Question

#define CHAR_BIT 8
union
{
     float input;   // assumes sizeof(float) == sizeof(int)
     int   output;
}    data;

data.input = 122.5;

bitset<sizeof(float) * CHAR_BIT>   bits(data.output);

int ieee[32];
for(int i = 0 ; i < 32 ; ++i){
    ieee[i] = (int)bits[i];
}

My intention is to fill the ieee array with the IEEE representation of a float, and is done (I’ve used code from another question), but there are 2 things I don’t understand:

1) Why do I have to use #define CHAR_BIT 8 for a correct output?

2) How can I fill the ieee array with the correct bit values?

Answer 1

1) Because you need to convert a size in bytes (that’s what sizeof(float) is) to a size in bits (which is what bitset expects).

2) Looks to me like you are already putting the correct bit value into ieee. Can you give an example and say why you think it’s wrong.