imag () function detail regarding -0.0000

[Daniel J Sebald]

I was cleaning up fftfilt() wanting to zero out the real component of 
the final result and noticed -0 showing up unexpected to me.  To achieve 
this I'm using something like y(:,xisimag) = imag(y(:,xisimag))*i, which 
to me is the most straightforward approach.  The appearance of the -0's 
boils down to the following behavior of multiplication:

octave:59> [1,-1,-0]'*i
ans =

   0 + 1i
  -0 - 1i
  -0 - 0i

Is this proper behavior?  I would think

ans =

   0 + 1i
   0 - 1i
   0 - 0i

would be more appropriate in this case.

Dan