Waldek Hebisch <[EMAIL PROTECTED]> writes: > Martin, first I _really_ prefer to get exceptions in normal code. > Ignoring exceptions is a great way to do not see bugs. Also, things > like infinities and (particularly nasty) "not a number" break normal > mathematical reasoning.
Don't worry, I see your point. One thing though: > (basically IEEE defined a new formal system quite unlike mathematical real > numbers). Is it really different? I thought that computation with +infinity and -infinity was OK - except for imagpart... I don't care about nan. (but read on, first...) > Note: trapping exceptions in plotted (user provided) function is > reasonable because we have no control on user code. Yes, this is what I want in the end. It would have been easier to commit the current code, but I guess that you are right. > I think that correct way is to rewrite plotting routine in a way > which avoids overflow. If we up front limit user values to some > reasonable range (like -10^150..10^150) this should be not that > hard. Of course, we will have to think a while about each formula, > but if we do not care much what happens with big numbers than > it will be easy to clip them. If some case turns out to be more > tricky, then it is quite likely that ignoring errors would > produce wrong value. I understand that clipping code may > be bulky and look ugly, but proper clipping/rounding is essential > part of the plotting algorithm. Unfortunately, it is not that easy, since trapNumericErrors does *not* trap all errors - most importantly, (float reallybiginteger 0.0d0) throws a type-error instead: ------------------------------------------------------------------------ )lisp (defmacro |mytrap| (form) `(handler-case (cons 0 ,form) (arithmetic-error () |$numericFailure|) (simple-type-error (var) (print (simple-condition-format-control var)) |$numericFailure|))) Value = |mytrap| (1) -> )lisp (|mytrap| (float (expt 10 400) 0.0d0)) "Too large to be represented as a ~S:~% ~S" Value = (1 . "failed") (1) )lisp (|mytrap| (float (- (expt 10 400)) 0.0d0)) "Too large to be represented as a ~S:~% ~S" Value = (1 . "failed") ------------------------------------------------------------------------ One problem I have is that myTrap distinguishes between positive and negative infinity. However, if I trap arithmetic-errors myself, I loose this possibility: I can only see in which operation the error occurred and cannot continue computation. So, maybe I should just return max()$FLOAT always and ignore infinities? I'll try that first. Martin --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group. To post to this group, send email to fricas-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/fricas-devel?hl=en -~----------~----~----~----~------~----~------~--~---
