I have a dataset in which one column is Branch-ID and other one is Branch Manager and it looks as follows in the given url. dataset
I want to combine the branch managers into one single column based on the branch-id. For example if Bob and Sandra are two different branch-managers but have the same branch id which is branch-id=1, then we should concatenate them together as Bob-Sandra and place them in a separately created column.
I have attached the expected output for the above dataset. expected_output_dataset
I am currently using Oracle Analytics Cloud Professional Version.
I don't know Oracle Analytics, but - if it has anything to do with an Oracle database and its capabilities, then listagg
helps.
Sample data in lines #1 - 10; query you might be interested in begins at line #11.
SQL> with test (account_id, branch_id, branch_manager) as
2 (select 1, 123, 'Sandra' from dual union all
3 select 3, 124, 'Martha' from dual union all
4 select 4, 125, 'John' from dual union all
5 select 6, 126, 'Andrew' from dual union all
6 select 7, 126, 'Mathew' from dual union all
7 select 2, 123, 'Michael' from dual union all
8 select 5, 125, 'David' from dual union all
9 select 8, 126, 'Mark' from dual
10 )
11 select a.account_id, a.branch_id, a.branch_manager,
12 b.concatenated_column
13 from test a join (select branch_id,
14 listagg(branch_manager, '-') within group (order by null) concatenated_column
15 from test
16 group by branch_id
17 ) b on b.branch_id = a.branch_id;
ACCOUNT_ID BRANCH_ID BRANCH_ CONCATENATED_COLUMN
---------- ---------- ------- -------------------------
1 123 Sandra Michael-Sandra
3 124 Martha Martha
4 125 John David-John
6 126 Andrew Andrew-Mark-Mathew
7 126 Mathew Andrew-Mark-Mathew
2 123 Michael Michael-Sandra
5 125 David David-John
8 126 Mark Andrew-Mark-Mathew
8 rows selected.
SQL>