sql server 中语法定义
创建数据库
create DataBase database_name
[on [primary]
[/<filespec/>[1,.......n]]
[,<filegroup>[1,.......n]]
]
[LOG ON {<filespec>[1,.......n]} ]
[FOR LOAD|FOR ATTACH]
<filespec>::=([
[NAME=logical_filename,]
FILENAME='os_file_name'
[,SIZE=size]
[,MAXSIZE={max_size|UNLIMITED}]
[,FILEGROUP=growth_increment ]
)[,.......n]
<filegroup>::=FILEGROUP filegroup_name<filespec>[,.......n]
被[]括中的内容表示可有或可不有
[1,....n]表示在括号前面的项可以出现1次或N次
被<>括中的内容表示里面的具体内容会在下文中详细说明
FOR ATTACH 是附加数据库文件
修改数据库
Alter DataBase database
{
Add File <filespace> [1,2,3....n] [To filegroup filegroup_name]
|Add Log File <filespace> [1,2,3....n]
|Remove File logical_file_name
|Add FileGroup filegroup_name
|Remove Filegroup filegroup_name
|Modify file <filesapce>
|Modify filegroup file_group_name filegroup_property
}
<filespec>::=([
[NAME=logical_filename,]
FILENAME='os_file_name'
[,SIZE=size]
[,MAXSIZE={max_size|UNLIMITED}]
[,FILEGROUP=growth_increment ]
)[,.......n]
Modify filegroup file_group_name filegroup_property 中的filegroup_property可以设置的值为：ReadOnly(文件组只读)
,ReadWrite（读写）,Default（将文件组设置为默认的文件组。只能设置一个默认文件组）
}
删除数据库
drop database database_name
创建表
CreateTable
[database_name.[owner].|owner.] table_name
(
{
<column_definition> | column_name as Computed_column_expression
|<table_constraint>
}[1,2,....,n]
)
[on {filegroup|default}]
[textimage_on {filegroup|default}]
<column_definition>::={column_name data_type}
[
[ default constrant_expression] | [identity [(seed,increment ) ] ]
]
[<column_constraint>][....n]
修改表
Alter Table table
Alter Column column_name new_data_type [(precision [, scale ] ) ] [null|no null]
|Add {[column_definition] | column_name as computed_column_expression}
|drop Column column_name[1,...,n]
|Add Constraint constraint_name constraint_definition [1,....,n]
|drop Constraint constraint_name [1,...,n]
删除表
drop table tablename[,...,n]
主键约束
constraint constraint_name primary key (CLUSTERED|NONCLUTERED) [(字段[1,...,n])]
唯一键约束
constraint constraint_name unique key (CLUSTERED|NONCLUTERED) [(字段[1,...,n])]
外键约束
constraint constraint_name foreige key [(字段[1,...,n])] references rel_table[(rel_column)] [on delete {cascade|no action}] [on update {CASCADE|NO ACTION}] [NOT FOR REPLICATION]
NOT FOR REPLICATION:如果添加了这个选项，那么在复制数据的时候将忽略该外键约束
cascade：将发生在主键上的改变级联到应用键
no action：同上。拒绝任何变动
核查约束
constraint constraint_name check('check_expression')
例如 constraint constraint1 check('sex' in ('男',‘女’))
创建规则
create RULE rule_name as condition_expression
condition_expression:@param logical_expression
绑定规则
sp_bindrule [@rulename=] 'rule_name',[@objname=] 'object_name'
例如 create rule salary
as
@para_name between 1000 and 5000
exec sq_bindrule 'salary','员工.工资' 员工工资范围在1000~5000之间
创建默认值
create Default default
as constraint_expression
绑定默认值
sq_binddefault [@defaultname=]'default_name',[@objname=]'object_name'
例如:
create default salary
as
1000
exec sq_binddefault 'salary','员工.工资'
========SELECT=============
Select [select_expression]select_list
[INTO new_table_name]
From table_list
[Where search_constraint]
[Group By group_expression group_by_list]
[Having search_conditions]
[Order by order_list [ASC|DESC]]
INTO new_table_name:将获得的信息插入到表中
select_expression(
distinct 不返回重复的数据行
top num 选择返回数据的顶部num行
top num precent 选择返回数据的顶部百分比
top num关键字==set rowcount n
)
from_search_conditions
(
比较（=,>,<,>=,<=,<>,!>,!<,!=）
范围（between .. and .. , not between .... and ...）
列表( in(A,B,C,D,E) )
模式（like expression）expression:%任何串字符;_任何一个字符;[A-Z]A到Z中的任何一个;[^A-Z]不是A和Z中的任何一个
)
where子句的执行顺序>groupby子句的执行顺序>having子句的执行顺序
group_expression
(
ALL 将不符合where子句的内容也进行显示。但在其返回的相应数据中以NULL标记
CUBE
ROLLUP
)
======Transaction编程===============
begin..and ..
