From: | Michael Meskes <meskes(at)postgresql(dot)org> |
---|---|
To: | Matthieu Imbert <matthieu(dot)imbert(at)ens-lyon(dot)fr> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: binary representation of datatypes |
Date: | 2008-10-21 13:53:31 |
Message-ID: | 20081021135331.GB6864@feivel.credativ.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, Oct 21, 2008 at 01:37:44PM +0200, Matthieu Imbert wrote:
> Yes microseconds are available in textual mode but i do want to use binary mode. Let me explain why:
> ...
> if i'm correct, it seems obvious that the second scenario is more efficient (and less ugly).
I wouldn't bet on scenario 2 being more efficient. For this you not only need
less conversions but also cheaper conversion. Now I haven't looked at this in
detail, but you might spend a lot of time doing stuff that has only a marginal
effect.
> In scenario 2, when talking about timestamp 'official' format, i mean timestamp expressed as number of microseconds since
> 2000-01-01. But of course, it only deserves this name 'official' if it is guaranteed to stay the same across postgresql versions and
> platforms
You shouldn't rely on this. Again I'd recommend using text.
Michael
--
Michael Meskes
Michael at Fam-Meskes dot De, Michael at Meskes dot (De|Com|Net|Org)
Michael at BorussiaFan dot De, Meskes at (Debian|Postgresql) dot Org
ICQ: 179140304, AIM/Yahoo: michaelmeskes, Jabber: meskes(at)jabber(dot)org
Go VfL Borussia! Go SF 49ers! Use Debian GNU/Linux! Use PostgreSQL!
From | Date | Subject | |
---|---|---|---|
Next Message | Merlin Moncure | 2008-10-21 14:01:17 | Re: binary representation of datatypes |
Previous Message | Michael Meskes | 2008-10-21 13:48:38 | Re: automatic parser generation for ecpg |