On Fri, 10 Oct 2008 23:11:04 +1100 "Alex Ghitza" <[EMAIL PROTECTED]> wrote:
> The following looks ok (see comments below, however): > > sage: 1/RR(0) > +infinity > sage: RR(0)^(-1) > +infinity > > But how about this? > > sage: 1/CC(0) > NaN - NaN*I > sage: CC(0)^(-1) > ZeroDivisionError... > > I don't really like either of these; I guess I would prefer the > answer to be UnsignedInfinity in both cases. For that matter, I'm > not quite sure why 1/RR(0) is +infinity rather than -infinity, so I > guess I would prefer for *that* to be UnsignedInfinity as well. > > On the other hand, I don't really use this sort of thing every day, > so maybe someone who's closer to this issue can chime in. I don't use this either, but here's my opinion anyway. :) I think division by zero should always raise an error. When you obtain infinity in an expression, you shouldn't expect a meaningful result from your computation anyway. If you expect infinity in your computation, you can ignore the exception, but I believe in most cases, this will indicate an error. Note that this is different from evaluating an expression like 1/gamma(-1) to 0, which we already handle properly (IMHO). This might also be a good place to discuss if #2515 should be marked as invalid. Cheers, Burcin --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-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/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
