改进查询以提高写作效率

Question

您将如何改进 SQL Server 中的以下查询以查找哪些课程中的男女比例超过 70%？ （我添加了所有相关表格的图表......并非所有表格都是必需的）

SELECT 
    F.CourseId, 
    CAST(ROUND(CAST(M.Men AS FLOAT) / CAST(F.Women AS FLOAT) * 100, 2) AS NUMERIC(12, 2)) AS Ratio
FROM 
    (SELECT 
         c.CourseId, COUNT(DISTINCT c.StudentId) AS Women
     FROM 
         Classrooms c
     INNER JOIN 
         Students s ON c.StudentId = s.StudentId
     WHERE 
         s.Gender = 'F'
     GROUP BY 
         c.CourseId) F
INNER JOIN 
    (SELECT 
         c.CourseId, COUNT(DISTINCT c.StudentId) AS Men
     FROM 
         Classrooms c
     INNER JOIN 
         Students s ON c.StudentId = s.StudentId
     WHERE  
         s.Gender = 'M'
     GROUP BY 
         c.CourseId) M ON F.CourseId = M.CourseId
WHERE 
    CAST(M.Men AS FLOAT) / CAST(F.Women AS FLOAT) * 100 > 70
ORDER BY 
    F.CourseId 

我想练习有效且更短（更好）的查询。

谢谢大家！

Answer 1

您可以使用单个查询，例如：

SELECT 
     c.CourseId, 
     COUNT(DISTINCT CASE WHEN s.Gender = 'F' THEN c.StudentId ELSE NULL END) AS Women,
     COUNT(DISTINCT CASE WHEN s.Gender = 'M' THEN c.StudentId ELSE NULL END) AS Man,
     CAST(COUNT(DISTINCT CASE WHEN s.Gender = 'M' THEN c.StudentId ELSE NULL END) AS FLOAT) / COUNT(DISTINCT c.StudentId) AS MensRate
FROM Classrooms c
INNER JOIN Students s ON c.StudentId = s.StudentId
GROUP BY c.CourseId
HAVING CAST(COUNT(DISTINCT CASE WHEN s.Gender = 'M' THEN c.StudentId ELSE NULL END) AS FLOAT) / COUNT(DISTINCT c.StudentId) > 0.7

MS SQL fiddle group by

意味着您应该将男士计数除以总计数以计算速率。