在这里，邻接表的实现与深度优先遍历图，使用递归。

#include<iostream>

using namespace std;

#define VERTEXNUM 5//结点数

struct  edgenode

{

    int to;

    int weight; // 边的权值

    edgenode *next;

};

struct vnode

{

    int from;

    edgenode *first;

};

void createGraph(vnode *adjilist, int start, int end,int weight);

void displayGraph(vnode *adjilist,int nodeNum);

void DFT(vnode *adjilist,int* vertexStatusArr,int nodeNum);

void DFTcore(vnode *adjilist,int i,int* vertexStatusArr);

int main(void){

		//创建图

	    vnode adjilist[VERTEXNUM];

		int nodeNum=VERTEXNUM;

		for(int i=0;i<nodeNum;i++)

			adjilist[i].first=NULL;

		int vertexStatusArr[VERTEXNUM]={0};

		cout<<vertexStatusArr[4]<<endl;

        createGraph(adjilist,0,3,0);

        createGraph(adjilist,0,4,0);

        createGraph(adjilist,3,1,0);

        createGraph(adjilist,3,2,0);

        createGraph(adjilist,4,1,0);

        printf("after create:\n");

        displayGraph(adjilist,nodeNum);

		//深度优先遍历

        DFT(adjilist,vertexStatusArr,nodeNum);

		system("pause");

        return 0;

}

//创建图,採取前插法

void createGraph(vnode *adjilist, int start, int end,int weight)

{

	adjilist[start].from=start;

	edgenode *p=new edgenode;

	p->to=end;

	p->weight=weight;

	p->next=adjilist[start].first;

	adjilist[start].first=p;

}

//打印存储的图

void displayGraph(vnode *adjilist,int nodeNum)

{

    int i,j;

	edgenode *p;

    for(i=0;i<nodeNum;i++)

	{

		p=adjilist[i].first;

		while(p)

		{

			cout<<'('<<adjilist[i].from<<','<<p->to<<')'<<endl;

			p=p->next;

		}

     }

}

//深度优先遍历

void DFT(vnode *adjilist, int* vertexStatusArr,int nodeNum)

{

        printf("start BFT graph:\n");

        int i;

        for(i=0;i<nodeNum;i++){

                DFTcore(adjilist,i,vertexStatusArr);

        }

        printf("\n");

}

void DFTcore(vnode *adjilist,int i,int* vertexStatusArr)

{

    if(vertexStatusArr[i] == 1)

        return;

    printf("%d ",i);

    vertexStatusArr[i] = 1;

	edgenode *p;

    for(p=adjilist[i].first;p;p=p->next)

	{

        DFTcore(adjilist, p->to, vertexStatusArr);

    }

}

时间复杂度O(V+E)。