mysql 表连接
表连接
当需同时显示多个表中的字段时,就可以用表连接来实现这样的功能。表连接分为内连接和外连接。它们区别在于内连接仅选出两张表中互相匹配的记录。而外连接会显示其他不匹配的记录。
列如有一张表存着员工姓名以及入职日期还有工资和部门id ,另一张表存着部门id和部门名
首先我们先创建这2个表。
创建员工信息表:
创建员工信息表:
create table emp(
ename varchar(20),
hiredate date,
sal decimal(10,2),
deptno int(3));
#创建部门信息表:
create table dept(
deptno int(3),
deptname varchar(20));
#插入多条信息成如下:
mysql> select * from emp;
+--------+------------+---------+--------+
| ename | hiredate | sal | deptno |
+--------+------------+---------+--------+
| zzx | 2000-01-01 | 2000.00 | 1 |
| lisa | 2003-02-01 | 4000.00 | 2 |
| bjguan | 2004-04-02 | 5000.00 | 1 |
| bzshen | 2005-04-01 | 4000.00 | 3 |
+--------+------------+---------+--------+
mysql> select * from dept;
+--------+----------+
| deptno | deptname |
+--------+----------+
| 1 | tech |
| 2 | sale |
| 3 | hr |
+--------+----------+
3 rows in set (0.00 sec)
笛卡尔积
笛卡尔积是不加任何约束连接的多表查询,语法:select * from 表1,表2 ;显示的行数是表一查询出来的行数x表二的行数:
mysql> select * from emp,dept;
+--------+------------+---------+--------+--------+----------+
| ename | hiredate | sal | deptno | deptno | deptname |
+--------+------------+---------+--------+--------+----------+
| zzx | 2000-01-01 | 2000.00 | 1 | 1 | tech |
| zzx | 2000-01-01 | 2000.00 | 1 | 2 | sale |
| zzx | 2000-01-01 | 2000.00 | 1 | 3 | hr |
| lisa | 2003-02-01 | 4000.00 | 2 | 1 | tech |
| lisa | 2003-02-01 | 4000.00 | 2 | 2 | sale |
| lisa | 2003-02-01 | 4000.00 | 2 | 3 | hr |
| bjguan | 2004-04-02 | 5000.00 | 1 | 1 | tech |
| bjguan | 2004-04-02 | 5000.00 | 1 | 2 | sale |
| bjguan | 2004-04-02 | 5000.00 | 1 | 3 | hr |
| bzshen | 2005-04-01 | 4000.00 | 3 | 1 | tech |
| bzshen | 2005-04-01 | 4000.00 | 3 | 2 | sale |
| bzshen | 2005-04-01 | 4000.00 | 3 | 3 | hr |
+--------+------------+---------+--------+--------+----------+
12 rows in set (0.00 sec)
内连接
语法:
select d1.ename,d2.deptname from emp d1 inner join dept d2 on d1.deptno=d2.deptno;
例如查询出所有员工的名字和部门名称:
#d1 d2是分别给两张表起的别名语法:表名+空格+别名
select d1.ename,d2.deptname from emp d1,dept d2 where d1.deptno=d2.deptno;
+--------+----------+
| ename | deptname |
+--------+----------+
| zzx | tech |
| lisa | sale |
| bjguan | tech |
| bzshen | hr |
+--------+----------+
4 rows in set (0.00 sec)
外连接
外连接又分左连接和右连接,具体定义如下:
1、左连接:包含所有左边表的记录甚至是右边表没有和它匹配的记录
2、右连接:包含所有右边表的记录甚至是左边表没有和它匹配的记录
左连接
我先在员工信息表中插入一条这样的信息:
insert into emp values('dony','2005-2-5',2000,4) ;#注意4这个部门id在部门表中是没有的
mysql> select * from emp;
+--------+------------+---------+--------+
| ename | hiredate | sal | deptno |
+--------+------------+---------+--------+
| zzx | 2000-01-01 | 2000.00 | 1 |
| lisa | 2003-02-01 | 4000.00 | 2 |
| bjguan | 2004-04-02 | 5000.00 | 1 |
| bzshen | 2005-04-01 | 4000.00 | 3 |
| dony | 2005-02-05 | 2000.00 | 4 |
+--------+------------+---------+--------+
5 rows in set (0.00 sec)
mysql> select * from dept;
+--------+----------+
| deptno | deptname |
+--------+----------+
| 1 | tech |
| 2 | sale |
| 3 | hr |
+--------+----------+
3 rows in set (0.00 sec)
左连接查询
语法:select 字段名 from 表1 left join 表2 on 连接条件;
select d1.ename,d2.deptname from emp d1 left join dept d2 on d1.deptno=d2.deptno;
mysql> select d1.ename,d2.deptname from emp d1 left join dept d2 on d1.deptno=d2.deptno;
+--------+----------+
| ename | deptname |
+--------+----------+
| zzx | tech |
| bjguan | tech |
| lisa | sale |
| bzshen | hr |
| dony | NULL |
+--------+----------+
5 rows in set (0.00 sec)
右连接
语法:select 字段名 from 表1 right join 表2 on 连接条件
select d1.ename ,d2.deptname from emp d1 right join dept d2 on d1.deptno = d2.deptno;
mysql> select d1.ename ,d2.deptname from emp d1 right join dept d2 on d1.deptno = d2.deptno;
+--------+----------+
| ename | deptname |
+--------+----------+
| zzx | tech |
| lisa | sale |
| bjguan | tech |
| bzshen | hr |
+--------+----------+
4 rows in set (0.00 sec)
总结:以谁为连接就以谁为准优先显示全它的信息决定行数,另一张表只显示匹配上的信息,匹配不上的根据行数显示去掉或者不空;
子查询
某些情况下,当进行查询的时候,需要的条件是另一个select语句的结果,这个时候就要用到子查询,用于子查询的关键字主要包括in、not in、=、!=、exists、not exists等。
例如,从emp表中查询出所有部门在dept表中的所有记录:
语法:select 字段 from 表1 where 字段 in (select 语句)
select * from emp where deptno in (select deptno from dept);
+--------+------------+---------+--------+
| ename | hiredate | sal | deptno |
+--------+------------+---------+--------+
| zzx | 2000-01-01 | 2000.00 | 1 |
| lisa | 2003-02-01 | 4000.00 | 2 |
| bjguan | 2004-04-02 | 5000.00 | 1 |
| bzshen | 2005-04-01 | 4000.00 | 3 |
+--------+------------+---------+--------+
4 rows in set (0.00 sec)
如果子查询记录数唯一,还可以用=代替 in
select * from emp where deptno=(select deptno from dept where deptno=1);
+--------+------------+---------+--------+
| ename | hiredate | sal | deptno |
+--------+------------+---------+--------+
| zzx | 2000-01-01 | 2000.00 | 1 |
| bjguan | 2004-04-02 | 5000.00 | 1 |
+--------+------------+---------+--------+
2 rows in set (0.00 sec)