Distinct Over multiple columns in JPA

Question

I have a courier_routes table which has two foreign keys to a locations table. The columns in the courier_routes table are start_location_id and end_location_id. (I'm using Oracle)

I want to find the all the distinct locations that the courier has been to in the courier_routes table.

Consider this example

|id |start_location_id| end_location_id|...
|1|1|2|
|2|1|3|
|3|4|5|
|4|2|1|

I want to get back ids 1 2 3 4 5

My SQL that works is

SELECT loc FROM (
SELECT start_location_id AS loc FROM courier_routes 
UNION
SELECT end_location_id AS loc FROM courier_routes);

I want to translate this into JPA. Hibernate does not support unions so I thought about doing sub queries. The problem with this is that it could affect performance.

I was thinking of something like (pseudo code...)

SELECT id FROM locations where id in (
    (SELECT start_location_id FROM courier_routes) 
       OR
    (SELECT end_location_id FROM courier_routes));

Is there an easier way? This table will get big so I really dont want to use sub-queries.

Should I just do a union in java by adding two separate query results into a set?

Thanks

Answer 1

You can do a join on the two columns in HQL :

SELECT distinct loc.id FROM locations loc, courier_routes c 
   where loc.id=c.start_location_id OR c.id=c.end_location_id

Hope this helps!