前言
最近开发的时候遇到一个任务,需要对重复的数据进行筛选,只取插入时间最早的一条数据。这里介绍一下解决这类去重问题的几种思路
先看样例数据
解决思路一:先group by找到每个人最新的数据插入时间(insert_time),再通过insert_time作为条件表关联的条件筛选出每个人最新的数据
1. 先group by找到每个人最新的数据插入时间(insert_time)
SELECT
T.u_name,
MAX( T.insert_time ) AS t_inserttime
user_test T
GROUP BY
T.u_name
2. 通过insert_time作为条件表关联的条件筛选出每个人最新的数据
SELECT
T1.id,
T1.u_name,
T1.u_sex,
T1.u_phone,
T1.insert_time,
T1.update_by
`user_test` T1,
( SELECT T2.u_name, MAX( T2.insert_time ) AS t_inserttime FROM user_test T2 GROUP BY T2.u_name ) T3
WHERE
T1.u_name = T3.u_name
AND T1.insert_time = T3.t_inserttime
结果如下:
解法2:通过row_number()over()函数解决 (适用于Oracle)
row_number() over()函数的主要功能是分组排序,实现类似group by + order by的效果
SELECT
( SELECT T.*, ROW_NUMBER ( ) OVER ( PARTITION BY T.u_name ORDER BY T.insert_time DESC ) RW FROM user_test T ) FI
WHERE
FI.RW =1
这里partition by
实现了根据用户名进行分组,order by
对结果集根据插入时间进行排序,row_number()
函数将每一组的行数单独标注了出来。最后我们取rw=1的数据,也就取到了每个重复用户数据的最新一条数据。
样板数据参考:
DROP TABLE IF EXISTS `user_test`;
CREATE TABLE `user_test` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`u_name` varchar(50) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT NULL,
`u_sex` varchar(20) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT NULL,
`u_phone` varchar(255) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT NULL,
`insert_time` datetime NULL DEFAULT NULL,
`update_by` varchar(50) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT NULL,
PRIMARY KEY (`id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 6 CHARACTER SET = utf8 COLLATE = utf8_general_ci ROW_FORMAT = Compact;
-- ----------------------------
-- Records of user_test
-- ----------------------------
INSERT INTO `user_test` VALUES (1, '小明', '男', '13288888888', '2020-10-28 09:44:16', 'admin');
INSERT INTO `user_test` VALUES (2, '小明', '男', '13288888888', '2020-10-28 09:45:01', 'admin');
INSERT INTO `user_test` VALUES (3, '小明', '男', '13288888888', '2020-10-28 09:45:35', 'admin');