SQL总结

1. 常用的数据类型

(1) char:      字符
(2) varchar:   字符串
(3) int:       整数
(4) float:     单精度(精度有一定的范围)
(5) double:    双精度
(6) date:      日期 年月日
(7) timestamp: 年月日, 时分秒
注意
(1) char与varchar区别
      如果将abc赋予char(10) ==> abcxxxxxxx
      如果将abc赋予vchar(10) ==> abc

2. SQL种类

(1)DML(增删改查): insert、update、delete、select
(2)DDL(创建db、创建table): create databases/drop databases

3. 创建table

(1) create table henrydb.henrydata(
    id int,
    name varchar(100),
    age int,
    //创建时间
    createtime timestamp,
    //创建人
    createuser varchar(100),
    //更新时间
    updatetime timestamp, 
    // 更新人
    updateuser varchar(100)
      );
   在DBeaver中显示的
   CREATE TABLE `henrydata` (
   `id` int(11) DEFAULT NULL,
   `name` varchar(100) DEFAULT NULL,
   `age` int(11) DEFAULT NULL,
   //默认是当前时间，当做update更新时会更新当前的时间
   `createtime` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
   `createuser` varchar(100) DEFAULT NULL,
   `updatetime` timestamp NOT NULL DEFAULT '0000-00-00 00:00:00',
   `updateuser` varchar(100) DEFAULT NULL
   ) ENGINE=InnoDB DEFAULT CHARSET=latin1
   注意：
   (1) 表名称字段名称尽量表达值的意思
   (2) 新建字段与之前字段形式保持一致

4. insert数据

(1) insert into henrydata(id, name, age)  values(1, 'henry', 18);

5. select

(1) select * from henrydata;

6. update

(1) update henrydata set age=20 where id=1;

7. delete

(1) delete from henrydata where id=1
注意: 确认delete后面是否有where语句

8. 改进版建表

/*生产上建表后面都加上prod*/
create table henrydb.henrydata_prod(
/*自增长, 主键约束*/
id int auto_increment primary key,
name varchar(100),
age int,
/*默认值约束 default*/
create_time timestamp default current_timestamp,
create_user varchar(100),
/*on update 只要更新mysql就会自动更新*/
update_time timestamp default current_timestamp on update current_timestamp,
update_user varchar(100)
/*设置默认字符集*/
)default charset=utf8;

注: 第一行与后四行是固定的

9. 在prod上增删改查数据

/*不需要主动输入id，因为id自增长，mysql会帮我们设置*/
insert into henrydb.henrydata_prod(name, age) values('henry', 20);

select id, name, age from henrydata_prod;

10. insert 的两种形式

(1) 指定字段
insert into henrydb.henrydata_prod(name, age) values('henry', 20);

(2) 不指定字段
insert into henrydb.henrydata_prod values(2, 'henry','2018-11-19 15:00:00', 'h', '2018-11-10 16:00:00', 'h');

常见错误

(1)Column count doesn't match value count at row 1
insert into henrydb.henrydata_prod values(2,'2018-11-19 15:00:00', 'h', '2018-11-10 16:00:00', 'h');

(2) Column count does not match value count at row 1
insert into henrydb.henrydata_prod values(1, 'henry','2018-11-19 15:00:00', 'h', '2018-11-10 16:00:00', 'h');
因为id为1已经存在，id为主键，主键是唯一的

11. 更新语句

(1) 将该表的age字段都设置为22
update henrydata_prod set age=22;

(2) update其age和address
/*修改多个字段用逗号分割*/
update henrydata_prod set age=25, address='zbc';

(3) 有条件更新(单个条件)
update henrydata_prod set age=25, address='zbc' where name='henry';

(4) 有条件更新(多个条件)
update henrydata_prod set address='abc' where name='henry' and age=21;

(5) 有条件更新或
update henrydata_prod set address='深圳' where name='henry' or age=25;

12. 删除语句

(1) 删除全部数据
delete from henrydata_prod;

(2) 有条件删除
delete from henrydata_prod where age=19;

13. 比较符号

>  <  =  >=  <=  <>

14. 模糊查询

like

示例: 查询名字中包含S的数据
select * from emp where ename like '%S%';

查询名字以S开头的数据
select * from emp where ename like 'S%';

查询名字以S结尾的数据
select * from emp where ename like '%S';

查询第二个字母是O的数据, 每个 "_"代表一个占位符
select * from emp where ename like '_O%';

15.排序

(1) 默认从小到大排序
select * from emp order by sal;

(2) 升序排列
select * from emp order by sal asc;

(3) 降序排列
select * from emp order by sal desc;

