Re: Why there is a PG_GETARG_UINT32 and PG_RETURN_UINT32?

Поиск
Список
Период
Сортировка
От Amit Kapila
Тема Re: Why there is a PG_GETARG_UINT32 and PG_RETURN_UINT32?
Дата
Msg-id 005801ce31bf$776782b0$66368810$@kapila@huawei.com
обсуждение исходный текст
Ответ на Re: Why there is a PG_GETARG_UINT32 and PG_RETURN_UINT32?  (Rodrigo Barboza <rodrigombufrj@gmail.com>)
Список pgsql-hackers
On Friday, April 05, 2013 10:00 AM Rodrigo Barboza wrote:

> I am creating my own uint32 type and faced this function.
> But my args are always of type my_int and one of the signed int types as
postgres doesn't hava unsigned.
> Could I use those functions in operations between those types?

It should not be a problem if your signed type doesn't have any negative
value.


> I can't see a place for this, I don't know if I am missing something....

>On Fri, Apr 5, 2013 at 1:12 AM, Amit Kapila <amit.kapila@huawei.com> wrote:
>>On Thursday, April 04, 2013 8:30 PM Rodrigo Barboza wrote:

>> Hi guys.
>> I am wondering when I can use the PG_GETARG_UINT32 and PG_RETURN_UINT32.
>> If postgres has no unsigned int type, what is the use of these macros?
> They are mainly used for contrib module functionality or some built-in
> functions which are not exposed.
> For example, bt_page_items() receives relation name (text) and block
number
> (int), but internally the block number
> is unit32 as max blocks can be oxFFFFFFFE.

With Regards,
Amit Kapila.





В списке pgsql-hackers по дате отправления:

Предыдущее
От: Amit Kapila
Дата:
Сообщение: Re: Page replacement algorithm in buffer cache
Следующее
От: Dimitri Fontaine
Дата:
Сообщение: Re: [PATCH] Exorcise "zero-dimensional" arrays