这是表的结构
CREATE TABLE [dbo].[Table_1](
    [ID] [int] IDENTITY(1,1) NOT NULL,
    [GDSCode] [nchar](10) COLLATE Chinese_PRC_CI_AS NULL,
    [ExHotelCode] [int] NULL,
    [ExRoomCode] [nchar](10) COLLATE Chinese_PRC_CI_AS NULL,
    [ExRatePlaneCode] [nchar](10) COLLATE Chinese_PRC_CI_AS NULL,
    [HotelID] [nchar](10) COLLATE Chinese_PRC_CI_AS NULL,
    [RoomID] [nchar](10) COLLATE Chinese_PRC_CI_AS NULL,
    [RatePlaneID] [nchar](10) COLLATE Chinese_PRC_CI_AS NULL,
 CONSTRAINT [PK_Table_1] PRIMARY KEY CLUSTERED 
(
    [ID] ASC
)WITH (IGNORE_DUP_KEY = OFF) ON [PRIMARY]
) ON [PRIMARY]

这是删除重复数据的SQL语句：
DELETE FROM Table_1 WHERE --c.ID=2
(GDSCode, a.ExHotelCode, a.ExRoomCode,a.ExRatePlaneCode,a.HotelID, a.RoomID,a.RatePlaneID) IN 
(SELECT c.GDSCode, c.ExHotelCode, c.ExRoomCode,c.ExRatePlaneCode, 
c.HotelID, c.RoomID,c.RatePlaneID FROM Table_1 c GROUP BY c.GDSCode,c.ExHotelCode, 
c.ExRoomCode,c.ExRatePlaneCode, c.HotelID, c.RoomID,c.RatePlaneID HAVING COUNT(*)>1)
AND 
a.ID NOT IN(SELECT MIN(ID) FROM Table_1 b GROUP BY b.GDSCode, b.ExHotelCode, 
b.ExRoomCode,b.ExRatePlaneCode, b.HotelID, b.RoomID,b.RatePlaneID HAVING COUNT(*)>1)

delete a from  TB a where  exists(select 1 from TB where Name=a.Name and ID<a.ID)--只保留一条记录

------解决方案--------------------
delete a from Table_1 a  
where  exists(select 1 from Table_1 where Name=a.Name and ID>a.ID)



------解决方案--------------------


处理表重复记录(查询和删除)_整理贴4

ORACLE中不熟悉...