Remove text from a column

MB_Dem

Hi all, I'm trying to figure out the easiest way to create a new column with slightly filtered strings.

My "Team" column contains strings such as:

• Team Name A - EMEA
• Team Name A - AMERICAS
• Team Name A - APAC
• Team Name B - EMEA
• Team Name B - AMERICAS
• Team Name B - GLOBAL

I want to create a new column, "Simplified Team" that contains the same text, but with the EMEA, AMERICAS, APAC, GLOBAL parts of the string removed.

Note: Not all teams have the region appended to them.

GrantSmith

You can use a formula tile in Magic ETL to do a regular expression replacement

REGEXP_REPLACE(`Team`, '( - EMEA)|( - AMERICAS)|( - GLOBAL)|( - APAC)', '')

Alternatively you can chain REPLACE functions together in a Beast Mode:

REPLACE(REPLACE(REPLACE(REPLACE(`Team`, ' - EMEA', ''), '- AMERICAS', ''), ' - GLOBAL', ''), '- APAC', '')

GrantSmith

If your format is split by a hyphen you could use SPLIT_PART to get the first part of your string:

SPLIT_PART(`Team`, ' - ', 1)

MB_Dem

That's getting me closer, but the team names have a bazillion hyphens in them. 😥 E.g. "Engineering - Controls - Americas" or "Marketing - Communication - Global" or "Super long team name - with more hyphens - than a team name needs - Americas"

GrantSmith

So you just want to drop the last section after the last hyphen?

MB_Dem

Ish. Essentially, I want a new "Simplified Team" column that copies "Team" and then either truncates the following strings, or does a search and replace of the strings below and leaves nothing behind.

• " - EMEA"
• " - AMERICAS"
• " - GLOBAL"
• " - APAC"

GrantSmith

You can use a formula tile in Magic ETL to do a regular expression replacement

REGEXP_REPLACE(`Team`, '( - EMEA)|( - AMERICAS)|( - GLOBAL)|( - APAC)', '')

Alternatively you can chain REPLACE functions together in a Beast Mode:

REPLACE(REPLACE(REPLACE(REPLACE(`Team`, ' - EMEA', ''), '- AMERICAS', ''), ' - GLOBAL', ''), '- APAC', '')