[boost] Re: mpl, type_traits and VC7.1

From: Andy Little (andy_at_[hidden])
Date: 2004-01-10 03:57:21

"David Abrahams" <dave_at_[hidden]> wrote
> "Fernando Cacciola" <fernando_cacciola_at_[hidden]> writes:

> > (2)
> > VC7.1 doesn't like the implementation of "mpl::equal_to" when the
operands
> > are "mpl::integral_c" of ENUMERATION type.

Arent enums different enough from integers to have their own 'enum_c'

1)
 enum-->int// ok , int --> enum //error.

2)
enums are best suited  for use where you need a Closed set
of arbitrarily different values. They should not really be used for maths,
because there is no requirement on enums that they go beyond a certain
range(C++3rdEd 4.8):

enum my{x = 1};    // acceptable that any 'my' is stored in 1 bit

// Might do what you want but If stored in 1 bit ...ouch!
static_cast<my>(x+x);

Hence  I am a bit uneasy using enums with integral_c. next,prior etc

3) enums can be differentiated:
#include <iostream>
enum my{x=1};
enum your{y=0};
void func(int x)
{
    std::cout <<"int" <<'\n';
}
void func(my m)
{
    std::cout <<"my" <<'\n';
}
void func(your y)
{
    std::cout <<"your" <<'\n';
}

int main()
{
    func(1);
    func(x);
    func(y);
    func(x+x); ///Ouch
    my  m = x;
    my mm = static_cast<my>(x+x) ;// Ouch
   int  mmm = x+y; //ok
}
/*
output:

int
my
your
int
*/

regards
Andy Little