[vtk-developers] isnan

[John Platt]

David,

 

You're probably right about MSVC 6 but for those of us who are stuck
with it, the following may be useful.

 

John.

 

static int IsNaN(const double x)

{

  unsigned int e_mask, f_mask;

  double as_double = x;

  unsigned int *as_int;

 

  vtkByteSwap::Swap8BE(&as_double);

  as_int = (unsigned int *)&as_double;

  vtkByteSwap::Swap4BE(as_int);

 

  e_mask = 0x7ff0;

  e_mask <<= 16;

 

  if ((*as_int & e_mask) != e_mask)

    {

    return 0;

    }

 

  f_mask = 1;

  f_mask <<= 20;

  f_mask--;

 

  return (*as_int & f_mask) != 0;

}

 

-----Original Message-----
From: vtk-developers-bounces+jcplatt=dsl.pipex.com at vtk.org
[mailto:vtk-developers-bounces+jcplatt=dsl.pipex.com at vtk.org] On Behalf
Of Thompson, David C
Sent: 14 January 2007 00:10
To: Michael Halle; vtk-developers at vtk.org
Subject: RE: [vtk-developers] isnan

 

> In the mean time, how about the question of where to put
> the actual isnan function in VTK, hopefully in a way that
> introduces as little overhead as possible?  Is vtkstd::isnan()
> the right place (since that's analogous to std::isnan() )?

Note that vtkMath already has static member functions
Inf(), NegInf(), and Nan() that return constants. If isnan
is placed in vtkstd, then these functions should also move
there. I would prefer vtkMath::IsNan() though. Also, I'm under
the impression that some of the compilers that Kitware wishes
to support (*cough*MSVC 6*cough*) don't have isnan() at all -- but
I could easily be wrong about that as I am not a Windows person.

    David 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtk-developers/attachments/20070114/f9b92b8d/attachment.html>