This is the mail archive of the gcc-patches@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] |
Other format: | [Raw text] |
I want to resurrect this patch that I didn't pursue for 4.8, because our <stdbool.h> violates this very explicit requirement in the C++11 and C++14 standards: 18.10 [support.runtime] p8 "The header <cstdbool> and the header <stdbool.h> shall not define macros named bool, true, or false." Is the gcc/ginclude/stdbool.h change OK for trunk? Tested x86_64-linux. I also looked in the Debian Code Search and it seems that all the code which cares whether bool is a macro follows one of these patterns: #ifdef bool #undef bool #ifdef bool /* Leave if macro is from C99 stdbool.h */ #ifndef __bool_true_false_are_defined #undef bool (The __bool_true_false_are_defined macro *is* defined in C++) #ifdef bool #error bool should not be defined #endif (That's in the libc++ testsuite for <stdbool.h>, we fail that test obviously) #ifdef bool #define HAS_BOOL #endif So although that search only covers FOSS code I don't think we're going to break much C++ code by doing this, and it's needed for C++11 anyway. On 5 February 2012 13:00, Jonathan Wakely wrote: > On 4 February 2012 23:35, Gerald Pfeifer wrote: >> For what it's worth, I strongly suggest that you only define those when >> __cpluplus is pre-C++11. >> >> There is simply too much software out there which will run into this > > Really? Why would any C++ code assume "bool" is defined as a macro? > It's been a keyword in C++ for longer than C99 has defined it as a macro. > >> and being aggressive in breaking (admittedly non-standard confirming >> programs) gives GCC a bad reputation and is not nice to our users to >> begin with. > > Fair enough, this revised patch still defines bool, true and false for > C++98 mode, but not for C++11 mode. > > gcc/ > * ginclude/stdbool.h (true, false, bool): Do not define for C++11. > > libstdc++/ > * testsuite/18_support/headers/cstdbool/macros.cc: New. > > Tested x86_64-linux, OK for trunk?
Attachment:
patch.txt
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |