Re: [Boost-users] isinf & Boost

From: Pete Bartlett (pete_at_[hidden])
Date: 2008-01-26 18:25:04

>Hi All,
>
>I have to use isinf in an application. The application
>was originally written using gcc (under linux) which
>supports isinf. I want to port it to Windows and VC++
>doesn't seem to support it. I googled around and I
>think the "omnipotent" Boost has something to help
>with this. Unfortunately, I couldn't get the details
>of how to make use of it. I would be grateful if
>someone can help me?
>
>Thank you in advance!
>
>metty
>

Try either John Maddock's Math Toolkit library or Johan Rade's floating
point utilities library. Both available from:

http://www.boost-consulting.com/vault/index.php?direction=0&order=&directory
=Math%20-%20Numerics

I believe the Math Toolkit will be in Boost 1.35 but recall finding both
libraries easy to lay over the top of 1.34. In John's library you need

Math_toolkit_headers.zip / boost / math / special_functions / fpclassify.hpp
/ bool template<typename T> isinf(T t);

I believe Johan's implementation is even more portable and robust than
John's, but perhaps less far through the review process:

Floating_point_utilities_v3.zip / boost / math / fpclassify.hpp / bool
template<typename T> isinf(T t);

Hth,
Pete