Re: [Boost-users] [Review] UUID library (mini-)review starts today, November 23rd

Subject: Re: [Boost-users] [Review] UUID library (mini-)review starts today, November 23rd
From: Scott McMurray (me22.ca+boost_at_[hidden])
Date: 2008-12-06 15:28:30

On Sat, Dec 6, 2008 at 15:18, Andy Tompkins <atompkins_at_[hidden]> wrote:
> On Wed, 03 Dec 2008 22:05:39 -0300, "Agustín K-ballo Bergé"
> <kaballo86_at_[hidden]> said:
>>
>> Andy Tompkins escribió:
>> > It could easily be made static.  It will _always_ return 16.
>> >
>> Even for architectures with chars bigger than 8 bits long? From a
>> quick read of the library source code, and boost::integer
>> documentation, it seems that the library would not compile there.
>
> Hmm, can anyone verify this?  I will address this if it is a problem.
>
> What are the guarantees of the size of a byte?  Is a byte always 8 bits?
>

By C++ definition, a byte is the size of a char, and contains at least 8 bits.

My understanding is that posix sockets require CHAR_BIT == 8, so
outside of DSP chips and other special hardware, that's almost always
the case.  Considering that UUIDs were originally designed for RPC --
typically over sockets -- and that the v1 algorithm is defined using
MAC addresses, it's probably a safe assumption.