如何删除数据集中重复的数据

有时对于表中的结果集，满足一定规则我们则认为是重复数据，而这些重复数据需要删除。如何删除呢？见下面的例子（里面有一个用法只有在sql server 2005以上才有这个写法）

如下只要companyName,invoiceNumber,customerNumber三者都相同，我们则认为是重复数据，下面的例子演示了如何删除。

declare @InvoiceListMaster table ( ID int identity primary key ,
                            companyName Nchar(20),
                            invoiceNumber int,
                            CustomerNumber int,
                            rmaNumber int )

insert    @InvoiceListMaster
select N'华为', 1001,100,200
union all
select N'华为', 1001,100,300
union all
select N'华为', 1001,100,301
union all
select N'中兴', 1002, 200,1     
union all
select N'中兴', 1002, 200,2

select * from @InvoiceListMaster

DELETE A
from (
		select rown = ROW_NUMBER( )over( partition by companyname,
								 	 				  invoicenumber,
													  customerNumber 
											 order by companyname,
													  invoicenumber,
													  customerNumber ),
			   companyname,
			   invoicenumber,
			   customerNumber
		from @InvoiceListMaster  )a
where exists ( select 1 
               from ( select rown = ROW_NUMBER( )over( partition by companyname,
								 	 				  invoicenumber,
													  customerNumber 
											 order by companyname,
													  invoicenumber,
													  customerNumber ),
			   companyname,
			   invoicenumber,
			   customerNumber
		　　　　from @InvoiceListMaster  ) b
       　　　　 where b.companyName = a.companyName
             　　　　 and b.invoiceNumber = a.invoiceNumber
             　　　　 and b.CustomerNumber = a.CustomerNumber
             　　　　 and a.rown > b.rown
     )

select * from @InvoiceListMaster