(4) 首先按照deptno进行升序排列，如果deptno相同的话，然后按照sal降序排序，
    如果这两项都相同，按照第一列排序 
    注： 在生产中，大部分情况下都用where进行筛选
select * from emp order by deptno asc, sal desc;

16. 限制多少行

限制两行，limit加在筛选之后
select * from emp order by deptno asc, sal desc limit 2;

17.聚合(group by)

select deptno, sum(sal) as salSum from emp group by deptno;

先对deptno进行分组，然后对job进行分组
注：

select deptno,  job, sum(sal) as salSum from emp group by deptno;

(
   在select中出现的字段如deptno和job，在group by也必须出现
   字段顺序在group by中可以变化，先按按个进行分组不会导致错误，
   但是 尽量还是与select的字段顺序保持一致
)

注意: group by 字段必须出现在 select 字段 (“工作中已经不止一次由此出错”)

18. 聚合函数

(1) 聚合函数: sum, count, avg, max, min

select deptno, job from emp 
where job = "SALESMAN"
group by deptno, job
having sun(sal) > 3000
order by sum(sal) desc
limit 3;  

19. having

select deptno, sum(sal) as salSum from emp 
group by deptno having sun(sql) > 9000;

注: 对查询结果进行聚合，然后对聚合后的结果进行筛选
    使用group by......having......

20 union

去除重复数据
select id, name from a
union
select id, address from b

不去除重复数据
select id, name from a
union all
select id, address from b

注意：
     1. 查询的结果显示的字段名称由第一张表决定
     2. 在生产中，避免使用select *, 使用select 想要查询的字段 
     3. 两张表对应位置的字段类型应保持一致

21. 子表

各个部门各个岗位的薪水和, 并且薪水和大于3000
注: 子表必须要起别名

select 
  t.*
from
(
    select
    deptno, job, sum(sal) as sum_sal 
    from emp
    group by deptno, job
) as t
where t.sum_sal > 3000

22. join

inner join: 内连接或等值连接 (不经常使用)
left  join: 左连接  (常用)
right join: 右连接

新建testa表
+------+--------+
| aid  | aname  |
+------+--------+
|    1 | henry1 |
|    2 | henry2 |
|    3 | henry3 |
|    4 | henry4 |
|    5 | henry5 |
+------+--------+

新建testb表
+------+--------+------+
| bid  | bname  | age  |
+------+--------+------+
|    1 | henry1 |   15 |
|    2 | henry2 |   16 |
|    3 | henry3 |   17 |
|    4 | henry4 |   18 |
|    7 | henry7 |   19 |
|    8 | henry8 |   20 |
|    9 | henry9 |   21 |
+------+--------+------+

left join

select 
a.aid, a.aname,
b.bid, b.bname
from
testa a left join testb b on a.aid = b.bid;

输出结果为
+------+--------+------+--------+
| aid  | aname  | bid  | bname  |
+------+--------+------+--------+
|    1 | henry1 |    1 | henry1 |
|    2 | henry2 |    2 | henry2 |
|    3 | henry3 |    3 | henry3 |
|    4 | henry4 |    4 | henry4 |
|    5 | henry5 | NULL | NULL   |
+------+--------+------+--------+

which means
left join : 以左表数据为主， 左表的数据会全部显示出来，
但是右表的数据只显示能和左表匹配(on + 字段)上的，
如果坐标某一record有数据，右表没有，右表的字段就为null，如上表所示

注: right join与left join正好相反 

inner join: 两张表匹配的数据行才显示

select 
a.aid, a.aname,
b.bid, b.bname
from
testa a left join testb b on a.aid = b.bid;

select 
a.aid, a.aname,
b.bid, b.bname
from
testa a left join testb b on a.aid = b.bid;

testa表
+------+--------+
| aid  | aname  |
+------+--------+
|    1 | henry1 |
|    2 | henry2 |
|    3 | henry3 |
|    4 | henry4 |
|    5 | henry5 |
+------+--------+

testb表
+------+---------+------+
| bid  | bname   | age  |
+------+---------+------+
|    1 | henry11 |   15 |
|    2 | henry22 |   16 |
|    3 | henry3  |   17 |
|    4 | henry4  |   18 |
|    7 | henry7  |   19 |
|    8 | henry8  |   20 |
|    9 | henry9  |   21 |
+------+---------+------+

select
a.aid, a.aname,
b.bid, b.bname
from
testa a left join testb b
on
a.aid = b.bid and a.anane = b.bname;

上sql语句的结果为
+------+--------+------+--------+
| aid  | aname  | bid  | bname  |
+------+--------+------+--------+
|    3 | henry3 |    3 | henry3 |
|    4 | henry4 |    4 | henry4 |
|    1 | henry1 | NULL | NULL   |
|    2 | henry2 | NULL | NULL   |
|    5 | henry5 | NULL | NULL   |
+------+--------+------+--------+

