]>
Commit | Line | Data |
---|---|---|
42c80841 NL |
1 | <?php |
2 | ||
3 | /* | |
4 | ||
5 | Copyright 2007 Jeroen van der Meer <http://jero.net/> | |
6 | Copyright 2009 Edward Z. Yang <edwardzyang@thewritingpot.com> | |
7 | ||
8 | Permission is hereby granted, free of charge, to any person obtaining a | |
9 | copy of this software and associated documentation files (the | |
10 | "Software"), to deal in the Software without restriction, including | |
11 | without limitation the rights to use, copy, modify, merge, publish, | |
12 | distribute, sublicense, and/or sell copies of the Software, and to | |
13 | permit persons to whom the Software is furnished to do so, subject to | |
14 | the following conditions: | |
15 | ||
16 | The above copyright notice and this permission notice shall be included | |
17 | in all copies or substantial portions of the Software. | |
18 | ||
19 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS | |
20 | OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF | |
21 | MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. | |
22 | IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY | |
23 | CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, | |
24 | TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE | |
25 | SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. | |
26 | ||
27 | */ | |
28 | ||
29 | // Tags for FIX ME!!!: (in order of priority) | |
30 | // XXX - should be fixed NAO! | |
31 | // XERROR - with regards to parse errors | |
32 | // XSCRIPT - with regards to scripting mode | |
33 | // XENCODING - with regards to encoding (for reparsing tests) | |
34 | // XDOM - DOM specific code (tagName is explicitly not marked). | |
35 | // this is not (yet) in helper functions. | |
36 | ||
37 | class HTML5_TreeBuilder { | |
38 | public $stack = array(); | |
39 | public $content_model; | |
40 | ||
41 | private $mode; | |
42 | private $original_mode; | |
43 | private $secondary_mode; | |
44 | private $dom; | |
45 | // Whether or not normal insertion of nodes should actually foster | |
46 | // parent (used in one case in spec) | |
47 | private $foster_parent = false; | |
48 | private $a_formatting = array(); | |
49 | ||
50 | private $head_pointer = null; | |
51 | private $form_pointer = null; | |
52 | ||
53 | private $flag_frameset_ok = true; | |
54 | private $flag_force_quirks = false; | |
55 | private $ignored = false; | |
56 | private $quirks_mode = null; | |
57 | // this gets to 2 when we want to ignore the next lf character, and | |
58 | // is decrement at the beginning of each processed token (this way, | |
59 | // code can check for (bool)$ignore_lf_token, but it phases out | |
60 | // appropriately) | |
61 | private $ignore_lf_token = 0; | |
62 | private $fragment = false; | |
63 | private $root; | |
64 | ||
65 | private $scoping = array('applet','button','caption','html','marquee','object','table','td','th', 'svg:foreignObject'); | |
66 | private $formatting = array('a','b','big','code','em','font','i','nobr','s','small','strike','strong','tt','u'); | |
67 | // dl and ds are speculative | |
68 | private $special = array('address','area','article','aside','base','basefont','bgsound', | |
69 | 'blockquote','body','br','center','col','colgroup','command','dc','dd','details','dir','div','dl','ds', | |
70 | 'dt','embed','fieldset','figure','footer','form','frame','frameset','h1','h2','h3','h4','h5', | |
71 | 'h6','head','header','hgroup','hr','iframe','img','input','isindex','li','link', | |
72 | 'listing','menu','meta','nav','noembed','noframes','noscript','ol', | |
73 | 'p','param','plaintext','pre','script','select','spacer','style', | |
74 | 'tbody','textarea','tfoot','thead','title','tr','ul','wbr'); | |
75 | ||
76 | private $pendingTableCharacters; | |
77 | private $pendingTableCharactersDirty; | |
78 | ||
79 | // Tree construction modes | |
80 | const INITIAL = 0; | |
81 | const BEFORE_HTML = 1; | |
82 | const BEFORE_HEAD = 2; | |
83 | const IN_HEAD = 3; | |
84 | const IN_HEAD_NOSCRIPT = 4; | |
85 | const AFTER_HEAD = 5; | |
86 | const IN_BODY = 6; | |
87 | const IN_CDATA_RCDATA = 7; | |
88 | const IN_TABLE = 8; | |
89 | const IN_TABLE_TEXT = 9; | |
90 | const IN_CAPTION = 10; | |
91 | const IN_COLUMN_GROUP = 11; | |
92 | const IN_TABLE_BODY = 12; | |
93 | const IN_ROW = 13; | |
94 | const IN_CELL = 14; | |
95 | const IN_SELECT = 15; | |
96 | const IN_SELECT_IN_TABLE= 16; | |
97 | const IN_FOREIGN_CONTENT= 17; | |
98 | const AFTER_BODY = 18; | |
99 | const IN_FRAMESET = 19; | |
100 | const AFTER_FRAMESET = 20; | |
101 | const AFTER_AFTER_BODY = 21; | |
102 | const AFTER_AFTER_FRAMESET = 22; | |
103 | ||
104 | /** | |
105 | * Converts a magic number to a readable name. Use for debugging. | |
106 | */ | |
107 | private function strConst($number) { | |
108 | static $lookup; | |
109 | if (!$lookup) { | |
110 | $lookup = array(); | |
111 | $r = new ReflectionClass('HTML5_TreeBuilder'); | |
112 | $consts = $r->getConstants(); | |
113 | foreach ($consts as $const => $num) { | |
114 | if (!is_int($num)) continue; | |
115 | $lookup[$num] = $const; | |
116 | } | |
117 | } | |
118 | return $lookup[$number]; | |
119 | } | |
120 | ||
121 | // The different types of elements. | |
122 | const SPECIAL = 100; | |
123 | const SCOPING = 101; | |
124 | const FORMATTING = 102; | |
125 | const PHRASING = 103; | |
126 | ||
127 | // Quirks modes in $quirks_mode | |
128 | const NO_QUIRKS = 200; | |
129 | const QUIRKS_MODE = 201; | |
130 | const LIMITED_QUIRKS_MODE = 202; | |
131 | ||
132 | // Marker to be placed in $a_formatting | |
133 | const MARKER = 300; | |
134 | ||
135 | // Namespaces for foreign content | |
136 | const NS_HTML = null; // to prevent DOM from requiring NS on everything | |
3ec62cf9 | 137 | const NS_XHTML = 'http://www.w3.org/1999/xhtml'; |
42c80841 NL |
138 | const NS_MATHML = 'http://www.w3.org/1998/Math/MathML'; |
139 | const NS_SVG = 'http://www.w3.org/2000/svg'; | |
140 | const NS_XLINK = 'http://www.w3.org/1999/xlink'; | |
141 | const NS_XML = 'http://www.w3.org/XML/1998/namespace'; | |
142 | const NS_XMLNS = 'http://www.w3.org/2000/xmlns/'; | |
143 | ||
144 | // Different types of scopes to test for elements | |
145 | const SCOPE = 0; | |
146 | const SCOPE_LISTITEM = 1; | |
147 | const SCOPE_TABLE = 2; | |
148 | ||
149 | public function __construct() { | |
150 | $this->mode = self::INITIAL; | |
151 | $this->dom = new DOMDocument; | |
152 | ||
153 | $this->dom->encoding = 'UTF-8'; | |
154 | $this->dom->preserveWhiteSpace = true; | |
155 | $this->dom->substituteEntities = true; | |
156 | $this->dom->strictErrorChecking = false; | |
157 | } | |
158 | ||
159 | // Process tag tokens | |
160 | public function emitToken($token, $mode = null) { | |
161 | // XXX: ignore parse errors... why are we emitting them, again? | |
162 | if ($token['type'] === HTML5_Tokenizer::PARSEERROR) return; | |
163 | if ($mode === null) $mode = $this->mode; | |
164 | ||
165 | /* | |
166 | $backtrace = debug_backtrace(); | |
167 | if ($backtrace[1]['class'] !== 'HTML5_TreeBuilder') echo "--\n"; | |
168 | echo $this->strConst($mode); | |
169 | if ($this->original_mode) echo " (originally ".$this->strConst($this->original_mode).")"; | |
170 | echo "\n "; | |
171 | token_dump($token); | |
172 | $this->printStack(); | |
173 | $this->printActiveFormattingElements(); | |
174 | if ($this->foster_parent) echo " -> this is a foster parent mode\n"; | |
175 | if ($this->flag_frameset_ok) echo " -> frameset ok\n"; | |
176 | */ | |
177 | ||
178 | if ($this->ignore_lf_token) $this->ignore_lf_token--; | |
179 | $this->ignored = false; | |
180 | // indenting is a little wonky, this can be changed later on | |
181 | switch ($mode) { | |
182 | ||
183 | case self::INITIAL: | |
184 | ||
185 | /* A character token that is one of U+0009 CHARACTER TABULATION, | |
186 | * U+000A LINE FEED (LF), U+000C FORM FEED (FF), or U+0020 SPACE */ | |
187 | if ($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
188 | /* Ignore the token. */ | |
189 | $this->ignored = true; | |
190 | } elseif ($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
191 | if ( | |
192 | $token['name'] !== 'html' || !empty($token['public']) || | |
193 | !empty($token['system']) || $token !== 'about:legacy-compat' | |
194 | ) { | |
195 | /* If the DOCTYPE token's name is not a case-sensitive match | |
196 | * for the string "html", or if the token's public identifier | |
197 | * is not missing, or if the token's system identifier is | |
198 | * neither missing nor a case-sensitive match for the string | |
199 | * "about:legacy-compat", then there is a parse error (this | |
200 | * is the DOCTYPE parse error). */ | |
201 | // DOCTYPE parse error | |
202 | } | |
203 | /* Append a DocumentType node to the Document node, with the name | |
204 | * attribute set to the name given in the DOCTYPE token, or the | |
205 | * empty string if the name was missing; the publicId attribute | |
206 | * set to the public identifier given in the DOCTYPE token, or | |
207 | * the empty string if the public identifier was missing; the | |
208 | * systemId attribute set to the system identifier given in the | |
209 | * DOCTYPE token, or the empty string if the system identifier | |
210 | * was missing; and the other attributes specific to | |
211 | * DocumentType objects set to null and empty lists as | |
212 | * appropriate. Associate the DocumentType node with the | |
213 | * Document object so that it is returned as the value of the | |
214 | * doctype attribute of the Document object. */ | |
215 | if (!isset($token['public'])) $token['public'] = null; | |
216 | if (!isset($token['system'])) $token['system'] = null; | |
217 | // XDOM | |
218 | // Yes this is hacky. I'm kind of annoyed that I can't appendChild | |
219 | // a doctype to DOMDocument. Maybe I haven't chanted the right | |
220 | // syllables. | |
221 | $impl = new DOMImplementation(); | |
222 | // This call can fail for particularly pathological cases (namely, | |
223 | // the qualifiedName parameter ($token['name']) could be missing. | |
224 | if ($token['name']) { | |
225 | $doctype = $impl->createDocumentType($token['name'], $token['public'], $token['system']); | |
226 | $this->dom->appendChild($doctype); | |
227 | } else { | |
228 | // It looks like libxml's not actually *able* to express this case. | |
229 | // So... don't. | |
230 | $this->dom->emptyDoctype = true; | |
231 | } | |
232 | $public = is_null($token['public']) ? false : strtolower($token['public']); | |
233 | $system = is_null($token['system']) ? false : strtolower($token['system']); | |
234 | $publicStartsWithForQuirks = array( | |
235 | "+//silmaril//dtd html pro v0r11 19970101//", | |
236 | "-//advasoft ltd//dtd html 3.0 aswedit + extensions//", | |
237 | "-//as//dtd html 3.0 aswedit + extensions//", | |
238 | "-//ietf//dtd html 2.0 level 1//", | |
239 | "-//ietf//dtd html 2.0 level 2//", | |
