documentation bug: atan2

[Andrew Wittenberg <andyw@xxxxxxxxxxxxxxxxxxxx>]

Hello again,

Many of you have probably already figured this out... it appears that in
the Ferret v5.22 documentation for atan2, the arguments X,Y are swapped
compared to what Ferret actually does.  The Fortran and Matlab versions of
ATAN2 are consistent with the Ferret program, so I believe the
documentation (not the program) should be changed (see details below).

Cheers,

Andrew

----------------------------------------------------------------------
Andrew Wittenberg          Program in Atmospheric and Oceanic Sciences
andyw@splash.princeton.edu                        Princeton University
----------------------------------------------------------------------

Ch3 Sec2.3.14. ATAN2 

Current documentation (3/20/01)
http://ferret.wrc.noaa.gov/Ferret/Documentation/Users_Guide/v52/Chapter3.htm#_VPINDEXENTRY_203
-------------------------------
ATAN2(X,Y)  2-argument trigonometric arctangent of Y/X (-pi,pi);
discontinuous at Y=0.  ATAN2( X,Y ) computes the 2-argument arctangent
of Y/X for all points within the current region 

[According to this, would have ATAN2(0,1)==pi/2, ATAN2(1,0)==0,
which is not what Ferret actually does.]

Ferret v5.22 program
--------------------
yes? list atan2(0,1)
             ATAN2(0,1)
          0.0000
yes? list atan2(1,0)
             ATAN2(1,0)
          1.571

Proposed new documentation
--------------------------
ATAN2(Y,X)  2-argument trigonometric arctangent of Y/X (-pi,pi);
discontinuous at Y=0.  ATAN2( Y,X ) computes the 2-argument arctangent
of Y/X for all points within the current region