From bug-octave-request at bevo dot che dot wisc dot edu Tue Jan 22 14:09:19 2002
Subject: arg(double) patch
From: Paul Kienzle <pkienzle at jazz dot ncnr dot nist dot gov>
To: bug-octave at bevo dot che dot wisc dot edu
Date: Tue, 22 Jan 2002 15:09:16 -0500

John,

I'm including a patch to lo-mappers so that 
	Y=arg(X) 
will match 
	for i=1:length(x)
	  Y(i)=arg(X(i)); 
	end 
even if X contains signed zeros.

Test with:

  x=[0+0i, 0-0i, -0+0i, -0-0i, 1i, NaN+0i, NaN-0i,NaN*1i];
  y=x; for i=1:length(y), y(i)=arg(y(i)); end
  [x;y;arg(x)].'

Note that there is a separate problem in the octave print
routines in that they are not showing signed zeros, but
that will be in a later patch.

Paul Kienzle
pkienzle at users dot sf dot net

Index: lo-mappers.cc
===================================================================
RCS file: /cvs/octave/liboctave/lo-mappers.cc,v
retrieving revision 1.19
diff -c -p -r1.19 lo-mappers.cc
*** lo-mappers.cc	2001/02/06 15:57:23	1.19
--- lo-mappers.cc	2002/01/22 19:58:49
*************** Software Foundation, 59 Temple Place - S
*** 62,75 ****
  double
  arg (double x)
  {
!   if (x < 0.0)
!     return M_PI;
!   else
! #if defined (HAVE_ISNAN)
!     return xisnan (x) ? octave_NaN : 0.0;
! #else
!     return 0.0;
! #endif
  }
  
  double
--- 62,68 ----
  double
  arg (double x)
  {
!   return atan2(0.0,x);
  }
  
  double



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------