240 | "-//ietf//dtd html 2.0 strict level 1//", | |
241 | "-//ietf//dtd html 2.0 strict level 2//", | |
242 | "-//ietf//dtd html 2.0 strict//", | |
243 | "-//ietf//dtd html 2.0//", | |
244 | "-//ietf//dtd html 2.1e//", | |
245 | "-//ietf//dtd html 3.0//", | |
246 | "-//ietf//dtd html 3.2 final//", | |
247 | "-//ietf//dtd html 3.2//", | |
248 | "-//ietf//dtd html 3//", | |
249 | "-//ietf//dtd html level 0//", | |
250 | "-//ietf//dtd html level 1//", | |
251 | "-//ietf//dtd html level 2//", | |
252 | "-//ietf//dtd html level 3//", | |
253 | "-//ietf//dtd html strict level 0//", | |
254 | "-//ietf//dtd html strict level 1//", | |
255 | "-//ietf//dtd html strict level 2//", | |
256 | "-//ietf//dtd html strict level 3//", | |
257 | "-//ietf//dtd html strict//", | |
258 | "-//ietf//dtd html//", | |
259 | "-//metrius//dtd metrius presentational//", | |
260 | "-//microsoft//dtd internet explorer 2.0 html strict//", | |
261 | "-//microsoft//dtd internet explorer 2.0 html//", | |
262 | "-//microsoft//dtd internet explorer 2.0 tables//", | |
263 | "-//microsoft//dtd internet explorer 3.0 html strict//", | |
264 | "-//microsoft//dtd internet explorer 3.0 html//", | |
265 | "-//microsoft//dtd internet explorer 3.0 tables//", | |
266 | "-//netscape comm. corp.//dtd html//", | |
267 | "-//netscape comm. corp.//dtd strict html//", | |
268 | "-//o'reilly and associates//dtd html 2.0//", | |
269 | "-//o'reilly and associates//dtd html extended 1.0//", | |
270 | "-//o'reilly and associates//dtd html extended relaxed 1.0//", | |
271 | "-//spyglass//dtd html 2.0 extended//", | |
272 | "-//sq//dtd html 2.0 hotmetal + extensions//", | |
273 | "-//sun microsystems corp.//dtd hotjava html//", | |
274 | "-//sun microsystems corp.//dtd hotjava strict html//", | |
275 | "-//w3c//dtd html 3 1995-03-24//", | |
276 | "-//w3c//dtd html 3.2 draft//", | |
277 | "-//w3c//dtd html 3.2 final//", | |
278 | "-//w3c//dtd html 3.2//", | |
279 | "-//w3c//dtd html 3.2s draft//", | |
280 | "-//w3c//dtd html 4.0 frameset//", | |
281 | "-//w3c//dtd html 4.0 transitional//", | |
282 | "-//w3c//dtd html experimental 19960712//", | |
283 | "-//w3c//dtd html experimental 970421//", | |
284 | "-//w3c//dtd w3 html//", | |
285 | "-//w3o//dtd w3 html 3.0//", | |
286 | "-//webtechs//dtd mozilla html 2.0//", | |
287 | "-//webtechs//dtd mozilla html//", | |
288 | ); | |
289 | $publicSetToForQuirks = array( | |
290 | "-//w3o//dtd w3 html strict 3.0//", | |
291 | "-/w3c/dtd html 4.0 transitional/en", | |
292 | "html", | |
293 | ); | |
294 | $publicStartsWithAndSystemForQuirks = array( | |
295 | "-//w3c//dtd html 4.01 frameset//", | |
296 | "-//w3c//dtd html 4.01 transitional//", | |
297 | ); | |
298 | $publicStartsWithForLimitedQuirks = array( | |
299 | "-//w3c//dtd xhtml 1.0 frameset//", | |
300 | "-//w3c//dtd xhtml 1.0 transitional//", | |
301 | ); | |
302 | $publicStartsWithAndSystemForLimitedQuirks = array( | |
303 | "-//w3c//dtd html 4.01 frameset//", | |
304 | "-//w3c//dtd html 4.01 transitional//", | |
305 | ); | |
306 | // first, do easy checks | |
307 | if ( | |
308 | !empty($token['force-quirks']) || | |
309 | strtolower($token['name']) !== 'html' | |
310 | ) { | |
311 | $this->quirks_mode = self::QUIRKS_MODE; | |
312 | } else { | |
313 | do { | |
314 | if ($system) { | |
315 | foreach ($publicStartsWithAndSystemForQuirks as $x) { | |
316 | if (strncmp($public, $x, strlen($x)) === 0) { | |
317 | $this->quirks_mode = self::QUIRKS_MODE; | |
318 | break; | |
319 | } | |
320 | } | |
321 | if (!is_null($this->quirks_mode)) break; | |
322 | foreach ($publicStartsWithAndSystemForLimitedQuirks as $x) { | |
323 | if (strncmp($public, $x, strlen($x)) === 0) { | |
324 | $this->quirks_mode = self::LIMITED_QUIRKS_MODE; | |
325 | break; | |
326 | } | |
327 | } | |
328 | if (!is_null($this->quirks_mode)) break; | |
329 | } | |
330 | foreach ($publicSetToForQuirks as $x) { | |
331 | if ($public === $x) { | |
332 | $this->quirks_mode = self::QUIRKS_MODE; | |
333 | break; | |
334 | } | |
335 | } | |
336 | if (!is_null($this->quirks_mode)) break; | |
337 | foreach ($publicStartsWithForLimitedQuirks as $x) { | |
338 | if (strncmp($public, $x, strlen($x)) === 0) { | |
339 | $this->quirks_mode = self::LIMITED_QUIRKS_MODE; | |
340 | } | |
341 | } | |
342 | if (!is_null($this->quirks_mode)) break; | |
343 | if ($system === "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd") { | |
344 | $this->quirks_mode = self::QUIRKS_MODE; | |
345 | break; | |
346 | } | |
347 | foreach ($publicStartsWithForQuirks as $x) { | |
348 | if (strncmp($public, $x, strlen($x)) === 0) { | |
349 | $this->quirks_mode = self::QUIRKS_MODE; | |
350 | break; | |
351 | } | |
352 | } | |
353 | if (is_null($this->quirks_mode)) { | |
354 | $this->quirks_mode = self::NO_QUIRKS; | |
355 | } | |
356 | } while (false); | |
357 | } | |
358 | $this->mode = self::BEFORE_HTML; | |
359 | } else { | |
360 | // parse error | |
361 | /* Switch the insertion mode to "before html", then reprocess the | |
362 | * current token. */ | |
363 | $this->mode = self::BEFORE_HTML; | |
364 | $this->quirks_mode = self::QUIRKS_MODE; | |
365 | $this->emitToken($token); | |
366 | } | |
367 | break; | |
368 | ||
369 | case self::BEFORE_HTML: | |
370 | ||
371 | /* A DOCTYPE token */ | |
372 | if($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
373 | // Parse error. Ignore the token. | |
374 | $this->ignored = true; | |
375 | ||
376 | /* A comment token */ | |
377 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
378 | /* Append a Comment node to the Document object with the data | |
379 | attribute set to the data given in the comment token. */ | |
380 | // XDOM | |
381 | $comment = $this->dom->createComment($token['data']); | |
382 | $this->dom->appendChild($comment); | |
383 | ||
384 | /* A character token that is one of one of U+0009 CHARACTER TABULATION, | |
385 | U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), | |
386 | or U+0020 SPACE */ | |
387 | } elseif($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
388 | /* Ignore the token. */ | |
389 | $this->ignored = true; | |
390 | ||
391 | /* A start tag whose tag name is "html" */ | |
392 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] == 'html') { | |
393 | /* Create an element for the token in the HTML namespace. Append it | |
394 | * to the Document object. Put this element in the stack of open | |
395 | * elements. */ | |
396 | // XDOM | |
397 | $html = $this->insertElement($token, false); | |
398 | $this->dom->appendChild($html); | |
399 | $this->stack[] = $html; | |
400 | ||
401 | $this->mode = self::BEFORE_HEAD; | |
402 | ||
403 | } else { | |
404 | /* Create an html element. Append it to the Document object. Put | |
405 | * this element in the stack of open elements. */ | |
406 | // XDOM | |
407 | $html = $this->dom->createElementNS(self::NS_HTML, 'html'); | |
408 | $this->dom->appendChild($html); | |
409 | $this->stack[] = $html; | |
410 | ||
411 | /* Switch the insertion mode to "before head", then reprocess the | |
412 | * current token. */ | |
413 | $this->mode = self::BEFORE_HEAD; | |
414 | $this->emitToken($token); | |
415 | } | |
416 | break; | |
417 | ||
418 | case self::BEFORE_HEAD: | |
419 | ||
420 | /* A character token that is one of one of U+0009 CHARACTER TABULATION, | |
421 | U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), | |
422 | or U+0020 SPACE */ | |
423 | if($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
424 | /* Ignore the token. */ | |
425 | $this->ignored = true; | |
426 | ||
427 | /* A comment token */ | |
428 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
429 | /* Append a Comment node to the current node with the data attribute | |
430 | set to the data given in the comment token. */ | |
431 | $this->insertComment($token['data']); | |
432 | ||
433 | /* A DOCTYPE token */ | |
434 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
435 | /* Parse error. Ignore the token */ | |
436 | $this->ignored = true; | |
437 | // parse error | |
438 | ||
439 | /* A start tag token with the tag name "html" */ | |
440 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html') { | |
441 | /* Process the token using the rules for the "in body" | |
442 | * insertion mode. */ | |
443 | $this->processWithRulesFor($token, self::IN_BODY); | |
444 | ||
445 | /* A start tag token with the tag name "head" */ | |
446 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'head') { | |
447 | /* Insert an HTML element for the token. */ | |
448 | $element = $this->insertElement($token); | |
449 | ||
450 | /* Set the head element pointer to this new element node. */ | |
451 | $this->head_pointer = $element; | |
452 | ||
453 | /* Change the insertion mode to "in head". */ | |
454 | $this->mode = self::IN_HEAD; | |
455 | ||
456 | /* An end tag whose tag name is one of: "head", "body", "html", "br" */ | |
457 | } elseif( | |
458 | $token['type'] === HTML5_Tokenizer::ENDTAG && ( | |
459 | $token['name'] === 'head' || $token['name'] === 'body' || | |
460 | $token['name'] === 'html' || $token['name'] === 'br' | |
461 | )) { | |
462 | /* Act as if a start tag token with the tag name "head" and no | |
463 | * attributes had been seen, then reprocess the current token. */ | |
464 | $this->emitToken(array( | |
465 | 'name' => 'head', | |
466 | 'type' => HTML5_Tokenizer::STARTTAG, | |
467 | 'attr' => array() | |
468 | )); | |
469 | $this->emitToken($token); | |
470 | ||
471 | /* Any other end tag */ | |
472 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG) { | |
473 | /* Parse error. Ignore the token. */ | |
474 | $this->ignored = true; | |
475 | ||
476 | } else { | |
477 | /* Act as if a start tag token with the tag name "head" and no | |
478 | * attributes had been seen, then reprocess the current token. | |
479 | * Note: This will result in an empty head element being | |
480 | * generated, with the current token being reprocessed in the | |
481 | * "after head" insertion mode. */ | |
482 | $this->emitToken(array( | |
483 | 'name' => 'head', | |
484 | 'type' => HTML5_Tokenizer::STARTTAG, | |
485 | 'attr' => array() | |
486 | )); | |
487 | $this->emitToken($token); | |
488 | } | |
489 | break; | |
490 | ||
491 | case self::IN_HEAD: | |
492 | ||
493 | /* A character token that is one of one of U+0009 CHARACTER TABULATION, | |
494 | U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), | |
495 | or U+0020 SPACE. */ | |
496 | if($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
497 | /* Insert the character into the current node. */ | |
498 | $this->insertText($token['data']); | |
499 | ||
500 | /* A comment token */ | |
501 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
502 | /* Append a Comment node to the current node with the data attribute | |
503 | set to the data given in the comment token. */ | |
504 | $this->insertComment($token['data']); | |
505 | ||
