The purpose of this Python package is to provide easy & extensive access to the D3 library from within a Python 3 Jupyter notebook.

It enables to code a generic D3 visualization in a separate .js file and load it from Python with a Pandas dataframe sent as input.

The package is on purpose very lightweight, to enable easy forking & publishing from/to bl.ocks.org, or any other kind of standalone deployment.

pip install p3d3

or download repository from Github and python setup.py install

Run:

import p3d3
p3d3.init()

to download & initialize the d3 and Vega engines. Then you can use the functions p3d3.p3d3() and p3d3.vegalite()

Examples on how to use P3D3 are available in the examples/ folder.

By default the D3 version used is 4.13.0, you can override it with p3d3.init(version="5.16.0"). All versions available on Cloudflare CDN can be used.

This package is in its early stages. You are very welcome to report issues & propose pull requests to improve it.