Question

我有一个这样的表：

t1    t2    t3
1     01    M6
1     02    M6
2     01    M1
2     01    M1
2     02    M1
3     04    M4
3     04    M4
4     01    M1
4     01    M2

我需要结束这样的事情：

t1    t2    t3
1     01    M6
1     02    M6
2     01    M1
2     02    M1
3     04    M4
4     01    M1
4     01    M2

换句话说，我需要：
对于每个t1，删除在t2上重复的行（保留其中一个），同时具有相同的t3值。如果它们在t2上重复但具有不同的t3，则不会被删除。

正在使用类似这样的东西进行测试，但是还不足够：

WITH cte AS (
    SELECT t1,
        RNum = ROW_NUMBER() OVER (PARTITION BY t2, t1 order by t3)
    FROM tbl
    )
DELETE FROM cte WHERE RNum > 1;

Answer 1

Execute a query like this:

DELETE from tbl WHERE ROWID in (
  SELECT DISTINCT dup.ROWID
  FROM tbl inner join tbl as dup ON tbl.t1 = dup.t1 AND tbl.t2 = dup.t2 AND tbl.t3 = dup.t3 AND dup.ROWID > tbl.ROWID
)

Answer 2

具有CTE：

with cte as (
  select 
    t1, t2, t3, 
    ROW_NUMBER() OVER (PARTITION BY t1, t2, t3 order by t1, t2, t3) rn from tbl
)
delete from cte
where rn > 1;

请参见demo

Answer 3

请在下面尝试。希望它会有所帮助。

 delete from a
 from
 (select t1, t2, t3
       ,ROW_NUMBER() over (partition by t1, t2, t3
                      order by t1, t2, t3) RowNumber 
  from tbl) a
  where a.RowNumber > 1

删除重复项（如果符合条件则保留一个）

3 个答案: