This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Project proposal
- From: Carlo Wood <carlo at alinoe dot com>
- To: Jeffrey Turner <jturner at mail dot alum dot rpi dot edu>
- Cc: libstdc++ at gcc dot gnu dot org, gccadmin at gcc dot gnu dot org, gcc at gcc dot gnu dot org, pedwards at disaster dot jaj dot com, law at gcc dot gnu dot org
- Date: Wed, 2 Jan 2002 02:24:37 +0100
- Subject: Re: Project proposal
- References: <200201011310.AA603324472@mail.alum.rpi.edu>
On Tue, Jan 01, 2002 at 01:10:53PM -0500, Jeffrey Turner wrote:
> In light of my recent comments on the preprocessor
> and in keeping with the spirit of C++ I am hereby
> volunteering to remove all #define statements from
> the bits, c_std and c_shadow directories (except
> for those in header guards). I will start with
> bits/std_limits.h, which I expect to have done w/in
> a week.
But, but, who all agrees with that?
I didn't see anyone even reply to your previous mail...
Actually, I think that using macro's for extensions
is not a bad thing at all, as long as they are not
part of the API. Since one cannot rely on extensions
to be available everywhere, a macro like
#define _IsUnused __attribute__ ((unused))
is a Good Thing(tm) even.
The "authorities" that are against macro's are
talking about the use of macro "functions", which
are type unsafe and have a C++ counterpart that
is type-safe (inline functions) and thus better.
There is no reason not to use macro's for other
things.
It would be a good thing though if the the macro's
that are a 'standard' within the gcc code would be
clearly documented somewhere - so that new developers
can easily get an overview of what they are supposed
to use.
> Sincerely,
>
> --Jeff Turner
>
> PS. I suspect I'll need CVS write access.
You must be kidding :)
--
Carlo Wood <carlo@alinoe.com>