When casting using (idouble) to set the imaginary part of a complex double, the imaginary part of a cdouble never gets set.

int main()
{

   double a=5;
   double b=-0.5;
   cdouble plx = a +b*1.0i;
   cdouble ply = a+ (idouble)b;
   printf("re:%f im:%f\n",plx.re,plx.im);
   printf("re:%f im:%f\n",ply.re,ply.im);

    return 1;
}

prints out   re:5.0 im:-0.5
                 re:5.0  im:0.0

On Thu, 19 Feb 2004 17:12:28 -0500, dickl wrote:

> When casting using (idouble) to set the imaginary part of a complex double, the imaginary part of a cdouble never gets set.

Confirmed but not a bug. Casting a number to idouble or double yields just the imaginary or the real part of that number.

How else should that casts be defined? What should the result of
`cast(idouble)(2+3i)' be?

So long.

Ok,  but it makes it a little tedious to build a complex number from individual variables.

It would be nice to have a way to say
cdouble var = a + (imaginary)b;



"Manfred Nowak" <svv1999@hotmail.com> wrote in message news:c13se2$2pue$1@digitaldaemon.com...
> On Thu, 19 Feb 2004 17:12:28 -0500, dickl wrote:
>
> > When casting using (idouble) to set the imaginary part of a complex
double,
> > the imaginary part of a cdouble never gets set.
>
> Confirmed but not a bug. Casting a number to idouble or double yields just the imaginary or the real part of that number.
>
> How else should that casts be defined? What should the result of
> `cast(idouble)(2+3i)' be?
>
> So long.

dickl wrote:

[...]
> It would be nice to have a way to say
> cdouble var = a + (imaginary)b;

 `cdouble var = a + b * 1i;'
is even shorter. Why to introduce something special, that needs more
effort? And dont forget the cast! In D the cast keyword is obligate. Only
dmd is so kind to forget about that.

So long.