Re: BRIN minmax multi - incorrect distance for infinite timestamp/date

From: Tomas Vondra <tomas(dot)vondra(at)enterprisedb(dot)com>
To: Ashutosh Bapat <ashutosh(dot)bapat(dot)oss(at)gmail(dot)com>
Cc: Dean Rasheed <dean(dot)a(dot)rasheed(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: BRIN minmax multi - incorrect distance for infinite timestamp/date
Date: 2023-10-19 11:21:51
Message-ID: 768756b0-7598-6b69-2646-6db6dbe8511b@enterprisedb.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 10/19/23 11:22, Ashutosh Bapat wrote:
> On Thu, Oct 19, 2023 at 2:31 PM Tomas Vondra
> <tomas(dot)vondra(at)enterprisedb(dot)com> wrote:
>
>>
>> Does that explain the algorithm? I'm not against clarifying the comment,
>> of course.
>
> Thanks a lot for this explanation. It's clear now.
>
>> I tried to do that, but I ran into troubles with the "date" tests. I
>> needed to build values that close to the min/max values, so I did
>> something like
>>
>> SELECT '4713-01-01 BC'::date + (i || ' days')::interval FROM
>> generate_series(1,10) s(i);
>>
>> And then the same for the max date, but that fails because of the
>> date/timestamp conversion in date plus operator.
>>
>> However, maybe two simple generate_series() would work ...
>>
>
> Something like this? select i::date from generate_series('4713-02-01
> BC'::date, '4713-01-01 BC'::date, '-1 day'::interval) i;

That works, but if you try the same thing with the largest date, that'll
fail

select i::date from generate_series('5874896-12-01'::date,
'5874897-01-01'::date,
'1 day'::interval) i;

ERROR: date out of range for timestamp

regards

--
Tomas Vondra
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrei Zubkov 2023-10-19 12:40:24 Re: [PATCH] Tracking statements entry timestamp in pg_stat_statements
Previous Message Marko Tiikkaja 2023-10-19 11:03:42 Re: [HACKERS] Allow INSTEAD OF DELETE triggers to modify the tuple for RETURNING