506 | /* A DOCTYPE token */ | |
507 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
508 | /* Parse error. Ignore the token. */ | |
509 | $this->ignored = true; | |
510 | // parse error | |
511 | ||
512 | /* A start tag whose tag name is "html" */ | |
513 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
514 | $token['name'] === 'html') { | |
515 | $this->processWithRulesFor($token, self::IN_BODY); | |
516 | ||
517 | /* A start tag whose tag name is one of: "base", "command", "link" */ | |
518 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
519 | ($token['name'] === 'base' || $token['name'] === 'command' || | |
520 | $token['name'] === 'link')) { | |
521 | /* Insert an HTML element for the token. Immediately pop the | |
522 | * current node off the stack of open elements. */ | |
523 | $this->insertElement($token); | |
524 | array_pop($this->stack); | |
525 | ||
526 | // YYY: Acknowledge the token's self-closing flag, if it is set. | |
527 | ||
528 | /* A start tag whose tag name is "meta" */ | |
529 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'meta') { | |
530 | /* Insert an HTML element for the token. Immediately pop the | |
531 | * current node off the stack of open elements. */ | |
532 | $this->insertElement($token); | |
533 | array_pop($this->stack); | |
534 | ||
535 | // XERROR: Acknowledge the token's self-closing flag, if it is set. | |
536 | ||
537 | // XENCODING: If the element has a charset attribute, and its value is a | |
538 | // supported encoding, and the confidence is currently tentative, | |
539 | // then change the encoding to the encoding given by the value of | |
540 | // the charset attribute. | |
541 | // | |
542 | // Otherwise, if the element has a content attribute, and applying | |
543 | // the algorithm for extracting an encoding from a Content-Type to | |
544 | // its value returns a supported encoding encoding, and the | |
545 | // confidence is currently tentative, then change the encoding to | |
546 | // the encoding encoding. | |
547 | ||
548 | /* A start tag with the tag name "title" */ | |
549 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'title') { | |
550 | $this->insertRCDATAElement($token); | |
551 | ||
552 | /* A start tag whose tag name is "noscript", if the scripting flag is enabled, or | |
553 | * A start tag whose tag name is one of: "noframes", "style" */ | |
554 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
555 | ($token['name'] === 'noscript' || $token['name'] === 'noframes' || $token['name'] === 'style')) { | |
556 | // XSCRIPT: Scripting flag not respected | |
557 | $this->insertCDATAElement($token); | |
558 | ||
559 | // XSCRIPT: Scripting flag disable not implemented | |
560 | ||
561 | /* A start tag with the tag name "script" */ | |
562 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'script') { | |
563 | /* 1. Create an element for the token in the HTML namespace. */ | |
564 | $node = $this->insertElement($token, false); | |
565 | ||
566 | /* 2. Mark the element as being "parser-inserted" */ | |
567 | // Uhhh... XSCRIPT | |
568 | ||
569 | /* 3. If the parser was originally created for the HTML | |
570 | * fragment parsing algorithm, then mark the script element as | |
571 | * "already executed". (fragment case) */ | |
572 | // ditto... XSCRIPT | |
573 | ||
574 | /* 4. Append the new element to the current node and push it onto | |
575 | * the stack of open elements. */ | |
576 | end($this->stack)->appendChild($node); | |
577 | $this->stack[] = $node; | |
578 | // I guess we could squash these together | |
579 | ||
580 | /* 6. Let the original insertion mode be the current insertion mode. */ | |
581 | $this->original_mode = $this->mode; | |
582 | /* 7. Switch the insertion mode to "in CDATA/RCDATA" */ | |
583 | $this->mode = self::IN_CDATA_RCDATA; | |
584 | /* 5. Switch the tokeniser's content model flag to the CDATA state. */ | |
585 | $this->content_model = HTML5_Tokenizer::CDATA; | |
586 | ||
587 | /* An end tag with the tag name "head" */ | |
588 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'head') { | |
589 | /* Pop the current node (which will be the head element) off the stack of open elements. */ | |
590 | array_pop($this->stack); | |
591 | ||
592 | /* Change the insertion mode to "after head". */ | |
593 | $this->mode = self::AFTER_HEAD; | |
594 | ||
595 | // Slight logic inversion here to minimize duplication | |
596 | /* A start tag with the tag name "head". */ | |
597 | /* An end tag whose tag name is not one of: "body", "html", "br" */ | |
598 | } elseif(($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'head') || | |
599 | ($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] !== 'html' && | |
600 | $token['name'] !== 'body' && $token['name'] !== 'br')) { | |
601 | // Parse error. Ignore the token. | |
602 | $this->ignored = true; | |
603 | ||
604 | /* Anything else */ | |
605 | } else { | |
606 | /* Act as if an end tag token with the tag name "head" had been | |
607 | * seen, and reprocess the current token. */ | |
608 | $this->emitToken(array( | |
609 | 'name' => 'head', | |
610 | 'type' => HTML5_Tokenizer::ENDTAG | |
611 | )); | |
612 | ||
613 | /* Then, reprocess the current token. */ | |
614 | $this->emitToken($token); | |
615 | } | |
616 | break; | |
617 | ||
618 | case self::IN_HEAD_NOSCRIPT: | |
619 | if ($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
620 | // parse error | |
621 | } elseif ($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html') { | |
622 | $this->processWithRulesFor($token, self::IN_BODY); | |
623 | } elseif ($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'noscript') { | |
624 | /* Pop the current node (which will be a noscript element) from the | |
625 | * stack of open elements; the new current node will be a head | |
626 | * element. */ | |
627 | array_pop($this->stack); | |
628 | $this->mode = self::IN_HEAD; | |
629 | } elseif ( | |
630 | ($token['type'] === HTML5_Tokenizer::SPACECHARACTER) || | |
631 | ($token['type'] === HTML5_Tokenizer::COMMENT) || | |
632 | ($token['type'] === HTML5_Tokenizer::STARTTAG && ( | |
633 | $token['name'] === 'link' || $token['name'] === 'meta' || | |
634 | $token['name'] === 'noframes' || $token['name'] === 'style'))) { | |
635 | $this->processWithRulesFor($token, self::IN_HEAD); | |
636 | // inverted logic | |
637 | } elseif ( | |
638 | ($token['type'] === HTML5_Tokenizer::STARTTAG && ( | |
639 | $token['name'] === 'head' || $token['name'] === 'noscript')) || | |
640 | ($token['type'] === HTML5_Tokenizer::ENDTAG && | |
641 | $token['name'] !== 'br')) { | |
642 | // parse error | |
643 | } else { | |
644 | // parse error | |
645 | $this->emitToken(array( | |
646 | 'type' => HTML5_Tokenizer::ENDTAG, | |
647 | 'name' => 'noscript', | |
648 | )); | |
649 | $this->emitToken($token); | |
650 | } | |
651 | break; | |
652 | ||
653 | case self::AFTER_HEAD: | |
654 | /* Handle the token as follows: */ | |
655 | ||
656 | /* A character token that is one of one of U+0009 CHARACTER TABULATION, | |
657 | U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), | |
658 | or U+0020 SPACE */ | |
659 | if($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
660 | /* Append the character to the current node. */ | |
661 | $this->insertText($token['data']); | |
662 | ||
663 | /* A comment token */ | |
664 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
665 | /* Append a Comment node to the current node with the data attribute | |
666 | set to the data given in the comment token. */ | |
667 | $this->insertComment($token['data']); | |
668 | ||
669 | } elseif ($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
670 | // parse error | |
671 | ||
672 | } elseif ($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html') { | |
673 | $this->processWithRulesFor($token, self::IN_BODY); | |
674 | ||
675 | /* A start tag token with the tag name "body" */ | |
676 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'body') { | |
677 | $this->insertElement($token); | |
678 | ||
679 | /* Set the frameset-ok flag to "not ok". */ | |
680 | $this->flag_frameset_ok = false; | |
681 | ||
682 | /* Change the insertion mode to "in body". */ | |
683 | $this->mode = self::IN_BODY; | |
684 | ||
685 | /* A start tag token with the tag name "frameset" */ | |
686 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'frameset') { | |
687 | /* Insert a frameset element for the token. */ | |
688 | $this->insertElement($token); | |
689 | ||
690 | /* Change the insertion mode to "in frameset". */ | |
691 | $this->mode = self::IN_FRAMESET; | |
692 | ||
693 | /* A start tag token whose tag name is one of: "base", "link", "meta", | |
694 | "script", "style", "title" */ | |
695 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && in_array($token['name'], | |
696 | array('base', 'link', 'meta', 'noframes', 'script', 'style', 'title'))) { | |
697 | // parse error | |
698 | /* Push the node pointed to by the head element pointer onto the | |
699 | * stack of open elements. */ | |
700 | $this->stack[] = $this->head_pointer; | |
701 | $this->processWithRulesFor($token, self::IN_HEAD); | |
702 | array_splice($this->stack, array_search($this->head_pointer, $this->stack, true), 1); | |
703 | ||
704 | // inversion of specification | |
705 | } elseif( | |
706 | ($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'head') || | |
707 | ($token['type'] === HTML5_Tokenizer::ENDTAG && | |
708 | $token['name'] !== 'body' && $token['name'] !== 'html' && | |
709 | $token['name'] !== 'br')) { | |
710 | // parse error | |
711 | ||
712 | /* Anything else */ | |
713 | } else { | |
714 | $this->emitToken(array( | |
715 | 'name' => 'body', | |
716 | 'type' => HTML5_Tokenizer::STARTTAG, | |
717 | 'attr' => array() | |
718 | )); | |
719 | $this->flag_frameset_ok = true; | |
720 | $this->emitToken($token); | |
721 | } | |
722 | break; | |
723 | ||
724 | case self::IN_BODY: | |
725 | /* Handle the token as follows: */ | |
726 | ||
727 | switch($token['type']) { | |
728 | /* A character token */ | |
729 | case HTML5_Tokenizer::CHARACTER: | |
730 | case HTML5_Tokenizer::SPACECHARACTER: | |
731 | /* Reconstruct the active formatting elements, if any. */ | |
732 | $this->reconstructActiveFormattingElements(); | |
733 | ||
734 | /* Append the token's character to the current node. */ | |
735 | $this->insertText($token['data']); | |
736 | ||
737 | /* If the token is not one of U+0009 CHARACTER TABULATION, | |
738 | * U+000A LINE FEED (LF), U+000C FORM FEED (FF), or U+0020 | |
739 | * SPACE, then set the frameset-ok flag to "not ok". */ | |
740 | // i.e., if any of the characters is not whitespace | |
741 | if (strlen($token['data']) !== strspn($token['data'], HTML5_Tokenizer::WHITESPACE)) { | |
742 | $this->flag_frameset_ok = false; | |
743 | } | |
744 | break; | |
745 | ||
746 | /* A comment token */ | |
747 | case HTML5_Tokenizer::COMMENT: | |
748 | /* Append a Comment node to the current node with the data | |
749 | attribute set to the data given in the comment token. */ | |
750 | $this->insertComment($token['data']); | |
751 | break; | |
752 | ||
753 | case HTML5_Tokenizer::DOCTYPE: | |
754 | // parse error | |
755 | break; | |
756 | ||
757 | case HTML5_Tokenizer::EOF: | |
758 | // parse error | |
759 | break; | |
760 | ||
761 | case HTML5_Tokenizer::STARTTAG: | |
762 | switch($token['name']) { | |
763 | case 'html': | |
764 | // parse error | |
765 | /* For each attribute on the token, check to see if the | |
766 | * attribute is already present on the top element of the | |
767 | * stack of open elements. If it is not, add the attribute | |
768 | * and its corresponding value to that element. */ | |
769 | foreach($token['attr'] as $attr) { | |
770 | if(!$this->stack[0]->hasAttribute($attr['name'])) { | |
771 | $this->stack[0]->setAttribute($attr['name'], $attr['value']); | |
772 | } | |
773 | } | |
774 | break; | |
775 | ||
776 | case 'base': case 'command': case 'link': case 'meta': case 'noframes': | |
777 | case 'script': case 'style': case 'title': | |
778 | /* Process the token as if the insertion mode had been "in | |
779 | head". */ | |
780 | $this->processWithRulesFor($token, self::IN_HEAD); | |
781 | break; | |
782 | ||
783 | /* A start tag token with the tag name "body" */ | |
784 | case 'body': | |
785 | /* Parse error. If the second element on the stack of open | |
786 | elements is not a body element, or, if the stack of open | |
787 | elements has only one node on it, then ignore the token. | |
788 | (fragment case) */ | |
789 | if(count($this->stack) === 1 || $this->stack[1]->tagName !== 'body') { | |
790 | $this->ignored = true; | |
791 | // Ignore | |
792 | ||
793 | /* Otherwise, for each attribute on the token, check to see | |
794 | if the attribute is already present on the body element (the | |
795 | second element) on the stack of open elements. If it is not, | |
796 | add the attribute and its corresponding value to that | |
797 | element. */ | |
798 | } else { | |
799 | foreach($token['attr'] as $attr) { | |
800 | if(!$this->stack[1]->hasAttribute($attr['name'])) { | |
801 | $this->stack[1]->setAttribute($attr['name'], $attr['value']); | |
802 | } | |
803 | } | |
804 | } | |
805 | break; | |
806 | ||
807 | case 'frameset': | |
808 | // parse error | |
809 | /* If the second element on the stack of open elements is | |
810 | * not a body element, or, if the stack of open elements | |
811 | * has only one node on it, then ignore the token. | |
812 | * (fragment case) */ | |
813 | if(count($this->stack) === 1 || $this->stack[1]->tagName !== 'body') { | |
814 | $this->ignored = true; | |
815 | // Ignore | |
816 | } elseif (!$this->flag_frameset_ok) { | |
817 | $this->ignored = true; | |
818 | // Ignore | |
819 | } else { | |
820 | /* 1. Remove the second element on the stack of open | |
821 | * elements from its parent node, if it has one. */ | |
822 | if($this->stack[1]->parentNode) { | |
823 | $this->stack[1]->parentNode->removeChild($this->stack[1]); | |
824 | } | |
825 | ||
826 | /* 2. Pop all the nodes from the bottom of the stack of | |
827 | * open elements, from the current node up to the root | |
828 | * html element. */ | |
829 | array_splice($this->stack, 1); | |
830 | ||
831 | $this->insertElement($token); | |
832 | $this->mode = self::IN_FRAMESET; | |
833 | } | |
834 | break; | |
835 | ||
836 | // in spec, there is a diversion here | |
837 | ||
838 | case 'address': case 'article': case 'aside': case 'blockquote': | |
839 | case 'center': case 'datagrid': case 'details': case 'dir': | |
840 | case 'div': case 'dl': case 'fieldset': case 'figure': case 'footer': | |
841 | case 'header': case 'hgroup': case 'menu': case 'nav': | |
842 | case 'ol': case 'p': case 'section': case 'ul': | |
843 | /* If the stack of open elements has a p element in scope, | |
844 | then act as if an end tag with the tag name p had been | |
845 | seen. */ | |
846 | if($this->elementInScope('p')) { | |
847 | $this->emitToken(array( | |
848 | 'name' => 'p', | |
849 | 'type' => HTML5_Tokenizer::ENDTAG | |
850 | )); | |
851 | } | |
852 | ||
853 | /* Insert an HTML element for the token. */ | |
854 | $this->insertElement($token); | |
855 | break; | |
856 | ||
857 | /* A start tag whose tag name is one of: "h1", "h2", "h3", "h4", | |
858 | "h5", "h6" */ | |
859 | case 'h1': case 'h2': case 'h3': case 'h4': case 'h5': case 'h6': | |
860 | /* If the stack of open elements has a p element in scope, | |
861 | then act as if an end tag with the tag name p had been seen. */ | |
862 | if($this->elementInScope('p')) { | |
863 | $this->emitToken(array( | |
864 | 'name' => 'p', | |
865 | 'type' => HTML5_Tokenizer::ENDTAG | |
866 | )); | |
867 | } | |
868 | ||
869 | /* If the current node is an element whose tag name is one | |
870 | * of "h1", "h2", "h3", "h4", "h5", or "h6", then this is a | |
871 | * parse error; pop the current node off the stack of open | |
872 | * elements. */ | |
873 | $peek = array_pop($this->stack); | |
874 | if (in_array($peek->tagName, array("h1", "h2", "h3", "h4", "h5", "h6"))) { | |
875 | // parse error | |
876 | } else { | |
877 | $this->stack[] = $peek; | |
878 | } | |
879 | ||
880 | /* Insert an HTML element for the token. */ | |
881 | $this->insertElement($token); | |
882 | break; | |
883 | ||
884 | case 'pre': case 'listing': | |
885 | /* If the stack of open elements has a p element in scope, | |
886 | then act as if an end tag with the tag name p had been seen. */ | |
887 | if($this->elementInScope('p')) { | |
888 | $this->emitToken(array( | |
889 | 'name' => 'p', | |
890 | 'type' => HTML5_Tokenizer::ENDTAG | |
891 | )); | |
892 | } | |
893 | $this->insertElement($token); | |
894 | /* If the next token is a U+000A LINE FEED (LF) character | |
895 | * token, then ignore that token and move on to the next | |
896 | * one. (Newlines at the start of pre blocks are ignored as | |
897 | * an authoring convenience.) */ | |
898 | $this->ignore_lf_token = 2; | |
899 | $this->flag_frameset_ok = false; | |
900 | break; | |
901 | ||
902 | /* A start tag whose tag name is "form" */ | |
903 | case 'form': | |
904 | /* If the form element pointer is not null, ignore the | |
905 | token with a parse error. */ | |
906 | if($this->form_pointer !== null) { | |
907 | $this->ignored = true; | |
908 | // Ignore. | |
909 | ||
910 | /* Otherwise: */ | |
911 | } else { | |
912 | /* If the stack of open elements has a p element in | |
913 | scope, then act as if an end tag with the tag name p | |
914 | had been seen. */ | |
915 | if($this->elementInScope('p')) { | |
916 | $this->emitToken(array( | |
917 | 'name' => 'p', | |
918 | 'type' => HTML5_Tokenizer::ENDTAG | |
919 | )); | |
920 | } | |
921 | ||
922 | /* Insert an HTML element for the token, and set the | |
923 | form element pointer to point to the element created. */ | |
924 | $element = $this->insertElement($token); | |
925 | $this->form_pointer = $element; | |
926 | } | |
927 | break; | |
928 | ||
929 | // condensed specification | |
930 | case 'li': case 'dc': case 'dd': case 'ds': case 'dt': | |
931 | /* 1. Set the frameset-ok flag to "not ok". */ | |
932 | $this->flag_frameset_ok = false; | |
933 | ||
934 | $stack_length = count($this->stack) - 1; | |
935 | for($n = $stack_length; 0 <= $n; $n--) { | |
936 | /* 2. Initialise node to be the current node (the | |
937 | bottommost node of the stack). */ | |
938 | $stop = false; | |
939 | $node = $this->stack[$n]; | |
940 | $cat = $this->getElementCategory($node); | |
941 | ||
942 | // for case 'li': | |
943 | /* 3. If node is an li element, then act as if an end | |
944 | * tag with the tag name "li" had been seen, then jump | |
945 | * to the last step. */ | |
946 | // for case 'dc': case 'dd': case 'ds': case 'dt': | |
947 | /* If node is a dc, dd, ds or dt element, then act as if an end | |
948 | * tag with the same tag name as node had been seen, then | |
949 | * jump to the last step. */ | |
950 | if(($token['name'] === 'li' && $node->tagName === 'li') || | |
951 | ($token['name'] !== 'li' && ($node->tagName == 'dc' || $node->tagName === 'dd' || $node->tagName == 'ds' || $node->tagName === 'dt'))) { // limited conditional | |
952 | $this->emitToken(array( | |
953 | 'type' => HTML5_Tokenizer::ENDTAG, | |
954 | 'name' => $node->tagName, | |
955 | )); | |
956 | break; | |
957 | } | |
958 | ||
959 | /* 4. If node is not in the formatting category, and is | |
960 | not in the phrasing category, and is not an address, | |
961 | div or p element, then stop this algorithm. */ | |
962 | if($cat !== self::FORMATTING && $cat !== self::PHRASING && | |
963 | $node->tagName !== 'address' && $node->tagName !== 'div' && | |
964 | $node->tagName !== 'p') { | |
965 | break; | |
966 | } | |
967 | ||
968 | /* 5. Otherwise, set node to the previous entry in the | |
969 | * stack of open elements and return to step 2. */ | |
970 | } | |
971 | ||
972 | /* 6. This is the last step. */ | |
973 | ||
974 | /* If the stack of open elements has a p element in scope, | |
975 | then act as if an end tag with the tag name p had been | |
976 | seen. */ | |
977 | if($this->elementInScope('p')) { | |
978 | $this->emitToken(array( | |
979 | 'name' => 'p', | |
980 | 'type' => HTML5_Tokenizer::ENDTAG | |
981 | )); | |
982 | } | |
983 | ||
984 | /* Finally, insert an HTML element with the same tag | |
985 | name as the token's. */ | |
986 | $this->insertElement($token); | |
987 | break; | |
988 | ||
989 | /* A start tag token whose tag name is "plaintext" */ | |
990 | case 'plaintext': | |
991 | /* If the stack of open elements has a p element in scope, | |
992 | then act as if an end tag with the tag name p had been | |
993 | seen. */ | |
994 | if($this->elementInScope('p')) { | |
995 | $this->emitToken(array( | |
996 | 'name' => 'p', | |
997 | 'type' => HTML5_Tokenizer::ENDTAG | |
998 | )); | |
999 | } | |
1000 | ||
1001 | /* Insert an HTML element for the token. */ | |
1002 | $this->insertElement($token); | |
1003 | ||
1004 | $this->content_model = HTML5_Tokenizer::PLAINTEXT; | |
1005 | break; | |
1006 | ||
1007 | // more diversions | |
1008 | ||
1009 | /* A start tag whose tag name is "a" */ | |
1010 | case 'a': | |
1011 | /* If the list of active formatting elements contains | |
1012 | an element whose tag name is "a" between the end of the | |
1013 | list and the last marker on the list (or the start of | |
1014 | the list if there is no marker on the list), then this | |
1015 | is a parse error; act as if an end tag with the tag name | |
1016 | "a" had been seen, then remove that element from the list | |
1017 | of active formatting elements and the stack of open | |
1018 | elements if the end tag didn't already remove it (it | |
1019 | might not have if the element is not in table scope). */ | |
1020 | $leng = count($this->a_formatting); | |
1021 | ||
1022 | for($n = $leng - 1; $n >= 0; $n--) { | |
1023 | if($this->a_formatting[$n] === self::MARKER) { | |
1024 | break; | |
1025 | ||
1026 | } elseif($this->a_formatting[$n]->tagName === 'a') { | |
1027 | $a = $this->a_formatting[$n]; | |
1028 | $this->emitToken(array( | |
1029 | 'name' => 'a', | |
1030 | 'type' => HTML5_Tokenizer::ENDTAG | |
1031 | )); | |
1032 | if (in_array($a, $this->a_formatting)) { | |
1033 | $a_i = array_search($a, $this->a_formatting, true); | |
1034 | if($a_i !== false) array_splice($this->a_formatting, $a_i, 1); | |
1035 | } | |
1036 | if (in_array($a, $this->stack)) { | |
1037 | $a_i = array_search($a, $this->stack, true); | |
1038 | if ($a_i !== false) array_splice($this->stack, $a_i, 1); | |
1039 | } | |
1040 | break; | |
1041 | } | |
1042 | } | |
1043 | ||
1044 | /* Reconstruct the active formatting elements, if any. */ | |
1045 | $this->reconstructActiveFormattingElements(); | |
1046 | ||
1047 | /* Insert an HTML element for the token. */ | |
1048 | $el = $this->insertElement($token); | |
1049 | ||
1050 | /* Add that element to the list of active formatting | |
1051 | elements. */ | |
1052 | $this->a_formatting[] = $el; | |
1053 | break; | |
1054 | ||
1055 | case 'b': case 'big': case 'code': case 'em': case 'font': case 'i': | |
1056 | case 's': case 'small': case 'strike': | |
1057 | case 'strong': case 'tt': case 'u': | |
1058 | /* Reconstruct the active formatting elements, if any. */ | |
1059 | $this->reconstructActiveFormattingElements(); | |
1060 | ||
1061 | /* Insert an HTML element for the token. */ | |
1062 | $el = $this->insertElement($token); | |
1063 | ||
1064 | /* Add that element to the list of active formatting | |
1065 | elements. */ | |
1066 | $this->a_formatting[] = $el; | |
1067 | break; | |
1068 | ||
1069 | case 'nobr': | |
1070 | /* Reconstruct the active formatting elements, if any. */ | |
1071 | $this->reconstructActiveFormattingElements(); | |
1072 | ||
1073 | /* If the stack of open elements has a nobr element in | |
1074 | * scope, then this is a parse error; act as if an end tag | |
1075 | * with the tag name "nobr" had been seen, then once again | |
1076 | * reconstruct the active formatting elements, if any. */ | |
1077 | if ($this->elementInScope('nobr')) { | |
1078 | $this->emitToken(array( | |
1079 | 'name' => 'nobr', | |
1080 | 'type' => HTML5_Tokenizer::ENDTAG, | |
1081 | )); | |
1082 | $this->reconstructActiveFormattingElements(); | |
1083 | } | |
1084 | ||
1085 | /* Insert an HTML element for the token. */ | |
1086 | $el = $this->insertElement($token); | |
1087 | ||
1088 | /* Add that element to the list of active formatting | |
1089 | elements. */ | |
1090 | $this->a_formatting[] = $el; | |
1091 | break; | |
1092 | ||
1093 | // another diversion | |
1094 | ||
1095 | /* A start tag token whose tag name is "button" */ | |
1096 | case 'button': | |
1097 | /* If the stack of open elements has a button element in scope, | |
1098 | then this is a parse error; act as if an end tag with the tag | |
1099 | name "button" had been seen, then reprocess the token. (We don't | |
1100 | do that. Unnecessary.) (I hope you're right! -- ezyang) */ | |
1101 | if($this->elementInScope('button')) { | |
1102 | $this->emitToken(array( | |
1103 | 'name' => 'button', | |
1104 | 'type' => HTML5_Tokenizer::ENDTAG | |
1105 | )); | |
1106 | } | |
1107 | ||
1108 | /* Reconstruct the active formatting elements, if any. */ | |
1109 | $this->reconstructActiveFormattingElements(); | |
1110 | ||
1111 | /* Insert an HTML element for the token. */ | |
1112 | $this->insertElement($token); | |
1113 | ||
1114 | /* Insert a marker at the end of the list of active | |
1115 | formatting elements. */ | |
1116 | $this->a_formatting[] = self::MARKER; | |
1117 | ||
1118 | $this->flag_frameset_ok = false; | |
1119 | break; | |
1120 | ||
1121 | case 'applet': case 'marquee': case 'object': | |
1122 | /* Reconstruct the active formatting elements, if any. */ | |
1123 | $this->reconstructActiveFormattingElements(); | |
1124 | ||
1125 | /* Insert an HTML element for the token. */ | |
1126 | $this->insertElement($token); | |
1127 | ||
1128 | /* Insert a marker at the end of the list of active | |
1129 | formatting elements. */ | |
1130 | $this->a_formatting[] = self::MARKER; | |
1131 | ||
1132 | $this->flag_frameset_ok = false; | |
1133 | break; | |
1134 | ||
1135 | // spec diversion | |
1136 | ||
1137 | /* A start tag whose tag name is "table" */ | |
1138 | case 'table': | |
1139 | /* If the Document is not set to quirks mode, and the | |
1140 | * stack of open elements has a p element in scope, then | |
1141 | * act as if an end tag with the tag name "p" had been | |
1142 | * seen. */ | |
1143 | if($this->quirks_mode !== self::QUIRKS_MODE && | |
1144 | $this->elementInScope('p')) { | |
1145 | $this->emitToken(array( | |
1146 | 'name' => 'p', | |
1147 | 'type' => HTML5_Tokenizer::ENDTAG | |
1148 | )); | |
1149 | } | |
1150 | ||
1151 | /* Insert an HTML element for the token. */ | |
1152 | $this->insertElement($token); | |
1153 | ||
1154 | $this->flag_frameset_ok = false; | |
1155 | ||
1156 | /* Change the insertion mode to "in table". */ | |
1157 | $this->mode = self::IN_TABLE; | |
1158 | break; | |
1159 | ||
1160 | /* A start tag whose tag name is one of: "area", "basefont", | |
1161 | "bgsound", "br", "embed", "img", "param", "spacer", "wbr" */ | |
1162 | case 'area': case 'basefont': case 'bgsound': case 'br': | |
1163 | case 'embed': case 'img': case 'input': case 'keygen': case 'spacer': | |
1164 | case 'wbr': | |
1165 | /* Reconstruct the active formatting elements, if any. */ | |
1166 | $this->reconstructActiveFormattingElements(); | |
1167 | ||
1168 | /* Insert an HTML element for the token. */ | |
1169 | $this->insertElement($token); | |
1170 | ||
1171 | /* Immediately pop the current node off the stack of open elements. */ | |
1172 | array_pop($this->stack); | |
1173 | ||
1174 | // YYY: Acknowledge the token's self-closing flag, if it is set. | |
1175 | ||
1176 | $this->flag_frameset_ok = false; | |
1177 | break; | |
1178 | ||
1179 | case 'param': case 'source': | |
1180 | /* Insert an HTML element for the token. */ | |
1181 | $this->insertElement($token); | |
1182 | ||
1183 | /* Immediately pop the current node off the stack of open elements. */ | |
1184 | array_pop($this->stack); | |
1185 | ||
1186 | // YYY: Acknowledge the token's self-closing flag, if it is set. | |
1187 | break; | |
1188 | ||
1189 | /* A start tag whose tag name is "hr" */ | |
1190 | case 'hr': | |
1191 | /* If the stack of open elements has a p element in scope, | |
1192 | then act as if an end tag with the tag name p had been seen. */ | |
1193 | if($this->elementInScope('p')) { | |
1194 | $this->emitToken(array( | |
1195 | 'name' => 'p', | |
1196 | 'type' => HTML5_Tokenizer::ENDTAG | |
1197 | )); | |
1198 | } | |
1199 | ||
1200 | /* Insert an HTML element for the token. */ | |
1201 | $this->insertElement($token); | |
1202 | ||
1203 | /* Immediately pop the current node off the stack of open elements. */ | |
1204 | array_pop($this->stack); | |
1205 | ||
1206 | // YYY: Acknowledge the token's self-closing flag, if it is set. | |
1207 | ||
1208 | $this->flag_frameset_ok = false; | |
1209 | break; | |
1210 | ||
1211 | /* A start tag whose tag name is "image" */ | |
1212 | case 'image': | |
1213 | /* Parse error. Change the token's tag name to "img" and | |
1214 | reprocess it. (Don't ask.) */ | |
1215 | $token['name'] = 'img'; | |
1216 | $this->emitToken($token); | |
1217 | break; | |
1218 | ||
1219 | /* A start tag whose tag name is "isindex" */ | |
1220 | case 'isindex': | |
1221 | /* Parse error. */ | |
1222 | ||
1223 | /* If the form element pointer is not null, | |
1224 | then ignore the token. */ | |
1225 | if($this->form_pointer === null) { | |
1226 | /* Act as if a start tag token with the tag name "form" had | |
1227 | been seen. */ | |
1228 | /* If the token has an attribute called "action", set | |
1229 | * the action attribute on the resulting form | |
1230 | * element to the value of the "action" attribute of | |
1231 | * the token. */ | |
1232 | $attr = array(); | |
1233 | $action = $this->getAttr($token, 'action'); | |
1234 | if ($action !== false) { | |
1235 | $attr[] = array('name' => 'action', 'value' => $action); | |
1236 | } | |
1237 | $this->emitToken(array( | |
1238 | 'name' => 'form', | |
1239 | 'type' => HTML5_Tokenizer::STARTTAG, | |
1240 | 'attr' => $attr | |
1241 | )); | |
1242 | ||
1243 | /* Act as if a start tag token with the tag name "hr" had | |
1244 | been seen. */ | |
1245 | $this->emitToken(array( | |
1246 | 'name' => 'hr', | |
1247 | 'type' => HTML5_Tokenizer::STARTTAG, | |
1248 | 'attr' => array() | |
1249 | )); | |
1250 | ||
1251 | /* Act as if a start tag token with the tag name "label" | |
1252 | had been seen. */ | |
1253 | $this->emitToken(array( | |
1254 | 'name' => 'label', | |
1255 | 'type' => HTML5_Tokenizer::STARTTAG, | |
1256 | 'attr' => array() | |
1257 | )); | |
1258 | ||
1259 | /* Act as if a stream of character tokens had been seen. */ | |
1260 | $prompt = $this->getAttr($token, 'prompt'); | |
1261 | if ($prompt === false) { | |
1262 | $prompt = 'This is a searchable index. '. | |
1263 | 'Insert your search keywords here: '; | |
1264 | } | |
1265 | $this->emitToken(array( | |
1266 | 'data' => $prompt, | |
1267 | 'type' => HTML5_Tokenizer::CHARACTER, | |
1268 | )); | |
1269 | ||
1270 | /* Act as if a start tag token with the tag name "input" | |
1271 | had been seen, with all the attributes from the "isindex" | |
1272 | token, except with the "name" attribute set to the value | |
1273 | "isindex" (ignoring any explicit "name" attribute). */ | |
1274 | $attr = array(); | |
1275 | foreach ($token['attr'] as $keypair) { | |
1276 | if ($keypair['name'] === 'name' || $keypair['name'] === 'action' || | |
1277 | $keypair['name'] === 'prompt') continue; | |
1278 | $attr[] = $keypair; | |
1279 | } | |
1280 | $attr[] = array('name' => 'name', 'value' => 'isindex'); | |
1281 | ||
1282 | $this->emitToken(array( | |
1283 | 'name' => 'input', | |
1284 | 'type' => HTML5_Tokenizer::STARTTAG, | |
1285 | 'attr' => $attr | |
1286 | )); | |
1287 | ||
1288 | /* Act as if an end tag token with the tag name "label" | |
1289 | had been seen. */ | |
1290 | $this->emitToken(array( | |
1291 | 'name' => 'label', | |
1292 | 'type' => HTML5_Tokenizer::ENDTAG | |
1293 | )); | |
1294 | ||
1295 | /* Act as if a start tag token with the tag name "hr" had | |
1296 | been seen. */ | |
1297 | $this->emitToken(array( | |
1298 | 'name' => 'hr', | |
1299 | 'type' => HTML5_Tokenizer::STARTTAG | |
1300 | )); | |
1301 | ||
1302 | /* Act as if an end tag token with the tag name "form" had | |
1303 | been seen. */ | |
1304 | $this->emitToken(array( | |
1305 | 'name' => 'form', | |
1306 | 'type' => HTML5_Tokenizer::ENDTAG | |
1307 | )); | |
1308 | } else { | |
1309 | $this->ignored = true; | |
1310 | } | |
1311 | break; | |
1312 | ||
1313 | /* A start tag whose tag name is "textarea" */ | |
1314 | case 'textarea': | |
1315 | $this->insertElement($token); | |
1316 | ||
1317 | /* If the next token is a U+000A LINE FEED (LF) | |
1318 | * character token, then ignore that token and move on to | |
1319 | * the next one. (Newlines at the start of textarea | |
1320 | * elements are ignored as an authoring convenience.) | |
1321 | * need flag, see also <pre> */ | |
1322 | $this->ignore_lf_token = 2; | |
1323 | ||
1324 | $this->original_mode = $this->mode; | |
1325 | $this->flag_frameset_ok = false; | |
1326 | $this->mode = self::IN_CDATA_RCDATA; | |
1327 | ||
1328 | /* Switch the tokeniser's content model flag to the | |
1329 | RCDATA state. */ | |
1330 | $this->content_model = HTML5_Tokenizer::RCDATA; | |
1331 | break; | |
1332 | ||
1333 | /* A start tag token whose tag name is "xmp" */ | |
1334 | case 'xmp': | |
1335 | /* If the stack of open elements has a p element in | |
1336 | scope, then act as if an end tag with the tag name | |
1337 | "p" has been seen. */ | |
1338 | if ($this->elementInScope('p')) { | |
1339 | $this->emitToken(array( | |
1340 | 'name' => 'p', | |
1341 | 'type' => HTML5_Tokenizer::ENDTAG | |
1342 | )); | |
1343 | } | |
1344 | ||
1345 | /* Reconstruct the active formatting elements, if any. */ | |
1346 | $this->reconstructActiveFormattingElements(); | |
1347 | ||
1348 | $this->flag_frameset_ok = false; | |
1349 | ||
1350 | $this->insertCDATAElement($token); | |
1351 | break; | |
1352 | ||
1353 | case 'iframe': | |
1354 | $this->flag_frameset_ok = false; | |
1355 | $this->insertCDATAElement($token); | |
1356 | break; | |
1357 | ||
1358 | case 'noembed': case 'noscript': | |
1359 | // XSCRIPT: should check scripting flag | |
1360 | $this->insertCDATAElement($token); | |
1361 | break; | |
1362 | ||
1363 | /* A start tag whose tag name is "select" */ | |
1364 | case 'select': | |
1365 | /* Reconstruct the active formatting elements, if any. */ | |
1366 | $this->reconstructActiveFormattingElements(); | |
1367 | ||
1368 | /* Insert an HTML element for the token. */ | |
1369 | $this->insertElement($token); | |
1370 | ||
1371 | $this->flag_frameset_ok = false; | |
1372 | ||
1373 | /* If the insertion mode is one of in table", "in caption", | |
1374 | * "in column group", "in table body", "in row", or "in | |
1375 | * cell", then switch the insertion mode to "in select in | |
1376 | * table". Otherwise, switch the insertion mode to "in | |
1377 | * select". */ | |
1378 | if ( | |
1379 | $this->mode === self::IN_TABLE || $this->mode === self::IN_CAPTION || | |
1380 | $this->mode === self::IN_COLUMN_GROUP || $this->mode ==+self::IN_TABLE_BODY || | |
1381 | $this->mode === self::IN_ROW || $this->mode === self::IN_CELL | |
1382 | ) { | |
1383 | $this->mode = self::IN_SELECT_IN_TABLE; | |
1384 | } else { | |
1385 | $this->mode = self::IN_SELECT; | |
1386 | } | |
1387 | break; | |
1388 | ||
1389 | case 'option': case 'optgroup': | |
1390 | if ($this->elementInScope('option')) { | |
1391 | $this->emitToken(array( | |
1392 | 'name' => 'option', | |
1393 | 'type' => HTML5_Tokenizer::ENDTAG, | |
1394 | )); | |
1395 | } | |
1396 | $this->reconstructActiveFormattingElements(); | |
1397 | $this->insertElement($token); | |
1398 | break; | |
1399 | ||
1400 | case 'rp': case 'rt': | |
1401 | /* If the stack of open elements has a ruby element in scope, then generate | |
1402 | * implied end tags. If the current node is not then a ruby element, this is | |
1403 | * a parse error; pop all the nodes from the current node up to the node | |
1404 | * immediately before the bottommost ruby element on the stack of open elements. | |
1405 | */ | |
1406 | if ($this->elementInScope('ruby')) { | |
1407 | $this->generateImpliedEndTags(); | |
1408 | } | |
1409 | $peek = false; | |
1410 | do { | |
1411 | if ($peek) { | |
1412 | // parse error | |
1413 | } | |
1414 | $peek = array_pop($this->stack); | |
1415 | } while ($peek->tagName !== 'ruby'); | |
1416 | $this->stack[] = $peek; // we popped one too many | |
1417 | $this->insertElement($token); | |
1418 | break; | |
1419 | ||
1420 | // spec diversion | |
1421 | ||
1422 | case 'math': | |
1423 | $this->reconstructActiveFormattingElements(); | |
1424 | $token = $this->adjustMathMLAttributes($token); | |
1425 | $token = $this->adjustForeignAttributes($token); | |
1426 | $this->insertForeignElement($token, self::NS_MATHML); | |
1427 | if (isset($token['self-closing'])) { | |
1428 | // XERROR: acknowledge the token's self-closing flag | |
1429 | array_pop($this->stack); | |
1430 | } | |
1431 | if ($this->mode !== self::IN_FOREIGN_CONTENT) { | |
1432 | $this->secondary_mode = $this->mode; | |
1433 | $this->mode = self::IN_FOREIGN_CONTENT; | |
1434 | } | |
1435 | break; | |
1436 | ||
1437 | case 'svg': | |
1438 | $this->reconstructActiveFormattingElements(); | |
1439 | $token = $this->adjustSVGAttributes($token); | |
1440 | $token = $this->adjustForeignAttributes($token); | |
1441 | $this->insertForeignElement($token, self::NS_SVG); | |
1442 | if (isset($token['self-closing'])) { | |
1443 | // XERROR: acknowledge the token's self-closing flag | |
1444 | array_pop($this->stack); | |
1445 | } | |
1446 | if ($this->mode !== self::IN_FOREIGN_CONTENT) { | |
1447 | $this->secondary_mode = $this->mode; | |
1448 | $this->mode = self::IN_FOREIGN_CONTENT; | |
1449 | } | |
1450 | break; | |
1451 | ||
1452 | case 'caption': case 'col': case 'colgroup': case 'frame': case 'head': | |
1453 | case 'tbody': case 'td': case 'tfoot': case 'th': case 'thead': case 'tr': | |
1454 | // parse error | |
1455 | break; | |
1456 | ||
1457 | /* A start tag token not covered by the previous entries */ | |
1458 | default: | |
1459 | /* Reconstruct the active formatting elements, if any. */ | |
1460 | $this->reconstructActiveFormattingElements(); | |
1461 | ||
1462 | $this->insertElement($token); | |
1463 | /* This element will be a phrasing element. */ | |
1464 | break; | |
1465 | } | |
1466 | break; | |
1467 | ||
1468 | case HTML5_Tokenizer::ENDTAG: | |
1469 | switch($token['name']) { | |
1470 | /* An end tag with the tag name "body" */ | |
1471 | case 'body': | |
1472 | /* If the stack of open elements does not have a body | |
1473 | * element in scope, this is a parse error; ignore the | |
1474 | * token. */ | |
1475 | if(!$this->elementInScope('body')) { | |
1476 | $this->ignored = true; | |
1477 | ||
1478 | /* Otherwise, if there is a node in the stack of open | |
1479 | * elements that is not either a dc element, a dd element, | |
1480 | * a ds element, a dt element, an li element, an optgroup | |
1481 | * element, an option element, a p element, an rp element, | |
1482 | * an rt element, a tbody element, a td element, a tfoot | |
1483 | * element, a th element, a thead element, a tr element, | |
1484 | * the body element, or the html element, then this is a | |
1485 | * parse error. | |
1486 | */ | |
1487 | } else { | |
1488 | // XERROR: implement this check for parse error | |
1489 | } | |
1490 | ||
1491 | /* Change the insertion mode to "after body". */ | |
1492 | $this->mode = self::AFTER_BODY; | |
1493 | break; | |
1494 | ||
1495 | /* An end tag with the tag name "html" */ | |
1496 | case 'html': | |
1497 | /* Act as if an end tag with tag name "body" had been seen, | |
1498 | then, if that token wasn't ignored, reprocess the current | |
1499 | token. */ | |
1500 | $this->emitToken(array( | |
1501 | 'name' => 'body', | |
1502 | 'type' => HTML5_Tokenizer::ENDTAG | |
1503 | )); | |
1504 | ||
1505 | if (!$this->ignored) $this->emitToken($token); | |
1506 | break; | |
1507 | ||
1508 | case 'address': case 'article': case 'aside': case 'blockquote': | |
1509 | case 'center': case 'datagrid': case 'details': case 'dir': | |
1510 | case 'div': case 'dl': case 'fieldset': case 'footer': | |
1511 | case 'header': case 'hgroup': case 'listing': case 'menu': | |
1512 | case 'nav': case 'ol': case 'pre': case 'section': case 'ul': | |
1513 | /* If the stack of open elements has an element in scope | |
1514 | with the same tag name as that of the token, then generate | |
1515 | implied end tags. */ | |
1516 | if($this->elementInScope($token['name'])) { | |
1517 | $this->generateImpliedEndTags(); | |
1518 | ||
1519 | /* Now, if the current node is not an element with | |
1520 | the same tag name as that of the token, then this | |
1521 | is a parse error. */ | |
1522 | // XERROR: implement parse error logic | |
1523 | ||
1524 | /* If the stack of open elements has an element in | |
1525 | scope with the same tag name as that of the token, | |
1526 | then pop elements from this stack until an element | |
1527 | with that tag name has been popped from the stack. */ | |
1528 | do { | |
1529 | $node = array_pop($this->stack); | |
1530 | } while ($node->tagName !== $token['name']); | |
1531 | } else { | |
1532 | // parse error | |
1533 | } | |
1534 | break; | |
1535 | ||
1536 | /* An end tag whose tag name is "form" */ | |
1537 | case 'form': | |
1538 | /* Let node be the element that the form element pointer is set to. */ | |
1539 | $node = $this->form_pointer; | |
1540 | /* Set the form element pointer to null. */ | |
1541 | $this->form_pointer = null; | |
1542 | /* If node is null or the stack of open elements does not | |
1543 | * have node in scope, then this is a parse error; ignore the token. */ | |
1544 | if ($node === null || !in_array($node, $this->stack)) { | |
1545 | // parse error | |
1546 | $this->ignored = true; | |
1547 | } else { | |
1548 | /* 1. Generate implied end tags. */ | |
1549 | $this->generateImpliedEndTags(); | |
1550 | /* 2. If the current node is not node, then this is a parse error. */ | |
1551 | if (end($this->stack) !== $node) { | |
1552 | // parse error | |
1553 | } | |
1554 | /* 3. Remove node from the stack of open elements. */ | |
1555 | array_splice($this->stack, array_search($node, $this->stack, true), 1); | |
1556 | } | |
1557 | ||
1558 | break; | |
1559 | ||
1560 | /* An end tag whose tag name is "p" */ | |
1561 | case 'p': | |
1562 | /* If the stack of open elements has a p element in scope, | |
1563 | then generate implied end tags, except for p elements. */ | |
1564 | if($this->elementInScope('p')) { | |
1565 | /* Generate implied end tags, except for elements with | |
1566 | * the same tag name as the token. */ | |
1567 | $this->generateImpliedEndTags(array('p')); | |
1568 | ||
1569 | /* If the current node is not a p element, then this is | |
1570 | a parse error. */ | |
1571 | // XERROR: implement | |
1572 | ||
1573 | /* Pop elements from the stack of open elements until | |
1574 | * an element with the same tag name as the token has | |
1575 | * been popped from the stack. */ | |
1576 | do { | |
1577 | $node = array_pop($this->stack); | |
1578 | } while ($node->tagName !== 'p'); | |
1579 | ||
1580 | } else { | |
1581 | // parse error | |
1582 | $this->emitToken(array( | |
1583 | 'name' => 'p', | |
1584 | 'type' => HTML5_Tokenizer::STARTTAG, | |
1585 | )); | |
1586 | $this->emitToken($token); | |
1587 | } | |
1588 | break; | |
1589 | ||
1590 | /* An end tag whose tag name is "li" */ | |
1591 | case 'li': | |
1592 | /* If the stack of open elements does not have an element | |
1593 | * in list item scope with the same tag name as that of the | |
1594 | * token, then this is a parse error; ignore the token. */ | |
1595 | if ($this->elementInScope($token['name'], self::SCOPE_LISTITEM)) { | |
1596 | /* Generate implied end tags, except for elements with the | |
1597 | * same tag name as the token. */ | |
1598 | $this->generateImpliedEndTags(array($token['name'])); | |
1599 | /* If the current node is not an element with the same tag | |
1600 | * name as that of the token, then this is a parse error. */ | |
1601 | // XERROR: parse error | |
1602 | /* Pop elements from the stack of open elements until an | |
1603 | * element with the same tag name as the token has been | |
1604 | * popped from the stack. */ | |
1605 | do { | |
1606 | $node = array_pop($this->stack); | |
1607 | } while ($node->tagName !== $token['name']); | |
1608 | } else { | |
1609 | // XERROR: parse error | |
1610 | } | |
1611 | break; | |
1612 | ||
1613 | /* An end tag whose tag name is "dc", "dd", "ds", "dt" */ | |
1614 | case 'dc': case 'dd': case 'ds': case 'dt': | |
1615 | if($this->elementInScope($token['name'])) { | |
1616 | $this->generateImpliedEndTags(array($token['name'])); | |
1617 | ||
1618 | /* If the current node is not an element with the same | |
1619 | tag name as the token, then this is a parse error. */ | |
1620 | // XERROR: implement parse error | |
1621 | ||
1622 | /* Pop elements from the stack of open elements until | |
1623 | * an element with the same tag name as the token has | |
1624 | * been popped from the stack. */ | |
1625 | do { | |
1626 | $node = array_pop($this->stack); | |
1627 | } while ($node->tagName !== $token['name']); | |
1628 | ||
1629 | } else { | |
1630 | // XERROR: parse error | |
1631 | } | |
1632 | break; | |
1633 | ||
1634 | /* An end tag whose tag name is one of: "h1", "h2", "h3", "h4", | |
1635 | "h5", "h6" */ | |
1636 | case 'h1': case 'h2': case 'h3': case 'h4': case 'h5': case 'h6': | |
1637 | $elements = array('h1', 'h2', 'h3', 'h4', 'h5', 'h6'); | |
1638 | ||
1639 | /* If the stack of open elements has in scope an element whose | |
1640 | tag name is one of "h1", "h2", "h3", "h4", "h5", or "h6", then | |
1641 | generate implied end tags. */ | |
1642 | if($this->elementInScope($elements)) { | |
1643 | $this->generateImpliedEndTags(); | |
1644 | ||
1645 | /* Now, if the current node is not an element with the same | |
1646 | tag name as that of the token, then this is a parse error. */ | |
1647 | // XERROR: implement parse error | |
1648 | ||
1649 | /* If the stack of open elements has in scope an element | |
1650 | whose tag name is one of "h1", "h2", "h3", "h4", "h5", or | |
1651 | "h6", then pop elements from the stack until an element | |
1652 | with one of those tag names has been popped from the stack. */ | |
1653 | do { | |
1654 | $node = array_pop($this->stack); | |
1655 | } while (!in_array($node->tagName, $elements)); | |
1656 | } else { | |
1657 | // parse error | |
1658 | } | |
1659 | break; | |
1660 | ||
1661 | /* An end tag whose tag name is one of: "a", "b", "big", "em", | |
1662 | "font", "i", "nobr", "s", "small", "strike", "strong", "tt", "u" */ | |
1663 | case 'a': case 'b': case 'big': case 'code': case 'em': case 'font': | |
1664 | case 'i': case 'nobr': case 's': case 'small': case 'strike': | |
1665 | case 'strong': case 'tt': case 'u': | |
1666 | // XERROR: generally speaking this needs parse error logic | |
1667 | /* 1. Let the formatting element be the last element in | |
1668 | the list of active formatting elements that: | |
1669 | * is between the end of the list and the last scope | |
1670 | marker in the list, if any, or the start of the list | |
1671 | otherwise, and | |
1672 | * has the same tag name as the token. | |
1673 | */ | |
1674 | while(true) { | |
1675 | for($a = count($this->a_formatting) - 1; $a >= 0; $a--) { | |
1676 | if($this->a_formatting[$a] === self::MARKER) { | |
1677 | break; | |
1678 | ||
1679 | } elseif($this->a_formatting[$a]->tagName === $token['name']) { | |
1680 | $formatting_element = $this->a_formatting[$a]; | |
1681 | $in_stack = in_array($formatting_element, $this->stack, true); | |
1682 | $fe_af_pos = $a; | |
1683 | break; | |
1684 | } | |
1685 | } | |
1686 | ||
1687 | /* If there is no such node, or, if that node is | |
1688 | also in the stack of open elements but the element | |
1689 | is not in scope, then this is a parse error. Abort | |
1690 | these steps. The token is ignored. */ | |
1691 | if(!isset($formatting_element) || ($in_stack && | |
1692 | !$this->elementInScope($token['name']))) { | |
1693 | $this->ignored = true; | |
1694 | break; | |
1695 | ||
1696 | /* Otherwise, if there is such a node, but that node | |
1697 | is not in the stack of open elements, then this is a | |
1698 | parse error; remove the element from the list, and | |
1699 | abort these steps. */ | |
1700 | } elseif(isset($formatting_element) && !$in_stack) { | |
1701 | unset($this->a_formatting[$fe_af_pos]); | |
1702 | $this->a_formatting = array_merge($this->a_formatting); | |
1703 | break; | |
1704 | } | |
1705 | ||
1706 | /* Otherwise, there is a formatting element and that | |
1707 | * element is in the stack and is in scope. If the | |
1708 | * element is not the current node, this is a parse | |
1709 | * error. In any case, proceed with the algorithm as | |
1710 | * written in the following steps. */ | |
1711 | // XERROR: implement me | |
1712 | ||
1713 | /* 2. Let the furthest block be the topmost node in the | |
1714 | stack of open elements that is lower in the stack | |
1715 | than the formatting element, and is not an element in | |
1716 | the phrasing or formatting categories. There might | |
1717 | not be one. */ | |
1718 | $fe_s_pos = array_search($formatting_element, $this->stack, true); | |
1719 | $length = count($this->stack); | |
1720 | ||
1721 | for($s = $fe_s_pos + 1; $s < $length; $s++) { | |
1722 | $category = $this->getElementCategory($this->stack[$s]); | |
1723 | ||
1724 | if($category !== self::PHRASING && $category !== self::FORMATTING) { | |
1725 | $furthest_block = $this->stack[$s]; | |
1726 | break; | |
1727 | } | |
1728 | } | |
1729 | ||
1730 | /* 3. If there is no furthest block, then the UA must | |
1731 | skip the subsequent steps and instead just pop all | |
1732 | the nodes from the bottom of the stack of open | |
1733 | elements, from the current node up to the formatting | |
1734 | element, and remove the formatting element from the | |
1735 | list of active formatting elements. */ | |
1736 | if(!isset($furthest_block)) { | |
1737 | for($n = $length - 1; $n >= $fe_s_pos; $n--) { | |
1738 | array_pop($this->stack); | |
1739 | } | |
1740 | ||
1741 | unset($this->a_formatting[$fe_af_pos]); | |
1742 | $this->a_formatting = array_merge($this->a_formatting); | |
1743 | break; | |
1744 | } | |
1745 | ||
1746 | /* 4. Let the common ancestor be the element | |
1747 | immediately above the formatting element in the stack | |
1748 | of open elements. */ | |
1749 | $common_ancestor = $this->stack[$fe_s_pos - 1]; | |
1750 | ||
1751 | /* 5. Let a bookmark note the position of the | |
1752 | formatting element in the list of active formatting | |
1753 | elements relative to the elements on either side | |
1754 | of it in the list. */ | |
1755 | $bookmark = $fe_af_pos; | |
1756 | ||
1757 | /* 6. Let node and last node be the furthest block. | |
1758 | Follow these steps: */ | |
1759 | $node = $furthest_block; | |
1760 | $last_node = $furthest_block; | |
1761 | ||
1762 | while(true) { | |
1763 | for($n = array_search($node, $this->stack, true) - 1; $n >= 0; $n--) { | |
1764 | /* 6.1 Let node be the element immediately | |
1765 | prior to node in the stack of open elements. */ | |
1766 | $node = $this->stack[$n]; | |
1767 | ||
1768 | /* 6.2 If node is not in the list of active | |
1769 | formatting elements, then remove node from | |
1770 | the stack of open elements and then go back | |
1771 | to step 1. */ | |
1772 | if(!in_array($node, $this->a_formatting, true)) { | |
1773 | array_splice($this->stack, $n, 1); | |
1774 | ||
1775 | } else { | |
1776 | break; | |
1777 | } | |
1778 | } | |
1779 | ||
1780 | /* 6.3 Otherwise, if node is the formatting | |
1781 | element, then go to the next step in the overall | |
1782 | algorithm. */ | |
1783 | if($node === $formatting_element) { | |
1784 | break; | |
1785 | ||
1786 | /* 6.4 Otherwise, if last node is the furthest | |
1787 | block, then move the aforementioned bookmark to | |
1788 | be immediately after the node in the list of | |
1789 | active formatting elements. */ | |
1790 | } elseif($last_node === $furthest_block) { | |
1791 | $bookmark = array_search($node, $this->a_formatting, true) + 1; | |
1792 | } | |
1793 | ||
1794 | /* 6.5 Create an element for the token for which | |
1795 | * the element node was created, replace the entry | |
1796 | * for node in the list of active formatting | |
1797 | * elements with an entry for the new element, | |
1798 | * replace the entry for node in the stack of open | |
1799 | * elements with an entry for the new element, and | |
1800 | * let node be the new element. */ | |
1801 | // we don't know what the token is anymore | |
1802 | // XDOM | |
1803 | $clone = $node->cloneNode(); | |
1804 | $a_pos = array_search($node, $this->a_formatting, true); | |
1805 | $s_pos = array_search($node, $this->stack, true); | |
1806 | $this->a_formatting[$a_pos] = $clone; | |
1807 | $this->stack[$s_pos] = $clone; | |
1808 | $node = $clone; | |
1809 | ||
1810 | /* 6.6 Insert last node into node, first removing | |
1811 | it from its previous parent node if any. */ | |
1812 | // XDOM | |
1813 | if($last_node->parentNode !== null) { | |
1814 | $last_node->parentNode->removeChild($last_node); | |
1815 | } | |
1816 | ||
1817 | // XDOM | |
1818 | $node->appendChild($last_node); | |
1819 | ||
1820 | /* 6.7 Let last node be node. */ | |
1821 | $last_node = $node; | |
1822 | ||
1823 | /* 6.8 Return to step 1 of this inner set of steps. */ | |
1824 | } | |
1825 | ||
1826 | /* 7. If the common ancestor node is a table, tbody, | |
1827 | * tfoot, thead, or tr element, then, foster parent | |
1828 | * whatever last node ended up being in the previous | |
1829 | * step, first removing it from its previous parent | |
1830 | * node if any. */ | |
1831 | // XDOM | |
1832 | if ($last_node->parentNode) { // common step | |
1833 | $last_node->parentNode->removeChild($last_node); | |
