一道经典的sql面试题!认为自己sql玩得挺转的进来!

学号(自动编号)         姓名      性别        年龄

0001                    xw        男          18

0002                    mc        女          16

0003                    ww        男          21

0004                    xw        男          18

请写出实现如下功能的SQL语句:
删除除了学号（自动编号）字段以外，其它字段都相同的冗余记录！

方法一、delete table where  学号(自动编号) not in
(select max(学号(自动编号)) as xh ,         姓名 ,     性别 ,       年龄
  from table
group by 姓名 ,     性别 ,       年龄) A
方法二、
delete table where  学号(自动编号) not in (
select max(a.学号) from table B,(select distinct 姓名 ,     性别 ,       年龄 from table ) A
 where A.姓名=b.姓名 and a.性别=b.性别 and a.年龄 = b.年龄
) C

以上方法不知道是否可以，没有在ＳＱＬ中测试。
 delete from tablename where 学号(自动编号) notexists (select distinct 姓名,性别,年龄from tablename)

 

lxmfll2000(lxm)　的，没分清in 和exists啦...

lhx1977(清水无鱼)的思路是正确的，调试后的写法如下：(注意in的子查询只能select出一列)
DELETE FROM table1
WHERE (学号 NOT IN
          (SELECT MAX(学号) AS xh
         FROM TABLE1
         GROUP BY 姓名, 性别, 年龄))

DELETE FROM Test2
where [学号] in
(select [学号]=min([学号]) from Test2 group by [姓名],[性别],[年龄] having count(*)>1)

create table #temp
(学号 int)
insert into #temp
(select [学号]=min([学号]) from Test2 group by [姓名],[性别],[年龄]
delete from Test2 where 学号 not in (select 学号 from #temp)
drop #temp

create table #temp
(学号 int)
insert into #temp
(select [学号]=min([学号]) from Test2 group by [姓名],[性别],[年龄])
delete from Test2 where 学号 not in (select 学号 from #temp)
drop table #temp
更正

declare @t table(学号 int identity(1,1),
姓名 varchar(30),性别 nvarchar(1),年龄 int)

insert into @t

select 'xw','男',18 union all
select 'mc','女',18 union all
select 'mc','女',18 union all
select 'mc','女',18 union all
select 'ww','男',21 union all
select 'xw','男',18 union all
select 'xw','男',18

select * from @t

delete from @t where 学号 not in (select min(学号) from @t group by 姓名,性别,年龄)

select * from @t

 

DELETE FROM Test2
where [学号] in(
select a.[学号]
from Test2 a,(select [姓名],[性别],[年龄] from Test2 group by [姓名],[性别],[年龄] having count(*)>1)b
where a.[姓名]=b.[姓名] and a.[性别]=b.[性别] and a.[年龄]=b.[年龄]
and
[学号] not in
(select [学号]=min([学号]) from Test2 group by [姓名],[性别],[年龄] having count(*)>1)
)

 

create table #table
(
[学号] varchar(5),
[姓名] varchar(20),
[性别] varchar(2),
[年龄] int
)
--drop table #table
insert into #table ([学号],[姓名],[性别],[年龄])
select '0001','xw','男',18 union all
select '0002','mc','女',18 union all
select '0003','mc','女',18 union all
select '0004','mc','女',18 union all
select '0005','ww','男',21 union all
select '0006','xw','男',18 union all
select '0007','xw','男',18

--显示需要保留的
select min(学号),[姓名],[性别],[年龄] from #table
group by [姓名],[性别],[年龄] order by  min(学号)

--删除不需要保留的
delete --select *
from #table
where [学号] not in (select min(学号) from #table
group by [姓名],[性别],[年龄])

declare @t table(学号 int identity(1,1),
姓名 varchar(30),性别 nvarchar(1),年龄 int)

insert into @t
select 'xw','男',18 union all
select 'mc','女',18 union all
select 'mc','女',18 union all
select 'mc','女',18 union all
select 'ww','男',21 union all
select 'xw','男',18 union all
select 'xw','男',18

select * from @t

delete from @t
where 学号
in
(
 select b.学号
 from @t b
 inner join
 (
  select 姓名,性别,年龄
  from @t
  group by  姓名,性别,年龄
  having count(1)>1
 ) a
 on a.姓名=b.姓名 and a.性别=b.性别 and a.年龄=b.年龄
)

select * from @t
--------------------
你问的地方不对 这里是。NET版块 毕竟比SQL版块差多了