[6a3a178] | 1 | /*! https://mths.be/regenerate v1.4.2 by @mathias | MIT license */
|
---|
| 2 | ;(function(root) {
|
---|
| 3 |
|
---|
| 4 | // Detect free variables `exports`.
|
---|
| 5 | var freeExports = typeof exports == 'object' && exports;
|
---|
| 6 |
|
---|
| 7 | // Detect free variable `module`.
|
---|
| 8 | var freeModule = typeof module == 'object' && module &&
|
---|
| 9 | module.exports == freeExports && module;
|
---|
| 10 |
|
---|
| 11 | // Detect free variable `global`, from Node.js/io.js or Browserified code,
|
---|
| 12 | // and use it as `root`.
|
---|
| 13 | var freeGlobal = typeof global == 'object' && global;
|
---|
| 14 | if (freeGlobal.global === freeGlobal || freeGlobal.window === freeGlobal) {
|
---|
| 15 | root = freeGlobal;
|
---|
| 16 | }
|
---|
| 17 |
|
---|
| 18 | /*--------------------------------------------------------------------------*/
|
---|
| 19 |
|
---|
| 20 | var ERRORS = {
|
---|
| 21 | 'rangeOrder': 'A range\u2019s `stop` value must be greater than or equal ' +
|
---|
| 22 | 'to the `start` value.',
|
---|
| 23 | 'codePointRange': 'Invalid code point value. Code points range from ' +
|
---|
| 24 | 'U+000000 to U+10FFFF.'
|
---|
| 25 | };
|
---|
| 26 |
|
---|
| 27 | // https://mathiasbynens.be/notes/javascript-encoding#surrogate-pairs
|
---|
| 28 | var HIGH_SURROGATE_MIN = 0xD800;
|
---|
| 29 | var HIGH_SURROGATE_MAX = 0xDBFF;
|
---|
| 30 | var LOW_SURROGATE_MIN = 0xDC00;
|
---|
| 31 | var LOW_SURROGATE_MAX = 0xDFFF;
|
---|
| 32 |
|
---|
| 33 | // In Regenerate output, `\0` is never preceded by `\` because we sort by
|
---|
| 34 | // code point value, so let’s keep this regular expression simple.
|
---|
| 35 | var regexNull = /\\x00([^0123456789]|$)/g;
|
---|
| 36 |
|
---|
| 37 | var object = {};
|
---|
| 38 | var hasOwnProperty = object.hasOwnProperty;
|
---|
| 39 | var extend = function(destination, source) {
|
---|
| 40 | var key;
|
---|
| 41 | for (key in source) {
|
---|
| 42 | if (hasOwnProperty.call(source, key)) {
|
---|
| 43 | destination[key] = source[key];
|
---|
| 44 | }
|
---|
| 45 | }
|
---|
| 46 | return destination;
|
---|
| 47 | };
|
---|
| 48 |
|
---|
| 49 | var forEach = function(array, callback) {
|
---|
| 50 | var index = -1;
|
---|
| 51 | var length = array.length;
|
---|
| 52 | while (++index < length) {
|
---|
| 53 | callback(array[index], index);
|
---|
| 54 | }
|
---|
| 55 | };
|
---|
| 56 |
|
---|
| 57 | var toString = object.toString;
|
---|
| 58 | var isArray = function(value) {
|
---|
| 59 | return toString.call(value) == '[object Array]';
|
---|
| 60 | };
|
---|
| 61 | var isNumber = function(value) {
|
---|
| 62 | return typeof value == 'number' ||
|
---|
| 63 | toString.call(value) == '[object Number]';
|
---|
| 64 | };
|
---|
| 65 |
|
---|
| 66 | // This assumes that `number` is a positive integer that `toString()`s nicely
|
---|
| 67 | // (which is the case for all code point values).
|
---|
| 68 | var zeroes = '0000';
|
---|
| 69 | var pad = function(number, totalCharacters) {
|
---|
| 70 | var string = String(number);
|
---|
| 71 | return string.length < totalCharacters
|
---|
| 72 | ? (zeroes + string).slice(-totalCharacters)
|
---|
| 73 | : string;
|
---|
| 74 | };
|
---|
| 75 |
|
---|
| 76 | var hex = function(number) {
|
---|
| 77 | return Number(number).toString(16).toUpperCase();
|
---|
| 78 | };
|
---|
| 79 |
|
---|
| 80 | var slice = [].slice;
|
---|
| 81 |
|
---|
| 82 | /*--------------------------------------------------------------------------*/
|
---|
| 83 |
|
---|
| 84 | var dataFromCodePoints = function(codePoints) {
|
---|
| 85 | var index = -1;
|
---|
| 86 | var length = codePoints.length;
|
---|
| 87 | var max = length - 1;
|
---|
| 88 | var result = [];
|
---|
| 89 | var isStart = true;
|
---|
| 90 | var tmp;
|
---|
| 91 | var previous = 0;
|
---|
| 92 | while (++index < length) {
|
---|
| 93 | tmp = codePoints[index];
|
---|
| 94 | if (isStart) {
|
---|
| 95 | result.push(tmp);
|
---|
| 96 | previous = tmp;
|
---|
| 97 | isStart = false;
|
---|
| 98 | } else {
|
---|
| 99 | if (tmp == previous + 1) {
|
---|
| 100 | if (index != max) {
|
---|
| 101 | previous = tmp;
|
---|
| 102 | continue;
|
---|
| 103 | } else {
|
---|
| 104 | isStart = true;
|
---|
| 105 | result.push(tmp + 1);
|
---|
| 106 | }
|
---|
| 107 | } else {
|
---|
| 108 | // End the previous range and start a new one.
|
---|
| 109 | result.push(previous + 1, tmp);
|
---|
| 110 | previous = tmp;
|
---|
| 111 | }
|
---|
| 112 | }
|
---|
| 113 | }
|
---|
| 114 | if (!isStart) {
|
---|
| 115 | result.push(tmp + 1);
|
---|
| 116 | }
|
---|
| 117 | return result;
|
---|
| 118 | };
|
---|
| 119 |
|
---|
| 120 | var dataRemove = function(data, codePoint) {
|
---|
| 121 | // Iterate over the data per `(start, end)` pair.
|
---|
| 122 | var index = 0;
|
---|
| 123 | var start;
|
---|
| 124 | var end;
|
---|
| 125 | var length = data.length;
|
---|
| 126 | while (index < length) {
|
---|
| 127 | start = data[index];
|
---|
| 128 | end = data[index + 1];
|
---|
| 129 | if (codePoint >= start && codePoint < end) {
|
---|
| 130 | // Modify this pair.
|
---|
| 131 | if (codePoint == start) {
|
---|
| 132 | if (end == start + 1) {
|
---|
| 133 | // Just remove `start` and `end`.
|
---|
| 134 | data.splice(index, 2);
|
---|
| 135 | return data;
|
---|
| 136 | } else {
|
---|
| 137 | // Just replace `start` with a new value.
|
---|
| 138 | data[index] = codePoint + 1;
|
---|
| 139 | return data;
|
---|
| 140 | }
|
---|
| 141 | } else if (codePoint == end - 1) {
|
---|
| 142 | // Just replace `end` with a new value.
|
---|
| 143 | data[index + 1] = codePoint;
|
---|
| 144 | return data;
|
---|
| 145 | } else {
|
---|
| 146 | // Replace `[start, end]` with `[startA, endA, startB, endB]`.
|
---|
| 147 | data.splice(index, 2, start, codePoint, codePoint + 1, end);
|
---|
| 148 | return data;
|
---|
| 149 | }
|
---|
| 150 | }
|
---|
| 151 | index += 2;
|
---|
| 152 | }
|
---|
| 153 | return data;
|
---|
| 154 | };
|
---|
| 155 |
|
---|
| 156 | var dataRemoveRange = function(data, rangeStart, rangeEnd) {
|
---|
| 157 | if (rangeEnd < rangeStart) {
|
---|
| 158 | throw Error(ERRORS.rangeOrder);
|
---|
| 159 | }
|
---|
| 160 | // Iterate over the data per `(start, end)` pair.
|
---|
| 161 | var index = 0;
|
---|
| 162 | var start;
|
---|
| 163 | var end;
|
---|
| 164 | while (index < data.length) {
|
---|
| 165 | start = data[index];
|
---|
| 166 | end = data[index + 1] - 1; // Note: the `- 1` makes `end` inclusive.
|
---|
| 167 |
|
---|
| 168 | // Exit as soon as no more matching pairs can be found.
|
---|
| 169 | if (start > rangeEnd) {
|
---|
| 170 | return data;
|
---|
| 171 | }
|
---|
| 172 |
|
---|
| 173 | // Check if this range pair is equal to, or forms a subset of, the range
|
---|
| 174 | // to be removed.
|
---|
| 175 | // E.g. we have `[0, 11, 40, 51]` and want to remove 0-10 → `[40, 51]`.
|
---|
| 176 | // E.g. we have `[40, 51]` and want to remove 0-100 → `[]`.
|
---|
| 177 | if (rangeStart <= start && rangeEnd >= end) {
|
---|
| 178 | // Remove this pair.
|
---|
| 179 | data.splice(index, 2);
|
---|
| 180 | continue;
|
---|
| 181 | }
|
---|
| 182 |
|
---|
| 183 | // Check if both `rangeStart` and `rangeEnd` are within the bounds of
|
---|
| 184 | // this pair.
|
---|
| 185 | // E.g. we have `[0, 11]` and want to remove 4-6 → `[0, 4, 7, 11]`.
|
---|
| 186 | if (rangeStart >= start && rangeEnd < end) {
|
---|
| 187 | if (rangeStart == start) {
|
---|
| 188 | // Replace `[start, end]` with `[startB, endB]`.
|
---|
| 189 | data[index] = rangeEnd + 1;
|
---|
| 190 | data[index + 1] = end + 1;
|
---|
| 191 | return data;
|
---|
| 192 | }
|
---|
| 193 | // Replace `[start, end]` with `[startA, endA, startB, endB]`.
|
---|
| 194 | data.splice(index, 2, start, rangeStart, rangeEnd + 1, end + 1);
|
---|
| 195 | return data;
|
---|
| 196 | }
|
---|
| 197 |
|
---|
| 198 | // Check if only `rangeStart` is within the bounds of this pair.
|
---|
| 199 | // E.g. we have `[0, 11]` and want to remove 4-20 → `[0, 4]`.
|
---|
| 200 | if (rangeStart >= start && rangeStart <= end) {
|
---|
| 201 | // Replace `end` with `rangeStart`.
|
---|
| 202 | data[index + 1] = rangeStart;
|
---|
| 203 | // Note: we cannot `return` just yet, in case any following pairs still
|
---|
| 204 | // contain matching code points.
|
---|
| 205 | // E.g. we have `[0, 11, 14, 31]` and want to remove 4-20
|
---|
| 206 | // → `[0, 4, 21, 31]`.
|
---|
| 207 | }
|
---|
| 208 |
|
---|
| 209 | // Check if only `rangeEnd` is within the bounds of this pair.
|
---|
| 210 | // E.g. we have `[14, 31]` and want to remove 4-20 → `[21, 31]`.
|
---|
| 211 | else if (rangeEnd >= start && rangeEnd <= end) {
|
---|
| 212 | // Just replace `start`.
|
---|
| 213 | data[index] = rangeEnd + 1;
|
---|
| 214 | return data;
|
---|
| 215 | }
|
---|
| 216 |
|
---|
| 217 | index += 2;
|
---|
| 218 | }
|
---|
| 219 | return data;
|
---|
| 220 | };
|
---|
| 221 |
|
---|
| 222 | var dataAdd = function(data, codePoint) {
|
---|
| 223 | // Iterate over the data per `(start, end)` pair.
|
---|
| 224 | var index = 0;
|
---|
| 225 | var start;
|
---|
| 226 | var end;
|
---|
| 227 | var lastIndex = null;
|
---|
| 228 | var length = data.length;
|
---|
| 229 | if (codePoint < 0x0 || codePoint > 0x10FFFF) {
|
---|
| 230 | throw RangeError(ERRORS.codePointRange);
|
---|
| 231 | }
|
---|
| 232 | while (index < length) {
|
---|
| 233 | start = data[index];
|
---|
| 234 | end = data[index + 1];
|
---|
| 235 |
|
---|
| 236 | // Check if the code point is already in the set.
|
---|
| 237 | if (codePoint >= start && codePoint < end) {
|
---|
| 238 | return data;
|
---|
| 239 | }
|
---|
| 240 |
|
---|
| 241 | if (codePoint == start - 1) {
|
---|
| 242 | // Just replace `start` with a new value.
|
---|
| 243 | data[index] = codePoint;
|
---|
| 244 | return data;
|
---|
| 245 | }
|
---|
| 246 |
|
---|
| 247 | // At this point, if `start` is `greater` than `codePoint`, insert a new
|
---|
| 248 | // `[start, end]` pair before the current pair, or after the current pair
|
---|
| 249 | // if there is a known `lastIndex`.
|
---|
| 250 | if (start > codePoint) {
|
---|
| 251 | data.splice(
|
---|
| 252 | lastIndex != null ? lastIndex + 2 : 0,
|
---|
| 253 | 0,
|
---|
| 254 | codePoint,
|
---|
| 255 | codePoint + 1
|
---|
| 256 | );
|
---|
| 257 | return data;
|
---|
| 258 | }
|
---|
| 259 |
|
---|
| 260 | if (codePoint == end) {
|
---|
| 261 | // Check if adding this code point causes two separate ranges to become
|
---|
| 262 | // a single range, e.g. `dataAdd([0, 4, 5, 10], 4)` → `[0, 10]`.
|
---|
| 263 | if (codePoint + 1 == data[index + 2]) {
|
---|
| 264 | data.splice(index, 4, start, data[index + 3]);
|
---|
| 265 | return data;
|
---|
| 266 | }
|
---|
| 267 | // Else, just replace `end` with a new value.
|
---|
| 268 | data[index + 1] = codePoint + 1;
|
---|
| 269 | return data;
|
---|
| 270 | }
|
---|
| 271 | lastIndex = index;
|
---|
| 272 | index += 2;
|
---|
| 273 | }
|
---|
| 274 | // The loop has finished; add the new pair to the end of the data set.
|
---|
| 275 | data.push(codePoint, codePoint + 1);
|
---|
| 276 | return data;
|
---|
| 277 | };
|
---|
| 278 |
|
---|
| 279 | var dataAddData = function(dataA, dataB) {
|
---|
| 280 | // Iterate over the data per `(start, end)` pair.
|
---|
| 281 | var index = 0;
|
---|
| 282 | var start;
|
---|
| 283 | var end;
|
---|
| 284 | var data = dataA.slice();
|
---|
| 285 | var length = dataB.length;
|
---|
| 286 | while (index < length) {
|
---|
| 287 | start = dataB[index];
|
---|
| 288 | end = dataB[index + 1] - 1;
|
---|
| 289 | if (start == end) {
|
---|
| 290 | data = dataAdd(data, start);
|
---|
| 291 | } else {
|
---|
| 292 | data = dataAddRange(data, start, end);
|
---|
| 293 | }
|
---|
| 294 | index += 2;
|
---|
| 295 | }
|
---|
| 296 | return data;
|
---|
| 297 | };
|
---|
| 298 |
|
---|
| 299 | var dataRemoveData = function(dataA, dataB) {
|
---|
| 300 | // Iterate over the data per `(start, end)` pair.
|
---|
| 301 | var index = 0;
|
---|
| 302 | var start;
|
---|
| 303 | var end;
|
---|
| 304 | var data = dataA.slice();
|
---|
| 305 | var length = dataB.length;
|
---|
| 306 | while (index < length) {
|
---|
| 307 | start = dataB[index];
|
---|
| 308 | end = dataB[index + 1] - 1;
|
---|
| 309 | if (start == end) {
|
---|
| 310 | data = dataRemove(data, start);
|
---|
| 311 | } else {
|
---|
| 312 | data = dataRemoveRange(data, start, end);
|
---|
| 313 | }
|
---|
| 314 | index += 2;
|
---|
| 315 | }
|
---|
| 316 | return data;
|
---|
| 317 | };
|
---|
| 318 |
|
---|
| 319 | var dataAddRange = function(data, rangeStart, rangeEnd) {
|
---|
| 320 | if (rangeEnd < rangeStart) {
|
---|
| 321 | throw Error(ERRORS.rangeOrder);
|
---|
| 322 | }
|
---|
| 323 | if (
|
---|
| 324 | rangeStart < 0x0 || rangeStart > 0x10FFFF ||
|
---|
| 325 | rangeEnd < 0x0 || rangeEnd > 0x10FFFF
|
---|
| 326 | ) {
|
---|
| 327 | throw RangeError(ERRORS.codePointRange);
|
---|
| 328 | }
|
---|
| 329 | // Iterate over the data per `(start, end)` pair.
|
---|
| 330 | var index = 0;
|
---|
| 331 | var start;
|
---|
| 332 | var end;
|
---|
| 333 | var added = false;
|
---|
| 334 | var length = data.length;
|
---|
| 335 | while (index < length) {
|
---|
| 336 | start = data[index];
|
---|
| 337 | end = data[index + 1];
|
---|
| 338 |
|
---|
| 339 | if (added) {
|
---|
| 340 | // The range has already been added to the set; at this point, we just
|
---|
| 341 | // need to get rid of the following ranges in case they overlap.
|
---|
| 342 |
|
---|
| 343 | // Check if this range can be combined with the previous range.
|
---|
| 344 | if (start == rangeEnd + 1) {
|
---|
| 345 | data.splice(index - 1, 2);
|
---|
| 346 | return data;
|
---|
| 347 | }
|
---|
| 348 |
|
---|
| 349 | // Exit as soon as no more possibly overlapping pairs can be found.
|
---|
| 350 | if (start > rangeEnd) {
|
---|
| 351 | return data;
|
---|
| 352 | }
|
---|
| 353 |
|
---|
| 354 | // E.g. `[0, 11, 12, 16]` and we’ve added 5-15, so we now have
|
---|
| 355 | // `[0, 16, 12, 16]`. Remove the `12,16` part, as it lies within the
|
---|
| 356 | // `0,16` range that was previously added.
|
---|
| 357 | if (start >= rangeStart && start <= rangeEnd) {
|
---|
| 358 | // `start` lies within the range that was previously added.
|
---|
| 359 |
|
---|
| 360 | if (end > rangeStart && end - 1 <= rangeEnd) {
|
---|
| 361 | // `end` lies within the range that was previously added as well,
|
---|
| 362 | // so remove this pair.
|
---|
| 363 | data.splice(index, 2);
|
---|
| 364 | index -= 2;
|
---|
| 365 | // Note: we cannot `return` just yet, as there may still be other
|
---|
| 366 | // overlapping pairs.
|
---|
| 367 | } else {
|
---|
| 368 | // `start` lies within the range that was previously added, but
|
---|
| 369 | // `end` doesn’t. E.g. `[0, 11, 12, 31]` and we’ve added 5-15, so
|
---|
| 370 | // now we have `[0, 16, 12, 31]`. This must be written as `[0, 31]`.
|
---|
| 371 | // Remove the previously added `end` and the current `start`.
|
---|
| 372 | data.splice(index - 1, 2);
|
---|
| 373 | index -= 2;
|
---|
| 374 | }
|
---|
| 375 |
|
---|
| 376 | // Note: we cannot return yet.
|
---|
| 377 | }
|
---|
| 378 |
|
---|
| 379 | }
|
---|
| 380 |
|
---|
| 381 | else if (start == rangeEnd + 1 || start == rangeEnd) {
|
---|
| 382 | data[index] = rangeStart;
|
---|
| 383 | return data;
|
---|
| 384 | }
|
---|
| 385 |
|
---|
| 386 | // Check if a new pair must be inserted *before* the current one.
|
---|
| 387 | else if (start > rangeEnd) {
|
---|
| 388 | data.splice(index, 0, rangeStart, rangeEnd + 1);
|
---|
| 389 | return data;
|
---|
| 390 | }
|
---|
| 391 |
|
---|
| 392 | else if (rangeStart >= start && rangeStart < end && rangeEnd + 1 <= end) {
|
---|
| 393 | // The new range lies entirely within an existing range pair. No action
|
---|
| 394 | // needed.
|
---|
| 395 | return data;
|
---|
| 396 | }
|
---|
| 397 |
|
---|
| 398 | else if (
|
---|
| 399 | // E.g. `[0, 11]` and you add 5-15 → `[0, 16]`.
|
---|
| 400 | (rangeStart >= start && rangeStart < end) ||
|
---|
| 401 | // E.g. `[0, 3]` and you add 3-6 → `[0, 7]`.
|
---|
| 402 | end == rangeStart
|
---|
| 403 | ) {
|
---|
| 404 | // Replace `end` with the new value.
|
---|
| 405 | data[index + 1] = rangeEnd + 1;
|
---|
| 406 | // Make sure the next range pair doesn’t overlap, e.g. `[0, 11, 12, 14]`
|
---|
| 407 | // and you add 5-15 → `[0, 16]`, i.e. remove the `12,14` part.
|
---|
| 408 | added = true;
|
---|
| 409 | // Note: we cannot `return` just yet.
|
---|
| 410 | }
|
---|
| 411 |
|
---|
| 412 | else if (rangeStart <= start && rangeEnd + 1 >= end) {
|
---|
| 413 | // The new range is a superset of the old range.
|
---|
| 414 | data[index] = rangeStart;
|
---|
| 415 | data[index + 1] = rangeEnd + 1;
|
---|
| 416 | added = true;
|
---|
| 417 | }
|
---|
| 418 |
|
---|
| 419 | index += 2;
|
---|
| 420 | }
|
---|
| 421 | // The loop has finished without doing anything; add the new pair to the end
|
---|
| 422 | // of the data set.
|
---|
| 423 | if (!added) {
|
---|
| 424 | data.push(rangeStart, rangeEnd + 1);
|
---|
| 425 | }
|
---|
| 426 | return data;
|
---|
| 427 | };
|
---|
| 428 |
|
---|
| 429 | var dataContains = function(data, codePoint) {
|
---|
| 430 | var index = 0;
|
---|
| 431 | var length = data.length;
|
---|
| 432 | // Exit early if `codePoint` is not within `data`’s overall range.
|
---|
| 433 | var start = data[index];
|
---|
| 434 | var end = data[length - 1];
|
---|
| 435 | if (length >= 2) {
|
---|
| 436 | if (codePoint < start || codePoint > end) {
|
---|
| 437 | return false;
|
---|
| 438 | }
|
---|
| 439 | }
|
---|
| 440 | // Iterate over the data per `(start, end)` pair.
|
---|
| 441 | while (index < length) {
|
---|
| 442 | start = data[index];
|
---|
| 443 | end = data[index + 1];
|
---|
| 444 | if (codePoint >= start && codePoint < end) {
|
---|
| 445 | return true;
|
---|
| 446 | }
|
---|
| 447 | index += 2;
|
---|
| 448 | }
|
---|
| 449 | return false;
|
---|
| 450 | };
|
---|
| 451 |
|
---|
| 452 | var dataIntersection = function(data, codePoints) {
|
---|
| 453 | var index = 0;
|
---|
| 454 | var length = codePoints.length;
|
---|
| 455 | var codePoint;
|
---|
| 456 | var result = [];
|
---|
| 457 | while (index < length) {
|
---|
| 458 | codePoint = codePoints[index];
|
---|
| 459 | if (dataContains(data, codePoint)) {
|
---|
| 460 | result.push(codePoint);
|
---|
| 461 | }
|
---|
| 462 | ++index;
|
---|
| 463 | }
|
---|
| 464 | return dataFromCodePoints(result);
|
---|
| 465 | };
|
---|
| 466 |
|
---|
| 467 | var dataIsEmpty = function(data) {
|
---|
| 468 | return !data.length;
|
---|
| 469 | };
|
---|
| 470 |
|
---|
| 471 | var dataIsSingleton = function(data) {
|
---|
| 472 | // Check if the set only represents a single code point.
|
---|
| 473 | return data.length == 2 && data[0] + 1 == data[1];
|
---|
| 474 | };
|
---|
| 475 |
|
---|
| 476 | var dataToArray = function(data) {
|
---|
| 477 | // Iterate over the data per `(start, end)` pair.
|
---|
| 478 | var index = 0;
|
---|
| 479 | var start;
|
---|
| 480 | var end;
|
---|
| 481 | var result = [];
|
---|
| 482 | var length = data.length;
|
---|
| 483 | while (index < length) {
|
---|
| 484 | start = data[index];
|
---|
| 485 | end = data[index + 1];
|
---|
| 486 | while (start < end) {
|
---|
| 487 | result.push(start);
|
---|
| 488 | ++start;
|
---|
| 489 | }
|
---|
| 490 | index += 2;
|
---|
| 491 | }
|
---|
| 492 | return result;
|
---|
| 493 | };
|
---|
| 494 |
|
---|
| 495 | /*--------------------------------------------------------------------------*/
|
---|
| 496 |
|
---|
| 497 | // https://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
|
---|
| 498 | var floor = Math.floor;
|
---|
| 499 | var highSurrogate = function(codePoint) {
|
---|
| 500 | return parseInt(
|
---|
| 501 | floor((codePoint - 0x10000) / 0x400) + HIGH_SURROGATE_MIN,
|
---|
| 502 | 10
|
---|
| 503 | );
|
---|
| 504 | };
|
---|
| 505 |
|
---|
| 506 | var lowSurrogate = function(codePoint) {
|
---|
| 507 | return parseInt(
|
---|
| 508 | (codePoint - 0x10000) % 0x400 + LOW_SURROGATE_MIN,
|
---|
| 509 | 10
|
---|
| 510 | );
|
---|
| 511 | };
|
---|
| 512 |
|
---|
| 513 | var stringFromCharCode = String.fromCharCode;
|
---|
| 514 | var codePointToString = function(codePoint) {
|
---|
| 515 | var string;
|
---|
| 516 | // https://mathiasbynens.be/notes/javascript-escapes#single
|
---|
| 517 | // Note: the `\b` escape sequence for U+0008 BACKSPACE in strings has a
|
---|
| 518 | // different meaning in regular expressions (word boundary), so it cannot
|
---|
| 519 | // be used here.
|
---|
| 520 | if (codePoint == 0x09) {
|
---|
| 521 | string = '\\t';
|
---|
| 522 | }
|
---|
| 523 | // Note: IE < 9 treats `'\v'` as `'v'`, so avoid using it.
|
---|
| 524 | // else if (codePoint == 0x0B) {
|
---|
| 525 | // string = '\\v';
|
---|
| 526 | // }
|
---|
| 527 | else if (codePoint == 0x0A) {
|
---|
| 528 | string = '\\n';
|
---|
| 529 | }
|
---|
| 530 | else if (codePoint == 0x0C) {
|
---|
| 531 | string = '\\f';
|
---|
| 532 | }
|
---|
| 533 | else if (codePoint == 0x0D) {
|
---|
| 534 | string = '\\r';
|
---|
| 535 | }
|
---|
| 536 | else if (codePoint == 0x2D) {
|
---|
| 537 | // https://mathiasbynens.be/notes/javascript-escapes#hexadecimal
|
---|
| 538 | // Note: `-` (U+002D HYPHEN-MINUS) is escaped in this way rather
|
---|
| 539 | // than by backslash-escaping, in case the output is used outside
|
---|
| 540 | // of a character class in a `u` RegExp. /\-/u throws, but
|
---|
| 541 | // /\x2D/u is fine.
|
---|
| 542 | string = '\\x2D';
|
---|
| 543 | }
|
---|
| 544 | else if (codePoint == 0x5C) {
|
---|
| 545 | string = '\\\\';
|
---|
| 546 | }
|
---|
| 547 | else if (
|
---|
| 548 | codePoint == 0x24 ||
|
---|
| 549 | (codePoint >= 0x28 && codePoint <= 0x2B) ||
|
---|
| 550 | codePoint == 0x2E || codePoint == 0x2F ||
|
---|
| 551 | codePoint == 0x3F ||
|
---|
| 552 | (codePoint >= 0x5B && codePoint <= 0x5E) ||
|
---|
| 553 | (codePoint >= 0x7B && codePoint <= 0x7D)
|
---|
| 554 | ) {
|
---|
| 555 | // The code point maps to an unsafe printable ASCII character;
|
---|
| 556 | // backslash-escape it. Here’s the list of those symbols:
|
---|
| 557 | //
|
---|
| 558 | // $()*+./?[\]^{|}
|
---|
| 559 | //
|
---|
| 560 | // This matches SyntaxCharacters as well as `/` (U+002F SOLIDUS).
|
---|
| 561 | // https://tc39.github.io/ecma262/#prod-SyntaxCharacter
|
---|
| 562 | string = '\\' + stringFromCharCode(codePoint);
|
---|
| 563 | }
|
---|
| 564 | else if (codePoint >= 0x20 && codePoint <= 0x7E) {
|
---|
| 565 | // The code point maps to one of these printable ASCII symbols
|
---|
| 566 | // (including the space character):
|
---|
| 567 | //
|
---|
| 568 | // !"#%&',/0123456789:;<=>@ABCDEFGHIJKLMNO
|
---|
| 569 | // PQRSTUVWXYZ_`abcdefghijklmnopqrstuvwxyz~
|
---|
| 570 | //
|
---|
| 571 | // These can safely be used directly.
|
---|
| 572 | string = stringFromCharCode(codePoint);
|
---|
| 573 | }
|
---|
| 574 | else if (codePoint <= 0xFF) {
|
---|
| 575 | string = '\\x' + pad(hex(codePoint), 2);
|
---|
| 576 | }
|
---|
| 577 | else { // `codePoint <= 0xFFFF` holds true.
|
---|
| 578 | // https://mathiasbynens.be/notes/javascript-escapes#unicode
|
---|
| 579 | string = '\\u' + pad(hex(codePoint), 4);
|
---|
| 580 | }
|
---|
| 581 |
|
---|
| 582 | // There’s no need to account for astral symbols / surrogate pairs here,
|
---|
| 583 | // since `codePointToString` is private and only used for BMP code points.
|
---|
| 584 | // But if that’s what you need, just add an `else` block with this code:
|
---|
| 585 | //
|
---|
| 586 | // string = '\\u' + pad(hex(highSurrogate(codePoint)), 4)
|
---|
| 587 | // + '\\u' + pad(hex(lowSurrogate(codePoint)), 4);
|
---|
| 588 |
|
---|
| 589 | return string;
|
---|
| 590 | };
|
---|
| 591 |
|
---|
| 592 | var codePointToStringUnicode = function(codePoint) {
|
---|
| 593 | if (codePoint <= 0xFFFF) {
|
---|
| 594 | return codePointToString(codePoint);
|
---|
| 595 | }
|
---|
| 596 | return '\\u{' + codePoint.toString(16).toUpperCase() + '}';
|
---|
| 597 | };
|
---|
| 598 |
|
---|
| 599 | var symbolToCodePoint = function(symbol) {
|
---|
| 600 | var length = symbol.length;
|
---|
| 601 | var first = symbol.charCodeAt(0);
|
---|
| 602 | var second;
|
---|
| 603 | if (
|
---|
| 604 | first >= HIGH_SURROGATE_MIN && first <= HIGH_SURROGATE_MAX &&
|
---|
| 605 | length > 1 // There is a next code unit.
|
---|
| 606 | ) {
|
---|
| 607 | // `first` is a high surrogate, and there is a next character. Assume
|
---|
| 608 | // it’s a low surrogate (else it’s invalid usage of Regenerate anyway).
|
---|
| 609 | second = symbol.charCodeAt(1);
|
---|
| 610 | // https://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
|
---|
| 611 | return (first - HIGH_SURROGATE_MIN) * 0x400 +
|
---|
| 612 | second - LOW_SURROGATE_MIN + 0x10000;
|
---|
| 613 | }
|
---|
| 614 | return first;
|
---|
| 615 | };
|
---|
| 616 |
|
---|
| 617 | var createBMPCharacterClasses = function(data) {
|
---|
| 618 | // Iterate over the data per `(start, end)` pair.
|
---|
| 619 | var result = '';
|
---|
| 620 | var index = 0;
|
---|
| 621 | var start;
|
---|
| 622 | var end;
|
---|
| 623 | var length = data.length;
|
---|
| 624 | if (dataIsSingleton(data)) {
|
---|
| 625 | return codePointToString(data[0]);
|
---|
| 626 | }
|
---|
| 627 | while (index < length) {
|
---|
| 628 | start = data[index];
|
---|
| 629 | end = data[index + 1] - 1; // Note: the `- 1` makes `end` inclusive.
|
---|
| 630 | if (start == end) {
|
---|
| 631 | result += codePointToString(start);
|
---|
| 632 | } else if (start + 1 == end) {
|
---|
| 633 | result += codePointToString(start) + codePointToString(end);
|
---|
| 634 | } else {
|
---|
| 635 | result += codePointToString(start) + '-' + codePointToString(end);
|
---|
| 636 | }
|
---|
| 637 | index += 2;
|
---|
| 638 | }
|
---|
| 639 | return '[' + result + ']';
|
---|
| 640 | };
|
---|
| 641 |
|
---|
| 642 | var createUnicodeCharacterClasses = function(data) {
|
---|
| 643 | // Iterate over the data per `(start, end)` pair.
|
---|
| 644 | var result = '';
|
---|
| 645 | var index = 0;
|
---|
| 646 | var start;
|
---|
| 647 | var end;
|
---|
| 648 | var length = data.length;
|
---|
| 649 | if (dataIsSingleton(data)) {
|
---|
| 650 | return codePointToStringUnicode(data[0]);
|
---|
| 651 | }
|
---|
| 652 | while (index < length) {
|
---|
| 653 | start = data[index];
|
---|
| 654 | end = data[index + 1] - 1; // Note: the `- 1` makes `end` inclusive.
|
---|
| 655 | if (start == end) {
|
---|
| 656 | result += codePointToStringUnicode(start);
|
---|
| 657 | } else if (start + 1 == end) {
|
---|
| 658 | result += codePointToStringUnicode(start) + codePointToStringUnicode(end);
|
---|
| 659 | } else {
|
---|
| 660 | result += codePointToStringUnicode(start) + '-' + codePointToStringUnicode(end);
|
---|
| 661 | }
|
---|
| 662 | index += 2;
|
---|
| 663 | }
|
---|
| 664 | return '[' + result + ']';
|
---|
| 665 | };
|
---|
| 666 |
|
---|
| 667 | var splitAtBMP = function(data) {
|
---|
| 668 | // Iterate over the data per `(start, end)` pair.
|
---|
| 669 | var loneHighSurrogates = [];
|
---|
| 670 | var loneLowSurrogates = [];
|
---|
| 671 | var bmp = [];
|
---|
| 672 | var astral = [];
|
---|
| 673 | var index = 0;
|
---|
| 674 | var start;
|
---|
| 675 | var end;
|
---|
| 676 | var length = data.length;
|
---|
| 677 | while (index < length) {
|
---|
| 678 | start = data[index];
|
---|
| 679 | end = data[index + 1] - 1; // Note: the `- 1` makes `end` inclusive.
|
---|
| 680 |
|
---|
| 681 | if (start < HIGH_SURROGATE_MIN) {
|
---|
| 682 |
|
---|
| 683 | // The range starts and ends before the high surrogate range.
|
---|
| 684 | // E.g. (0, 0x10).
|
---|
| 685 | if (end < HIGH_SURROGATE_MIN) {
|
---|
| 686 | bmp.push(start, end + 1);
|
---|
| 687 | }
|
---|
| 688 |
|
---|
| 689 | // The range starts before the high surrogate range and ends within it.
|
---|
| 690 | // E.g. (0, 0xD855).
|
---|
| 691 | if (end >= HIGH_SURROGATE_MIN && end <= HIGH_SURROGATE_MAX) {
|
---|
| 692 | bmp.push(start, HIGH_SURROGATE_MIN);
|
---|
| 693 | loneHighSurrogates.push(HIGH_SURROGATE_MIN, end + 1);
|
---|
| 694 | }
|
---|
| 695 |
|
---|
| 696 | // The range starts before the high surrogate range and ends in the low
|
---|
| 697 | // surrogate range. E.g. (0, 0xDCFF).
|
---|
| 698 | if (end >= LOW_SURROGATE_MIN && end <= LOW_SURROGATE_MAX) {
|
---|
| 699 | bmp.push(start, HIGH_SURROGATE_MIN);
|
---|
| 700 | loneHighSurrogates.push(HIGH_SURROGATE_MIN, HIGH_SURROGATE_MAX + 1);
|
---|
| 701 | loneLowSurrogates.push(LOW_SURROGATE_MIN, end + 1);
|
---|
| 702 | }
|
---|
| 703 |
|
---|
| 704 | // The range starts before the high surrogate range and ends after the
|
---|
| 705 | // low surrogate range. E.g. (0, 0x10FFFF).
|
---|
| 706 | if (end > LOW_SURROGATE_MAX) {
|
---|
| 707 | bmp.push(start, HIGH_SURROGATE_MIN);
|
---|
| 708 | loneHighSurrogates.push(HIGH_SURROGATE_MIN, HIGH_SURROGATE_MAX + 1);
|
---|
| 709 | loneLowSurrogates.push(LOW_SURROGATE_MIN, LOW_SURROGATE_MAX + 1);
|
---|
| 710 | if (end <= 0xFFFF) {
|
---|
| 711 | bmp.push(LOW_SURROGATE_MAX + 1, end + 1);
|
---|
| 712 | } else {
|
---|
| 713 | bmp.push(LOW_SURROGATE_MAX + 1, 0xFFFF + 1);
|
---|
| 714 | astral.push(0xFFFF + 1, end + 1);
|
---|
| 715 | }
|
---|
| 716 | }
|
---|
| 717 |
|
---|
| 718 | } else if (start >= HIGH_SURROGATE_MIN && start <= HIGH_SURROGATE_MAX) {
|
---|
| 719 |
|
---|
| 720 | // The range starts and ends in the high surrogate range.
|
---|
| 721 | // E.g. (0xD855, 0xD866).
|
---|
| 722 | if (end >= HIGH_SURROGATE_MIN && end <= HIGH_SURROGATE_MAX) {
|
---|
| 723 | loneHighSurrogates.push(start, end + 1);
|
---|
| 724 | }
|
---|
| 725 |
|
---|
| 726 | // The range starts in the high surrogate range and ends in the low
|
---|
| 727 | // surrogate range. E.g. (0xD855, 0xDCFF).
|
---|
| 728 | if (end >= LOW_SURROGATE_MIN && end <= LOW_SURROGATE_MAX) {
|
---|
| 729 | loneHighSurrogates.push(start, HIGH_SURROGATE_MAX + 1);
|
---|
| 730 | loneLowSurrogates.push(LOW_SURROGATE_MIN, end + 1);
|
---|
| 731 | }
|
---|
| 732 |
|
---|
| 733 | // The range starts in the high surrogate range and ends after the low
|
---|
| 734 | // surrogate range. E.g. (0xD855, 0x10FFFF).
|
---|
| 735 | if (end > LOW_SURROGATE_MAX) {
|
---|
| 736 | loneHighSurrogates.push(start, HIGH_SURROGATE_MAX + 1);
|
---|
| 737 | loneLowSurrogates.push(LOW_SURROGATE_MIN, LOW_SURROGATE_MAX + 1);
|
---|
| 738 | if (end <= 0xFFFF) {
|
---|
| 739 | bmp.push(LOW_SURROGATE_MAX + 1, end + 1);
|
---|
| 740 | } else {
|
---|
| 741 | bmp.push(LOW_SURROGATE_MAX + 1, 0xFFFF + 1);
|
---|
| 742 | astral.push(0xFFFF + 1, end + 1);
|
---|
| 743 | }
|
---|
| 744 | }
|
---|
| 745 |
|
---|
| 746 | } else if (start >= LOW_SURROGATE_MIN && start <= LOW_SURROGATE_MAX) {
|
---|
| 747 |
|
---|
| 748 | // The range starts and ends in the low surrogate range.
|
---|
| 749 | // E.g. (0xDCFF, 0xDDFF).
|
---|
| 750 | if (end >= LOW_SURROGATE_MIN && end <= LOW_SURROGATE_MAX) {
|
---|
| 751 | loneLowSurrogates.push(start, end + 1);
|
---|
| 752 | }
|
---|
| 753 |
|
---|
| 754 | // The range starts in the low surrogate range and ends after the low
|
---|
| 755 | // surrogate range. E.g. (0xDCFF, 0x10FFFF).
|
---|
| 756 | if (end > LOW_SURROGATE_MAX) {
|
---|
| 757 | loneLowSurrogates.push(start, LOW_SURROGATE_MAX + 1);
|
---|
| 758 | if (end <= 0xFFFF) {
|
---|
| 759 | bmp.push(LOW_SURROGATE_MAX + 1, end + 1);
|
---|
| 760 | } else {
|
---|
| 761 | bmp.push(LOW_SURROGATE_MAX + 1, 0xFFFF + 1);
|
---|
| 762 | astral.push(0xFFFF + 1, end + 1);
|
---|
| 763 | }
|
---|
| 764 | }
|
---|
| 765 |
|
---|
| 766 | } else if (start > LOW_SURROGATE_MAX && start <= 0xFFFF) {
|
---|
| 767 |
|
---|
| 768 | // The range starts and ends after the low surrogate range.
|
---|
| 769 | // E.g. (0xFFAA, 0x10FFFF).
|
---|
| 770 | if (end <= 0xFFFF) {
|
---|
| 771 | bmp.push(start, end + 1);
|
---|
| 772 | } else {
|
---|
| 773 | bmp.push(start, 0xFFFF + 1);
|
---|
| 774 | astral.push(0xFFFF + 1, end + 1);
|
---|
| 775 | }
|
---|
| 776 |
|
---|
| 777 | } else {
|
---|
| 778 |
|
---|
| 779 | // The range starts and ends in the astral range.
|
---|
| 780 | astral.push(start, end + 1);
|
---|
| 781 |
|
---|
| 782 | }
|
---|
| 783 |
|
---|
| 784 | index += 2;
|
---|
| 785 | }
|
---|
| 786 | return {
|
---|
| 787 | 'loneHighSurrogates': loneHighSurrogates,
|
---|
| 788 | 'loneLowSurrogates': loneLowSurrogates,
|
---|
| 789 | 'bmp': bmp,
|
---|
| 790 | 'astral': astral
|
---|
| 791 | };
|
---|
| 792 | };
|
---|
| 793 |
|
---|
| 794 | var optimizeSurrogateMappings = function(surrogateMappings) {
|
---|
| 795 | var result = [];
|
---|
| 796 | var tmpLow = [];
|
---|
| 797 | var addLow = false;
|
---|
| 798 | var mapping;
|
---|
| 799 | var nextMapping;
|
---|
| 800 | var highSurrogates;
|
---|
| 801 | var lowSurrogates;
|
---|
| 802 | var nextHighSurrogates;
|
---|
| 803 | var nextLowSurrogates;
|
---|
| 804 | var index = -1;
|
---|
| 805 | var length = surrogateMappings.length;
|
---|
| 806 | while (++index < length) {
|
---|
| 807 | mapping = surrogateMappings[index];
|
---|
| 808 | nextMapping = surrogateMappings[index + 1];
|
---|
| 809 | if (!nextMapping) {
|
---|
| 810 | result.push(mapping);
|
---|
| 811 | continue;
|
---|
| 812 | }
|
---|
| 813 | highSurrogates = mapping[0];
|
---|
| 814 | lowSurrogates = mapping[1];
|
---|
| 815 | nextHighSurrogates = nextMapping[0];
|
---|
| 816 | nextLowSurrogates = nextMapping[1];
|
---|
| 817 |
|
---|
| 818 | // Check for identical high surrogate ranges.
|
---|
| 819 | tmpLow = lowSurrogates;
|
---|
| 820 | while (
|
---|
| 821 | nextHighSurrogates &&
|
---|
| 822 | highSurrogates[0] == nextHighSurrogates[0] &&
|
---|
| 823 | highSurrogates[1] == nextHighSurrogates[1]
|
---|
| 824 | ) {
|
---|
| 825 | // Merge with the next item.
|
---|
| 826 | if (dataIsSingleton(nextLowSurrogates)) {
|
---|
| 827 | tmpLow = dataAdd(tmpLow, nextLowSurrogates[0]);
|
---|
| 828 | } else {
|
---|
| 829 | tmpLow = dataAddRange(
|
---|
| 830 | tmpLow,
|
---|
| 831 | nextLowSurrogates[0],
|
---|
| 832 | nextLowSurrogates[1] - 1
|
---|
| 833 | );
|
---|
| 834 | }
|
---|
| 835 | ++index;
|
---|
| 836 | mapping = surrogateMappings[index];
|
---|
| 837 | highSurrogates = mapping[0];
|
---|
| 838 | lowSurrogates = mapping[1];
|
---|
| 839 | nextMapping = surrogateMappings[index + 1];
|
---|
| 840 | nextHighSurrogates = nextMapping && nextMapping[0];
|
---|
| 841 | nextLowSurrogates = nextMapping && nextMapping[1];
|
---|
| 842 | addLow = true;
|
---|
| 843 | }
|
---|
| 844 | result.push([
|
---|
| 845 | highSurrogates,
|
---|
| 846 | addLow ? tmpLow : lowSurrogates
|
---|
| 847 | ]);
|
---|
| 848 | addLow = false;
|
---|
| 849 | }
|
---|
| 850 | return optimizeByLowSurrogates(result);
|
---|
| 851 | };
|
---|
| 852 |
|
---|
| 853 | var optimizeByLowSurrogates = function(surrogateMappings) {
|
---|
| 854 | if (surrogateMappings.length == 1) {
|
---|
| 855 | return surrogateMappings;
|
---|
| 856 | }
|
---|
| 857 | var index = -1;
|
---|
| 858 | var innerIndex = -1;
|
---|
| 859 | while (++index < surrogateMappings.length) {
|
---|
| 860 | var mapping = surrogateMappings[index];
|
---|
| 861 | var lowSurrogates = mapping[1];
|
---|
| 862 | var lowSurrogateStart = lowSurrogates[0];
|
---|
| 863 | var lowSurrogateEnd = lowSurrogates[1];
|
---|
| 864 | innerIndex = index; // Note: the loop starts at the next index.
|
---|
| 865 | while (++innerIndex < surrogateMappings.length) {
|
---|
| 866 | var otherMapping = surrogateMappings[innerIndex];
|
---|
| 867 | var otherLowSurrogates = otherMapping[1];
|
---|
| 868 | var otherLowSurrogateStart = otherLowSurrogates[0];
|
---|
| 869 | var otherLowSurrogateEnd = otherLowSurrogates[1];
|
---|
| 870 | if (
|
---|
| 871 | lowSurrogateStart == otherLowSurrogateStart &&
|
---|
| 872 | lowSurrogateEnd == otherLowSurrogateEnd &&
|
---|
| 873 | otherLowSurrogates.length === 2
|
---|
| 874 | ) {
|
---|
| 875 | // Add the code points in the other item to this one.
|
---|
| 876 | if (dataIsSingleton(otherMapping[0])) {
|
---|
| 877 | mapping[0] = dataAdd(mapping[0], otherMapping[0][0]);
|
---|
| 878 | } else {
|
---|
| 879 | mapping[0] = dataAddRange(
|
---|
| 880 | mapping[0],
|
---|
| 881 | otherMapping[0][0],
|
---|
| 882 | otherMapping[0][1] - 1
|
---|
| 883 | );
|
---|
| 884 | }
|
---|
| 885 | // Remove the other, now redundant, item.
|
---|
| 886 | surrogateMappings.splice(innerIndex, 1);
|
---|
| 887 | --innerIndex;
|
---|
| 888 | }
|
---|
| 889 | }
|
---|
| 890 | }
|
---|
| 891 | return surrogateMappings;
|
---|
| 892 | };
|
---|
| 893 |
|
---|
| 894 | var surrogateSet = function(data) {
|
---|
| 895 | // Exit early if `data` is an empty set.
|
---|
| 896 | if (!data.length) {
|
---|
| 897 | return [];
|
---|
| 898 | }
|
---|
| 899 |
|
---|
| 900 | // Iterate over the data per `(start, end)` pair.
|
---|
| 901 | var index = 0;
|
---|
| 902 | var start;
|
---|
| 903 | var end;
|
---|
| 904 | var startHigh;
|
---|
| 905 | var startLow;
|
---|
| 906 | var endHigh;
|
---|
| 907 | var endLow;
|
---|
| 908 | var surrogateMappings = [];
|
---|
| 909 | var length = data.length;
|
---|
| 910 | while (index < length) {
|
---|
| 911 | start = data[index];
|
---|
| 912 | end = data[index + 1] - 1;
|
---|
| 913 |
|
---|
| 914 | startHigh = highSurrogate(start);
|
---|
| 915 | startLow = lowSurrogate(start);
|
---|
| 916 | endHigh = highSurrogate(end);
|
---|
| 917 | endLow = lowSurrogate(end);
|
---|
| 918 |
|
---|
| 919 | var startsWithLowestLowSurrogate = startLow == LOW_SURROGATE_MIN;
|
---|
| 920 | var endsWithHighestLowSurrogate = endLow == LOW_SURROGATE_MAX;
|
---|
| 921 | var complete = false;
|
---|
| 922 |
|
---|
| 923 | // Append the previous high-surrogate-to-low-surrogate mappings.
|
---|
| 924 | // Step 1: `(startHigh, startLow)` to `(startHigh, LOW_SURROGATE_MAX)`.
|
---|
| 925 | if (
|
---|
| 926 | startHigh == endHigh ||
|
---|
| 927 | startsWithLowestLowSurrogate && endsWithHighestLowSurrogate
|
---|
| 928 | ) {
|
---|
| 929 | surrogateMappings.push([
|
---|
| 930 | [startHigh, endHigh + 1],
|
---|
| 931 | [startLow, endLow + 1]
|
---|
| 932 | ]);
|
---|
| 933 | complete = true;
|
---|
| 934 | } else {
|
---|
| 935 | surrogateMappings.push([
|
---|
| 936 | [startHigh, startHigh + 1],
|
---|
| 937 | [startLow, LOW_SURROGATE_MAX + 1]
|
---|
| 938 | ]);
|
---|
| 939 | }
|
---|
| 940 |
|
---|
| 941 | // Step 2: `(startHigh + 1, LOW_SURROGATE_MIN)` to
|
---|
| 942 | // `(endHigh - 1, LOW_SURROGATE_MAX)`.
|
---|
| 943 | if (!complete && startHigh + 1 < endHigh) {
|
---|
| 944 | if (endsWithHighestLowSurrogate) {
|
---|
| 945 | // Combine step 2 and step 3.
|
---|
| 946 | surrogateMappings.push([
|
---|
| 947 | [startHigh + 1, endHigh + 1],
|
---|
| 948 | [LOW_SURROGATE_MIN, endLow + 1]
|
---|
| 949 | ]);
|
---|
| 950 | complete = true;
|
---|
| 951 | } else {
|
---|
| 952 | surrogateMappings.push([
|
---|
| 953 | [startHigh + 1, endHigh],
|
---|
| 954 | [LOW_SURROGATE_MIN, LOW_SURROGATE_MAX + 1]
|
---|
| 955 | ]);
|
---|
| 956 | }
|
---|
| 957 | }
|
---|
| 958 |
|
---|
| 959 | // Step 3. `(endHigh, LOW_SURROGATE_MIN)` to `(endHigh, endLow)`.
|
---|
| 960 | if (!complete) {
|
---|
| 961 | surrogateMappings.push([
|
---|
| 962 | [endHigh, endHigh + 1],
|
---|
| 963 | [LOW_SURROGATE_MIN, endLow + 1]
|
---|
| 964 | ]);
|
---|
| 965 | }
|
---|
| 966 |
|
---|
| 967 | index += 2;
|
---|
| 968 | }
|
---|
| 969 |
|
---|
| 970 | // The format of `surrogateMappings` is as follows:
|
---|
| 971 | //
|
---|
| 972 | // [ surrogateMapping1, surrogateMapping2 ]
|
---|
| 973 | //
|
---|
| 974 | // i.e.:
|
---|
| 975 | //
|
---|
| 976 | // [
|
---|
| 977 | // [ highSurrogates1, lowSurrogates1 ],
|
---|
| 978 | // [ highSurrogates2, lowSurrogates2 ]
|
---|
| 979 | // ]
|
---|
| 980 | return optimizeSurrogateMappings(surrogateMappings);
|
---|
| 981 | };
|
---|
| 982 |
|
---|
| 983 | var createSurrogateCharacterClasses = function(surrogateMappings) {
|
---|
| 984 | var result = [];
|
---|
| 985 | forEach(surrogateMappings, function(surrogateMapping) {
|
---|
| 986 | var highSurrogates = surrogateMapping[0];
|
---|
| 987 | var lowSurrogates = surrogateMapping[1];
|
---|
| 988 | result.push(
|
---|
| 989 | createBMPCharacterClasses(highSurrogates) +
|
---|
| 990 | createBMPCharacterClasses(lowSurrogates)
|
---|
| 991 | );
|
---|
| 992 | });
|
---|
| 993 | return result.join('|');
|
---|
| 994 | };
|
---|
| 995 |
|
---|
| 996 | var createCharacterClassesFromData = function(data, bmpOnly, hasUnicodeFlag) {
|
---|
| 997 | if (hasUnicodeFlag) {
|
---|
| 998 | return createUnicodeCharacterClasses(data);
|
---|
| 999 | }
|
---|
| 1000 | var result = [];
|
---|
| 1001 |
|
---|
| 1002 | var parts = splitAtBMP(data);
|
---|
| 1003 | var loneHighSurrogates = parts.loneHighSurrogates;
|
---|
| 1004 | var loneLowSurrogates = parts.loneLowSurrogates;
|
---|
| 1005 | var bmp = parts.bmp;
|
---|
| 1006 | var astral = parts.astral;
|
---|
| 1007 | var hasLoneHighSurrogates = !dataIsEmpty(loneHighSurrogates);
|
---|
| 1008 | var hasLoneLowSurrogates = !dataIsEmpty(loneLowSurrogates);
|
---|
| 1009 |
|
---|
| 1010 | var surrogateMappings = surrogateSet(astral);
|
---|
| 1011 |
|
---|
| 1012 | if (bmpOnly) {
|
---|
| 1013 | bmp = dataAddData(bmp, loneHighSurrogates);
|
---|
| 1014 | hasLoneHighSurrogates = false;
|
---|
| 1015 | bmp = dataAddData(bmp, loneLowSurrogates);
|
---|
| 1016 | hasLoneLowSurrogates = false;
|
---|
| 1017 | }
|
---|
| 1018 |
|
---|
| 1019 | if (!dataIsEmpty(bmp)) {
|
---|
| 1020 | // The data set contains BMP code points that are not high surrogates
|
---|
| 1021 | // needed for astral code points in the set.
|
---|
| 1022 | result.push(createBMPCharacterClasses(bmp));
|
---|
| 1023 | }
|
---|
| 1024 | if (surrogateMappings.length) {
|
---|
| 1025 | // The data set contains astral code points; append character classes
|
---|
| 1026 | // based on their surrogate pairs.
|
---|
| 1027 | result.push(createSurrogateCharacterClasses(surrogateMappings));
|
---|
| 1028 | }
|
---|
| 1029 | // https://gist.github.com/mathiasbynens/bbe7f870208abcfec860
|
---|
| 1030 | if (hasLoneHighSurrogates) {
|
---|
| 1031 | result.push(
|
---|
| 1032 | createBMPCharacterClasses(loneHighSurrogates) +
|
---|
| 1033 | // Make sure the high surrogates aren’t part of a surrogate pair.
|
---|
| 1034 | '(?![\\uDC00-\\uDFFF])'
|
---|
| 1035 | );
|
---|
| 1036 | }
|
---|
| 1037 | if (hasLoneLowSurrogates) {
|
---|
| 1038 | result.push(
|
---|
| 1039 | // It is not possible to accurately assert the low surrogates aren’t
|
---|
| 1040 | // part of a surrogate pair, since JavaScript regular expressions do
|
---|
| 1041 | // not support lookbehind.
|
---|
| 1042 | '(?:[^\\uD800-\\uDBFF]|^)' +
|
---|
| 1043 | createBMPCharacterClasses(loneLowSurrogates)
|
---|
| 1044 | );
|
---|
| 1045 | }
|
---|
| 1046 | return result.join('|');
|
---|
| 1047 | };
|
---|
| 1048 |
|
---|
| 1049 | /*--------------------------------------------------------------------------*/
|
---|
| 1050 |
|
---|
| 1051 | // `regenerate` can be used as a constructor (and new methods can be added to
|
---|
| 1052 | // its prototype) but also as a regular function, the latter of which is the
|
---|
| 1053 | // documented and most common usage. For that reason, it’s not capitalized.
|
---|
| 1054 | var regenerate = function(value) {
|
---|
| 1055 | if (arguments.length > 1) {
|
---|
| 1056 | value = slice.call(arguments);
|
---|
| 1057 | }
|
---|
| 1058 | if (this instanceof regenerate) {
|
---|
| 1059 | this.data = [];
|
---|
| 1060 | return value ? this.add(value) : this;
|
---|
| 1061 | }
|
---|
| 1062 | return (new regenerate).add(value);
|
---|
| 1063 | };
|
---|
| 1064 |
|
---|
| 1065 | regenerate.version = '1.4.2';
|
---|
| 1066 |
|
---|
| 1067 | var proto = regenerate.prototype;
|
---|
| 1068 | extend(proto, {
|
---|
| 1069 | 'add': function(value) {
|
---|
| 1070 | var $this = this;
|
---|
| 1071 | if (value == null) {
|
---|
| 1072 | return $this;
|
---|
| 1073 | }
|
---|
| 1074 | if (value instanceof regenerate) {
|
---|
| 1075 | // Allow passing other Regenerate instances.
|
---|
| 1076 | $this.data = dataAddData($this.data, value.data);
|
---|
| 1077 | return $this;
|
---|
| 1078 | }
|
---|
| 1079 | if (arguments.length > 1) {
|
---|
| 1080 | value = slice.call(arguments);
|
---|
| 1081 | }
|
---|
| 1082 | if (isArray(value)) {
|
---|
| 1083 | forEach(value, function(item) {
|
---|
| 1084 | $this.add(item);
|
---|
| 1085 | });
|
---|
| 1086 | return $this;
|
---|
| 1087 | }
|
---|
| 1088 | $this.data = dataAdd(
|
---|
| 1089 | $this.data,
|
---|
| 1090 | isNumber(value) ? value : symbolToCodePoint(value)
|
---|
| 1091 | );
|
---|
| 1092 | return $this;
|
---|
| 1093 | },
|
---|
| 1094 | 'remove': function(value) {
|
---|
| 1095 | var $this = this;
|
---|
| 1096 | if (value == null) {
|
---|
| 1097 | return $this;
|
---|
| 1098 | }
|
---|
| 1099 | if (value instanceof regenerate) {
|
---|
| 1100 | // Allow passing other Regenerate instances.
|
---|
| 1101 | $this.data = dataRemoveData($this.data, value.data);
|
---|
| 1102 | return $this;
|
---|
| 1103 | }
|
---|
| 1104 | if (arguments.length > 1) {
|
---|
| 1105 | value = slice.call(arguments);
|
---|
| 1106 | }
|
---|
| 1107 | if (isArray(value)) {
|
---|
| 1108 | forEach(value, function(item) {
|
---|
| 1109 | $this.remove(item);
|
---|
| 1110 | });
|
---|
| 1111 | return $this;
|
---|
| 1112 | }
|
---|
| 1113 | $this.data = dataRemove(
|
---|
| 1114 | $this.data,
|
---|
| 1115 | isNumber(value) ? value : symbolToCodePoint(value)
|
---|
| 1116 | );
|
---|
| 1117 | return $this;
|
---|
| 1118 | },
|
---|
| 1119 | 'addRange': function(start, end) {
|
---|
| 1120 | var $this = this;
|
---|
| 1121 | $this.data = dataAddRange($this.data,
|
---|
| 1122 | isNumber(start) ? start : symbolToCodePoint(start),
|
---|
| 1123 | isNumber(end) ? end : symbolToCodePoint(end)
|
---|
| 1124 | );
|
---|
| 1125 | return $this;
|
---|
| 1126 | },
|
---|
| 1127 | 'removeRange': function(start, end) {
|
---|
| 1128 | var $this = this;
|
---|
| 1129 | var startCodePoint = isNumber(start) ? start : symbolToCodePoint(start);
|
---|
| 1130 | var endCodePoint = isNumber(end) ? end : symbolToCodePoint(end);
|
---|
| 1131 | $this.data = dataRemoveRange(
|
---|
| 1132 | $this.data,
|
---|
| 1133 | startCodePoint,
|
---|
| 1134 | endCodePoint
|
---|
| 1135 | );
|
---|
| 1136 | return $this;
|
---|
| 1137 | },
|
---|
| 1138 | 'intersection': function(argument) {
|
---|
| 1139 | var $this = this;
|
---|
| 1140 | // Allow passing other Regenerate instances.
|
---|
| 1141 | // TODO: Optimize this by writing and using `dataIntersectionData()`.
|
---|
| 1142 | var array = argument instanceof regenerate ?
|
---|
| 1143 | dataToArray(argument.data) :
|
---|
| 1144 | argument;
|
---|
| 1145 | $this.data = dataIntersection($this.data, array);
|
---|
| 1146 | return $this;
|
---|
| 1147 | },
|
---|
| 1148 | 'contains': function(codePoint) {
|
---|
| 1149 | return dataContains(
|
---|
| 1150 | this.data,
|
---|
| 1151 | isNumber(codePoint) ? codePoint : symbolToCodePoint(codePoint)
|
---|
| 1152 | );
|
---|
| 1153 | },
|
---|
| 1154 | 'clone': function() {
|
---|
| 1155 | var set = new regenerate;
|
---|
| 1156 | set.data = this.data.slice(0);
|
---|
| 1157 | return set;
|
---|
| 1158 | },
|
---|
| 1159 | 'toString': function(options) {
|
---|
| 1160 | var result = createCharacterClassesFromData(
|
---|
| 1161 | this.data,
|
---|
| 1162 | options ? options.bmpOnly : false,
|
---|
| 1163 | options ? options.hasUnicodeFlag : false
|
---|
| 1164 | );
|
---|
| 1165 | if (!result) {
|
---|
| 1166 | // For an empty set, return something that can be inserted `/here/` to
|
---|
| 1167 | // form a valid regular expression. Avoid `(?:)` since that matches the
|
---|
| 1168 | // empty string.
|
---|
| 1169 | return '[]';
|
---|
| 1170 | }
|
---|
| 1171 | // Use `\0` instead of `\x00` where possible.
|
---|
| 1172 | return result.replace(regexNull, '\\0$1');
|
---|
| 1173 | },
|
---|
| 1174 | 'toRegExp': function(flags) {
|
---|
| 1175 | var pattern = this.toString(
|
---|
| 1176 | flags && flags.indexOf('u') != -1 ?
|
---|
| 1177 | { 'hasUnicodeFlag': true } :
|
---|
| 1178 | null
|
---|
| 1179 | );
|
---|
| 1180 | return RegExp(pattern, flags || '');
|
---|
| 1181 | },
|
---|
| 1182 | 'valueOf': function() { // Note: `valueOf` is aliased as `toArray`.
|
---|
| 1183 | return dataToArray(this.data);
|
---|
| 1184 | }
|
---|
| 1185 | });
|
---|
| 1186 |
|
---|
| 1187 | proto.toArray = proto.valueOf;
|
---|
| 1188 |
|
---|
| 1189 | // Some AMD build optimizers, like r.js, check for specific condition patterns
|
---|
| 1190 | // like the following:
|
---|
| 1191 | if (
|
---|
| 1192 | typeof define == 'function' &&
|
---|
| 1193 | typeof define.amd == 'object' &&
|
---|
| 1194 | define.amd
|
---|
| 1195 | ) {
|
---|
| 1196 | define(function() {
|
---|
| 1197 | return regenerate;
|
---|
| 1198 | });
|
---|
| 1199 | } else if (freeExports && !freeExports.nodeType) {
|
---|
| 1200 | if (freeModule) { // in Node.js, io.js, or RingoJS v0.8.0+
|
---|
| 1201 | freeModule.exports = regenerate;
|
---|
| 1202 | } else { // in Narwhal or RingoJS v0.7.0-
|
---|
| 1203 | freeExports.regenerate = regenerate;
|
---|
| 1204 | }
|
---|
| 1205 | } else { // in Rhino or a web browser
|
---|
| 1206 | root.regenerate = regenerate;
|
---|
| 1207 | }
|
---|
| 1208 |
|
---|
| 1209 | }(this));
|
---|