1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
|
<?php
namespace HTML5\Parser;
/**
* The HTML5 tokenizer.
*
* The tokenizer's role is reading data from the scanner and gathering it into
* semantic units. From the tokenizer, data is emitted to an event handler,
* which may (for example) create a DOM tree.
*
* The HTML5 specification has a detailed explanation of tokenizing HTML5. We
* follow that specification to the maximum extent that we can. If you find
* a discrepancy that is not documented, please file a bug and/or submit a
* patch.
*
* This tokenizer is implemented as a recursive descent parser.
*
* Within the API documentation, you may see references to the specific section
* of the HTML5 spec that the code attempts to reproduce. Example: 8.2.4.1.
* This refers to section 8.2.4.1 of the HTML5 CR specification.
*
* @see http://www.w3.org/TR/2012/CR-html5-20121217/
*/
class Tokenizer {
protected $scanner;
protected $events;
protected $tok;
/**
* Buffer for text.
*/
protected $text = '';
// When this goes to false, the parser stops.
protected $carryOn = TRUE;
/**
* Create a new tokenizer.
*
* Typically, parsing a document involves creating a new tokenizer, giving
* it a scanner (input) and an event handler (output), and then calling
* the Tokenizer::parse() method.`
*
* @param \HTML5\Parser\Scanner $scanner
* A scanner initialized with an input stream.
* @param \HTML5\Parser\EventHandler $eventHandler
* An event handler, initialized and ready to receive
* events.
*/
public function __construct($scanner, $eventHandler) {
$this->scanner = $scanner;
$this->events = $eventHandler;
}
/**
* Main entry point.
*/
public function parse() {
$p = 0;
do {
$p = $this->scanner->position();
$this->consumeData();
// FIXME: Add infinite loop protection.
}
while ($this->carryOn);
}
/**
* Consume a character and make a move.
* HTML5 8.2.4.1
*/
protected function consumeData() {
// Character Ref
$this->characterReference() ||
$this->tagOpen() ||
$this->eof() ||
$this->characterData();
return $this->carryOn;
}
/**
* This buffers the current token as character data.
*/
protected function characterData() {
$tok = $this->scanner->current();
// This should never happen...
if ($tok === FALSE) {
return FALSE;
}
// Null
if ($tok === "\00") {
$this->parseError("Received NULL character.");
}
// fprintf(STDOUT, "Writing '%s'", $tok);
$this->buffer($tok);
$this->scanner->next();
return TRUE;
}
protected function eof() {
if ($this->scanner->current() === FALSE) {
//fprintf(STDOUT, "EOF");
$this->flushBuffer();
$this->events->eof();
$this->carryOn = FALSE;
return TRUE;
}
return FALSE;
}
/**
* Handle character references (aka entities).
*
* HTML5 8.2.4.2
*
* @param boolean $inAttribute
* Set to TRUE if the text is inside of an attribute value.
* FALSE otherwise.
*/
protected function characterReference($inAttribute = FALSE) {
// If it fails this, it's definitely not an entity.
if ($this->scanner->current() != '&') {
return FALSE;
}
// Next char after &.
$tok = $this->scanner->next();
$entity = '';
$start = $this->scanner->position();
// Whitespace: Ignore
switch ($tok) {
case NULL:
case "\t":
case "\n":
case "\f":
case ' ':
case '&':
case '<':
// Don't consume; just return. Spec says return nothing, but I
// think we have to append '&' to the string.
$this->buffer('&');
return FALSE;
case '#':
// Consume and read a number
$tok = $this->scanner->next();
// Hexidecimal encoding.
// X[0-9a-fA-F]+;
// x[0-9a-fA-F]+;
if ($tok == 'x' || $tok == 'X') {
$tok = $this->scanner->next(); // Consume x
$hex = $this->scanner->getHex();
if (empty($hex)) {
//throw new ParseError("Expected &#xHEX;, got &#x" . $tok);
$this->parseError("Expected &#xHEX;, got &#x%s", $tok);
return FALSE;
}
$entity = CharacterReference::lookupHex($hex);
}
// Decimal encoding.
// [0-9]+;
else {
$numeric = $this->scanner->getNumeric();
if (empty($numeric)) {
//throw ParseError("Expected &#DIGITS;, got $#" . $tok);
$this->parseError("Expected &#DIGITS;, got $#%s", $tok);
return FALSE;
}
$entity = CharacterReference::lookupDecimal($numeric);
}
break;
default:
// Attempt to consume a string up to a ';'.
// [a-zA-Z0-9]+;
$cname = $this->scanner->getAsciiAlpha();
$entity = CharacterReference::lookupName($cname);
if ($entity == NULL) {
$this->parseError("No match in entity table for '%s'", $entity);
}
}
// The scanner has advanced the cursor for us.
$tok = $this->scanner->current();
// We have an entity. We're done here.
if ($tok == ';') {
$this->buffer($entity);
$this->scanner->next();
return TRUE;
}
// If in an attribute, then failing to match ; means unconsume the
// entire string. Otherwise, failure to match is an error.
if ($inAttribute) {
$this->scanner->unconsume($this->scanner->position() - $start);
$this->buffer('&');
return FALSE;
}
//throw new ParseError("Expected &ENTITY;, got &ENTITY (no trailing ;) " . $tok);
$this->parseError("Expected &ENTITY;, got &ENTITY%s (no trailing ;) ", $tok);
}
/**
* 8.2.4.8
*/
protected function tagOpen() {
if ($this->scanner->current() != '<') {
return FALSE;
}
$this->scanner->next();
return $this->markupDeclaration() ||
$this->endTag() ||
$this->tagName() ||
$this->processingInstruction() ||
// This always returns false.
$this->parseError("Illegal tag opening") ||
$this->characterData();
}
protected function markupDeclaration() {
if ($this->scanner->current() != '!') {
return FALSE;
}
$tok = $this->scanner->next();
// Comment:
if ($tok == '-' && $this->scanner->peek() == '-') {
}
// FINISH
return TRUE;
}
protected function rcdata() {
// Ampersand
// <
// Null
// EOF
// Character
}
protected function rawtext() {
// < is a literal
// NULL is an error
// EOF
// Character data
}
protected function scriptData() {
// < is a literal
// NULL is an error
// EOF
// Character data
}
/**
* 8.2.4.7
*/
protected function plaintext() {
// NULL -> parse error
// EOF -> eof
// -> Character data
}
/**
* Consume an end tag.
* 8.2.4.9
*/
protected function endTag() {
if ($this->scanner->current() != '/') {
return FALSE;
}
$tok = $this->scanner->next();
// a-zA-Z -> tagname
// > -> parse error
// EOF -> parse error
// -> parse error
if (!ctype_alpha($tok)) {
$this->parseError("Expected tag name, got '%s'", $tok);
if ($tok == "\0" || $tok === FALSE) {
return FALSE;
}
return $this->bogusComment('</');
}
$name = $this->scanner->charsUntil("\n\f \t>");
// Trash whitespace.
$this->scanner->whitespace();
if ($this->scanner->current() != '>') {
$this->parseError("Expected >, got '%s'", $this->scanner->current());
// We just trash stuff until we get to the next tag close.
$this->scanner->charsUntil('>');
}
$this->events->endTag($name);
$this->scanner->next();
return TRUE;
}
/**
* 8.2.4.10
*/
protected function tagName() {
$name = $this->scanner->current();
$tok = $this->scanner->next();
switch ($tok) {
case "\n":
case "\t":
case "\f":
case ' ':
return $this->beforeAttribute();
case '/':
return $this->selfClosingTag();
case '>':
}
return FALSE;
// tab, lf, ff, space -> before attr name
// / -> self-closing tag
// > -> current tag is done, data-state
// NULL parse error
// EOF -> parse error
// -> append to tagname
}
/**
* 8.2.4.11
*/
protected function rcdataLessThan() {
// / -> empty the tmp buffer and go to end-tag
// ->rcdata
}
/**
* 8.2.4.12
*/
protected function rcdataEndTag() {
// A-Za-z: append to tagname
// -> rcdata state
}
/**
* 8.2.4.13
*/
protected function rcdataEndTagName() {
// tab, lf, ff, space -> before attribute or treat as anything
// / -> self-closing tag
// > -> end tag, back to data
// A-Za-z -> append to tagname
// -> rcdata state
}
/**
* 8.2.4.14
*/
protected function rawtextLessThan() {
// / -> rawtext endtag state
// -> rawtext
}
/**
* 8.2.4.15
*/
protected function rawtextEndTagOpen() {
// A-Za-z -> rawtext
// ->rawtext
}
protected function rawtextEndTagName() {
// tab, lf, ff, space -> before attr name
//
}
protected function scriptLessThan(){
}
protected function scriptEndTagOpen() {
}
protected function scriptEndTagName() {
}
protected function scriptEscapeStart() {
}
protected function scriptEscapeStartDash() {
}
protected function scriptEscaped() {
}
protected function scriptEscapedDash() {
}
protected function scriptEscapedDashDash() {
}
protected function scriptEscapedLessThan() {
}
protected function scriptEscapedEndTagOpen() {
}
protected function scriptEscapedEndTagName() {
}
protected function scriptDoubleEscapeStart() {
}
protected function scriptDoubleEscaped() {
}
protected function scriptDoubleEscapedDash() {
}
protected function scriptDoubleEscapedDashDash() {
}
protected function scriptDoubleEscapedLessThan() {
}
protected function scriptDoubleEscapeEnd() {
}
protected function beforeAttributeName() {
}
protected function attributeName() {
}
protected function afterAttributeName() {
}
protected function beforeAttributeValue() {
}
protected function attributeValueDoubleQuote() {
}
protected function attributeValueSingleQuote() {
}
protected function attributeValueUnquoted() {
}
protected function characterReferenceInAttributeValue() {
}
protected function afterAttributeValueQuoted() {
}
protected function selfCloseingStartTag() {
}
/**
* Consume malformed markup as if it were a comment.
* 8.2.4.44
*
* The spec requires that the ENTIRE tag-like thing be enclosed inside of
* the comment. So this will generate comments like:
*
* <!--</+foo>-->
*
* @param string $leading
* Prepend any leading characters. This essentially
* negates the need to backtrack, but it's sort of
* a hack.
*/
protected function bogusComment($leading = '') {
// TODO: This can be done more efficiently when the
// scanner exposes a readUntil() method.
$comment = $leading;
$tok = $this->scanner->current();
do {
$comment .= $tok;
$tok = $this->scanner->next();
} while ($tok !== FALSE && $tok != '>');
$this->flushBuffer();
$this->events->comment($comment . $tok);
$this->scanner->next();
return TRUE;
}
protected function commentStart() {
}
protected function commentStartDash() {
}
protected function comment() {
$tok = $this->scanner->current();
$comment = '';
// <!-->. Emit an empty comment because 8.2.4.46 says to.
if ($tok == '>') {
// Parse error. Emit the comment token.
$this->parseError("Expected comment data, got '>'");
$this->events->comment('');
$this->scanner->next();
return TRUE;
}
// Replace NULL with the replacement char.
if ($tok == "\0") {
$tok = UTF8Utils::FFFD;
}
while (!$this->isCommentEnd()) {
$comment .= $tok;
$tok = $this->scanner->next();
}
$this->events->comment($comment);
$this->scanner->next();
return TRUE;
}
protected function isCommentEnd() {
// If it doesn't start with -, not the end.
if($this->scanner->current() != '-') {
return FALSE;
}
// EOF
if($this->scanner->Current() === FALSE) {
// Hit the end.
$this->events->parseError("Unexpected EOF in a comment.");
return TRUE;
}
// Advance one, and test for '->'
if ($this->scanner->next() == '-'
&& $this->scanner->peek() == '>') {
$this->scanner->next(); // Consume the last '>'
return TRUE;
}
// Unread '-';
$this->scanner->unconsume(1);
return FALSE;
}
protected function doctype() {
}
protected function beforeDoctype() {
}
protected function doctypeName() {
}
protected function afterDoctypeName() {
}
protected function doctypePublicKeyword() {
}
protected function beforeDoctypePublicId() {
}
protected function doctypePublicIdDoubleQuoted() {
}
protected function doctypePublicIdSingleQuoted() {
}
protected function afterDoctypePublicId() {
}
protected function betweenDoctypePublicAndSystem() {
}
protected function afterDoctypeSystemKeyword() {
}
protected function beforeDoctypeSystemIdentifier() {
}
protected function doctypeSystemIdDoubleQuoted() {
}
protected function doctypeSystemIdSingleQuoted() {
}
protected function afterDoctypeSystemId() {
}
protected function bogusDoctype() {
}
protected function cdataSection() {
}
// ================================================================
// Non-HTML5
// ================================================================
/**
* Handle a processing instruction.
*
* XML processing instructions are supposed to be ignored in HTML5,
* treated as "bogus comments". However, since we're not a user
* agent, we allow them. We consume until ?> and then issue a
* EventListener::processingInstruction() event.
*/
protected function processingInstruction() {
}
// ================================================================
// UTILITY FUNCTIONS
// ================================================================
/**
* Send a TEXT event with the contents of the text buffer.
*
* This emits an EventHandler::text() event with the current contents of the
* temporary text buffer. (The buffer is used to group as much PCDATA
* as we can instead of emitting lots and lots of TEXT events.)
*/
protected function flushBuffer() {
if (empty($this->text)) {
return;
}
$this->events->text($this->text);
$this->text = '';
}
/**
* Add text to the temporary buffer.
*
* @see flushBuffer()
*/
protected function buffer($str) {
$this->text .= $str;
}
/**
* Emit a parse error.
*
* A parse error always returns FALSE because it never consumes any
* characters.
*/
protected function parseError($msg) {
$args = func_get_args();
if (count($args) > 1) {
array_shift($args);
$msg = vsprintf($msg, $args);
}
$line = $this->scanner->currentLine();
$col = $this->scanner->columnOffset();
$this->events->parseError($msg, $line, $col);
return FALSE;
}
}
|