[su_junwei]

Dear everyone

I made a simple test about the class scalar, the code
are listed below


#include "IOstreams.H"
#include "scalar.H"
using namespace Foam;
int main()
{
scalar s=10;
scalar ss=exp(s);
#ifdef DP
Info<<ss<<endl;
#endif
}


but the compiler always complained with message below
test.C:87: error: call of overloaded 'exp(Foam::scalar&)' is ambiguous
/usr/include/bits/mathcalls.h:101: note: candidates are: double exp(double)
/home/sujunwei/OpenFOAM/OpenFOAM-1.4.1/src/OpenFOAM/lnInclude/Scalar.H:179: note: Foam::doubleScalar Foam::exp(Foam::doubleScalar)
/home/sujunwei/OpenFOAM/OpenFOAM-1.4.1/src/OpenFOAM/lnInclude/Scalar.H:179: note: Foam::floatScalar Foam::exp(Foam::floatScalar)

anything wrong ?
but code with
scalar ss=expf(s);
it passed !

it seems that the function macro

#define transFunc(func) \
inline Scalar func(const Scalar s) \
{ \
return ::func##f(s); \
}

in floatScalar.H was activated by default ?

compilation with double precision is default in OpenFOAM ?