1
// Copyright 2010 The Closure Library Authors. All Rights Reserved.
2
//
3
// Licensed under the Apache License, Version 2.0 (the "License");
4
// you may not use this file except in compliance with the License.
5
// You may obtain a copy of the License at
6
//
7
//      http://www.apache.org/licenses/LICENSE-2.0
8
//
9
// Unless required by applicable law or agreed to in writing, software
10
// distributed under the License is distributed on an "AS-IS" BASIS,
11
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12
// See the License for the specific language governing permissions and
13
// limitations under the License.
14

15
/**
16
 * @fileoverview Functions for encoding strings according to MIME
17
 * standards, especially RFC 1522.
18
 */
19
goog.provide('goog.i18n.mime');
20
goog.provide('goog.i18n.mime.encode');
21

22
goog.require('goog.array');
23

24

25
/**
26
 * Regular expression for matching those characters that are outside the
27
 * range that can be used in the quoted-printable encoding of RFC 1522:
28
 * anything outside the 7-bit ASCII encoding, plus ?, =, _ or space.
29
 * @type {RegExp}
30
 * @private
31
 */
32
goog.i18n.mime.NONASCII_ = /[^!-<>@-^`-~]/g;
33

34

35
/**
36
 * Like goog.i18n.NONASCII_ but also omits double-quotes.
37
 * @type {RegExp}
38
 * @private
39
 */
40
goog.i18n.mime.NONASCII_NOQUOTE_ = /[^!#-<>@-^`-~]/g;
41

42

43
/**
44
 * Encodes a string for inclusion in a MIME header. The string is encoded
45
 * in UTF-8 according to RFC 1522, using quoted-printable form.
46
 * @param {string} str The string to encode.
47
 * @param {boolean=} opt_noquote Whether double-quote characters should also
48
 *     be escaped (should be true if the result will be placed inside a
49
 *     quoted string for a parameter value in a MIME header).
50
 * @return {string} The encoded string.
51
 */
52
goog.i18n.mime.encode = function(str, opt_noquote) {
53
  var nonascii =
54
      opt_noquote ? goog.i18n.mime.NONASCII_NOQUOTE_ : goog.i18n.mime.NONASCII_;
55

56
  if (str.search(nonascii) >= 0) {
57
    str = '=?UTF-8?Q?' +
58
        str.replace(
59
            nonascii,
60
            /**
61
             * @param {string} c The matched char.
62
             * @return {string} The quoted-printable form of utf-8 encoding.
63
             */
64
            function(c) {
65
              var i = c.charCodeAt(0);
66
              if (i == 32) {
67
                // Special case for space, which can be encoded as _ not =20
68
                return '_';
69
              }
70
              var a = goog.array.concat('', goog.i18n.mime.getHexCharArray(c));
71
              return a.join('=');
72
            }) +
73
        '?=';
74
  }
75
  return str;
76
};
77

78

79
/**
80
 * Get an array of UTF-8 hex codes for a given character.
81
 * @param {string} c The matched character.
82
 * @return {!Array<string>} A hex array representing the character.
83
 */
84
goog.i18n.mime.getHexCharArray = function(c) {
85
  var i = c.charCodeAt(0);
86
  var a = [];
87
  // First convert the UCS-2 character into its UTF-8 bytes
88
  if (i < 128) {
89
    a.push(i);
90
  } else if (i <= 0x7ff) {
91
    a.push(0xc0 + ((i >> 6) & 0x3f), 0x80 + (i & 0x3f));
92
  } else if (i <= 0xffff) {
93
    a.push(
94
        0xe0 + ((i >> 12) & 0x3f), 0x80 + ((i >> 6) & 0x3f), 0x80 + (i & 0x3f));
95
  } else {
96
    // (This is defensive programming, since ecmascript isn't supposed
97
    // to handle code points that take more than 16 bits.)
98
    a.push(
99
        0xf0 + ((i >> 18) & 0x3f), 0x80 + ((i >> 12) & 0x3f),
100
        0x80 + ((i >> 6) & 0x3f), 0x80 + (i & 0x3f));
101
  }
102
  // Now convert those bytes into hex strings (don't do anything with
103
  // a[0] as that's got the empty string that lets us use join())
104
  for (i = a.length - 1; i >= 0; --i) {
105
    a[i] = a[i].toString(16);
106
  }
107
  return a;
108
};
109

110