amdextract

Uses AST to extract AMD modules, their parts and an optimized output without unused dependencies while keeping the original format.

example

source.js

define('module1', ['p1', 'p2'], function (a, b) {
  /**
   * b is not used in this scope.
   */
  return (function(b) {
    return b;
  })(a);
});

define('module2', ['p1', 'p2', 'p3', 'p4'], function (a, b, c) {
    return b;
});

example.js

var fs = require('fs');
var amdextract = require('amdextract');

var content = fs.readFileSync('source.js');
var result = amdextract.parse(content);

console.log(result.results.length + ' modules detected.');

result.results.forEach(function (r) {
    console.log('Unused paths: ' + r.unusedPaths.join(', '));
});

console.log('\nOptimized output:');
console.log(result.optimizedContent);

output

2 modules detected.
Unused paths: p2
Unused paths: p1, p3, p4

Optimized output:

define('module1', ['p1'], function (a) {
  /**
   * b is not used in this scope.
   */
  return (function(b) {
    return b;
  })(a);
});

define('module2', ['p2'], function (b) {
    return b;
});

methods

parse(content[, options])

content

Type: string JavaScript source for parsing.

options

excepts

Type: Array Default value: []

An array of strings or RegExps that represent dependency names that should not take into account.

exceptsPaths

Type: Array Default value: []

An array of strings or RegExps that represent dependency paths that should not take into account.

NOTE: exceptsPaths can also be declared before each module definition as a comment of strings of module paths separated by commas. This only applies on the underlying module definition.

source.js

/* exceptsPaths: p3 */
define(['p1', 'p2', 'p3'], function (a, b, c) {
    return b;
});

optimized-source.js

/* exceptsPaths: p3 */
define(['p2', 'p3'], function (b, c) {
    return b;
});

removeUnusedDependencies

Type: Boolean Default value: false

Removes unused dependencies from content and returns optimized content as optimizedContent property of result.

returns

Returns an object with the following properties:

results

Type: Array

An array of hash objects witch have this properties for each AMD module detected in content:

• moduleId
• paths
• dependencies
• unusedPaths
• unusedDependencies

optimizedContent

Type: String

Optimized content (original content without unused dependencies). This property is available when the value of option removeUnusedDependencies is true.

Release History

• 2015-12-27   v2.1.14   Remove "for in" loop in traverse.
• 2015-12-26   v2.1.13   Remove unused function "extendRange".
• 2015-12-26   v2.1.12   Fixed indentation of samples in README.md.
• 2015-12-26   v2.1.11   Fixed a bug in "optimizedContent" property.
• 2015-07-25   v2.1.10   Update travis CI.
• 2015-07-25   v2.1.9   Update dependencies.
• 2015-04-15   v2.1.8   Update README.
• 2014-08-22   v2.1.6   Fix a bug when there are comments between paths and dependencies.
• 2014-08-16   v2.1.0   Entirely uses AST.
• 2014-07-21   v2.0.3   Fix an issue related to comment detection.
• 2014-07-21   v2.0.2   Fix an issue related to RegExp literals.
• 2014-07-20   v2.0.0   Uses AST to find out unused dependencies.
• 2014-04-19   v1.1.0   Keeps the original separators of paths and dependencies.
• 2014-04-09   v1.0.4   Fix a bug when dependencies length is grater than paths length.
• 2014-03-29   v1.0.3   Fix a bug when specifying exceptsPaths in options only.
• 2014-03-17   v1.0.0   First release.