C ++:使用一对(cpp_int，int)整数作为unordered_map中的键(其中cpp_int是升压多精度整数) [英] C++: Using pair of (cpp_int, int) integers as key in unordered_map (where cpp_int are boost multiprecision integers)

问题描述

对于我的无序地图，我想用作(cpp_int, int)的键对，其中cpp_int是

For my unordered map I would like to use as keys pairs of (cpp_int, int) where cpp_int are boost multiprecision integers:

#include <boost/multiprecision/cpp_int.hpp>
#include <unordered_map>

using boost::multiprecision::cpp_int;

std::unordered_map<std::pair<cpp_int, int>, double> myMap

在此站点上搜索时，我发现了很多关于将std::pair<int,int>的自定义哈希函数用作键的建议，但是我找不到如何处理std::pair<cpp_int, int>的方法.

Searching on this site I have found many suggestions of using a custom hash function for std::pair<int,int> as keys, but I couldn't found how to deal with std::pair<cpp_int, int>.

更新:为澄清起见，我尝试了在网络上找到的哈希函数(用于(int，int):

Update: To clarify, I have tried a hash function I found on the web (for (int,int):

#include <boost/multiprecision/cpp_int.hpp>
#include <unordered_map>
using boost::multiprecision::cpp_int;
typedef std::pair<cpp_int, int> MyPair;

struct MyHash {
public:
        size_t operator()(MyPair x) const throw() {
             size_t h = x.first * 1 + x.second * 100000;
             return h;
        }
};

void function()
{
    std::unordered_map<MyPair, double, MyHash> M;
}

无法编译:

error: cannot convert ‘boost::enable_if_c<true, boost::multiprecision::detail::expression<boost::multiprecision::detail::multiply_add, boost::multiprecision::detail::expression<boost::multiprecision::detail::terminal, boost::multiprecision::number<boost::multiprecision::backends::cpp_int_backend<> >, void, void, void>, boost::multiprecision::detail::expression<boost::multiprecision::detail::terminal, int, void, void, void>, int, void> >::type {aka boost::multiprecision::detail::expression<boost::multiprecision::detail::multiply_add, boost::multiprecision::detail::expression<boost::multiprecision::detail::terminal, boost::multiprecision::number<boost::multiprecision::backends::cpp_int_backend<> >, void, void, void>, boost::multiprecision::detail::expression<boost::multiprecision::detail::terminal, int, void, void, void>, int, void>}’ to ‘size_t {aka long unsigned int}’ in initialization
              size_t h = x.first * 1 + x.second * 100000;
                                                  ^

我的问题是:如何使用(cpp_int，int)对作为unordered_map中的键?

非常感谢您！

更新2: 感谢@sehe向我指出了他的回答(其中他为cpp_int提供了哈希函数).结合此答案(它显示了如何将两个哈希函数组合为一对)，我提出了以下内容解决方案(编译正常，我需要对我的问题进行测试以查看其是否有效):

Update 2: Thanks to @sehe for pointing me to his answer (in which he provided a hash function for cpp_int). Combining with this answer (which shows how to combine two hash functions for a pair), I've come up with the following solution (it compiles fine, I'll need to test on my problem to see if it works):

#include <boost/archive/binary_oarchive.hpp>
#include <boost/multiprecision/cpp_int.hpp>
#include <boost/multiprecision/cpp_int/serialize.hpp>
#include <boost/iostreams/device/back_inserter.hpp>
#include <boost/iostreams/stream_buffer.hpp>
#include <boost/iostreams/stream.hpp>

