Formatted input: Detecting numeric overflow?

I couldn't find anything about how the formatted input operator handles numeric overflows. For example, how can you detect an overflow in the following code:

Code:

---

int x;
std::cin >> x;

---

Lol I tried to find the solution for you looked around for about 2hrs. I see alot of ways to test if a multiplication or addition is above limits using.

Code:

---

if((x*x) > std::numeric_limits<int>::max())

---

but I couldn't find any way to combine this with

Code:

---

std::cin >> x;

---

So my only guess would be putting the input into a string or larger bit variable like a double and then checking if it's greater than the int max. Such as

Code:

---

#include <limits>
#include <iostream>
using namespace std;

int main(int argc, char *argv[])
{
  double a;
  std::cout<<" max int value = "<<std::numeric_limits<int>::max()<<endl;
  std::cin>>a;
  while((a) > std::numeric_limits<int>::max() )
  {
    std::cout << "overflow" << std::endl;
    std::cout << "please input a non overflow number."<<endl;
    std::cin>>a;
  }
  else
    std::cout<< "Your number is within the limits of an int"<<endl;
 
  int b = a;
  std::cout<<b<<endl;
  system("pause");
  return 0;
}

---

I'm sure this wouldn't be the most efficient method, but I've tried alot of combinations and the only way to glitch it out that I've found is inputing a character instead of a number, or going over the limits of double. (which a string would probably solve, but of course the code would get a little lengthy.) Of course this obviously doesn't check for a negative overflow either.

There is some discussion on the matter here: C++ Standard Library Active Issues List

It seems that failbit gets set on numeric overflow, but there's no way to explicitly check for overflows.