SELECT语句,去除某个字段的重复信息,例如: 表名:table id uid username message dateline 1 6 a 111 1284240714(时间戳) 2 6 a 222 1268840565 3 8 b 444 1266724527 4 9 c 555 1266723391 执行语句(去除username字段重复信息并按时间排序): SELECT * FROM table a INNER JOIN ( SELECT max( dateline ) AS dateline FROM table GROUP BY uid ) b ON a.dateline = b.dateline GROUP BY id ORDER BY a.dateline DESC 结果: id uid username message dateline 1 6 a 111 1284240714(时间戳) 3 8 b 444 1266724527 4 9 c 555 1266723391 此语句用于显示最新记录信息,在一个区域内不允许某个信息(例如:用户)同时出现多次(一次以上)。 后记:效率问题 开始用了个这语句: select * from table where dateline IN ( select max(dateline) from table GROUP BY uid ) ORDER BY dateline DESC IN:当处理数据量比较大的时候,就没效率可言了,所以优化成内联,计算下快了6倍多。。。 继续条效率就加索引了~~