在每天的多个条目中只获取最后一行

Question

我被困在一个查询中，为每一个实体获得24行的查询中的最后一个条目。我已经阅读了this thred中的sollution相当类似的问题，但我无法将其纳入到我现有的查询中。

我现有的查询：

SELECT distinct b.Udc_id as EAN, sax.attrib_01, c.channel_no, a.local_read_time, a.read, a.Status 
from register_reads a, lookup_mv_channels1 c, lookup_mv_sdps b 
LEFT outer JOIN s_asset@mudr2sebl sa 
ON sa.x_udc_asset_id = b.Udc_id 
LEFT outer JOIN s_asset_xm@mudr2sebl sax 
ON sax.par_row_id = sa.row_id 
WHERE c.Channel_id = a.channel_id 
AND b.meter_id = c.meter_id 
AND sax.attrib_02 = 'Y' 
AND a.local_read_time >= (SYSDATE) - 1 
order by b.Udc_id, a.local_read_time 

从这个意愿，例如结果是：

||EAN | ATTRIB_01 | channel_no | LOCAL_READ_TIME | READ | STATUS|| 
||1234| ready  |  1  |09/11/2015 01:00:01|  0| val || 
||1234| ready  |  1  |09/11/2015 02:00:01| 10| val || 
||1234| ready  |  1  |  ...  | ...| val || 
||1234| ready  |  1  |09/11/2015 23:59:01| 60| val || 
||3456| ready  |  1  |09/11/2015 01:00:01| 204| val || 
||3456| ready  |  1  |  ...  | ...| val || 
||3456| ready  |  1  |09/11/2015 23:58:01| 400| val || 

我想达到的结果是：反馈后

||EAN | ATTRIB_01 | channel_no | LOCAL_READ_TIME | READ | STATUS|| 
||1234| ready  |  1  |09/11/2015 23:59:01| 60| val || 
||3456| ready  |  1  |09/11/2015 23:58:01| 400| val || 

变化从@levelonehuman：

SELECT b.Udc_id as EAN, 
     c.physical_channel_no, 
     a.local_read_time, 
     a.cum_read, 
     a.validation_Status 
FROM register_reads a, lookup_mv_channels1 c, lookup_mv_sdps b 
    INNER JOIN (SELECT b.Udc_id, 
         MAX(a.LOCAL_READ_TIME) AS LOCAL_READ_TIME 
       FROM register_reads, lookup_mv_sdps b 
       GROUP BY b.Udc_id) AS t2 
    ON b.Udc_id = t2.Udc_id AND b.LOCAL_READ_TIME = t2.LOCAL_READ_TIME 
; 

Answer 1

注：这是MSSQL

您需要添加额外的连接S表示无论其他表，但可以使用MAX(LOCAL_READ_TIME)和GROUP BY EAN让你的最后一行：

SELECT EAN, 
     ATTRIB_01, 
     channel_no, 
     LOCAL_READ_TIME, 
     READ, 
     STATUS 
FROM TableName t1 
    INNER JOIN ( SELECT EAN, 
          MAX(LOCAL_READ_TIME) AS LOCAL_READ_TIME 
        FROM TableName 
        GROUP BY EAN 
    ) AS t2 ON t1.EAN = t2.EAN AND t1.LOCAL_READ_TIME = t2.LOCAL_READ_TIME 

的INNER JOIN部分将选择在TableName行是有EAN和MAX(LOCAL_READ_TIME) - 那么，你会回来参加到TableName（同一张表），根据JOIN标准选择其余数据。

实际上，这会过滤掉不符合EAN + MAX(LOCAL_READ_TIME)条件的行。