Question

如何从以下列方式设置的表中删除重复项？

unique_ID | worker_ID | date | type_ID

一个worker可以有多个与它们相关联的type_ID，我想删除任何重复的类型。如果有重复，我想删除具有最新条目的类型。

Answer 1

窗口函数row_number（）的教科书候选：

;WITH x AS (
    SELECT unique_ID
          ,row_number() OVER (PARTITION BY worker_ID,type_ID ORDER BY date) AS rn
    FROM   tbl
    )
DELETE FROM tbl
FROM   x
WHERE  tbl.unique_ID = x.unique_ID
AND    x.rn > 1

这也解决了(worker_ID,type_ID)上的一组欺骗共享date的情况。
请参阅简化的demo on data.SE。

使用更简单的版本

进行更新

事实证明，这可以简化：在SQL Server中，您可以直接从CTE中删除：

;WITH x AS (
    SELECT unique_ID
          ,row_number() OVER (PARTITION BY worker_ID,type_ID ORDER BY date) AS rn
    FROM   tbl
    )
DELETE x
WHERE  rn > 1

Answer 2

delete from table t
 where exists ( select 1 from table t2 
                 where t2.worker_id = t.worker_id
                   and t2.type_id = t.type_id
                   and t2.date < t.date )

HTH

Answer 3

DELETE FROM @t WHERE unique_Id IN 
(
    SELECT unique_Id FROM 
    (   
        SELECT  unique_Id
                ,Type_Id 
                ,ROW_NUMBER() OVER (PARTITION BY worker_Id, type_Id ORDER BY date) AS rn 
        FROM @t 
    ) Q 
    WHERE rn > 1
)

要测试......

DECLARE @t TABLE
(
    unique_ID  INT IDENTITY,
    worker_ID  INT,
    date  DATETIME,
    type_ID INT
)

INSERT INTO @t VALUES (1, DATEADD(DAY, 1, GETDATE()), 1)
INSERT INTO @t VALUES (1, GETDATE(), 1)
INSERT INTO @t VALUES (2, GETDATE(), 1)
INSERT INTO @t VALUES (1, DATEADD(DAY, 2, GETDATE()), 1)
INSERT INTO @t VALUES (1, DATEADD(DAY, 3, GETDATE()), 2)

SELECT * FROM @t

DELETE FROM @t WHERE unique_Id IN 
(
    SELECT unique_Id FROM 
    (   
        SELECT  unique_Id
                ,Type_Id 
                ,ROW_NUMBER() OVER (PARTITION BY worker_Id, type_Id ORDER BY date) AS rn 
        FROM @t 
    ) Q 
    WHERE rn > 1
)

SELECT * FROM @t

Answer 4

您可以使用此查询

delete from worker where unique_id in (
select max(unique_id)  from worker group by  worker_ID , type_ID having count(type_id)>1)

这里我假设工人是你的表名

SQL：删除重复项

4 个答案:

使用更简单的版本