'Splitting the string into columns to extract values using BigQuery
How can i split the string by a specific character and extract the value of each. The idea is that i need to extract each word between the line including the start/end of the string as this information represents something. Is there a regex pattern ? or a way to split the info into columns ?
Name
A|B|C|D|E|F|G
Name col1 col2 col3 col4 col5 col6 col7
A|B|C|D|E|F|G A B C D E F G
I am using BigQuery for this and couldn't find a way to get the info of all of those. I tried the regex code which only works for the case where we have A|B|C.
I have to compare each column value and then create conditions using case when
CODE:
select
regexp_extract(name, "\\w+\\S(x|y)") as c2, -- gives either x or y
left(regexp_substr(name, "\\w+\\S\\w+\\S\\w+"),1) as c1,
right(regexp_extract(name, "\\w+\\S\\w+\\S\\w+"),1) as c3
from Table
Solution 1:[1]
Consider below approach
select * from (
select *
from your_table, unnest(split(name, '|')) value with offset
)
pivot(any_value(value) as col for offset in (0,1,2,3,4,5,6))
if applied to dummy data as in your question - output is
Solution 2:[2]
This seems like a use case for SPLIT().
select split(name,"|")[safe_offset(0)] as c1, split(name,"|")[safe_offset(1)] as c2, ..
from table
see https://cloud.google.com/bigquery/docs/reference/standard-sql/string_functions#split
Added use of safe_offset instead of offset per Array index 74 is out of bounds (overflow) google big query
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | Mikhail Berlyant |
| Solution 2 |

