litegraph.js
A graph node engine and editor written in Javascript similar to PD or UDK Blueprints, comes with its own editor in HTML5 Canvas2D. The engine can run client side or server side using Node. It allows to export graphs as JSONs to be included in applications independently.
Top Related Projects
Quick Overview
LiteGraph.js is a JavaScript library for creating and visualizing node-based graphs. It provides a user-friendly interface for creating, connecting, and manipulating nodes in a graph-like structure, making it ideal for visual programming, data flow diagrams, and other node-based applications.
Pros
- Easy to use and integrate into web applications
- Highly customizable with support for custom node types and styles
- Supports real-time graph execution and data flow
- Lightweight and performant, suitable for complex graphs
Cons
- Limited documentation and examples for advanced usage
- May require additional effort to implement complex node behaviors
- Not actively maintained, with infrequent updates
- Lacks built-in support for some advanced graph features (e.g., subgraphs)
Code Examples
Creating a basic graph:
const graph = new LGraph();
const canvas = new LGraphCanvas("#mycanvas", graph);
const nodeA = LiteGraph.createNode("basic/const");
nodeA.pos = [100, 100];
graph.add(nodeA);
const nodeB = LiteGraph.createNode("basic/watch");
nodeB.pos = [400, 100];
graph.add(nodeB);
nodeA.connect(0, nodeB, 0);
Adding a custom node type:
function MyCustomNode() {
this.addInput("in", "number");
this.addOutput("out", "number");
}
MyCustomNode.prototype.onExecute = function() {
const input = this.getInputData(0);
this.setOutputData(0, input * 2);
};
LiteGraph.registerNodeType("custom/doubler", MyCustomNode);
Executing the graph:
graph.start();
// Update graph in animation loop
function update() {
graph.runStep();
requestAnimationFrame(update);
}
requestAnimationFrame(update);
Getting Started
- Include LiteGraph.js and LiteGraph.css in your HTML:
<link rel="stylesheet" type="text/css" href="litegraph.css">
<script src="litegraph.js"></script>
- Create a canvas element in your HTML:
<canvas id="mycanvas" width="1024" height="720"></canvas>
- Initialize LiteGraph in your JavaScript:
const graph = new LGraph();
const canvas = new LGraphCanvas("#mycanvas", graph);
// Add nodes and connections here
graph.start();
Competitor Comparisons
JavaScript framework for visual programming
Pros of Rete
- More active development and community support
- Better TypeScript integration and type definitions
- Modular architecture with plugins for extended functionality
Cons of Rete
- Steeper learning curve for beginners
- Less comprehensive documentation compared to LiteGraph.js
- Smaller built-in node library
Code Comparison
LiteGraph.js:
var node = LiteGraph.createNode("basic/const");
node.pos = [100,100];
graph.add(node);
Rete:
const node = new Rete.Node('Example');
node.addInput(new Rete.Input('input', 'Input', numSocket));
node.addControl(new NumControl(this.editor, 'num'));
editor.addNode(node);
Both libraries allow for creating and adding nodes to a graph, but Rete's approach is more object-oriented and modular. LiteGraph.js uses a more straightforward, functional style, which may be easier for beginners to grasp quickly.
Rete offers more flexibility in terms of customization and extensibility, while LiteGraph.js provides a simpler API for basic graph operations. The choice between the two depends on the project's complexity, required features, and the developer's familiarity with node-based programming concepts.
Hybrid visual and textual functional programming.
Pros of Enso
- More comprehensive visual programming environment with a broader scope
- Supports multiple programming languages and data processing capabilities
- Active development with regular updates and a growing community
Cons of Enso
- Steeper learning curve due to its more complex nature
- Requires more system resources to run compared to LiteGraph.js
- Less suitable for lightweight, browser-based applications
Code Comparison
LiteGraph.js:
var node = LiteGraph.createNode("basic/const");
node.pos = [100,100];
graph.add(node);
Enso:
main =
Http.get "https://api.example.com/data"
|> Json.decode
|> Table.fromJson
|> Table.filter (row -> row.value > 10)
|> Chart.scatter .x .y
While LiteGraph.js focuses on creating and connecting nodes in a graph, Enso's code demonstrates a more complex data processing pipeline with built-in functions for HTTP requests, JSON parsing, and data visualization.
Literate scientific computing and communication for the web
Pros of Iodide
- Focuses on interactive scientific computing and data visualization in the browser
- Supports multiple programming languages, including Python and R
- Provides a notebook-like interface for data exploration and analysis
Cons of Iodide
- Less suitable for general-purpose node-based programming
- May have a steeper learning curve for users not familiar with scientific computing
- Limited to web-based environments
Code Comparison
Iodide (Python code in a notebook cell):
import numpy as np
import matplotlib.pyplot as plt
x = np.linspace(0, 10, 100)
y = np.sin(x)
plt.plot(x, y)
plt.show()
Litegraph.js (JavaScript code for creating a node):
LiteGraph.registerNodeType("math/sin", {
title: "Sin",
inputs: [["in", "number"]],
outputs: [["out", "number"]],
onExecute: function() {
this.setOutputData(0, Math.sin(this.getInputData(0)));
}
});
While Iodide is geared towards scientific computing with a notebook interface, Litegraph.js provides a more flexible node-based programming environment for various applications. Iodide excels in data analysis and visualization, whereas Litegraph.js is better suited for creating visual programming interfaces and flow-based systems.
Flow-based programming for JavaScript
Pros of NoFlo
- More comprehensive flow-based programming framework
- Supports multiple runtime environments (browser, Node.js, etc.)
