https://github.com/dkosmari/libxint

libxint is a C++20 headers-only library for fixed-width large integers.
https://github.com/dkosmari/libxint

cpp cpp20

Last synced: 5 days ago
JSON representation

libxint is a C++20 headers-only library for fixed-width large integers.

• Host: GitHub
• URL: https://github.com/dkosmari/libxint
• Owner: dkosmari
• License: gpl-3.0
• Created: 2023-03-20T14:32:49.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-04-01T03:20:26.000Z (8 months ago)
• Last Synced: 2024-04-02T03:50:55.244Z (8 months ago)
• Topics: cpp, cpp20
• Language: C++
• Homepage:
• Size: 485 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: ChangeLog
  • License: COPYING
  • Authors: AUTHORS

Awesome Lists containing this project

README

        
libxint - a C++ library for extended size unsigned integers

===========================================================

This is a headers-only C++20 library to emulate large fixed-size integers.

It's licensed under the GPLv3+ license.

Installation

------------

When building from a tarball,  you can skip step 0.

  0. `./bootstrap`

  1. `./configure`

  2. `make`

  3. `sudo make install`

To uninstall:

    sudo make uninstall

This is a standard Automake package; run `./configure --help` or check `INSTALL` for more

details.

Usage

-----

First, include the header `libxint/uint.hpp` in your program.

The type `xint::uint` is now available; the template argument `B` is the number

of desired bits. It can be used as any ordinary unsigned integer:


```cpp

#include 

#include 

using namespace std;

using namespace xint::literals;

int main()

{

    xint::uint<1024> x = 42;

    cout << x << endl;

    x <<= 1000;

    cout << x << endl;

    ++x;

    cout << x << endl;

    cout << popcount(x) << endl;

    auto y = 0xdeadbeef0000babecafe0000dec0de_uint;

    cout << hex << y << dec << " has " << y.num_bits << " bits." << endl;

}

```

Implementation details

----------------------

This extended integer is implemented through multiple native integers, called limbs.

The size of each limb can be controlled by defining the `XINT_LIMB_SIZE` macro. Valid

sizes are 8, 16 and 32 (default is 32.) For instance:

    g++ -std=c++20 -DXINT_LIMB_SIZE=16 my-program.cpp

The limbs can be accessed directly through the `limbs()` and `limb(i)` member functions.

NOTE: it's not possible to specify a bit size that is not a multiple of the limb size. No

attempt is made to mask out the "excess" bits (for performance reasons) so the results of

all operations will be inconsistent. A `static_assert()` exists to prevent invalid limb

sizes.