Message-ID: <86a93a08-8056-9c9c-7311-019b08458450@postgresfriends.org>
Date: Mon, 19 Dec 2022 17:53:46 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.4.2
Subject: Re: GROUP BY ALL
To: Andrey Borodin <amborodin86@gmail.com>,
 pgsql-hackers <pgsql-hackers@postgresql.org>
References: 
 <CAAhFRxjyTO5BHn9y1oOSEp0TtpTDTTTb7HJBNhTG+i3-hXC0XQ@mail.gmail.com>
From: Vik Fearing <vik@postgresfriends.org>
Content-Language: en-US
In-Reply-To: 
 <CAAhFRxjyTO5BHn9y1oOSEp0TtpTDTTTb7HJBNhTG+i3-hXC0XQ@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: 
 <https://www.postgresql.org/message-id/86a93a08-8056-9c9c-7311-019b08458450%40postgresfriends.org>
Precedence: bulk

On 12/19/22 05:19, Andrey Borodin wrote:
> Hi hackers!
> 
> I saw a thread in a social network[0] about GROUP BY ALL. The idea seems useful.
> I always was writing something like
>      select datname, usename, count(*) from pg_stat_activity group by 1,2;
> and then rewriting to
>      select datname, usename, query, count(*) from pg_stat_activity group by 1,2;
> and then "aaahhhh, add a number at the end".
> 
> With the proposed feature I can write just
>      select datname, usename, count(*) from pg_stat_activity group by all;


We already have GROUP BY ALL, but it doesn't do this.


> PFA very dummy implementation just for a discussion. I think we can
> add all non-aggregating targets.
> 
> What do you think?


I think this is a pretty terrible idea.  If we want that kind of 
behavior, we should just allow the GROUP BY to be omitted since without 
grouping sets, it is kind of redundant anyway.

I don't know what my opinion is on that.
-- 
Vik Fearing