- Larger ecosystem with extensive component libraries
Cons of NoFlo
- Steeper learning curve due to its more complex architecture
- Less focus on visual graph editing compared to LiteGraph.js
- Requires more setup and configuration for basic usage
Code Comparison
NoFlo:
const noflo = require('noflo');
const graph = new noflo.Graph('MyGraph');
graph.addNode('Add', 'math/Add');
graph.addNode('Multiply', 'math/Multiply');
graph.addEdge('Add', 'sum', 'Multiply', 'multiplicand');
LiteGraph.js:
const graph = new LGraph();
const addNode = LiteGraph.createNode("basic/sum");
const multiplyNode = LiteGraph.createNode("basic/mult");
graph.add(addNode);
graph.add(multiplyNode);
addNode.connect(0, multiplyNode, 0);
Both libraries provide ways to create and manipulate graphs, but NoFlo's approach is more focused on flow-based programming concepts, while LiteGraph.js is more oriented towards visual node-based editing. NoFlo offers a more extensive framework for complex applications, while LiteGraph.js provides a simpler, more lightweight solution for visual graph editing.
Convert designs to code with AI
Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.
Try Visual CopilotREADME
litegraph.js
A library in Javascript to create graphs in the browser similar to Unreal Blueprints. Nodes can be programmed easily and it includes an editor to construct and tests the graphs.
It can be integrated easily in any existing web applications and graphs can be run without the need of the editor.
Try it in the demo site.
Features
- Renders on Canvas2D (zoom in/out and panning, easy to render complex interfaces, can be used inside a WebGLTexture)
- Easy to use editor (searchbox, keyboard shortcuts, multiple selection, context menu, ...)
- Optimized to support hundreds of nodes per graph (on editor but also on execution)
- Customizable theme (colors, shapes, background)
- Callbacks to personalize every action/drawing/event of nodes
- Subgraphs (nodes that contain graphs themselves)
- Live mode system (hides the graph but calls nodes to render whatever they want, useful to create UIs)
- Graphs can be executed in NodeJS
- Highly customizable nodes (color, shape, slots vertical or horizontal, widgets, custom rendering)
- Easy to integrate in any JS application (one single file, no dependencies)
- Typescript support
Nodes provided
Although it is easy to create new node types, LiteGraph comes with some default nodes that could be useful for many cases:
- Interface (Widgets)
- Math (trigonometry, math operations)
- Audio (AudioAPI and MIDI)
- 3D Graphics (Postprocessing in WebGL)
- Input (read Gamepad)
Installation
You can install it using npm
npm install litegraph.js
Or downloading the build/litegraph.js
and css/litegraph.css
version from this repository.
First project
<html>
<head>
<link rel="stylesheet" type="text/css" href="litegraph.css">
<script type="text/javascript" src="litegraph.js"></script>
</head>
<body style='width:100%; height:100%'>
<canvas id='mycanvas' width='1024' height='720' style='border: 1px solid'></canvas>
<script>
var graph = new LGraph();
var canvas = new LGraphCanvas("#mycanvas", graph);
var node_const = LiteGraph.createNode("basic/const");
node_const.pos = [200,200];
graph.add(node_const);
node_const.setValue(4.5);
var node_watch = LiteGraph.createNode("basic/watch");
node_watch.pos = [700,200];
graph.add(node_watch);
node_const.connect(0, node_watch, 0 );
graph.start()
</script>
</body>
</html>
How to code a new Node type
Here is an example of how to build a node that sums two inputs:
//node constructor class
function MyAddNode()
{
this.addInput("A","number");
this.addInput("B","number");
this.addOutput("A+B","number");
this.properties = { precision: 1 };
}
//name to show
MyAddNode.title = "Sum";
//function to call when the node is executed
MyAddNode.prototype.onExecute = function()
{
var A = this.getInputData(0);
if( A === undefined )
A = 0;
var B = this.getInputData(1);
if( B === undefined )
B = 0;
this.setOutputData( 0, A + B );
}
//register in the system
LiteGraph.registerNodeType("basic/sum", MyAddNode );
or you can wrap an existing function:
function sum(a,b)
{
return a+b;
}
LiteGraph.wrapFunctionAsNode("math/sum",sum, ["Number","Number"],"Number");
Server side
It also works server-side using NodeJS although some nodes do not work in server (audio, graphics, input, etc).
var LiteGraph = require("./litegraph.js").LiteGraph;
var graph = new LiteGraph.LGraph();
var node_time = LiteGraph.createNode("basic/time");
graph.add(node_time);
var node_console = LiteGraph.createNode("basic/console");
node_console.mode = LiteGraph.ALWAYS;
graph.add(node_console);
node_time.connect( 0, node_console, 1 );
graph.start()
Projects using it
comfyUI
webglstudio.org
MOI Elephant
Mynodes
Utils
It includes several commands in the utils folder to generate doc, check errors and build minifyed version.
Demo
The demo includes some examples of graphs. In order to try them you can visit demo site or install it on your local computer, to do so you need git
, node
and npm
. Given those dependencies are installed, run the following commands to try it out:
$ git clone https://github.com/jagenjo/litegraph.js.git
$ cd litegraph.js
$ npm install
$ node utils/server.js
Example app listening on port 80!
Open your browser and point it to http://localhost:8000/. You can select a demo from the dropdown at the top of the page.
Feedback
You can write any feedback to javi.agenjo@gmail.com
Contributors
- atlasan
- kriffe
- rappestad
- InventivetalentDev
- NateScarlet
- coderofsalvation
- ilyabesk
- gausszhou
Top Related Projects
Convert designs to code with AI
Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.
Try Visual Copilot