README.md

Randomized rolling hash functions in C++

License: Apache 2.0

What is this?

This is a set of C++ classes implementing various recursive n-gram hashing techniques, also called rolling hashing (http://en.wikipedia.org/wiki/Rolling_hash), including:

• Randomized Karp-Rabin (sometimes called Rabin-Karp)
• Hashing by Cyclic Polynomials (also known as Buzhash)
• Hashing by Irreducible Polynomials

This library is used by khmer: the in-memory nucleotide sequence k-mer engine.

These are randomized hash functions, meaning that each time you create a new hasher instance, you will get new hash values for a given input.

Code sample

    const uint n(3);//hash all sequences of 3 characters
    const uint L(7); // you need 7 bits
    CyclicHash<uint32> hf(n,L );// if you want 64-bit values replace uint32 by uint64
    for(uint32 k = 0; k<n;++k) {
              chartype c = ... ; // grab some character
              hf.eat(c); // feed it to the hasher
    }
    while(...) { // go over your string
       hf.hashvalue; // at all times, this contains the hash value
       chartype c = ... ;// point to the next character
       chartype out = ...; // character we want to forget
       hf.update(out,c); // update hash value
    }
    hf.reset(); // you can now hash a new string

Requirements

A recent GNU GCC C++ compiler or a recent CLANG.

What should I do after I download it?

type:

    make

then

    ./unit

then

    ./speedtesting

Nim version

See Cyclic-Polynomial-Hash for a similar library written in Nim.

References

• Daniel Lemire, Owen Kaser: Recursive n-gram hashing is pairwise independent, at best, Computer Speech & Language, Volume 24, Issue 4, October 2010, Pages 698-710 http://arxiv.org/abs/0705.4676
• Daniel Lemire, The universality of iterated hashing over variable-length strings, Discrete Applied Mathematics 160 (4-5), 2012. http://arxiv.org/abs/1008.1715
• Owen Kaser and Daniel Lemire, Strongly universal string hashing is fast, Computer Journal (2014) 57 (11): 1624-1638. http://arxiv.org/abs/1202.4961

This work has been used in genomics, see

• Ilia Minkin, Son Pham, Paul Medvedev, TwoPaCo: an efficient algorithm to build the compacted de Bruijn graph from many complete genomes, Bioinformatics (to appear). https://doi.org/10.1093/bioinformatics/btw609 and http://github.com/medvedevgroup/TwoPaCo