Check duplicate counting

gistfile1.sql

select * from (
  select count(*), d.timestamp, d.receiver_name, d.transmitter_id from detection d
  join (
    select timestamp, receiver_name, transmitter_id from detection
    where receiver_download_id = 127971115
      and duplicate = true
  ) dups

  on d.timestamp = dups.timestamp
  and d.receiver_name = dups.receiver_name
  and d.transmitter_id = dups.transmitter_id

  group by d.timestamp, d.receiver_name, d.transmitter_id
  order by d.timestamp, d.receiver_name, d.transmitter_id
) dup_counts
where count = 1

@dnahodil I believe this query confirms that duplicate counting is correct.
The algorithm is basically, for every duplicate in the particular upload, count how many detections in total match on each timestamp/receiver/transmitter combination.

There are no results for this query, as expected. Any results would indicate that for each duplicate in the upload, there is only one matching detection - i.e. not really a duplicate.