删除重复保留一条要点
- 思路一：建临时表存去重结果，DROP 原表，RENAME 回去
- 临时表用 GROUP BY 取每组一条，结构可自定义
- 思路二：DELETE 删掉 id 不在'每组最小 id'集合里的行
- 删前务必先 SELECT 确认要删的范围，防止误伤
- 重建表方式会丢掉原表索引，记得补建
- 删除后用唯一索引兜底，避免再次产生重复
删除重复保留一条示例
-- 方法一：搬到临时表再换回
CREATE TABLE member_tmp
SELECT username, email, MAX(phone) AS phone
FROM member GROUP BY username, email;
DROP TABLE member;
ALTER TABLE member_tmp RENAME TO member;
-- 方法二：删掉每组里非最小 id 的重复行
DELETE FROM member
WHERE id NOT IN (
SELECT min_id FROM (
SELECT MIN(id) AS min_id
FROM member
GROUP BY username, email
) AS t
);
避坑提醒：用 NOT IN 子查询删副本最稳：先按重复键取每组最小 id，不在集合里的就是多余副本，整组只留最小那个。
