AWS Redshift：列到行（从长到长）

I'm writing a query which result will be displayed as dashboard on redash. The query is basically a sum of several columns of table within a specific time range:

我正在编写一个查询，结果将在redash上显示为仪表板。查询基本上是特定时间范围内的几列表的总和：

SELECT
   sum(col_A),
   sum(col_B),
   sum(col_C)
FROM
   table_X
WHERE
   timestamp > '2018-01-01'
   AND timestamp < '2018-02-01'

The result is something like this:

结果是这样的：

col_A    col_B    col_C    
123      456      789

However, to render the dasboard on redash properly, I need the result in long format:

但是，要正确渲染redboard，我需要长格式的结果：

column   sum
col_A    123
col_B    456
col_C    789

How can I cast the result from wide to long format?

如何将结果从长格式转换为长格式？

1 个解决方案

#1

A simple way is to use union all:

一个简单的方法是使用union all：

SELECT sum(col_A) FROM table_X WHERE timestamp > '2018-01-01' AND timestamp < '2018-02-01'
UNION ALL
SELECT sum(col_B) FROM table_X WHERE timestamp > '2018-01-01' AND timestamp < '2018-02-01'
UNION ALL
SELECT sum(col_C) FROM table_X WHERE timestamp > '2018-01-01' AND timestamp < '2018-02-01';

You may need to be careful about the ordering. I would include a second column specifying which is being aggregated.

您可能需要注意订购。我将包括第二列，指定正在聚合的列。

Three separate queries is not quite as bad as it sounds. Redshift is a columnar database, so only the columns references should be read.

三个单独的查询并不像听起来那么糟糕。 Redshift是一个柱状数据库，因此只应读取列引用。

#1