http://codeforces.com/contest/1204/problem/C
The main characters have been omitted to be short.
You are given a directed unweighted graph without loops with nn vertexes and a path in it (that path is not necessary simple) given by a sequence p1,p2,…,pm of mm vertexes; for each 1≤i<m there is an arc from pipi to pi+1.
Define the sequence v1,v2,…,vk of k vertexes as good, if v is a subsequence of p, v1=p1, vk=pm, and p is one of the shortest paths passing through the vertexes v1, ……, vk in that order.
A sequence aa is a subsequence of a sequence bb if aa can be obtained from bb by deletion of several (possibly, zero or all) elements. It is obvious that the sequence pp is good but your task is to find the shortest good subsequence.
If there are multiple shortest good subsequences, output any of them.
Input
The first line contains a single integer nn (2≤n≤100) — the number of vertexes in a graph.
The next nn lines define the graph by an adjacency matrix: the jj-th character in the ii-st line is equal to 1 if there is an arc from vertex ii to the vertex jj else it is equal to 00. It is guaranteed that the graph doesn't contain loops.
The next line contains a single integer mm (2≤m≤10^6) — the number of vertexes in the path.
The next line contains mm integers p1,p2,…,pm (1≤pi≤n) — the sequence of vertexes in the path. It is guaranteed that for any 1≤i<m there is an arc from pipi to pi+1.
Output
In the first line output a single integer kk (2≤k≤m) — the length of the shortest good subsequence. In the second line output kk integers v1, ……, vk (1≤vi≤n) — the vertexes in the subsequence. If there are multiple shortest subsequences, print any. Any two consecutive numbers should be distinct.
Examples
input
Copy
4 0110 0010 0001 1000 4 1 2 3 4
output
Copy
3 1 2 4
input
Copy
4 0110 0010 1001 1000 20 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
output
Copy
11 1 2 4 2 4 2 4 2 4 2 4
input
Copy
3 011 101 110 7 1 2 3 1 3 2 1
output
Copy
7 1 2 3 1 3 2 1
input
Copy
4 0110 0001 0001 1000 3 1 2 4
output
Copy
2 1 4
Note
Below you can see the graph from the first example:
The given path is passing through vertexes 1, 2, 3, 4. The sequence 1−2−4 is good because it is the subsequence of the given path, its first and the last elements are equal to the first and the last elements of the given path respectively, and the shortest path passing through vertexes 1, 2 and 4 in that order is 1−2−3−4. Note that subsequences 1−4 and 1−3−4 aren't good because in both cases the shortest path passing through the vertexes of these sequences is 1−3−4.
In the third example, the graph is full so any sequence of vertexes in which any two consecutive elements are distinct defines a path consisting of the same number of vertexes.
In the fourth example, the paths 1−2−4 and 1−3−4 are the shortest paths passing through the vertexes 1 and 4.
题意:按照所给序列在图上走最短路,要求输出最短的子序列,使得在图上走的最短路不变(说的不太清楚,就是有点迷)
思路:因为n最大才100,先floyd预处理出图中任意两点之间的最短路,然后在序列上求最短路(即下标之差),如果和图中最短路相同,那就不需要限制,如果图中最短路比序列中的最短路小,说明没有按照序列顺序走,所以用前面那个点限制一下
#include<bits/stdc++.h>
#pragma GCC optimize(3)
#define max(a,b) a>b?a:b
using namespace std;
typedef long long ll;
const int N=1e6+5;
const int Inf=0x3f3f3f3f;
char mp[105][105];
int G[105][105];
int p[N];
int q[N];
int main(){
int n;
scanf("%d",&n);
for(int i=1;i<=n;i++){
scanf("%s",mp[i]+1);
}
for(int i=1;i<=n;i++){
for(int j=1;j<=n;j++){
if(i==j) G[i][j]=0;
else if(mp[i][j]=='1') G[i][j]=1;
else G[i][j]=Inf;
}
}
for(int k=1;k<=n;k++){
for(int i=1;i<=n;i++){
for(int j=1;j<=n;j++){
if(G[i][k]<Inf&&G[k][j]<Inf)
G[i][j]=min(G[i][j],G[i][k]+G[k][j]);
}
}
}
int m;
scanf("%d",&m);
for(int i=1;i<=m;i++) scanf("%d",&p[i]);
int ans=0;
q[++ans]=p[1];
int l=1,r=1;
while(r<=m){
if(r-l==G[p[l]][p[r]]) r++;
else{
q[++ans]=p[r-1];
l=r-1;
}
}
q[++ans]=p[m];
printf("%d\n",ans);
for(int i=1;i<=ans;i++) printf("%d%c",q[i]," \n"[i==ans]);
return 0;
}