This repository serves as the codebase for the virtual office project. Currently, the system works on any device that has a private IP address on the Stanford campus network, and we expect that the signaling server would be able to host ~50 people at the same time at its maximum.
For Chrome, type the following in the URL field
chrome://flags/#unsafely-treat-insecure-origin-as-secure
Enable the Insecure origins treated as secure
section and enter http://localhost:6001
.
WARNING: For Chrome, you might need to restart your browser after you change the settings above. Save the tabs!
For Firefix, type about:config
and set the values of media.devices.insecure.enabled
and media.getusermedia.insecure.enabled
to be true
.
NOTE: Qizheng is hosting and maintaining the signaling server now, so you can skip this step.
The signaling server is built with aiohttp and socketio. Run the following commands to install the required libraries first.
cd signaling
pip install -r requirements.txt
python server.py
A websocket server will be hosted at the IP address and at the port number indicated in web.run_app(). Note that this server does not have to be hosted on any client's local computer.
To set up personal info that will be displayed during the chat, do
cd goto-lunch-web
touch clientconfig.js
In clientconfig.js
, put the following lines
const USER_ID = "your name";
const DEVICE = "your device";
let USER_INFO = {"user_id": USER_ID, "device": DEVICE};
Note that this information will be broadcast to other users when you get online.
To start the web chat client, do
cd goto-lunch-web
python -m http.server 6001
A web chat window will be started and hosted at localhost:6001
. Feel free to change the port number.
The implementation of the signaling server is adapted based on https://github.com/pfertyk/webrtc-working-example.