Dijkstra算法迷宫地图里多节点Python实现

以下是Dijkstra算法在迷宫地图里多节点的Python实现：

import heapq

def dijkstra(maze, start, end):
    # 初始化起点和终点
    start_node = (start, 0)
    end_node = (end, float('inf'))

    # 初始化距离字典和visited集合
    distances = {node: float('inf') for row in maze for node in row}
    distances[start_node[0]] = start_node[1]

    visited = set()

    # 初始化堆并添加起点
    heap = []
    heapq.heappush(heap, start_node)

    while heap:
        # 弹出距离最小的节点
        current_node = heapq.heappop(heap)[0]

        if current_node == end_node[0]:
            return distances[end_node[0]]

        if current_node in visited:
            continue

        visited.add(current_node)

        # 遍历四个方向
        for i, j in [(0, 1), (1, 0), (0, -1), (-1, 0)]:
            row = current_node[0][0] + i
            col = current_node[0][1] + j

            if 0 <= row < len(maze) and 0 <= col < len(maze[0]) and maze[row][col] != 1:
                neighbor = ((row, col), 1 if maze[row][col] == 2 else 0)
                distance = distances[current_node] + neighbor[1]

                if distance < distances[neighbor[0]]:
                    distances[neighbor[0]] = distance
                    heapq.heappush(heap, (neighbor, distance))

    return -1

其中，`maze`是一个二维列表，表示迷宫地图。0表示可以通过的空地，1表示障碍物，2表示终点。`start`和`end`是起点和终点的坐标。函数返回起点到终点的最短距离。