在开发数据库系统的过程中，经常要写很多的存储过程。为了统一格式和简化开发过程，我编写一些存储过程，用来自动生成存储过程。下面就为您简单介绍一下它们。其中一个用于生成Insert过程，另一个用于生成Update过程。


Sp_GenInsert

该过程运行后，它为给定的表生成一个完整的Insert过程。如果原来的表有标识列，您得将生成的过程中的SETIDNTITY_INSERTON语句手工删除。

语法如下

sp_GenInsert<TableName>,<StoredProcedureName>

以northwind数据库为例

sp_GenInsert'Employees','INS_Employees'

最后会生成一个Insert存储过程。利用它，您可以作进一步的开发。


Sp_GenUpdate

它会为一个表生成update存储过程。语法如下：

sp_GenUpdate<TableName>,<PrimaryKey>,<StoredProcedureName>

以northwind数据库为例

sp_GenUpdate'Employees','EmployeeID','UPD_Employees'

运行后生成如下所示的存储过程：

CreateProcedureUPD_Employees

@EmployeeIDint

@LastNamenvarchar(40),

@FirstNamenvarchar(20),

@Titlenvarchar(60),

@TitleofCourtesynvarchar(50),

@BirthDatedatetime,

@HireDatedatetime,

@Addressnvarchar(120),

@Citynvarchar(30),

@Regionnvarchar(30),

@PostalCodenvarchar(20),

@Countrynvarchar(30),

@HomePhonenvarchar(48),

@Extensionnvarchar(8),

@Photeimage,

@Notesntext,

@ReportsToint,

@PhotoPathnvarchar(510)

AS

UPDATEEmployees

SET

LastName=@LastName,

FirstName=@FirstName,

Title=@Title,

TitleofCourtesy=@TitleofCourtesy,

BirthDate=@BirthDate,

HireDate=@HireDate,

Address=@Address,

City=@City,

Regin=@Regin,

PostalCode=@PostCode,

Country=@Country,

HomePhone=@HomePhone,

Extension=@Extension,

Photo=@Photo

Notes=@Notes，

ReportsTo=@ReportsTo,

PhotoPath=@PhotoPath

WHEREEmployeeID=@EmployeeID


使用以上的两个存储过程，节省了我不少时间。特别是在改变了表结构后，重新构造各个存储过程的过程中。您可以改写这两个程序，来自动生成别的存储过程。

SQL Server编写存储过程小工具
以下是两个存储过程的源程序

CREATE procedure sp_GenInsert
@TableName varchar(130),
@ProcedureName varchar(130)
as
set nocount on

declare @maxcol int,
@TableID int
--itlearner.com
set @TableID = object_id(@TableName)

select @MaxCol = max(colorder)
from syscolumns
where id = @TableID

select 'Create Procedure ' + rtrim(@ProcedureName) as type,0 as colorder into #TempProc
union
select convert(char(35),'@' + syscolumns.name)
+ rtrim(systypes.name)
+ case when rtrim(systypes.name) in ('binary','char','nchar','nvarchar','varbinary','varchar') then '(' + rtrim(convert(char(4),syscolumns.length)) + ')'
when rtrim(systypes.name) not in ('binary','char','nchar','nvarchar','varbinary','varchar') then ' '
end
+ case when colorder < @maxcol then ','
when colorder = @maxcol then ' '
end
as type,
colorder
from syscolumns
join systypes on syscolumns.xtype = systypes.xtype
where id = @TableID and systypes.name <> 'sysname'
union
select 'AS',@maxcol + 1 as colorder
union
select 'INSERT INTO ' + @TableName,@maxcol + 2 as colorder
union
select '(',@maxcol + 3 as colorder
union
select syscolumns.name
+ case when colorder < @maxcol then ','
when colorder = @maxcol then ' '
end
as type,
colorder + @maxcol + 3 as colorder
from syscolumns
join systypes on syscolumns.xtype = systypes.xtype
where id = @TableID and systypes.name <> 'sysname'
union
select ')',(2 * @maxcol) + 4 as colorder
union
select 'VALUES',(2 * @maxcol) + 5 as colorder
union
select '(',(2 * @maxcol) + 6 as colorder
union
select '@' + syscolumns.name
+ case when colorder < @maxcol then ','
when colorder = @maxcol then ' '
end
as type,
colorder + (2 * @maxcol + 6) as colorder
from syscolumns
join systypes on syscolumns.xtype = systypes.xtype
where id = @TableID and systypes.name <> 'sysname'
union
select ')',(3 * @maxcol) + 7 as colorder
order by colorder

select type from #tempproc order by colorder

drop table #tempproc

SQL Server编写存储过程小工具
功能：为给定表创建Update存储过程
语法： sp_GenUpdate <Table Name>,<Primary Key>,<Stored Procedure Name>
以northwind 数据库为例
sp_GenUpdate 'Employees','EmployeeID','UPD_Employees'

注释：如果您在Master系统数据库中创建该过程，那您就可以在您服务器上所有的数据库中使用该过程。

===========================================================*/
CREATE procedure sp_GenUpdate
@TableName varchar(130),
@PrimaryKey varchar(130),
@ProcedureName varchar(130)
as
set nocount on

declare @maxcol int,
@TableID int
--itlearner.com
set @TableID = object_id(@TableName)

select @MaxCol = max(colorder)
from syscolumns
where id = @TableID

select 'Create Procedure ' + rtrim(@ProcedureName) as type,0 as colorder into #TempProc
union
select convert(char(35),'@' + syscolumns.name)
+ rtrim(systypes.name)
+ case when rtrim(systypes.name) in ('binary','char','nchar','nvarchar','varbinary','varchar') then '(' + rtrim(convert(char(4),syscolumns.length)) + ')'
when rtrim(systypes.name) not in ('binary','char','nchar','nvarchar','varbinary','varchar') then ' '
end
+ case when colorder < @maxcol then ','
when colorder = @maxcol then ' '
end
as type,
colorder
from syscolumns
join systypes on syscolumns.xtype = systypes.xtype
where id = @TableID and systypes.name <> 'sysname'
union
select 'AS',@maxcol + 1 as colorder
union
select 'UPDATE ' + @TableName,@maxcol + 2 as colorder
union
select 'SET',@maxcol + 3 as colorder
union
select syscolumns.name + ' = @' + syscolumns.name
+ case when colorder < @maxcol then ','
when colorder = @maxcol then ' '
end
as type,
colorder + @maxcol + 3 as colorder
from syscolumns
join systypes on syscolumns.xtype = systypes.xtype
where id = @TableID and syscolumns.name <> @PrimaryKey and systypes.name <> 'sysname'
union
select 'WHERE ' + @PrimaryKey + ' = @' + @PrimaryKey,(2 * @maxcol) + 4 as colorder
order by colorder

select type from #tempproc order by colorder

drop table #tempproc