SQL：如何在外键上使用DISTINCT而不在聚合函数中使用它

Question

So I want to find the average amount everyone who purchased flowers, earns. However, I need to account for duplicates, because some people may have purchased flowers multiple times, and that would mess up the average.

But when I put DISTINCT sales.customerid it wants me to put it in an aggregate function, but when I do that, it separates the averages. It doesn't put it into one average.

SELECT DISTINCT sales.customerid, AVG(moneyearned) AS averageearned
FROM customer,  sales
WHERE customer.customerid = sales.customerid 
AND (purchaseflower = TRUE);

What am I doing wrong?

Answer 1

You might use group by instead of DISTINCT

By the Way, INNER JOIN is better than CROSS JOIN on this case.

SELECT customer.CustomerID, AVG(moneyearned) AS averageearned
FROM customer INNER JOIN sales
ON customer.CustomerID = sales.customerid 
WHERE purchaseflower = TRUE
GROUP BY customer.CustomerID;

SQLFiddle

Answer 2

Use GROUP BY clause instead of Distinct this will group all duplicate customers.

SELECT sales.customerid, AVG(moneyearned) AS averageearned
FROM customer,  sales
WHERE customer.customerid = sales.customerid 
AND (purchaseflower = TRUE)
GROUP BY sales.customerid;

But remember all attributes in SELECT which are not in aggregated function should be present inside GROUP BY clause.