From: Tom Lane <tgl@sss.pgh.pa.us>
To: Dino Maric <dinom@hey.com>
cc: pgsql-bugs@lists.postgresql.org
Subject: Re: Confusion about the range types
In-reply-to: <d1bbdc3da826051cc37efc1094673a2357ace614@hey.com>
References: <d1bbdc3da826051cc37efc1094673a2357ace614@hey.com>
Comments: In-reply-to Dino Maric <dinom@hey.com>
	message dated "Sat, 07 Jan 2023 16:17:19 +0100"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <3452151.1673107986.1@sss.pgh.pa.us>
Date: Sat, 07 Jan 2023 11:13:06 -0500
Message-ID: <3452152.1673107986@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/3452152.1673107986%40sss.pgh.pa.us>
Precedence: bulk

Dino Maric <dinom@hey.com> writes:
> When I insert range like this:
> INSERT INTO public.tests
> VALUES (int4range(7,8,'[]'))
> After when querying table my return value for this column is not [7,8]
> but it is [7,9).
> I found this behaviour confusing, because I want to insert 7-8 ranges
> (including upper value) and then present that range to a user.

This is the effect of canonicalization, as explained here:

https://www.postgresql.org/docs/current/rangetypes.html#RANGETYPES-DISCRETE

If you don't like it you can make a range type with a different
canonicalization function, or no such function, but that might
have odd effects on the behavior of range comparison operators.

			regards, tom lane