Re: apparently wrong description of 'arg' function in octave-3.0.5

[Ben Abbott]

On Apr 17, 2011, at 8:55 AM, Sergei Steshenko wrote:

> Hello,
> 
> in octave-3.0.5 'help arg' produces:
> 
> "
> -- Mapping Function:  arg (Z)
> -- Mapping Function:  angle (Z)
>     Compute the argument of Z, defined as THETA = `atan (Y/X)'.  in
>     radians.
> ".
> 
> According to Wiki and my memories from school:
> 
> http://en.wikipedia.org/wiki/Inverse_trigonometric_functions :
> 
> arctangent    y = arctan x    x = tan y       all real numbers        −π/2 < 
> y < π/2  −90° < y < 90°
> 
> , i.e. the ranges equivalently are:
> 
> −π/2 < y < π/2        −90° < y < 90°
> .
> 
> The above ranges are insufficient for complex numbers - the range in
> radians should be either
> 
> -pi .. pi
> 
> or
> 
> 0 .. 2 * pi.
> 
> In reality the 'arg' function works correctly:
> 
> "
> octave:23> 180 / pi * arg(-1 + 0.01i)
> ans =  179.43
> octave:24> 180 / pi * arg(-1 - 0.01i)
> ans = -179.43
> ",
> 
> i.e. it implements -pi .. pi (-180 .. 180 degrees) range - opposed to
> what its built-in help message says (because of 'atan' and it's
> -pi/2 .. pi/2 range).
> 
> Is the description fixed in later 'octave' versions ?

The current sources include ...

 -- Mapping Function:  arg (Z)
 -- Mapping Function:  angle (Z)
     Compute the argument of Z, defined as, THETA = `atan2 (Y, X)', in
     radians.

Ben