Careercup - Google面试题 - 4716965625069568

2014-05-06 00:17

题目链接

原题：

Given a 2-D matrix represents the room, obstacle and guard like the following (0 is room, B->obstacle, G-> Guard): 
0 0 0 
B G G 
B 0 0 

calculate the steps from a room to nearest Guard and set the matrix, like this 
2 1 1 
B G G 
B 1 1 
Write the algorithm, with optimal solution.

题目：有一个二维矩阵表示的迷宫，其中G表示保安人员，B表示不可穿越的墙，其他位置均为空地。如果每次可以向东南西北四个方向移动一格，请计算出每个空格离最近的保安有多远。题目给出了一个示例。

解法：既然矩阵里可能有多个保安，我的思路是以各个保安为中心，进行广度优先搜索，搜索的过程中随时更新每个空格位置的最短距离，保证全部搜索完之后所有的结果都是最小的。单次BFS的时间代价是O(n^2)级别的。

代码：

 1 // http://www.careercup.com/question?id=4716965625069568
 2 #include <queue>
 3 #include <vector>
 4 using namespace std;
 5 
 6 class Solution {
 7 public:
 8     void solve(vector<vector<int> > &matrix) {
 9         // -1 for guard
10         // -2 for blockade
11         // 0 for room
12         // > 0 for distance
13         
14         n = (int)matrix.size();
15         if (n == 0) {
16             return;
17         }
18         m = (int)matrix[0].size();
19         if (m == 0) {
20             return;
21         }
22         
23         int i, j;
24         
25         for (i = 0; i < n; ++i) {
26             for (j = 0; j < m; ++j) {
27                 if (matrix[i][j] == -1) {
28                     doBFS(matrix, i, j);
29                 }
30             }
31         }
32     };
33 private:
34     int n, m;
35     
36     bool inBound(int x, int y) {
37         return x >= 0 && x <= n - 1 && y >= 0 && y <= m - 1;
38     }
39     
40     void doBFS(vector<vector<int> > &matrix, int x, int y) {
41         queue<int> q;
42         static int dd[4][2] = {{-1, 0}, {+1, 0}, {0, -1}, {0, +1}};
43         int tmp;
44         int xx, yy;
45         int i;
46         int dist;
47         
48         q.push(x * m + y);
49         while (!q.empty()) {
50             tmp = q.front();
51             q.pop();
52             x = tmp / m;
53             y = tmp % m;
54             dist = matrix[x][y] > 0 ? matrix[x][y] : 0;
55             for (i = 0; i < 4; ++i) {
56                 xx = x + dd[i][0];
57                 yy = y + dd[i][1];
58                 if (!inBound(xx, yy) || matrix[xx][yy] < 0 || 
59                     (matrix[xx][yy] > 0 && matrix[xx][yy] <= dist + 1)) {
60                     // out of boundary
61                     // a guard or a blockade
62                     // the distance is no shorter
63                     continue;
64                 }
65                 matrix[xx][yy] = dist + 1;
66                 q.push(xx * m + yy);
67             }
68         }
69     }
70 };

 

转载于:https://www.cnblogs.com/zhuli19901106/p/3710623.html