PathForge — Real-Time Pathfinding Visualizer

Live Interactive Web Visualizer: https://albert101255.github.io/PathForge/

PathForge — Pathfinding Visualizer

Live Interactive Visualizer: https://albert101255.github.io/PathForge/

An interactive C++17 pathfinding visualizer built with SFML (with a companion interactive web visualizer on GitHub Pages). It lets you draw obstacles, paint weighted terrain, choose a search algorithm, and watch the algorithm explore the grid before reconstructing a path.

Algorithms

• A* — uses a heuristic to guide exploration toward the target
• Dijkstra — computes shortest paths for non-negative edge costs
• Breadth-First Search (BFS) — finds a shortest path in an unweighted grid
• Depth-First Search (DFS) — explores deeply but does not guarantee a shortest path

A* is often faster than uninformed search when the heuristic is informative, but performance depends on the map, heuristic and implementation.

What this project demonstrates

• C++17
• Graph/grid traversal
• Queues and priority queues
• Heuristics
• Object-oriented design
• SFML rendering/input
• CMake builds
• Separation between algorithm logic and visualization

Controls

Action	Input
Draw wall	Left-click / drag
Erase wall	Right-click
Place start	S + left-click
Place end	E + left-click
Run	Space
Reset path	R
Clear grid	C
Select algorithm	1 A*, 2 Dijkstra, 3 BFS, 4 DFS
Adjust speed	+ / -

Architecture

Grid / Cell state
      |
      v
Pathfinding algorithm
      |
      v
Search-step updates
      |
      v
SFML renderer / UI

The IPathfinder abstraction keeps the algorithm implementation separate from the visualization layer so different search strategies can be selected without rewriting the UI.

Complexity notes

For a graph with V vertices and E edges:

• BFS: O(V + E)
• DFS: O(V + E)
• Dijkstra with a binary heap: approximately O((V + E) log V)
• A*: worst-case behavior depends on the heuristic and graph, while a useful heuristic can significantly reduce explored nodes in practice

Build

Linux

sudo apt update
sudo apt install build-essential cmake libsfml-dev

Windows / MSYS2

pacman -S mingw-w64-ucrt-x86_64-sfml

Compile & Test

cmake -S . -B build
cmake --build build
ctest --test-dir build --output-on-failure

Project status

Includes a headless unit test suite (tests/test_pathfinding.cpp) covering Grid state management, BFS, DFS, Dijkstra, and A* across open grids, obstacles, unreachable targets, start == end edge cases, and weighted terrain.

License

MIT License — see LICENSE.

Author

Albert