Problem with a GROUP BY clause

Hi,

I am running Postgresql ver 6.5.1 on an SGI (Irix).

I am trying to write a query to return some top scoring items, using a
GROUP BY clause, however, I am having trouble with "empty" results.  For
example:

htg=> SELECT LocusID, Score FROM Match
htg-> WHERE SeqID=4358;
locusid|score
-------+-----
(0 rows)

In the case of SeqID=4358, there are no matching LocusIDs, so 0 rows are
returned. This is fine. There may be cases, however, when there are many
matching LocusIDs, so I use a GROUP BY clause. This usually works, but in cases
like the above one, where there are no matching LocusIDs, it returns the wrong
result. Ie:

htg=> SELECT LocusID, max(Score) FROM Match
htg-> WHERE SeqID=4358
htg-> GROUP BY LocusID;
locusid|max
-------+---
       |
(1 row)

There is no row, yet Postgresql reports there is "one" which it returns as
empty.  This also makes the Perl DBI::DBD interface unhappy. It seems to return
an undefined value which can't even be tested with "if(! defined $value)"!

Unless I really misunderstand "GROUP BY", I think this is a bug.

Mark

--
Mark Dalphin                          email: mdalphin@amgen.com
Mail Stop: 29-2-A                     phone: +1-805-447-4951 (work)
One Amgen Center Drive                       +1-805-375-0680 (home)
Thousand Oaks, CA 91320                 fax: +1-805-499-9955 (work)