多表查询
本节重点：
- 多表连接查询
- 符合条件连接查询
- 子查询
准备工作：准备两张表，部门表（department）、员工表（employee）
1create table department( 2id int, 3name varchar(20) 4); 5 6create table employee( 7id int primary key auto_increment, 8name varchar(20), 9sex enum('male','female') not null default 'male', 10age int, 11dep_id int 12); 13 14#插入数据 15insert into department values 16(200,'技术'), 17(201,'人力资源'), 18(202,'销售'), 19(203,'运营'); 20 21insert into employee(name,sex,age,dep_id) values 22('egon','male',18,200), 23('alex','female',48,201), 24('wupeiqi','male',38,201), 25('yuanhao','female',28,202), 26('nvshen','male',18,200), 27('xiaomage','female',18,204) 28; 29 30# 查看表结构和数据 31mysql> desc department; 32+-------+-------------+------+-----+---------+-------+ 33| Field | Type | Null | Key | Default | Extra | 34+-------+-------------+------+-----+---------+-------+ 35| id | int(11) | YES | | NULL | | 36| name | varchar(20) | YES | | NULL | | 37+-------+-------------+------+-----+---------+-------+ 382 rows in set (0.19 sec) 39 40mysql> desc employee; 41+--------+-----------------------+------+-----+---------+----------------+ 42| Field | Type | Null | Key | Default | Extra | 43+--------+-----------------------+------+-----+---------+----------------+ 44| id | int(11) | NO | PRI | NULL | auto_increment | 45| name | varchar(20) | YES | | NULL | | 46| sex | enum('male','female') | NO | | male | | 47| age | int(11) | YES | | NULL | | 48| dep_id | int(11) | YES | | NULL | | 49+--------+-----------------------+------+-----+---------+----------------+ 505 rows in set (0.01 sec) 51 52mysql> select * from department; 53+------+--------------+ 54| id | name | 55+------+--------------+ 56| 200 | 技术 | 57| 201 | 人力资源 | 58| 202 | 销售 | 59| 203 | 运营 | 60+------+--------------+ 614 rows in set (0.02 sec) 62 63mysql> select * from employee; 64+----+----------+--------+------+--------+ 65| id | name | sex | age | dep_id | 66+----+----------+--------+------+--------+ 67| 1 | egon | male | 18 | 200 | 68| 2 | alex | female | 48 | 201 | 69| 3 | wupeiqi | male | 38 | 201 | 70| 4 | yuanhao | female | 28 | 202 | 71| 5 | nvshen | male | 18 | 200 | 72| 6 | xiaomage | female | 18 | 204 | 73+----+----------+--------+------+--------+ 746 rows in set (0.00 sec)
ps:观察两张表，发现department表中id=203部门在employee中没有对应的员工，发现employee中id=6的员工在department表中没有对应关系。
一、多表连接查询
两张表的准备工作已完成，比如现在我要查询的员工信息以及该员工所在的部门。从该题中，我们看出既要查员工又要查该员工的部门，肯定要将两张表进行连接查询，多表连接查询。
重点：外链接语法
语法：
1SELECT 字段列表 2 FROM 表1 INNER|LEFT|RIGHT JOIN 表2 3 ON 表1.字段 = 表2.字段;
（1）先看第一种情况交叉连接：不适用任何匹配条件。生成笛卡尔积（关于笛卡尔积的含义，大家百度自行补脑）。
1mysql> select * from employee,department; 2+----+----------+--------+------+--------+------+--------------+ 3| id | name | sex | age | dep_id | id | name | 4+----+----------+--------+------+--------+------+--------------+ 5| 1 | egon | male | 18 | 200 | 200 | 技术 | 6| 1 | egon | male | 18 | 200 | 201 | 人力资源 | 7| 1 | egon | male | 18 | 200 | 202 | 销售 | 8| 1 | egon | male | 18 | 200 | 203 | 运营 | 9| 2 | alex | female | 48 | 201 | 200 | 技术 | 10| 2 | alex | female | 48 | 201 | 201 | 人力资源 | 11| 2 | alex | female | 48 | 201 | 202 | 销售 | 12| 2 | alex | female | 48 | 201 | 