Artificial-Engineering/Nidium

Name: Nidium

Owner: Artificial Engineering

Description: nidium is an ongoing effort for a mobile hw-accelerated rendering engine to create apps and games. Embedding Mozilla JavaScript VM, Google Skia, Facebook Yoga. https://twitter.com/nidiumproject for updates

Created: 2018-01-27 15:42:00.0

Updated: 2018-01-27 15:42:03.0

Pushed: 2018-01-30 07:45:33.0

Homepage: https://www.nidium.com

Size: 12068

Language: C++

GitHub Committers

UserMost Recent Commit# Commits

Other Committers

UserEmailMost Recent Commit# Commits

README


nidium

Introduction

nidium is an ongoing effort for a general purpose rendering engine to create apps and games that run on both desktop and mobile. It also offers a way to create server-side applications through nidium-server which shares all the non graphics related code base.

nidium helps you create graphical software with Javascript. It's not a NodeJS, QT, Chromium, or WebKit derivative. It has been designed from scratch and has a small codebase written in C++.

It leverages the combination of Skia Graphics from Google, and Mozilla's JavaScript Engine (SpiderMonkey) together with many more awesome libraries.

It supports various common and well known APIs such as:

And some other non standard:

It can seamlessly run various JavaScript libraries like Three.JS, PixiJS, Phaser (and probably a lot more without much modification).

What problem(s) is nidium trying to solve

To put it simply, nidium aims to expose something close to web technology but in an easily hackable environment.

We're not trying to copy the Web. Think of it as a sandbox where you can quickly prototype new things, without any dependency to blink/webkit (e.g. like Electron does). This allows us to add various optimizations and non standard things. Moreover, nidium is small (nidium statically linked with all of its dependencies is about 20 MB).

Also, nidium aims to target a large spectrum of devices and our short term goal is to run it on low-power devices (mobile, rpi, …)

By the way, it uses SpiderMonkey as its JavaScript engine, and we believe that Mozilla needs some love too!

Work in progress

nidium is still a work in progress and several key features are missing:

Layout engine

nidium ships with its own layout engine. That is, every element has its own memory buffer (retained mode), pretty similar to HTML elements.

The layout engine allows several “layout” operations to be made on each element (relative position, opacity, drag'n'drop, margin, scrolling, overflow, and so on).

It also implements a simple HTML DOM compatility layer and is mature enough to run a MVVM framework like Vue.js and probably React with little to no effort.

Network and event loop

nidium uses its own high-performance library (libapenetwork) which handles all the networking operations in a non-blocking, async way.

Download nidium

You can download nidium binaries for Linux and OSX from the download page.

Building nidium

To build nidium you need at least 5.7GB of disk space. A build from scratch may take 30 to 90 minutes, depending of the speed of your computer.

On a debian based system, a few commands will get you started.

t-get install libpci-dev python2.7 git make patch clang pkg-config libgtk2.0-dev libgtk-3-dev mesa-common-dev libglu1-mesa-dev libosmesa6-dev yasm libasound2 libasound2-dev libbz2-1.0
t clone https://github.com/nidium/NidiumTools.git
t clone --recursive https://github.com/nidium/Nidium.git
port PYTHONPATH=$(pwd)/NidiumTools/src
 Nidium
configure_frontend
bin/nidium
Documentation

We strive to have excellent documentation, both for our getting started guide and for the API Reference.

Videos - nidium in action
Bug reports and Collaboration

Feel free to report any bug or issue to us.

If you feel that something is not clear, that could be an issue worth reporting and solving.

License

Copyright 2017 Nidium Inc. All rights reserved. Use of this source code is governed by a MIT license that can be found in the LICENSE file.


This work is supported by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number U24TR002306. This work is solely the responsibility of the creators and does not necessarily represent the official views of the National Institutes of Health.