Question

我正在使用SQL Server数据库。

鉴于以下查询，我试图获得每个班级的最高分。 Scores表有两行50行，所以我想要总共2行。但是，因为我有Scores.Id，所以它返回Scores的每一行，因为Scores.Id是唯一的。当然，简单的解决方案是删除Scores.Id列，除了我需要知道Scores.Id以进行其他查找。

SELECT 
    Class_Id, Scores.Id, MAX(Scores.ClassScore) AS Score
FROM  
    Classes 
INNER JOIN
    Scores ON Classes.Id = Scores.Class_Id                
GROUP BY 
    Scores.Class_Id, Scores.Id

Answer 1

这是你想要的吗？

[$parse:syntax] Syntax Error: Token '{' is unexpected, expecting [)] at column 45 of the expression [vm.solderEcriture('debit',ecriture,formName_{{$index}}.$valid)] starting at [{{$index}}.$valid)].

根据您想要的信息，加入select s.* from (select s.*, max(s.classscore) over (partition by s.class_id) as max_classscore from scores s ) s where classscore = max_classscore;是多余的。

Answer 2

class_id的最高分数只是

select class_id, max(classScore) score from scores
group by class_id

如果您需要知道得分表中哪些行具有最高得分，您可以进行加入。如果每个班级的最高分数超过一个，则可能会有两行以上。

select id, class_id, classscore from
scores s
inner join
(
    select class_id, max(classScore) score from scores
    group by class_id
)
t
on t.class_id = s.class_id and t.score = s.classScore

或者我可能会使用cte

with maxScores as
(
    select class_id, max(classScore) score from scores
    group by class_id
)
select id, class_id, classscore from
scores s
on maxScores.class_id = s.class_id and maxScores.score = s.classScore

Answer 3

这看起来像是rank窗口函数的作业：

SELECT Class_Id, Id, Score
FROM   (SELECT Class_Id, Scores.Id AS Id, Score, 
               RANK() OVER (PARTITION BY Class_Id, Scores.Id
                            ORDER BY Score DESC) AS rk
        FROM   Classes 
        JOIN   Scores ON Classes.Id = Scores.Class_Id) t
WHERE  rk = 1

通过返回太多行限制最大值和组

3 个答案: