找出每一行的最大值

如何解决找出每一行的最大值

| 这是我的问题的简化版本。有两个表，一个有机器，第二个有关于其状态更改的信息（带有日期参考）（实际上不是日期，而是班次索引，这就是为什么我必须进行另一个查询才能获取日期本身的原因）。 我想获取每台计算机的最后状态。 我的查询是：

SELECT * FROM machines  
LEFT JOIN (SELECT * FROM statechange,dates WHERE....ORDER BY date DESC LIMIT 0,1) AS state
ON state.mid=machines.mid 

但由于查询，我只能得到一台机器。 不创建视图怎么办？     

解决方法

        

select m.*,s.`date` from machines as m
inner join (
select mid,`date`
from statechange
order by `date` desc) as s
on m.mid = s.mid
group by m.mid

    ,        通过执行以下操作，您将获得每台计算机的最新状态更改。

SELECT mid,max(date)
FROM statechange
GROUP BY mid

然后加入该连接，以获取每台计算机的当前状态。 （猜测列名。）

SELECT m.*,sc.state,sc.date
FROM machines m
INNER JOIN statechange sc 
    ON ( sc.mid = m.mid )
INNER JOIN (
            SELECT mid,MAX(date) date
            FROM statechange
            GROUP BY mid 
            ) s
    ON ( (s.mid = sc.mid) and (s.date = sc.date) );

我认为查看每台计算机的当前状态可能很有用。     ,        假设

statechange

表的主键是Max有意义的东西（如整数）

Select ...
From Machines As M
    Left Join   (
                Select S1.mid,Max( S1.PrimaryKeyCol ) As PrimaryKeyCol
                From statechange As S1
                    Join dates As D1
                        On D1.statechangeFK = S1.PrimaryKeyCol
                Where D1.date = (
                                Select Max( D2.date )
                                From dates As D2
                                Where D2.statechangeFK = D1.statechangeFK
                                )
                Group By S1.mid
                ) As LastStateChange
        On LastStateChange.mid = M.mid

    Left Join statechange As S
        On S.PrimaryKeyCol = LastStateChange.PrimaryKeyCol

如果您想从statechange表返回列，则最后一个联接。 根据其他信息进行修订

Select ...
From Machines As M
    Left Join   (
                Select S1.mid,Max( S1.PrimaryKeyCol ) As PrimaryKeyCol
                From dates As D1
                    Join statechange As S1
                        On S1.DatesForeignKeyCol = D1.PrimaryKeyCol
                Where D1.date = (
                                Select Max( D2.date )
                                From dates As D2
                                    Join statechange As S2
                                        On S2.DatesForeignKeyCol = D2.PrimaryKeyCol
                                Where S2.mid = S1.mid
                                )
                Group By S1.mid
                ) As LastStateChange
        On LastStateChange.mid = M.mid

    Left Join statechange As S
        On S.PrimaryKeyCol = LastStateChange.PrimaryKeyCol

    Left Join dates As D
        On D.PrimaryKeyCol = S.DatesForeignKeyCol

我使用上述方法的原因之一是要考虑以下情况：对于相同的

mid

和

date

值，您可能有多个

statechange

行。在这种情况下，以上方法将使用主键列作为决胜局。