Question

SELECT `loanac`.`id`, `loanac`.`area`
FROM loanac, loantrans
WHERE `loanac`.`lacc`=`loantrans`.`account`
GROUP BY `loanac`.`lacc`
HAVING SUM(IFNULL(`loantrans`.`out`,0)) > SUM(IFNULL(`loantrans`.`in`,0))

QUERY RESULT：

----------------------------------------
id      area
----------------------------------------
5       A
165     C
166     C
167     C
232     D
91      B

这个查询给了我所有的id＆amp;对于已经输入＆gt;的lacc的区域，此查询是OK＆amp;工作，我只需要从此查询中获取输出GROUP BY区域和COUNT个id。有人可以帮忙吗？

我想要的是

----------------------------------------
COUNT(id)      GROUP BY(area)
----------------------------------------
1               A
5               B
9               C
1               D

注意：条件不会改变目标：想要在相同条件下查找/计算（loanac.area）分组的数量（loanac.area）

Answer 1

我想你想要这样的东西。将area添加到GROUP BY行，并使用COUNT（id）来获取交易次数。

SELECT count(`loanac`.`id`),
       SUM(IFNULL(`loantrans`.`out`,0)) AS transout,
       SUM(IFNULL(`loantrans`.`in`,0)) AS transin,
       `loanac`.`area`,
       `loanac`.`lacc`
FROM loanac, loantrans
WHERE `loanac`.`lacc`=`loantrans`.`account`
GROUP BY `loanac`.`lacc`,`loanac`.`area`
HAVING transout > transin

编辑：要提供您正在寻找的输出


SELECT count(loanac.id), `loanac`.`area`
FROM loanac, loantrans
WHERE `loanac`.`lacc`=`loantrans`.`account`
GROUP BY `loanac`.`lacc`,`loanac`.`area`
HAVING SUM(IFNULL(`loantrans`.`out`,0)) > SUM(IFNULL(`loantrans`.`in`,0))

Answer 2

我认为这符合您的要求：

SELECT COUNT(id), area FROM
  (SELECT id, area FROM loanac
     JOIN loantrans ON loanac.lacc = loantrans.account
     GROUP BY loanac.lacc
     HAVING SUM(IFNULL(loantrans.out, 0)) > SUM(IFNULL(loantrans.in,0))) t
  GROUP BY area;

上面，我有一个loanac.area组，因为这就是原始示例的样子。如果您想按loanac.lacc进行分组，只需将上面的GROUP BY更改为GROUP BY loanac.lacc。

编辑：在评论中链接的测试数据库中加载并运行查询会产生：

+-----------+------+
| COUNT(id) | area |
+-----------+------+
|        86 | A    |
|        81 | B    |
|        64 | C    |
|        66 | D    |
|        20 | E    |
|         3 | F    |
+-----------+------+

如何使用“COUNT”和“GROUP BY”＆amp; “哪里”

2 个答案: