This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: [v3] long long, c99 support
- To: Gabriel Dos Reis <Gabriel dot Dos-Reis at cmla dot ens-cachan dot fr>
- Subject: Re: [v3] long long, c99 support
- From: Benjamin Kosnik <bkoz at redhat dot com>
- Date: Fri, 20 Apr 2001 01:35:54 -0700 (PDT)
- cc: gcc-patches at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org
Hey Gaby.
> I'd rather have c99 inside std::. We can't make the assumption that
> c99 isn't a sensible user-reserved identifier name (generated codes
> tends to looks really ugly). If afterwards, it turns out that c99 is
> really a reserved name, then it is easy to use namespace alias to do
> the right thing.
>
> namespace std
> {
> namespace c99 { }
> using namespace c99;
> }
How is this any better, just out of curiosity? Perhaps it's late. Note that
all the c_std/bits headers have to be changed too, not just this one bit.
> Well, I prefer to leave <math.h> as is: to test C++ as proper.
> One add additional test for testing <cmath> but the test with <math.h>
> should not be removed.
Yeah. I know, me too. The reality is that the <math.h> solution is
relatively far-off, whereas <cmath> is now mostly fixed. It would be cool
to have something that is either XFAIL in the testsuite or fixed, so that
people who don't care about the "C" bits won't stress (ie backend people
doing ports will then pay attention to the test results.)
-benjamin