How to modify Oracle SQL query to Snowflake - oracle

Question

Using Oracle SQL, there is a function, noted below, that will allow you to create a "list" of names, phone numbers, etc., without using
multiple DUAL queries and UNION/UNION ALL to get more than one record.
The query below produces a list in this case of 10 names.
SELECT COLUMN_VALUE USERNAME
FROM TABLE(SYS.DBMS_DEBUG_VC2COLL(
'WARNER,JEFF',
'MALITO,CARL',
'MOODY,JEANNE',
'PHILLIPS,HUGH & KELLY',
'PATSANTARAS,VICTORIA',
'BROWN,ROLAND',
'RADOSEVICH,MIKE',
'RIDER,JACK',
'MACLEOD,LENARD',
'SCOTT,DAN' ))
However, when trying to run this same query in Snowflake, it will not work.
I receive this error: SQL compilation error: Invalid identifier SYS.DBMS_DEBUG_VC2COLL
Is there a "Snowflake version" of this query that can be used?

Answer 1

Here are some options, you can see which works best for you.
This works if you can get your SQL to look similar:
SELECT $1::VARCHAR AS column_value
FROM (VALUES ('WARNER,JEFF'), ('MACLEOD,LENARD'), ('SCOTT,DAN'));
This also works if you can get your list to be in a single string, delimited by a pipe or similar:
SELECT value::VARCHAR AS column_value
FROM LATERAL FLATTEN(INPUT=>SPLIT('WARNER,JEFF|MACLEOD,LENARD|SCOTT,DAN', '|'));
If you have the strings in the format 'a','b' and find it painful to do one of the above, I'd do something like this:
SELECT value::VARCHAR AS column_value
FROM LATERAL FLATTEN(INPUT=>SPLIT(ARRAY_TO_STRING(ARRAY_CONSTRUCT('WARNER,JEFF', 'MALITO,CARL', 'MOODY,JEANNE'), '|'), '|'));
Similar to the above suggestions, you can try this:
SELECT VALUE::VARCHAR as column_name
FROM TABLE(FLATTEN(INPUT => ARRAY_CONSTRUCT('WARNER,JEFF', 'MALITO,CARL', 'MOODY,JEANNE'), MODE => 'array'));