1 | /*! https://mths.be/punycode v1.4.1 by @mathias */
|
2 |
|
3 | /** Highest positive signed 32-bit float value */
|
4 | var maxInt = 2147483647; // aka. 0x7FFFFFFF or 2^31-1
|
5 |
|
6 | /** Bootstring parameters */
|
7 | var base = 36;
|
8 | var tMin = 1;
|
9 | var tMax = 26;
|
10 | var skew = 38;
|
11 | var damp = 700;
|
12 | var initialBias = 72;
|
13 | var initialN = 128; // 0x80
|
14 | var delimiter = '-'; // '\x2D'
|
15 |
|
16 | /** Regular expressions */
|
17 | var regexPunycode = /^xn--/;
|
18 | var regexNonASCII = /[^\x20-\x7E]/; // unprintable ASCII chars + non-ASCII chars
|
19 | var regexSeparators = /[\x2E\u3002\uFF0E\uFF61]/g; // RFC 3490 separators
|
20 |
|
21 | /** Error messages */
|
22 | var errors = {
|
23 | 'overflow': 'Overflow: input needs wider integers to process',
|
24 | 'not-basic': 'Illegal input >= 0x80 (not a basic code point)',
|
25 | 'invalid-input': 'Invalid input'
|
26 | };
|
27 |
|
28 | /** Convenience shortcuts */
|
29 | var baseMinusTMin = base - tMin;
|
30 | var floor = Math.floor;
|
31 | var stringFromCharCode = String.fromCharCode;
|
32 |
|
33 | /*--------------------------------------------------------------------------*/
|
34 |
|
35 | /**
|
36 | * A generic error utility function.
|
37 | * @private
|
38 | * @param {String} type The error type.
|
39 | * @returns {Error} Throws a `RangeError` with the applicable error message.
|
40 | */
|
41 | function error(type) {
|
42 | throw new RangeError(errors[type]);
|
43 | }
|
44 |
|
45 | /**
|
46 | * A generic `Array#map` utility function.
|
47 | * @private
|
48 | * @param {Array} array The array to iterate over.
|
49 | * @param {Function} callback The function that gets called for every array
|
50 | * item.
|
51 | * @returns {Array} A new array of values returned by the callback function.
|
52 | */
|
53 | function map(array, fn) {
|
54 | var length = array.length;
|
55 | var result = [];
|
56 | while (length--) {
|
57 | result[length] = fn(array[length]);
|
58 | }
|
59 | return result;
|
60 | }
|
61 |
|
62 | /**
|
63 | * A simple `Array#map`-like wrapper to work with domain name strings or email
|
64 | * addresses.
|
65 | * @private
|
66 | * @param {String} domain The domain name or email address.
|
67 | * @param {Function} callback The function that gets called for every
|
68 | * character.
|
69 | * @returns {Array} A new string of characters returned by the callback
|
70 | * function.
|
71 | */
|
72 | function mapDomain(string, fn) {
|
73 | var parts = string.split('@');
|
74 | var result = '';
|
75 | if (parts.length > 1) {
|
76 | // In email addresses, only the domain name should be punycoded. Leave
|
77 | // the local part (i.e. everything up to `@`) intact.
|
78 | result = parts[0] + '@';
|
79 | string = parts[1];
|
80 | }
|
81 | // Avoid `split(regex)` for IE8 compatibility. See #17.
|
82 | string = string.replace(regexSeparators, '\x2E');
|
83 | var labels = string.split('.');
|
84 | var encoded = map(labels, fn).join('.');
|
85 | return result + encoded;
|
86 | }
|
87 |
|
88 | /**
|
89 | * Creates an array containing the numeric code points of each Unicode
|
90 | * character in the string. While JavaScript uses UCS-2 internally,
|
91 | * this function will convert a pair of surrogate halves (each of which
|
92 | * UCS-2 exposes as separate characters) into a single code point,
|
93 | * matching UTF-16.
|
94 | * @see `punycode.ucs2.encode`
|
95 | * @see <https://mathiasbynens.be/notes/javascript-encoding>
|
96 | * @memberOf punycode.ucs2
|
97 | * @name decode
|
98 | * @param {String} string The Unicode input string (UCS-2).
|
99 | * @returns {Array} The new array of code points.
|
100 | */
|
101 | function ucs2decode(string) {
|
102 | var output = [],
|
103 | counter = 0,
|
104 | length = string.length,
|
105 | value,
|
106 | extra;
|
107 | while (counter < length) {
|
108 | value = string.charCodeAt(counter++);
|
109 | if (value >= 0xD800 && value <= 0xDBFF && counter < length) {
|
110 | // high surrogate, and there is a next character
|
111 | extra = string.charCodeAt(counter++);
|
112 | if ((extra & 0xFC00) == 0xDC00) { // low surrogate
|
113 | output.push(((value & 0x3FF) << 10) + (extra & 0x3FF) + 0x10000);
|
114 | } else {
|
115 | // unmatched surrogate; only append this code unit, in case the next
|
116 | // code unit is the high surrogate of a surrogate pair
|
117 | output.push(value);
|
118 | counter--;
|
119 | }
|
120 | } else {
|
121 | output.push(value);
|
122 | }
|
123 | }
|
124 | return output;
|
125 | }
|
126 |
|
127 | /**
|
128 | * Creates a string based on an array of numeric code points.
|
129 | * @see `punycode.ucs2.decode`
|
130 | * @memberOf punycode.ucs2
|
131 | * @name encode
|
132 | * @param {Array} codePoints The array of numeric code points.
|
133 | * @returns {String} The new Unicode string (UCS-2).
|
134 | */
|
135 | function ucs2encode(array) {
|
136 | return map(array, function(value) {
|
137 | var output = '';
|
138 | if (value > 0xFFFF) {
|
139 | value -= 0x10000;
|
140 | output += stringFromCharCode(value >>> 10 & 0x3FF | 0xD800);
|
141 | value = 0xDC00 | value & 0x3FF;
|
142 | }
|
143 | output += stringFromCharCode(value);
|
144 | return output;
|
145 | }).join('');
|
146 | }
|
147 |
|
148 | /**
|
149 | * Converts a basic code point into a digit/integer.
|
150 | * @see `digitToBasic()`
|
151 | * @private
|
152 | * @param {Number} codePoint The basic numeric code point value.
|
153 | * @returns {Number} The numeric value of a basic code point (for use in
|
154 | * representing integers) in the range `0` to `base - 1`, or `base` if
|
155 | * the code point does not represent a value.
|
156 | */
|
157 | function basicToDigit(codePoint) {
|
158 | if (codePoint - 48 < 10) {
|
159 | return codePoint - 22;
|
160 | }
|
161 | if (codePoint - 65 < 26) {
|
162 | return codePoint - 65;
|
163 | }
|
164 | if (codePoint - 97 < 26) {
|
165 | return codePoint - 97;
|
166 | }
|
167 | return base;
|
168 | }
|
169 |
|
170 | /**
|
171 | * Converts a digit/integer into a basic code point.
|
172 | * @see `basicToDigit()`
|
173 | * @private
|
174 | * @param {Number} digit The numeric value of a basic code point.
|
175 | * @returns {Number} The basic code point whose value (when used for
|
176 | * representing integers) is `digit`, which needs to be in the range
|
177 | * `0` to `base - 1`. If `flag` is non-zero, the uppercase form is
|
178 | * used; else, the lowercase form is used. The behavior is undefined
|
179 | * if `flag` is non-zero and `digit` has no uppercase form.
|
180 | */
|
181 | function digitToBasic(digit, flag) {
|
182 | // 0..25 map to ASCII a..z or A..Z
|
183 | // 26..35 map to ASCII 0..9
|
184 | return digit + 22 + 75 * (digit < 26) - ((flag != 0) << 5);
|
185 | }
|
186 |
|
187 | /**
|
188 | * Bias adaptation function as per section 3.4 of RFC 3492.
|
189 | * https://tools.ietf.org/html/rfc3492#section-3.4
|
190 | * @private
|
191 | */
|
192 | function adapt(delta, numPoints, firstTime) {
|
193 | var k = 0;
|
194 | delta = firstTime ? floor(delta / damp) : delta >> 1;
|
195 | delta += floor(delta / numPoints);
|
196 | for ( /* no initialization */ ; delta > baseMinusTMin * tMax >> 1; k += base) {
|
197 | delta = floor(delta / baseMinusTMin);
|
198 | }
|
199 | return floor(k + (baseMinusTMin + 1) * delta / (delta + skew));
|
200 | }
|
201 |
|
202 | /**
|
203 | * Converts a Punycode string of ASCII-only symbols to a string of Unicode
|
204 | * symbols.
|
205 | * @memberOf punycode
|
206 | * @param {String} input The Punycode string of ASCII-only symbols.
|
207 | * @returns {String} The resulting string of Unicode symbols.
|
208 | */
|
209 | export function decode(input) {
|
210 | // Don't use UCS-2
|
211 | var output = [],
|
212 | inputLength = input.length,
|
213 | out,
|
214 | i = 0,
|
215 | n = initialN,
|
216 | bias = initialBias,
|
217 | basic,
|
218 | j,
|
219 | index,
|
220 | oldi,
|
221 | w,
|
222 | k,
|
223 | digit,
|
224 | t,
|
225 | /** Cached calculation results */
|
226 | baseMinusT;
|
227 |
|
228 | // Handle the basic code points: let `basic` be the number of input code
|
229 | // points before the last delimiter, or `0` if there is none, then copy
|
230 | // the first basic code points to the output.
|
231 |
|
232 | basic = input.lastIndexOf(delimiter);
|
233 | if (basic < 0) {
|
234 | basic = 0;
|
235 | }
|
236 |
|
237 | for (j = 0; j < basic; ++j) {
|
238 | // if it's not a basic code point
|
239 | if (input.charCodeAt(j) >= 0x80) {
|
240 | error('not-basic');
|
241 | }
|
242 | output.push(input.charCodeAt(j));
|
243 | }
|
244 |
|
245 | // Main decoding loop: start just after the last delimiter if any basic code
|
246 | // points were copied; start at the beginning otherwise.
|
247 |
|
248 | for (index = basic > 0 ? basic + 1 : 0; index < inputLength; /* no final expression */ ) {
|
249 |
|
250 | // `index` is the index of the next character to be consumed.
|
251 | // Decode a generalized variable-length integer into `delta`,
|
252 | // which gets added to `i`. The overflow checking is easier
|
253 | // if we increase `i` as we go, then subtract off its starting
|
254 | // value at the end to obtain `delta`.
|
255 | for (oldi = i, w = 1, k = base; /* no condition */ ; k += base) {
|
256 |
|
257 | if (index >= inputLength) {
|
258 | error('invalid-input');
|
259 | }
|
260 |
|
261 | digit = basicToDigit(input.charCodeAt(index++));
|
262 |
|
263 | if (digit >= base || digit > floor((maxInt - i) / w)) {
|
264 | error('overflow');
|
265 | }
|
266 |
|
267 | i += digit * w;
|
268 | t = k <= bias ? tMin : (k >= bias + tMax ? tMax : k - bias);
|
269 |
|
270 | if (digit < t) {
|
271 | break;
|
272 | }
|
273 |
|
274 | baseMinusT = base - t;
|
275 | if (w > floor(maxInt / baseMinusT)) {
|
276 | error('overflow');
|
277 | }
|
278 |
|
279 | w *= baseMinusT;
|
280 |
|
281 | }
|
282 |
|
283 | out = output.length + 1;
|
284 | bias = adapt(i - oldi, out, oldi == 0);
|
285 |
|
286 | // `i` was supposed to wrap around from `out` to `0`,
|
287 | // incrementing `n` each time, so we'll fix that now:
|
288 | if (floor(i / out) > maxInt - n) {
|
289 | error('overflow');
|
290 | }
|
291 |
|
292 | n += floor(i / out);
|
293 | i %= out;
|
294 |
|
295 | // Insert `n` at position `i` of the output
|
296 | output.splice(i++, 0, n);
|
297 |
|
298 | }
|
299 |
|
300 | return ucs2encode(output);
|
301 | }
|
302 |
|
303 | /**
|
304 | * Converts a string of Unicode symbols (e.g. a domain name label) to a
|
305 | * Punycode string of ASCII-only symbols.
|
306 | * @memberOf punycode
|
307 | * @param {String} input The string of Unicode symbols.
|
308 | * @returns {String} The resulting Punycode string of ASCII-only symbols.
|
309 | */
|
310 | export function encode(input) {
|
311 | var n,
|
312 | delta,
|
313 | handledCPCount,
|
314 | basicLength,
|
315 | bias,
|
316 | j,
|
317 | m,
|
318 | q,
|
319 | k,
|
320 | t,
|
321 | currentValue,
|
322 | output = [],
|
323 | /** `inputLength` will hold the number of code points in `input`. */
|
324 | inputLength,
|
325 | /** Cached calculation results */
|
326 | handledCPCountPlusOne,
|
327 | baseMinusT,
|
328 | qMinusT;
|
329 |
|
330 | // Convert the input in UCS-2 to Unicode
|
331 | input = ucs2decode(input);
|
332 |
|
333 | // Cache the length
|
334 | inputLength = input.length;
|
335 |
|
336 | // Initialize the state
|
337 | n = initialN;
|
338 | delta = 0;
|
339 | bias = initialBias;
|
340 |
|
341 | // Handle the basic code points
|
342 | for (j = 0; j < inputLength; ++j) {
|
343 | currentValue = input[j];
|
344 | if (currentValue < 0x80) {
|
345 | output.push(stringFromCharCode(currentValue));
|
346 | }
|
347 | }
|
348 |
|
349 | handledCPCount = basicLength = output.length;
|
350 |
|
351 | // `handledCPCount` is the number of code points that have been handled;
|
352 | // `basicLength` is the number of basic code points.
