Ha ha. But I know that one thing *I *find frustrating is that atan2
doesn't simplify very well; but part of the problem with *that* is that
atan2(0,0) needs to be special-cased, e.g. carg(x) => atan2(0,x) can't be
simplified to unit_step(-x)*%pi or (if x<0 then %pi else 0) because that is
wrong for x=0 -- not to mention that unit_step doesn't simplify very well
either, e.g. unit_step(x)^2 doesn't simplify to unit_step(x); similarly,
atan2(a^2*x,x) is undefined for x=0, so is not equivalent to atan2(a^2,1).
But carg(0) (reasonably, I think) gives 0. Maybe we should allow these
cases, the same way we simplify x/x=>1....
Maybe I will look at this some day....
-s
On Thu, Jun 28, 2012 at 5:01 PM, Henry Baker <hbaker1 at pipeline.com> wrote:
> I haven't been using them for a while, so I don't recall all the details.
>
> Something about realpart,imagpart,rectfrom going off on an arctangent...
>
> At 08:07 PM 6/27/2012, Stavros Macrakis wrote:
> >Henry,
> >
> >Have you reported the problems you're having with real/imagpart? I think
> it would be useful to document all the problematic cases so that next time
> someone works on it, they can be improved.
> >
> > -s
> >
> >On Wed, Jun 27, 2012 at 9:29 AM, Henry Baker <hbaker1 at pipeline.com>
> wrote:
> >I do a lot of algebra with complex variables, with some expressions
> growing to thousands of terms.
> >
> >Due to the amount of space taken up by Maxima writing out
> "conjugate(foo)" in full for every conjugated
> >variable, I'd love to shorten this to " foo' " (or equivalent). I'm not
> so worried about
> >_input_ of conjugate as _output_.
> >
> >Is there any "easy"/"trivial" way to accomplish this in Maxima ?
> >
> >---
> >On a separate note, I've almost given up on using "realpart()",
> "imagpart()" and "rectform()"
> >due to unpredictable behavior. I use
> >
> >mrealpart(x):=(x+conjugate(x))/2;
> >
> >mimagpart(x):=(x-conjugate(x))/2/%i;
> >
> >instead.
> >
> >So I'm relatively happy with the functional operation of "conjugate()";
> I'd just like to cut down
> >on the volume of output characters.
> >
> >Thanks in advance for any help.
> >
> >Henry Baker
> >Santa Barbara, CA
>
>