POJ 3009 dfs暴搜
来源:互联网 发布:mac地址ip地质作用 编辑:程序博客网 时间:2024/06/05 14:59
Curling 2.0
Time Limit: 1000MS Memory Limit: 65536K
Total Submissions: 15172 Accepted: 6306
Description
On Planet MM-21, after their Olympic games this year, curling is getting popular. But the rules are somewhat different from ours. The game is played on an ice game board on which a square mesh is marked. They use only a single stone. The purpose of the game is to lead the stone from the start to the goal with the minimum number of moves.
Fig. 1 shows an example of a game board. Some squares may be occupied with blocks. There are two special squares namely the start and the goal, which are not occupied with blocks. (These two squares are distinct.) Once the stone begins to move, it will proceed until it hits a block. In order to bring the stone to the goal, you may have to stop the stone by hitting it against a block, and throw again.
Fig. 1: Example of board (S: start, G: goal)
The movement of the stone obeys the following rules:
At the beginning, the stone stands still at the start square.
The movements of the stone are restricted to x and y directions. Diagonal moves are prohibited.
When the stone stands still, you can make it moving by throwing it. You may throw it to any direction unless it is blocked immediately(Fig. 2(a)).
Once thrown, the stone keeps moving to the same direction until one of the following occurs:
The stone hits a block (Fig. 2(b), (c)).
The stone stops at the square next to the block it hit.
The block disappears.
The stone gets out of the board.
The game ends in failure.
The stone reaches the goal square.
The stone stops there and the game ends in success.
You cannot throw the stone more than 10 times in a game. If the stone does not reach the goal in 10 moves, the game ends in failure.
Fig. 2: Stone movements
Under the rules, we would like to know whether the stone at the start can reach the goal and, if yes, the minimum number of moves required.
With the initial configuration shown in Fig. 1, 4 moves are required to bring the stone from the start to the goal. The route is shown in Fig. 3(a). Notice when the stone reaches the goal, the board configuration has changed as in Fig. 3(b).
Fig. 3: The solution for Fig. D-1 and the final board configuration
Input
The input is a sequence of datasets. The end of the input is indicated by a line containing two zeros separated by a space. The number of datasets never exceeds 100.
Each dataset is formatted as follows.
the width(=w) and the height(=h) of the board
First row of the board
…
h-th row of the board
The width and the height of the board satisfy: 2 <= w <= 20, 1 <= h <= 20.
Each line consists of w decimal numbers delimited by a space. The number describes the status of the corresponding square.
0 vacant square
1 block
2 start position
3 goal position
The dataset for Fig. D-1 is as follows:
6 6
1 0 0 2 1 0
1 1 0 0 0 0
0 0 0 0 0 3
0 0 0 0 0 0
1 0 0 0 0 1
0 1 1 1 1 1
Output
For each dataset, print a line having a decimal integer indicating the minimum number of moves along a route from the start to the goal. If there are no such routes, print -1 instead. Each line should not have any character other than this number.
