Missing overload of evaluate function for scalars

[Marcin Copik]

I noticed that evaluate can't be apply to expressions which result is a scalar value, for example:

blaze::DynamicVector<double> v1(10), v2(10);
auto x = blaze::evaluate(blaze::trans(v1) * v2);

does not compile since evaluate is defined only for Matrix and Vector types. I fixed it by adding another overload

template< typename T, typename = std::enable_if_t<std::is_floating_point<T>::value || std::is_integral<T>::value> >
inline T evaluate( const T & result)
{
   return result;
}

It seems to work. I'd like to contribute a patch but I'm not sure where this function should be declared. Should I create a new file math/expressions/Scalar.h?

[Klaus Iglberger]

Commit ca310df adds a generic evaluate() function for fundamental and complex data types. The new evaluate() function is immediately available via cloning the Blaze repository and will be officially released in Blaze 3.3.

[Marcin Copik]

Thanks @eagle42 !