SQL常用语句

• 表操作

create table t_person1(id int not null,name nvarchar(50),age int null)

drop table t_person1

alter table t_person1 add department vchar(20)

• 数据更新

update Student set Name='李鹏' where Id=5

update a set a.Bonus=500 from Student a inner join Gender b on a.Gender=b.Id where b.GenderName='男'

update vw_Gender set Bonus=200 where GenderName='女'

select * from Student

• 数据插入

insert into Student(Name,Age,Bonus,Gender)values('李强',18,300,1)

• 数据删除

delete from Student where Id=11  //与drop table不同的是delete只是删除数据 表还在

delete from Student where Id in(8,9,10)

• 数据检索

select * from pud

select top 3 * from pud

select pudname,pudprice from pud

select pudname as N'姓名' from pud

select pudprice=pudprice+1 from pud

select count/max/min/avg/sum(money) from t_person1  //聚合函数不能出现在where子句

select * from pud where id=259

select * from pud where pudprice>100

select * from pud where pudname='Kiddy安全座椅限时特价套装：限量版宝宝安全提篮－9个月到4岁儿童安全座椅'

select * from pud where pudname like '%母爱%'

select * from pud where pudname like '%母爱%' and pudprice>100

select * from pud where pudname like '%2%'

select * from pud where pudname like '2%'

select * from pud where pudname like '%2'

select * from pud where CHARINDEX('2',pudname)>0

select * from pud where id in(253,254,255,256)

select id,pudname,pudprice from pud where id in(253,254,255,256)

select 1+1 / select newid() / select getdate()

• 多表关联数据检索

1. 添加外键关系: 表设计器中右键 → 关系 → 添加 → …

select a.Name,a.Age,b.GenderName from Student a, Gender b where a.Gender=b.Id

select a.Name,a.Age,b.GenderName from Student a inner join Gender b on a.Gender=b.Id

select a.Name,a.Age,b.GenderName from Student a left join Gender b on a.Gender=b.Id

select a.Name,a.Age,b.GenderName from Student a right join Gender b on a.Gender=b.Id

(创建视图)

select Name, Age, GenderName from vw_Gender

• 联合结果集

1. union all: 禁止结果集重复数据抑制/一般保留all

select number,name,age from t_person1

　　union 

　　select fnumber,fname,fage from t_person2  //union要求上下结果集列数 数据类型必须相同

select number,name,department from t_person1

　　union

　　select number,name,‘无部门’ from t_person2  //结果集补足列

• 子查询

select * from

　　(

　　　　select row_number() over(order by salary desc) as rownum,number,name,salary,age from t_employee

　　) as a

　　where a.rownum>=3 and a.rownum<=7  //限制结果集 返回第3行到第7行的数据

• 数据排序

1. 升序(ASC/默认) / 降序(DESC)

select top 3 * from pud order by pudprice desc

select top 3 * from pud order by pudprice asc

select top 3 * from pud order by LEN(pudname) desc

select top 3 * from pud order by LEN(pudname) asc

select top 3 * from pud order by salary ASC,age DESC

select top 3 * from pud

　　where age>20

　　order by salary ASC,age DESC  //where在order by之前

• 数据分组

select * from t_person1 group by age
select age,avg(salary) from t_person1 group by age  //没有放在group by子句中的列不能出现在select语句后(聚合函数除外)

• having语句

1. having: 对组进行过滤/放在group by之后/where语句中不能出现聚合函数而必须使用having

select age count(*) as N'人数' from t_person1

　　group by age

　　having count(*)>1  //having中的列必须包含在聚合函数或order by子句中

• 抑制重复数据

select GenderName from

(

select b.GenderName from Student a inner join Gender b on a.Gender=b.Id

)as c group by GenderName

select GenderName from vw_Gender group by GenderName

select distinct(GenderName) from vw_Gender

• 限制结果集行数

1. 作用: 分页

select top 5 * from t_person1 order by salary desc  //取工资前五

select top 3 * from t_person1

　　where number not in (select top 5 number from t_person1 order by salary desc;)

　　order by salary desc  //取工资第六开始的前三

select row_number over(order by salary desc) as rownum, number,name,age,salary from t_employee  //SQLServer2005之后增加了row_number函数简化实现

• 通配符过滤/模糊匹配

1. Like(where)

select * from t_person1 where name like ‘_erry’  //单字符通配符(_)

select * from t_person1 where name like ‘k%’  //0或多字符通配符(%)

• 多值匹配

select * from t_person1 where age=23 or age=25 or age=27

select * from t_person1 where age in (23,25,27)

select * from t_person1 where age>=23 and age<=27

select * from t_person1 where age between 23 and 27

• 空值处理

1. 数据库(狭义)中一个列如果没有指定值则为null(表示‘不知道’而不是没有/不同于C#中的null(空/不指向任何对象))

select * from t_person1 where age is null  //SQL中特殊空值判断语法(is null/is not null)

• 游标循环

pl/sql类似于C、C++ 这类语言一次只能处理一条数据库记录 如果想用这类语言处理返回的多条数据就要用到游标

declare @id int

declare Student_Cursor cursor for select Id from Student

open Student_Cursor

fetch next from Student_Cursor into @id

while @@fetch_status=0

begin

update Student   --单独对返回的每条数据进行处理

set CreateDaTe=RandTime('2011-9-8','2011-9-11',rand())

where Id=@id

fetch next from Student_Cursor into @id

end

close Student_Cursor

deallocate Student_Cursor

• 其他

declare @str nvarchar(50)

set @str='明天|你要|嫁给|我了'

if charindex('2',@str)>0

begin

print('存在')

end

else

begin

print('不存在')

end