Re: Partial Mode in Aggregate Functions

From: David Rowley <dgrowleyml(at)gmail(dot)com>
To: Marcos Pegoraro <marcos(at)f10(dot)com(dot)br>
Cc: "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Partial Mode in Aggregate Functions
Date: 2026-02-26 13:03:19
Message-ID: CAApHDvoez7-dfCEt+w8QXSHxTXgDewAumVAw_jjZ2NHBEDRF6g@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, 27 Feb 2026 at 00:44, Marcos Pegoraro <marcos(at)f10(dot)com(dot)br> wrote:
>
> Em qui., 26 de fev. de 2026 às 01:57, David Rowley <dgrowleyml(at)gmail(dot)com> escreveu:
>>
>> As for the proposed patch, I'm strongly against it. If people are
>> confused about what partial aggregation is, then let's modify the
>> documentation to explain what it means.
>
> Well, I started this because the way it's written is confusing, so much so that I showed a link [1] from someone who speaks English better than me saying that PARTIAL is FILTER.
> And from the user point of view, who is reading just that page, partial means incomplete, a subset of, etc. So I think it's easy for the reader to misunderstand. Whether Parallel Mode is the best term to use here, I don't know, but I'm convinced that the way it is written is easy to confuse.

Does the wording I proposed make it easier to understand why the word
"partial" is in the name?

> [1] - https://www.cybertec-postgresql.com/en/partial-aggregation-the-beautiful-way/

I've never heard anyone using "partial" for FILTER before. The blog is
from 2015, and partial aggregates got committed in 2016. I hope nobody
would use the partial terminology for FILTER today, as we now have
something else using that name.

David

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Srirama Kucherlapati 2026-02-26 13:15:35 RE: AIX support
Previous Message Matt Blewitt 2026-02-26 12:57:24 [PATCH] Fix PITR pause bypass when initial XLOG_RUNNING_XACTS has subxid overflow