Question

我有一个包含许多列的表，我必须根据一列找到副本。

即。如果我在customer_name中找到重复的Customer_name

我必须删除源表中的所有重复。
将所有这些行发送到具有相同结构的其他表。

Answer 1

我猜每行都有一个唯一的Id主键。

这将插入您的重复行表：

Insert into duplicateRowsTable
select * from myTable t1
where (select count(*) from myTable t2 where t1.customerId = t2.customerId) > 1

您从duplicateRowsTable中删除了好行：

delete from duplicatesTable
where --this is not the faulty row for each customerId

最后你从第一张桌子上删除了：

delete from myTable
where id IN (select id from duplicatesTable)

Answer 2

如果您有两个这样的表：

CREATE TABLE t1 (ID int, customerName varchar(64))
CREATE TABLE t2 (ID int, customerName varchar(64))

你可以做这样的事情:( ID列只是为了设置一个基础来保存，你可以根据需要改变它）

--First Copy

WITH CTE_T1
AS
(
SELECT
  ID,
  customerName,
  ROW_NUMBER() OVER(PARTITION BY customerName ORDER BY ID) as OrderOfCustomer
  FROM 
  t1
)
INSERT INTO t2
SELECT ID, customerName FROM cte_T1
WHERE OrderOfCustomer > 1;

--Then Delete

WITH CTE_T1
AS
(
SELECT
  ID,
  customerName,
  ROW_NUMBER() OVER(PARTITION BY customerName ORDER BY ID) as OrderOfCustomer
  FROM 
  t1
)
DELETE FROM CTE_T1
WHERE OrderOfCustomer > 1

这是一个SQLFiddle来展示它是如何运作的。

Answer 3

试试这个：

用于移动重复项

INSERT Into DuplicatesTable
SELECT *
FROM 
(SELECT *, ROW_NUMBER() OVER(PARTITION BY Customer_name ORDER BY Customer_name) As RowID,
FROM SourceTable)  as temp
WHERE RowID > 1

用于侦测：

WITH TableCTE
AS
(
SELECT *,
ROW_NUMBER() OVER(PARTITION BY Customer_name ORDER BY Customer_name) AS RowID
FROM SourceTable
)
DELETE
FROM TableCTE
WHERE RowID> 1

将有故障的行发送到其他表

3 个答案: