The main purpose of this tool is to automate image optimization and analysis. This project is recent, so use it with care, I'm listening to all feedback (we can talk via twitter, don't follow me I never tweet).

What is the difference with ImageMin? I think that if you deal with large applications, then go on ImageMin, but if you are building small static websites and you want to optimize your resources easily, then you could try lepto.

You give to lepto your input and output directories, the plugins you want to use and their options. Then lepto does his job, you keep your original files and the structure of the input directory. Some plugins can even collect data (like primary colors) from your images and save them in a JSON file.

If you want to learn more about image optimizations, I recommend to you the amazing images.guide by Addy Osmani.

Design and illustrations by Dorian Colin.

Get started with CLI / NPM scripts

You can follow this Get-started dev.to article: Learn How to Automate your Images Optimization Process with Lepto.

I recommend you to use lepto via lepto-cli, so it can easily be integrated into your build process with npm scripts.

$ npm i -g lepto-cli

Then you can follow the setup process:

$ lepto setup

It will guide you to create a configuration file.

Check out lepto-cli repository for more information.

See below for Node.js API usage.

GUI

You can access the GUI if you launched lepto from the CLI, by default at the address http://localhost:4490. You can change the port with the option guiPort.

The purpose of the GUI is to add more precise quality settings to files one by one. You can easily play with the quality slider and see the result at the same time, so you can choose the most suitable option for each of your resources.

You can also edit your filters and plugins configuration thought the interface.

To save the changes and relaunch lepto's process, click on the Save button or press ⌘S / Ctrl+S.

Plugins

You have to say to lepto how it will process your files. So you set in your config a filters array containing several groups of plugins associated with their glob. For each image, Lepto will go through the list of filters and test if the file matches each of the plugins groups. If it matches the same plugin several times, the last one will overwrite the parameters of the previous ones, but the plugin will keep its place in the order of process. Example:

/* letpo.config.json */
{
  "input": "assets/input",
  "output": "assets/output",
  "watch": true,
  "filters": [
    {
      "glob": "**/*.*",
      "use": [
        {
          "name": "lepto.jpeg",
          "quality": 80
        }, {
          "name": "lepto-resize",
          "maxWidth": 1200
        }
      ]
    }, {
      "glob": "*.jpg",
      "use": [
        {
          "name": "lepto-resize",
          "height": 100
        }, {
          "name": "lepto.jpeg",
          "quality": 60
        }
      ]
    }
  ]
}

The file photo.jpg, will be processed with this list of plugins:

[
  {
    "name": "lepto.jpeg",
    "quality": 60
  }, {
    "name": "lepto-resize",
    "height": 100
  }
]

So the first order of appearance is preserved, but the settings are overridden by the last ones.

If you want to use a plugin more than one time, you can add # at the end of its name, so it's creating like a "new plugin", eg: "lepto-resize#retina".

You can disable a plugin by setting its disabled option to true.

Built-in plugins

Lepto carries some built-in plugins, their name is prefixed by "lepto.". These plugins don't create more files than they receive. Their only goal is to optimize files size, they can't output a larger file.

"lepto.jpeg"

It uses sharp. Default config:

{
  "name": "lepto.jpeg",
  "quality": 80, /* From 1 to 100 */
  "progressive": true,
  "forceExt": null /* You can force the same extension for all jpgs file, eg: replace all .jpeg images by setting forceExt to "jpg" */
}

"lepto.png"

It uses node-pngquant. Default config:

{
  "name": "lepto.png",
  "quality": "70-80", /* From 0 (worst) to 100 (better) */
  "colors": 256, /* From 2 to 256 */
  "speed": 3 /* From 1 (faster but heavier) to 10 (slower but lighter) */
}

"lepto.gif"

It uses ImageMin's implementation of gifsicle: gifsicle. Default config:

{
  "name": "lepto.gif",
  "colors": 256 /* From 2 to 256 */
}

"lepto.svg"

It uses svgo, and his config follows the SVGO's config.

Additional plugins

• lepto-resize To resize and create retina alternatives
• lepto-webp To create .webp alternatives
• lepto-vibrant-color To collect the vibrant colors from your images using node-vibrant and save them inside your data JSON file. So you can set a placeholder background color to your images while they are loading.

