() {
UserPO userPO = new UserPO();
userPO.setId(12L);
userPO.setUserName("baobao333");
userPO.setStatus(UserStatusEnum.ENABLE);
userMapper.update(userPO);
}
测试删除
@Test
void testDelete() {
userMapper.deleteById(12L);
}
测试查询
@Test
void testQuery() {
List<UserPO> all = userMapper.getAll();
System.out.println(all);
UserPO userPO = userMapper.getById(11L);
System.out.println(userPO);
}
这里我们会发现像userName，createTime这种多单词组合的属性都没有查出来，原因是mysql中属性名称是类似user_name，create_time这种下划线形式，有以下两种解决方案：
- 在
UserMapper.xml中手动定义resultMap
-
配置驼峰转下划线字段映射。有3种配置方式：
-
在yaml中用
config‐location指定mybatis全局xml配置文件的位置，然后在全局配置文件中配置map-underscore-to-camel-casemybatis: config‐location: classpath:mybatis/mybatis‐config.xml # 指定全局配置文件的位置 mapper‐locations: classpath:mapper/*Mapper.xml # 指定sql映射文件的位置 -
编写Mybatis配置类，在配置类中配置
@Configuration public class MybatisConfig { // 自定义配置mybatis全局配置文件中的属性 @Bean public ConfigurationCustomizer customizer(){ return configuration -> { // 设置驼峰名称映射 configuration.setMapUnderscoreToCamelCase(true); }; } } -
在yaml配置文件中配置
configuration.map-underscore-to-camel-casemybatis: mapper-locations: classpath:mapper/*Mapper.xml configuration: # mybatis全局配置文件中有的属性都可以在这里配置 map-underscore-to-camel-case: true
-
yaml中可以支持的mybatis相关的所有配置如下：
| Property | Description |
|---|---|
config-location |
Location of MyBatis xml config file. |
check-config-location |
Indicates whether perform presence check of the MyBatis xml config file. |
mapper-locations |
Locations of Mapper xml config file. |
type-aliases-package |
Packages to search for type aliases. (Package delimiters are “,; \t\n”) |
type-aliases-super-type |
The super class for filtering type alias. If this not specifies, the MyBatis deal as type alias all classes that searched from type-aliases-package. |
type-handlers-package |
Packages to search for type handlers. (Package delimiters are “,; \t\n”) |
executor-type |
Executor type: SIMPLE, REUSE, BATCH |
default-scripting-language-driver |
The default scripting language driver class. This feature requires to use together with mybatis-spring 2.0.2+. |
configuration-properties |
Externalized properties for MyBatis configuration. Specified properties can be used as placeholder on MyBatis config file and Mapper file. For detail see the MyBatis reference page. |
lazy-initialization |
Whether enable lazy initialization of mapper bean. Set true to enable lazy initialization. This feature requires to use together with mybatis-spring 2.0.2+. |
configuration.* |
Property keys for Configuration bean provided by MyBatis Core. About available nested properties see the MyBatis reference page. NOTE: This property cannot be used at the same time with the config-location. |
scripting-language-driver.thymeleaf.* |
Property keys for ThymeleafLanguageDriverConfig bean provided by MyBatis Thymeleaf. About available nested properties see the MyBatis Thymeleaf reference page. |
scripting-language-driver.freemarker.* |
Properties keys for FreeMarkerLanguageDriverConfig bean provided by MyBatis FreeMarker. About available nested properties see the MyBatis FreeMarker reference page. This feature requires to use together with mybatis-freemarker 1.2.0+. |
scripting-language-driver.velocity.* |
Properties keys for VelocityLanguageDriverConfig bean provided by MyBatis Velocity. About available nested properties see the MyBatis Velocity reference page. This feature requires to use together with mybatis-velocity 2.1.0+. |
如果我们的Mapper接口很多，在每个接口上都标注
@Mapper非常麻烦，我们可以选择在配置类上标注@MapperScan，指定Mapper接口所在的包名，这样就省去了在每个Mapper接口上标注@Mapper
3.注解版
可以省去UserMapper.xml文件，直接在UserMapper.java接口中利用注解编写sql语句
@Mapper
public interface UserMapper {
// 设置自增主键的值到department的id属性
@Options(useGeneratedKeys = true,keyProperty = "id")
@Insert("insert into tb_user (name, age, email) values(#{name}, #{age}, #{email})")
void insert(UserPO userPO);
@Update("update tb_user set name = #{name} where id = #{id}")
void update(UserPO userPO);
@Delete("delete from tb_user where id = #{userId}")
void deleteById(Long userId);
@Select("select * from tb_user")
List<UserPO> getAll();
