在mysql中选择group by的最小值

Question

我的数据库方案,

CREATE TABLE `result` (
`ID` bigint(21) NOT NULL AUTO_INCREMENT,
`datetime` datetime DEFAULT NULL,
`recordnum` int(11) DEFAULT NULL,
`recordtype` int(11) DEFAULT NULL,
`minvalue` int(11) DEFAULT NULL,
`maxvalue` int(11) DEFAULT NULL,
`data1` int(11) DEFAULT NULL,
`data2` int(11) DEFAULT NULL,
`area` varchar(32) DEFAULT NULL,
PRIMARY KEY (`ID`),
UNIQUE KEY `res` (`datetime`,`recordnum`,`area`)
) ENGINE=MyISAM AUTO_INCREMENT=7 DEFAULT CHARSET=latin1;

数据样本

INSERT INTO `result` VALUES ('1', '2013-03-26 12:03:31', '2', '1', '5', '10', '100', '200','Zone1');

INSERT INTO `result` VALUES ('2', '2013-03-26 12:03:31', '2', '1', '2', '7', '20', '20','Zone2');

INSERT INTO `result` VALUES ('3', '2013-03-26 12:03:31', '5', '2', '1', '2', '8', '15','Zone2');

INSERT INTO `result` VALUES ('4', '2013-03-26 12:00:31', '2', '1', '2', '7', '100', '200','Zone1');

INSERT INTO `result` VALUES ('5', '2013-03-26 12:00:31', '2', '1', '3', '9', '100', '200','Zone2');

INSERT INTO `result` VALUES ('6', '2013-03-26 12:00:31', '5', '2', '7', '2', '4', '10', 'Zone2');

INSERT INTO `result` VALUES ('7', '2013-03-25 12:03:31', '2', '1', '5', '10', '100', '200','Zone1');

INSERT INTO `result` VALUES ('8', '2013-03-25 12:03:31', '2', '1', '2', '7', '20', '20','Zone2');

INSERT INTO `result` VALUES ('9', '2013-03-25 12:03:31', '5', '2', '1', '2', '8', '15','Zone2');

INSERT INTO `result` VALUES ('10', '2013-03-25 12:00:31', '2', '1', '2', '7', '100', '200','Zone1');

INSERT INTO `result` VALUES ('11', '2013-03-25 12:00:31', '2', '1', '3', '9', '100', '200','Zone2');

INSERT INTO `result` VALUES ('12', '2013-03-25 12:00:31', '5', '2', '7', '2', '4', '10','Zone2');


INSERT INTO `result` VALUES ('13', '2013-03-25 11:03:31', '2', '1', '3', '10', '100','200', 'Zone1');

INSERT INTO `result` VALUES ('14', '2013-03-25 11:03:31', '2', '1', '8', '7', '20', '20','Zone2');

INSERT INTO `result` VALUES ('15', '2013-03-25 11:03:31', '5', '2', '2', '2', '8', '15','Zone2');

INSERT INTO `result` VALUES ('16', '2013-03-25 11:00:31', '2', '1', '1', '7', '100', '200','Zone1');

INSERT INTO `result` VALUES ('17', '2013-03-25 11:00:31', '2', '1', '9', '9', '100', '200','Zone2');

INSERT INTO `result` VALUES ('18', '2013-03-25 11:00:31', '5', '2', '12', '2', '4', '10','Zone2');

我想要的是在每个记录号和区域以及日期的最小值最小值时查询每个值

  +----+---------------------+-----------+-------+----------+----------+-------+-------+------------+
    | ID | datetime            | recordnum | area  |minvalue  | maxvalue | data1 | data2 | recordtype |
    +----+---------------------+-----------+-------+----------+----------+-------+-------+------------+
    |  2 | 2013-03-26 12:03:31 |         2 | Zone2 |        2 |        7 |    20 |    20 |          1 |
    |  3 | 2013-03-26 12:03:31 |         5 | Zone2 |        1 |        2 |     8 |    15 |          2 |
    |  4 | 2013-03-26 12:00:31 |         2 | Zone1 |        2 |        7 |   100 |   200 |          1 |
    |  8 | 2013-03-25 12:03:31 |         2 | Zone2 |        2 |        7 |    20 |    20 |          1 |
    |  9 | 2013-03-25 12:03:31 |         5 | Zone2 |        1 |        2 |     8 |    15 |          2 |
    | 16 | 2013-03-25 11:00:31 |         2 | Zone1 |        1 |        7 |   100 |   200 |          1 |
    +----+---------------------+-----------+-------+----------+----------+-------+-------+------------+

