我有三张桌子: 商店 果味 3. vegeta
我确实离开了stores
表,fruity
和vegeta
作为联盟。
我需要按order_id和vendor_id进行分组的计数和总和,但它们仍然有重复的行。
以下是表格:
mysql> SELECT * FROM stores;
+-----+-----------------+
| id | store_name |
+-----+-----------------+
| 701 | Machette Grill |
| 702 | Mateau Conserva |
+-----+-----------------+
mysql> SELECT * FROM fruity;
+------+----------+-----------+----------+-------+
| id | order_id | vendor_id | store_id | sales |
+------+----------+-----------+----------+-------+
| 1816 | 86 | 1 | 701 | 1000 |
| 1817 | 86 | 11 | 701 | 1000 |
| 1818 | 86 | 12 | 701 | 1000 |
| 1819 | 86 | 1 | 702 | 1000 |
| 1820 | 86 | 1 | 702 | 1000 |
| 1821 | 86 | 11 | 702 | 1000 |
| 1822 | 86 | 12 | 702 | 1000 |
| 1823 | 86 | 1 | 702 | 1000 |
| 1824 | 86 | 1 | 702 | 1000 |
| 1825 | 86 | 1 | 702 | 1000 |
| 1826 | 86 | 11 | 702 | 1000 |
| 1827 | 86 | 12 | 702 | 1000 |
| 1828 | 86 | 1 | 701 | 1000 |
+------+----------+-----------+----------+-------+
mysql> SELECT * FROM vegeta;
+----+----------+-----------+----------+-------+
| id | order_id | vendor_id | store_id | sales |
+----+----------+-----------+----------+-------+
| 15 | 86 | 11 | 701 | 2000 |
| 16 | 86 | 12 | 702 | 2000 |
| 17 | 86 | 11 | 701 | 2000 |
| 18 | 86 | 12 | 702 | 2000 |
| 19 | 86 | 11 | 701 | 2000 |
| 20 | 86 | 12 | 702 | 2000 |
+----+----------+-----------+----------+-------+
我在下面运行的代码:
SELECT
s.order_id,
s.store_id,
c.store_name,
s.vendor_id,
s.fruity_count,
s.vegeta_count,
s.fruity_sum,
s.vegeta_sum
FROM stores AS c
LEFT JOIN (
SELECT
order_id AS order_id,
store_id AS store_id,
vendor_id AS vendor_id,
COUNT(sales) AS fruity_count,
0 AS vegeta_count,
SUM(sales) AS fruity_sum,
0 AS vegeta_sum
FROM fruity
WHERE order_id = 86
GROUP BY store_id,vendor_id
UNION
SELECT
order_id AS order_id,
store_id AS store_id,
vendor_id AS vendor_id,
0 AS fruity_count,
COUNT(sales) AS vegeta_count,
0 AS fruity_sum,
SUM(sales) AS vegeta_sum
FROM vegeta
WHERE order_id = 86
GROUP BY store_id,vendor_id) AS s ON s.store_id = c.id
WHERE s.order_id = 86
ORDER BY s.store_id ASC;
我只需要如下所示的结果:
+----------+----------+-----------------+-----------+--------------+--------------+------------+------------+
| order_id | store_id | store_name | vendor_id | fruity_count | vegeta_count | fruity_sum | vegeta_sum |
+----------+----------+-----------------+-----------+--------------+--------------+------------+------------+
| 86 | 701 | Machette Grill | 1 | 2 | 0 | 2000 | 0 |
| 86 | 701 | Machette Grill | 11 | 1 | 3 | 1000 | 6000 |
| 86 | 701 | Machette Grill | 12 | 1 | 0 | 1000 | 0 |
| 86 | 702 | Mateau Conserva | 12 | 2 | 3 | 2000 | 6000 |
| 86 | 702 | Mateau Conserva | 1 | 5 | 0 | 5000 | 0 |
| 86 | 702 | Mateau Conserva | 11 | 2 | 0 | 2000 | 0 |
+----------+----------+-----------------+-----------+--------------+--------------+------------+------------+
我想要做的就是移动重复的行来填充vegeta_count
和vegeta_sum
。不计算新行。
请帮忙。 谢谢。
答案 0 :(得分:0)
您应该在外部查询中使用group by
,如下所示:
SELECT
s.order_id,
s.store_id,
c.store_name,
s.vendor_id,
SUM(s.fruity_count) AS fruity_count,
SUM(s.vegeta_count) AS vegeta_count,
SUM(s.fruity_sum) AS fruity_sum,
SUM(s.vegeta_sum) AS vegeta_sum
FROM stores AS c
LEFT JOIN (
SELECT
order_id AS order_id,
store_id AS store_id,
vendor_id AS vendor_id,
COUNT(sales) AS fruity_count,
0 AS vegeta_count,
SUM(sales) AS fruity_sum,
0 AS vegeta_sum
FROM fruity
WHERE order_id = 86
GROUP BY store_id,vendor_id
UNION
SELECT
order_id AS order_id,
store_id AS store_id,
vendor_id AS vendor_id,
0 AS fruity_count,
COUNT(sales) AS vegeta_count,
0 AS fruity_sum,
SUM(sales) AS vegeta_sum
FROM vegeta
WHERE order_id = 86
GROUP BY store_id,vendor_id) AS s ON s.store_id = c.id
WHERE s.order_id = 86
GROUP BY store_id,vendor_id
ORDER BY s.store_id ASC