From: Sergei Kornilov <sk@zsrv.org>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Alvaro Herrera <alvherre@alvh.no-ip.org>,
	Ildar Musin <i.musin@postgrespro.ru>,
	Stephen Frost <sfrost@snowman.net>,
	Robert Haas <robertmhaas@gmail.com>,
	"pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
In-Reply-To: <1142.1520362313@sss.pgh.pa.us>
References: <880511520351578@web5j.yandex.ru>
 <20180306172528.xnelm3xcq2qeyd2g@alvherre.pgsql>
 <1715451520361128@web26o.yandex.ru> <1142.1520362313@sss.pgh.pa.us>
Subject: Re: using index or check in ALTER TABLE SET NOT NULL
MIME-Version: 1.0
Message-Id: <2091981520364460@web24g.yandex.ru>
Date: Tue, 06 Mar 2018 22:27:40 +0300
Content-Transfer-Encoding: 7bit
Content-Type: text/plain
Precedence: bulk


Hello

> Do you actually need test output proving that this code path was taken
> rather than the default one? Seems like looking at the code coverage
> report might be enough.
I not known. In v4 i use DEBUG1 message and do not check code path in tests at all: by full table scan or by constraint, i tested only command result to not break behavior.
Today Ildar Musin proposed to test code path through NotNullImpliedByRelConstraints function. This is my first patch and I do not have the confidence to write a test. So I looked more closely at the alter table tests, found several info in attach partition and updated my patch.

> I did not see any INFO messages in a quick test of ALTER TABLE ATTACH
> PARTITION, but if there are any lurking in there, they probably need
> to be downgraded.
In src/test/regress/expected/alter_table.out i found 7 test with
> INFO: partition constraint for table "..." is implied by existing constraints
and 5 with
> INFO:  updated partition constraint for default partition "..." is implied by existing constraints
ereport's are in ValidatePartitionConstraints function src/backend/commands/tablecmds.c

regards, Sergei