最近一直在做Dnn模块的开发,过程中碰到这么一个问题,需要同时插入N条数据,不想在程序里控制,但是SQL Sever又不支持数组参数.所以只能用变通的办法了.利用SQL Server强大的字符串处理传把数组格式化为类似"1,2,3,4,5,6"。
然后在存储过程中用SubString配合CharIndex把分割开来
详细的存储过程
CREATE PROCEDURE dbo.ProductListUpdateSpecialList
@ProductId_Array varChar(800),
@ModuleId int
AS
DECLARE @PointerPrev int
DECLARE @PointerCurr int
DECLARE @TId int
Set @PointerPrev=1
set @PointerCurr=1
begin transaction
Set NoCount ON
delete from ProductListSpecial where ModuleId=@ModuleId
Set @PointerCurr=CharIndex(',',@ProductId_Array,@PointerPrev+1)
set @TId=cast(SUBSTRING(@ProductId_Array,@PointerPrev,@PointerCurr-@PointerPrev) as int)
Insert into ProductListSpecial (ModuleId,ProductId) Values(@ModuleId,@TId)
SET @PointerPrev = @PointerCurr
while (@PointerPrev+1 < LEN(@ProductId_Array))
Begin
Set @PointerCurr=CharIndex(',',@ProductId_Array,@PointerPrev+1)
if(@PointerCurr>0)
Begin
set @TId=cast(SUBSTRING(@ProductId_Array,@PointerPrev+1,@PointerCurr-@PointerPrev-1) as int)
Insert into ProductListSpecial (ModuleId,ProductId) Values(@ModuleId,@TId)
SET @PointerPrev = @PointerCurr
End
else
Break
End
set @TId=cast(SUBSTRING(@ProductId_Array,@PointerPrev+1,LEN(@ProductId_Array)-@PointerPrev) as int)
Insert into ProductListSpecial (ModuleId,ProductId) Values(@ModuleId,@TId)
Set NoCount OFF
if @@error=0
begin
commit transaction
end
else
begin
rollback transaction
end
GO
网友Bizlogic对此的改进方法:
应该用SQL2000 OpenXML更简单，效率更高，代码更可读：
CREATE Procedure [dbo].[ProductListUpdateSpecialList]
(
@ProductId_Array NVARCHAR(2000),
@ModuleId INT
)
AS
delete from ProductListSpecial where ModuleId=@ModuleId
-- If empty, return
IF (@ProductId_Array IS NULL OR LEN(LTRIM(RTRIM(@ProductId_Array))) = 0)
RETURN
DECLARE @idoc int
EXEC sp_xml_preparedocument @idoc OUTPUT, @ProductId_Array
Insert into ProductListSpecial (ModuleId,ProductId)
Select
@ModuleId,C.[ProductId]
FROM
OPENXML(@idoc, '/Products/Product', 3)
with (ProductId int ) as C
where
C.[ProductId] is not null
EXEC sp_xml_removedocument @idoc
类别：数据库技术 来源：本站原创 作者：hpping 日期：2009-10-20 14:33
上一条：简单实现Standby SQL Server 数据库的方法
下一条：超越批量处理与MapReduce：如何让Hadoop走得更远