This approach involves using Breadth-First Search (BFS) to precompute the minimum Manhattan distance from each cell to any thief in the grid, and then using this information to find the maximum safeness factor for reaching the bottom-right corner.
Time Complexity: O(n2 log n), where n is the grid size, due to the Dijkstra-like process.
Space Complexity: O(n2) for storing distances and safeness factors.
1import java.util.*;
2
3public class SafestPathFinder {
4 private static final int[][] directions = {{0, 1}, {1, 0}, {0, -1}, {-1, 0}};
5
6 public int maxSafenessFactor(int[][] grid) {
7 int n = grid.length;
8 int[][] distance = new int[n][n];
9 for (int[] row : distance) Arrays.fill(row, Integer.MAX_VALUE);
10 Queue<int[]> queue = new LinkedList<>();
11
12 // BFS to calculate distance to the nearest thief
13 for (int r = 0; r < n; r++) {
14 for (int c = 0; c < n; c++) {
15 if (grid[r][c] == 1) {
16 queue.offer(new int[]{r, c});
17 distance[r][c] = 0;
18 }
19 }
20 }
21
22 while (!queue.isEmpty()) {
23 int[] pos = queue.poll();
24 int r = pos[0], c = pos[1];
25 for (int[] dir : directions) {
26 int nr = r + dir[0], nc = c + dir[1];
27 if (nr >= 0 && nr < n && nc >= 0 && nc < n && distance[nr][nc] == Integer.MAX_VALUE) {
28 distance[nr][nc] = distance[r][c] + 1;
29 queue.offer(new int[]{nr, nc});
30 }
31 }
32 }
33
34 // Max-heap-like Dijkstra's for the safest path
35 PriorityQueue<int[]> pq = new PriorityQueue<>((a, b) -> b[2] - a[2]); // Max-heap based on safeness factor
36 pq.offer(new int[]{0, 0, distance[0][0]});
37 boolean[][] visited = new boolean[n][n];
38
39 while (!pq.isEmpty()) {
40 int[] cur = pq.poll();
41 int r = cur[0], c = cur[1], safeness = cur[2];
42
43 if (r == n - 1 && c == n - 1) return safeness;
44 visited[r][c] = true;
45
46 for (int[] dir : directions) {
47 int nr = r + dir[0], nc = c + dir[1];
48 if (nr >= 0 && nr < n && nc >= 0 && nc < n && !visited[nr][nc]) {
49 int newSafeness = Math.min(safeness, distance[nr][nc]);
50 pq.offer(new int[]{nr, nc, newSafeness});
51 }
52 }
53 }
54
55 return 0;
56 }
57
58 public static void main(String[] args) {
59 SafestPathFinder spf = new SafestPathFinder();
60 int[][] grid = {{0, 0, 1}, {0, 0, 0}, {0, 0, 0}};
61 System.out.println(spf.maxSafenessFactor(grid)); // Output: 2
62 }
63}
64
The Java implementation uses BFS to compute how far each cell is from the closest thief, followed by priority-based processing to ensure optimal safeness to the target using Dijkstra's inspired strategy.
This approach considers processing from both the starting and ending points in a bidirectional BFS style, potentially meeting in the middle for optimized distance calculations and safeness factor determination.
Time Complexity: O(n2) due to simultaneous BFS processing.
Space Complexity: O(n2) as distances from both ends are computed.
1#include <stdio.h>
2#include <stdlib.h>
3#include <limits.h>
4#include <stdbool.h>
5#include <string.h>
6
7#define MAX 400
8
9int grid[MAX][MAX];
10int n;
11
12int direction[4][2] = {{0,1}, {1,0}, {0,-1}, {-1,0}};
13
14int bidirectionalSafenessFactor(int grid[][MAX], int n) {
15 // Distance from start and end
16 int distanceStart[MAX][MAX];
17 int distanceEnd[MAX][MAX];
18
19 memset(distanceStart, INT_MAX, sizeof(distanceStart));
20 memset(distanceEnd, INT_MAX, sizeof(distanceEnd));
21
22 // BFS start queue
23 int queueStart[MAX*MAX][2];
24 int frontStart = 0, rearStart = 0;
25
26 queueStart[rearStart][0] = 0;
27 queueStart[rearStart][1] = 0;
28 rearStart++;
29 distanceStart[0][0] = 0;
30
31 // BFS end queue
32 int queueEnd[MAX*MAX][2];
33 int frontEnd = 0, rearEnd = 0;
34
35 queueEnd[rearEnd][0] = n-1;
36 queueEnd[rearEnd][1] = n-1;
37 rearEnd++;
38 distanceEnd[n-1][n-1] = 0;
39
40 while (frontStart < rearStart && frontEnd < rearEnd) {
41 // Expand from start
42 int r = queueStart[frontStart][0], c = queueStart[frontStart][1];
43 frontStart++;
44
45 for (int d = 0; d < 4; ++d) {
46 int nr = r + direction[d][0], nc = c + direction[d][1];
47 if (nr >= 0 && nr < n && nc >= 0 && nc < n && distanceStart[nr][nc] == INT_MAX) {
48 distanceStart[nr][nc] = distanceStart[r][c] + 1;
49 queueStart[rearStart][0] = nr;
50 queueStart[rearStart][1] = nc;
51 rearStart++;
52 }
53 }
54
55 // Expand from end
56 r = queueEnd[frontEnd][0], c = queueEnd[frontEnd][1];
57 frontEnd++;
58
59 for (int d = 0; d < 4; ++d) {
60 int nr = r + direction[d][0], nc = c + direction[d][1];
61 if (nr >= 0 && nr < n && nc >= 0 && nc < n && distanceEnd[nr][nc] == INT_MAX) {
62 distanceEnd[nr][nc] = distanceEnd[r][c] + 1;
63 queueEnd[rearEnd][0] = nr;
64 queueEnd[rearEnd][1] = nc;
65 rearEnd++;
66 }
67 }
68
69 // Check overlap
70 for (int i = 0; i < n; ++i) {
71 for (int j = 0; j < n; ++j) {
72 if (distanceStart[i][j] != INT_MAX && distanceEnd[i][j] != INT_MAX) {
73 return distanceStart[i][j] + distanceEnd[i][j];
74 }
75 }
76 }
77 }
78
79 return -1;
80}
81
82int main() {
83 int gridExample[MAX][MAX] = {{0, 0, 1}, {0, 0, 0}, {0, 0, 0}};
84 int n = 3;
85 printf("%d\n", bidirectionalSafenessFactor(gridExample, n));
86 return 0;
87}
88
The C solution processes the grid using bidirectional BFS starting from both the top-left corner and the bottom-right corner simultaneously, aiming to explore overlapping paths and maximize their safeness factor.