How to get bit representation of type double in C++?

Question

C++ / C#

You need to get a bit representation of type double in C++. How can this be done with a built-in function from somewhere, or by manually getting the sign, mantissa, and exponent?

Answer 1

The idea is that double has 64 bits. So if we use type punning and overlay a 64-bit uint64_t double on top, we get a 64-bit integer containing the bits of the original double.
*reinterpret_cast(&a)
Next, we use the standard trick with std::bitset to translate a number into a string containing its binary representation
. And then a couple of useful facts:
- we output the same number, but with a minus sign. You can see that the sign is stored in the first bit.
- similarly, we can see in which bits the exponent is stored by looking at a, 2*a, 4*a

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

void printBits(const double a)
{
  cout << bitset<64>(*reinterpret_cast<const uint64_t*>(&a)) << endl;
}

int main() 
{
    const double a = 3.14;
    printBits(a);
    printBits(-a);
    printBits(a*2);
    printBits(a*4);
    return 0;
}
0100000000001001000111101011100001010001111010111000010100011111
1100000000001001000111101011100001010001111010111000010100011111
0100000000011001000111101011100001010001111010111000010100011111
0100000000101001000111101011100001010001111010111000010100011111

Ideone