From: Sergei Kornilov <sk@zsrv.org>
To: Peter Eisentraut <peter.eisentraut@2ndquadrant.com>,
	Tom Lane <tgl@sss.pgh.pa.us>,
	Stephen Frost <sfrost@snowman.net>
Cc: Robert Haas <robertmhaas@gmail.com>,
	"pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
In-Reply-To: <b74455c1-f89e-02e3-471a-8fdd9275fab9@2ndquadrant.com>
References: <81911511895540@web58j.yandex.ru>
	 <CA+TgmoZKR4dTR7KyoM4jfJN2J-eNoBRG2_XHdZa1f0VUbK1xXQ@mail.gmail.com>
	 <20171129152135.GN4628@tamriel.snowman.net> <9878.1511970441@sss.pgh.pa.us>
	 <370481511970756@web3j.yandex.ru>
 <b74455c1-f89e-02e3-471a-8fdd9275fab9@2ndquadrant.com>
Subject: Re: using index or check in ALTER TABLE SET NOT NULL
MIME-Version: 1.0
Message-Id: <3796271522999680@web16g.yandex.ru>
Date: Fri, 06 Apr 2018 10:28:00 +0300
Content-Transfer-Encoding: 8bit
Content-Type: text/plain; charset=utf-8
Precedence: bulk

Hello, Peter!
I agree my patch seems as strange workaround. And better will be SET NOT NULL NOT VALID feature. But this change is very complicated for me
For now my patch is small, simple and provides way for users.

regards, Sergei

06.04.2018, 06:29, "Peter Eisentraut" <peter.eisentraut@2ndquadrant.com>:
> On 11/29/17 10:52, Sergei Kornilov wrote:
>>  My target problem of adding NOT NULL to big relation without long downtime can be done with ADD CONSTRAINT NOT VALID, VALIDATE it in second transaction, then SET NOT NULL by my patch and drop unneeded constraint.
>
> It seems to me that this is a workaround for not properly cataloguing
> NOT NULL constraints. If we fixed that, you could do SET NOT NULL NOT
> VALID and go from there. Maybe we should look again into fixing that.
> That would solve so many problems.
>
> --
> Peter Eisentraut http://www.2ndQuadrant.com/
> PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services