This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: ostream and long long


Hmm... I see your point, but I don't see how it would work... It would
somehow need to be invoked only if the code were actually gonna run at
compile time, because this being in a header file means it will always
get compiled.


What I'd like to know (and I've posted it (with no replies) on gcc-help)
is if there is a way to change the bool type to 32bits on my particular
platform ... it used to be, but now some support libraries break because
it deals with pointers to bool.... (plz, email me privatly or post a
reply in gcc-help.. this is OT)

On 24 Jul 2001 21:52:33 -0400, Phil Edwards wrote:
> 
> I wish the compiler had some sort of __builtin_compile_error(), like #error
> but only triggered if the code segment in question is actually compiled.
> 
>   ostream& operator<< (ostream&, long long)
>   {
>   #ifdef _GLIBCPP_USE_LONG_LONG
>       ... what we currently have ...
>   #else
>       __builtin_compile_error("support for `long long' not configured, see "
>                               "<the location of v3's configopts.html>");
>   #endif
>   }
> 
> 
> Phil
> 




Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]