From: KES <kes-kes@yandex.ru>
Envelope-From: kes-kes@yandex.com
To: Tom Lane <tgl@sss.pgh.pa.us>,
	Bruce Momjian <bruce@momjian.us>
Cc: PostgreSQL-development <pgsql-hackers@postgresql.org>
In-Reply-To: <25606.1533843069@sss.pgh.pa.us>
References: <153121527691.1408.5686988620817799073@wrigleys.postgresql.org>
 <20180807173659.GD7297@momjian.us>
 <31616681533725753@sas1-d856b3d759c7.qloud-c.yandex.net>
 <20180808130040.GA2611@momjian.us> <12107.1533736288@sss.pgh.pa.us>
 <37593701533809465@sas1-87f9feb8d943.qloud-c.yandex.net>
 <20180809190913.GB14011@momjian.us> <25606.1533843069@sss.pgh.pa.us>
Subject: Re: Typo in doc or wrong EXCLUDE implementation
MIME-Version: 1.0
Date: Fri, 10 Aug 2018 12:10:47 +0300
Message-Id: <43021481533892247@sas1-ea1d14049a51.qloud-c.yandex.net>
Content-Transfer-Encoding: 8bit
Content-Type: text/plain; charset=utf-8
Precedence: bulk

huh, maybe you are right, I missread that. English is not my native language.
Actually I come there from FK constraints. 

Would it be sufficient for FK require not UNIQUEs, but **allow** "EXCLUDE with operators that act like equality"?

09.08.2018, 22:31, "Tom Lane" <tgl@sss.pgh.pa.us>:
> Bruce Momjian <bruce@momjian.us> writes:
>>  On Thu, Aug 9, 2018 at 01:11:05PM +0300, KES wrote:
>>>  Why surprising? It is
>>>  [documented](https://www.postgresql.org/docs/current/static/sql-create
>>>  table.html#sql-createtable-exclude):
>>>>  If all of the specified operators test for equality, this is
>>>>  equivalent to a UNIQUE constraint, although an ordinary unique
>>>>  constraint will be faster.
>
>>>  Thus the UNIQUE constraint is just particular case of exclusion
>>>  constraint, is not?
>
>>  Well, for me a UNIQUE constraint guarantees each discrete value is
>>  unique, while exclusion constraint says discrete or ranges or geometric
>>  types don't overlap. I realize equality is a special case of discrete,
>>  but having such cases be marked as UNIQUE seems too confusing.
>
> I think the OP is reading "equivalent" literally, as meaning that
> an EXCLUDE with operators that act like equality is treated as being
> the same as UNIQUE for *every* purpose. We're not going there, IMO,
> so probably we need to tweak the doc wording a little. Perhaps
> writing "functionally equivalent" would be better? Or instead of
> "is equivalent to", write "imposes the same restriction as"?
>
>                         regards, tom lane