oracle 选择查询:删除重复的重复行,而不按分组

9lowa7mx  于 2023-11-17  发布在  Oracle
关注(0)|答案(5)|浏览(119)

我有一个这样的表,其中Student(id,attempt)数据如下所示:
| ID|尝试|
| --|--|
| 1 | 1 |
| 1 | 2 |
| 1 | 3 |
| 1 | 4 |
| 2 | 1 |
| 2 | 2 |
我想找到学生的最高尝试例如:student 1在他的第三次尝试中通过了考试
最终输出必须像
| ID|尝试|
| --|--|
| 1 | 4 |
| 2 | 2 |
我不想使用where,group by,因为我正在创建一个视图,并且将连接多个表

INNER JOIN (SELECT id, MAX(attempts)as MaxCreated
                FROM SHRTCKG
                GROUP BY id, attempts) u
        ON x.id = u.attempts
        AND x.id = u.attempts)

字符串

4urapxun

4urapxun1#

您正在查找每个ID的最大尝试次数,并显示按ID排序的结果?这是最基本的聚合:

select id, max(attempt)
from shrtckg
group by id
order by id;

字符串

kiayqfof

kiayqfof2#

这里是使用SQL Server的查询。这基本上是每个id的最大尝试值。

select 
       id , 
       max(attempt) as max_attempt from Student as a 
group by id;

字符串

vof42yt1

vof42yt13#

如果你不想使用WHERE和GROUP BY(虽然不知道为什么),那么解决方案可以是MODEL子句(带Distinct):

WITH  --  S a m p l e   D a t a : 
    tbl (ID, ATTEMPT) As
        (   Select 1,   1 From Dual Union All 
            Select 1,   2 From Dual Union All
            Select 1,   3 From Dual Union All
            Select 1,   4 From Dual Union All
            Select 2,   1 From Dual Union All
            Select 2,   2 From Dual 
        )

字符串
代码如下:

-- M a i n   S Q L :
SELECT  Distinct ID, COUNT_ATTEMPT "ATTEMPT"
FROM    (   Select  ID, ATTEMPT, 0 "COUNT_ATTEMPT" From tbl )
MODEL   Dimension By    ( ID, ATTEMPT   )
        Measures        ( COUNT_ATTEMPT )
RULES   (   COUNT_ATTEMPT[ANY, ANY] = Count(COUNT_ATTEMPT)[CV(ID), ATTEMPT > 0] )


... ...这是什么?

--    R e s u l t :
--          ID    ATTEMPT
--  ---------- ----------
--           1          4
--           2          2

qvtsj1bj

qvtsj1bj4#

你可以像这样使用标量子查询:

select distinct id, 
(select MAX(attempt) 
  from mytable t2
  where t2.id = t.id) as attempt
from mytable t

字符串

icnyk63a

icnyk63a5#

如果不能使用GROUP BY,我们可以使用窗口函数MAX()

select distinct id, MAX(attempt) over (partition by id) as attempt
from mytable

字符串
Demo here

相关问题