Getting started

Installation

Installing from pip

pip install panopti

Installing from source

git clone https://github.com/ArmanMaesumi/panopti

# build frontend viewer
cd panopti/frontend 
npm install
npm run build
cd ..

# install python package
pip install .

Core dependencies (included in pip install):

pip install numpy eventlet requests flask flask-socketio python-socketio[client] tomli

Optional dependencies:

pip install matplotlib      # for colormap utilities
pip install plotly==5.22.0  # for plotly figure support

Running a Panopti server

The Panopti server runs in an isolated process as your user code, hence you should run it in a separate terminal. You may use the convenience module (recommended):

python -m panopti.run_server --host localhost --port 8080

or manually through your own script

import panopti
panopti.start_server(host='localhost', port=8080, debug=False)

Warning

You should not run start_server from your main script. The server is meant to run in its own process, which avoids slowing down your code.

The viewer will then be available at http://http://localhost:8080/?viewer_id=your_client_id

Note: Your application will register its own viewer_id, which will affect the link above (see below).

Connecting to a Panopti server

Once your server is running, your main application can connect to the instance via:

import panopti
viewer_id = "client" # choose a unique id here
viewer = panopti.connect(server_url="http://localhost:8080", viewer_id=viewer_id) 
# now in your browser open: http://localhost:8080/?viewer_id=client

# ... your code here ...

viewer.hold() # prevent the script from terminating

The server uses viewer_id to distinguish between different scripts that may be running concurrently. This argument must be specified even when just one session is active.

Hello World with Panopti

Now let's add a simple mesh to the viewer.

import panopti
import trimesh # only for io
viewer = panopti.connect(server_url="http://localhost:8080", viewer_id='client') 

mesh = trimesh.load('demosthenes.obj')
verts, faces = mesh.vertices, mesh.faces

viewer.add_mesh(
    vertices=verts,
    faces=faces,
    name="Statue"
)

viewer.hold() # prevent the script from terminating

Done! For more elaborate examples, See the Examples tab, or:

Working remotely (SSH)

Panopti seamlessly integrates into remote work setups. Please refer to the Remote SSH page for details.