minhashes.hpp

Go to the documentation of this file.

// SPDX-FileCopyrightText: 2006-2024, Knut Reinert & Freie Universität Berlin
// SPDX-FileCopyrightText: 2016-2024, Knut Reinert & MPI für molekulare Genetik
// SPDX-License-Identifier: BSD-3-Clause
 
#pragma once
 
#include <cstddef> // for size_t
#include <cstdint> // for uint64_t, uint32_t
#include <span>    // for span
#include <vector>  // for vector
 
#include <cereal/access.hpp> // for access
#include <cereal/cereal.hpp> // for make_nvp, CEREAL_NVP
 
#include <hibf/misc/md_vector.hpp>
#include <hibf/platform.hpp>
 
namespace seqan::hibf::sketch
{
 
struct minhashes
{
    static constexpr uint64_t register_id_mask{15}; 
    static constexpr size_t num_sketches{16};
    static constexpr size_t sketch_size{40};
 
    md_vector<uint64_t> table{}; // Each element (vector<uint64_t>) is a minhash.
 
    minhashes() = default;                              
    minhashes(minhashes const &) = default;             
    minhashes & operator=(minhashes const &) = default; 
    minhashes(minhashes &&) = default;                  
    minhashes & operator=(minhashes &&) = default;      
    ~minhashes() = default;                             
    minhashes(std::vector<uint64_t> const & smallest_values);
 
    bool is_valid() const;
 
    void fill_incomplete_sketches(std::span<uint64_t> const & more_smallest_values);
 
    static void push_to_heap_if_smaller(uint64_t const value, std::vector<uint64_t> & heap);
 
private:
    friend class cereal::access;
 
    template <typename archive_t>
    void serialize(archive_t & archive)
    {
        uint32_t version{1};
        archive(CEREAL_NVP(version));
 
        // other members are const static currently
        archive(CEREAL_NVP(table));
    }
};
 
} // namespace seqan::hibf::sketch