From: Tom Lane <tgl@sss.pgh.pa.us>
To: John Lumby <johnlumby@hotmail.com>
cc: "pgsql-docs@lists.postgresql.org" <pgsql-docs@lists.postgresql.org>,
        "David G. Johnston" <david.g.johnston@gmail.com>
Subject: Re: description of Aggregate Expressions
In-reply-to: 
 <DM6PR06MB55621D8E8B9C786DAA6BA077A35F0@DM6PR06MB5562.namprd06.prod.outlook.com>
References: 
 <2A91BEF8171A5349931391E0C721CC5375705457@CPEMS-KPN501.KPNCNL.LOCAL>
 <251962385fdc49da0da54e22588aabd6fbc2e0fa.camel@osdev.nl>
 <2A91BEF8171A5349931391E0C721CC53757054D3@CPEMS-KPN501.KPNCNL.LOCAL>
 <2CB4B3C9-F43F-49D8-A594-93DBE99EF514@elevated-dev.com>
 <7308f0f3a1674a39bada8571189a3735@CPEMS-KPN509.KPNCNL.LOCAL>
 <91CEBEAA-7260-4503-832F-E5B27B91BE1F@elevated-dev.com>
 <2A91BEF8171A5349931391E0C721CC5375705528@CPEMS-KPN501.KPNCNL.LOCAL>
 <E813B7F6-B641-4739-A209-F0352A780EF8@elevated-dev.com>
 <2A91BEF8171A5349931391E0C721CC53757055C3@CPEMS-KPN501.KPNCNL.LOCAL>
 <DM6PR06MB55622DFEE2755236B8A810A0A35C0@DM6PR06MB5562.namprd06.prod.outlook.com>
 <CAKFQuwZk3Kt4uNktMgLYj5B1Rzrm1R4EMVZQRka189AhnPkM=Q@mail.gmail.com>
 <DM6PR06MB55621D8E8B9C786DAA6BA077A35F0@DM6PR06MB5562.namprd06.prod.outlook.com>
Comments: In-reply-to John Lumby <johnlumby@hotmail.com>
	message dated "Fri, 06 Dec 2019 18:43:39 +0000"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <381.1575659855.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Fri, 06 Dec 2019 14:17:35 -0500
Message-ID: <382.1575659855@sss.pgh.pa.us>
Precedence: bulk

John Lumby <johnlumby@hotmail.com> writes:
> On 12/05/19 18:06, David G. Johnston wrote:
>> Then I would have to say the proposed solution to this edge case is wor=
se than the problem.  I also don't expect there to be a clean solution to =
dealing with the complexities of expressions at the syntax diagram level.

> Yes,  I see what I suggested is not ideal either.     But I think someth=
ing needs to be changed.

> How about replacing "expression [ , ... ]"  by "parameter_list" in the d=
escription, and then stating that parameter_list can be either a single ex=
pression or ,   if the particular aggregate function accepts it (for which=
,   consult that function's reference),    a comma-separated list of expre=
ssions.

That's just as wrong.  As we tried to explain before, the
parenthesized-list syntax is a row constructor, so it only works
in cases where the aggregate function can accept a composite
argument type.  Most don't.

Moreover, the very same thing holds in *any* expression context,
not only aggregate arguments.  So if we took this seriously there
would have to be a lot of other places plastered with equally
confusing/unhelpful addenda.

			regards, tom lane