Re: Infinite Interval

From: Joseph Koshakow <koshy44(at)gmail(dot)com>
To: jian he <jian(dot)universality(at)gmail(dot)com>
Cc: Ashutosh Bapat <ashutosh(dot)bapat(dot)oss(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Infinite Interval
Date: 2023-01-14 21:22:49
Message-ID: CAAvxfHcm1TPwH_zaGWuFoL8pZBestbRZTU6Z=-RvAdSXTPbKfg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Ok, I've updated the patch to handle every function that inputs or
outputs intervals, as well as added some tests. In the process I
noticed that some of the existing date/timestamp/timestamptz don't
handle infinite values properly. For example,
postgres=# SELECT age('infinity'::timestamp);
age
--------------------------------------------------
-292253 years -11 mons -26 days -04:00:54.775807
(1 row)

It might be worth going through all those functions separately
and making sure they are correct.

I also added some overflow handling to make_interval.

I also added handling of infinite timestamp subtraction.

At this point the patch is ready for review again except for the one
outstanding question of: Should finite checks on intervals only look at
months or all three fields?

- Joe

Attachment Content-Type Size
v11-0001-Support-infinite-interval.patch text/x-patch 87.4 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2023-01-14 21:23:47 Re: Removing redundant grouping columns
Previous Message Justin Pryzby 2023-01-14 20:57:07 Re: Support for dumping extended statistics