Question

我有三个表product，cost和account

CREATE TABLE `product` (
  `id` int(11) unsigned NOT NULL AUTO_INCREMENT,
  `account_id` int(11) DEFAULT NULL,
  `name` varchar(255) NOT NULL DEFAULT '',
  `datestamp` date NOT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=13 DEFAULT CHARSET=utf8;

CREATE TABLE `cost` (
  `account_id` int(11) NOT NULL,
  `amount` float NOT NULL,
  `datestamp` date NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

CREATE TABLE `account` (
  `id` int(11) unsigned NOT NULL AUTO_INCREMENT,
  `name` varchar(255) DEFAULT NULL,
  `email` varchar(255) DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=103 DEFAULT CHARSET=utf8;

+-----+----------+-----------------+
| id  | name     | email           |
+-----+----------+-----------------+
| 101 | Acme Inc | jondoe@host.com |
+-----+----------+-----------------+

+------------+--------+------------+
| account_id | amount | datestamp  |
+------------+--------+------------+
|        101 |      5 | 2014-01-01 |
|        101 |     10 | 2014-01-03 |
|        101 |     15 | 2014-01-06 |
+------------+--------+------------+

+----+------------+---------+------------+
| id | account_id | name    | datestamp  |
+----+------------+---------+------------+
|  1 |        101 | Alfa    | 2014-01-01 |
|  2 |        101 | Bravo   | 2014-01-02 |
|  3 |        101 | Charlie | 2014-01-03 |
|  4 |        101 | Delta   | 2014-01-04 |
|  5 |        101 | Echo    | 2014-01-06 |
|  6 |        101 | Foxtrot | 2014-01-08 |
+----+------------+---------+------------+

我想返回每个单位成本和总数量帐户将产品行与基于日期戳的成本相匹配并按成本和帐户对这些进行分组，以便我可以加以总结。

就我的查询而言，这是我所坚持的尝试将产品与合适的价格相匹配。

select
    a.id as account_id,
    count(p.id) as qty,
    c.amount as cost_per_unit,
    c.datestamp as cost_date,
    p.datestamp as product_date
from product p
left join account a on a.id = p.account_id
left join cost c on 1
where p.datestamp <is within right range based on cost date stamps>
group by c.amount, a.id
;

因此，如果2014-01-01的金额为5美元，2014-01-3和15美元的金额为10美元在2014-01-06每个产品日期在10美元之前盖章将是5美元，然后15美元将是10美元等等..

有关如何处理此问题的任何建议吗？

- 编辑基于下面的答案，我能够得到我需要的东西

SELECT 
    a.id AS account_id,
    COUNT(p.id) AS qty,
    c1.amount AS cost_per_unit,
    c1.datestamp AS cost_date
FROM cost c1
    INNER JOIN account a ON a.id = c1.account_id
    LEFT JOIN cost c2 ON c2.datestamp = 
    (
        select datestamp 
        from cost 
        where account_id = a.id 
        and datestamp > c1.datestamp 
        order by datestamp ASC 
        LIMIT 1
    )
    LEFT JOIN product p ON p.datestamp >= c1.datestamp
WHERE p.datestamp < c2.datestamp OR c2.datestamp IS NULL
GROUP BY a.id, c1.amount
ORDER BY c1.datestamp ASC
;

Answer 1

在生产环境中这可能是非常低效的，但是为了您的修补享受，这个查询应该可以帮助您完成大部分工作。我知道你能够在你到目前为止做得很好的时候擦亮边缘：）

SELECT a.id AS account_id,
    COUNT(p.id) AS qty,
    c2.amount AS previous_cost_per_unit,
    c1.amount AS cost_per_unit,
    c1.datestamp AS cost_date,
    p.datestamp AS product_date
FROM product p
    INNER JOIN account a ON a.id = p.account_id
    INNER JOIN cost c1 ON c1.account_id = p.account_id AND c1.datestamp = p.datestamp
    LEFT JOIN cost c2 ON c2.datestamp = (SELECT MAX(datestamp) FROM cost WHERE account_id = c1.account_id AND datestamp < c1.datestamp)
GROUP BY account_id, product_date

mysql产品行的总和，其中数据标记在成本数据范围内

1 个答案: