使用 BigQuery SQL 查找字符串的一部分并在分隔符之间提取数据

Question

我有这样一个专栏：

String_to_Extract
A~S1_B~S2_C~S11
A~S1_B~S3_C~S12
C~S13_A~S11_B~S4

“~”前的部分应该是列名。 “~”之后的部分应该是行值。这是由 "_" 分隔的。因此，结果应如下所示：

String_to_Extract	A	B	C
A~S1_B~S2_C~S11	S1	S2	S11
A~S1_B~S3_C~S12	S1	S3	S12
C~S13_A~S11_B~S4	S11	S4	S13

这是我的方法：

SELECT
String_to_Extract,
SUBSTRING(String_to_Extract, INSTR(Advertiser, "A~")+2, ?) AS A,
SUBSTRING(String_to_Extract, INSTR(Advertiser, "B~")+2, ?) AS B,
SUBSTRING(String_to_Extract, INSTR(Advertiser, "C~")+2, ?) AS C,
From Table

如何获取每一列的 ~ 和下一个 _ 之间的部分？

很乐意提供帮助！

Answer 1

一种方法使用 REGEXP_EXTRACT:

SELECT
    REGEXP_EXTRACT(String_to_Extract, r"(?:^|_)A~([^_]+)") AS A,
    REGEXP_EXTRACT(String_to_Extract, r"(?:^|_)B~([^_]+)") AS B,
    REGEXP_EXTRACT(String_to_Extract, r"(?:^|_)C~([^~]+)") AS C
FROM yourTable;

Answer 2

考虑以下方法 (BigQuery)

select * from (
  select String_to_Extract, col_val[offset(0)] as col, col_val[offset(1)] as val
  from your_table, unnest(split(String_to_Extract, '_')) kv,
  unnest([struct(split(kv, '~') as col_val)])
)
pivot (any_value(val) for col in ('A', 'B', 'C'))

如果应用于您问题中的示例数据 - 输出为

Answer 3

您也可以使用这种方法，先对拆分后的项目排序，然后选择值：


select 
   split(ordered[safe_offset(0)], '~')[safe_offset(1)] as A,
   split(ordered[safe_offset(1)], '~')[safe_offset(1)] as B,
   split(ordered[safe_offset(2)], '~')[safe_offset(1)] as C
 from (
    select 
        array(select _ from unnest(split(Advertiser, '_') ) as _ order by 1) as ordered
    from dataset.table
)

使用 BigQuery SQL 查找字符串的一部分并在分隔符之间提取数据

Finding part of string and extracting data between delimiter using BigQuery SQL

mysql

sql

split

substring

google-bigquery