Regular Expression Match between occurrence of character

Question

I have the following string:

3#White House, District Of Columbia, United States#US#USDC#DC001#38.8951#-77.0364#531871#382

as you can see, the s

Answer 1

My use-case resembles a simple SPLIT(string,"#") operation but regex gives me a bit more flexibility

Obviously REGEXP_EXTRACT() is the way to go here - but wanted to throw different option to show flexibility in using split too - just one of an option

#standardSQL
WITH `project.dataset.table` AS (
  SELECT '3#White House, District Of Columbia, United States#US#USDC#DC001#38.8951#-77.0364#531871#382' locations
)
SELECT 
  REGEXP_EXTRACT(locations, r'^(?:[^#]*#){2}([^#]*(?:#[^#]*){3})') value_via_regexp,
  (SELECT STRING_AGG(part, '#' ORDER BY pos) FROM UNNEST(SPLIT(locations, '#')) part WITH OFFSET pos WHERE pos BETWEEN 2 AND 5) value_via_split_unnest
FROM `project.dataset.table`      

with result as

Row     value_via_regexp            value_via_split_unnest   
1       US#USDC#DC001#38.8951       US#USDC#DC001#38.8951