Question

我有表格MyTable（编号，位置编号，用户名，日期，排名）

我需要从此表中删除重复项，但左行带有最小ID

select min(cr.id)
from MyTable mt
group by mt.LocationId, mt.UserName, mt.Date, mt.Rank
having count(mt.id) > 1

这是从组中选择最小值的查询...但是不知道如何从重复项中排除它们

select mt.LocationId, mt.UserName, mt.Date, mt.Rank
from MyTable mt
group by mt.LocationId, mt.UserName, mt.Date, mt.Rank
having count(mt.id) > 1

Answer 1

您可以使用CTE和ROW_NUMBER() ROW_NUMBER() OVER ()函数正在查找具有与重复列相同值的行。此列组合的首次出现被分配为rn为1，下一个为2、3，依此类推。按Id升序排列，最小将为rn = 1，因此删除where rn > 1，删除重复项并保留ID最小的行。

with cte as (
    select
        *
        ,rn = ROW_NUMBER() over (PARTITION BY LocationId, UserName, Date, Rank order by id asc)
    from MyTable 
)
delete from cte where rn > 1

Answer 2

我会使用row_number()函数：

delete mt
from (select mt.*,
             row_number() over (partition by mt.LocationId, mt.UserName, mt.Date, mt.Rank order by mt.id) as seq
      from MyTable mt
     ) mt
where seq > 1;

Answer 3

您可以使用row_number窗口功能

with t1 as
(
select * , 
row_number() over(partition by mt.LocationId, mt.UserName, mt.Date, mt.Rank order by Id) as rn from MyTable mt
) delete from t1 where t1.rn!=1

Answer 4

使用row_number（）

Base

从组中排除最小值

4 个答案: