vispydeprecated

A Custom Jupyter Widget Library using WebGL for the VisPy Python Library

Usage no npm install needed!

<script type="module">
  import vispy from 'https://cdn.skypack.dev/vispy';
</script>

README

vispy.js

This is a light JavaScript version of Vispy. This is used by the python library to create a jupyter notebook and jupyter lab extension. This allows the python code to communicate with the client's browser to display WebGL visuals. This is accomplished by sending GLIR (Graphics Language Intermediate Representation) over a web socket to the client. See the vispy python library for more information.

This repository also includes javascript files for not only GLIR parsing and handling, but a standalone javascript implementation of the vispy "gloo" interface for a high-level object oriented OpenGL workflow.

Right now this javascript library is considered experimental and the limitations of it should be considered before using it in an operational setting. Due to the design of this library and the need for passing GLIR commands, some operations can be slow and cause a backlog of commands to occur. An example of such an operation would be using a timer to immediately trigger animation-like redraws of a canvas with no delay in between frames.

This repositories structure follows that of the ipywidgets project set up in this cookiecutter project. See those templates for more information about their purpose. This library is kept in its own repository, but is also available in the vispy python library's repository as a git submodule.

Installation

This library currently only serves as a jupyter extension/widget and should be installed by following the instructions of the vispy python library. See the installation instructions for more details.

In the future, this library may include additional entry points to act as a standalone javascript library, but support for this has not been implemented at this time.

Developer Installation

Prerequisites

npm install --save vispy

You can also automatically rebuild the package by running:

npm run watch