Hamming.hpp

/* SPDX-License-Identifier: MIT */
/* Copyright © 2021 Max Bachmann */
 
#pragma once
#include <limits>
#include <rapidfuzz/details/common.hpp>
#include <rapidfuzz/distance/Hamming_impl.hpp>
 
namespace rapidfuzz {
 
template <typename InputIt1, typename InputIt2>
size_t hamming_distance(InputIt1 first1, InputIt1 last1, InputIt2 first2, InputIt2 last2, bool pad_ = true,
                        size_t score_cutoff = std::numeric_limits<size_t>::max())
{
    return detail::Hamming::distance(first1, last1, first2, last2, pad_, score_cutoff, score_cutoff);
}
 
template <typename Sentence1, typename Sentence2>
size_t hamming_distance(const Sentence1& s1, const Sentence2& s2, bool pad_ = true,
                        size_t score_cutoff = std::numeric_limits<size_t>::max())
{
    return detail::Hamming::distance(s1, s2, pad_, score_cutoff, score_cutoff);
}
 
template <typename InputIt1, typename InputIt2>
size_t hamming_similarity(InputIt1 first1, InputIt1 last1, InputIt2 first2, InputIt2 last2, bool pad_ = true,
                          size_t score_cutoff = 0)
{
    return detail::Hamming::similarity(first1, last1, first2, last2, pad_, score_cutoff, score_cutoff);
}
 
template <typename Sentence1, typename Sentence2>
size_t hamming_similarity(const Sentence1& s1, const Sentence2& s2, bool pad_ = true, size_t score_cutoff = 0)
{
    return detail::Hamming::similarity(s1, s2, pad_, score_cutoff, score_cutoff);
}
 
template <typename InputIt1, typename InputIt2>
double hamming_normalized_distance(InputIt1 first1, InputIt1 last1, InputIt2 first2, InputIt2 last2,
                                   bool pad_ = true, double score_cutoff = 1.0)
{
    return detail::Hamming::normalized_distance(first1, last1, first2, last2, pad_, score_cutoff,
                                                score_cutoff);
}
 
template <typename Sentence1, typename Sentence2>
double hamming_normalized_distance(const Sentence1& s1, const Sentence2& s2, bool pad_ = true,
                                   double score_cutoff = 1.0)
{
    return detail::Hamming::normalized_distance(s1, s2, pad_, score_cutoff, score_cutoff);
}
 
template <typename InputIt1, typename InputIt2>
Editops hamming_editops(InputIt1 first1, InputIt1 last1, InputIt2 first2, InputIt2 last2, bool pad_ = true,
                        size_t score_hint = std::numeric_limits<size_t>::max())
{
    return detail::hamming_editops(detail::make_range(first1, last1), detail::make_range(first2, last2), pad_,
                                   score_hint);
}
 
template <typename Sentence1, typename Sentence2>
Editops hamming_editops(const Sentence1& s1, const Sentence2& s2, bool pad_ = true,
                        size_t score_hint = std::numeric_limits<size_t>::max())
{
    return detail::hamming_editops(detail::make_range(s1), detail::make_range(s2), pad_, score_hint);
}
 
template <typename InputIt1, typename InputIt2>
double hamming_normalized_similarity(InputIt1 first1, InputIt1 last1, InputIt2 first2, InputIt2 last2,
                                     bool pad_ = true, double score_cutoff = 0.0)
{
    return detail::Hamming::normalized_similarity(first1, last1, first2, last2, pad_, score_cutoff,
                                                  score_cutoff);
}
 
template <typename Sentence1, typename Sentence2>
double hamming_normalized_similarity(const Sentence1& s1, const Sentence2& s2, bool pad_ = true,
                                     double score_cutoff = 0.0)
{
    return detail::Hamming::normalized_similarity(s1, s2, pad_, score_cutoff, score_cutoff);
}
 
template <typename CharT1>
struct CachedHamming : public detail::CachedDistanceBase<CachedHamming<CharT1>, size_t, 0,
                                                         std::numeric_limits<int64_t>::max()> {
    template <typename Sentence1>
    explicit CachedHamming(const Sentence1& s1_, bool pad_ = true)
        : CachedHamming(detail::to_begin(s1_), detail::to_end(s1_), pad_)
    {}
 
    template <typename InputIt1>
    CachedHamming(InputIt1 first1, InputIt1 last1, bool pad_ = true) : s1(first1, last1), pad(pad_)
    {}
 
private:
    friend detail::CachedDistanceBase<CachedHamming<CharT1>, size_t, 0, std::numeric_limits<int64_t>::max()>;
    friend detail::CachedNormalizedMetricBase<CachedHamming<CharT1>>;
 
    template <typename InputIt2>
    size_t maximum(const detail::Range<InputIt2>& s2) const
    {
        return std::max(s1.size(), s2.size());
    }
 
    template <typename InputIt2>
    size_t _distance(const detail::Range<InputIt2>& s2, size_t score_cutoff, size_t score_hint) const
    {
        return detail::Hamming::distance(s1, s2, pad, score_cutoff, score_hint);
    }
 
    std::vector<CharT1> s1;
    bool pad;
};
 
#ifdef RAPIDFUZZ_DEDUCTION_GUIDES
template <typename Sentence1>
explicit CachedHamming(const Sentence1& s1_, bool pad_ = true) -> CachedHamming<char_type<Sentence1>>;
 
template <typename InputIt1>
CachedHamming(InputIt1 first1, InputIt1 last1, bool pad_ = true) -> CachedHamming<iter_value_t<InputIt1>>;
#endif
 
} // namespace rapidfuzz