factorpoly is a standalone single-variate polynomial factorizer, written in C++, based on code taken from factor-by-gnfs
Clone the code from Github:
git clone https://github.com/ChrisCGH/factorpoly.git
Change directory to factorpoly and build using make:
cd factorpoly
make
This should create the executable factorpoly in the same directory.
The supplied makefile is written for Linux, and assumes that the standard toolchain for building C++ programs is already installed (e.g. make, gcc, g++). In addition the code makes use of the GNU MP Bignum library, and assumes that the development library for GMP is installed. Building on other systems is left as an exercise, but the code compiles with g++ and -std=c++14, so should compile with a recent C++ compiler.
factorpoly can be built as a Docker image suitable for use with OpenFaas using the supplied Dockerfile. The latest image can be found here.
To run the standalone factorpoly simply read a polynomial from stdin:
$ echo "X^55 + 1" | ./factorpoly
(X + 1)(X^10 - X^9 + X^8 - X^7 + X^6 - X^5 + X^4 - X^3 + X^2 - X + 1)(X^4 - X^3 + X^2 - X + 1)(X^40 + X^39 - X^35 - X^34 + X^30 - X^28 - X^25 + X^23 + X^20 + X^17 - X^15 - X^12 + X^10 - X^6 - X^5 + X + 1)
If the environment variable USE_MATHJAX is set, then the output is an html page which uses mathjax to format the polynomial factors:
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width">
<title>factorpoly output</title>
<script type="text/javascript" async
src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.2/MathJax.js?config=TeX-MML-AM_CHTML">
</script>
</head>
<body>
<p>
$$(X + 1)(X^4 - X^3 + X^2 - X + 1)(X^{10} - X^9 + X^8 - X^7 + X^6 - X^5 + X^4 - X^3 + X^2 - X + 1)(X^{40} + X^{39} - X^{35} - X^{34} + X^{30} - X^{28} - X^{25} + X^{23} + X^{20} + X^{17} - X^{15} - X^{12} + X^{10} - X^6 - X^5 + X + 1)$$
</p>
</body>
</html>
factorpoly can be run as an OpenFaas function, by deploying the function from the image ccard/factorpoly. Once deployed, the function can be run e.g. by using curl:
curl http://localhost:8080/function/factorpoly -X POST --data-binary 'X^55 + 1'
The algorithm used by factorpoly is an implementation of Algorithm 3.5.7 in "A Course in Computational Algebraic Number Theory" by Henri Cohen (http://www.springer.com/gp/book/9783540556404)