Subject: Re: Compiling NetBSD with another compiler.
To: None <email@example.com>
From: Thor Lancelot Simon <firstname.lastname@example.org>
Date: 10/13/2007 11:43:26
On Sat, Oct 13, 2007 at 05:33:02PM +0300, Aleksey Cheusov wrote:
> Just because you ALREADY do this for many years. Actually EVERY macro
> definition in system headers is a BAD NAME.
No. The relevant standards control which names the system header files
may use, including which ones may be (or must be, as in some of your
examples) preprocessor macros.
Some of the namespace of legal C names is reserved for us; we are
_required_ to consume some of it; some is reserved for the user, and
thus forbidden to us.
The C namespace is not flat, as others have pointed out: the fact that
a given token is reserved to us as a function name does not mean that
it is not forbidden to us as the name of an element in a structure. The
way you are suggesting we implement function renaming would violate that