poj 1935 journey 树DP

Journey

Time Limit: 1000MS		Memory Limit: 30000K
Total Submissions: 1425		Accepted: 523

Description

There are n cities in Byteland (numbered from 1 to n), connected by bidirectional roads. The king of Byteland is not very generous, so there are only n-1 roads, but they connect the cities in such a way that it is possible to travel from each city to any other city. 
One day, a traveller Byterider arrived in the city number k. He was planning to make a journey starting in the city k and visiting on his way cities m1 , m2 , ..., mj (not necessarily in this order) --the numbers m i are all different and they are also different from k. Byterider -- like every traveller -- has only a limited amount of money, so he would like to visit all the cities that he has planned to visit using the shortest possible path (starting in the city k). A path is one road or a sequence of roads, where every next road begins in the city where the previous one ends. Help Byterider to determine the length of the shortest path for his journey.

Input

The first line of the standard input contains two integers n and k separated by a single space (2 <= n <= 50000, 1 <= k <= n), n is the number of cities in Byteland and k is the number of the first city on Byterider's path. Each of the following n-1 lines contains the description of one road in Byteland. Line (i + 1) (for 1 <= i <= n-1) contains three integers ai , bi and di separated by single spaces (1 <= ai ; bi <= n, 1 <= di <= 1000), ai and bi are the cities connected by the road, and di is the length of the road. Line (n + 1) contains one integer j -- the number of cities which Byterider would like to visit (1 <= j <= n-1). The next line contains j different integers m i separated by single spaces -- the numbers of the cities that Byterider would like to visit (1 <= mi <= n, mi != k).

Output

The first and only line of the standard output should contain exactly one integer: the length of the shortest path for Byterider's journey.

Sample Input

4 2
1 2 1
4 2 2
2 3 3
2
1 3

Sample Output

5

Source

CEOI 2004 sample task

#include<iostream>
#include<cstring>
#include<cstdio>
using namespace std;


int head[50005];
int next[100005];
int point[100005];
int weight[100005];


struct node
{
int dis; //存储访问该子树但不最终返回根所需长度
int dis_and_return; //存储访问该子树并最终返回根所需长度
bool is_target; //存储该节点是否为目标
bool have_target; //存储该节点为根的子树中是否包含目标
};


node nd[50005];


int n,k;
bool fl[50005];


bool dfs(int v)
{
int ret=0;
fl[v]=1;
nd[v].dis=0x1f1f1f1f; //为找最小，初始化成最大值
nd[v].dis_and_return=0; //初始化成0
for(int e=head[v];e!=-1;e=next[e])
{
if(fl[point[e]])continue;
if(!dfs(point[e]))
{
continue;
}
nd[v].dis_and_return+=(weight[e]*2+nd[point[e]].dis_and_return);
}
if(nd[v].dis_and_return==0) //为0说明所有子树都不包含目标
{
nd[v].dis=0; //将该子树的dis赋为0
if(nd[v].is_target==1) //如果该节点为目标，则返回1，否则返回0
{
return 1;
}
else
{
return 0;
}
}
nd[v].have_target=1; //该子树中有目标点
for(int e=head[v];e!=-1;e=next[e])
{
if(nd[point[e]].have_target==1&&nd[v].dis_and_return-nd[point[e]].dis_and_return-weight[e]+nd[point[e]].dis<nd[v].dis)
{
nd[v].dis=nd[v].dis_and_return-nd[point[e]].dis_and_return-weight[e]+nd[point[e]].dis; //寻找该子树的dis_and_return
}
}
return 1;
}


int main()
{
while(~scanf("%d%d",&n,&k))
{
memset(head,-1,sizeof(head));
memset(next,-1,sizeof(next));
memset(point,-1,sizeof(point));
memset(weight,-1,sizeof(weight));
memset(fl,0,sizeof(fl));
for(int i=1;i<n;i++)
{
int a,b,w;
scanf("%d%d%d",&a,&b,&w);
next[i]=head[a];
point[i]=b;
weight[i]=w;
head[a]=i;


next[i+n]=head[b];
point[i+n]=a;
weight[i+n]=w;
head[b]=i+n;
}
int j;
scanf("%d",&j);
for(int i=1;i<=n;i++)
{
nd[i].have_target=0;
nd[i].is_target=0;
}
for(int i=1;i<=j;i++)
{
int tar;
scanf("%d",&tar);
nd[tar].is_target=1;
nd[tar].have_target=1;
}
dfs(k);
printf("%d\n",nd[k].dis);
}
return 0;
}