From: Tom Lane <tgl@sss.pgh.pa.us>
To: Stefan Fehrenbach <stefan.fehrenbach@gmail.com>
cc: pgsql-performance@lists.postgresql.org
Subject: Re: When can joins be avoided?
In-reply-to: 
 <CAB9vtGACGyRjg132fu52rTRUu_m1Xz0ezPNdeba76DbC_Nsy3g@mail.gmail.com>
References: 
 <CAB9vtGACGyRjg132fu52rTRUu_m1Xz0ezPNdeba76DbC_Nsy3g@mail.gmail.com>
Comments: In-reply-to Stefan Fehrenbach <stefan.fehrenbach@gmail.com>
	message dated "Wed, 16 Nov 2022 11:39:13 +0100"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <1826809.1668696560.1@sss.pgh.pa.us>
Date: Thu, 17 Nov 2022 09:49:20 -0500
Message-ID: <1826810.1668696560@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/1826810.1668696560%40sss.pgh.pa.us>
Precedence: bulk

Stefan Fehrenbach <stefan.fehrenbach@gmail.com> writes:
> I'm interested in the rules that Postgres uses to avoid joins. Are
> these documented somewhere? If I had to look at the source code, where
> would I start?

src/backend/optimizer/plan/analyzejoins.c

> They don't seem to match my intuition about which joins
> could be avoided.

I believe only left joins to single tables can be elided ATM.
It's too hard to prove uniqueness of the join key in more-
complicated cases.

			regards, tom lane