Paul Mensonides wrote: > ----- Original Message ----- > From: "Aleksey Gurtovoy" <[EMAIL PROTECTED]> > > > > #define NUMBERS \ > > > (0)(1)(2)(3)(4)(5)(6)(7)(8)(9) \ > > > (10)(11)(12)(13)(14)(15)(16)(17)(18)(19) \ > > > /* ... */ > > > > > > #define RANGE(first, length) \ > > > BOOST_PP_SEQ_SUBSEQ( NUMBERS, first, length ) \ > > > /**/ > > > > > > BOOST_PP_SEQ_FOLD_LEFT(AUX_MAX_ARITY_OP, -1, RANGE(0, 5)) > > > > > > > Looks good! How efficient is it? For instance, if NUMBERS > > is a sequence from 0 to 255? > > It should be very efficient. The length of the input doesn't > matter, the length of the output range matters slightly.
Oh, good! > Actually, if you only need a range starting from zero, you can > use BOOST_PP_SEQ_FIRST_N instead of BOOST_PP_SEQ_SUBSEQ which will > be more efficient yet. OK, will have it in mind. Thanks, Aleksey _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost