This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: std::cout and long long problem
- To: Alexandre Oliva <aoliva at redhat dot com>
- Subject: Re: std::cout and long long problem
- From: Benjamin Kosnik <bkoz at redhat dot com>
- Date: Mon, 12 Feb 2001 11:55:58 -0800 (PST)
- cc: Gabriel Dos Reis <gdr at codesourcery dot com>, Craig Rodrigues <rodrigc at mediaone dot net>, libstdc++ at gcc dot gnu dot org
Seems kind of odd to have long long support and only enable some of it. I
feel your frustration, but we have to come up with a non-kludgy way to
make all long long support work, and make it work correctly with things
like -ansi. Unfortunately, this means making more than just iostreams work.
It's the "harder but more correct" route...
> WRT OS bits, wouldn't it be reasonable to arrange for any C++
> translation unit to implicitly #include the libstdc++ config bits
> upfront? Adding `-include c++config.h' to the C++ specs doesn't seem
> too much trouble to me.
This might be the way to go.
I'm working on __EXCEPTIONS
related issues right now and depending on how that stuff goes I might be
more up for this approach. It uses a similar mechanism.
-benjamin