Question

我有一张表格如下：

Create Table Movies (
    ID int,
    Actor varchar (50),
    MovieName varchar (50),
    ReleaseDate Date
)

让我们插入一些样本数据

Insert into Movies (ID,Actor,MovieName, ReleaseDate)
Select 1,'Sean Connery', 'Gone gone wild', '2/1/2014'
Union
Select 2,'Roberta Jones', 'My first love', '2/1/2014'
Union
Select 3,'Michael Young', 'Booba spirit', '2/1/2014'
Union
Select 4,'Sean Conenry', 'Gone gone wild 2', '10/1/2014'
Union
Select 5,'Roberta Jones', 'My Second Love', '2/3/2015'
Union
Select 6,'Michael Young', 'Boboa spirit 2', '2/2/2016'
Union
Select 7,'Sean Connery', 'Gone gone wild 3', '2/1/2015'
Union
Select 8,'Sean Connery', 'Gone gone wild 4', '2/1/2016

我想知道演员，每个演员每年的发行日期和电影数量，但只返回电影数量最多的一年。

这是一个在SQL Server上运行良好的查询

Select * from Movies

select Actor, Release, NbrMovies
From (Select Actor, Release, NbrMovies, rank() over (partition by Actor Order By NbrMovies desc) as rn 
      From (Select Actor, year(ReleaseDate) as Release, count(*) as NbrMovies
            From Movies
            Group by Actor, year(ReleaseDate)
           ) x 
     ) y
where rn = 1

我想知道是否有一种方法可以在没有任何窗口函数（如rank（）或row_number（）的情况下编写查询，因为我也将它用于Oracle。

由于

Answer 1

您的查询适用于SQL Server和Oracle。但是，您可以删除一层子查询：

select Actor, ReleaseYear, NbrMovies
from (Select Actor, year(ReleaseDate) as ReleaseYear,   
             count(*) as NbrMovies,
             rank() over (partition by Actor Order By count(*) desc) as rn 
      From Movies m
      Group by Actor, year(ReleaseDate)
     ) m 
where rn = 1;

查询每年/演员的最大值

1 个答案: