2

我有一个包含多个重复字段的模式，这些字段都是记录类型。例如，架构是这样的：

name: string,

class: record(repeated),

class.id :string

class.title: string,

school: record(repeated),

school.id: string,

school.title: string.

如何在标准 SQL 中查询并返回包含班级和学校列表的学生行（班级和学校完全不相关）。

实际上，一个简单的 SELECT * 有效，但我想选择每一列并重命名它们，即 SELECT name as student_name、classes、schools。我还想过滤一些班级和学校，即只显示满足某些约束的班级。

示例：姓名：'Tom'，班级：[{id：1，title：math}，{id：2，title：physics}，{id：3，title：math}]，学校：[{id：1， title: Yale}, {id: 2, title: Havard}, {id: 3, title: Yale}] 我想按 class.title 和 school.title 过滤以返回这样的行： name: 'Tom',班级：[{id：1，title：math}，{id：3，title：math}]，学校：[{id：1，title：耶鲁}，{id：3，title：耶鲁}]

4

1 回答 1

0

以下是 BigQuery 标准 SQL 

#standardSQL
SELECT name as student_name, 
  ARRAY(SELECT AS STRUCT * FROM UNNEST(class) class WHERE class.title = 'math') classes,
  ARRAY(SELECT AS STRUCT * FROM UNNEST(school) school WHERE school.title = 'Yale') schools
FROM `yourproject.yourdataset.yourtable`

您可以使用您问题中的虚拟数据测试/玩上面的内容： 

#standardSQL
WITH `yourproject.yourdataset.yourtable` AS (
  SELECT 
    'Tom' name, 
    [
      STRUCT(1 AS id, 'math' AS title), 
      (2, 'physics'), 
      (3, 'math')
    ] class, 
    [
      STRUCT(1 AS id, 'Yale' AS title), 
      (2, 'Havard'), 
      (3, 'Yale')
    ] school 
)
SELECT name as student_name, 
  ARRAY(SELECT AS STRUCT * FROM UNNEST(class) class WHERE class.title = 'math') classes,
  ARRAY(SELECT AS STRUCT * FROM UNNEST(school) school WHERE school.title = 'Yale') schools
FROM `yourproject.yourdataset.yourtable`

结果如下 

name    class.id    class.title school.id   school.title     
Tom     1           math        1           Yale     
        3           math        3           Yale

以上应该让您了解如何在重复字段中进行过滤

于 2018-01-24T05:40:45.577 回答