#include <boost/functional/hash.hpp>
#include <boost/multiprecision/cpp_int.hpp>
#include <unordered_map>


using boost::multiprecision::cpp_int;

typedef std::pair<cpp_int, int> MyPair;



namespace mp_hashing {
    namespace io = boost::iostreams;

    struct hash_sink {
        hash_sink(size_t& seed_ref) : _ptr(&seed_ref) {}

        typedef char         char_type;
        typedef io::sink_tag category;

        std::streamsize write(const char* s, std::streamsize n) {
            boost::hash_combine(*_ptr, boost::hash_range(s, s+n));
            return n;
        }
      private:
        size_t* _ptr;
    };

    template <typename T> struct hash_impl {
        size_t operator()(T const& v) const {
            using namespace boost;
            size_t seed = 0;
            {
                iostreams::stream<hash_sink> os(seed);
                archive::binary_oarchive oa(os, archive::no_header | archive::no_codecvt);
                oa << v;
            }
            return seed;
        }
    };
}

namespace std {
    template <typename backend>
    struct hash<boost::multiprecision::number<backend> >
        : mp_hashing::hash_impl<boost::multiprecision::number<backend> >
    {};
}


struct pair_hash {
    template <class T1, class T2>
    std::size_t operator () (const std::pair<T1,T2> &p) const {
        auto h1 = std::hash<T1>{}(p.first);
        auto h2 = std::hash<T2>{}(p.second);

        // Mainly for demonstration purposes, i.e. works but is overly simple
        // In the real world, use sth. like boost.hash_combine
        return h1 ^ h2;
    }
};

void function()
    {
        std::unordered_map<MyPair, double, pair_hash> M;
    }

推荐答案

是的，您采用了我之前提供的Multiprecision哈希，并为std::pair添加了哈希.我不喜欢手动推散列组合(好的普通散列组合并非微不足道).

Yes, you took the Multiprecision hash I contributed earlier and added the hash for std::pair. I'm not a fan of handrolling the hash combination (good general hash combination is not trivial).

所以我要对boost::hash_combine做同样的事情:

So I'd do the same with boost::hash_combine:

template <typename K, typename V>
struct hash<std::pair<K, V> > 
{
    size_t operator()(std::pair<K, V> const& pair) const {
        size_t seed = std::hash<K>{}(pair.first);
        boost::hash_combine(seed, pair.second);
        return seed;
    }
};

在MSVC RexTester上直播

#include <iostream>
#include <iomanip>

#include <boost/archive/binary_oarchive.hpp>
#include <boost/multiprecision/cpp_int.hpp>
#include <boost/multiprecision/cpp_int/serialize.hpp>
#include <boost/iostreams/device/back_inserter.hpp>
#include <boost/iostreams/stream_buffer.hpp>
#include <boost/iostreams/stream.hpp>

#include <boost/functional/hash.hpp>

namespace mp_hashing {
    namespace io = boost::iostreams;

    struct hash_sink {
        hash_sink(size_t& seed_ref) : _ptr(&seed_ref) {}

        typedef char         char_type;
        typedef io::sink_tag category;

        std::streamsize write(const char* s, std::streamsize n) {
            boost::hash_combine(*_ptr, boost::hash_range(s, s+n));
            return n;
        }
      private:
        size_t* _ptr;
    };

    template <typename T> struct hash_impl {
        size_t operator()(T const& v) const {
            using namespace boost;
            size_t seed = 0;
            {
                iostreams::stream<hash_sink> os(seed);
                archive::binary_oarchive oa(os, archive::no_header | archive::no_codecvt);
                oa << v;
            }
            return seed;
        }
    };
}

#include <unordered_map>
#include <boost/unordered_map.hpp>

namespace std {
    template <typename backend> 
    struct hash<boost::multiprecision::number<backend> > 
        : mp_hashing::hash_impl<boost::multiprecision::number<backend> > 
    {};

    template <typename K, typename V>
    struct hash<std::pair<K, V> > 
    {
        size_t operator()(std::pair<K, V> const& pair) const {
            size_t seed = std::hash<K>{}(pair.first);
            boost::hash_combine(seed, pair.second);
            return seed;
        }
    };
}

int main() {
    using boost::multiprecision::cpp_int;

    std::unordered_map<std::pair<cpp_int, int>, int>  m {
        { { cpp_int(1) << 111, -1 }, 1 },
        { { cpp_int(2) << 222, -2 }, 2 },
        { { cpp_int(3) << 333, -3 }, 3 },
    };

    for (auto& p : m)
        std::cout << p.first.first << " -> " << p.second << "\n";
}

这篇关于C ++:使用一对(cpp_int，int)整数作为unordered_map中的键(其中cpp_int是升压多精度整数)的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！