Skip to content

gramener/gramex-network

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

34 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

@gramex/network

A force-directed network or graph visualization.

See the workshop for a step-by-step tutorial.

Example

Given this table of countries and religions:

Country-religion dataset screenshot

... we can render the following network:

Religion network visual

Here is the source code for the network above

Installation

Install via npm:

npm install @gramex/network@2

Use locally as an ES module:

<script type="module">
  import { network } from "./node_modules/@gramex/network/dist/network.js";
</script>

Use locally as a script:

<script src="./node_modules/@gramex/network/dist/network.min.js"></script>
<script>
  gramex.network(...)
</script>

Use via CDN as an ES Module:

<script type="module">
  import { network } from "https://cdn.jsdelivr.net/npm/@gramex/network@2";
</script>

Use via CDN as a script:

<script src="https://cdn.jsdelivr.net/npm/@gramex/network@2/dist/network.min.js"></script>
<script>
  gramex.network(...)
</script>

Use a node-link dataset

The network() function accepts a { nodes, links } object.

  • nodes is an array of objects.
  • links is an array of { source, target } objects that to the node by index number or by reference.
{
  "nodes": [{ "name": "Alice" }, { "name": "Bob" }, { "name": "Carol" }],
  "links": [
    { "source": 0, "target": 1 },
    { "source": 1, "target": 2 }
  ]
}

If nodes has an id key, you can specify the links using id:

{
  "nodes": [{ "id": "Alice" }, { "id": "Bob" }, { "id": "Carol" }],
  "links": [
    { "source": "Alice", "target": "Bob" },
    { "source": "Bob", "target": "Carol" }
  ]
}

Here is a simple network that draws the above dataset:

const graph = network("#network", data);

Example

See how to use network()

Style nodes and links

Update the attributes and styles of the returned .nodes and .links joins to style the nodes and links.

By default, the nodes are rendered as <circle> elements and links as <line> elements.

You can apply the D3 .attr, .classed, .style, .text, and any other selection methods to style the elements.

You can use any node/link keys in the styles. For example:

const graph = network("#network", data);
graph.nodes.attr("r", (d) => d.depth);

Example

See how to style nodes and links

Add tooltips

You can use Bootstrap tooltips.

  1. Add a data-bs-toggle="tooltip" title="..." attribute to each feature using update
  2. Call new bootstrap.Tooltip(element, {selector: '[data-bs-toggle="tooltip"]'}) to initialize tooltips

Example

See how to add tooltips

Dragging and pinning

Dragging a node pins it where you release it. Double-click a pinned node to unpin. To unpin all nodes, double-click on the background.

When dragging, the node gets a dragging class. When pinned, it gets a pinned class. You can use this to style nodes that are dragged or pinned. For example:

.dragging {
  stroke: black;
  stroke-width: 5;
}
.pinned {
  stroke: black;
  stroke-width: 3;
}

Example

See how to highlight dragged and pinned nodes

Filter nodes and links

To dynamically filter nodes and links, pass a subset of the SAME nodes and links.

Make sure the nodes and links are the same objects as the original nodes and links. This ensures that the simulation is not restarted.

In this example, when you move the slider, the country - religion links are filtered based on population. Any isolated nodes are also removed.

Example

See how to filter nodes and links

Tabular data

Any tabular data can be converted into a node-link structure. For example, take this table of countries and religions:

Country-religion dataset screenshot

Convert this into a flat array of objects like this:

const data = [
  { Country: "USA", Religion: "Christian", Population: 100 },
  { Country: "UK", Religion: "Christian", Population: 90 },
  { Country: "Iran", Religion: "Muslim", Population: 80 },
];

Now you can convert it to a node-link dataset using { nodes, links } = kpartite(data, keys, values). It accepts 3 parameters:

  1. data - array of objects containing the data.
  2. keys - object of {key: column} pairs or an array of [key, column] pairs.
    • key is a string node type
    • column is the string name of a field in data, or a function(object) that returns the field, or a static value.
  3. values - object of accessor functions for link values that are aggregated across links and nodes

For example:

const { nodes, links } = kpartite(
  data,
  {
    Country: "Country", // Create a node for each country
    Religion: (d) => d.Religion, // Create a node for each religion
  },
  {
    count: 1, // Count the number of links between countries and religions
    Population: "Population", // Sum the population of countries and religions
  },
);

This creates the following nodes:

[
  {
    key: "Country",
    value: "USA",
    id: '["Country","USA"]',
    count: 1,
    Population: 100,
  },
  {
    key: "Religion",
    value: "Christian",
    id: '["Religion","Christian"]',
    count: 2,
    Population: 190,
  },
  // ... etc.
];

... and the following links:

[
  {
    source: {}, // link to USA source node
    target: {}, // link to Christian target node
    id: '["[\\"Country\\",\\"USA\\"]","[\\"Religion\\",\\"Christian\\"]"]',
    count: 1,
    Population: 100,
  },
  // ... etc.
];

Tabular unipartite data

If you have a relationships between the same entity (e.g. people connected with people, countries trading with countries), use kpartite() as follows:

// Create nodes and links, mapping the "name" key with both "Column A" and "Column B"
const { nodes, links } = kpartite(
  data,
  [
    ["name", "Column A"],
    ["name", "Column B"],
  ],
  { count: 1 },
);

Here is an example with the Friends sexual partners data. (Interestingly, Rachel is the only one who doesn't share a sexual partner with any of the others.)

