Question

我有3个表分别称为交易，客户和票证。每张票都有一个类别（剧院，电影院或音乐会）。我想找到每个类别中购买次数最多的人的名字：

例如：

Category | Name |  Purchases
 Theater   George      5
 Cinema     Tom        12
 Cinema     Mary       12
 Concert    John       7

WITH PurchasesSum AS (
    SELECT Category,Customers.Name,COUNT(Transactions.Customer_ID) AS Purchases,
        RANK() OVER (ORDER BY COUNT(Transactions.Customer_ID) DESC) AS Rank
    FROM Transactions
    INNER JOIN Customers
        ON Transactions.Customer_ID= Customers.ID
    INNER JOIN Tickets
    ON Transactions.Ticket_num = Tickets.Ticket_num
    GROUP BY Name,Category
)

SELECT Name, Purchases,Category
FROM PurchasesSum 
ORDER BY Category ASC;

到目前为止，我到了这一步，但它返回了所有这些
例如：

Category | Name |  Purchases
 Cinema     Tom        12
 Cinema     Mary       12
 Cinema     John       3
 Concert    John       7
 Concert    Tom        1
 Theater   George      5
 Theater   Mary        2

Answer 1

您非常亲密。将OVER的{{1}}子句扩展到rank()，以便排名相对于类别。然后将PARTITION BY category子句放在外部查询过滤器中，排名为1。

WHERE

Answer 2

在Postges中，我建议使用DISTINCT ON。它不仅更简洁，而且通常更快：

SELECT DISTINCT ON (Category) Category, c.Name, 
       COUNT(*) AS Purchases,
FROM Transactions t JOIN
     Customers c
     ON t.Customer_ID = c.ID JOIN
     Tickets ti
     ON t.Ticket_num = ti.Ticket_num
GROUP BY Category, c.Name
ORDER BY Category, COUNT(*) DESC;

如何从表格中获取每个类别的名字？ PostgreSQL

2 个答案: