From: Tom Lane <tgl@sss.pgh.pa.us>
To: Greg Hennessy <greg.hennessy@gmail.com>
cc: pgsql-general@lists.postgresql.org
Subject: Re: help debugging an issue with selectivity
In-reply-to: <8e4f0d76-9e77-4819-96f7-cbc64a9d2b5a@gmail.com>
References: <CA+FnnTxGOqF1DN4e70KXRJakuPV_V7_vGonQTX4eEOEE92PCKw@mail.gmail.com> <CAKFQuwakY0c3x=2o=gQ6Cr7jQ=DgE_snL1cHf8Oj+irO1fmjEQ@mail.gmail.com> <8e4f0d76-9e77-4819-96f7-cbc64a9d2b5a@gmail.com>
Comments: In-reply-to Greg Hennessy <greg.hennessy@gmail.com>
	message dated "Tue, 17 Mar 2026 13:29:39 -0400"
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-ID: <602083.1773801455.1@sss.pgh.pa.us>
Content-Transfer-Encoding: 8bit
Date: Tue, 17 Mar 2026 22:37:35 -0400
Message-ID: <602084.1773801455@sss.pgh.pa.us>
Archived-At: <https://www.postgresql.org/message-id/602084.1773801455%40sss.pgh.pa.us>
Precedence: bulk

Greg Hennessy <greg.hennessy@gmail.com> writes:
> I am not sure if this belongs in pgsql-general or pgsql-hackers, I am 
> trying first in psgl-general.

Doesn't matter a lot, but as Rob noted, you should not hijack an
existing thread.  Start your own thread, don't reply to an unrelated
message.

> When I actually execute this, I get output to the terminal of the form:
> WARNING:  join_selectivity: operator id 16818 jointype 0 0
> WARNING:  HERE0 pgq3c_seljoin.... 5.873266e-12
> WARNING:  datum result 4438812783922730423 0.000000
> WARNING:  HERE0 pgq3c_seljoin.... 5.873266e-12
> WARNING:  join_selectivity: 0.000000 16818 jointype 0
> WARNING:  join_selectivity: s1 0.000000
> WARNING:  clause_selectivity: s1 0.000000

> where it seems to me the q3c code is returning a non zero value, but in 
> the guts of postgres what is found is a zero value.

I don't see any discrepancy in your results, only in your choices of
printf specifiers.  "%f" defaults to "%.6f", that is, print only six
digits after the decimal point.  So a value down around 1e-12 is
going to print as all zeroes.

			regards, tom lane