ORACLE / Select the records that only has the value X on Column_B

Question

I have the following scenario:

Table1

ID  Column_B
10001   X
10001   Y
10002   X
10003   X
10003   Y
10004   X
10005   X
10005   Y

I need to do a query that shows the IDs that has only X on Column_B, so the result would be:

10002   X
10004   X

If I run the select below, it shows also the records that has Y on Column_B:

select ID, Column_B from Table1 where Column_B = 'X'

ID  Column B
10001   X
10002   X
10003   X
10004   X
10005   X

I couldn't find anything similar to this case, anywhere. Losing my mind here trying to figure it out. It seems to be so simple. Damn.

Answer 1

You can use aggregation:

select id
from t
group by id
having min(column_b) = max(column_b) and min(column_b) = 'X';

Answer 2

Use relational algebra - in this case, MINUS :

SELECT ID
  FROM TABLE1
  WHERE COLUMN_B = 'X'
MINUS
  SELECT ID
    FROM TABLE1
    WHERE COLUMN_B = 'Y';

You start with a list of all the ID's which are associated with 'X', and subtract off the list of ID's which are associated with 'Y'.

SQLFiddle here

Best of luck.