Re: BUG #2994: avg() calculates wrong on Interval-type

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Peter Eisentraut <peter_e(at)gmx(dot)net>
Cc: pgsql-bugs(at)postgresql(dot)org, "Frank F(dot) Burmo" <fburmo(at)online(dot)no>
Subject: Re: BUG #2994: avg() calculates wrong on Interval-type
Date: 2007-02-13 00:00:34
Message-ID: 16483.1171324834@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> Moreover, my system thinks that the average of '1 day 12 hours' and '8
> hours' is 22 hours, and also that the average of '2 days' and '3 days'
> is 2 days and 12 hours, so it does make the assumption that 1 day = 24
> hours.

Yeah, that's what it does with fractional days, because it doesn't have
a lot of choice --- we could perhaps change the day field from integer
to fractional, but I don't know what 0.5 day really means, so I'm not
sure that that'd be an improvement. What it won't do is up-convert
hours to days without being told to (via justify_hours).

The issue in its simplest form is:

regression=# select '3 days'::interval / 2;
?column?
----------------
1 day 12:00:00
(1 row)

regression=# select '72 hours'::interval / 2;
?column?
----------
36:00:00
(1 row)

regression=# select '3 days'::interval * 2;
?column?
----------
6 days
(1 row)

regression=# select '72 hours'::interval * 2;
?column?
-----------
144:00:00
(1 row)

regards, tom lane

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Bruce Momjian 2007-02-13 02:01:19 Re: [BUGS] wrong behavior using to_char()
Previous Message Anthony Taylor 2007-02-12 23:59:31 BUG #2996: to_char( timestamp, 'DD-Mon-YYYY HH24:MI:SS.MS' ) reports .1000 ms