SQL学习

常用的数据库

关系型数据库

MySQL，Oracle，SqlServer

非关系型数据库

Redis，MongoDB

两者关系

为什么要使用NoSQL

关系型数据库和非关系型数据库

数据库的数据类型

时间类型区别

DATETIM和TIMESTAMP类型所占的存储空间不同，前者8个字节，后者4个字节。

这样造成的后果是两者能表示的时间范围不同。

前者范围为1000-01-01 00:00:00 ~ 9999-12-31 23:59:59，后者范围为1970-01-01 08:00:01到2038-01-19 11:14:07。

所以可以看到TIMESTAMP支持的范围比DATATIME要小,容易出现超出的情况。

更多资料

mysql中timestamp,datetime,int类型的区别与优劣

MySQL数据类型详解

MySQL关键字

须知

• SQL 对大小写不敏感

• ''单引号包裹用来表示是字符，不过大多数数据库也支持""双引号

• *表示所有表内字段

表示例

user

course

class

grade

基础语法

SELECT

从某个表里筛选

SELECT 列名称 FROM 表名称

select * from user

select user_id,user_name from user

SELECT DISTINCT

把返回结果中重复的值去掉

SELECT DISTINCT 列名称 FROM 表名称

select user_id from grade

select distinct user_id from grade

WHERE

增加筛选的条件

SELECT 列名称 FROM 表名称 WHERE 列 运算符 值

select user_id from user where user_name = 'frank'

select user_id,user_name from user where user_name like '%r%'

select user_id from user where user_name in ('frank','barry')

AND&OR

AND 和 OR 可在 WHERE 子语句中把两个或多个条件结合起来

select score from grade where course_id = 2 and user_id = 2

select score from grade where course_id = 2 and (user_id = 1 or user_id = 2)

ORDER BY

用于根据指定的列对结果集进行排序，默认是升序，字母顺序或数字顺序

select score from grade order by score desc

select user_name from user order by user_name

select user_id,score from grade order by user_id desc , score asc

INSERT INTO

用于向表格中插入新的行

INSERT INTO 表名称 VALUES (值1, 值2,….)

INSERT INTO table_name (列1, 列2,…) VALUES (值1, 值2,….)

insert into user values (4,'tom',16063385);
select user_name from user

insert into user (user_id,user_name) values (5,'jerry');
select * from user

UPDATE

用于修改表中的数据

UPDATE 表名称 SET 列名称 = 新值 WHERE 列名称 = 某值

update grade set score = 100 where user_id = 1 and course_id = 3;
select user_id,course_id,score from grade where user_id = 1

update grade set course_id = 1,score = 99 where user_id = 1 and course_id = 2;
select user_id,course_id,score from grade where user_id = 1

DELETE

用于删除表中的行

DELETE FROM 表名称 WHERE 列名称 = 值

DELETE FROM table_name 删除表内所有的行

DELETE * FROM table_name 删除表内所有的行

delete from user where user_name = 'peter';
select user_id,user_name from user

进阶语法

LIMIT

用于规定要返回的记录的数目

SELECT column_name(s)
FROM table_name
LIMIT number

select user_name from user limit 2

LIKE

用于在 WHERE 子句中搜索列中的指定模式

SELECT column_name(s)
FROM table_name
WHERE column_name LIKE pattern

提示：”%” 可用于定义通配符（模式中缺少的字母）。

select user_id,user_name from user like '%r'   --结尾是r

select user_id,user_name from user not like 'f%'  --开头不是f

通配符

在搜索数据库中的数据时，SQL 通配符可以替代一个或多个字符。

SQL 通配符必须与 LIKE 运算符一起使用。

select user_name from user like '[fb]%'

AS别名

为列名称和表名称指定别名

表的别名

SELECT column_name(s)
FROM table_name
AS alias_name

列的别名

SELECT column_name AS alias_name
FROM table_name

select user_id, user_name as name from user where name = 'frank'

连接查询

之前都是在一个表里进行查询，当查询涉及两个表以上时，就叫连接查询

select user.user_name,course.course_name from user,course

select user.user_name,class.class_name from user,class where user.class_id = class.class_id

JOIN

专门用于查询两张及两张表以上的数据

INNER JOIN

两张表里都有匹配的话，就返回那一行

SELECT column_name(s)
FROM table_name1
INNER JOIN table_name2
ON table_name1.column_name=table_name2.column_name

select user.user_name,grade.score from user inner join grade on user.user_id = grade.user_id

