From: Tom Lane <tgl@sss.pgh.pa.us>
To: Tomas Vondra <tomas.vondra@2ndquadrant.com>
cc: David Rowley <david.rowley@2ndquadrant.com>,
 Sergei Kornilov <sk@zsrv.org>,
        Andres Freund <andres@anarazel.de>,
        Robert Haas <robertmhaas@gmail.com>,
        David Steele <david@pgmasters.net>,
        PostgreSQL mailing lists <pgsql-hackers@postgresql.org>,
        Alvaro Herrera <alvherre@alvh.no-ip.org>,
        Stephen Frost <sfrost@snowman.net>,
        Ildar Musin <i.musin@postgrespro.ru>
Subject: Re: using index or check in ALTER TABLE SET NOT NULL
In-reply-to: <20190802195438.4vuotpz6pdq6r6jg@development>
References: <20190501162021.q7r5cys6uwdb5bsp@alap3.anarazel.de>
 <25570.1556728681@sss.pgh.pa.us>
 <20481181556734692@sas2-80c42a3bfdfe.qloud-c.yandex.net>
 <CAKJS1f9Gaor9+wGWFwaSrm1=N5cPHiM=LGWcCDeuPymC5dkyhw@mail.gmail.com>
 <19795.1556759294@sss.pgh.pa.us>
 <CAKJS1f9dYoeqqOwXW4wqD8tVGNki2AjzOQy=Xo=a_nNAMdYD=g@mail.gmail.com>
 <CAKJS1f-Y2CAV7y_Vay_bmE=oyZPgtm2qXRinozYOJ0SeopjYtw@mail.gmail.com>
 <CAKJS1f_3qanbQSJ8ndkkxCAuHuByJLXR6zuQetOmnmBzWbHg3Q@mail.gmail.com>
 <10303161560180908@sas1-fb8a605c4548.qloud-c.yandex.net>
 <CAKJS1f-MBwNBQWn=pMrhuk=k=9Hmw8BPAC5ryuX0fXdNMqeiFg@mail.gmail.com>
 <20190802195438.4vuotpz6pdq6r6jg@development>
Comments: In-reply-to Tomas Vondra <tomas.vondra@2ndquadrant.com>
	message dated "Fri, 02 Aug 2019 21:54:38 +0200"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <24687.1564776140.1@sss.pgh.pa.us>
Date: Fri, 02 Aug 2019 16:02:20 -0400
Message-ID: <24688.1564776140@sss.pgh.pa.us>
Precedence: bulk

Tomas Vondra <tomas.vondra@2ndquadrant.com> writes:
> I think there's a consensus to change INFO to DEBUG1 in pg12, and then
> maybe imlpement something like VERBOSE mode in the future. Objections?

ISTM the consensus is "we'd rather reduce the verbosity, but we don't
want to give up test coverage".  So what's blocking this is lack of
a patch to show that there's another way to verify what code path
was taken.

> As for the reduction of test coverage, can't we deduce whether a
> constraint was used from data in pg_stats or something like that?

Not sure how exactly ... and we've already learned that pg_stats
isn't too reliable.

			regards, tom lane