Config

Default config:

{
  "input": null, /* Input directory */
  "output": null, /* Output directory */
  "filters": [], /* The list of filters associated with their plugins */

  "ignore": null, /* An array or an unique glob string */
  "watch": false, /* Watch for input file changes */
  "watchConfig": false, /* Watch for config file change to automatically update it */
  "followUnlink": false, /* Remove output files when the source file is deleted from the input directory */
  "processAll": true, /* Process all files on launch, recommended if followUnlink activated */
  "logLevel": "all", /* all 0-3 (0: silent, 1: only errors, ..., 3: all) */
  
  "gui": true, /* The GUI can be disabled */
  "openGui": false, /* Automatically open the GUI in your default browser */
  "guiPort": "4490", /* GUI port */

  "dataOutput": null, /* Path of your data json file, eg: output/data.json */
  "dataRootPath": null /* Relative path removed from file names inside the data json file */
}

Lepto watch files by default when launched with lepto-cli.

Node.js API

$ npm i -D lepto

You simply have to call lepto() by giving it your config.

const lepto = require('lepto');

const runner = lepto({
  input: 'assets/input',
  output: 'assets/output',
  filters: [

  ]
  /* ...config */
})

Now you can listen to events with the on(event, callback) method, the events are all for all events, success, info, warn, error and processed-file.

The success, info and warn events gives an object with a msg inside, eg: { msg: 'Info message' }.

The error just gives a string of the error message.

The processed-file gives an object with information about the file process:

{
  adj: 'new', /* file watch event: '' (initial process), 'new' or 'changed' */
  input: 'icons/github.png',
  inputSize: 1000000, /* sizes in bytes */
  output: [ 'icons/github.png', 'icons/github@2x.png' ],
  outputSizes: [ 20000, 50000 ],
  timeSpent: 300 /* process time in ms */
}

Example of events integration:

runner.on('error', msg => {
  console.error(msg);
});
runner.on('processed-file', data => {
  /* deal with data */
});
runner.on('all', (data, event) => { /* When listening to 'all' events, the callback receive the event name as a second argument */
  if (typeof data.msg !== 'undefined') {
    console.log(`${event}: ${data.msg}`);
  }
});

Contributing

I really like to save people time. That's why I created this tool:

• To help developers easily optimize their images
• To make sites load faster

So if you have any suggestion that could help people use this tool faster, tell me!

Lepto Build process

Run npm test for testing the tool.

There is only a build step for the GUI part that can be launched with the npm start command. It will watch for CSS and js files changes from the gui/src/ directory and compiles them into gui/dist/ with Babel and PostCSS.

Because you could ask yourself the question: I love React but I didn't use it for the GUI because I had planned to deal with many contenteditable elements that are terrible to work with React. The part of the is messy I admit it, I have to tidy up.

Lepto Plugin writing

A lepto plugin has to deal with multiples output files associated to one input file, a plugin is a function called with the plugins options that must return a function that will process the files. This last function receives an input object, a fulfill method and an object of utils methods.

The input object looks like that:

input = {
  input: 'icons/social/github.png',
  outputs: [
    {
      dir: 'icons/social',
      filename: 'github.png',
      buffer: <Buffer>
    },
    {
      dir: 'icons/social',
      filename: 'github@2x.png',
      buffer: <Buffer>
    }
  ],
  data: {}
};

If the plugin is the first called, it will receive only one output, additional outputs are created by others plugins.

A data object is shared between plugins during the process of files, his content will be saved to a JSON file chosen by the user.

Because lepto plugins have to deal with multiples outputs Buffer and often with an async process, I suggest you this model:

const namePlugin = (opts={}) => {
  return function name(input, fulfill, utils) {
    const next = () => {
      finish--;
      if (finish <= 0) {
        fulfill(input);
      }
    };

    let finish = input.outputs.length;
    for (const i in input.outputs) {
      if (Object.prototype.hasOwnProperty.call(input.outputs, i)) {
        optimizer(input.outputs[i].buffer).then(function optimizerThen(i) {
          return function optimizerSuccess(buffer) {
            input.outputs[i].buffer = buffer;
            next();
          };
        }(i));
      }
    }
  };
};

module.exports = namePlugin;

Utils functions:

• utils.size(Buffer) return an object like { width: 100, height: 100 }.
• utils.sharp(Buffer) sharp node module.
• utils.mime(Buffer) return the mime type as a string, eg: "image/jpeg", learn more here.
• utils.base(String) return the base name of a file name, eg: "IMG001.JPG" > "IMG001".
• utils.ext(String) return the extension of a file name, eg: "IMG001.JPG" > "JPG".

You can inspire yourself with the built-in plugins.

License

This project is licensed under the MIT license.