203 | 运营 | 13| 3 | wupeiqi | male | 38 | 201 | 200 | 技术 | 14| 3 | wupeiqi | male | 38 | 201 | 201 | 人力资源 | 15| 3 | wupeiqi | male | 38 | 201 | 202 | 销售 | 16| 3 | wupeiqi | male | 38 | 201 | 203 | 运营 | 17| 4 | yuanhao | female | 28 | 202 | 200 | 技术 | 18| 4 | yuanhao | female | 28 | 202 | 201 | 人力资源 | 19| 4 | yuanhao | female | 28 | 202 | 202 | 销售 | 20| 4 | yuanhao | female | 28 | 202 | 203 | 运营 | 21| 5 | nvshen | male | 18 | 200 | 200 | 技术 | 22| 5 | nvshen | male | 18 | 200 | 201 | 人力资源 | 23| 5 | nvshen | male | 18 | 200 | 202 | 销售 | 24| 5 | nvshen | male | 18 | 200 | 203 | 运营 | 25| 6 | xiaomage | female | 18 | 204 | 200 | 技术 | 26| 6 | xiaomage | female | 18 | 204 | 201 | 人力资源 | 27| 6 | xiaomage | female | 18 | 204 | 202 | 销售 | 28| 6 | xiaomage | female | 18 | 204 | 203 | 运营 |
（2）内连接：只连接匹配的行
1#找两张表共有的部分，相当于利用条件从笛卡尔积结果中筛选出了匹配的结果 2#department没有204这个部门，因而employee表中关于204这条员工信息没有匹配出来 3mysql> select employee.id,employee.name,employee.age,employee.sex,department.name from employee inner join department on employee.dep_id=department.id; 4+----+---------+------+--------+--------------+ 5| id | name | age | sex | name | 6+----+---------+------+--------+--------------+ 7| 1 | egon | 18 | male | 技术 | 8| 2 | alex | 48 | female | 人力资源 | 9| 3 | wupeiqi | 38 | male | 人力资源 | 10| 4 | yuanhao | 28 | female | 销售 | 11| 5 | nvshen | 18 | male | 技术 | 12+----+---------+------+--------+--------------+ 135 rows in set (0.00 sec) 14 15#上述sql等同于 16mysql> select employee.id,employee.name,employee.age,employee.sex,department.name from employee,department where employee.dep_id=department.id;
（3）外链接之左连接：优先显示左表全部记录
1#以左表为准，即找出所有员工信息，当然包括没有部门的员工 2#本质就是：在内连接的基础上增加左边有，右边没有的结果 3mysql> select employee.id,employee.name,department.name as depart_name from employee left join department on employee.dep_id=department.id; 4+----+----------+--------------+ 5| id | name | depart_name | 6+----+----------+--------------+ 7| 1 | egon | 技术 | 8| 5 | nvshen | 技术 | 9| 2 | alex | 人力资源 | 10| 3 | wupeiqi | 人力资源 | 11| 4 | yuanhao | 销售 | 12| 6 | xiaomage | NULL | 13+----+----------+--------------+ 146 rows in set (0.00 sec)
（4） 外链接之右连接：优先显示右表全部记录
1#以右表为准，即找出所有部门信息，包括没有员工的部门 2#本质就是：在内连接的基础上增加右边有，左边没有的结果 3mysql> select employee.id,employee.name,department.name as depart_name from employee right join department on employee.dep_id=department.id; 4+------+---------+--------------+ 5| id | name | depart_name | 6+------+---------+--------------+ 7| 1 | egon | 技术 | 8| 2 | alex | 人力资源 | 9| 3 | wupeiqi | 人力资源 | 10| 4 | yuanhao | 销售 | 11| 5 | nvshen | 技术 | 12| NULL | NULL | 运营 | 13+------+---------+--------------+ 146 rows in set (0.00 sec)
（5） 全外连接：显示左右两个表全部记录(了解)
