On Wed, 18 Aug 2010 17:47:10 -0700 (PDT) kcrisman <kcris...@gmail.com> wrote:
> > > On Aug 18, 7:02 pm, Markus Dangl <s...@q1cc.net> wrote: > > Hi, > > > > i've looked through this group and sage-support and could not find > > anything related. I'm using sage 4.5.2 on 64 bit Linux (Ubuntu) - if > > this is even relevant. > > > > The simplify_radical() method for symbolic expression seems to > > assume that sqrt(z*z) == abs(z) in all cases. This is wrong for > > complex numbers, where only sqrt(z * z.conjugate()) == abs(z). I'm > > relatively new to sage, so please excuse me if i missed it, but > > there seems to be no way to tell "simplify_radical()" to respect > > complex numbers. If i am wrong, this is not a bug, hence the post > > here before opening a bug report. > > Thanks for your report, Markus. I don't have time to look into this > in general, but the quick answer almost certainly is also the correct > one - namely, that the various simplify_*() methods implicitly make > assumptions. simplify_log(), at least, has good documentation about > this (thanks Robert M.!), and some of the others will too. The > simplify methods are Sage's best guess at what simplifications might > be useful to people from the plethora that Maxima provides for us, but > certainly not all of them are correct in full generality. You may > wish to look at Maxima's documentation for its various simplification > routines as well. > > That doesn't mean this isn't a bug, but if it's consistently assuming > z is real, for this simplification and the documentation doesn't say > so, it's more likely to be lacking documentation for this. If in > other places z could be complex and it behaves appropriately, maybe it > is a bug. For quite some time, maxima assumes that all variables are real by default. This is #6862 on trac: http://trac.sagemath.org/sage_trac/ticket/6862 I don't think there is anyone working on this. Burcin -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
