A support chat application for integration with qualtrics through embedding <iframe>
Used to communicate with research participants during live lab data collections in the PRISE research project
https://www.gu.se/forskning/prise
Prise chat uses a python flask backend with socket.io for message transport. The docker container fires up a gunicorn server with eventlet for socket support.
Prise chat was developed to tackle the impact of COVID-19 had on the project and hence the software is tailored to fit needs of Prise. If anyone want to enhance the application please feel free to make a pull request 💖
- Support for multiple users answering questions (implemented by a claim-chat feature)
- The application randomly suggests a user to claim a chat to monitor.
- It is not possible to save chats (This is a feature not a bug)
- The user model is extremely simple (But safe)
- Uses sqlalchemy ORM (defaults to sqlite but any other DB supported by the ORM should work out of the box)
- No internalization (Admin UI in english chatbox in Swedish )
PRISE CHAT is deployed for production in University of Gothenburgs openshift platform. Secure communication setting like TSL and CORS are configured with openshift routing. Please see openshift docs for routing setup.
Other deployment options is of course possible kubernetes or custom *nix server
First build the image by running:
docker build -t chat:latest .
Spin it up by running:
docker run --name chat -d -p 8000:5000 -e SECRET_KEY=my_very_secret_key --rm chat:latest
Point your browser at http://127.0.0.1:8000
Environment variables can be set either by .env or .flaskenv if you run dev or non docker deployment.
SECRET_KEY = 'A long random string' FLASK_APP_BRAND = "Brand name of the app e.g. My data collection chat"
TBD
Yes please PR's are welcome 💖