java代码

以下是Java代码实现，假设邻接表的数据结构为`List<List<Integer>> adjList`，起点、终点、经过顶点分别为`int u, v, k`：

public int countPaths(List<List<Integer>> adjList, int u, int v, int k) {
    int[] count = new int[1];
    boolean[] visited = new boolean[adjList.size()];
    visited[u] = true;
    dfs(adjList, u, v, k, visited, count);
    return count[0];
}

private void dfs(List<List<Integer>> adjList, int curr, int dest, int k, boolean[] visited, int[] count) {
    if (curr == k) {
        for (int neighbor : adjList.get(k)) {
            if (!visited[neighbor]) {
                visited[neighbor] = true;
                dfs(adjList, neighbor, dest, k, visited, count);
                visited[neighbor] = false;
            }
        }
    } else if (curr == dest) {
        count[0]++;
    } else {
        for (int neighbor : adjList.get(curr)) {
            if (!visited[neighbor]) {
                visited[neighbor] = true;
                dfs(adjList, neighbor, dest, k, visited, count);
                visited[neighbor] = false;
            }
        }
    }
}

其中，count数组用于记录经过顶点k的路径数量，visited数组用于记录每个顶点的访问状态，dfs方法为深度优先搜索算法实现。