SpringBoot+MyBatis多表联合查询

写在前面

联合查询在实际工作中用的并不多，因为很多表的数据比较大，或者说未来比较大的表，都要谨慎使用联合查询

数据准备

建表语句

create table m_user
(id         bigint       not nullprimary key,username   varchar(64)  null,avatar     varchar(255) null,email      varchar(64)  null,password   varchar(64)  null,status     int(5)       not null,created    datetime     null,last_login datetime     null
);create index UK_USERNAMEon m_user (username);

create table m_blog
(id          bigint       not nullprimary key,user_id     bigint       not null,title       varchar(255) not null,description varchar(255) not null,content     longtext     null,created     datetime     not null on update CURRENT_TIMESTAMP,status      tinyint      null
)charset = utf8mb4;

插入数据

插入数据(数据很多都是没用的，那都是我平时自己测试用的，但是不影响使用

INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (1, '文章一', 'https://image-1300566513.cos.ap-guangzhou.myqcloud.com/upload/images/5a9f48118166308daba8b6da7e466aab.jpg', 'zhangxi@qq.com', '96e79218965eb72c92a549dd5a330112', 0, '2021-06-02 17:52:01', null);
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (2, '张liu', 'new', 'zhangliu@qq.com', '123456', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (3, '王wu', 'old', 'wangwu@qq.com', '2222222', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (4, 'si', 'sisi', 'sisisi', 'sisisisi', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (5, 'wu', 'wuwu', 'wuwuwu', 'wuwuwuwuwu', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (6, 'liu', 'liuliu', 'liuliuliu', 'liuliuliuliu', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (7, 'qi', 'qiqi', 'qiqiqi', 'qiqiqiqi', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (8, 'ba', 'baba', 'bababa', 'babababa', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (9, 'jiu', 'jiujiu', 'jiujiujiu', 'jiujiujiujiu', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (10, 'shi', 'shishi', 'shishishi', 'shishishishi', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (11, 'shiyi', 'shiyiyi', 'shiyiyiyi', 'shiyiyiyiyi', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (12, 'shier', 'shierer', 'shiererer', 'shierererer', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (13, 'shisan', 'shisansan', 'shisansansan', 'shisansansansan', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (14, 'shisi', 'shisisi', 'shisisisi', 'shisisisisi', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (15, 'shiwu', 'shiwuwu', 'shiwuwuwu', 'shiwuwuwuwu', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (16, 'shiliu', 'shiliuyliu', 'shiliuliuliu', 'shiliuliuliuliu', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (17, 'shiqi', 'shiqiqi', 'shiqiqiqi', 'shiqiqiqiqi', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (18, 'shiba', 'shibaba', 'shibabababa', 'shibababababa', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (19, 'shijiu', 'shijiujiu', 'shijiujiujiu', 'shijiujiujiujiu', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (21, 'ershi', 'ershishi', 'ershishishi', 'ershishishishi', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');
INSERT INTO testgroup.m_user (id, username, avatar, email, password, status, created, last_login) VALUES (22, 'ershiyi', 'ershiyiyi', 'ershiyiyiyi', 'ershiyiyiyiyi', 0, '2021-06-02 17:52:01', '2021-06-02 17:52:01');

INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (1, 1, '文章一', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-06-17 17:36:11', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (2, 1, '文章一2', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-06-17 17:36:11', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (3, 2, '文章一3', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-06-18 11:38:56', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (4, 3, '文章一4', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-06-18 11:38:56', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (5, 2, '文章一5', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-06-18 11:38:56', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (6, 4, '文章一6', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-06-18 11:38:56', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (7, 4, '文章一7', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-09-29 17:45:29', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (8, 5, '文章一8', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-11-14 14:12:37', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (9, 5, '文章一8', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-11-14 14:12:37', 0);
INSERT INTO testgroup.m_blog (id, user_id, title, description, content, created, status) VALUES (11, 6, '文章一8', '第一篇文章', '第一篇文章第一篇文章第一篇文章', '2021-11-14 14:12:37', 0);

entity层

package com.newcrud.entity;import lombok.Data;@Data
public class Blog {private Integer id;private Integer user_id;private String title;private String description;private String content;private String created;private Integer status;
}

package com.newcrud.entity;import lombok.Data;@Data
public class User {private Integer id;private String username;private String avatar;private String email;private String password;private Integer status;private String created;private String last_login;
}

多表联合查询的思路

首先我们来举一个例子,就像上面两张表

1、作者和多条博客为一对多的关系，也就是一条m_user对应了多条m_blog的数据

2、一条博客和作者为一对一的关系，也就是一条m_blog对应了一条m_user的数据

对于一对一，我们可以在m_blog的entity层加入一个User实例，但是对于一对多，我们就需要在m_user的entity层加入一个List，里面的内容为User。然后我们再去编写mybatis的xml文件，对于一对一，我们在xml文件中需要用到一个关键词为：association，意思为关联，对于一对多，我们在xml文件中需要用到一个关键词为：collection，意思为集合。

一对一

一篇博客对应一个作者

entity

首先，我们在m_bloh的entity里加入User

package com.newcrud.entity;import lombok.Data;@Data
public class Blog {private Integer id;private Integer user_id;private String title;private String description;private String content;private String created;private Integer status;private User user;
}

mapper

package com.newcrud.mapper;import com.newcrud.entity.Blog;
import org.apache.ibatis.annotations.Mapper;import java.util.List;
@Mapper
public interface BlogMapper {Blog getBlogById(Integer id);
}

service

package com.newcrud.service;import com.newcrud.entity.Blog;import java.util.List;public interface BlogService {Blog getBlogById(Integer id);
}

impl

package com.newcrud.service.impl;import com.newcrud.entity.Blog;
import com.newcrud.mapper.BlogMapper;
import com.newcrud.service.BlogService;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;import java.util.List;@Service
public class BlogServiceImpl implements BlogService {@AutowiredBlogMapper blogMapper;@Overridepublic Blog getBlogById (Integer id){return blogMapper.getBlogById(id);}
}

mybatis的xml文件

<?xml version="1.0" encoding="UTF-8" ?>
<!DOCTYPE mapperPUBLIC "-//mybatis.org//DTD Mapper 3.0//EN""http://mybatis.org/dtd/mybatis-3-mapper.dtd">
<!--这里一定要把namespace写全了，要不然会找不到对应的UserMapper里的每一个方法-->
<mapper namespace="com.newcrud.mapper.BlogMapper"><resultMap id="blog" type="Blog"><id column="id" property="id" javaType="Integer"></id><result column="user_id" property="user_id" javaType="Integer"></result><result column="title" property="title" javaType="String"></result><result column="description" property="description" javaType="String"></result><result column="content" property="content" javaType="String"></result><result column="created" property="created" javaType="String"></result><result column="status" property="status" javaType="Integer"></result><!--通过association关键字将User的属性引入进来，几乎每一个表对应的xml都有一个关于本表所有字段的resultMap，其实直接把User的全表resultMap内容复制过来即可--><association property="user" javaType="User"><id column="id" property="id" javaType="Integer"></id><result column="username" property="username" javaType="String"></result><result column="avatar" property="avatar" javaType="String"></result><result column="email" property="email" javaType="String"></result><result column="password" property="password" javaType="String"></result><result column="status" property="status" javaType="Integer"></result><result column="created" property="created" javaType="String"></result><result column="last_login" property="last_login" javaType="String"></result></association></resultMap><select id="getBlogById" resultType="Blog" resultMap="blog" parameterType="Integer">select a.* from m_user a,m_blog b<where><trim suffixOverrides="and" prefixOverrides="and">b.user_id=a.id and b.id= #{id}</trim></where></select>
</mapper>

测试类

package com.newcrud.service.impl;import com.newcrud.entity.Blog;
import com.newcrud.entity.User;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.boot.test.context.SpringBootTest;
import org.springframework.test.context.testng.AbstractTransactionalTestNGSpringContextTests;
import org.testng.annotations.Test;@SpringBootTestpublic class BlogServiceImplTest extends AbstractTransactionalTestNGSpringContextTests {@AutowiredBlogServiceImpl blogService;@Testpublic void getBlogByIdTest(){Blog blog= blogService.getBlogById(1);User user=blog.getUser();System.out.println(user);}
}

结果

User(id=1, username=文章一, avatar=https://image-1300566513.cos.ap-guangzhou.myqcloud.com/upload/images/5a9f48118166308daba8b6da7e466aab.jpg, email=zhangxi@qq.com, password=96e79218965eb72c92a549dd5a330112, status=0, created=2021-06-02 17:52:01, last_login=null)

如果xml文件不这样写呢

我们把m_user从xml文件中摘出来

    <select id="getBlogById" resultType="Blog" resultMap="blog" parameterType="Integer">select * from m_blog <where><trim suffixOverrides="and" prefixOverrides="and">id= #{id}</trim></where></select>

同样的测试类，结果

User(id=1, username=null, avatar=null, email=null, password=null, status=0, created=2021-06-17 17:36:11, last_login=null)

emmmm,经过分析得知，这个结果是将blog表里和user的表里有相同的字段，直接将blog的字段结果填充到了user的数据里，为什么这样谁能给我解释一下。。。

