🌍 GIS Network Builder

GIS Network Builder is a modular, high-performance desktop GIS workstation built with Python. It is specifically designed for councils, local authorities, and civil engineers to digitise urban assets, model road networks, and perform advanced spatial analysis.

🚀 Features

🛠️ Advanced Digitising & UI

1. Modern Enterprise UI: Built with CustomTkinter, featuring a native dark theme and rounded, high-end design elements.
2. Automated Data Import: Pull real-world drivable streets from OpenStreetMap around any address, with road names, classes, speed limits and one-way streets.
3. Standardised Asset Management: Dropdowns for rapid, error-free entry of standard council assets (Streetlights, Potholes, Drains, Manhole Covers, Bus Stops and more), road classes and road directions.
4. Intelligent Snapping: An adjustable tolerance (in metres) snaps new clicks to the nearest asset point, line vertex or polygon corner, or part-way along an existing road, so digitised networks connect properly.
5. Undo: Ctrl+Z undoes the last shape you added or deleted.

🧠 Network & Spatial Analysis

1. Real-World Routing: Calculates the fastest path using travel time (length ÷ speed limit) rather than just physical distance. Respects one-way streets and treats line ends, T-junctions and shared vertices as junctions.
2. Time-Based Isochrones: Visualise service coverage areas based on a travel-time budget (e.g., 5-minute response zones).
3. Spatial Geofencing: Run "Select by Location" queries to find all assets and roads inside or crossing a zone.
4. Choropleth Mapping: Colour-code zones by any numeric attribute.

📊 Data Visualisation & Export

1. Custom Map Layers: Switch between Google Maps, Google Satellite, Google Hybrid and OpenStreetMap base maps.
2. Interactive Web Export: Generate standalone HTML maps (OpenStreetMap background) with interactive popups and a Point Density Heatmap.
3. Flexible Workspaces: Save entire sessions, including custom colours and attributes, to JSON Project Workspaces.
4. Industry Standard I/O: Load Shapefile, KML and GeoJSON; export to Shapefile (.shp) for CAD/GIS or CSV for Microsoft Excel.

📖 Usage Guide

The left-hand panel has three tabs: Draw, Tools and Layers.

1. Drawing and Editing

• Draw shapes: On the Draw tab, choose 📍 Draw Point, 📏 Draw Line or ⬡ Draw Polygon, then left-click on the map. For lines and polygons, click ✓ Finish Line or ✓ Finish Polygon to complete the shape.
• Snapping: Set the Snapping Tolerance slider on the Layers tab (0–50 m; 0 turns snapping off). Clicks snap to the nearest existing point or vertex, or onto an existing road, so a new road started part-way along another one becomes a T-junction. Clicking an existing point marker while drawing uses its exact position.
• Crossroads: Roads that simply cross without a shared vertex are not joined, because they could be a bridge or an underpass. To make a crossroads, click on the other road while drawing so the new road gets a vertex on it.
• One-way streets: Click a road and set its Direction to "One-Way (A to B)" (A = the first point drawn) or "One-Way (B to A)". One-way roads show an arrowhead in the direction of travel, and routing and isochrones follow it.
• Edit data: Click any shape to open the Dynamic Editor. Here you can add custom fields, change values, pick a custom colour or delete the shape. A road's Speed (mph) and Length_m are used for routing.
• Undo: Click ↩️ Undo or press Ctrl+Z.

2. Getting a Road Network

Routing and isochrones only use the roads in your project. The roads on the background map are just a picture. To add roads, either:

• Import real streets: On the Layers tab, click 🌍 Import OSM Data and enter a place (e.g., "Moseley, Birmingham"). Drivable roads within about 1.5 km of it are downloaded. Stick to neighbourhoods rather than whole cities.
• Draw your own: Use 📏 Draw Line with snapping switched on so the roads join up.

3. Network Analysis

• Fastest route: On the Tools tab, click 🟢 Set Start Point and click the map, then 🔴 Set End Point and click again, then ⚡ Calculate Route. The route is drawn in purple with the estimated travel time. If the start or end point is more than 200 m from your road network, a note tells you the route uses the nearest junction instead.
• Isochrones (service areas): Click ⏳ Calculate Isochrone, click a start point on the map and enter a travel-time budget in seconds (e.g., 300 for 5 minutes). Reachable roads are highlighted and outlined.
• Clear: ✖ Clear Route removes the route, the isochrone and the start/end markers.

