On Thu, 12 Dec 2013, Florian Weimer wrote: > > #define PP_NARG( ...) PP_NARG_(__VA_ARGS__,PP_RSEQ_N()) > > #define PP_NARG_(...) PP_ARG_N(__VA_ARGS__) > > #define PP_ARG_N(_1,_2,_3,_4,_5,_6,_7,_8,_9,[..],_61,_62,_63,N,...) N > > #define PP_RSEQ_N() 63,62,61,60,[..],9,8,7,6,5,4,3,2,1,0 > > Hmm. I think this returns the 64th argument if the argument list is longer > than 63. I don't want to silently produce wrong results if some arbitrary > limit is exceeded. You can establish an upper bound on the number of arguments using the following device: #define PP_STRIP_1(_, ...) __VA_ARGS__ #define PP_STRIP_2(_, ...) PP_STRIP_1(__VA_ARGS__) #define PP_STRIP_3(_, ...) PP_STRIP_2(__VA_ARGS__) PP_STRIP_3(a, b, c, d, e) -> d, e PP_STRIP_3(a) -> (empty) So you can check that arg list count does not exceed 63 by examining sizeof(STR((PP_STRIP_63(__VA_ARGS__)))) One downside to this is that the above simple device relies on a GNU extension to the C preprocessor and produces a warning with -pedantic. Alexander
