mysql select 从两列查询总和和分组

Question

我有一张名为 ward 的桌子

+---------+-----------+
| ward_id | ward_name |
+---------+-----------+
|       3 | Ward 1    |
|       4 | Ward 2    |
+---------+-----------+

+----+----------+----------------+---------+---------+--------------+------------+
| id | username | date_allocated | ward_id | shop_id | food_name    | amount     |
+----+----------+----------------+---------+---------+--------------+------------+
|  1 | STORE1   | 2020-04-06     |       4 |       2 | Beans        | 1000 Bags  |
|  2 | STAMP    | 2020-04-21     |       4 |       2 | Maize Grains | 60 bags    |
|  3 | STORE1   | 2020-04-21     |       4 |       2 | Maize Flour  | 120 bags   |
|  4 | STORE1   | 2020-04-21     |       4 |       2 | Wheat Flour  | 120 bags   |
|  5 | STAMP    | 2020-04-24     |       3 |       2 | Maize Flour  | 50 bales   |
|  6 | STAMP    | 2020-04-25     |       4 |       2 | Maize Flour  | 1000 Bales |
|  7 | STAMP    | 2020-04-25     |       4 |       2 | Wheat Flour  | 150 pkts   |
+----+----------+----------------+---------+---------+--------------+------------+

用过的食物

+----+----------+------------+---------+---------+---------+-------------+-----------+
| id | username | date_spent | idno    | ward_id | shop_id | food_name   | amount    |
+----+----------+------------+---------+---------+---------+-------------+-----------+
|  5 | STAMP    | 2020-04-21 | 3000000 |       4 |       2 | Beans       | 50 pkts   |
|  6 | STAMP    | 2020-04-21 | 3000000 |       4 |       2 | Maize Flour | 1000 Bags |
|  7 | STAMP    | 2020-04-24 | 3000000 |       3 |       2 | Wheat Flour | 4 Packets |
|  8 | STAMP    | 2020-04-25 | 3000000 |       3 |       3 | Maize Flour | 50 bales  |
|  9 | STAMP    | 2020-04-25 | 3000000 |       3 |       3 | Wheat Flour | 50 pkts   |
| 10 | STAMP    | 2020-04-25 | 3000000 |       4 |       2 | Wheat Flour | 120 Bales |
+----+----------+------------+---------+---------+---------+-------------+-----------+

我创建了一个查询

SELECT ward.ward_name, ward.ward_id, 
       food_allocation.food_name, food_allocation.ward_id, spent_food.food_name, 
       sum(food_allocation.tot1) as food,
       sum(spent_food.tot2) as spent,
       sum(food_allocation.tots-spent_food.tots) as Balance
FROM ward
INNER JOIN (
  SELECT ward_id,food_name,
         sum(amount) tot1,
         sum(amount) tots
  FROM food_allocation 
  GROUP BY food_name,ward_id
) food_allocation ON ward.ward_id = food_allocation.ward_id
INNER JOIN (
  SELECT ward_id,food_name,
         sum(amount) tot2,
         sum(amount) tots
  FROM spent_food 
  GROUP BY food_name, ward_id
) spent_food ON food_allocation.ward_id = spent_food.ward_id 
GROUP BY food_allocation.ward_id, food_allocation.food_name 
ORDER BY food_allocation.ward_id;

我在下面得到这个错误的数字

+-----------+---------+--------------+---------+-------------+------+-------+---------+
| ward_name | ward_id | food_name    | ward_id | food_name   | food | spent | Balance |
+-----------+---------+--------------+---------+-------------+------+-------+---------+
| Ward 1    |       3 | Maize Flour  |       3 | Maize Flour |  100 |   104 |      -4 |
| Ward 2    |       4 | Beans        |       4 | Beans       | 3000 |  1170 |    1830 |
| Ward 2    |       4 | Maize Flour  |       4 | Beans       | 3360 |  1170 |    2190 |
| Ward 2    |       4 | Maize Grains |       4 | Beans       |  180 |  1170 |    -990 |
| Ward 2    |       4 | Wheat Flour  |       4 | Beans       |  810 |  1170 |    -360 |
+-----------+---------+--------------+---------+-------------+------+-------+---------+

我的预期答案是

+-----------+---------+--------------+---------+-------------+------+-------+---------+
| ward_name | ward_id | food_name    | ward_id |     food | spent | Balance |
+-----------+---------+--------------+---------+-------------+------+-------+---------+
| Ward 1    |       3 | Maize Flour  |       3 |       50 |    50 |      0 |
| Ward 2    |       4 | Beans        |       4 |     1000 |    50 |    950 |
| Ward 2    |       4 | Maize Flour  |       4 |     1120 |  1000 |    120 |
| Ward 2    |       4 | Maize Grains |       4 |       60 |     0 |     60 |
| Ward 2    |       4 | Wheat Flour  |       4 |      270 |   120 |    150 |
+-----------+---------+--------------+---------+-------------+------+-------+---------+

Answer 1

当您在最后一张桌子上加入 Sped_food 时，您必须在正确的地方加入他们，即 ward_id 和 food_name。

仅供参考，我缩短了表格的名称，这样我的文字就更少了，而且阅读起来好多了

SELECT w.ward_name, w.ward_id, 
       f.food_name
       , MAX(f.ward_id)
       , MAX(s.food_name), 
       sum(f.tot1) as food,
       sum(s.tot2) as spent,
       sum(f.tots-s.tots) as Balance
FROM ward w
INNER JOIN (
  SELECT ward_id,food_name,
         sum(amount) tot1,
         sum(amount) tots
  FROM food_allocation 
  GROUP BY food_name,ward_id
) f  ON w.ward_id = f.ward_id
INNER JOIN (
  SELECT ward_id,food_name,
         sum(amount) tot2,
         sum(amount) tots
  FROM spent_food 
  GROUP BY food_name, ward_id
) s ON f.ward_id = s.ward_id  AND f.food_name = s.food_name
GROUP BY w.ward_name, w.ward_id, f.food_name 
ORDER BY f.ward_id;

Answer 2

您必须将连接转换为左连接，并在正确的列上连接查询。
另外，由于您在每个子查询中进行聚合，因此不需要最终聚合，因此我删除了最后一个GROUP BY ：

SELECT w.ward_name, 
       w.ward_id, 
       fa.food_name,
       coalesce(fa.tot1,0) as food,
       coalesce(sf.tot2,0) as spent,
       coalesce(fa.tots,0)-coalesce(sf.tots,0) as Balance
FROM ward w
LEFT JOIN (
  SELECT ward_id,food_name,
         sum(amount) tot1,
         sum(amount) tots
  FROM food_allocation 
  GROUP BY food_name,ward_id
) fa  ON w.ward_id = fa.ward_id
LEFT JOIN (
  SELECT ward_id,food_name,
         sum(amount) tot2,
         sum(amount) tots
  FROM spent_food 
  GROUP BY food_name, ward_id
) sf ON fa.ward_id = sf.ward_id  AND fa.food_name = sf.food_name
ORDER BY fa.ward_id, fa.food_name;

见演示。
结果：

> ward_name | ward_id | food_name    | food | spent | Balance
> :-------- | ------: | :----------- | ---: | ----: | ------:
> Ward 1    |       3 | Maize Flour  |   50 |    50 |       0
> Ward 2    |       4 | Beans        | 1000 |    50 |     950
> Ward 2    |       4 | Maize Flour  | 1120 |  1000 |     120
> Ward 2    |       4 | Maize Grains |   60 |     0 |      60
> Ward 2    |       4 | Wheat Flour  |  270 |   120 |     150