Minidenticons

Super lightweight SVG identicon generator. No dependencies.

^{(using the minidenticon() function only)}

^{(using the included custom element)}

Why

• Generate identicons (pixelated avatars) on the client from usernames instead of fetching images from a server. Much faster, saves bandwidth and GDPR compliant!
• Replace dull "initials" avatars by an easily remembered graphical avatar
• Give a visual representation of hashes or any ID string

Live Demo 🎮

Play with it here.

Basic usage with the included custom element

Minidenticons uses ES modules, now widely supported in browsers. Import the minidenticonSvg custom element from the minidenticons.min.js file. This file can be located in a CDN (example below) or copied in any directory of your website (for better performance and to be GDPR compliant, since you don’t have to connect to a third party server).

<script type="module">
  import { minidenticonSvg } from 'https://cdn.jsdelivr.net/npm/[email protected]/minidenticons.min.js'
</script>

Then simply use minidenticon-svg tags with a username attribute 😂

<minidenticon-svg username="laurent"></minidenticon-svg>

^{(Note that like for all elements except void elements, the closing tag </minidenticon-svg> is required)}

For instance with the laurent username you will get the following identicon (without the border):

Styling

For easy CSS styling:

• Minidenticons are SVG images that will take all the space available. The picture above is resized.

• The background is transparent.

• There is white space around the colored square matrix to allow uncropped circle avatars.

So with the following CSS:

minidenticon-svg svg {
  border-radius: 50%;
  background-color: grey;
  height: 48px;
  width: 48px;
};

You will get:

By default the color saturation is set to 95% and the lightness is set to 45%. But you can change these values with the saturation and/or lightness attributes, for instance:

<minidenticon-svg username="laurent" saturation="60" lightness="50"></minidenticon-svg>

Play with the demo to find a combination of saturation and lightness that matches your website theme colors: light, dark or whatever 😎

Performance

• Custom element identicons are memoized (stored in memory so that it does not need to be recalculated).
• To see how long it takes for your browser to generate 100 identicon custom elements (for a big page) try out the online browser benchmark.

Advanced usage with the minidenticon() function

Instead of using the custom element, you can also use the minidenticon() function to generate SVG strings in your client (or your server).

minidenticon(seed: string, saturation?: number|string, lightness?: number|string): string

The minidenticon() function will return a SVG string generated from its seed string argument. The seed argument can be a username, but actually any string used as an identifier.

Optional saturation and lightness arguments should be percentages; that is, numbers (or strings) between 0 and 100.

Note that the minidenticon() function itself is not memoized.

NodeJS

Be sure to use a NodeJS version greater or equal to 15.14.0.

Installation

npm install minidenticons

Import

import { minidenticon } from 'minidenticons'

The minidenticonSvg custom element should be tree-shaken from your bundle, for an even smaller size of minidenticons 😁

Performance

The minidenticon() function is fast. You can see by yourself if you run node benchmark/node at the root of a Minidenticons git clone. On my machine I get the following result:

Time to generate 10000 minidenticon SVG strings for 15 characters random seeds:
8 milliseconds (10 runs average)

React

The following React component example inserts the identicon into an img tag src attribute. It also uses React useMemo to memoize the identicon.

import { minidenticon } from 'minidenticons'
import { useMemo } from 'react'

const MinidenticonImg = ({ username, saturation, lightness, ...props }) => {
  const svgURI = useMemo(
    () => 'data:image/svg+xml;utf8,' + encodeURIComponent(minidenticon(username, saturation, lightness)),
    [username, saturation, lightness]
  )
  return (<img src={svgURI} alt={username} {...props} />)
}

You can then use this component with img props such as width and height along with minidenticons ones. All props except username are optional.

<MinidenticonImg username="laurent" saturation="90" width="150" height="150" />

For a TypeScript version of this example see the original issue comment by Dan Yishai.

Workbox

In this example using Workbox, images with a path ending with minidenticons/<username>.svg are generated by the service worker and cached for one year.

import { minidenticon } from 'minidenticons'
import { registerRoute } from 'workbox-routing'

registerRoute(
  /minidenticons\/[^\/]+\.svg$/,
  async ({ url }) => {
    const username = url.pathname.match(/([^\/]+)\.svg$/)[1]
    return new Response(
      minidenticon(username),
      { headers: { "Content-Type": "image/svg+xml", "Cache-Control": "max-age=31536000" } }
    )
  }
)

Elm

For Elm enthusiasts there is a Minidenticons package on the Elm package repository: minidenticons-elm.

Collisions

You will always get the same identicon for a given username. But it is not impossible to have different usernames with the same identicon. That's a collision.

Generated identicons are 5×5 matrices with vertical symmetry, and can have 9 different hues for the same saturation and lightness. This means there are 2^(3×5) × 9 = 294,912 different identicons possible, but duplicate identicons are inevitable when using a lot of them. It shouldn’t matter as identicons should not be used solely to identify an user, and should always be coupled to a unique username 😉

The npm test command results below show that you have less than a 2 percent chance to generate a duplicate identicon when already using 10,000 of them.

0 collisions out of 100 (0.00%)
0 collisions out of 200 (0.00%)
0 collisions out of 300 (0.00%)
0 collisions out of 400 (0.00%)
0 collisions out of 500 (0.00%)
0 collisions out of 600 (0.00%)
0 collisions out of 700 (0.00%)
0 collisions out of 800 (0.00%)
1 collisions out of 900 (0.11%)
1 collisions out of 1000 (0.10%)
8 collisions out of 2000 (0.40%)
14 collisions out of 3000 (0.47%)
22 collisions out of 4000 (0.55%)
37 collisions out of 5000 (0.74%)
58 collisions out of 6000 (0.97%)
75 collisions out of 7000 (1.07%)
99 collisions out of 8000 (1.24%)
129 collisions out of 9000 (1.43%)
163 collisions out of 10000 (1.63%)

License

MIT

Stargazers ❤️