SQL从查询结果中删除重复项

时间:2015-12-07 09:21:17

标签: mysql sql greatest-n-per-group

这是我的SQL:

 SELECT * FROM unity_database.unit_uptime_daily
        inner join unity_database.units on unity_database.units.id = unity_database.unit_uptime_daily.unit_id
        where unity_database.units.location_id = 1

以下是结果第一部分的屏幕截图: enter image description here

我试图只为每个具有最新日期的unit_id显示一个结果,所以在这个例子中,unit_id为30的单位应该只显示一次,它应该是底行。

有什么想法吗?

3 个答案:

答案 0 :(得分:1)

WITH X AS
(
SELECT * 
       ,ROW_NUMBER() OVER (PARTITION BY Unit_Id ORDER BY [Timestamp] DESC) rn
FROM unity_database.unit_uptime_daily
inner join unity_database.units 
on unity_database.units.id = unity_database.unit_uptime_daily.unit_id
where unity_database.units.location_id = 1
)
SELECT * FROM X Where rn = 1

OR

SELECT * 
FROM (
    SELECT * 
           ,ROW_NUMBER() OVER (PARTITION BY Unit_Id ORDER BY [Timestamp] DESC) rn
    FROM unity_database.unit_uptime_daily
    inner join unity_database.units 
    on unity_database.units.id = unity_database.unit_uptime_daily.unit_id
    where unity_database.units.location_id = 1
     )A
WHERE rn = 1

答案 1 :(得分:1)

这是您可以考虑的MySQL解决方案:

SELECT t1.id, t1.unit_id, t1.uptime, t1.total_update, t2.timestamp
FROM
(
    SELECT *
    FROM unity_database.unit_uptime_daily ud1 INNER JOIN unity_database.units ud2
        ON ud2.id = ud1.unit_id
    WHERE ud2.location_id = 1
) t1
INNER JOIN
(
    SELECT ud1.unit_id, MAX(ud1.timestamp) AS timestamp
    FROM unity_database.unit_uptime_daily ud1 INNER JOIN unity_database.units ud2
        ON ud2.id = ud1.unit_id
    WHERE ud2.location_id = 1
    GROUP BY ud1.unit_id
) t2
ON t1.unit_id = t2.unit_id AND t1.timestamp = t2.timestamp

答案 2 :(得分:0)

如果这个答案能够为您提供您期望的结果:

 SELECT MAX(id) AS ID, unit_id, uptime, total_update, MAX(timestamp) as timesstamp FROM unity_database.unit_uptime_daily
    inner join unity_database.units on unity_database.units.id = unity_database.unit_uptime_daily.unit_id
 where unity_database.units.location_id = 1
 GROUP BY unit_id, uptime, total_update
PS:我没有测试,但我很肯定它会起作用