Proposed bug fix for Bug 505443



>>>>> "willisb" == willisb  <willisb@unk.edu> writes:

    willisb> Perhaps

    willisb> sin(x/2) = signum(sin(x/2)) * sqrt(1-cos(x)) / sqrt(2),

I tried that and, since halfangles is true, the sin(x/2) inside the
signum gets expanded out, which then just makes it wrong too.

    willisb> However, Maxima's signum function doesn't "know" anything about numbers
    willisb> off the real axis;  I don't think a function should be fixed-up on
    willisb> the negative real axis only to mess it up everywhere off the real axis.

Not sure what to do about that because I don't think maxima does a
very good job about complex numbers in general.  (My general feeling,
no definitive examples.)

Ray