4. Spatial Analysis & Visualisation

• Geofencing: On the Tools tab, click 🎯 Geofence Selection, then click on a zone. Points inside it and roads inside or crossing it are highlighted. Click Clear Selection to reset.
• Choropleth: On the Layers tab, click 🎨 Generate Choropleth and enter the name of a numeric attribute on your zones (add one in the editor first, e.g., Population or Area_sqm). Zones are shaded from light to dark blue by value.
• Visibility: The switches on the Layers tab show or hide points, roads and zones.

5. Search and Navigation

• Search: Type an address or landmark into the search bar above the map and press Enter (or click Fly To).
• Attribute Table: On the Layers tab, click 📊 Attribute Table to see all data at once. Double-click a row to zoom the map to that feature.

6. Saving and Exporting

• 💾 Save Project (JSON) / 📂 Load Project (JSON): Save or reload your whole workspace.
• 📂 Load GIS File: Add shapes from a Shapefile, KML or GeoJSON. Data is reprojected to WGS84 automatically and height values are dropped. Only each feature's name is imported for now.
• 💾 Save SHP: Writes Points.shp, Lines.shp and Polygons.shp to a shapefiles folder in the folder the app was started from.
• 📄 Export to CSV: Exports every feature's attributes (no coordinates).
• 🌐 Web Map Export: Creates Network_WebMap.html and opens it in your browser.

🗺️ Map Tiles & Data Licences

• OpenStreetMap: Tiles come from OpenStreetMap's volunteer-run servers, which have a tile usage policy. The app follows it by identifying itself (the GISNetworkBuilder/1.0 User-Agent set in map_tiles.py), caching tiles on disk (%LOCALAPPDATA%\GISNetworkBuilder\tiles on Windows), only fetching tiles around the area being viewed, and showing "© OpenStreetMap contributors" on the map. These servers aren't meant for heavy or commercial use.
• Google: The Google base maps are loaded straight from Google's tile servers without an API key, which Google Maps Platform's terms don't permit. For council or commercial use, switch to a licensed provider such as Ordnance Survey's OS Maps API.
• Road data: Imported roads are © OpenStreetMap contributors, available under the Open Database Licence.

🏗️ Project Architecture

• GUI Framework: CustomTkinter
• Map Widget: TkinterMapView
• Spatial Logic: Shapely & GeoPandas
• Network Engine: NetworkX (directed graph weighted by travel time)
• API Integrations: OpenStreetMap via OSMnx; Nominatim geocoding via Geopy
• Web Export: Folium

The project follows a modular structure that separates the user interface from spatial processing, file handling, and export logic.

GISNetworkBuilder/
│
├── main.py              # Core application and UI coordinator
├── network_engine.py    # Snapping, graph building, routing and isochrones (NetworkX)
├── map_tiles.py         # Tile downloads that follow the OpenStreetMap tile usage policy
├── web_export.py        # HTML/Folium map generation and heatmap export
├── ui_datatable.py      # Attribute table window
├── ui_popups.py         # Dynamic Editor popup for points, lines and zones
├── file_handler.py      # Reading SHP, KML and GeoJSON files
├── requirements.txt     # Python dependencies (pinned, tested on Python 3.12)
└── README.md            # Project documentation

🛠️ Installation

Tested with Python 3.12 on Windows.

1. Clone the repo:

   git clone https://github.com/gingermuaz/GISNetworkBuilder.git
   cd GISNetworkBuilder

2. Create a virtual environment and install the dependencies:

   python -m venv .venv
   .venv\Scripts\activate          # macOS/Linux: source .venv/bin/activate
   pip install -r requirements.txt

3. Run the app:

   python main.py

❓ Troubleshooting

• "There are no roads in this project to route along": Import streets (Layers → 🌍 Import OSM Data) or draw some roads first. Routing doesn't use the background map.
• "No path exists between these points": The roads aren't joined up (check snapping, and remember roads that cross without a shared vertex aren't connected), or a one-way street blocks the journey.
• OpenStreetMap tiles show "403 Access blocked": Make sure you're running the latest map_tiles.py. If it continues, add a contact email to USER_AGENT in map_tiles.py, or switch to a tile provider that issues API keys.