if ...else ...
while logical_expression begin...[break|continue]...and
case param
when expression1 then
when expression2 then
when expression3 then
else
expression4
end
waitfor delay 时间间隔|time 时间值
goto
游标
Declare 游标名 CURSOR
[LOCAL|GLOBLE]
[FORWARD_ONLY|SCROLL]
[READONLY]
FOR 选择语句
[FOR UPDATE [OF 字段1，字段2]]
游标的读取方法
FIRST
LAST
PRIOR
NEXT
RELATIVE 相对
ABSOLUTE 绝对
例子
declare mycursor CURCOR
for
select * from customers
open mycursor
fetch next from mycursor
while @@fetch_status=0
begin
fetch next from mycursor
end
close mycursor
deallocate mycursor
fetch next from cursorName into @parameter1,@parameter2 从游标中读取数据存入变量中
======创建索引=========
create [unique] [CLUSTERED|NONCLUSTERED] Index index_name
on Table(column,...,n)
[
WITH [PAD_INDEX]
[,FILLFACTOR=filefactor]
[,IGNORE_DUP_KEY]
[,DROP_EXISTING]
[,STATISTICE_NORECOMPUTE]
]
[on filegroup]
FILLFACTOR填充程度 值1~100
IGNORE_DUP_KEY 是否忽略重复的值
DROP_EXISTING 指定应除去名称相同的索引然后重建
删除索引 drop index table.index[1,....,n]
=========创建触发器==============
create trigger trigger_name
on table|view
[with encryption]
{for|after|instead of} {[insert][,][update][,][delete]}
as
sql_statement [....n]
(缓存表 inserted,deleted)update=delete+insert
exec sp_helptrigger 'table'[,type]
exec sp_helptext 'trigger_name'
exec sp_help 'trigger_name'
exec sp_rename 'old_trigger_name','new_trigger_name'
drop trigger trigger_name[,...,n]
例如
create trigger trigger_salary_change
on pmanager
after update
as
if update(工资)
begin
if(select max(abs(inserted.工资-deleted.工资))
from inserted join deleted
on inserted.负责人ID =deleted.负责人ID)>2000
begin
print '工资变动不能超过2000'
ROLLBACK TRANSACTION
end
end
else
print '工资没有变！'
===========存储过程===========
create procedure procedure_name [;name]
[
{@parameter data_type} [varying] [=default] [output]
] [,....n]
[with {recomplie|encryption|recompile,encryption}]
[for replication]
as
sql_statement [,...,n]
执行存储过程
exec {[@return_status=] procedure_name [;name]}
[[@parameter=] {value|@variable [output] | [default] } ][,...,n]
例如
exec @returnvalue=GetAvgPbiaodi '参数','参数'
exec sq_helptext GetAvgPbiaodi
exec sq_depends GetAvgPbiaodi
exec sq_help GetAvgPbiaodi
exec sql_rename oldname newname