在上面sql语句中，只有3, 4符合条件，其他的testb都不符合条件，所以只有testa表的数据

多表join
eg. 查询员工的所属部门是什么？在什么地方？工资等级是什么？

select
empno, ename, e.deptno, dname, loc, (sal+comm) as total_sal, grade
from 
emp e left join dept d on e.deptno = d.deptno
left join salgrade on (sal + comm) between losal and hisal;

查询结果为
+-------+--------+--------+------------+----------+-----------+-------+
| empno | ename  | deptno | dname      | loc      | total_sal | grade |
+-------+--------+--------+------------+----------+-----------+-------+
|  7499 | ALLEN  |     30 | SALES      | CHICAGO  |   1900.00 |     3 |
|  7521 | WARD   |     30 | SALES      | CHICAGO  |   1750.00 |     3 |
|  7844 | TURNER |     30 | SALES      | CHICAGO  |   1500.00 |     3 |
|  7654 | MARTIN |     30 | SALES      | CHICAGO  |   2650.00 |     4 |
|  7782 | CLARK  |     10 | ACCOUNTING | NEW YORK |      NULL |  NULL |
|  7839 | KING   |     10 | ACCOUNTING | NEW YORK |      NULL |  NULL |
|  7934 | MILLER |     10 | ACCOUNTING | NEW YORK |      NULL |  NULL |
|  7369 | SMITH  |     20 | RESEARCH   | DALLAS   |      NULL |  NULL |
|  7566 | JONES  |     20 | RESEARCH   | DALLAS   |      NULL |  NULL |
|  7788 | SCOTT  |     20 | RESEARCH   | DALLAS   |      NULL |  NULL |
|  7876 | ADAMS  |     20 | RESEARCH   | DALLAS   |      NULL |  NULL |
|  7902 | FORD   |     20 | RESEARCH   | DALLAS   |      NULL |  NULL |
|  7698 | BLAKE  |     30 | SALES      | CHICAGO  |      NULL |  NULL |
|  7900 | JAMES  |     30 | SALES      | CHICAGO  |      NULL |  NULL |
+-------+--------+--------+------------+----------+-----------+-------+

由于NULL + 任意数值结果还是NULL，所以需要对NULL进行处理

改进后sql语句
select
empno, ename, e.deptno, dname, loc, (sal+IFNULL(comm,0)) as total_sal, grade
from 
emp e left join dept d on e.deptno = d.deptno
left join salgrade on (sal + IFNULL(comm,0)) between losal and hisal;

查询结果为
+-------+--------+--------+------------+----------+-----------+-------+
| empno | ename  | deptno | dname      | loc      | total_sal | grade |
+-------+--------+--------+------------+----------+-----------+-------+
|  7369 | SMITH  |     20 | RESEARCH   | DALLAS   |    800.00 |     1 |
|  7876 | ADAMS  |     20 | RESEARCH   | DALLAS   |   1100.00 |     1 |
|  7900 | JAMES  |     30 | SALES      | CHICAGO  |    950.00 |     1 |
|  7934 | MILLER |     10 | ACCOUNTING | NEW YORK |   1300.00 |     2 |
|  7499 | ALLEN  |     30 | SALES      | CHICAGO  |   1900.00 |     3 |
|  7521 | WARD   |     30 | SALES      | CHICAGO  |   1750.00 |     3 |
|  7844 | TURNER |     30 | SALES      | CHICAGO  |   1500.00 |     3 |
|  7782 | CLARK  |     10 | ACCOUNTING | NEW YORK |   2450.00 |     4 |
|  7566 | JONES  |     20 | RESEARCH   | DALLAS   |   2975.00 |     4 |
|  7788 | SCOTT  |     20 | RESEARCH   | DALLAS   |   3000.00 |     4 |
|  7902 | FORD   |     20 | RESEARCH   | DALLAS   |   3000.00 |     4 |
|  7654 | MARTIN |     30 | SALES      | CHICAGO  |   2650.00 |     4 |
|  7698 | BLAKE  |     30 | SALES      | CHICAGO  |   2850.00 |     4 |
|  7839 | KING   |     10 | ACCOUNTING | NEW YORK |   5000.00 |     5 |
+-------+--------+--------+------------+----------+-----------+-------+

注：
1. join可以出现多次
2. on后面语句也可以加 >, <, <>, between
3. 由于NULL + 数值还是NULL， 需要IFNULL()函数对其进行处理

23. 简单举例

求出各个/每部门的薪水和
注意: 看见各个/每就用group by

select
deptno, sun(sal)
from emp
group by deptno

各个部门各个岗位的薪水和

select
deptno, job, sun(sal)
from emp
group by deptno, job