需求：

因系统设计问题，导致实际业务中插入数据库的值有重复的情况，现在需要把重复的数据删除，只保留最后插入的数据。（数据库设计时，数据表有默认唯一的自增长ID）

解决思路：

1.先把重复重复记录的主键查询出来

2.查询重复内容的最大自增长ID

3.刷选出主键属于重复，且ID不在最大的自增长ID的数据，并进行删除

具体代码：

delete from [Emails] where ID not in (
  SELECT 
     max(ID) as ID
  FROM [Emails]
  group by [EmailAddress]
  having COUNT([EmailAddress])>1
  ) and [EmailAddress] in (
    SELECT 
      [EmailAddress]
  FROM [Emails]
  group by [EmailAddress]
  having COUNT([EmailAddress])>1
  )