Re: [PORTS] Port Bug Report: Result not "GROUPED BY" from SQL: select im,bn,count(adr) FROM logtmp GROUP BY im,bn;

From: "Vadim B(dot) Mikheev" <vadim(at)sable(dot)krasnoyarsk(dot)su>
To: Arkadiy Vavilin <ar(at)bizlink(dot)ru>
Cc: pgsql-hackers(at)postgreSQL(dot)org
Subject: Re: [PORTS] Port Bug Report: Result not "GROUPED BY" from SQL: select im,bn,count(adr) FROM logtmp GROUP BY im,bn;
Date: 1998-01-27 08:40:58
Message-ID: 34CD9D9A.9AB9E39A@sable.krasnoyarsk.su
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Unprivileged user wrote:
>
> Category : runtime: back-end: SQL
> Severity : critical
>
> Summary: Result not "GROUPED BY" from SQL:
> select im,bn,count(adr) FROM logtmp GROUP BY im,bn;

I found from where it comes: nodeGroup.c' funcs assume that slots from
subplan (node Sort) have ttc_shouldFree = FALSE. This was right before
6.2 and is right for in-memory sorts, but wrong for disk ones.
Node Mergejoin uses some tricks to deal with this - #define MarkInnerTuple...
At the moment, I haven't time to fix this bug - will do this latter
if no one else.

Vadim

Browse pgsql-hackers by date

  From Date Subject
Next Message Brett McCormick 1998-01-27 08:44:50 weird query plans involving hash [join] & no indicies
Previous Message Goran Thyni 1998-01-27 08:39:42 Re: [HACKERS] postmaster crash and .s.pgsql file