SQL-删除所有重复项并仅保留一个

我有一张像下面这样的桌子

Order_Id (PK), Customer_ID, Order_Amount
  A01             AA01         100
  A01             AA01         100 (duplicate)
  A02             AA01         200
  A03             AA02         300

我想删除副本并保留剩余的一个

我可以使用下面的选项选择副本

select *
from 
(
select order_id, customer_id, order_amount, row_number () over (partition by order_id order by order_Id) as rnk
from mytable
) where rnk > 1

但是从mytable中删除此记录的下一步是什么？我需要创建一个临时表吗

---

谢谢

如果所有数据都相同，我认为红移无法识别行。所以，我认为你最好的办法是重新创建表格：

create table temp_mytable as
    select distinct *
    from mytable;

truncate table mytable;

insert into mytable
    select *
    from distinct mytable;

---

如果您的表确实有主键，那么可以使用其他方法删除行。

如果所有数据都相同，我认为Redshift无法识别行。所以，我认为你最好的办法是重新创建表格：

create table temp_mytable as
    select distinct *
    from mytable;

truncate table mytable;

insert into mytable
    select *
    from distinct mytable;

如果您的表确实有主键，那么可以使用其他方法删除行。

应该可以

SELECT *
FROM 
(
  SELECT 
     order_id, 
     customer_id, 
     order_amount, 
     row_number () OVER (partition by order_id order by order_Id) as rnk
  FROM mytable
  GROUP BY order_id
) WHERE rnk > 1

这应该是一种工作

SELECT *
FROM 
(
  SELECT 
     order_id, 
     customer_id, 
     order_amount, 
     row_number () OVER (partition by order_id order by order_Id) as rnk
  FROM mytable
  GROUP BY order_id
) WHERE rnk > 1

---

请尝试下面的查询，它应符合您的要求

delete from CUSTOMER
where rowid not in
(select max(rowid) from CUSTOMER group by order_amount);

---

请尝试下面的查询，它应符合您的要求

delete from CUSTOMER
where rowid not in
(select max(rowid) from CUSTOMER group by order_amount);

---

order\u id

如果重复则不是主键。

order\u id

如果重复则不是主键。