一对多

entity

关键：private List blogs; 是一个List

package com.newcrud.entity;import lombok.Data;import java.util.List;@Data
public class User {private Integer id;private String username;private String avatar;private String email;private String password;private Integer status;private String created;private String last_login;/*** 由于一个作者对应了多篇博客，所以这里用List来加入Blog* */private List<Blog> blogs;
}

最好呢，将Blog的entity还原成下面这个样子，我们暂时先不还原，看看结果如何。

package com.newcrud.entity;import lombok.Data;@Data
public class Blog {private Integer id;private Integer user_id;private String title;private String description;private String content;private String created;private Integer status;
}

mapper

package com.newcrud.mapper;import com.newcrud.entity.User;
import org.apache.ibatis.annotations.Mapper;
import org.apache.ibatis.annotations.Param;import java.util.List;/*** 这里有两种方法，一种是像我们一样使用@Mapper，一种是在CrudApplication上面新增了@MapperScan备注。直接扫描了整个包* https://www.cnblogs.com/JackpotHan/p/10286496.html* 不过不知道为啥，不加上这个好像是不行，没办法auto，稍微有那么点尴尬* **/
@Mapper
public interface UserMapper {/*** 一对多联合查询* */User getUserAndBlog(Integer id);
}

service

package com.newcrud.service;import com.newcrud.entity.User;
import com.newcrud.outInParam.PageRequest;
import com.newcrud.outInParam.PageResult;import java.util.List;public interface UserService {/*** 一对多联合查询* */User getUserAndBlog(Integer id);
}

impl

package com.newcrud.service.impl;import com.github.pagehelper.PageHelper;
import com.github.pagehelper.PageInfo;
import com.newcrud.entity.User;
import com.newcrud.mapper.UserMapper;
import com.newcrud.outInParam.PageRequest;
import com.newcrud.outInParam.PageResult;
import com.newcrud.service.UserService;
import com.newcrud.utils.PageUtils;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;import java.util.List;@Service
public class UserServiceImpl  implements UserService {@Autowiredprivate UserMapper userMapper;/*** 一对多联合查询* */@Overridepublic User getUserAndBlog(Integer id){return userMapper.getUserAndBlog(id);}

xml

<?xml version="1.0" encoding="UTF-8" ?>
<!DOCTYPE mapperPUBLIC "-//mybatis.org//DTD Mapper 3.0//EN""http://mybatis.org/dtd/mybatis-3-mapper.dtd">
<mapper namespace="com.newcrud.mapper.UserMapper"><resultMap id="userAndBlog" type="User"><!--重点1:这里的column对应的值为sql语句中对应字段的别名--><id column="uid" property="id" javaType="Integer"></id><result column="username" property="username" javaType="String"></result><result column="avatar" property="avatar" javaType="String"></result><result column="email" property="email" javaType="String"></result><result column="password" property="password" javaType="String"></result><result column="ustatus" property="status" javaType="Integer"></result><result column="ucreated" property="created" javaType="String"></result><result column="last_login" property="last_login" javaType="String"></result><!--重点2:这里一定要用ofType不能用javaType了，因为对应的Blog是一个List，如果用javaType的话就会报错数据类型不匹配--><collection property="blogs" ofType="Blog" ><id column="bid" property="id" javaType="Integer"></id><result column="user_id" property="user_id" javaType="Integer"></result><result column="title" property="title" javaType="String"></result><result column="description" property="description" javaType="String"></result><result column="content" property="content" javaType="String"></result><result column="bcreated" property="created" javaType="String"></result><result column="bstatus" property="status" javaType="Integer"></result></collection></resultMap><select id="getUserAndBlog" resultMap="userAndBlog" parameterType="Integer"><!--重点3：我们查询出了一条记录所有的值，但是两个表有很多字段都是重名的怎么办，那就需要为同名字段起一个别名，别忘了修改resultMap对应字段的column值哦-->select a.id uid,a.username,a.avatar,a.email,a.password,a.status ustatus,a.created ucreated,a.last_login,b.id bid,b.user_id,b.title,b.description,b.content,b.created bcreated,b.status bstatus from m_user a , m_blog b where a.id=b.user_id and a.id=#{id}</select>
</mapper>

测试类

package com.newcrud.service.impl;import com.newcrud.entity.User;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.boot.test.context.SpringBootTest;
import org.springframework.test.context.testng.AbstractTestNGSpringContextTests;
import org.testng.annotations.Test;import java.util.List;import static org.testng.Assert.*;@SpringBootTest
public class UserServiceImplTest  extends AbstractTestNGSpringContextTests {@AutowiredUserServiceImpl userService;@Testpublic void testGetUserById() {User users = userService.getUserAndBlog(1);System.out.println(users);}

