为 Google Big Query 替换正则表达式

Question

I have some data for which i want to select the LAST VALUE BEFORE THE DELIMITER

Example- A -> B -> C -> D

In this case, i want to select "C" so i used a positive look ahead but BQ doesnt allow look aheads/behinds

Here is the Regex, (?=[^\n>] -[^\n-] $)[^\n-]*

Can someone help me replace the look ahead?

Answer 1

You don't need any lookarounds and can use a simple regex like following and extract your needed contents from group1.

^.*->(.*)->.*$

Explanation:

• ^ - Matches start of text
• .*-> - Matches any text greedily and second last delimiter
• (.*) -> Captures the text between second last and last delimiter in group1
• -> -> Matches last delimiter
• .*$ -> Matches some text optionally after last delimiter and end of line

Check this demo

Answer 2

Consider below options (using regex and split)

select col, 
  regexp_extract(col, r'^.*->(.*)->.*$') as extract_with_regex, 
  array_reverse(split(col, ' -> '))[offset(1)] as extract_with_split
from your_table

if applied to sample data in your question - output is