1
|
// regjsparser
|
2
|
//
|
3
|
// ==================================================================
|
4
|
//
|
5
|
// See ECMA-262 Standard: 15.10.1
|
6
|
//
|
7
|
// NOTE: The ECMA-262 standard uses the term "Assertion" for /^/. Here the
|
8
|
// term "Anchor" is used.
|
9
|
//
|
10
|
// Pattern ::
|
11
|
// Disjunction
|
12
|
//
|
13
|
// Disjunction ::
|
14
|
// Alternative
|
15
|
// Alternative | Disjunction
|
16
|
//
|
17
|
// Alternative ::
|
18
|
// [empty]
|
19
|
// Alternative Term
|
20
|
//
|
21
|
// Term ::
|
22
|
// Anchor
|
23
|
// Atom
|
24
|
// Atom Quantifier
|
25
|
//
|
26
|
// Anchor ::
|
27
|
// ^
|
28
|
// $
|
29
|
// \ b
|
30
|
// \ B
|
31
|
// ( ? = Disjunction )
|
32
|
// ( ? ! Disjunction )
|
33
|
//
|
34
|
// Quantifier ::
|
35
|
// QuantifierPrefix
|
36
|
// QuantifierPrefix ?
|
37
|
//
|
38
|
// QuantifierPrefix ::
|
39
|
// *
|
40
|
// +
|
41
|
// ?
|
42
|
// { DecimalDigits }
|
43
|
// { DecimalDigits , }
|
44
|
// { DecimalDigits , DecimalDigits }
|
45
|
//
|
46
|
// Atom ::
|
47
|
// PatternCharacter
|
48
|
// .
|
49
|
// \ AtomEscape
|
50
|
// CharacterClass
|
51
|
// ( Disjunction )
|
52
|
// ( ? : Disjunction )
|
53
|
//
|
54
|
// PatternCharacter ::
|
55
|
// SourceCharacter but not any of: ^ $ \ . * + ? ( ) [ ] { } |
|
56
|
//
|
57
|
// AtomEscape ::
|
58
|
// DecimalEscape
|
59
|
// CharacterEscape
|
60
|
// CharacterClassEscape
|
61
|
//
|
62
|
// CharacterEscape[U] ::
|
63
|
// ControlEscape
|
64
|
// c ControlLetter
|
65
|
// HexEscapeSequence
|
66
|
// RegExpUnicodeEscapeSequence[?U] (ES6)
|
67
|
// IdentityEscape[?U]
|
68
|
//
|
69
|
// ControlEscape ::
|
70
|
// one of f n r t v
|
71
|
// ControlLetter ::
|
72
|
// one of
|
73
|
// a b c d e f g h i j k l m n o p q r s t u v w x y z
|
74
|
// A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
|
75
|
//
|
76
|
// IdentityEscape ::
|
77
|
// SourceCharacter but not IdentifierPart
|
78
|
// <ZWJ>
|
79
|
// <ZWNJ>
|
80
|
//
|
81
|
// DecimalEscape ::
|
82
|
// DecimalIntegerLiteral [lookahead ∉ DecimalDigit]
|
83
|
//
|
84
|
// CharacterClassEscape ::
|
85
|
// one of d D s S w W
|
86
|
//
|
87
|
// CharacterClass ::
|
88
|
// [ [lookahead ∉ {^}] ClassRanges ]
|
89
|
// [ ^ ClassRanges ]
|
90
|
//
|
91
|
// ClassRanges ::
|
92
|
// [empty]
|
93
|
// NonemptyClassRanges
|
94
|
//
|
95
|
// NonemptyClassRanges ::
|
96
|
// ClassAtom
|
97
|
// ClassAtom NonemptyClassRangesNoDash
|
98
|
// ClassAtom - ClassAtom ClassRanges
|
99
|
//
|
100
|
// NonemptyClassRangesNoDash ::
|
101
|
// ClassAtom
|
102
|
// ClassAtomNoDash NonemptyClassRangesNoDash
|
103
|
// ClassAtomNoDash - ClassAtom ClassRanges
|
104
|
//
|
105
|
// ClassAtom ::
|
106
|
// -
|
107
|
// ClassAtomNoDash
|
108
|
//
|
109
|
// ClassAtomNoDash ::
|
110
|
// SourceCharacter but not one of \ or ] or -
|
111
|
// \ ClassEscape
|
112
|
//
|
113
|
// ClassEscape ::
|
114
|
// DecimalEscape
|
115
|
// b
|
116
|
// CharacterEscape
|
117
|
// CharacterClassEscape
|
118
|
|
119
|
(function() {
|
120
|
|
121
|
function parse(str, flags) {
|
122
|
function addRaw(node) {
|
123
|
node.raw = str.substring(node.range[0], node.range[1]);
|
124
|
return node;
|
125
|
}
|
126
|
|
127
|
function updateRawStart(node, start) {
|
128
|
node.range[0] = start;
|
129
|
return addRaw(node);
|
130
|
}
|
131
|
|
132
|
function createAnchor(kind, rawLength) {
|
133
|
return addRaw({
|
134
|
type: 'anchor',
|
135
|
kind: kind,
|
136
|
range: [
|
137
|
pos - rawLength,
|
138
|
pos
|
139
|
]
|
140
|
});
|
141
|
}
|
142
|
|
143
|
function createValue(kind, codePoint, from, to) {
|
144
|
return addRaw({
|
145
|
type: 'value',
|
146
|
kind: kind,
|
147
|
codePoint: codePoint,
|
148
|
range: [from, to]
|
149
|
});
|
150
|
}
|
151
|
|
152
|
function createEscaped(kind, codePoint, value, fromOffset) {
|
153
|
fromOffset = fromOffset || 0;
|
154
|
return createValue(kind, codePoint, pos - (value.length + fromOffset), pos);
|
155
|
}
|
156
|
|
157
|
function createCharacter(matches) {
|
158
|
var _char = matches[0];
|
159
|
var first = _char.charCodeAt(0);
|
160
|
if (hasUnicodeFlag) {
|
161
|
var second;
|
162
|
if (_char.length === 1 && first >= 0xD800 && first <= 0xDBFF) {
|
163
|
second = lookahead().charCodeAt(0);
|
164
|
if (second >= 0xDC00 && second <= 0xDFFF) {
|
165
|
// Unicode surrogate pair
|
166
|
pos++;
|
167
|
return createValue(
|
168
|
'symbol',
|
169
|
(first - 0xD800) * 0x400 + second - 0xDC00 + 0x10000,
|
170
|
pos - 2, pos);
|
171
|
}
|
172
|
}
|
173
|
}
|
174
|
return createValue('symbol', first, pos - 1, pos);
|
175
|
}
|
176
|
|
177
|
function createDisjunction(alternatives, from, to) {
|
178
|
return addRaw({
|
179
|
type: 'disjunction',
|
180
|
body: alternatives,
|
181
|
range: [
|
182
|
from,
|
183
|
to
|
184
|
]
|
185
|
});
|
186
|
}
|
187
|
|
188
|
function createDot() {
|
189
|
return addRaw({
|
190
|
type: 'dot',
|
191
|
range: [
|
192
|
pos - 1,
|
193
|
pos
|
194
|
]
|
195
|
});
|
196
|
}
|
197
|
|
198
|
function createCharacterClassEscape(value) {
|
199
|
return addRaw({
|
200
|
type: 'characterClassEscape',
|
201
|
value: value,
|
202
|
range: [
|
203
|
pos - 2,
|
204
|
pos
|
205
|
]
|
206
|
});
|
207
|
}
|
208
|
|
209
|
function createReference(matchIndex) {
|
210
|
return addRaw({
|
211
|
type: 'reference',
|
212
|
matchIndex: parseInt(matchIndex, 10),
|
213
|
range: [
|
214
|
pos - 1 - matchIndex.length,
|
215
|
pos
|
216
|
]
|
217
|
});
|
218
|
}
|
219
|
|
220
|
function createGroup(behavior, disjunction, from, to) {
|
221
|
return addRaw({
|
222
|
type: 'group',
|
223
|
behavior: behavior,
|
224
|
body: disjunction,
|
225
|
range: [
|
226
|
from,
|
227
|
to
|
228
|
]
|
229
|
});
|
230
|
}
|
231
|
|
232
|
function createQuantifier(min, max, from, to) {
|
233
|
if (to == null) {
|
234
|
from = pos - 1;
|
235
|
to = pos;
|
236
|
}
|
237
|
|
238
|
return addRaw({
|
239
|
type: 'quantifier',
|
240
|
min: min,
|
241
|
max: max,
|
242
|
greedy: true,
|
243
|
body: null, // set later on
|
244
|
range: [
|
245
|
from,
|
246
|
to
|
247
|
]
|
248
|
});
|
249
|
}
|
250
|
|
251
|
function createAlternative(terms, from, to) {
|
252
|
return addRaw({
|
253
|
type: 'alternative',
|
254
|
body: terms,
|
255
|
range: [
|
256
|
from,
|
257
|
to
|
258
|
]
|
259
|
});
|
260
|
}
|
261
|
|
262
|
function createCharacterClass(classRanges, negative, from, to) {
|
263
|
return addRaw({
|
264
|
type: 'characterClass',
|
265
|
body: classRanges,
|
266
|
negative: negative,
|
267
|
range: [
|
268
|
from,
|
269
|
to
|
270
|
]
|
271
|
});
|
272
|
}
|
273
|
|
274
|
function createClassRange(min, max, from, to) {
|
275
|
// See 15.10.2.15:
|
276
|
if (min.codePoint > max.codePoint) {
|
277
|
bail('invalid range in character class', min.raw + '-' + max.raw, from, to);
|
278
|
}
|
279
|
|
280
|
return addRaw({
|
281
|
type: 'characterClassRange',
|
282
|
min: min,
|
283
|
max: max,
|
284
|
range: [
|
285
|
from,
|
286
|
to
|
287
|
]
|
288
|
});
|
289
|
}
|
290
|
|
291
|
function flattenBody(body) {
|
292
|
if (body.type === 'alternative') {
|
293
|
return body.body;
|
294
|
} else {
|
295
|
return [body];
|
296
|
}
|
297
|
}
|
298
|
|
299
|
function isEmpty(obj) {
|
300
|
return obj.type === 'empty';
|
301
|
}
|
302
|
|
303
|
function incr(amount) {
|
304
|
amount = (amount || 1);
|
305
|
var res = str.substring(pos, pos + amount);
|
306
|
pos += (amount || 1);
|
307
|
return res;
|
308
|
}
|
309
|
|
310
|
function skip(value) {
|
311
|
if (!match(value)) {
|
312
|
bail('character', value);
|
313
|
}
|
314
|
}
|
315
|
|
316
|
function match(value) {
|
317
|
if (str.indexOf(value, pos) === pos) {
|
318
|
return incr(value.length);
|
319
|
}
|
320
|
}
|
321
|
|
322
|
function lookahead() {
|
323
|
return str[pos];
|
324
|
}
|
325
|
|
326
|
function current(value) {
|
327
|
return str.indexOf(value, pos) === pos;
|
328
|
}
|
329
|
|
330
|
function next(value) {
|
331
|
return str[pos + 1] === value;
|
332
|
}
|
333
|
|
334
|
function matchReg(regExp) {
|
335
|
var subStr = str.substring(pos);
|
336
|
var res = subStr.match(regExp);
|
337
|
if (res) {
|
338
|
res.range = [];
|
339
|
res.range[0] = pos;
|
340
|
incr(res[0].length);
|
341
|
res.range[1] = pos;
|
342
|
}
|
343
|
return res;
|
344
|
}
|
345
|
|
346
|
function parseDisjunction() {
|
347
|
// Disjunction ::
|
348
|
// Alternative
|
349
|
// Alternative | Disjunction
|
350
|
var res = [], from = pos;
|
351
|
res.push(parseAlternative());
|
352
|
|
353
|
while (match('|')) {
|
354
|
res.push(parseAlternative());
|
355
|
}
|
356
|
|
357
|
if (res.length === 1) {
|
358
|
return res[0];
|
359
|
}
|
360
|
|
361
|
return createDisjunction(res, from, pos);
|
362
|
}
|
363
|
|
364
|
function parseAlternative() {
|
365
|
var res = [], from = pos;
|
366
|
var term;
|
367
|
|
368
|
// Alternative ::
|
369
|
// [empty]
|
370
|
// Alternative Term
|
371
|
while (term = parseTerm()) {
|
372
|
res.push(term);
|
373
|
}
|
374
|
|
375
|
if (res.length === 1) {
|
376
|
return res[0];
|
377
|
}
|
378
|
|
379
|
return createAlternative(res, from, pos);
|
380
|
}
|
381
|
|
382
|
function parseTerm() {
|
383
|
// Term ::
|
384
|
// Anchor
|
385
|
// Atom
|
386
|
// Atom Quantifier
|
387
|
|
388
|
if (pos >= str.length || current('|') || current(')')) {
|
389
|
return null; /* Means: The term is empty */
|
390
|
}
|
391
|
|
392
|
var anchor = parseAnchor();
|
393
|
|
394
|
if (anchor) {
|
395
|
return anchor;
|
396
|
}
|
397
|
|
398
|
var atom = parseAtom();
|
399
|
if (!atom) {
|
400
|
bail('Expected atom');
|
401
|
}
|
402
|
var quantifier = parseQuantifier() || false;
|
403
|
if (quantifier) {
|
404
|
quantifier.body = flattenBody(atom);
|
405
|
// The quantifier contains the atom. Therefore, the beginning of the
|
406
|
// quantifier range is given by the beginning of the atom.
|
407
|
updateRawStart(quantifier, atom.range[0]);
|
408
|
return quantifier;
|
409
|
}
|
410
|
return atom;
|
411
|
}
|
412
|
|
413
|
function parseGroup(matchA, typeA, matchB, typeB) {
|
414
|
var type = null, from = pos;
|
415
|
|
416
|
if (match(matchA)) {
|
417
|
type = typeA;
|
418
|
} else if (match(matchB)) {
|
419
|
type = typeB;
|
420
|
} else {
|
421
|
return false;
|
422
|
}
|
423
|
|
424
|
var body = parseDisjunction();
|
425
|
if (!body) {
|
426
|
bail('Expected disjunction');
|
427
|
}
|
428
|
skip(')');
|
429
|
var group = createGroup(type, flattenBody(body), from, pos);
|
430
|
|
431
|
if (type == 'normal') {
|
432
|
// Keep track of the number of closed groups. This is required for
|
433
|
// parseDecimalEscape(). In case the string is parsed a second time the
|
434
|
// value already holds the total count and no incrementation is required.
|
435
|
if (firstIteration) {
|
436
|
closedCaptureCounter++;
|
437
|
}
|
438
|
}
|
439
|
return group;
|
440
|
}
|
441
|
|
442
|
function parseAnchor() {
|
443
|
// Anchor ::
|
444
|
// ^
|
445
|
// $
|
446
|
// \ b
|
447
|
// \ B
|
448
|
// ( ? = Disjunction )
|
449
|
// ( ? ! Disjunction )
|
450
|
var res, from = pos;
|
451
|
|
452
|
if (match('^')) {
|
453
|
return createAnchor('start', 1 /* rawLength */);
|
454
|
} else if (match('$')) {
|
455
|
return createAnchor('end', 1 /* rawLength */);
|
456
|
} else if (match('\\b')) {
|
457
|
return createAnchor('boundary', 2 /* rawLength */);
|
458
|
} else if (match('\\B')) {
|
459
|
return createAnchor('not-boundary', 2 /* rawLength */);
|
460
|
} else {
|
461
|
return parseGroup('(?=', 'lookahead', '(?!', 'negativeLookahead');
|
462
|
}
|
463
|
}
|
464
|
|
465
|
function parseQuantifier() {
|
466
|
// Quantifier ::
|
467
|
// QuantifierPrefix
|
468
|
// QuantifierPrefix ?
|
469
|
//
|
470
|
// QuantifierPrefix ::
|
471
|
// *
|
472
|
// +
|
473
|
// ?
|
474
|
// { DecimalDigits }
|
475
|
// { DecimalDigits , }
|
476
|
// { DecimalDigits , DecimalDigits }
|
477
|
|
478
|
var res, from = pos;
|
479
|
var quantifier;
|
480
|
var min, max;
|
481
|
|
482
|
if (match('*')) {
|
483
|
quantifier = createQuantifier(0);
|
484
|
}
|
485
|
else if (match('+')) {
|
486
|
quantifier = createQuantifier(1);
|
487
|
}
|
488
|
else if (match('?')) {
|
489
|
quantifier = createQuantifier(0, 1);
|
490
|
}
|
491
|
else if (res = matchReg(/^\{([0-9]+)\}/)) {
|
492
|
min = parseInt(res[1], 10);
|
493
|
quantifier = createQuantifier(min, min, res.range[0], res.range[1]);
|
494
|
}
|
495
|
else if (res = matchReg(/^\{([0-9]+),\}/)) {
|
496
|
min = parseInt(res[1], 10);
|
497
|
quantifier = createQuantifier(min, undefined, res.range[0], res.range[1]);
|
498
|
}
|
499
|
else if (res = matchReg(/^\{([0-9]+),([0-9]+)\}/)) {
|
500
|
min = parseInt(res[1], 10);
|
501
|
max = parseInt(res[2], 10);
|
502
|
if (min > max) {
|
503
|
bail('numbers out of order in {} quantifier', '', from, pos);
|
504
|
}
|
505
|
quantifier = createQuantifier(min, max, res.range[0], res.range[1]);
|
506
|
}
|
507
|
|
508
|
if (quantifier) {
|
509
|
if (match('?')) {
|
510
|
quantifier.greedy = false;
|
511
|
quantifier.range[1] += 1;
|
512
|
}
|
513
|
}
|
514
|
|
515
|
return quantifier;
|
516
|
}
|
517
|
|
518
|
function parseAtom() {
|
519
|
// Atom ::
|
520
|
// PatternCharacter
|
521
|
// .
|
522
|
// \ AtomEscape
|
523
|
// CharacterClass
|
524
|
// ( Disjunction )
|
525
|
// ( ? : Disjunction )
|
526
|
|
527
|
var res;
|
528
|
|
529
|
// jviereck: allow ']', '}' here as well to be compatible with browser's
|
530
|
// implementations: ']'.match(/]/);
|
531
|
// if (res = matchReg(/^[^^$\\.*+?()[\]{}|]/)) {
|
532
|
if (res = matchReg(/^[^^$\\.*+?(){[|]/)) {
|
533
|
// PatternCharacter
|
534
|
return createCharacter(res);
|
535
|
}
|
536
|
else if (match('.')) {
|
537
|
// .
|
538
|
return createDot();
|
539
|
}
|
540
|
else if (match('\\')) {
|
541
|
// \ AtomEscape
|
542
|
res = parseAtomEscape();
|
543
|
if (!res) {
|
544
|
bail('atomEscape');
|
545
|
}
|
546
|
return res;
|
547
|
}
|
548
|
else if (res = parseCharacterClass()) {
|
549
|
return res;
|
550
|
}
|
551
|
else {
|
552
|
// ( Disjunction )
|
553
|
// ( ? : Disjunction )
|
554
|
return parseGroup('(?:', 'ignore', '(', 'normal');
|
555
|
}
|
556
|
}
|
557
|
|
558
|
function parseUnicodeSurrogatePairEscape(firstEscape) {
|
559
|
if (hasUnicodeFlag) {
|
560
|
var first, second;
|
561
|
if (firstEscape.kind == 'unicodeEscape' &&
|
562
|
(first = firstEscape.codePoint) >= 0xD800 && first <= 0xDBFF &&
|
563
|
current('\\') && next('u') ) {
|
564
|
var prevPos = pos;
|
565
|
pos++;
|
566
|
var secondEscape = parseClassEscape();
|
567
|
if (secondEscape.kind == 'unicodeEscape' &&
|
568
|
(second = secondEscape.codePoint) >= 0xDC00 && second <= 0xDFFF) {
|
569
|
// Unicode surrogate pair
|
570
|
firstEscape.range[1] = secondEscape.range[1];
|
571
|
firstEscape.codePoint = (first - 0xD800) * 0x400 + second - 0xDC00 + 0x10000;
|
572
|
firstEscape.type = 'value';
|
573
|
firstEscape.kind = 'unicodeCodePointEscape';
|
574
|
addRaw(firstEscape);
|
575
|
}
|
576
|
else {
|
577
|
pos = prevPos;
|
578
|
}
|
579
|
}
|
580
|
}
|
581
|
return firstEscape;
|
582
|
}
|
583
|
|
584
|
function parseClassEscape() {
|
585
|
return parseAtomEscape(true);
|
586
|
}
|
587
|
|
588
|
function parseAtomEscape(insideCharacterClass) {
|
589
|
// AtomEscape ::
|
590
|
// DecimalEscape
|
591
|
// CharacterEscape
|
592
|
// CharacterClassEscape
|
593
|
|
594
|
var res, from = pos;
|
595
|
|
596
|
res = parseDecimalEscape();
|
597
|
if (res) {
|
598
|
return res;
|
599
|
}
|
600
|
|
601
|
// For ClassEscape
|
602
|
if (insideCharacterClass) {
|
603
|
if (match('b')) {
|
604
|
// 15.10.2.19
|
605
|
// The production ClassEscape :: b evaluates by returning the
|
606
|
// CharSet containing the one character <BS> (Unicode value 0008).
|
607
|
return createEscaped('singleEscape', 0x0008, '\\b');
|
608
|
} else if (match('B')) {
|
609
|
bail('\\B not possible inside of CharacterClass', '', from);
|
610
|
}
|
611
|
}
|
612
|
|
613
|
res = parseCharacterEscape();
|
614
|
|
615
|
return res;
|
616
|
}
|
617
|
|
618
|
|
619
|
function parseDecimalEscape() {
|
620
|
// DecimalEscape ::
|
621
|
// DecimalIntegerLiteral [lookahead ∉ DecimalDigit]
|
622
|
// CharacterClassEscape :: one of d D s S w W
|
623
|
|
624
|
var res, match;
|
625
|
|
626
|
if (res = matchReg(/^(?!0)\d+/)) {
|
627
|
match = res[0];
|
628
|
var refIdx = parseInt(res[0], 10);
|
629
|
if (refIdx <= closedCaptureCounter) {
|
630
|
// If the number is smaller than the normal-groups found so
|
631
|
// far, then it is a reference...
|
632
|
return createReference(res[0]);
|
633
|
} else {
|
634
|
// ... otherwise it needs to be interpreted as a octal (if the
|
635
|
// number is in an octal format). If it is NOT octal format,
|
636
|
// then the slash is ignored and the number is matched later
|
637
|
// as normal characters.
|
638
|
|
639
|
// Recall the negative decision to decide if the input must be parsed
|
640
|
// a second time with the total normal-groups.
|
641
|
backrefDenied.push(refIdx);
|
642
|
|
643
|
// Reset the position again, as maybe only parts of the previous
|
644
|
// matched numbers are actual octal numbers. E.g. in '019' only
|
645
|
// the '01' should be matched.
|
646
|
incr(-res[0].length);
|
647
|
if (res = matchReg(/^[0-7]{1,3}/)) {
|
648
|
return createEscaped('octal', parseInt(res[0], 8), res[0], 1);
|
649
|
} else {
|
650
|
// If we end up here, we have a case like /\91/. Then the
|
651
|
// first slash is to be ignored and the 9 & 1 to be treated
|
652
|
// like ordinary characters. Create a character for the
|
653
|
// first number only here - other number-characters
|
654
|
// (if available) will be matched later.
|
655
|
res = createCharacter(matchReg(/^[89]/));
|
656
|
return updateRawStart(res, res.range[0] - 1);
|
657
|
}
|
658
|
}
|
659
|
}
|
660
|
// Only allow octal numbers in the following. All matched numbers start
|
661
|
// with a zero (if the do not, the previous if-branch is executed).
|
662
|
// If the number is not octal format and starts with zero (e.g. `091`)
|
663
|
// then only the zeros `0` is treated here and the `91` are ordinary
|
664
|
// characters.
|
665
|
// Example:
|
666
|
// /\091/.exec('\091')[0].length === 3
|
667
|
else if (res = matchReg(/^[0-7]{1,3}/)) {
|
668
|
match = res[0];
|
669
|
if (/^0{1,3}$/.test(match)) {
|
670
|
// If they are all zeros, then only take the first one.
|
671
|
return createEscaped('null', 0x0000, '0', match.length + 1);
|
672
|
} else {
|
673
|
return createEscaped('octal', parseInt(match, 8), match, 1);
|
674
|
}
|
675
|
} else if (res = matchReg(/^[dDsSwW]/)) {
|
676
|
return createCharacterClassEscape(res[0]);
|
677
|
}
|
678
|
return false;
|
679
|
}
|
680
|
|
681
|
function parseCharacterEscape() {
|
682
|
// CharacterEscape ::
|
683
|
// ControlEscape
|
684
|
// c ControlLetter
|
685
|
// HexEscapeSequence
|
686
|
// UnicodeEscapeSequence
|
687
|
// IdentityEscape
|
688
|
|
689
|
var res;
|
690
|
if (res = matchReg(/^[fnrtv]/)) {
|
691
|
// ControlEscape
|
692
|
var codePoint = 0;
|
693
|
switch (res[0]) {
|
694
|
case 't': codePoint = 0x009; break;
|
695
|
case 'n': codePoint = 0x00A; break;
|
696
|
case 'v': codePoint = 0x00B; break;
|
697
|
case 'f': codePoint = 0x00C; break;
|
698
|
case 'r': codePoint = 0x00D; break;
|
699
|
}
|
700
|
return createEscaped('singleEscape', codePoint, '\\' + res[0]);
|
701
|
} else if (res = matchReg(/^c([a-zA-Z])/)) {
|
702
|
// c ControlLetter
|
703
|
return createEscaped('controlLetter', res[1].charCodeAt(0) % 32, res[1], 2);
|
704
|
} else if (res = matchReg(/^x([0-9a-fA-F]{2})/)) {
|
705
|
// HexEscapeSequence
|
706
|
return createEscaped('hexadecimalEscape', parseInt(res[1], 16), res[1], 2);
|
707
|
} else if (res = matchReg(/^u([0-9a-fA-F]{4})/)) {
|
708
|
// UnicodeEscapeSequence
|
709
|
return parseUnicodeSurrogatePairEscape(
|
710
|
createEscaped('unicodeEscape', parseInt(res[1], 16), res[1], 2)
|
711
|
);
|
712
|
} else if (hasUnicodeFlag && (res = matchReg(/^u\{([0-9a-fA-F]+)\}/))) {
|
713
|
// RegExpUnicodeEscapeSequence (ES6 Unicode code point escape)
|
714
|
return createEscaped('unicodeCodePointEscape', parseInt(res[1], 16), res[1], 4);
|
715
|
} else {
|
716
|
// IdentityEscape
|
717
|
return parseIdentityEscape();
|
718
|
}
|
719
|
}
|
720
|
|
721
|
// Taken from the Esprima parser.
|
722
|
function isIdentifierPart(ch) {
|
723
|
// Generated by `tools/generate-identifier-regex.js`.
|
724
|
var NonAsciiIdentifierPart = new RegExp('[\xAA\xB5\xBA\xC0-\xD6\xD8-\xF6\xF8-\u02C1\u02C6-\u02D1\u02E0-\u02E4\u02EC\u02EE\u0300-\u0374\u0376\u0377\u037A-\u037D\u037F\u0386\u0388-\u038A\u038C\u038E-\u03A1\u03A3-\u03F5\u03F7-\u0481\u0483-\u0487\u048A-\u052F\u0531-\u0556\u0559\u0561-\u0587\u0591-\u05BD\u05BF\u05C1\u05C2\u05C4\u05C5\u05C7\u05D0-\u05EA\u05F0-\u05F2\u0610-\u061A\u0620-\u0669\u066E-\u06D3\u06D5-\u06DC\u06DF-\u06E8\u06EA-\u06FC\u06FF\u0710-\u074A\u074D-\u07B1\u07C0-\u07F5\u07FA\u0800-\u082D\u0840-\u085B\u08A0-\u08B2\u08E4-\u0963\u0966-\u096F\u0971-\u0983\u0985-\u098C\u098F\u0990\u0993-\u09A8\u09AA-\u09B0\u09B2\u09B6-\u09B9\u09BC-\u09C4\u09C7\u09C8\u09CB-\u09CE\u09D7\u09DC\u09DD\u09DF-\u09E3\u09E6-\u09F1\u0A01-\u0A03\u0A05-\u0A0A\u0A0F\u0A10\u0A13-\u0A28\u0A2A-\u0A30\u0A32\u0A33\u0A35\u0A36\u0A38\u0A39\u0A3C\u0A3E-\u0A42\u0A47\u0A48\u0A4B-\u0A4D\u0A51\u0A59-\u0A5C\u0A5E\u0A66-\u0A75\u0A81-\u0A83\u0A85-\u0A8D\u0A8F-\u0A91\u0A93-\u0AA8\u0AAA-\u0AB0\u0AB2\u0AB3\u0AB5-\u0AB9\u0ABC-\u0AC5\u0AC7-\u0AC9\u0ACB-\u0ACD\u0AD0\u0AE0-\u0AE3\u0AE6-\u0AEF\u0B01-\u0B03\u0B05-\u0B0C\u0B0F\u0B10\u0B13-\u0B28\u0B2A-\u0B30\u0B32\u0B33\u0B35-\u0B39\u0B3C-\u0B44\u0B47\u0B48\u0B4B-\u0B4D\u0B56\u0B57\u0B5C\u0B5D\u0B5F-\u0B63\u0B66-\u0B6F\u0B71\u0B82\u0B83\u0B85-\u0B8A\u0B8E-\u0B90\u0B92-\u0B95\u0B99\u0B9A\u0B9C\u0B9E\u0B9F\u0BA3\u0BA4\u0BA8-\u0BAA\u0BAE-\u0BB9\u0BBE-\u0BC2\u0BC6-\u0BC8\u0BCA-\u0BCD\u0BD0\u0BD7\u0BE6-\u0BEF\u0C00-\u0C03\u0C05-\u0C0C\u0C0E-\u0C10\u0C12-\u0C28\u0C2A-\u0C39\u0C3D-\u0C44\u0C46-\u0C48\u0C4A-\u0C4D\u0C55\u0C56\u0C58\u0C59\u0C60-\u0C63\u0C66-\u0C6F\u0C81-\u0C83\u0C85-\u0C8C\u0C8E-\u0C90\u0C92-\u0CA8\u0CAA-\u0CB3\u0CB5-\u0CB9\u0CBC-\u0CC4\u0CC6-\u0CC8\u0CCA-\u0CCD\u0CD5\u0CD6\u0CDE\u0CE0-\u0CE3\u0CE6-\u0CEF\u0CF1\u0CF2\u0D01-\u0D03\u0D05-\u0D0C\u0D0E-\u0D10\u0D12-\u0D3A\u0D3D-\u0D44\u0D46-\u0D48\u0D4A-\u0D4E\u0D57\u0D60-\u0D63\u0D66-\u0D6F\u0D7A-\u0D7F\u0D82\u0D83\u0D85-\u0D96\u0D9A-\u0DB1\u0DB3-\u0DBB\u0DBD\u0DC0-\u0DC6\u0DCA\u0DCF-\u0DD4\u0DD6\u0DD8-\u0DDF\u0DE6-\u0DEF\u0DF2\u0DF3\u0E01-\u0E3A\u0E40-\u0E4E\u0E50-\u0E59\u0E81\u0E82\u0E84\u0E87\u0E88\u0E8A\u0E8D\u0E94-\u0E97\u0E99-\u0E9F\u0EA1-\u0EA3\u0EA5\u0EA7\u0EAA\u0EAB\u0EAD-\u0EB9\u0EBB-\u0EBD\u0EC0-\u0EC4\u0EC6\u0EC8-\u0ECD\u0ED0-\u0ED9\u0EDC-\u0EDF\u0F00\u0F18\u0F19\u0F20-\u0F29\u0F35\u0F37\u0F39\u0F3E-\u0F47\u0F49-\u0F6C\u0F71-\u0F84\u0F86-\u0F97\u0F99-\u0FBC\u0FC6\u1000-\u1049\u1050-\u109D\u10A0-\u10C5\u10C7\u10CD\u10D0-\u10FA\u10FC-\u1248\u124A-\u124D\u1250-\u1256\u1258\u125A-\u125D\u1260-\u1288\u128A-\u128D\u1290-\u12B0\u12B2-\u12B5\u12B8-\u12BE\u12C0\u12C2-\u12C5\u12C8-\u12D6\u12D8-\u1310\u1312-\u1315\u1318-\u135A\u135D-\u135F\u1380-\u138F\u13A0-\u13F4\u1401-\u166C\u166F-\u167F\u1681-\u169A\u16A0-\u16EA\u16EE-\u16F8\u1700-\u170C\u170E-\u1714\u1720-\u1734\u1740-\u1753\u1760-\u176C\u176E-\u1770\u1772\u1773\u1780-\u17D3\u17D7\u17DC\u17DD\u17E0-\u17E9\u180B-\u180D\u1810-\u1819\u1820-\u1877\u1880-\u18AA\u18B0-\u18F5\u1900-\u191E\u1920-\u192B\u1930-\u193B\u1946-\u196D\u1970-\u1974\u1980-\u19AB\u19B0-\u19C9\u19D0-\u19D9\u1A00-\u1A1B\u1A20-\u1A5E\u1A60-\u1A7C\u1A7F-\u1A89\u1A90-\u1A99\u1AA7\u1AB0-\u1ABD\u1B00-\u1B4B\u1B50-\u1B59\u1B6B-\u1B73\u1B80-\u1BF3\u1C00-\u1C37\u1C40-\u1C49\u1C4D-\u1C7D\u1CD0-\u1CD2\u1CD4-\u1CF6\u1CF8\u1CF9\u1D00-\u1DF5\u1DFC-\u1F15\u1F18-\u1F1D\u1F20-\u1F45\u1F48-\u1F4D\u1F50-\u1F57\u1F59\u1F5B\u1F5D\u1F5F-\u1F7D\u1F80-\u1FB4\u1FB6-\u1FBC\u1FBE\u1FC2-\u1FC4\u1FC6-\u1FCC\u1FD0-\u1FD3\u1FD6-\u1FDB\u1FE0-\u1FEC\u1FF2-\u1FF4\u1FF6-\u1FFC\u200C\u200D\u203F\u2040\u2054\u2071\u207F\u2090-\u209C\u20D0-\u20DC\u20E1\u20E5-\u20F0\u2102\u2107\u210A-\u2113\u2115\u2119-\u211D\u2124\u2126\u2128\u212A-\u212D\u212F-\u2139\u213C-\u213F\u2145-\u2149\u214E\u2160-\u2188\u2C00-\u2C2E\u2C30-\u2C5E\u2C60-\u2CE4\u2CEB-\u2CF3\u2D00-\u2D25\u2D27\u2D2D\u2D30-\u2D67\u2D6F\u2D7F-\u2D96\u2DA0-\u2DA6\u2DA8-\u2DAE\u2DB0-\u2DB6\u2DB8-\u2DBE\u2DC0-\u2DC6\u2DC8-\u2DCE\u2DD0-\u2DD6\u2DD8-\u2DDE\u2DE0-\u2DFF\u2E2F\u3005-\u3007\u3021-\u302F\u3031-\u3035\u3038-\u303C\u3041-\u3096\u3099\u309A\u309D-\u309F\u30A1-\u30FA\u30FC-\u30FF\u3105-\u312D\u3131-\u318E\u31A0-\u31BA\u31F0-\u31FF\u3400-\u4DB5\u4E00-\u9FCC\uA000-\uA48C\uA4D0-\uA4FD\uA500-\uA60C\uA610-\uA62B\uA640-\uA66F\uA674-\uA67D\uA67F-\uA69D\uA69F-\uA6F1\uA717-\uA71F\uA722-\uA788\uA78B-\uA78E\uA790-\uA7AD\uA7B0\uA7B1\uA7F7-\uA827\uA840-\uA873\uA880-\uA8C4\uA8D0-\uA8D9\uA8E0-\uA8F7\uA8FB\uA900-\uA92D\uA930-\uA953\uA960-\uA97C\uA980-\uA9C0\uA9CF-\uA9D9\uA9E0-\uA9FE\uAA00-\uAA36\uAA40-\uAA4D\uAA50-\uAA59\uAA60-\uAA76\uAA7A-\uAAC2\uAADB-\uAADD\uAAE0-\uAAEF\uAAF2-\uAAF6\uAB01-\uAB06\uAB09-\uAB0E\uAB11-\uAB16\uAB20-\uAB26\uAB28-\uAB2E\uAB30-\uAB5A\uAB5C-\uAB5F\uAB64\uAB65\uABC0-\uABEA\uABEC\uABED\uABF0-\uABF9\uAC00-\uD7A3\uD7B0-\uD7C6\uD7CB-\uD7FB\uF900-\uFA6D\uFA70-\uFAD9\uFB00-\uFB06\uFB13-\uFB17\uFB1D-\uFB28\uFB2A-\uFB36\uFB38-\uFB3C\uFB3E\uFB40\uFB41\uFB43\uFB44\uFB46-\uFBB1\uFBD3-\uFD3D\uFD50-\uFD8F\uFD92-\uFDC7\uFDF0-\uFDFB\uFE00-\uFE0F\uFE20-\uFE2D\uFE33\uFE34\uFE4D-\uFE4F\uFE70-\uFE74\uFE76-\uFEFC\uFF10-\uFF19\uFF21-\uFF3A\uFF3F\uFF41-\uFF5A\uFF66-\uFFBE\uFFC2-\uFFC7\uFFCA-\uFFCF\uFFD2-\uFFD7\uFFDA-\uFFDC]');
|
725
|
|
726
|
return (ch === 36) || (ch === 95) || // $ (dollar) and _ (underscore)
|
727
|
(ch >= 65 && ch <= 90) || // A..Z
|
728
|
(ch >= 97 && ch <= 122) || // a..z
|
729
|
(ch >= 48 && ch <= 57) || // 0..9
|
730
|
(ch === 92) || // \ (backslash)
|
731
|
((ch >= 0x80) && NonAsciiIdentifierPart.test(String.fromCharCode(ch)));
|
732
|
}
|
733
|
|
734
|
function parseIdentityEscape() {
|
735
|
// IdentityEscape ::
|
736
|
// SourceCharacter but not IdentifierPart
|
737
|
// <ZWJ>
|
738
|
// <ZWNJ>
|
739
|
|
740
|
var ZWJ = '\u200C';
|
741
|
var ZWNJ = '\u200D';
|
742
|
|
743
|
var tmp;
|
744
|
|
745
|
if (!isIdentifierPart(lookahead())) {
|
746
|
tmp = incr();
|
747
|
return createEscaped('identifier', tmp.charCodeAt(0), tmp, 1);
|
748
|
}
|
749
|
|
750
|
if (match(ZWJ)) {
|
751
|
// <ZWJ>
|
752
|
return createEscaped('identifier', 0x200C, ZWJ);
|
753
|
} else if (match(ZWNJ)) {
|
754
|
// <ZWNJ>
|
755
|
return createEscaped('identifier', 0x200D, ZWNJ);
|
756
|
}
|
757
|
|
758
|
return null;
|
759
|
}
|
760
|
|
761
|
function parseCharacterClass() {
|
762
|
// CharacterClass ::
|
763
|
// [ [lookahead ∉ {^}] ClassRanges ]
|
764
|
// [ ^ ClassRanges ]
|
765
|
|
766
|
var res, from = pos;
|
767
|
if (res = matchReg(/^\[\^/)) {
|
768
|
res = parseClassRanges();
|
769
|
skip(']');
|
770
|
return createCharacterClass(res, true, from, pos);
|
771
|
} else if (match('[')) {
|
772
|
res = parseClassRanges();
|
773
|
skip(']');
|
774
|
return createCharacterClass(res, false, from, pos);
|
775
|
}
|
776
|
|
777
|
return null;
|
778
|
}
|
779
|
|
780
|
function parseClassRanges() {
|
781
|
// ClassRanges ::
|
782
|
// [empty]
|
783
|
// NonemptyClassRanges
|
784
|
|
785
|
var res;
|
786
|
if (current(']')) {
|
787
|
// Empty array means nothing insinde of the ClassRange.
|
788
|
return [];
|
789
|
} else {
|
790
|
res = parseNonemptyClassRanges();
|
791
|
if (!res) {
|
792
|
bail('nonEmptyClassRanges');
|
793
|
}
|
794
|
return res;
|
795
|
}
|
796
|
}
|
797
|
|
798
|
function parseHelperClassRanges(atom) {
|
799
|
var from, to, res;
|
800
|
if (current('-') && !next(']')) {
|
801
|
// ClassAtom - ClassAtom ClassRanges
|
802
|
skip('-');
|
803
|
|
804
|
res = parseClassAtom();
|
805
|
if (!res) {
|
806
|
bail('classAtom');
|
807
|
}
|
808
|
to = pos;
|
809
|
var classRanges = parseClassRanges();
|
810
|
if (!classRanges) {
|
811
|
bail('classRanges');
|
812
|
}
|
813
|
from = atom.range[0];
|
814
|
if (classRanges.type === 'empty') {
|
815
|
return [createClassRange(atom, res, from, to)];
|
816
|
}
|
817
|
return [createClassRange(atom, res, from, to)].concat(classRanges);
|
818
|
}
|
819
|
|
820
|
res = parseNonemptyClassRangesNoDash();
|
821
|
if (!res) {
|
822
|
bail('nonEmptyClassRangesNoDash');
|
823
|
}
|
824
|
|
825
|
return [atom].concat(res);
|
826
|
}
|
827
|
|
828
|
function parseNonemptyClassRanges() {
|
829
|
// NonemptyClassRanges ::
|
830
|
// ClassAtom
|
831
|
// ClassAtom NonemptyClassRangesNoDash
|
832
|
// ClassAtom - ClassAtom ClassRanges
|
833
|
|
834
|
var atom = parseClassAtom();
|
835
|
if (!atom) {
|
836
|
bail('classAtom');
|
837
|
}
|
838
|
|
839
|
if (current(']')) {
|
840
|
// ClassAtom
|
841
|
return [atom];
|
842
|
}
|
843
|
|
844
|
// ClassAtom NonemptyClassRangesNoDash
|
845
|
// ClassAtom - ClassAtom ClassRanges
|
846
|
return parseHelperClassRanges(atom);
|
847
|
}
|
848
|
|
849
|
function parseNonemptyClassRangesNoDash() {
|
850
|
// NonemptyClassRangesNoDash ::
|
851
|
// ClassAtom
|
852
|
// ClassAtomNoDash NonemptyClassRangesNoDash
|
853
|
// ClassAtomNoDash - ClassAtom ClassRanges
|
854
|
|
855
|
var res = parseClassAtom();
|
856
|
if (!res) {
|
857
|
bail('classAtom');
|
858
|
}
|
859
|
if (current(']')) {
|
860
|
// ClassAtom
|
861
|
return res;
|
862
|
}
|
863
|
|
864
|
// ClassAtomNoDash NonemptyClassRangesNoDash
|
865
|
// ClassAtomNoDash - ClassAtom ClassRanges
|
866
|
return parseHelperClassRanges(res);
|
867
|
}
|
868
|
|
869
|
function parseClassAtom() {
|
870
|
// ClassAtom ::
|
871
|
// -
|
872
|
// ClassAtomNoDash
|
873
|
if (match('-')) {
|
874
|
return createCharacter('-');
|
875
|
} else {
|
876
|
return parseClassAtomNoDash();
|
877
|
}
|
878
|
}
|
879
|
|
880
|
function parseClassAtomNoDash() {
|
881
|
// ClassAtomNoDash ::
|
882
|
// SourceCharacter but not one of \ or ] or -
|
883
|
// \ ClassEscape
|
884
|
|
885
|
var res;
|
886
|
if (res = matchReg(/^[^\\\]-]/)) {
|
887
|
return createCharacter(res[0]);
|
888
|
} else if (match('\\')) {
|
889
|
res = parseClassEscape();
|
890
|
if (!res) {
|
891
|
bail('classEscape');
|
892
|
}
|
893
|
|
894
|
return parseUnicodeSurrogatePairEscape(res);
|
895
|
}
|
896
|
}
|
897
|
|
898
|
function bail(message, details, from, to) {
|
899
|
from = from == null ? pos : from;
|
900
|
to = to == null ? from : to;
|
901
|
|
902
|
var contextStart = Math.max(0, from - 10);
|
903
|
var contextEnd = Math.min(to + 10, str.length);
|
904
|
|
905
|
// Output a bit of context and a line pointing to where our error is.
|
906
|
//
|
907
|
// We are assuming that there are no actual newlines in the content as this is a regular expression.
|
908
|
var context = ' ' + str.substring(contextStart, contextEnd);
|
909
|
var pointer = ' ' + new Array(from - contextStart + 1).join(' ') + '^';
|
910
|
|
911
|
throw SyntaxError(message + ' at position ' + from + (details ? ': ' + details : '') + '\n' + context + '\n' + pointer);
|
912
|
}
|
913
|
|
914
|
var backrefDenied = [];
|
915
|
var closedCaptureCounter = 0;
|
916
|
var firstIteration = true;
|
917
|
var hasUnicodeFlag = (flags || "").indexOf("u") !== -1;
|
918
|
var pos = 0;
|
919
|
|
920
|
// Convert the input to a string and treat the empty string special.
|
921
|
str = String(str);
|
922
|
if (str === '') {
|
923
|
str = '(?:)';
|
924
|
}
|
925
|
|
926
|
var result = parseDisjunction();
|
927
|
|
928
|
if (result.range[1] !== str.length) {
|
929
|
bail('Could not parse entire input - got stuck', '', result.range[1]);
|
930
|
}
|
931
|
|
932
|
// The spec requires to interpret the `\2` in `/\2()()/` as backreference.
|
933
|
// As the parser collects the number of capture groups as the string is
|
934
|
// parsed it is impossible to make these decisions at the point when the
|
935
|
// `\2` is handled. In case the local decision turns out to be wrong after
|
936
|
// the parsing has finished, the input string is parsed a second time with
|
937
|
// the total number of capture groups set.
|
938
|
//
|
939
|
// SEE: https://github.com/jviereck/regjsparser/issues/70
|
940
|
for (var i = 0; i < backrefDenied.length; i++) {
|
941
|
if (backrefDenied[i] <= closedCaptureCounter) {
|
942
|
// Parse the input a second time.
|
943
|
pos = 0;
|
944
|
firstIteration = false;
|
945
|
return parseDisjunction();
|
946
|
}
|
947
|
}
|
948
|
|
949
|
return result;
|
950
|
}
|
951
|
|
952
|
var regjsparser = {
|
953
|
parse: parse
|
954
|
};
|
955
|
|
956
|
if (typeof module !== 'undefined' && module.exports) {
|
957
|
module.exports = regjsparser;
|
958
|
} else {
|
959
|
window.regjsparser = regjsparser;
|
960
|
}
|
961
|
|
962
|
}());
|