如何在MySql中减小此排名查询的大小？

Question

我有一个排名查询，可以对挑战中团队的表现进行排名。

数据层次结构如下： 团队有成员 成员有活动 活动有活动类型 挑战有活动类型

如果我想在单个挑战中对所有团队的表现进行排名，这个查询效果很好：

SELECT     t.teamID, t.teamName, 
        scoring.challengeID, 
        outerchallenge.name AS ChallengeName, outerchallenge.description AS ChallengeDescription, outerchallenge.startDate, outerchallenge.endDate, 
        scoring.standardValueSum, scoring.standardUnit, scoring.rank 
FROM challenge outerchallenge 
    LEFT JOIN ( 
        SELECT teamID, challengeID, standardValueSum, standardUnit, FIND_IN_SET(standardValueSum, scores ) AS rank 
        FROM ( 
            SELECT teammember.teamID, mc.challengeID, sum(v.standardValue) standardValueSum, v.standardUnit 
            FROM v_activitystats v 
                INNER JOIN memberchallenge mc ON v.memberID = mc.memberID AND v.standardValue > 0 
                INNER JOIN teammember ON v.memberID = teammember.memberID 
                INNER JOIN challenge c ON mc.challengeID = c.challengeID 
                INNER JOIN challengeactivitytype cat ON c.challengeID = cat.challengeID AND cat.activityTypeID = v.activityTypeID 
            WHERE v.activityDate BETWEEN c.startDate AND c.endDate 
                AND c.challengeID = 33  
            GROUP BY standardUnit, challengeID, teamID 
            ) vstats 
    CROSS JOIN ( 
        SELECT GROUP_CONCAT( DISTINCT standardValueSum ORDER BY standardValueSum DESC ) AS scores 
        FROM ( 
            SELECT teammember.teamID, mc.challengeID, sum(v.standardValue) standardValueSum 
            FROM v_activitystats v 
                INNER JOIN memberchallenge mc ON v.memberID = mc.memberID AND v.standardValue > 0 
                INNER JOIN teammember ON v.memberID = teammember.memberID 
                INNER JOIN challenge c ON mc.challengeID = c.challengeID 
                INNER JOIN challengeactivitytype cat ON c.challengeID = cat.challengeID AND cat.activityTypeID = v.activityTypeID 
            WHERE v.activityDate BETWEEN c.startDate AND c.endDate 
                AND c.challengeID = 33  
            GROUP BY challengeID, teamID 
            ) vstats 
        ) scores 
    ) scoring 

    ON outerchallenge.challengeID = scoring.challengeID 
        INNER JOIN team t ON scoring.teamID = t.teamID 

以下是格式化查询：http://mysql.pastebin.com/XggRL5kX

ChallengeID，团队，排名 99红队1 99 Blue Team 2

再一次，这适用于特定的挑战，（ID = 33）

我希望获得具有相同排名的查询，但是针对多个挑战，例如那些已经结束的挑战。

我尝试了这个查询：

SELECT rankings.teamID, stuff.teamName, rankings.challengeID, 
        rankings.ChallengeName, rankings.ChallengeDescription, rankings.startDate, rankings.endDate, 
        rankings.standardValueSum, rankings.standardUnit, rankings.rank 
FROM challenge chal 
    LEFT JOIN ( 
    SELECT t.teamID, t.teamName, scoring.challengeID, 
        outerchallenge.name AS ChallengeName, outerchallenge.description AS ChallengeDescription, outerchallenge.startDate, outerchallenge.endDate, 
        scoring.standardValueSum, scoring.standardUnit, scoring.rank 
    FROM challenge outerchallenge 
        LEFT JOIN ( 
            SELECT teamID, challengeID, standardValueSum, standardUnit, FIND_IN_SET(standardValueSum, scores ) AS rank 
            FROM ( 
                SELECT teammember.teamID, mc.challengeID, sum(v.standardValue) standardValueSum, v.standardUnit 
                FROM v_activitystats v 
                    INNER JOIN memberchallenge mc ON v.memberID = mc.memberID AND v.standardValue > 0 
                    INNER JOIN teammember ON v.memberID = teammember.memberID 
                    INNER JOIN challenge c ON mc.challengeID = c.challengeID 
                    INNER JOIN challengeactivitytype cat ON c.challengeID = cat.challengeID AND cat.activityTypeID = v.activityTypeID 
                WHERE v.activityDate BETWEEN c.startDate AND c.endDate 
                GROUP BY standardUnit, challengeID, teamID ) vstats 
            CROSS JOIN ( 
                SELECT GROUP_CONCAT( DISTINCT standardValueSum ORDER BY standardValueSum DESC ) AS scores 
                FROM ( 
                    SELECT teammember.teamID, mc.challengeID, sum(v.standardValue) standardValueSum 
                    FROM v_activitystats v 
                        INNER JOIN memberchallenge mc ON v.memberID = mc.memberID AND v.standardValue > 0 
                        INNER JOIN teammember ON v.memberID = teammember.memberID 
                        INNER JOIN challenge c ON mc.challengeID = c.challengeID 
                        INNER JOIN challengeactivitytype cat ON c.challengeID = cat.challengeID AND cat.activityTypeID = v.activityTypeID 
                    WHERE v.activityDate BETWEEN c.startDate AND c.endDate 
                    GROUP BY challengeID, teamID 
                ) vstats 
            ) scores 
        ) scoring ON outerchallenge.challengeID = scoring.challengeID 
        INNER JOIN team t ON scoring.teamID = t.teamID 
) rankings ON chal.challengeID = rankings.challengeID
WHERE chal.endDate <= current_date()

以下是格式化查询：http://mysql.pastebin.com/mSZwtDm3

但是，不是每个挑战都有第一名和第二名，排名都是跨越所有挑战。喜欢这个

ChallengeID，团队，排名 99红队1 99蓝队2 134红队3 134蓝队4 443红队5 442 Blue Team 6

所以，我想，我正在评估错误位置的排名，但我对如何使这项工作有点想法。我怎样才能得到这样的结果： ChallengeID，团队，排名 99红队1 99蓝队2 134红队1 134蓝队2 443红队1 443蓝队2

Answer 1

听起来你正在寻找的是Oracle子句“PARTITION BY”，它可以进行子组，然后允许你像你试图做的那样排名多次。

可能最简单的方法是，不是试图显示由mySQL生成的排名，而是创建自己的排名。

这是一个比我能够制定的更好的例子，嘿，这几乎是凌晨4点！ http://www.xaprb.com/blog/2006/12/02/how-to-number-rows-in-mysql/