Question

我有一个包含大约200个属性的表。对于每个属性，我想知道所有可能的值。大多数属性都有2到20个可能的值。

如果属性可能有太多不同的值（例如邮政编码），那么我只想知道计数。（请注意，这不需要是一个独特的计数，我更多地将计数步骤作为占位符，因此我保持源表中的属性顺序比其他任何东西。）

我想保存此输出到csv文件，但如果需要为单个属性保存多个值，则可以接受不同的输出文件类型。

对我要实现的目标的缩短和简化查询是：

COPY(
SELECT
(SELECT COUNT(id) FROM complete_file_301115) AS id,
(SELECT DISTINCT (title) FROM complete_file_301115) AS title,
(SELECT COUNT(firstname) FROM complete_file_301115) AS firstname,
(SELECT COUNT(lastname) FROM complete_file_301115) AS lastname,
(SELECT COUNT(address1) FROM complete_file_301115) AS address1,
(SELECT COUNT(address2) FROM complete_file_301115) AS address2,
(SELECT DISTINCT (city) FROM complete_file_301115) AS city,
(SELECT DISTINCT (county) FROM complete_file_301115) AS county,
(SELECT COUNT(postcode) FROM complete_file_301115) AS postcode,
(SELECT DISTINCT (gender) FROM complete_file_301115) AS gender,
(SELECT DISTINCT (job) FROM complete_file_301115) AS job,
(SELECT DISTINCT (children) FROM complete_file_301115) AS children) TO
'/Users/Shared/test123.csv' (FORMAT CSV, DELIMITER ',', HEADER, ENCODING LATIN1);

当我运行查询时，Postgres给了我错误：

ERROR:  more than one row returned by a subquery used as an expression

我知道这个错误基本上是说我不能为csv中的单个属性写多个值。

我的问题是，无论如何都要保存一个输出文件，其中包含每个属性的所有可能值，例如

id：1,2,3,4
标题：先生，夫人，小姐，女士，博士等。

非常感谢提前

Answer 1

您可以将其转换为文字。它适合你吗？

 ...
 (array_to_string( array(SELECT DISTINCT (city) FROM complete_file_301115), ',') ) AS city,
 ...

多个选择不同的查询输出到csv文件

1 个答案: