From: Tom Lane <tgl@sss.pgh.pa.us>
To: David Rowley <david.rowley@2ndquadrant.com>
cc: Tomas Vondra <tomas.vondra@2ndquadrant.com>,
        Simon Riggs <simon@2ndquadrant.com>,
        PostgreSQL-development <pgsql-hackers@postgresql.org>
Subject: Re: Performance improvement for joins where outer side is unique
In-reply-to: 
 <CAKJS1f-qN8NVT456XT6-NW+Vc_n0-jfCVtaQio0JoOWzP6F2ig@mail.gmail.com>
References: <20150320.121140.148610378.horiguchi.kyotaro@lab.ntt.co.jp>
 <CAPpHfdviphQFjk-gaGZGGxLAyt=YuVdJLnntsxOKnU296s9u-Q@mail.gmail.com>
 <CAKJS1f-=1y0-f5FFV2_r7LtS+CL2Lw-nhJjXB3O_0FmBqnK=3A@mail.gmail.com>
 <27152324.eFKJb4Khb5@dinodell> <8608754a32d1325c2f83ba8b1df71585@xs4all.nl>
 <55DA1F9F.5040204@2ndquadrant.com>
 <CAKJS1f-9UTo6fA+h8rojd30CujSKscZamLn5YnKw5+ASbV0Wwg@mail.gmail.com>
 <5085.1440375543@sss.pgh.pa.us>
 <CAKJS1f_8hrzTDCu0pwERWj_dLZu-6DYFQJAtNT+vM6UeOn0kHw@mail.gmail.com>
 <8907.1440383377@sss.pgh.pa.us>
 <CAKJS1f8JCt15ngEP8sZO-JYYT8mS6HUP_CT0T6S_0ogk1kQt+Q@mail.gmail.com>
 <CAKJS1f_4hm3qwKeHYKftd+BQuYBgdNR+rk+XP_JW9EY-ALHZbQ@mail.gmail.com>
 <56718B15.6060507@2ndquadrant.com>
 <CAKJS1f_i7dGh9ytc0DuaX3JJaN7Gp9AGXNbdtWjj0PkPzNzUeg@mail.gmail.com>
 <5671FF0D.2010403@2ndquadrant.com>
 <CANP8+j+vxfEf3dqhnzKAnDopnnUM-Zjb90gNT=haEQNQePj9Mw@mail.gmail.com>
 <CAKJS1f8MpcL4_oS+n-vmnisd_8OMuZR2vyLETOF7ABSLk_iNZg@mail.gmail.com>
 <56A25A72.2090505@2ndquadrant.com> <CAKJS1f-V7xEh6+!
 YHrCRNaZzbk7mbgPvi9MdXTCD8eJ=ROuq2Mg@mail.gmail.com>
 <24304.1457482780@sss.pgh.pa.us>
 <CAKJS1f-qN8NVT456XT6-NW+Vc_n0-jfCVtaQio0JoOWzP6F2ig@mail.gmail.com>
Comments: In-reply-to David Rowley <david.rowley@2ndquadrant.com>
	message dated "Wed, 09 Mar 2016 13:38:52 +1300"
Date: Fri, 11 Mar 2016 16:32:58 -0500
Message-ID: <13653.1457731978@sss.pgh.pa.us>
Precedence: bulk
Sender: pgsql-hackers-owner@postgresql.org

So I started re-reading this thread in preparation for looking at the
patch, and this bit in your initial message jumped out at me:

> In all of our join algorithms in the executor, if the join type is SEMI,
> we skip to the next outer row once we find a matching inner row. This is
> because we don't want to allow duplicate rows in the inner side to
> duplicate outer rows in the result set. Obviously this is required per SQL
> spec. I believe we can also skip to the next outer row in this case when
> we've managed to prove that no other row can possibly exist that matches
> the current outer row, due to a unique index or group by/distinct clause
> (for subqueries).

I wondered why, instead of inventing an extra semantics-modifying flag,
we couldn't just change the jointype to *be* JOIN_SEMI when we've
discovered that the inner side is unique.

Now of course this only works if the join type was INNER to start with.
If it was a LEFT join, you'd need an "outer semi join" jointype which
we haven't got at the moment.  But I wonder whether inventing that
jointype wouldn't let us arrive at a less messy handling of things in
the executor and EXPLAIN.  I'm not very enamored of plastering this
"match_first_tuple_only" flag on every join, in part because it doesn't
appear to have sensible semantics for other jointypes such as JOIN_RIGHT.
And I'd really be happier to see the information reflected by join type
than a new line in EXPLAIN, also.

			regards, tom lane


-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers