BigQuery GroupBy 与 STRUCT答案

【问题标题】：BigQuery GroupBy with STRUCTBigQuery GroupBy 与 STRUCT
【发布时间】：2018-05-25 08:11:34
【问题描述】：

在 BigQuery 中，我可以使用标准 SQL 成功运行以下查询：

SELECT 
  COUNT(*) AS totalCount,
  city,
  DATE_TRUNC(timeInterval.intervalStart, YEAR) AS start
FROM 
  sandbox.CountByCity
GROUP BY 
    city, start

但是当我将 start 值嵌套在 STRUCT 中时它会失败，就像这样......

SELECT 
  COUNT(*) AS totalCount,
  city,
  STRUCT(
    DATE_TRUNC(timeInterval.intervalStart, YEAR) AS start
  ) as timeSpan
FROM 
  sandbox.CountByCity
GROUP BY 
    city, timeSpan.start

在这种情况下，我收到以下错误消息：

无法从 [10:11] 的 SELECT 列表别名 timeSpan 中对字段引用进行 GROUP BY

编写查询以使start 值嵌套在STRUCT 中的正确方法是什么？

【问题讨论】：

标签： sql google-bigquery

【解决方案1】：

您可以使用ANY_VALUE 执行此操作。您获得的结构值是明确定义的，因为整个组的值是相同的：

SELECT 
  COUNT(*) AS totalCount,
  city,
  ANY_VALUE(STRUCT(
    DATE_TRUNC(timeInterval.intervalStart, YEAR) AS start
  )) as timeSpan
FROM 
  sandbox.CountByCity
GROUP BY 
    city, DATE_TRUNC(timeInterval.intervalStart, YEAR);

这是一个使用一些示例数据的示例：

WITH `sandbox.CountByCity` AS (
  SELECT 'Seattle' AS city, STRUCT(DATE '2017-12-11' AS intervalStart) AS timeInterval UNION ALL
  SELECT 'Seattle', STRUCT(DATE '2016-11-10' AS intervalStart) UNION ALL
  SELECT 'Seattle', STRUCT(DATE '2017-03-24' AS intervalStart) UNION ALL
  SELECT 'Kirkland', STRUCT(DATE '2017-02-01' AS intervalStart)
)
SELECT 
  COUNT(*) AS totalCount,
  city,
  ANY_VALUE(STRUCT(
    DATE_TRUNC(timeInterval.intervalStart, YEAR) AS start
  )) as timeSpan
FROM 
  `sandbox.CountByCity`
GROUP BY 
    city, DATE_TRUNC(timeInterval.intervalStart, YEAR);

您还可以考虑使用submitting a feature request 启用GROUP BY 和STRUCT 类型。

【讨论】：

【解决方案2】：

不知道你到底为什么想要这个——但相信这是出于某种原因——所以在下面试试（至少在形式上它会按照你的要求做）

#standardSQL
SELECT
  totalCount, 
  city, 
  STRUCT(start) timeSpan
FROM (
  SELECT 
    COUNT(*) AS totalCount,
    city,
    DATE_TRUNC(timeInterval.intervalStart, YEAR) AS start
  FROM `sandbox.CountByCity`
  GROUP BY city, start
)

【讨论】：