1#外连接：在内连接的基础上增加左边有右边没有的和右边有左边没有的结果 2#注意：mysql不支持全外连接 full JOIN 3#强调：mysql可以使用此种方式间接实现全外连接 4语法：select * from employee left join department on employee.dep_id = department.id 5 union all 6 select * from employee right join department on employee.dep_id = department.id; 7 8 mysql> select * from employee left join department on employee.dep_id = department.id 9 union 10 select * from employee right join department on employee.dep_id = department.id 11 ; 12+------+----------+--------+------+--------+------+--------------+ 13| id | name | sex | age | dep_id | id | name | 14+------+----------+--------+------+--------+------+--------------+ 15| 1 | egon | male | 18 | 200 | 200 | 技术 | 16| 5 | nvshen | male | 18 | 200 | 200 | 技术 | 17| 2 | alex | female | 48 | 201 | 201 | 人力资源 | 18| 3 | wupeiqi | male | 38 | 201 | 201 | 人力资源 | 19| 4 | yuanhao | female | 28 | 202 | 202 | 销售 | 20| 6 | xiaomage | female | 18 | 204 | NULL | NULL | 21| NULL | NULL | NULL | NULL | NULL | 203 | 运营 | 22+------+----------+--------+------+--------+------+--------------+ 237 rows in set (0.01 sec) 24 25#注意 union与union all的区别：union会去掉相同的纪录
二、符合条件连接查询
示例1：以内连接的方式查询employee和department表，并且employee表中的age字段值必须大于25，即找出年龄大于25岁的员工以及员工所在的部门
1select employee.name,department.name from employee inner join department 2 on employee.dep_id = department.id 3 where age > 25;
示例2：以内连接的方式查询employee和department表，并且以age字段的升序方式显示。
1select employee.id,employee.name,employee.age,department.name from employee,department 2 where employee.dep_id = department.id 3 and age > 25 4 order by age asc;
三、子查询
1#1：子查询是将一个查询语句嵌套在另一个查询语句中。 2#2：内层查询语句的查询结果，可以为外层查询语句提供查询条件。 3#3：子查询中可以包含：IN、NOT IN、ANY、ALL、EXISTS 和 NOT EXISTS等关键字 4#4：还可以包含比较运算符：= 、 !=、> 、<等
1子查询(嵌套查询): 查多次, 多个select 2 3注意: 第一次的查询结果可以作为第二次的查询的 条件 或者 表名 使用. 4 5子查询中可以包含：IN、NOT IN、ANY、ALL、EXISTS 和 NOT EXISTS等关键字. 还可以包含比较运算符：= 、 !=、> 、<等. 6 7 1.作为表名使用 8select * from (select * from person) as 表名; 9 10ps:大家需要注意的是: 一条语句中可以有多个这样的子查询,在执行时,最里层括号(sql语句) 具有优先执行权.<br>注意: as 后面的表名称不能加引号('') 11 2.求最大工资那个人的姓名和薪水 12 13复制代码 141.求最大工资 15select max(salary) from person; 162.求最大工资那个人叫什么 17select name,salary from person where salary=53000; 18 19合并 20select name,salary from person where salary=(select max(salary) from person); 21复制代码 22复制代码 233. 求工资高于所有人员平均工资的人员 24 25复制代码 261.求平均工资 27select avg(salary) from person; 28 292.工资大于平均工资的 人的姓名、工资 30select name,salary from person where salary > 21298.625; 31 32合并 33select name,salary from person where salary >(select avg(salary) from person); 34复制代码
子查询示例
例子：
（1）带in关键字的子查询
1#查询平均年龄在25岁以上的部门名 2select id,name from department 3 where id in 4 (select dep_id from employee group by dep_id having avg(age) > 25); 5# 查看技术部员工姓名 6select name from employee 7 where dep_id in 8 (select id from department where name='技术'); 9#查看不足1人的部门名 10select name from department 11 where id not in 12 (select dep_id from employee group by dep_id);
（2）带比较运算符的子查询
1#比较运算符：=、!=、>、>=、<、<=、<> 2#查询大于所有人平均年龄的员工名与年龄 3mysql> select name,age from employee where age > (select avg(age) from employee); 4+---------+------+ 5| name | age | 6+---------+------+ 7| alex | 48 | 8| wupeiqi | 38 | 9+---------+------+ 10 11#查询大于部门内平均年龄的员工名、年龄 12思路： 13 （1）先对员工表(employee)中的人员分组(group by)，查询出dep_id以及平均年龄。 (2)将查出的结果作为临时表,再对根据临时表的dep_id和employee的dep_id作为筛选条件将employee表和临时表进行内连接。 14 (3)最后再将employee员工的年龄是大于平均年龄的员工名字和年龄筛选。 15 16 17 18mysql> select t1.name,t1.age from employee as t1 19 inner join 20 (select dep_id,avg(age) as avg_age from employee group by dep_id) as t2 21 on t1.dep_id = t2.dep_id 22 where t1.age > t2.avg_age; 23+------+------+ 24| name | age | 25+------+------+ 26| alex | 48 |
