postgres - retain duplicates from query

Question

I just ran into a bug in our code based on the way we're matching data against a Postgres array.

Here's the issue:

In the database, we have a surcharge column that lives in the appointment table where the ids of all surcharges attached to an appointment are stored. In some cases, a surcharge may be added twice and therefore would look like {1,1} in the database.

I was using a simple query like the following to retrieve the surcharge data associated with the surcharges that have been attached to the appointment:

SELECT s.* FROM surcharge s, appointment a WHERE s.id = ANY(a.surcharges);

Unfortunately, I didn't test this case where more than one of the same surcharge exists. The result is that only 1 record is being returned rather than the 2 I really need.

To clarify, an example would be an appointment with appointment.surcharges = {1,1} , I'd actually want to be able to retrieve the record from surcharges TWICE because it has two surcharges associated with it. I then use those records to create billing-related transactions associated with the appointment.

All that said, is there a simple way to allow a query to return the duplicates that I need?

I can handle this situation in code, but it would be great to run a simple query to handle this situation.

I'm running Postgres 9.5.

Thank you in advance!

Answer 1

I think a plain inner join to an unnested table of surcharges should give you what you want:

SELECT s.*
FROM surcharge s
INNER JOIN
(
    SELECT UNNEST(surcharges) AS surcharges
    FROM appointment
) a
    ON s.id = a.surcharges;