SQLFiddle

Question

我正在研究api以返回建筑项目列表。

SELECT 
`sample`.`Development Type` AS 'Development Type',
`sample`.`Postcode` AS 'Postcode',
`sample`.`Value Desc` AS 'Value Desc'
FROM
`sample`
WHERE `sample`.`Postcode` LIKE '%SW1%'

，这将返回一个建筑项目列表。

Development Type          | Postcode      | Value Desc

Refurbishment & Extension | SW1E 6LB      | 400,000
Refurbishment & Extension | SW1E 6LB      | 400,000
Refurbishment & Extension | SW1X 8AL      | 75,000
Refurbishment & Extension | SW1X 8AL      | 75,000 
Refurbishment & Extension | SW1P 3RE      | 225,000
Refurbishment & Extension | SW1P 3RE      | 225,000
Extension                 | SW1X 7QU      | 200,000
Extension                 | SW1X 7QU      | 200,000
Refurbishment             | SW1X 8AL      | 75,000

我需要修改查询以获得这样的输出。

Development Type          | Sum Value Desc

Refurbishment & Extension | 700,000
Extension                 | 200,000
Refurbishment             | 75,000

- 所以按开发类型分组 - 删除重复的结果（可能是我应该在查询时执行的标记，总计数千来处理逗号）

Answer 1

如果不是因为每个记录似乎重复出现，这将是一个简单的GROUP BY查询。但这可以通过首先使用带有SELECT DISTINCT的子查询来删除重复项，然后从此子查询中分组以获得所需的总和来解决：

SELECT t.`Development Type`, FORMAT(SUM(t.`Value Desc`), 0)
FROM
(
    SELECT DISTINCT `Development Type`, `Postcode`, `Value Desc`
    FROM sample
) t
GROUP BY t.`Development Type`
ORDER BY SUM(t.`Value Desc`) DESC

请按照以下链接查看正在运行的演示：

SQLFiddle

<强>更新

由于您的源数据可能是CSV格式，因此您可以将外部SELECT语句更改为以下内容以处理逗号：

SELECT t.`Development Type`,
       FORMAT(SUM(CAST(REPLACE(`Value Desc`, ',' , '') AS UNSIGNED)), 0)

您也可以将ORDER BY子句更改为：

ORDER BY SUM(CAST(REPLACE(`Value Desc`, ',' , '') AS UNSIGNED)) DESC

SQL分组按开发类型和值的总和

1 个答案:

SQLFiddle