Sign in to follow this  

RSA and Elgamal signatures

Recommended Posts

Hi. I'm looking for simple RSA and Elgamal sources in c# or c++. Everything i've found on the net was uncommented or just has miles of code. It's for my friend so i don't really need (and don't want) to start from scratch, plus program should be done in few days. Thanx for your help.

Share this post

Link to post
Share on other sites
RSA itself isn't that much code.
The problem is the large numbers that you need to use to make it secure.
I'd look for a big-num library that can handle the number of bits that you need, it probably already got the functions needed for RSA if not there's only a few that need to be implemented (and they are not that complicated).
The "problem" with all the implementations out there is that they are based on some quite heavy-weight big-num libraries (often written in C with a C++ wrapper).
I actually wrote my own library in a few days, only supporting basic operations and only unsigned numbers (wich is fine for RSA).

I'm sorry but I'm not allowed to give away my code, but it's got quite a few dependecies on the rest of the code base so it's not that suitable anyway.

But my recommendation is to search for a light-weight big-num libary instead (maybe C# have one built in?).

Once you have that the RSA part is fairly easy, this is my implementation:

class Rsa
template <class T> struct Key
T m_exponent;
T m_modulo;
template <class T> static void genKeys(Key<T>& publicKey, Key<T>& privateKey, Random* const rng = null)
static const unsigned int maxBits = (sizeof(T) * 4);
static const unsigned int minBits = ((sizeof(T) * 33) >> 4);
static const unsigned int eBits = ((minBits + 3) >> 2) + 1;
T e = PrimeSearch::find<T>(eBits, maxBits, rng);
T p;
do {
p = PrimeSearch::find<T>(minBits, maxBits, rng);
}while ((p % e) == T(1));
unsigned int bits = p.highestUsedBit();
T q;
do {
q = PrimeSearch::find<T>(maxBits - bits + 1, maxBits - bits + 24, rng);
}while (((q % e) == T(1)) && ((p * q).highestUsedBit() <= maxBits) && ((p * q).highestUsedBit() >= (maxBits + 24)));
T n = p * q;
T m = (q - T(1)) * (p - T(1));
T d = e.inv(m);
T nd = m - d;
T test = q + T(7);
T k = test.powModN(e, n);
T de = k.powModN(d, n);
if (de != test)
d = nd;
publicKey.m_exponent = e;
publicKey.m_modulo = n;
privateKey.m_exponent = d;
privateKey.m_modulo = n;
template <class T> static T crypt(const T& m, const Key<T>& key)
return (m + (key.m_modulo >> T(2))).powModN(key.m_exponent, key.m_modulo);
template <class T> static T decrypt(const T& m, const Key<T>& key)
return m.powModN(key.m_exponent, key.m_modulo) - (key.m_modulo >> T(2));

Share this post

Link to post
Share on other sites
Check out the Crypto Toolset for Python. I can't recall, but I guess you'll find the C sources there as well.

Share this post

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this