Find duplicates of one column but only return results if another column has same value

Question

I want to find the values of a given column that repeat (duplicate), but only return the repeats if another column in the row is the same.

Perhaps an example would be clearer. Given a table called filters, I have 3 columns (id, name, filter_type):

DESCRIBE filters;
+-------------------+--------------+------+-----+---------+----------------+
| Field             | Type         | Null | Key | Default | Extra          |
+-------------------+--------------+------+-----+---------+----------------+
| id                | int(11)      | NO   | PRI | NULL    | auto_increment |
| name              | varchar(255) | YES  |     | NULL    |                |
| filter_type       | varchar(255) | YES  |     | NULL    |                |
| created_at        | datetime     | YES  |     | NULL    |                |
| updated_at        | datetime     | YES  |     | NULL    |                |
+-------------------+--------------+------+-----+---------+----------------+

Now I want to select all filter_types that repeat. Easy enough:

SELECT COUNT(*) as count, filter_type 
FROM filters 
GROUP BY filter_type_type HAVING count > 1; 

+-------+-----------------+
| count | filter_type     |
+-------+-----------------+
|     5 | Contact         |
+-------+-----------------+

But I want to only return the repeat if the name column values are the SAME for all of the group. So in the result above, we have repeat 'Contact'. It repeats 5 times. But I only want to return this result if all 5 records have the same name column value.

SELECT name FROM filters WHERE filter_type = 'Contact';
+---------------------------+
| name                      |
+---------------------------+
| All                       |
| All                       |
| New                       |
| All                       |
| New                       |
+---------------------------+

Since the name values are not the same for the 5 records, I do not want to return the 'Contact' filter_type. But if they were all the same, then I would want to return it.

I was thinking to use subqueries, but not sure how to connect them. How could I do this?

Answer 1

You can use aggregation for this, along with a having clause:

select filter_type
from filters
group by filter_type
having count(*) > 1 and
       min(name) = max(name)