题解 | #删除emp_no重复的记录,只保留最小的id对应的记录。#
删除emp_no重复的记录,只保留最小的id对应的记录。
http://www.nowcoder.com/practice/3d92551a6f6d4f1ebde272d20872cf05
看到这道题的时候,觉得不是特别难,不就是根据emp_no分组查一下最小的id,然后只要id不在这个范围内的都删除掉就行了,一执行,哦吼,sql报错了,原因是MySQL中不允许在子查询的同时删除本表数据
delete
from titles_test
where id not in (
select min(id) as id
from titles_test
group by titles_test.emp_no
);
于是,你可以换种思路,将子查询的结果放入临时表中
delete
from titles_test
where id not in (
select id
from (
select min(id) as id
from titles_test
group by titles_test.emp_no
) as a
);
还有一种思路:就是创建一个视图,将表titles_test的emp_no个数大于1的id放入该视图,然后删除的时候从视图查出要删除的id即可。
create view temp(id) as
(
select max(id)
from titles_test
group by emp_no
having count(*) > 1);
delete
from titles_test
where id in (select id from temp);