Re: [lmi] Cating doubles to enums

[Greg Chicares]

On 2018-11-04 20:36, Vadim Zeitlin wrote:
> On Sun, 4 Nov 2018 17:49:17 +0000 Greg Chicares <address@hidden> wrote:
> 
> GC> Therefore, what do you think of, say,
> GC>   double query    (database_key);
> GC>   void   query_ref(T&, database_key);
> GC>   void   query_ref(vector<T>&, database_key);
> GC> which distinguishes the names without denying the similarity of
> GC> the operations?
> 
>  I'm fine with this.
> 
> GC> I tried different second lexemes:
> GC>   query_value  meaningless--they all retrieve values
> GC>   query_into   seems obscure
> 
>  But my preferred version would be this one.

Then its name shall be called query_into<>(), and the implementation
for scalars is:

template<typename T>
void product_database::query_into(T& dst, e_database_key k) const
{
    double d = [result of calling a more fundamental overload]
    if constexpr(std::is_enum_v<T>)
        {
        dst = static_cast<T>(bourn_cast<std::underlying_type_t<T>>(d));
        }
    else
        {
        static_assert(std::is_integral_v<T>);
        dst = bourn_cast<T>(d);
        }
}

which falls just short of my threshold for adding a unit test module.