Leetcode - Course Schedule

[分析]
拓扑排序的经典应用。review时发现自己搞反了课程直接的依赖关系，但却真切地被Accept，
图上画了画，发现搞反后若输出排序结果就是和原来相反，但不影响本题的正确性。
网上搜了搜逆拓扑，发现其是有可用之处的，参见下面这篇博文：
[url]http://blog.csdn.net/guodongxiaren/article/details/37988833[/url]
另一篇博文提出拓扑排序是所有节点dfs的逆后序，也就是每个节点任务完成的时间的逆序排序 ，这个洞见也挺有意思的[url]https://m.oschina.net/blog/419291[/url]

public class Solution {
    // method 2: 二维数组表示有向图
    public boolean canFinish2(int numCourses, int[][] prerequisites) {
        if (prerequisites == null || prerequisites.length == 0 || prerequisites[0].length == 0)
            return true;
        // build graph
        ArrayList<HashSet<Integer>> graph = new ArrayList<HashSet<Integer>>(numCourses);
        int[] indegree = new int[numCourses];
        for (int i = 0; i < numCourses; i++) {
            graph.add(new HashSet<Integer>());
        }
        int m = prerequisites.length;
        for (int i = 0; i < m; i++) {
            if (graph.get(prerequisites[i][1]).add(prerequisites[i][0]))
                indegree[prerequisites[i][0]]++;
        }
        LinkedList<Integer> queue = new LinkedList<Integer>();
        for (int i = 0; i < numCourses; i++) {
            if (indegree[i] == 0) queue.offer(i);
        }
        int counter = 0;
        while (!queue.isEmpty()) {
            counter++;
            int curr = queue.poll();
            for (int w : graph.get(curr)) {
                if (--indegree[w] == 0) queue.offer(w);
            }
        }
        return counter == numCourses;
    }
    // method 1
    class Vertex {
        int value;
        int indegree;
        HashSet<Vertex> adjacent;
        public Vertex(int val) {
            this.value = val;
            indegree = 0;
            adjacent = new HashSet<Vertex>();
        }
    }
    public boolean canFinish(int numCourses, int[][] prerequisites) {
        if (prerequisites == null || prerequisites.length == 0 || prerequisites[0].length == 0)
            return true;
        HashMap<Integer, Vertex> graph = buildGraph(numCourses, prerequisites);
        LinkedList<Vertex> queue = new LinkedList<Vertex>();
        for(Vertex node : graph.values()) {
            if (node.indegree == 0)
                queue.offer(node);
        }
        int counter = 0;
        while (!queue.isEmpty()) {
            Vertex node = queue.poll();
            counter++;
            for (Vertex w : node.adjacent) {
                w.indegree--;
                if (w.indegree == 0) queue.offer(w);
            }
        }
        return counter == graph.size();
    }
    private HashMap<Integer, Vertex> buildGraph(int numCourses, int[][] prerequisites) {
        HashMap<Integer, Vertex> graph = new HashMap<Integer, Vertex>();
        for (int i = 0; i < numCourses; i++) {
            graph.put(i, new Vertex(i));
        }
        int m = prerequisites.length;
        for (int i = 0; i < m; i++) {
            if(graph.get(prerequisites[i][1]).adjacent.add(graph.get(prerequisites[i][0]))) {
                graph.get(prerequisites[i][0]).indegree++;
            }
        }
        return graph;
    }
}