如何在 Bigquery 的一个 select 语句中使用两个 group by 语句？答案

【问题标题】：How to use two group by statement in one select statement in Bigquery?如何在 Bigquery 的一个 select 语句中使用两个 group by 语句？
【发布时间】：2020-03-11 15:17:04
【问题描述】：

我有一个三列的表。 (Column_A, Column_B, Column_C)

表中数据结构如下：

Column_A  |  Column_B  |  Column_C
----------------------------------
  Jeeva   |  football  |   1
  Jeeva   |  football  |   2
  Jeeva   |  football  |   3
  Jeeva   |  handball  |   5
  Jeeva   |  handball  |   1
  Jeeva   |  handball  |   7
  Jeeva   |  handball  |   2
  Jeeva   |  voleyball |   1
  Jeeva   |  voleyball |   2
 Bharathi |  voleyball |   10
 Bharathi |  voleyball |   7
 Bharathi |  voleyball |   2

我想要以下结果：

Column_A  |  Column_B  |  Column_C
----------------------------------
  Jeeva   |  football  |   3
  Jeeva   |  handball  |   7
  Jeeva   |  voleyball |   2
 Bharathi |  voleyball |   10

期望的结果：对于 Column_A 中的用户在 Column_B 中参加过的每项运动，在 Column_C 中找到最大得分值

我在Bigquery中写了如下查询，

SELECT DISTINCT a.* FROM `project.dataset.individual_results` a 
RIGHT JOIN
(SELECT column_a, column_b, MAX(column_c) AS max_val FROM `project.dataset.individual_results` GROUP BY column_a GROUP BY column_b) b 
ON a.column_a = b.column_a AND a.column_b= b.max_val WHERE a.column_c IS NOT NULL

由于第二个 Group By 语句，我收到错误消息。

错误信息：

Syntax error: Expected ")" but got keyword GROUP

如何解决这个问题并获得想要的结果？

【问题讨论】：

标签： google-bigquery

【解决方案1】：

GROUP BY column_a GROUP BY column_b - 应该只是GROUP BY column_a, column_b。

但我也认为您根本不需要 JOIN，第二个子查询为您提供了您描述的确切结果：

SELECT column_a, column_b, MAX(column_c) AS max_val 
FROM `project.dataset.individual_results` 
GROUP BY column_a, column_b

【讨论】：