结果

User(id=1, username=文章一, avatar=https://image-1300566513.cos.ap-guangzhou.myqcloud.com/upload/images/5a9f48118166308daba8b6da7e466aab.jpg, email=zhangxi@qq.com, password=96e79218965eb72c92a549dd5a330112, status=0, created=2021-06-02 17:52:01, last_login=null, blogs=[Blog(id=1, user_id=1, title=文章一, description=第一篇文章, content=第一篇文章第一篇文章第一篇文章, created=2021-06-17 17:36:11, status=0, user=null), Blog(id=2, user_id=1, title=文章一2, description=第一篇文章, content=第一篇文章第一篇文章第一篇文章, created=2021-06-17 17:36:11, status=0, user=null)])

有没有注意到blog里有一段user=null，我们把entity层的Blog还原就行了

优化

一个小小的优化，一是我们并不需要返回那么多字段信息，二是实际工作中这个xml里的sql，其实更多的是使用union的方式，而不是像我们上面的那种

entity

package com.newcrud.entity;import lombok.Data;import java.util.List;@Data
public class User {private Integer id;private String username;private String avatar;private String email;private String password;private Integer status;private String created;private String last_login;/*** 由于一个作者对应了多篇博客，所以这里用List来加入Blog* */private List<Blog> blogs;
}

mapper

package com.newcrud.mapper;import com.newcrud.entity.User;
import org.apache.ibatis.annotations.Mapper;
import org.apache.ibatis.annotations.Param;import java.util.List;/*** 这里有两种方法，一种是像我们一样使用@Mapper，一种是在CrudApplication上面新增了@MapperScan备注。直接扫描了整个包* https://www.cnblogs.com/JackpotHan/p/10286496.html* 不过不知道为啥，不加上这个好像是不行，没办法auto，稍微有那么点尴尬* **/
@Mapper
public interface UserMapper {/*** 一对多联合查询-union方式* */User getAllUserAndBlog(Integer id);
}

service

package com.newcrud.service;import com.newcrud.entity.User;
import com.newcrud.outInParam.PageRequest;
import com.newcrud.outInParam.PageResult;import java.util.List;public interface UserService {/*** 一对多联合查询-union方式* */User getAllUserAndBlog(Integer id);
}

impl

package com.newcrud.service.impl;import com.github.pagehelper.PageHelper;
import com.github.pagehelper.PageInfo;
import com.newcrud.entity.User;
import com.newcrud.mapper.UserMapper;
import com.newcrud.outInParam.PageRequest;
import com.newcrud.outInParam.PageResult;
import com.newcrud.service.UserService;
import com.newcrud.utils.PageUtils;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;import java.util.List;@Service
public class UserServiceImpl  implements UserService {@Autowiredprivate UserMapper userMapper;/*** 一对多联合查询-union方式* */@Overridepublic User getAllUserAndBlog(Integer id){return userMapper.getAllUserAndBlog(id);}
}

mybatis的xml文件

<?xml version="1.0" encoding="UTF-8" ?>
<!DOCTYPE mapperPUBLIC "-//mybatis.org//DTD Mapper 3.0//EN""http://mybatis.org/dtd/mybatis-3-mapper.dtd">
<mapper namespace="com.newcrud.mapper.UserMapper">
<resultMap id="AllUserAndBlog" type="User"><id column="uid" property="id" javaType="Integer"></id><result column="username" property="username" javaType="String"></result><result column="email" property="email" javaType="String"></result><collection property="blogs" ofType="Blog"><id column="bid" property="id" javaType="Integer"></id><result column="title" property="title" javaType="String"></result><result column="description" property="description" javaType="String"></result></collection></resultMap><select id="getAllUserAndBlog" resultMap="AllUserAndBlog" parameterType="Integer">select a.id uid,a.username,a.email,b.id bid,b.title,b.description  from m_user a left outer join m_blog b on a.id = b.user_id where a.id=#{id}</select>
</mapper>

测试类

package com.newcrud.service.impl;import com.newcrud.entity.Blog;
import com.newcrud.entity.User;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.boot.test.context.SpringBootTest;
import org.springframework.test.context.testng.AbstractTestNGSpringContextTests;
import org.testng.annotations.Test;import javax.jws.soap.SOAPBinding;
import java.util.List;import static org.testng.Assert.*;@SpringBootTest
public class UserServiceImplTest  extends AbstractTestNGSpringContextTests {@AutowiredUserServiceImpl userService;@Testpublic void testgetAllUserAndBlog(){User user = userService.getAllUserAndBlog(1);System.out.println(user);}}

结果

User(id=1, username=文章一, avatar=null, email=zhangxi@qq.com, password=null, status=null, created=null, last_login=null, blogs=[Blog(id=1, user_id=null, title=文章一, description=第一篇文章, content=null, created=null, status=null), Blog(id=2, user_id=null, title=文章一2, description=第一篇文章, content=null, created=null, status=null)])