LEFT JOIN

LEFT JOIN 关键字会从左表那里返回所有的行，即使在右表中没有匹配的行

这里的左表就是指下面的table_name1，右表指的是下面的table_name2

SELECT column_name(s)
FROM table_name1
LEFT JOIN table_name2
ON table_name1.column_name=table_name2.column_name

insert into user(user_id,user_name) values(4,'shell');
select user.user_name,class.class_name from user 
left join class on user.class_id = class.class_id

RIGHT JOIN

RIGHT JOIN 关键字会右表那里返回所有的行，即使在左表中没有匹配的行。

这里的左表就是指下面的table_name1，右表指的是下面的table_name2

SELECT column_name(s)
FROM table_name1
RIGHT JOIN table_name2
ON table_name1.column_name=table_name2.column_name

insert into class values (16090134,'新建班级',50);
select user.user_name,class.class_name from user 
right join class on user.class_id = class.class_id

FULL JOIN

只要其中某个表存在匹配，就会返回行

SELECT column_name(s)
FROM table_name1
FULL JOIN table_name2
ON table_name1.column_name=table_name2.column_name

select user.user_name,class.class_name from user
full join class on user.class_id = class.class_id

子查询

一个select语句中嵌套另一个select语句

MySQL子查询

子查询返回的是一个结果

SELECT column_name(s)

FROM table_name

WHERE column_name 运算符 (SELECT子查询)

select user_name from user 
where class_id = (select class_id from class where class_name = '软工3班')

IN / NOT IN

当子查询返回的结果有多个的时候，判断是不是在子查询的结果里

SELECT column_name(s)

FROM table_name

WHERE column_name IN (SELECT子查询)

select user_name from user where user_id in (select user_id from grade where score > 80)

EXISTS / NOT EXISTS

判断子查询的结果存不存在

SELECT column_name(s)

FROM table_name

WHER EXISTS (SELECT子查询)

select user_name from user 
where exists (select 1 from grade where user.user_id = grade.user_id and grade.score > 80)

ANY / SOME

当子查询返回的结果有多个的时候，表示任何的意思，ANY和SOME的结果一样，如where n > any (1,2) 等同 where n > 1 or n > 2, n>1或n>2都行

SELECT column_name(s)

FROM table_name

WHERE column_name 运算符 ANY (SELECT子查询)

select user_name from user where class_id =  
any (select class_id from class where class_name = '软工3班' and class_name = '会计4班')

ALL

当子查询返回的结果有多个的时候，表示所有的意思，如where n > all (1,2) 等同 where n > 1 and n > 2, 必须满足n > 1 和 n > 2

SELECT column_name(s)

FROM table_name

WHERE column_name 运算符 ALL (SELECT子查询)

select score from grade where score > all (select score from grade where user_id = 3)

函数

AVG()

求该列的平均值

SELECT AVG(column_name) FROM table_name

select avg(score) as avg_score from grade

select user_name from user where user_id 
in (select user_id from grade where score > (select avg(score) from grade))

COUNT()

返回匹配指定条件的行数

SELECT COUNT(column_name) FROM table_name

select count(user_id) as user_sum from grade

select count(distinct user_id) as user_sum from grade

MAX()

返回一列中的最大值

SELECT MAX(column_name) FROM table_name

select max(score) from grade

MIN()

返回一列中的最小值

SELECT MIN(column_name) FROM table_name

select min(score) from grade

SUM()

返回数值列的总数

SELECT SUM(column_name) FROM table_name

select sum(score) from grade

GROUP BY

根据一个或多个列对结果集进行分组

SELECT column_name, aggregate_function(column_name)
FROM table_name
WHERE column_name operator value
GROUP BY column_name

select user_id,avg(score) as avg from grade group by user_id

select user_id,course_id,sum(score) from grade group by user_id,course_id

HAVING

和where类似，不过筛选条件可以有函数

SELECT column_name, aggregate_function(column_name)
FROM table_name
WHERE column_name operator value
GROUP BY column_name
HAVING aggregate_function(column_name) operator value

select user_id,avg(score) from grade where user_id > 1
group by user_id
having avg(score) > 70

1. charlist ↩