On Jul 11, 2025 at 23:57 +0800, Tom Lane , wrote:
> "David G. Johnston" writes:
> > On Friday, July 11, 2025, Zhang Mingli wrote:
> > > So, are both result sets technically correct given the absence of an ORDER
> > > BY clause?
>
> > The system is behaving within the requirements of the specific
the ORDER BY clause on windows function, were the regression
test results become deterministic.
Thanks in advance
Dinesh
From: Tom Lane
Sent: Friday, July 11, 2025 9:27 PM
To: David G. Johnston
Cc: Zhang Mingli ; PostgreSQL Hackers
Subject: Re: [Question] W
"David G. Johnston" writes:
> On Friday, July 11, 2025, Zhang Mingli wrote:
>> So, are both result sets technically correct given the absence of an ORDER
>> BY clause?
> The system is behaving within the requirements of the specification. The
> query itself is bugged code that the query author
On Friday, July 11, 2025, Zhang Mingli wrote:
>
> Referring to the SQL 2011 standard, it states that if ORDER BY is
> omitted, the order of rows in the partition is undefined.
> While using a window function without ORDER BY is valid, the resulting
> output seems unpredictable.
>
> So, are both r
Hi,
I am reaching out to discuss the behavior of window functions in Postgres,
specifically regarding the use of the OVER() clause without an ORDER BY
specification.
In our recent tests, we observed that the results can be unstable.
For example, when executing the following query:
SELECT sum(u
