... args NAG f95: returns NaN I can't find anything in the IEEE floating-point standard nor in Fortran 2003 that is relevant. For the usual definition of IEEE NaN, returning NaN is probably the right thing to do. It might be that new functions that ignore NaN should be defined. Since there is already MAX0 and MAX1, how about MAX2? (I like it better than MAXNA.) -- glen

....2007 to mozno dac pomedzi zarty prosze zobaczyc http://jizdnirady.atlas.cz/ConnRes.asp?tt=a&from=3&from=0&from=6149&froml=-1 &to=3&to=0&to=8892&tol=-1&date=1.5.2007&time=4:00&min1=5&max1=240&alg=1&isde p=1&beds=0&chn=5&stop=0&odcht=0&odch=50&link=AA6B jezeli link niepokaze sie to prosto Praha - Wroclaw dnia 1.5.2007 na www.idos.cz ranek godzina 4 Albo zlikwidujÄ… R251...

....5.2007 to mozno dac pomedzi zarty prosze zobaczyc http://jizdnirady.atlas.cz/ConnRes.asp?tt=a&from=3&from=0&from=6149&froml=-1&to=3&to=0&to=8892&tol=-1&date=1.5.2007&time=4:00&min1=5&max1=240&alg=1&isdep=1&beds=0&chn=5&stop=0&odcht=0&odch=50&link=AA6B jezeli link niepokaze sie to prosto Praha - Wroclaw dnia 1.5.2007 na www.idos.cz ranek godzina 4 _________________ -- Wysłano z...

...> glen herrmannsfeldt <gah@seniti.ugcs.caltech.edu> wrote: For an extra complication (for comp.lang.fortran), what should MAX1(NaN,0) return? That's easy. It is illegal code because MAX1 requires default real arguments, but 0 is integer. Presumably that's not what you actually meant, but that's ...

glen herrmannsfeldt <gah@seniti.ugcs.caltech.edu> wrote: For an extra complication (for comp.lang.fortran), what should MAX1(NaN,0) return? That's easy. It is illegal code because MAX1 requires default real arguments, but 0 is integer. Presumably that's not what you actually meant, but that's my answer to ...

... MaxNum, etc. How languages expose those functions to users is up to the language. Personally I would not call them 'min' or 'max'... For an extra complication (for comp.lang.fortran), what should MAX1(NaN,0) return? Note that C, or at least C89, doesn't have max and min, but they are commonly done using preprocessor macros. They are then dependent on the result of the or =...