Example

See how to generate unipartite data

Here is an example with the Arxiv paper coauthors data of Computational Lingustics papers that have "LLM" in their title, and authored 2+ papers together.

Example

See how to generate unipartite data

Forces

By default, the network uses the following forces:

  • link: a d3.forceLink(links) that links the nodes and links
  • charge: a d3.forceManyBody() that pushes nodes apart
  • x: a d3.forceX(width / 2) that centers the nodes horizontally
  • y: a d3.forceY(height / 2) that centers the nodes vertically

To disable these or modify the forces, pass a force object with the required forces. For example, this network uses a d3.forceCenter() force instead of x and y:

forces: {
  x: false,
  y: false,
  center: ({ width, height }) => d3.forceCenter(width / 2, height / 2),
}

This network uses a d3.forceCollide() to prevent nodes from overlapping, and a stronger d3.forceManyBody() replacing the default charge force to push nodes apart:

forces: {
  charge: () => d3.forceManyBody().strength(-20),
  collide: () => d3.forceCollide().radius(d => 22 - d.depth * 4).iterations(3),
},

Each force is a function({ nodes, links, width, height }) where width and height are the size of the SVG. It should return a D3 force.

Example

See how to modify forces

Brushing

Passing a brush function enables brushing. The brush function is called with the selected nodes as parameters. You can use this to update other visualizations.

Example

See how to handle brushing

Node shapes and labels

By default, nodes are <circle> elements. Change nodeTag for a different shape. For example, to use <text> elements:

const graph = network("#network", { nodes, links, nodeTag: "text" });
graph.nodes.text((d) => d.id);

Here is a detailed example on how to draw labels with text and a rectangle background:

Example

See how to use different node shapes

Arrows

To add arrows, add a triangle <marker> to your SVG's <defs> like below:

<svg>
  <defs>
    <marker
      id="triangle"
      viewBox="0 0 10 10"
      refX="20"
      refY="5"
      markerUnits="strokeWidth"
      markerWidth="8"
      markerHeight="8"
      orient="auto"
    >
      <path d="M 0 0 L 10 5 L 0 10 z" fill="rgba(0,0,0,0.2)" />
    </marker>
  </defs>
</svg>

Here:

  • refX is the distance from the end of the line to the tip of the arrow
  • refY is the distance from the center of the line to the center of the arrow
  • markerWidth and markerHeight are the size of the arrow
  • orient is the direction of the arrow. auto means the arrow points in the direction of the line

Then add graph.links.attr("marker-end", "url(#triangle)") to your code.

Example

See how to draw arrows on links

Curved links

To draw curved links, set linkCurvature to a number between -1 and 1. 0 is a straight line. 1 is a half-circle. -1 is a half-circle in the opposite direction.

Example

See how to draw curved links

Stop and restart simulation

To stop or restart the simulation, call graph.simulation.stop() or graph.simulation.alphaTarget(0.3).restart().

Example

Bring your own D3

If you already have D3 loaded, or want to use a specific version / instance of D3, pass it to network(el, { d3 }):

Example

See how to use your own D3 version

React usage

Use the following pattern when using `network()`` with React:

const { useEffect } = React;
function App() {
  useEffect(() => network("#network", { ... }), []);
  return React.createElement("svg", { id: "network", width: 600, height: 380 });
}
const root = ReactDOM.createRoot(document.querySelector("#root"));
root.render(React.createElement(React.StrictMode, null, React.createElement(App)));

Example

See how to use network with React

Here are instructions to create a React Component:

npx create-react-app network-react
cd network-react
npm install d3 @gramex/network

Create src/NetworkComponent.js with this code:

See NetworkComponent.js code

Modify src/App.js as follows:

See App.js code

Then run npm start or npm run build.

Explore the code.

API

See API documentation

Release notes

  • 2.1.0: 31 May 2024. Support stop and restart
  • 2.0.0: 24 Nov 2023. Synchronous API. Bring your own D3. To migrate from v1:
    • Replace await network(...) with network(...)
    • Pass d3 as a param, e.g. network(el, { d3 })
  • 1.0.8: 14 Sep 2023. Stop past simulations on re-run. Use MIT license
  • 1.0.7: 7 Sep 2023. Support any node shape.
  • 1.0.6: 6 Sep 2023. Enable styles on pinned nodes
  • 1.0.5: 6 Sep 2023. Initial release

Authors

License

MIT