Skip to content

Latest commit

 

History

History
executable file
·
86 lines (65 loc) · 3.33 KB

README.md

File metadata and controls

executable file
·
86 lines (65 loc) · 3.33 KB
Raw

integer

A C++ Arbitrary Precision Integer Implementation

Copyright (c) 2013 - 2017 Jason Lee @ calccrypto at gmail.com

Please see LICENSE file for license.

Build Status

With much help from:

  • Auston Sterling - Initial debugging and coding help and FFT multiplication
  • Corbin @ Code Review (StackExchange) - wrote a sizeable chunk of code and suggestions
  • Keith Nicholas @ Code Review (StackExchange)
  • mugwort-rc - tons of edits
  • ROBOKITTY @ Code Review (StackExchange)
  • Winston Ewert @ Code Review (StackExchange) - suggested many improvements

Although this class can theoretically hold an infinitely large value, the actual limit of how large the integers can be is std::deque <Z>().max_size() * sizeof(Z) * 8 bits, which should be more than enough for any purpose.

Usage:

#include <iostream>
#include "integer.h"

int main(){
    integer a = -1234;         // standard integer input
    integer b(  "5678",  16);  // string input (no base prefixes: '0b' or '0x')
    integer c( "-9abC",  16);  // hex input cases can be mixed
    integer d("Hello!", 256);  // ASCII strings can be used too!

    integer sum = a + b + c + d + 1;           // 79600447923724 (decimal)

    std::cout << sum             << std::endl; // print directly
    std::cout << sum.str(16)     << std::endl; // print with a specific base
    std::cout << std::hex << sum << std::endl; // print using std::oct, std::dec, and std::hex

    return 0;
}

Compilation

c++ <some arguments> integer.cpp <other arguments>

  • g++ and clang++ should both work
  • C++11 is required

Internal Operations

  • Data is stored in big-endian, so value[0] is the most significant digit, and value[value.size() - 1] is the least significant digit.

    • By default, the container holding the value is std::deque <uint8_t>.

    • Changing the internal representation to a std::string makes integer run slower than using a std::deque <uint8_t>

  • Negative values are stored as their positive value, with a bool that says the value is negative.

  • Arithmetic (+, -, *, /, %) and Comparison (>, >=, <, <=, ==, !=) operators operate on the absolute values of the operands from within private functions. Signs are handled in the operators themselves.

    • Multiple algorithms for subtraction, multiplication, and division have been implemented and commented out. They should all work, but are there for educational purposes. If one is faster than another on different systems, by all means change which algorithm is used. Just make sure all related functions are changed as well.

  • Bitwise operations on negative values will use the two's complement version of the number.

Notes

  • Conversions for strings of values in bases [2, 16] and 256 are provided. If others are required, add the conversions to integer::integer(Iterator, const Iterator&, const uint16_t &) and std::string integer::str(const uint16_t &, const unsigned int &) const

  • The constructor using iterators allows for creating values from any integral base larger than 1.

  • Hexadecimal output strings use lowercase characters.

  • Base256 strings are assumed to be positive when read into integer. Use operator-() to negate the value.