From: Tom Lane <tgl@sss.pgh.pa.us>
To: Alvaro Herrera <alvherre@2ndquadrant.com>
cc: David Rowley <dgrowleyml@gmail.com>, Nagaraj Raj <nagaraj.sf@yahoo.com>,
        Pg Bugs <pgsql-bugs@postgresql.org>
Subject: Re: ERROR: insufficient columns in the PRIMARY KEY constraint
 definition
In-reply-to: <20200929223907.GA14950@alvherre.pgsql>
References: <20200929223907.GA14950@alvherre.pgsql>
Comments: In-reply-to Alvaro Herrera <alvherre@2ndquadrant.com>
	message dated "Tue, 29 Sep 2020 19:39:07 -0300"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <725316.1601419418.1@sss.pgh.pa.us>
Date: Tue, 29 Sep 2020 18:43:38 -0400
Message-ID: <725317.1601419418@sss.pgh.pa.us>
Precedence: bulk

Alvaro Herrera <alvherre@2ndquadrant.com> writes:
> On 2020-Sep-29, Tom Lane wrote:
>> Meh.  If you've read that bit you probably already understand that
>> pkeys are unique constraints.  I think the problem is with the error
>> text not the docs.

> Maybe mention PKs in parens:

> "Unique constraints (and[, by extension,] primary key constraints) on
> partitioned tables must include all the partition key columns. This
> limitation exists because PostgreSQL can only enforce uniqueness in each
> partition individually.".

I don't object to clarifying that (and that wording seems fine), but I
think fixing the error message is more important.  We'd not be having this
discussion if the OP had found that documentation.

			regards, tom lane