Set and get multiple information in bit format

Question

I have these bits 0010 1110 0101 0111.

Currently the value of bits from 7 to 11 (right to left) is 10011. I want to set it to 10110 for example. How do I do that?

Why unsigned long long? `std::vector` would work. Or I believe there is a standard library bit field. — Weak to Enuma Elish, Sep 15 '15 at 01:07
Boolean Arithmetic and uint64_t. You will need to mask bits and shift them (either into position or shift them to the right for normal processing). — Thomas Matthews, Sep 15 '15 at 01:12
I don't understand, are you asking how to mask bits and extract them? (The information is already on the internet.) — Thomas Matthews, Sep 15 '15 at 01:13
Are you asking someone to write the code for you, or did you have a specific issue you encountered along the way? What have you tried? — Weak to Enuma Elish, Sep 15 '15 at 01:14
You could also consider [bit fields](http://en.cppreference.com/w/cpp/language/bit_field).... — Tony Delroy, Sep 15 '15 at 01:18
Remember, you are trading executable speed for data size here. Extracting bits takes executable time. Storing as uncompressed fields in a structure will improve your execution time, but waste more bits. — Thomas Matthews, Sep 15 '15 at 01:22

score 2 · Answer 1 · answered Sep 15 '15 at 01:21

This topic is common in the world of embedded systems. Usually, manufacturers of hardware devices use bit fields to represent information, such as statuses.

Inserting Into Your Number

This involves left shifting your number (such as birth year) into the appropriate position then ORing the value with your number:

unsigned int value;
//...
value |= (birth_year << 1);

Extracting or Getting the Number:

You will need to AND the number with a mask so that only the important bits are extracted. For example, retrieving gender:

unsigned int gender;
unsigned int value;
gender = value & 1;
// or 
gender = value & (~0);

You may need to right shift the bits to get the correct value, such as after extracting the birth year, right shift it by 1.

Bit Field Structure

You can let the Compiler figure all this by using bit fields in a structure, something like:

struct Compressed_Number
{
  unsigned int  gender : 1;
  unsigned int  birth_year : 11;
  //..
};

I personally prefer the Boolean Arithmetic version because you always know the bit positions.

To be complete, you need to mask out the field that you're inserting, so something like: `value = (value & ((~((1< — Russ Schultz, Sep 15 '15 at 02:23

Set and get multiple information in bit format

1 Answers1

Inserting Into Your Number

Extracting or Getting the Number:

Bit Field Structure