Train Display

A tiny Express + WebSocket app that shows a live list of running model trains. An admin page lets you register trains as running or stopped, and the public display updates instantly. Runs are stored in a local SQLite database so you can review daily activity later.

How it works

• Pages: / shows the public display, /admin is for adding/stopping trains, and /activity shows all runs for a selected day.
• Live updates: Changes in the admin UI broadcast over WebSockets so the visitor display stays in sync without refreshes.
• Data: Trains and run history are kept in trains.db (SQLite) via better-sqlite3. Each update also writes a run record so you can report on past days.
• API: /api/running manages the current running list (GET/POST/PATCH/DELETE) and /api/reports/runs?date=YYYY-MM-DD returns historical runs.

Operating the app

1. Start the server (see platform-specific steps below).
2. Visit /admin to add a running train. Required fields: name, railway, country (emoji flag), power type, train type, year range, and location.
3. The public display at / updates immediately. Use the Stop action in /admin to end a run and remove it from the live board.
4. Visit /activity and pick a date to see all runs for that day with durations.

Running on macOS

1. Install Node.js (e.g., brew install node).
2. Install dependencies: npm install.
3. Start the server: npm start (optionally PORT=4000 npm start to change the port).
4. Open http://localhost:3000 (or your chosen port) and use /admin to manage trains.

Running on a dedicated Raspberry Pi

1. Install prerequisites:

   • Node.js 18+ (for ARM64/ARMv7 you can use NodeSource: curl -fsSL https://deb.nodesource.com/setup_20.x | sudo -E bash - then sudo apt-get install -y nodejs).
   • Build tools for better-sqlite3: sudo apt-get install -y build-essential python3 make g++.

2. In the project directory, install dependencies: npm install --production.

3. Start the server: PORT=3000 npm start. The app listens on all interfaces, so other devices on the network can connect.

4. Optional: create a systemd unit for auto-start on boot:

   [Unit]
   Description=Train Display
   After=network.target
   
   [Service]
   WorkingDirectory=/home/pi/whatisrunning
   ExecStart=/usr/bin/env PORT=3000 /usr/bin/node /home/pi/whatisrunning/server.js
   Restart=always
   Environment=NODE_ENV=production
   
   [Install]
   WantedBy=multi-user.target

   Save as /etc/systemd/system/traindisplay.service, then run sudo systemctl daemon-reload && sudo systemctl enable --now traindisplay.

Database files (trains.db) live alongside the code; back up that file if you need to preserve history across reinstalls.