拓扑排序

  若图为有向无环图，则可进行拓扑排序。拓扑排序的结果为DFS后序遍历的倒序。选课是拓扑排序的经典应用场景之一，即：选修一门课程之前须先修完该课程的前置课程。

class Graph(object):
    def __init__(self, points_nums, is_directed):
        self.__points_nums = points_nums
        self.__adj = [[] for _ in range(points_nums)]
        self.__directed = is_directed
        self.__in_degree = [0 for _ in range(points_nums)]

    def add_edge(self, point1, point2):
        self.__adj[point1].append(point2)
        self.__in_degree[point2] += 1
        if not self.__directed:
            self.__adj[point2].append(point1)

    def get_adj(self, point):
        return self.__adj[point]

    def get_point_nums(self):
        return self.__points_nums

    def get_in_degree(self, point):
        return self.__in_degree[point]


class DepthFirstOrder(object):      
    def __init__(self, graph):
        self.__graph = graph
        self.__reverse_postorder = []
        self.__marked[False for _ in range(graph.get_point_nums())]
        for point in range(graph.get_point_nums()):
            if not self.__marked[point] and graph.get_in_degree(point) == 0:
                self.__dfs(self.graph, point)

    def __dfs(self, point):
        self.__marked[point] = True
        for adj_point in self.__graph.get_adj(point):
            if not self.__mark[adj_point]:
                self.__dfs(adj_point)
        self.__reverse_postorder.insert(0, point)