Sample Input
2 1
3 2
6 6
1 0 0 2 1 0
1 1 0 0 0 0
0 0 0 0 0 3
0 0 0 0 0 0
1 0 0 0 0 1
0 1 1 1 1 1
6 1
1 1 2 1 1 3
6 1
1 0 2 1 1 3
12 1
2 0 1 1 1 1 1 1 1 1 1 3
13 1
2 0 1 1 1 1 1 1 1 1 1 1 3
0 0
Sample Output
1
4
-1
4
10
-1
模拟冰壶的运动,开始想用bfs,后来发现系统内存消耗太大。直接dfs回溯即可,我用了最小堆来保存找到的步数,堆顶元素即为结果。
#include<cstdio>#include<iostream>#include<bitset>#include<queue>using namespace std;typedef struct po{ int x,y,h; bool operator < (const po &a) const{ return a.h<h; }}Po;int G[30][30];int m,n,CNT = 0;Po move[] = {-1,0,0, 1,0,0, 0,-1,0, 0,1,0};priority_queue<Po>pq;bool IsTrue(Po k){ if(k.x>=0&&k.x<m&&k.y>=0&&k.y<n){ if(G[k.x][k.y] == 0||G[k.x][k.y] == 3) return true; } return false;}void dfs(Po k){ if(k.h>10){ pq.push(k); return ; } if(!pq.empty()) if(k.h>=pq.top().h) return ; if(G[k.x][k.y] == 3){ pq.push(k); return ; } int i; for(i = 0;i<4;i++){ Po term = k; term.x += move[i].x; term.y += move[i].y; if(IsTrue(term)){ if(G[term.x][term.y] == 3){ term.h++; pq.push(term); return ; } switch (i){ case 0: while(term.x - 1>=0){ if(G[term.x-1][term.y] == 0){ term.x--; }else if(G[term.x-1][term.y] == 1){ G[term.x-1][term.y] = 0; term.h++; dfs(term); G[term.x-1][term.y] = 1; break; }else if(G[term.x-1][term.y] == 3){ term.h++; pq.push(term); return ; } } break; case 1: while(term.x + 1<m){ if(G[term.x+1][term.y] == 0){ term.x++; }else if(G[term.x+1][term.y] == 1){ G[term.x+1][term.y] = 0; term.h++; dfs(term); G[term.x+1][term.y] = 1; break; }else if(G[term.x+1][term.y] == 3){ term.h++; pq.push(term); return ; } } break; case 2: while(term.y - 1>=0){ if(G[term.x][term.y-1] == 0){ term.y--; }else if(G[term.x][term.y-1] == 1){ G[term.x][term.y-1] = 0; term.h++; dfs(term); G[term.x][term.y-1] = 1; break; }else if(G[term.x][term.y-1] == 3){ term.h++; pq.push(term); return ; } } break; case 3: while(term.y + 1<n){ if(G[term.x][term.y+1] == 0){ term.y++; }else if(G[term.x][term.y+1] == 1){ G[term.x][term.y+1] = 0; term.h++; dfs(term); G[term.x][term.y+1] = 1; break; }else if(G[term.x][term.y+1] == 3){ term.h++; pq.push(term); return ; } } break; } } }}int main(){// freopen("data.in","r",stdin); while(scanf("%d%d",&n,&m)&&(m||n)){ CNT++; int i,j; Po S; for(i = 0;i<m;i++) for(j = 0;j<n;j++){ scanf("%d",&G[i][j]); if(G[i][j] == 2){ S.x = i; S.y = j; S.h = 0; G[i][j] = 0; } } dfs(S); if(pq.empty()) printf("-1\n"); else{ int hh = pq.top().h; if(hh>10) printf("-1\n"); else printf("%d\n",hh); } while(!pq.empty()) pq.pop(); } return 0;}
- POJ 3009 dfs暴搜
- POJ 3009 dfs
- POJ 3009 DFS +剪枝
- DFS poj 3009
- poj 3009 dfs
- poj 3009 (DFS)
- poj 3009 dfs + 恢复
- poj 3009 dfs+剪枝
- poj 3009 dfs
- poj 3009 (dfs)
- poj 3009 dfs
- poj 3009 DFS +回溯
- poj 3009 DFS
- poj 3009(dfs)
- POJ 3009(Dfs)
- POJ 3009 DFS+剪枝
- poj 3009 dfs
- poj 3009 DFS
- Swift-- 方法
- [TwistedFate]属性property
- zoj 3891 K-hash(后缀自动机)
- hdoj2053(switch game
- Intent 传数据
- POJ 3009 dfs暴搜
- 外键约束和级联操作
- Java学习笔记(1):重载方法
- 【编程开发】opencv实现对Mat中某一列或某一行的元素进行normalization
- READING NOTE: Two-Stream Convolutional Networks for Action Recognition in Videos
- 用于Android ART虚拟机JNI调用的NativeBridge介绍
- 配置ADT
- C中堆和栈的区别
- 最长递增子序列详解(longest increasing subsequence)