From: Junio C Hamano <gitster@pobox.com>
Date: Thu, 23 Jul 2009 05:15:13 GMT
Subject: Re: [PATCH 2/3] refactor: use bitsizeof() instead of 8 * sizeof()
Message-ID: <7vljmg56am.fsf@alter.siamese.dyndns.org>
In-Reply-To: <20090723051112.GA9372@coredump.intra.peff.net>

Jeff King <peff@peff.net> writes:

> On Wed, Jul 22, 2009 at 10:09:56PM -0700, Junio C Hamano wrote:
>
>> Jeff King <peff@peff.net> writes:
>> 
>> > On Wed, Jul 22, 2009 at 11:34:34PM +0200, Pierre Habouzit wrote:
>> >
>> >>  #define ARRAY_SIZE(x) (sizeof(x)/sizeof(x[0]))
>> >> +#define bitsizeof(x)  (CHAR_BIT * sizeof(x))
>> >
>> > Isn't our style to use all-caps for macros? I.e., BITSIZEOF?
>> 
>> Perhaps but I'd say similarity between sizeof() and bitsizeof() calls for
>> consistency in the lowercase in this particular case.
>
> Fair enough. I just wanted to point it out in case nobody thought of it.

I've actually thought of using elemof() to call what is known as
ARRAY_SIZE() when we introduced it, but now it is way _too_ late ;-)