Question

我有一个具有以下架构的表

 test_name, subject_1, subject_2

它有类似的条目

  ('A',['a','b','c'],['b','d']),('B',['d','a','b'],['a','b']),etc

现在，我希望像 set(subject_1)-set(subject_2) 那样执行集合减法。因此，对于 'A'，输出将是 ['a','c']。对于'B'，['d']。

任何帮助！！我尝试在互联网上搜索，但徒劳无功。

Answer 1

在 Athena / Presto 中，集合基于 Array。

WITH example_table AS
 (SELECT 'A' AS test_name, ARRAY['a','b','c'] AS subject_1, ARRAY['b','d'] AS subject_2 UNION ALL
  SELECT 'B', ARRAY['d','a','b'], ARRAY['a','b'] )

SELECT test_name, 
       array_except(
             subject_1,
             array_intersect(subject_1,subject_2)
       ) AS diff
FROM example_table

WITH 部分用于为主查询创建临时表。
Diff 是 array_intersect 和 array_except

从第二个 AWS Athena 中减去第一个数组的元素

1 个答案: