Re: formatting.c

Поиск
Список
Период
Сортировка
От Bruce Momjian
Тема Re: formatting.c
Дата
Msg-id 20140913142645.GE25518@momjian.us
обсуждение исходный текст
Ответ на Re: formatting.c  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
On Fri, Sep 12, 2014 at 11:48:03PM -0400, Tom Lane wrote:
> Bruce Momjian <bruce@momjian.us> writes:
> > Is everyone OK with me renaming some variables, structures, and macros? 
> > It will make back-patching harder, but will allow us to properly maintain
> > that file.
> 
> The back-patching problem could be addressed by back-patching the
> renaming.  If it's purely a mechanical thing, there should be minimal risk
> no?  I would claim that messed-up back-patches would have non-negligible
> risk too, so you can't say that not back-patching is clearly safer.

Uh, yes, it is mechanical and could be backpatched as well.

> Now, whether the renaming actually makes things any clearer is something
> I reserve judgment on.

What the code does now is call almost everything a "number", including
the format values, binary and string representations of the number, etc.
You can take a look at reverted commit
f68dc5d86b9f287f80f4417f5a24d876eb13771d to see an example of the
renaming;  that is probably only half the job.

--  Bruce Momjian  <bruce@momjian.us>        http://momjian.us EnterpriseDB
http://enterprisedb.com
 + Everyone has their own god. +



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Fabrízio de Royes Mello
Дата:
Сообщение: Re: [GSoC2014] Patch ALTER TABLE ... SET LOGGED
Следующее
От: Andres Freund
Дата:
Сообщение: Re: [REVIEW] Re: Compression of full-page-writes