---
id: cyc
title: Cycle Detection
author: Siyong Huang
prerequisites:
- Silver - Functional Graphs
- Gold - Breadth First Search
description: A simple cycle is a non-empty path of distinct edges that start and end at the same vertex such that no vertex appears more than once. Describes how to detect cycles in both directed and undirected graphs.
---
import { Problem } from "../models";
export const metadata = {
problems: {
und: [
new Problem("CSES", "Round Trip", "1669", "Easy", false, ["Cycle"]),
],
dir: [
new Problem("CSES", "Round Trip II", "1678", "Easy", false, ["Cycle"]),
],
general: [
new Problem("CSES", "Graph Girth", "1707", "Easy", false, ["Cycle"]),
],
}
};
*Cycle detection* determines properties of cycles in a directed or undirected graph, such as whether each node of the graph is part of a cycle or just checking whether a cycle exists.
## Undirected Graphs
BFS-Cycle?
## Directed Graphs
The same general idea is implemented below to find any cycle in a directed graph (if one exists).
```cpp
//UNTESTED
bool visited[MAXN], on_stack[MAXN];
vector adj[MAXN];
vector cycle;
bool dfs(int n)
{
visited[n] = on_stack[n] = true;
for(int u:adj[n])
{
if(on_stack[u])
return cycle.push_back(v), cycle.push_back(u), on_stack[n] = on_stack[u] = false, true;
else if(!visited[u])
{
if(dfs(u))
if(on_stack[n])
return cycle.push_back(n), on_stack[n] = false, true;
else
return false;
if(!cycle.empty())
return false;
}
}
on_stack[n] = false;
return false;
}
int main()
{
//take input, etc
for(int i = 1;cycle.empty() && i <= N;i++)
dfs(i);
if(cycle.empty())
printf("No cycle found!\n");
else
{
reverse(cycle.begin(), cycle.end());
printf("Cycle of length %u found!\n", cycle.size());
for(int n : cycle) printf("%d ", n);
printf("\n");
}
}
```
## Problems
VT-HSPC 2019?