1  /*!
 2   * encodeurl
 3   * Copyright(c) 2016 Douglas Christopher Wilson
 4   * MIT Licensed
 5   */
 6  
 7  'use strict'
 8  
 9  /**
10   * Module exports.
11   * @public
12   */
13  
14  module.exports = encodeUrl
15  
16  /**
17   * RegExp to match non-URL code points, *after* encoding (i.e. not including "%")
18   * and including invalid escape sequences.
19   * @private
20   */
21  
22  var ENCODE_CHARS_REGEXP = /(?:[^\x21\x23-\x3B\x3D\x3F-\x5F\x61-\x7A\x7C\x7E]|%(?:[^0-9A-Fa-f]|[0-9A-Fa-f][^0-9A-Fa-f]|$))+/g
23  
24  /**
25   * RegExp to match unmatched surrogate pair.
26   * @private
27   */
28  
29  var UNMATCHED_SURROGATE_PAIR_REGEXP = /(^|[^\uD800-\uDBFF])[\uDC00-\uDFFF]|[\uD800-\uDBFF]([^\uDC00-\uDFFF]|$)/g
30  
31  /**
32   * String to replace unmatched surrogate pair with.
33   * @private
34   */
35  
36  var UNMATCHED_SURROGATE_PAIR_REPLACE = '$1\uFFFD$2'
37  
38  /**
39   * Encode a URL to a percent-encoded form, excluding already-encoded sequences.
40   *
41   * This function will take an already-encoded URL and encode all the non-URL
42   * code points. This function will not encode the "%" character unless it is
43   * not part of a valid sequence (`%20` will be left as-is, but `%foo` will
44   * be encoded as `%25foo`).
45   *
46   * This encode is meant to be "safe" and does not throw errors. It will try as
47   * hard as it can to properly encode the given URL, including replacing any raw,
48   * unpaired surrogate pairs with the Unicode replacement character prior to
49   * encoding.
50   *
51   * @param {string} url
52   * @return {string}
53   * @public
54   */
55  
56  function encodeUrl (url) {
57    return String(url)
58      .replace(UNMATCHED_SURROGATE_PAIR_REGEXP, UNMATCHED_SURROGATE_PAIR_REPLACE)
59      .replace(ENCODE_CHARS_REGEXP, encodeURI)
60  }