Received: from sss.pgh.pa.us (sss.pgh.pa.us [209.114.132.154])
	by mail.postgresql.org (8.11.1/8.11.1) with ESMTP id eBJ12m630809
	for <pgsql-docs@postgresql.org>; Mon, 18 Dec 2000 20:02:48 -0500 (EST)
	(envelope-from tgl@sss.pgh.pa.us)
Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1])
	by sss.pgh.pa.us (8.11.1/8.11.1) with ESMTP id eBJ12nC02642;
	Mon, 18 Dec 2000 20:02:49 -0500 (EST)
To: "He Weiping(Laser Henry)" <laser@zhengmai.com.cn>
cc: pgsql-docs@postgresql.org
Subject: Re: some errors and/or bugs? 
In-reply-to: <3A3EB05E.C564D84A@zhengmai.com.cn> 
References: <3A3EB05E.C564D84A@zhengmai.com.cn>
Comments: In-reply-to "He Weiping(Laser Henry)" <laser@zhengmai.com.cn>
	message dated "Tue, 19 Dec 2000 08:48:30 +0800"
Date: Mon, 18 Dec 2000 20:02:49 -0500
Message-ID: <2639.977187769@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>
X-Archive-Number: 200012/55
X-Sequence-Number: 102

"He Weiping(Laser Henry)" <laser@zhengmai.com.cn> writes:
> laser_zh=# select round(42.5);
>  round
> -------
>     42
> (1 row)
>
> is it correct? I think result should be 43.

It's correct, if the underlying float arithmetic is IEEE-compliant.
Rounding for exact half-integral values is supposed to be "round to
nearest even".  So 42.5 goes to 42, but 43.5 goes to 44.

I notice our NUMERIC rounding code does not do that ... perhaps it
should.

Also, the numeric round() function is just plain broken:

regression=# select round(42.5::numeric);
ERROR:  Function 'numeric_round(numeric, int4)' does not exist
        Unable to identify a function that satisfies the given argument types
        You may need to add explicit typecasts
regression=# select round(42.5::numeric,0);
 round
-------
    43
(1 row)

Looks like a silly oversight in the pg_proc entry for round(numeric).

			regards, tom lane