@Select("select * from tb_user where id = #{userId}")
UserPO getById(Long userId);
}
这种方式实际中不推荐，对于复杂sql语句的支持不友好
枚举处理
默认情况下，Mybatis会将枚举的名称字符串作为值保存到数据库指定字段，查询时再映射成枚举类型
如果我们想自定义插入时的值为枚举的代码，而不是名称字符串，需要自己实现枚举类型处理器
首先改造之前的用户状态枚举类，实现自定义代码的设置
@AllArgsConstructor
@Getter
public enum UserStatusEnum {
/**
* 启用
*/
ENABLE(0, "启用"),
/**
* 禁用
*/
DISABLE(1, "禁用");
private final int code;
private final String desc;
public static UserStatusEnum getByCode(int code) {
UserStatusEnum[] values = UserStatusEnum.values();
for (UserStatusEnum instance : values) {
if (instance.getCode() == code) {
return instance;
}
}
return null;
}
}
然后编写自定义枚举处理类，实现TypeHandler接口，泛型指定我们需要处理的java类型。在类上标注@MappedTypes注解
@MappedTypes(UserStatusEnum.class)
public class UserStatusEnumTypeHandler implements TypeHandler<UserStatusEnum> {
/**
* 定义当前数据如何保存到数据库中
*/
@Override
public void setParameter(PreparedStatement preparedStatement, int i, UserStatusEnum userStatusEnum, JdbcType jdbcType) throws SQLException {
preparedStatement.setInt(i, userStatusEnum.getCode());
}
/**
* 从数据库取出的数据如何映射成java类
*/
@Override
public UserStatusEnum getResult(ResultSet resultSet, String columnName) throws SQLException {
return UserStatusEnum.getByCode(resultSet.getInt(columnName));
}
@Override
public UserStatusEnum getResult(ResultSet resultSet, int columnIndex) throws SQLException {
return UserStatusEnum.getByCode(resultSet.getInt(columnIndex));
}
@Override
public UserStatusEnum getResult(CallableStatement callableStatement, int columnIndex) throws SQLException {
return UserStatusEnum.getByCode(callableStatement.getInt(columnIndex));
}
}
然后在yaml配置中指定自定义TypeHandler所在的包名
此时插入数据库的数据就变成了自定义枚举的code值(当然前提是要修改数据库表中status字段的类型)
整合分页插件
首先引入PageHelper依赖
<dependency>
<groupId>com.github.pagehelpergroupId>
<artifactId>pagehelper-spring-boot-starterartifactId>
<version>1.3.0version>
dependency>
在yaml中配置分页插件参数
pagehelper:
helperDialect: mysql # 数据库方言
reasonable: true # 分页合理化参数，默认值为false。当该参数设置为 true 时，pageNum<=0 时会查询第一页， pageNum>pages（超过总数时），会查询最后一页。默认false 时，直接根据参数进行查询
编写分页测试代码
@Test
void testPage() {
// 分页，查询第2页，每页3条数据
PageHelper.startPage(2, 3);
// 紧跟在startPage之后的方法会被分页
List<UserPO> list = userMapper.getAll();
// 分页结果包装成分页信息
PageInfo<UserPO> pageInfo = new PageInfo<>(list);
// 遍历结果集
for (UserPO userPO : list) {
System.out.println(userPO);
}
// 打印分页信息
System.out.println("当前页码：" + pageInfo.getPageNum());
System.out.println("每页数据条数：" + pageInfo.getPageSize());
System.out.println("总条数：" + pageInfo.getTotal());
System.out.println("前一页页码：" + pageInfo.getPrePage());
System.out.println("后一页页码：" + pageInfo.getNextPage());
System.out.println("是否首页：" + pageInfo.isIsFirstPage());
System.out.println("是否尾页：" + pageInfo.isIsLastPage());
System.out.println("是否有前一页：" + pageInfo.isHasPreviousPage());
System.out.println("是否有后一页：" + pageInfo.isHasNextPage());
}
批量插入或更新
我们首先尝试用最简单的循环插入单条数据来实现批量插入
@Test
void testBatchInsert() {
long start = System.currentTimeMillis();
// 插入10000条数据
for (int i = 0; i < 10000; i++) {
UserPO userPO = new UserPO();
userPO.setUserName("baobao" + i);
userPO.setAge(18 + i);
userMapper.insert(userPO);
}
System.out.println(System.currentTimeMillis() - start);
}
最终的执行时间是13.328秒，非常慢
我们可以利用foreach标签优化批量插入速度。首先新建Mapper的批量插入方法
// UserMapper.java
void insertBatch(List<UserPO> userPOList);
<insert id="insertBatch">
insert into tb_user(user_name, age) values
<foreach collection="userPOList" item="user" separator=",">
(#{user.userName}, #{user.age})
foreach>
insert>
利用这个insertBatch测试批量插入数据
@Test
void testBatchInsert() {
long start = System.currentTimeMillis();
List<UserPO> userPOList = new ArrayList<>();
for (int i = 0; i < 10000; i++) {
UserPO userPO = new UserPO();
userPO.setUserName("baobao" + i);
userPO.setAge(18 + i);
userPOList.add(userPO);
}
userMapper.insertBatch(userPOList);
System.out.println(System.currentTimeMillis() - start);
}
耗时935毫秒，速度提升非常明显
但是当我们继续增大一次性插入的数据量，foreach的速度也将变得很慢。比如将插入数据量增加到80000条，此时耗时54.861秒
我们可以开启Batch类型的SqlSession来优化，yaml中配置mybatis.executor-type: batch
此时插入速度提升到31.826秒
进一步优化，配置mybatis.configuration.default-executor-type: batch
此时插入速度提升到1.986秒，效果非常明显