From: | August Zajonc <augustz(at)augustz(dot)com> |
---|---|
To: | Andrew Dunstan <andrew(at)dunslane(dot)net> |
Cc: | Dennis Bjorklund <db(at)zigo(dot)dhs(dot)org>, "Joshua D(dot) Drake" <jd(at)commandprompt(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Money type todos? |
Date: | 2007-03-21 15:43:16 |
Message-ID: | 46015294.1090608@augustz.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Andrew Dunstan wrote:
> August Zajonc wrote:
>> The only other nice thing would be user defined precision, but
>> can live without that as most currencies work under nnn.mm.
>
> That's useless for our system at least. The minimum scale we use for
> money values is 5. I guess we can just continue to use numeric though.
>
Emailing off list with Andrew I'm reminded of one other feature that
would be potentially valuable in a money type, and that's being able to
specify the method of rounding.
There end up being a number of different rounding methods out there, and
in some cases it is very valuable to critical to follow a certain
method, usually dictated by externalities such as trading partners etc.
This could probably be a compile time flag even for postgresql and is
obviously a low priority behind speed.
Symmetric Arithmetic Rounding
Bankers Rounding
Round-half-down
Not sure its worth doing and if so probably a low priority, just
mentioning it for completeness.
- August
From | Date | Subject | |
---|---|---|---|
Next Message | Shane Ambler | 2007-03-21 15:47:32 | Re: Money type todos? |
Previous Message | Magnus Hagander | 2007-03-21 15:36:29 | Re: [COMMITTERS] pgsql: Native shared memory implementation for win32. |