Graph visualization library that enables to customize nodes using HTML

Visit live demo

Instead of connecting nodes dirrectly this library uses concept of ports, which provide greater fexibility at managing connections. Port is an entity of the node to which connection can be attached to.

This library fits for tasks where easy nodes customization and interactiveness are required.

Features:

• easy nodes customization using HTML
• wide configuration options out of the box
• draggable and scalable canvas with draggable nodes
• exhaustive set of examples
• typescript support
• mobile devices support

Getting started:

npm i @html-graph/core

import {
  CanvasCore,
  TransformableCanvas,
  DraggableNodesCanvas,
  ApiPortPayload,
} from "@html-graph/core";

const canvasElement = document.getElementById("canvas")!;

const canvas = new TransformableCanvas(
  new DraggableNodesCanvas(
    new CanvasCore({
      background: { type: "dots" },
    }),
  ),
);

function createNode(
  name: string,
  frontPortId: string,
  backPortId: string,
): [HTMLElement, Record<string, ApiPortPayload>] {
  const node = document.createElement("div");
  node.classList.add("node");

  const frontPort = document.createElement("div");
  frontPort.classList.add("port");
  node.appendChild(frontPort);

  const text = document.createElement("div");
  text.innerText = name;
  node.appendChild(text);

  const backPort = document.createElement("div");
  backPort.classList.add("port");
  node.appendChild(backPort);

  return [node, { [frontPortId]: frontPort, [backPortId]: backPort }];
}

const [node1, ports1] = createNode("Node 1", "port-1-1", "port-1-2");
const [node2, ports2] = createNode("Node 2", "port-2-1", "port-2-2");

canvas
  .attach(canvasElement)
  .addNode({ element: node1, x: 200, y: 400, ports: ports1 })
  .addNode({ element: node2, x: 600, y: 500, ports: ports2 })
  .addConnection({ from: "port-1-2", to: "port-2-1" });

Refer to Examples for more.

Run examples locally

Use node version >= 20

npm install

npm run start

or

docker compose up

Open http://localhost:3100