Kojo360/UG-Navigation

An algorithm and a Java application to help users find the best route from one location to another on the University of Ghana (UG) campus # UG-Navigation An algorithm and Java application to help users find the best route from one location to another on the University of Ghana (UG) campus. ## Setup ### Prerequisites - **Python 3.6+** (for generating road graph CSV files) - **JDK 17+** (for compiling and running Java) Verify your tools: ```bash python --version javac -version java -version ``` If `javac`/`java` is not found, install a JDK and add its `bin` folder to your PATH. ## Quick Start (Recommended) Run from the project root. ### 1) Build road graph data ```bash python scripts/build_road_graph.py --geojson "export (3).geojson" --mode drive ``` This creates: - `data/road_drive_nodes.csv` - `data/road_drive_edges.csv` ### 2) Compile Java sources ```bash javac -d . src/ug/campus/*.java ``` ### 3) Run the app ```bash java -cp . ug.campus.Main ``` You will be prompted for: - source name - destination name - optional landmark keyword ### Non-interactive use POI names are resolved first and then snapped to the selected road graph, so POI IDs are never used directly as road-node IDs. Use flags for scripts and tools: ```bash java -cp . ug.campus.Main --mode walk --from "Commonwealth Hall" --to "Balme Library" --via "Great Hall" --format json ``` Available options: `--mode drive|walk`, `--from `, `--to `, `--via `, and `--format text|json`. `--from` and `--to` must be provided together; otherwise the app uses the interactive prompts. ## Platform Notes ### Windows PowerShell (optional JAVA_HOME usage) If you prefer using `JAVA_HOME` directly: ```powershell $env:JAVA_HOME = 'D:\jdk-26.0.2' & "$env:JAVA_HOME\bin\javac" -d . src/ug/campus/*.java & "$env:JAVA_HOME\bin\java" -cp . ug.campus.Main ``` ### Linux/macOS If Java is installed and on PATH, the same `javac` / `java` commands above work directly. ## Behavior Notes - `Main` loads the requested drive or walk graph (`road_drive_*` or `road_walk_*`). - If a place name does not match exactly, the app suggests similar names. ## N …