196. 删除重复的电子邮箱

SQL架构

表: Person

+-------------+---------+
| Column Name | Type    |
+-------------+---------+
| id          | int     |
| email       | varchar |
+-------------+---------+
id是该表的主键列。
该表的每一行包含一封电子邮件。电子邮件将不包含大写字母。

编写一个 SQL 删除语句来 删除 所有重复的电子邮件，只保留一个id最小的唯一电子邮件。

以 任意顺序 返回结果表。 （注意： 仅需要写删除语句，将自动对剩余结果进行查询）

查询结果格式如下所示。

示例 1:

输入: 
Person 表:
+----+------------------+
| id | email            |
+----+------------------+
| 1  | john@example.com |
| 2  | bob@example.com  |
| 3  | john@example.com |
+----+------------------+
输出: 
+----+------------------+
| id | email            |
+----+------------------+
| 1  | john@example.com |
| 2  | bob@example.com  |
+----+------------------+
解释: john@example.com重复两次。我们保留最小的Id = 1。

DELETE
from
Person
where id in
(SELECT#过滤后剩下的email的id
ttt.p2d
from
(SELECT#剔除重复email中 ID编号排第一的
p2.id as p2d,dense_rank() over(partition by p2.email order by p2.id) as dr
from
Person as p2,

(SELECT   #选出重复的email
email as e1, count(email) as c1
from
Person
group by
email) as p1
where 
p2.email = p1.e1 and p1.c1>1
group by 
p2.id) as ttt
where ttt.dr>1 )

DELETE FROM PERSON
WHERE ID 
NOT IN
(SELECT * FROM
(
    SELECT MIN(ID)
    FROM PERSON
    GROUP BY EMAIL
) AS P1)