本文源码：GitHub·点这里 || GitEE·点这里

一、连接查询

图解示意图

1、建表语句

部门和员工关系表：

CREATE TABLE `tb_dept` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT ‘主键ID‘,
  `deptName` varchar(30) DEFAULT NULL COMMENT ‘部门名称‘,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=6 DEFAULT CHARSET=utf8;
CREATE TABLE `tb_emp` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT ‘主键ID‘,
  `empName` varchar(20) DEFAULT NULL COMMENT ‘员工名称‘,
  `deptId` int(11) DEFAULT ‘0‘ COMMENT ‘部门ID‘,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=8 DEFAULT CHARSET=utf8;

2、七种连接查询

• 图1：左外连接

select t1.*,t2.empName,t2.deptId
from tb_dept t1 LEFT JOIN tb_emp t2 on t1.id=t2.deptId;

• 图2：右外连接

select t1.*,t2.empName,t2.deptId
from tb_dept t1 RIGHT JOIN tb_emp t2 on t1.id=t2.deptId;

• 图3：内连接

select t1.*,t2.empName,t2.deptId
from tb_dept t1 inner join tb_emp t2 on t1.id=t2.deptId;

• 图4：左连接

查询tb_dept表特有的地方。

select t1.*,t2.empName,t2.deptId
from tb_dept t1 LEFT JOIN tb_emp t2 on t1.id=t2.deptId
WHERE t2.deptId IS NULL;

• 图5：右连接

查询tb_emp表特有的地方。

select t1.*,t2.empName,t2.deptId
from tb_dept t1 RIGHT JOIN tb_emp t2 on t1.id=t2.deptId
WHERE t1.id IS NULL;

• 图6：全连接

select t1.*,t2.empName,t2.deptId
from tb_dept t1 LEFT JOIN tb_emp t2 on t1.id=t2.deptId
UNION
select t1.*,t2.empName,t2.deptId
from tb_dept t1 RIGHT JOIN tb_emp t2 on t1.id=t2.deptId

• 图7：全不连接

查询两张表互不关联到的数据。

select t1.*,t2.empName,t2.deptId
from tb_dept t1 RIGHT JOIN tb_emp t2 on t1.id=t2.deptId
WHERE t1.id IS NULL
UNION
select t1.*,t2.empName,t2.deptId
from tb_dept t1 LEFT JOIN tb_emp t2 on t1.id=t2.deptId
WHERE t2.deptId IS NULL

二、时间日期查询

1、建表语句

CREATE TABLE `ms_consume` (
  `id` int(11) NOT NULL AUTO_INCREMENT COMMENT ‘主键ID‘,
  `user_id` int(11) NOT NULL COMMENT ‘用户ID‘,
  `user_name` varchar(20) NOT NULL COMMENT ‘用户名‘,
  `consume_money` decimal(20,2) DEFAULT ‘0.00‘ COMMENT ‘消费金额‘,
  `create_time` datetime DEFAULT CURRENT_TIMESTAMP COMMENT ‘创建时间‘,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=9 DEFAULT CHARSET=utf8 COMMENT=‘消费表‘;

2、日期统计案例

• 日期范围内首条数据

场景：产品日常运营活动中，经常见到这样规则：活动时间内，首笔消费满多少，优惠多少。

SELECT * FROM
    (
        SELECT * FROM ms_consume
        WHERE
            create_time
        BETWEEN ‘2019-12-10 00:00:00‘ AND ‘2019-12-18 23:59:59‘
        ORDER BY create_time
    ) t1
GROUP BY t1.user_id ;

• 日期之间时差

场景：常用的倒计时场景

SELECT t1.*,
       timestampdiff(SECOND,NOW(),t1.create_time) second_diff
FROM ms_consume t1 WHERE t1.id=‘9‘ ;

