使用了并查集+优先队列.

具体解释等周末再描述 :shuijiao:

以下是代码:

import java.util.*;

public class Kruskal {
private int[] points;
private void initPoints(int n){
points = new int[n+1];
for(int i = 0;i<=n;i++){
points[i] = i;
}
}

private void union(int a ,int b){
    int A = find(a);
    int B = find(b);
    if(A != B){
        points[A] = B;
    }
}
private boolean connect(int a,int b){
    int A = find(a);
    int B = find(b);
    return A==B;
}

private int find(int a){
    if(points[a] == a){
        return points[a];
    }
    else{
        return points[a] = find(points[a]);
    }
}

private void doKruskal(int pointCnt,int[][] edges){
    PriorityQueue<int[]> pq = new PriorityQueue<>(new Comparator<int[]>() {
        @Override
        public int compare(int[] ints, int[] t1) {
            return ints[2]-t1[2];
        }
    });
    List<int[]> newEdges = new ArrayList<>();
    Collections.addAll(pq, edges);
    initPoints(pointCnt);
    int[] minWeightEdge;
    while(!pq.isEmpty()){
        minWeightEdge = pq.poll();
        if(!connect(minWeightEdge[0],minWeightEdge[1])){
            union(minWeightEdge[0],minWeightEdge[1]);
            newEdges.add(new int[]{minWeightEdge[0],minWeightEdge[1]});
        }
    }
    for(int[] newEdge:newEdges){
        System.out.println(Arrays.toString(newEdge));
    }
}

public static void main(String[] args) {
    Kruskal k = new Kruskal();
    int[][] edges = new int[][]{
            {1,2,12},
            {1,4,16},
            {1,3,14},
            {2,4,7},
            {2,5,10},
            {3,4,9},
            {3,6,8},
            {4,5,6},
            {4,6,2},
            {5,6,5},
            {5,7,3},
            {6,7,4}
    };
    k.doKruskal(7,edges);
}

}