Re: BUG #15812: Select statement of a very big number, with adivision operator seems to round up.

Поиск
Список
Период
Сортировка
От Dean Rasheed
Тема Re: BUG #15812: Select statement of a very big number, with adivision operator seems to round up.
Дата
Msg-id CAEZATCWLQ3p6sHkQ7mF9hMtkX4xaJDxmgO1mHh760BZ_B9W5Fg@mail.gmail.com
обсуждение исходный текст
Ответ на RE: BUG #15812: Select statement of a very big number, with adivision operator seems to round up.  (Kaleb Akalework <kaleb.akalework@asg.com>)
Список pgsql-bugs
On Fri, 17 May 2019 at 17:36, Kaleb Akalework <kaleb.akalework@asg.com> wrote:
>
> I tried Andres suggestion in his last email and that seems to work as a work around. Please see below.
>
> SELECT BIG_NUM, FLOOR(BIG_NUM /10000000000.0000000000),  BIG_NUM/10000000000.0000000000 from test_table
>
> "3691635539999999999"   "369163553"     "369163553.9999999999"
> "3691635530099999999"   "369163553"     "369163553.0099999999"
> "3691635530999999999"   "369163553"     "369163553.0999999999"
>
> But should this be a bug? Can their a better support of this, instead of having the query writer to know how many
decimalnumbers to put to get the correct type?
 
>
> It seems a little awkward and error prone to have to type .0000.... etc?
>

I would suggest using the numeric div() function, which divides a pair
of numeric values, and returns the truncated integer result. I.e.,
div(big_num, 10000000000). For example:

SELECT div(3691635539999999999::numeric(20, 0), 10000000000);
returns 369163553.

See https://www.postgresql.org/docs/current/functions-math.html

Regards,
Dean



В списке pgsql-bugs по дате отправления:

Предыдущее
От: Kaleb Akalework
Дата:
Сообщение: RE: BUG #15812: Select statement of a very big number, with adivision operator seems to round up.
Следующее
От: Andres Freund
Дата:
Сообщение: Re: BUG #15804: Assertion failure when using logging_collector withEXEC_BACKEND