• 查询今日数据

-- 方式一
SELECT * FROM ms_consume
WHERE DATE_FORMAT(NOW(),‘%Y-%m-%d‘)=DATE_FORMAT(create_time,‘%Y-%m-%d‘);
-- 方式二
SELECT * FROM ms_consume
WHERE TO_DAYS(now())=TO_DAYS(create_time) ;

• 时间范围统计

场景：统计近七日内，消费次数大于两次的用户。

SELECT user_id,user_name,COUNT(user_id) userIdSum
FROM ms_consume WHERE create_time>date_sub(NOW(), interval ‘7‘ DAY)
GROUP BY user_id  HAVING userIdSum>1;

• 日期范围内平均值

场景：指定日期范围内的平均消费，并排序。

SELECT * FROM
    (
        SELECT user_id,user_name,
            AVG(consume_money) avg_money
        FROM ms_consume t
        WHERE t.create_time BETWEEN ‘2019-12-10 00:00:00‘
                            AND ‘2019-12-18 23:59:59‘
        GROUP BY user_id
    ) t1
ORDER BY t1.avg_money DESC;

三、树形表查询

1、建表语句

CREATE TABLE ms_city_sort (
    `id` INT (11) NOT NULL AUTO_INCREMENT COMMENT ‘主键ID‘,
    `city_name` VARCHAR (50) NOT NULL DEFAULT ‘‘ COMMENT ‘城市名称‘,
    `city_code` VARCHAR (50) NOT NULL DEFAULT ‘‘ COMMENT ‘城市编码‘,
    `parent_id` INT (11) NOT NULL DEFAULT ‘0‘ COMMENT ‘父级ID‘,
    `state` INT (11) NOT NULL DEFAULT ‘1‘ COMMENT ‘状态：1启用,2停用‘,
    `create_time` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP COMMENT ‘创建时间‘,
    `update_time` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP COMMENT ‘修改时间‘,
    PRIMARY KEY (id)
) ENGINE = INNODB DEFAULT CHARSET = utf8 COMMENT = ‘城市分类管理‘;

2、直接SQL查询

SELECT t1.*, t2.parentName
FROM ms_city_sort t1
LEFT JOIN (
    SELECT
        m1.id,m2.city_name parentName
    FROM
        ms_city_sort m1,ms_city_sort m2
    WHERE m1.parent_id = m2.id
    AND m1.parent_id > 0
) t2 ON t1.id = t2.id;

3、函数查询

• 查询父级名称

DROP FUNCTION IF EXISTS get_city_parent_name;
CREATE FUNCTION `get_city_parent_name`(pid INT)
RETURNS varchar(50) CHARSET utf8
begin
    declare parentName VARCHAR(50) DEFAULT NULL;
    SELECT city_name FROM ms_city_sort WHERE id=pid into parentName;
    return parentName;
end

SELECT t1.*,get_city_parent_name(t1.parent_id) parentName FROM ms_city_sort t1 ;

• 查询根节点子级

DROP FUNCTION IF EXISTS get_root_child;
CREATE FUNCTION `get_root_child`(rootId INT)
    RETURNS VARCHAR(1000) CHARSET utf8
    BEGIN
        DECLARE resultIds VARCHAR(500);
        DECLARE nodeId VARCHAR(500);
        SET resultIds = ‘%‘;
        SET nodeId = cast(rootId as CHAR);
        WHILE nodeId IS NOT NULL DO
            SET resultIds = concat(resultIds,‘,‘,nodeId);
            SELECT group_concat(id) INTO nodeId
            FROM ms_city_sort WHERE FIND_IN_SET(parent_id,nodeId)>0;
        END WHILE;
        RETURN resultIds;
END  ;

SELECT * FROM ms_city_sort WHERE FIND_IN_SET(id,get_root_child(5)) ORDER BY id ;

四、源代码地址

GitHub·地址
https://github.com/cicadasmile/mysql-data-base
GitEE·地址
https://gitee.com/cicadasmile/mysql-data-base

原文地址：https://blog.51cto.com/14439672/2461529