有這樣一張表，表數(shù)據(jù)及結(jié)果如下：
| school_id | school_name | total_student | test_takers | |
|---|---|---|---|---|
| 1239 | Abraham Lincoln High School | 55 | 50 | |
| 1240 | Abraham Lincoln High School | 70 | 35 | |
| 1241 | Acalanes High School | 120 | 120 | |
| 1242 | Academy Of The Canyons | 30 | 30 | |
| 1243 | Agoura High School | 89 | 40 | |
| 1244 | Agoura High School | 100 | 50 |
我們可以看出，school_name的字段值有重復(fù)數(shù)據(jù)(Abraham Lincoln High School 和Agoura High School分別出現(xiàn)兩次)，那么如何刪除這兩條數(shù)據(jù)，從而只讓這兩個(gè)數(shù)值出現(xiàn)一次呢？ 具體實(shí)現(xiàn)方法如下：
1、刪除重復(fù)記錄,保存Id最小的一條
delete FROM `test` WHERE `school_name` in (SELECT `school_name`
FROM `test`
GROUP BY `school_name`
HAVING COUNT( * ) >1) and school_id not in (select min(school_id) from test group by school_id having count(* )>1)
先使用GROUP BY having語法查詢出重復(fù)的數(shù)據(jù)，然后刪除重復(fù)數(shù)據(jù)并保留school_id最小的一條.
2、刪除重復(fù)記錄,保存Id最大的一條
delete FROM `test` WHERE `school_name` in (SELECT `school_name`
FROM `test`
GROUP BY `school_name`
HAVING COUNT( * ) >1) and school_id not in (select max(school_id) from test group by school_id having count(* )>1)
原理和上面一樣。
