Re: [boost] [Boost-users] [Review] ITL review starts today, February 18th

Subject: Re: [boost] [Boost-users] [Review] ITL review starts today, February 18th
From: Steven Watanabe (watanabesj_at_[hidden])
Date: 2010-03-04 12:50:19

AMDG

Joachim Faulhaber wrote:
> Well, I've looked at my code. It looks correct. The problem seems to be, that
> numeric_limits< interval<double >::size_type >::infinity()
> maps infinity() to 0.
>
> if(numeric_limits< std::size_t >::infinity()==0)
>    cout << "infinity()==0";
>
> Sine the test code, that I have done for that only checks for that value
> numeric_limits< interval<double >::size_type >::infinity() against
> cardinality(),
> I didn't detect the problem.
>
> Is there a standard way to deal with this kind of problem?
>   

Check numeric_limits<...>::has_infinity?

18.2.1.2:
"static T infinity() throw();
Representation of positive infinity, if available.
Meaningful for all specializations for which has_infinity != false. 
Required in specializations
for which is_iec559 != false."

In Christ,
Steven Watanabe