(3)带EXISTS关键字的子查询
1#EXISTS关字键字表示存在。在使用EXISTS关键字时，内层查询语句不返回查询的记录。而是返回一个真假值。True或False 2#当返回True时，外层查询语句将进行查询；当返回值为False时，外层查询语句不进行查询 3#department表中存在dept_id=203，Ture 4mysql> select * from employee where exists (select id from department where id=200); 5+----+----------+--------+------+--------+ 6| id | name | sex | age | dep_id | 7+----+----------+--------+------+--------+ 8| 1 | egon | male | 18 | 200 | 9| 2 | alex | female | 48 | 201 | 10| 3 | wupeiqi | male | 38 | 201 | 11| 4 | yuanhao | female | 28 | 202 | 12| 5 | nvshen | male | 18 | 200 | 13| 6 | xiaomage | female | 18 | 204 | 14+----+----------+--------+------+--------+ 15#department表中存在dept_id=205，False 16mysql> select * from employee where exists (select id from department where id=204); 17Empty set (0.00 sec)
小练习：
查询每个部门最新入职的那位员工
1#创建表 2create table employee( 3id int not null unique auto_increment, 4name varchar(20) not null, 5sex enum('male','female') not null default 'male', #大部分是男的 6age int(3) unsigned not null default 28, 7hire_date date not null, 8post varchar(50), 9post_comment varchar(100), 10salary double(15,2), 11office int, #一个部门一个屋子 12depart_id int 13); 14 15 16#查看表结构 17mysql> desc employee; 18+--------------+-----------------------+------+-----+---------+----------------+ 19| Field | Type | Null | Key | Default | Extra | 20+--------------+-----------------------+------+-----+---------+----------------+ 21| id | int(11) | NO | PRI | NULL | auto_increment | 22| name | varchar(20) | NO | | NULL | | 23| sex | enum('male','female') | NO | | male | | 24| age | int(3) unsigned | NO | | 28 | | 25| hire_date | date | NO | | NULL | | 26| post | varchar(50) | YES | | NULL | | 27| post_comment | varchar(100) | YES | | NULL | | 28| salary | double(15,2) | YES | | NULL | | 29| office | int(11) | YES | | NULL | | 30| depart_id | int(11) | YES | | NULL | | 31+--------------+-----------------------+------+-----+---------+----------------+ 32 33#插入记录 34#三个部门：教学，销售，运营 35insert into employee(name,sex,age,hire_date,post,salary,office,depart_id) values 36('egon','male',18,'20170301','办事处外交大使',7300.33,401,1), #以下是教学部 37('alex','male',78,'20150302','teacher',1000000.31,401,1), 38('wupeiqi','male',81,'20130305','teacher',8300,401,1), 39('yuanhao','male',73,'20140701','teacher',3500,401,1), 40('liwenzhou','male',28,'20121101','teacher',2100,401,1), 41('jingliyang','female',18,'20110211','teacher',9000,401,1), 42('jinxin','male',18,'19000301','teacher',30000,401,1), 43('成龙','male',48,'20101111','teacher',10000,401,1), 44 45('歪歪','female',48,'20150311','sale',3000.13,402,2),#以下是销售部门 46('丫丫','female',38,'20101101','sale',2000.35,402,2), 47('丁丁','female',18,'20110312','sale',1000.37,402,2), 48('星星','female',18,'20160513','sale',3000.29,402,2), 49('格格','female',28,'20170127','sale',4000.33,402,2), 50 51('张野','male',28,'20160311','operation',10000.13,403,3), #以下是运营部门 52('程咬金','male',18,'19970312','operation',20000,403,3), 53('程咬银','female',18,'20130311','operation',19000,403,3), 54('程咬铜','male',18,'20150411','operation',18000,403,3), 55('程咬铁','female',18,'20140512','operation',17000,403,3)