需要 Hive 相当于 oracle 的 regexp_extract 来提取模式之间的文本

Question

My data looks like this:

bizunit
nam-bu1-us-credit
nam-bu2-us-debit
latam-bu3-mx-debit

Now I want to extract nam & latam into a separate column called region and extract bu1 , bu2 , bu3 into a separate column called business unit and us , mx into a separate column called country .

What hive function and SQL would I use? Please share a sample hive SQL to split the above data into 3 columns to be extracted as above.

Answer 1

CREATE TABLE myTably(
  myText string
  )
INSERT INTO TABLE concat_test VALUES 
  ('nam-bu1-us-credit'), 
  ('nam-bu2-us-debit'), 
  ('latam-bu3-mx-debit');

Here's the query to extract region, business unit and country

select 
regexp_extract(myText, '([^-]+)-([^-]+)-([^-]+)-', 1),
regexp_extract(myText, '([^-]+)-([^-]+)-([^-]+)-', 2),
regexp_extract(myText, '([^-]+)-([^-]+)-([^-]+)-', 3) from MyTable