Question

因此一直在处理一些数据，目前已按照

的行进行输出

客户|原因
客户1 |答案1，答案3，答案2，答案4，答案5，答案1，答案3，答案1

Big Query标准sql中是否存在要消除此字符串中重复项并以下面的输出结尾的内容？

客户|原因
客户1 |答案1，答案3，答案2，答案4，答案5

预先感谢

Answer 1

假设我正确理解了这个问题，则需要类似以下内容的

', '

这将在DISTINCT分隔符上拆分字符串，然后将子字符串聚合为一个新字符串，并使用(oal)关键字删除了重复项。

Answer 2

投票支持Elliott的答案时-想要添加另一个选项（BigQuery标准SQL）：

#standardSQL
WITH `project.dataset.table` AS (
  SELECT 'Customer1' customer, 'Answer1, Answer3, Answer2, Answer4, Answer5, Answer1, Answer3, Answer1' answers 
)
SELECT * REPLACE(
  ARRAY_TO_STRING(ARRAY(SELECT DISTINCT answer
    FROM UNNEST(SPLIT(answers, ', ')) AS answer 
  ), ', ') AS answers)   
FROM `project.dataset.table`

产生所需的结果

Row customer    answers  
1   Customer1   Answer1, Answer3, Answer2, Answer4, Answer5

如果出于某种原因您希望对这些值进行排序-您只需添加如下一行

#standardSQL
WITH `project.dataset.table` AS (
  SELECT 'Customer1' customer, 'Answer1, Answer3, Answer2, Answer4, Answer5, Answer1, Answer3, Answer1' answers 
)
SELECT * REPLACE(
  ARRAY_TO_STRING(ARRAY(SELECT DISTINCT answer
    FROM UNNEST(SPLIT(answers, ', ')) AS answer 
    ORDER BY answer
  ), ', ') AS answers)   
FROM `project.dataset.table`

结果为

Row customer    answers  
1   Customer1   Answer1, Answer2, Answer3, Answer4, Answer5

注意：最有可能需要订购与您问题中的特定用例无关-在其他情况下可以方便使用

是否可以从Big Query中的字符串中删除重复项？

2 个答案: