1 | 'use strict';
|
---|
2 | // based on https://github.com/bestiejs/punycode.js/blob/master/punycode.js
|
---|
3 | var uncurryThis = require('../internals/function-uncurry-this');
|
---|
4 |
|
---|
5 | var maxInt = 2147483647; // aka. 0x7FFFFFFF or 2^31-1
|
---|
6 | var base = 36;
|
---|
7 | var tMin = 1;
|
---|
8 | var tMax = 26;
|
---|
9 | var skew = 38;
|
---|
10 | var damp = 700;
|
---|
11 | var initialBias = 72;
|
---|
12 | var initialN = 128; // 0x80
|
---|
13 | var delimiter = '-'; // '\x2D'
|
---|
14 | var regexNonASCII = /[^\0-\u007E]/; // non-ASCII chars
|
---|
15 | var regexSeparators = /[.\u3002\uFF0E\uFF61]/g; // RFC 3490 separators
|
---|
16 | var OVERFLOW_ERROR = 'Overflow: input needs wider integers to process';
|
---|
17 | var baseMinusTMin = base - tMin;
|
---|
18 |
|
---|
19 | var $RangeError = RangeError;
|
---|
20 | var exec = uncurryThis(regexSeparators.exec);
|
---|
21 | var floor = Math.floor;
|
---|
22 | var fromCharCode = String.fromCharCode;
|
---|
23 | var charCodeAt = uncurryThis(''.charCodeAt);
|
---|
24 | var join = uncurryThis([].join);
|
---|
25 | var push = uncurryThis([].push);
|
---|
26 | var replace = uncurryThis(''.replace);
|
---|
27 | var split = uncurryThis(''.split);
|
---|
28 | var toLowerCase = uncurryThis(''.toLowerCase);
|
---|
29 |
|
---|
30 | /**
|
---|
31 | * Creates an array containing the numeric code points of each Unicode
|
---|
32 | * character in the string. While JavaScript uses UCS-2 internally,
|
---|
33 | * this function will convert a pair of surrogate halves (each of which
|
---|
34 | * UCS-2 exposes as separate characters) into a single code point,
|
---|
35 | * matching UTF-16.
|
---|
36 | */
|
---|
37 | var ucs2decode = function (string) {
|
---|
38 | var output = [];
|
---|
39 | var counter = 0;
|
---|
40 | var length = string.length;
|
---|
41 | while (counter < length) {
|
---|
42 | var value = charCodeAt(string, counter++);
|
---|
43 | if (value >= 0xD800 && value <= 0xDBFF && counter < length) {
|
---|
44 | // It's a high surrogate, and there is a next character.
|
---|
45 | var extra = charCodeAt(string, counter++);
|
---|
46 | if ((extra & 0xFC00) === 0xDC00) { // Low surrogate.
|
---|
47 | push(output, ((value & 0x3FF) << 10) + (extra & 0x3FF) + 0x10000);
|
---|
48 | } else {
|
---|
49 | // It's an unmatched surrogate; only append this code unit, in case the
|
---|
50 | // next code unit is the high surrogate of a surrogate pair.
|
---|
51 | push(output, value);
|
---|
52 | counter--;
|
---|
53 | }
|
---|
54 | } else {
|
---|
55 | push(output, value);
|
---|
56 | }
|
---|
57 | }
|
---|
58 | return output;
|
---|
59 | };
|
---|
60 |
|
---|
61 | /**
|
---|
62 | * Converts a digit/integer into a basic code point.
|
---|
63 | */
|
---|
64 | var digitToBasic = function (digit) {
|
---|
65 | // 0..25 map to ASCII a..z or A..Z
|
---|
66 | // 26..35 map to ASCII 0..9
|
---|
67 | return digit + 22 + 75 * (digit < 26);
|
---|
68 | };
|
---|
69 |
|
---|
70 | /**
|
---|
71 | * Bias adaptation function as per section 3.4 of RFC 3492.
|
---|
72 | * https://tools.ietf.org/html/rfc3492#section-3.4
|
---|
73 | */
|
---|
74 | var adapt = function (delta, numPoints, firstTime) {
|
---|
75 | var k = 0;
|
---|
76 | delta = firstTime ? floor(delta / damp) : delta >> 1;
|
---|
77 | delta += floor(delta / numPoints);
|
---|
78 | while (delta > baseMinusTMin * tMax >> 1) {
|
---|
79 | delta = floor(delta / baseMinusTMin);
|
---|
80 | k += base;
|
---|
81 | }
|
---|
82 | return floor(k + (baseMinusTMin + 1) * delta / (delta + skew));
|
---|
83 | };
|
---|
84 |
|
---|
85 | /**
|
---|
86 | * Converts a string of Unicode symbols (e.g. a domain name label) to a
|
---|
87 | * Punycode string of ASCII-only symbols.
|
---|
88 | */
|
---|
89 | var encode = function (input) {
|
---|
90 | var output = [];
|
---|
91 |
|
---|
92 | // Convert the input in UCS-2 to an array of Unicode code points.
|
---|
93 | input = ucs2decode(input);
|
---|
94 |
|
---|
95 | // Cache the length.
|
---|
96 | var inputLength = input.length;
|
---|
97 |
|
---|
98 | // Initialize the state.
|
---|
99 | var n = initialN;
|
---|
100 | var delta = 0;
|
---|
101 | var bias = initialBias;
|
---|
102 | var i, currentValue;
|
---|
103 |
|
---|
104 | // Handle the basic code points.
|
---|
105 | for (i = 0; i < input.length; i++) {
|
---|
106 | currentValue = input[i];
|
---|
107 | if (currentValue < 0x80) {
|
---|
108 | push(output, fromCharCode(currentValue));
|
---|
109 | }
|
---|
110 | }
|
---|
111 |
|
---|
112 | var basicLength = output.length; // number of basic code points.
|
---|
113 | var handledCPCount = basicLength; // number of code points that have been handled;
|
---|
114 |
|
---|
115 | // Finish the basic string with a delimiter unless it's empty.
|
---|
116 | if (basicLength) {
|
---|
117 | push(output, delimiter);
|
---|
118 | }
|
---|
119 |
|
---|
120 | // Main encoding loop:
|
---|
121 | while (handledCPCount < inputLength) {
|
---|
122 | // All non-basic code points < n have been handled already. Find the next larger one:
|
---|
123 | var m = maxInt;
|
---|
124 | for (i = 0; i < input.length; i++) {
|
---|
125 | currentValue = input[i];
|
---|
126 | if (currentValue >= n && currentValue < m) {
|
---|
127 | m = currentValue;
|
---|
128 | }
|
---|
129 | }
|
---|
130 |
|
---|
131 | // Increase `delta` enough to advance the decoder's <n,i> state to <m,0>, but guard against overflow.
|
---|
132 | var handledCPCountPlusOne = handledCPCount + 1;
|
---|
133 | if (m - n > floor((maxInt - delta) / handledCPCountPlusOne)) {
|
---|
134 | throw new $RangeError(OVERFLOW_ERROR);
|
---|
135 | }
|
---|
136 |
|
---|
137 | delta += (m - n) * handledCPCountPlusOne;
|
---|
138 | n = m;
|
---|
139 |
|
---|
140 | for (i = 0; i < input.length; i++) {
|
---|
141 | currentValue = input[i];
|
---|
142 | if (currentValue < n && ++delta > maxInt) {
|
---|
143 | throw new $RangeError(OVERFLOW_ERROR);
|
---|
144 | }
|
---|
145 | if (currentValue === n) {
|
---|
146 | // Represent delta as a generalized variable-length integer.
|
---|
147 | var q = delta;
|
---|
148 | var k = base;
|
---|
149 | while (true) {
|
---|
150 | var t = k <= bias ? tMin : k >= bias + tMax ? tMax : k - bias;
|
---|
151 | if (q < t) break;
|
---|
152 | var qMinusT = q - t;
|
---|
153 | var baseMinusT = base - t;
|
---|
154 | push(output, fromCharCode(digitToBasic(t + qMinusT % baseMinusT)));
|
---|
155 | q = floor(qMinusT / baseMinusT);
|
---|
156 | k += base;
|
---|
157 | }
|
---|
158 |
|
---|
159 | push(output, fromCharCode(digitToBasic(q)));
|
---|
160 | bias = adapt(delta, handledCPCountPlusOne, handledCPCount === basicLength);
|
---|
161 | delta = 0;
|
---|
162 | handledCPCount++;
|
---|
163 | }
|
---|
164 | }
|
---|
165 |
|
---|
166 | delta++;
|
---|
167 | n++;
|
---|
168 | }
|
---|
169 | return join(output, '');
|
---|
170 | };
|
---|
171 |
|
---|
172 | module.exports = function (input) {
|
---|
173 | var encoded = [];
|
---|
174 | var labels = split(replace(toLowerCase(input), regexSeparators, '\u002E'), '.');
|
---|
175 | var i, label;
|
---|
176 | for (i = 0; i < labels.length; i++) {
|
---|
177 | label = labels[i];
|
---|
178 | push(encoded, exec(regexNonASCII, label) ? 'xn--' + encode(label) : label);
|
---|
179 | }
|
---|
180 | return join(encoded, '.');
|
---|
181 | };
|
---|