最小生成树算法——Kruskal

Kruskal算法的原理是先将图中的所有边按照权从小到大排序，然后循环取边，判断添加上该边后是子图中否有闭合回路，如果没有，则添加该边，否则舍弃该边。直到所有的边都遍历一遍。我认为该算法的核心是排序和判断闭合。
下面是具体的代码：

#include <iostream>
#include <vector>
#include <algorithm>
#include <set>

struct path {
    int start ;
    int end ;
    int weight ;
};

static bool compare( path pi, path pj){
    return pi.weight < pj.weight;
}

static int find( const std::vector <int>& array, int i){
    while(array[i])
        i = array[i];

    return i;
}

int main(){
    std:: vector<int> vertexs;
    std:: vector<path > paths;

    std::cout << "请输入顶点数：" << std::flush;
    int sum; std::cin >> sum;
    int tmp;
    for(int i = 0; i != sum; ++i){
        vertexs.push_back(tmp);
    }

    std::cout << "请输入边数：" << std::flush;
    std::cin >> sum;

    for(int i = 0; i != sum; ++i){
        path p;
        std::cin >> p. start >> p.end >> p.weight;
        paths.push_back(p);
    }

    std::sort(paths.begin(), paths.end(), compare);

    std:: vector<int> array(vertexs.size(), 0);

    for(size_t i = 0; i != paths.size(); ++i){
        int m = find(array, paths[i].start );
        int n = find(array, paths[i].end );

        if(m != n){
            array[m] = n;
            std::cout << paths[i].start << "->" << paths[i]. end << std::endl;
        }
    }
}

本文链接：http://blog.csdn.net/girlkoo/article/details/17435951
本文作者：girlkoo