Notes of deprecation: There exist much more general and safer concepts in the standard library like std::byte, std::any, or std::variant

BIts TO character Pack (bitop) Library

C++ lacks of dynamic type casting, which represents a difficulty when dealing with unknown types at compile time. E.g. do you want to process data, which can have various types (int, float, double,...) and want a general interface to process that data? Then bitop is the library of your choice.

You can save any built in data type:

int i;
double d;
float f;

charPack cp_i(i);
charPack cp_d(d);
charPack cp_f(f);

cp_i.asInt() == i;            // evaluates to true
cp_d.asDouble_IEEE754() == d; // evaluates to true
cp_f.asFloat_IEEE754() == f;  // evaluates to true

The library assumes the usage of IEEE754 floating point numbers.
Moreover you can save data of unknown type if you know the size:

int i;
double d;
float f;

void* p_i = &i;
void* p_d = &d;
void* p_f = &f;

charPack cp_i(p_i, sizeof(int)*8);    // size in bits
charPack cp_d(p_d, sizeof(double)*8); // size in bits
charPack cp_f(p_f, sizeof(float)*8);  // size in bits

cp_i.asInt() == i;            // evaluates to true
cp_d.asDouble_IEEE754() == d; // evaluates to true
cp_f.asFloat_IEEE754() == f;  // evaluates to true

The given class can save bits from a certain memory location and converts the bits to a vector of characters to save them internally. You can access every bit of the vector<unsigned char>, where the highest index denotes the highest byte, e.g. you save 16 bits and convert them to an integer:

Representation	Values	Values
Memory	`0 1 0 1 0 0 0 1`	`1 0 0 0 0 1 0 1`
Vector element	[1]	[0]
If converted to int	2^14 + 2^12 + 2^8 +	2^7 + 2^2 + 2^0

The code comments obey doxygen rules, thus building your own documentation is recommended.

codecircuit/bitop

BIts TO character Pack (bitop) Library