如何在 SQL 中将表连接到子查询结果？

Question

我有以下查询：

SELECT customer_id, product_id, COUNT(product_id) CUSTOMER_PRODUCT_COUNT
  2  FROM Order_Details
  3  Group BY customer_id, product_id
  4  HAVING COUNT(product_id) > 1;

结果如下表：

CUSTOMER_ID PRODUCT_ID CUSTOMER_PRODUCT_COUNT
---------- ---------- ----------------------
C1         P3                              2

我正在尝试将其用作子查询，以将 customer_ID 从 Customer 表连接到 customer_name。 product_id 列也存在于另一个名为 Order_Details 的表中，我也在尝试加入该表，以便提供客户的地址：

到目前为止，我提出了以下查询，它给出了一个错误：

SELECT c.customer_name, a.country, a.zone, a.district, a.city
  2  FROM (
  3     SELECT customer_id, product_id, COUNT(product_id) CUSTOMER_PRODUCT_COUNT
  4     FROM Order_Details
  5     Group BY customer_id, product_id
  6     HAVING COUNT(product_id) > 1) AS o
  7  JOIN Customer c ON c.customer_id = o.customer_id
  8  JOIN Address a ON a.address_id = o.address_id;

Answer 1

这个查询看起来正确：

SELECT c.customer_name, a.country, a.zone, a.district, a.city
FROM (SELECT customer_id, product_id, COUNT(product_id) CUSTOMER_PRODUCT_COUNT
      FROM Order_Details
      GROUP BY customer_id, product_id
      HAVING COUNT(product_id) > 1
     ) AS o JOIN
     Customer c
     ON c.customer_id = o.customer_id JOIN
     Address a
     ON a.address_id = o.address_id;

除了对o.address_id 的引用（我猜想是c.address_id）之外，我只能想到两个可能的问题：

• 列引用不正确。
• 您使用的是 Oracle，它不允许在 FROM 子句中使用 as。

无论如何，我都会这样写查询：

SELECT c.customer_name, a.country, a.zone, a.district, a.city
FROM (SELECT customer_id, product_id, COUNT(*) as CUSTOMER_PRODUCT_COUNT
      FROM Order_Details od
      GROUP BY customer_id, product_id
      HAVING COUNT(*) > 1
     ) o JOIN
     Customer c
     ON c.customer_id = o.customer_id JOIN
     Address a
     ON a.address_id = c.address_id;

没有必要计算非NULLvalues in a column when you simply want to count rows. That is whatCOUNT(*)`。