JSPM

  • ESM via JSPM
  • ES Module Entrypoint
  • Export Map
  • Keywords
  • License
  • Repository URL
  • TypeScript Types
  • README
  • Created
  • Published
  • Downloads 562
  • Score
    100M100P100Q97488F
  • License MIT

Loader module for webpack to execute your custom procedure. It works as your custom loader.

Package Exports

  • skeleton-loader

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (skeleton-loader) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

skeleton-loader

npm GitHub issues David license

Loader module for webpack to execute your custom procedure. It works as your custom loader.

By default, skeleton-loader only outputs the input content. When you specify a function, skeleton-loader executes your function with the input content, and outputs its result. The function does something, it might edit the content, it might parse the content and indicate something in a console, it might do anything else.

That is, you can specify a function in webpack configuration instead of writing new custom loader.

skeleton-loader is useful when:

  • You couldn't find a loader you want.
  • You don't want to write a special loader for your project.
  • You want to add something to the result of another loader.
  • You want to do additional editing.
  • etc.

For example:

// webpack.config.js
module.exports = {
  entry: './app.js',
  output: {
    filename: 'bundle.js'
  },
  module: {
    rules: [{
      test: /\.js$/,
      loader: 'skeleton-loader',
      options: {
        procedure: function(content) {
          // Change the input content, and output it.
          return (content + '').replace(/foo/g, 'bar');
        }
      }
    }]
  }
};
// webpack.config.js
// ...
test: /\.html$/,
// ...
// skeleton-loader options
options: {
  procedure: function(content) {
    // Remove all elements for testing from HTML.
    return (content + '').replace(/<div class="test">[^]*?<\/div>/g, '');
  },
  toCode: true
}
// webpack.config.js
// ...
test: /\.json$/,
// ...
// skeleton-loader options
options: {
  procedure: function(content) {
    var appConfig = JSON.parse(content);
    // Check and change JSON.
    console.log(appConfig.foo);
    appConfig.bar = 'PUBLISH';
    return appConfig;
  },
  toCode: true
}
// webpack.config.js
// ...
// skeleton-loader options
options: {
  // Asynchronous mode
  procedure: function(content, sourceMap, callback) {
    setTimeout(function() {
      callback(null, 'Edited: ' + content);
    }, 5000);
  }
}

Installation

npm install --save-dev skeleton-loader

Usage

Documentation:

Options

You can specify options via query parameters or an options (or skeletonLoader for webpack v1) object in webpack configuration.

procedure

Type: function
Default: undefined

A function to do something with the input content. The result of the function is output.
The function is passed the following arguments:

  • content
    The content of the resource file as string, or something that is passed by previous loader. That is, if another loader is chained in loaders list, the content that is passed by that loader might not be string.
  • sourceMap
    An optional value SourceMap as JavaScript object that might be passed by previous loader.
  • callback
    A callback function for asynchronous mode.
  • options
    Reference to current options. Also, this might contain options.resourceOptions.

In the function, this refers to the loader context. It has resourcePath, query, etc. See: https://webpack.js.org/api/loaders/#the-loader-context

In synchronous mode, the procedure function has to return the content. The content is output as JavaScript code, or passed to next loader if it is chained.

For example:

// webpack.config.js
// ...
// skeleton-loader options
options: {
  procedure: function(content, sourceMap, callback, options) {

    // Do something with content.
    console.log('Size: ' + content.length);
    content = (content + '').replace(/foo/g, 'bar'); // content might be not string.

    // Check the resource file by using context.
    if (this.resourcePath === '/abc/resource.js') {

      // Change current option.
      options.toCode = true;
    }

    // Return the content to output.
    return content;
  }
}

If the procedure function returns nothing (or returns undefined or null), the loader works in asynchronous mode. To return a SourceMap, it must be asynchronous mode.
In asynchronous mode, the procedure function has to call the callback function when it finished.

The callback function accepts the following arguments:

  • error
    An error object, when your procedure failed.
  • content
    The content that is output as JavaScript code, or passed to next loader if it is chained.
  • sourceMap
    An optional value SourceMap as JavaScript object that is output, or passed to next loader if it is chained.

For example:

// webpack.config.js
// ...
// skeleton-loader options
options: {
  procedure: function(content, sourceMap, callback) {
    // Do something asynchronously.
    require('fs').readFile('data.txt', function(error, data) {
      if (error) {
        // Failed
        callback(error);
      } else {
        // Done
        callback(null, data + content);
      }
    });
  }
}

options.resourceOptions

The options argument has resourceOptions property if a query string is specified with the resource file, and it is an object that is parsed query string.
This is useful for specifying additional parameters when importing the resource files. For example, you can specify the behavior with resource files.

var
  all = require('file.html'),
  noHead = require('file.html?removeHead=yes'),;
// webpack.config.js
// ...
// skeleton-loader options
options: {
  procedure: function(content, sourceMap, callback, options) {
    if (options.resourceOptions && options.resourceOptions.removeHead) {
      content = content.replace(/<head[^]*?<\/head>/, ''); // Remove <head>
    }
    return content;
  }
}

The query string is parsed in the same way as loader-utils.

toCode

Type: boolean
Default: false

When the content is not JavaScript code (e.g. HTML, CSS, JSON, etc.), a loader that is specified as a final loader has to convert the content to JavaScript code and output it to allow another code to import the content.
If true is specified for toCode option, the content is converted to JavaScript code.
If the loader is specified as not a final loader, this option is ignored (i.e. the content is not converted, and it is passed to next loader).

For example:

// webpack.config.js
module.exports = {
  // ...
  module: {
    rules: [
      // HTML code is converted to JavaScript string.
      // It works same as raw-loader.
      {test: /\.html$/, loader: 'skeleton-loader?toCode=true'},

      // JSON data is converted to JavaScript object.
      // It works same as json-loader.
      {
        test: /\.json$/,
        loader: 'skeleton-loader',
        options: {
          procedure: function(content) { return JSON.parse(content); },
          toCode: true
        }
      }
    ]
  }
};
// app.js
var html = require('file.html');
element.innerHTML = html;

var obj = require('file.json');
console.log(obj.array1[3]);

cacheable

Type: boolean
Default: true

Make the result cacheable.
A cacheable loader must have a deterministic result, when inputs and dependencies haven't changed. This means the loader shouldn't have other dependencies than specified with context.addDependency.
Note that the default value is true.