这个查询给出了很好的结果,但仍然显示重复记录,请帮忙

SELECT  a.*
FROM    result a
        JOIN
        (
            SELECT  datetime as mindatetime,DATE(datetime) dateOnly,area, recordNum, MIN(minvalue) min_val
            FROM    result
            GROUP   BY dateOnly,area, recordNum
        ) b  ON a.area = b.area AND
                a.recordNUM = b.recordNum AND
                a.minvalue = b.min_val ;
+----+---------------------+-----------+------------+----------+----------+-------+-------+-------+
| ID | datetime            | recordnum | recordtype | minvalue | maxvalue | data1 | data2 | area  |
+----+---------------------+-----------+------------+----------+----------+-------+-------+-------+
|  2 | 2013-03-26 12:03:31 |         2 |          1 |        2 |        7 |    20 |    20 | Zone2 |
|  2 | 2013-03-26 12:03:31 |         2 |          1 |        2 |        7 |    20 |    20 | Zone2 |
|  3 | 2013-03-26 12:03:31 |         5 |          2 |        1 |        2 |     8 |    15 | Zone2 |
|  3 | 2013-03-26 12:03:31 |         5 |          2 |        1 |        2 |     8 |    15 | Zone2 |
|  4 | 2013-03-26 12:00:31 |         2 |          1 |        2 |        7 |   100 |   200 | Zone1 |
|  8 | 2013-03-25 12:03:31 |         2 |          1 |        2 |        7 |    20 |    20 | Zone2 |
|  8 | 2013-03-25 12:03:31 |         2 |          1 |        2 |        7 |    20 |    20 | Zone2 |
|  9 | 2013-03-25 12:03:31 |         5 |          2 |        1 |        2 |     8 |    15 | Zone2 |
|  9 | 2013-03-25 12:03:31 |         5 |          2 |        1 |        2 |     8 |    15 | Zone2 |
| 10 | 2013-03-25 12:00:31 |         2 |          1 |        2 |        7 |   100 |   200 | Zone1 |
| 16 | 2013-03-25 11:00:31 |         2 |          1 |        1 |        7 |   100 |   200 | Zone1 |
+----+---------------------+-----------+------------+----------+----------+-------+-------+-------+

Answer 1

如何解决这个问题有很多变化.一种是使用一个单独的子查询,minvalue每个子查询得到最小值area and recordNum.然后,子查询的结果将连接回原始表,以便您可以获取行中的所有列.

SELECT  a.*
FROM    result a
        INNER JOIN
        (
            SELECT  area, recordNum, MIN(minvalue) min_val
            FROM    result
            GROUP   BY area, recordNum
        ) b  ON a.area = b.area AND
                a.recordNUM = b.recordNum AND
                a.minvalue = b.min_val

• SQLFiddle演示

OUTPUT

????????????????????????????????????????????????????????????????????????????????????????????????????????????
? ID ?           DATETIME           ? RECORDNUM ? RECORDTYPE ? MINVALUE ? MAXVALUE ? DATA1 ? DATA2 ? AREA  ?
????????????????????????????????????????????????????????????????????????????????????????????????????????????
?  2 ? March, 26 2013 12:03:31+0000 ?         2 ?          1 ?        2 ?        7 ?    20 ?    20 ? Zone2 ?
?  3 ? March, 26 2013 12:03:31+0000 ?         5 ?          2 ?        1 ?        2 ?     8 ?    15 ? Zone2 ?
?  4 ? March, 26 2013 12:00:31+0000 ?         2 ?          1 ?        2 ?        7 ?   100 ?   200 ? Zone1 ?
????????????????????????????????????????????????????????????????????????????????????????????????????????????