Slint Frontend Showcase for an AI-powered Coffee Machine

Embedded frontend showcase made with slint for controlling an AI-powered coffee machine, which detects coffee cups and selects a product based on the type of cup. The frontend itself communicates with the backend (the coffee machine) via a websocket connection. A mock backend is included in this repository.

Prerequisites

Slint uses Rust, so Rust has to be installed. For development (mock websocket/REST Server) NodeJS is needed.

• Rust compiler (1.75 or newer)
• Node.js
• npm

Build locally

• Precondition: Websocket and REST Server must run, otherwise the application will fail.

  You can also start the Mock websocket and REST Server:

  npm install --include dev
  cd websocketserver
  node manual_coreservice_mock.js
  

• Build and run

  cargo run
  

Adapt websocket endpoints

The application expects the websocket server endpoint to be reachable on the same device at ws:://localhost:3000/frontend.

To use a different websocket server endpoint, you can place a config.yaml file in /usr/share/coffee-app, with the custom address set in the parameter websocket_address. E.g.:

echo 'websocket_address: ws://192.168.3.10:3000/frontend' >| /usr/share/coffee-app/config.yaml

Licensing

The code in this repository is licensed under MIT, see LICENSE, except for the fonts in ui/assets/fonts/ which are licensed under OFL.

Note that this project uses slint, which is licensed under GPLv3. If this software is distributed together with that library, then it must be distributed under the conditions of GPL. If this is not an option, then a different license can be purchased for slint.