|
353 |
|
354 | // Finish the basic string - if it is not empty - with a delimiter
|
355 | if (basicLength) {
|
356 | output.push(delimiter);
|
357 | }
|
358 |
|
359 | // Main encoding loop:
|
360 | while (handledCPCount < inputLength) {
|
361 |
|
362 | // All non-basic code points < n have been handled already. Find the next
|
363 | // larger one:
|
364 | for (m = maxInt, j = 0; j < inputLength; ++j) {
|
365 | currentValue = input[j];
|
366 | if (currentValue >= n && currentValue < m) {
|
367 | m = currentValue;
|
368 | }
|
369 | }
|
370 |
|
371 | // Increase `delta` enough to advance the decoder's <n,i> state to <m,0>,
|
372 | // but guard against overflow
|
373 | handledCPCountPlusOne = handledCPCount + 1;
|
374 | if (m - n > floor((maxInt - delta) / handledCPCountPlusOne)) {
|
375 | error('overflow');
|
376 | }
|
377 |
|
378 | delta += (m - n) * handledCPCountPlusOne;
|
379 | n = m;
|
380 |
|
381 | for (j = 0; j < inputLength; ++j) {
|
382 | currentValue = input[j];
|
383 |
|
384 | if (currentValue < n && ++delta > maxInt) {
|
385 | error('overflow');
|
386 | }
|
387 |
|
388 | if (currentValue == n) {
|
389 | // Represent delta as a generalized variable-length integer
|
390 | for (q = delta, k = base; /* no condition */ ; k += base) {
|
391 | t = k <= bias ? tMin : (k >= bias + tMax ? tMax : k - bias);
|
392 | if (q < t) {
|
393 | break;
|
394 | }
|
395 | qMinusT = q - t;
|
396 | baseMinusT = base - t;
|
397 | output.push(
|
398 | stringFromCharCode(digitToBasic(t + qMinusT % baseMinusT, 0))
|
399 | );
|
400 | q = floor(qMinusT / baseMinusT);
|
401 | }
|
402 |
|
403 | output.push(stringFromCharCode(digitToBasic(q, 0)));
|
404 | bias = adapt(delta, handledCPCountPlusOne, handledCPCount == basicLength);
|
405 | delta = 0;
|
406 | ++handledCPCount;
|
407 | }
|
408 | }
|
409 |
|
410 | ++delta;
|
411 | ++n;
|
412 |
|
413 | }
|
414 | return output.join('');
|
415 | }
|
416 |
|
417 | /**
|
418 | * Converts a Punycode string representing a domain name or an email address
|
419 | * to Unicode. Only the Punycoded parts of the input will be converted, i.e.
|
420 | * it doesn't matter if you call it on a string that has already been
|
421 | * converted to Unicode.
|
422 | * @memberOf punycode
|
423 | * @param {String} input The Punycoded domain name or email address to
|
424 | * convert to Unicode.
|
425 | * @returns {String} The Unicode representation of the given Punycode
|
426 | * string.
|
427 | */
|
428 | export function toUnicode(input) {
|
429 | return mapDomain(input, function(string) {
|
430 | return regexPunycode.test(string) ?
|
431 | decode(string.slice(4).toLowerCase()) :
|
432 | string;
|
433 | });
|
434 | }
|
435 |
|
436 | /**
|
437 | * Converts a Unicode string representing a domain name or an email address to
|
438 | * Punycode. Only the non-ASCII parts of the domain name will be converted,
|
439 | * i.e. it doesn't matter if you call it with a domain that's already in
|
440 | * ASCII.
|
441 | * @memberOf punycode
|
442 | * @param {String} input The domain name or email address to convert, as a
|
443 | * Unicode string.
|
444 | * @returns {String} The Punycode representation of the given domain name or
|
445 | * email address.
|
446 | */
|
447 | export function toASCII(input) {
|
448 | return mapDomain(input, function(string) {
|
449 | return regexNonASCII.test(string) ?
|
450 | 'xn--' + encode(string) :
|
451 | string;
|
452 | });
|
453 | }
|
454 | export var version = '1.4.1';
|
455 | /**
|
456 | * An object of methods to convert from JavaScript's internal character
|
457 | * representation (UCS-2) to Unicode code points, and back.
|
458 | * @see <https://mathiasbynens.be/notes/javascript-encoding>
|
459 | * @memberOf punycode
|
460 | * @type Object
|
461 | */
|
462 |
|
463 | export var ucs2 = {
|
464 | decode: ucs2decode,
|
465 | encode: ucs2encode
|
466 | };
|
467 | export default {
|
468 | version: version,
|
469 | ucs2: ucs2,
|
470 | toASCII: toASCII,
|
471 | toUnicode: toUnicode,
|
472 | encode: encode,
|
473 | decode: decode
|
474 | }
|