Mysql 删除重复数据保留一条有效数据

==============================

©Copyright 蕃薯耀 2023-01-31

蕃薯耀的博客_CSDN博客-蕃薯耀分享,java,js领域博主

DELETE FROM SZ_Building WHERE id NOT IN (  SELECT t.min_id FROM (    SELECT MIN(id) AS min_id FROM SZ_Building GROUP BY BLDG_NO  ) t);

原理：

1. 根据字段对数据进行分组，查询出所有分组的最小ID（即要保留的不重复数据）
2. 将查询出来的数据（所有不重复的数据）存放到临时表中
3. 从原来的表中删除ID不在临时表中的重复数据

当表的数据量达到百万级以上时，删除数据很慢，建议这样处理：

• 先把min(id)存到临时表（SZ_Building_id）：

create table SZ_Building_idSELECT MIN(id) AS id FROM SZ_Building GROUP BY BLDG_NO;

• 给临时表的id设置成主键或者创建索引（加快后面的查询速度）

alter table add index ind_xxName(字段名：如ID);

• 删除id不在临时表的数据。

delete from ​SZ_Building where id not in(  select id from SZ_Building_id);

DELETE FROM SZ_Water_Level WHERE id NOT IN (  SELECT t.min_id FROM (    SELECT MIN(id) AS min_id FROM SZ_Water_Level GROUP BY CZBM,SJ,SW  ) t);

SELECT * FROM SZ_Building WHERE id NOT IN (  SELECT t.min_id FROM (    SELECT MIN(id) AS min_id FROM SZ_Building GROUP BY BLDG_NO  ) t);

或

SELECT * FROM SZ_Building WHERE BLDG_NO IN (SELECT BLDG_NO FROM SZ_Building GROUP BY BLDG_NO HAVING COUNT(1)>1) AND id NOT IN (SELECT MIN(id) FROM SZ_Building GROUP BY BLDG_NO HAVING COUNT(1)>1);

 ==============================

©Copyright 蕃薯耀 2023-01-31

蕃薯耀的博客_CSDN博客-蕃薯耀分享,java,js领域博主

来源地址：https://blog.csdn.net/w995223851/article/details/128816753