Horton

Re: Wrong aggregate result when sorting by a NULL value

Andres Freund <andres@anarazel.de>

View thread View on postgr.es
From: Andres Freund <andres@anarazel.de>
To: Andrew Gierth <andrew@tao11.riddles.org.uk>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Ondřej Bouda <obouda@email.cz>, pgsql-bugs@postgresql.org
Date: 2018-11-03T21:50:09Z
Lists: pgsql-bugs 
Hi,

On 2018-11-02 22:57:00 -0700, Andres Freund wrote:
> On 2018-11-02 17:34:43 +0000, Andrew Gierth wrote:
> > >>>>> "Andres" == Andres Freund <andres@anarazel.de> writes:
> > 
> >  >> Andres, do you have time to look at this right now?
> > 
> >  Andres> Thanks for bisecting.  I'll take a look later today.
> > 
> > Looks like this:
> > 
> > +           scratch.d.agg_strict_input_check.nargs = numInputs;
> > 
> > should have been pertrans->numTransInputs instead?
> 
> That looks like it's precisely the reason. I'll push something tomorrow
> PST morning, crediting both you (diagnosis) and Tom (bisecting)
> obviously, unless you prefer to do so yourself.
> 
> It's a bit sad that we don't have any tests that test this :/. I'll add
> something for the specific case, but that'll obviously not be
> exhaustive.

And pushed. Thanks Ondřej for the report, thanks Tom & Andrew for
identifying the issue.

Greetings,

Andres Freund

Commits

  1. Fix unused-variable warning.

  2. Prevent generating EEOP_AGG_STRICT_INPUT_CHECK operations when nargs == 0.

  3. Fix STRICT check for strict aggregates with NULL ORDER BY columns.

  4. Expression evaluation based aggregate transition invocation.