Question

我有一张名为“付款”的表格如下：

+----+--------+
| ID | Amount |
+----+--------+
|  1 | 1.00   |
|  2 | 3.00   |
|  3 | -15.00 |
|  3 | 20.00  |
+----+--------+

“金额”表示转手的现金 - 正数的记录表示已收到现金，而负数的记录表示现金返还为变更。

以下查询：

SELECT ID
     , CASE WHEN Amount >= 0 THEN Amount END AS CashTaken
     , CASE WHEN Amount < 0 THEN Amount END AS CashBack  
  FROM Payments
 WHERE TicketID IN ('1', '2', '3')

...给我这个结果集：

+----+-----------+----------+
| ID | CashTaken | CashBack |
+----+-----------+----------+
|  1 | 1.00      | NULL     |
|  2 | 3.00      | NULL     |
|  3 | NULL      | -15.00   |
|  3 | 20.00     | NULL     |
+----+-----------+----------+

如何按ID对这些记录进行分组，以便我可以获得以下内容？

+----+-----------+----------+
| ID | CashTaken | CashBack |
+----+-----------+----------+
|  1 | 1.00      | NULL     |
|  2 | 3.00      | NULL     |
|  3 | 20.00     | -15.00   |
+----+-----------+----------+

我有点尴尬，因为我觉得这很简单，就在我面前。

Answer 1

您可以尝试使用MIN，MAX：

SELECT TicketID 
     , MAX(CASE WHEN Amount >= 0 THEN Amount END) AS CashTaken
     , MIN(CASE WHEN Amount < 0 THEN Amount END) AS CashBack  
FROM Payments
WHERE TicketID IN ('1', '2', '3')
GROUP BY TicketID

使用给定数据执行示例：

DECLARE @Payments TABLE (TicketID INT, Amount DECIMAL(5,2));

INSERT INTO @Payments(TicketID, Amount) 

SELECT 1, 1.00 UNION
SELECT 2, 3.00 UNION
SELECT 3, -15.00 UNION
SELECT 3, 20.00;

SELECT TicketID
     , MAX(CASE WHEN Amount >= 0 THEN Amount END) AS CashTaken
     , MIN(CASE WHEN Amount < 0 THEN Amount END) AS CashBack  
FROM @Payments
WHERE TicketID IN ('1', '2', '3')
GROUP BY TicketID

结果：

TicketID    CashTaken   CashBack
---------------------------------
1           1.00       NULL
2           3.00       NULL
3           20.00     -15.00

需要“分组”记录，但我没有聚合函数

1 个答案: