[CppAD] compile error with boost ublas

[Brad Bell]

I have been able to reproduce Casper's problem below and added a bug 
report; see
     https://github.com/coin-or/CppAD/blob/master/bug/boost_lu.sh

The error occurs in boost code of the form
      return norm_inf (e1 - e2) < epsilon
I think the problem here is that epsilon is coming from std::numeric_limits.

It seems that there has been a change in the C++ standard. The old 
standard used to say that one should not place anything in the std 
namespace. The new standard seems to allow for one specializing 
std::numeric_limits<Type> where Type is a user defined type. It seems 
that boost is using this; see
http://www.boost.org/doc/libs/1_56_0/libs/multiprecision/doc/html/boost_multiprecision/tut/limits/constants.html


On 4/13/2015 11:15 PM, Peter Caspers wrote:
> Hi Brad,
>
> the following code does not compile ( I tried with boost_1_57_0 and
> gcc 4.8.2 ). Can you check this please ?
>
> Thanks
> Peter
>
> #include <cppad/cppad.hpp>
> #include <boost/numeric/ublas/lu.hpp>
>
> int main() {
>      typedef CppAD::AD<double> T;
>      boost::numeric::ublas::matrix<T> a(5,5);
>      boost::numeric::ublas::permutation_matrix<std::size_t> pert(5);
>      // lu decomposition
>      const std::size_t s = lu_factorize(a, pert);
>      return 0;
> }
> _______________________________________________
> CppAD mailing list
> CppAD at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/cppad
>