转载-MYSQL查询重复记录的步骤
程序员文章站
2022-05-07 09:45:42
...
转载---MYSQL查询重复记录的方法 MYSQL查询重复记录的方法很多,下面就为您介绍几种最常用的MYSQL查询重复记录的方法,希望对您学习MYSQL查询重复记录方面能有所帮助。 1、查找表中多余的重复记录,重复记录是根据单个字段(peopleId)来判断 select * from p
转载---MYSQL查询重复记录的方法MYSQL查询重复记录的方法很多,下面就为您介绍几种最常用的MYSQL查询重复记录的方法,希望对您学习MYSQL查询重复记录方面能有所帮助。
1、查找表中多余的重复记录,重复记录是根据单个字段(peopleId)来判断
select * from people
where peopleId in (select peopleId from people group by peopleId having count(peopleId) > 1)
2、删除表中多余的重复记录,重复记录是根据单个字段(peopleId)来判断,只留有rowid最小的记录
delete from people
where peopleId in (select peopleId from people group by peopleId having count(peopleId) > 1)
and rowid not in (select min(rowid) from people group by peopleId having count(peopleId )>1)
3、查找表中多余的重复记录(多个字段)
select * from vitae a
where (a.peopleId,a.seq) in (select peopleId,seq from vitae group by peopleId,seq having count(*) > 1)
4、删除表中多余的重复记录(多个字段),只留有rowid最小的记录
delete from vitae a
where (a.peopleId,a.seq) in (select peopleId,seq from vitae group by peopleId,seq having count(*) > 1)
and rowid not in (select min(rowid) from vitae group by peopleId,seq having count(*)>1)
5、查找表中多余的重复记录(多个字段),不包含rowid最小的记录
select * from vitae a
where (a.peopleId,a.seq) in (select peopleId,seq from vitae group by peopleId,seq having count(*) > 1)
and rowid not in (select min(rowid) from vitae group by peopleId,seq having count(*)>1)
上一篇: 通俗易懂的数据库设计三大范式