1834 | } | |
1835 | if (in_array($common_ancestor->tagName, array('table', 'tbody', 'tfoot', 'thead', 'tr'))) { | |
1836 | $this->fosterParent($last_node); | |
1837 | /* Otherwise, append whatever last node ended up being | |
1838 | * in the previous step to the common ancestor node, | |
1839 | * first removing it from its previous parent node if | |
1840 | * any. */ | |
1841 | } else { | |
1842 | // XDOM | |
1843 | $common_ancestor->appendChild($last_node); | |
1844 | } | |
1845 | ||
1846 | /* 8. Create an element for the token for which the | |
1847 | * formatting element was created. */ | |
1848 | // XDOM | |
1849 | $clone = $formatting_element->cloneNode(); | |
1850 | ||
1851 | /* 9. Take all of the child nodes of the furthest | |
1852 | block and append them to the element created in the | |
1853 | last step. */ | |
1854 | // XDOM | |
1855 | while($furthest_block->hasChildNodes()) { | |
1856 | $child = $furthest_block->firstChild; | |
1857 | $furthest_block->removeChild($child); | |
1858 | $clone->appendChild($child); | |
1859 | } | |
1860 | ||
1861 | /* 10. Append that clone to the furthest block. */ | |
1862 | // XDOM | |
1863 | $furthest_block->appendChild($clone); | |
1864 | ||
1865 | /* 11. Remove the formatting element from the list | |
1866 | of active formatting elements, and insert the new element | |
1867 | into the list of active formatting elements at the | |
1868 | position of the aforementioned bookmark. */ | |
1869 | $fe_af_pos = array_search($formatting_element, $this->a_formatting, true); | |
1870 | array_splice($this->a_formatting, $fe_af_pos, 1); | |
1871 | ||
1872 | $af_part1 = array_slice($this->a_formatting, 0, $bookmark - 1); | |
1873 | $af_part2 = array_slice($this->a_formatting, $bookmark); | |
1874 | $this->a_formatting = array_merge($af_part1, array($clone), $af_part2); | |
1875 | ||
1876 | /* 12. Remove the formatting element from the stack | |
1877 | of open elements, and insert the new element into the stack | |
1878 | of open elements immediately below the position of the | |
1879 | furthest block in that stack. */ | |
1880 | $fe_s_pos = array_search($formatting_element, $this->stack, true); | |
1881 | array_splice($this->stack, $fe_s_pos, 1); | |
1882 | ||
1883 | $fb_s_pos = array_search($furthest_block, $this->stack, true); | |
1884 | $s_part1 = array_slice($this->stack, 0, $fb_s_pos + 1); | |
1885 | $s_part2 = array_slice($this->stack, $fb_s_pos + 1); | |
1886 | $this->stack = array_merge($s_part1, array($clone), $s_part2); | |
1887 | ||
1888 | /* 13. Jump back to step 1 in this series of steps. */ | |
1889 | unset($formatting_element, $fe_af_pos, $fe_s_pos, $furthest_block); | |
1890 | } | |
1891 | break; | |
1892 | ||
1893 | case 'applet': case 'button': case 'marquee': case 'object': | |
1894 | /* If the stack of open elements has an element in scope whose | |
1895 | tag name matches the tag name of the token, then generate implied | |
1896 | tags. */ | |
1897 | if($this->elementInScope($token['name'])) { | |
1898 | $this->generateImpliedEndTags(); | |
1899 | ||
1900 | /* Now, if the current node is not an element with the same | |
1901 | tag name as the token, then this is a parse error. */ | |
1902 | // XERROR: implement logic | |
1903 | ||
1904 | /* Pop elements from the stack of open elements until | |
1905 | * an element with the same tag name as the token has | |
1906 | * been popped from the stack. */ | |
1907 | do { | |
1908 | $node = array_pop($this->stack); | |
1909 | } while ($node->tagName !== $token['name']); | |
1910 | ||
1911 | /* Clear the list of active formatting elements up to the | |
1912 | * last marker. */ | |
1913 | $keys = array_keys($this->a_formatting, self::MARKER, true); | |
1914 | $marker = end($keys); | |
1915 | ||
1916 | for($n = count($this->a_formatting) - 1; $n > $marker; $n--) { | |
1917 | array_pop($this->a_formatting); | |
1918 | } | |
1919 | } else { | |
1920 | // parse error | |
1921 | } | |
1922 | break; | |
1923 | ||
1924 | case 'br': | |
1925 | // Parse error | |
1926 | $this->emitToken(array( | |
1927 | 'name' => 'br', | |
1928 | 'type' => HTML5_Tokenizer::STARTTAG, | |
1929 | )); | |
1930 | break; | |
1931 | ||
1932 | /* An end tag token not covered by the previous entries */ | |
1933 | default: | |
1934 | for($n = count($this->stack) - 1; $n >= 0; $n--) { | |
1935 | /* Initialise node to be the current node (the bottommost | |
1936 | node of the stack). */ | |
1937 | $node = $this->stack[$n]; | |
1938 | ||
1939 | /* If node has the same tag name as the end tag token, | |
1940 | then: */ | |
1941 | if($token['name'] === $node->tagName) { | |
1942 | /* Generate implied end tags. */ | |
1943 | $this->generateImpliedEndTags(); | |
1944 | ||
1945 | /* If the tag name of the end tag token does not | |
1946 | match the tag name of the current node, this is a | |
1947 | parse error. */ | |
1948 | // XERROR: implement this | |
1949 | ||
1950 | /* Pop all the nodes from the current node up to | |
1951 | node, including node, then stop these steps. */ | |
1952 | // XSKETCHY | |
1953 | do { | |
1954 | $pop = array_pop($this->stack); | |
1955 | } while ($pop !== $node); | |
1956 | break; | |
1957 | ||
1958 | } else { | |
1959 | $category = $this->getElementCategory($node); | |
1960 | ||
1961 | if($category !== self::FORMATTING && $category !== self::PHRASING) { | |
1962 | /* Otherwise, if node is in neither the formatting | |
1963 | category nor the phrasing category, then this is a | |
1964 | parse error. Stop this algorithm. The end tag token | |
1965 | is ignored. */ | |
1966 | $this->ignored = true; | |
1967 | break; | |
1968 | // parse error | |
1969 | } | |
1970 | } | |
1971 | /* Set node to the previous entry in the stack of open elements. Loop. */ | |
1972 | } | |
1973 | break; | |
1974 | } | |
1975 | break; | |
1976 | } | |
1977 | break; | |
1978 | ||
1979 | case self::IN_CDATA_RCDATA: | |
1980 | if ( | |
1981 | $token['type'] === HTML5_Tokenizer::CHARACTER || | |
1982 | $token['type'] === HTML5_Tokenizer::SPACECHARACTER | |
1983 | ) { | |
1984 | $this->insertText($token['data']); | |
1985 | } elseif ($token['type'] === HTML5_Tokenizer::EOF) { | |
1986 | // parse error | |
1987 | /* If the current node is a script element, mark the script | |
1988 | * element as "already executed". */ | |
1989 | // probably not necessary | |
1990 | array_pop($this->stack); | |
1991 | $this->mode = $this->original_mode; | |
1992 | $this->emitToken($token); | |
1993 | } elseif ($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'script') { | |
1994 | array_pop($this->stack); | |
1995 | $this->mode = $this->original_mode; | |
1996 | // we're ignoring all of the execution stuff | |
1997 | } elseif ($token['type'] === HTML5_Tokenizer::ENDTAG) { | |
1998 | array_pop($this->stack); | |
1999 | $this->mode = $this->original_mode; | |
2000 | } | |
2001 | break; | |
2002 | ||
2003 | case self::IN_TABLE: | |
2004 | $clear = array('html', 'table'); | |
2005 | ||
2006 | /* A character token */ | |
2007 | if ($token['type'] === HTML5_Tokenizer::CHARACTER || | |
2008 | $token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
2009 | /* Let the pending table character tokens | |
2010 | * be an empty list of tokens. */ | |
2011 | $this->pendingTableCharacters = ""; | |
2012 | $this->pendingTableCharactersDirty = false; | |
2013 | /* Let the original insertion mode be the current | |
2014 | * insertion mode. */ | |
2015 | $this->original_mode = $this->mode; | |
2016 | /* Switch the insertion mode to | |
2017 | * "in table text" and | |
2018 | * reprocess the token. */ | |
2019 | $this->mode = self::IN_TABLE_TEXT; | |
2020 | $this->emitToken($token); | |
2021 | ||
2022 | /* A comment token */ | |
2023 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
2024 | /* Append a Comment node to the current node with the data | |
2025 | attribute set to the data given in the comment token. */ | |
2026 | $this->insertComment($token['data']); | |
2027 | ||
2028 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
2029 | // parse error | |
2030 | ||
2031 | /* A start tag whose tag name is "caption" */ | |
2032 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2033 | $token['name'] === 'caption') { | |
2034 | /* Clear the stack back to a table context. */ | |
2035 | $this->clearStackToTableContext($clear); | |
2036 | ||
2037 | /* Insert a marker at the end of the list of active | |
2038 | formatting elements. */ | |
2039 | $this->a_formatting[] = self::MARKER; | |
2040 | ||
2041 | /* Insert an HTML element for the token, then switch the | |
2042 | insertion mode to "in caption". */ | |
2043 | $this->insertElement($token); | |
2044 | $this->mode = self::IN_CAPTION; | |
2045 | ||
2046 | /* A start tag whose tag name is "colgroup" */ | |
2047 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2048 | $token['name'] === 'colgroup') { | |
2049 | /* Clear the stack back to a table context. */ | |
2050 | $this->clearStackToTableContext($clear); | |
2051 | ||
2052 | /* Insert an HTML element for the token, then switch the | |
2053 | insertion mode to "in column group". */ | |
2054 | $this->insertElement($token); | |
2055 | $this->mode = self::IN_COLUMN_GROUP; | |
2056 | ||
2057 | /* A start tag whose tag name is "col" */ | |
2058 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2059 | $token['name'] === 'col') { | |
2060 | $this->emitToken(array( | |
2061 | 'name' => 'colgroup', | |
2062 | 'type' => HTML5_Tokenizer::STARTTAG, | |
2063 | 'attr' => array() | |
2064 | )); | |
2065 | ||
2066 | $this->emitToken($token); | |
2067 | ||
2068 | /* A start tag whose tag name is one of: "tbody", "tfoot", "thead" */ | |
2069 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && in_array($token['name'], | |
2070 | array('tbody', 'tfoot', 'thead'))) { | |
2071 | /* Clear the stack back to a table context. */ | |
2072 | $this->clearStackToTableContext($clear); | |
2073 | ||
2074 | /* Insert an HTML element for the token, then switch the insertion | |
2075 | mode to "in table body". */ | |
2076 | $this->insertElement($token); | |
2077 | $this->mode = self::IN_TABLE_BODY; | |
2078 | ||
2079 | /* A start tag whose tag name is one of: "td", "th", "tr" */ | |
2080 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2081 | in_array($token['name'], array('td', 'th', 'tr'))) { | |
2082 | /* Act as if a start tag token with the tag name "tbody" had been | |
2083 | seen, then reprocess the current token. */ | |
2084 | $this->emitToken(array( | |
2085 | 'name' => 'tbody', | |
2086 | 'type' => HTML5_Tokenizer::STARTTAG, | |
2087 | 'attr' => array() | |
2088 | )); | |
2089 | ||
2090 | $this->emitToken($token); | |
2091 | ||
2092 | /* A start tag whose tag name is "table" */ | |
2093 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2094 | $token['name'] === 'table') { | |
2095 | /* Parse error. Act as if an end tag token with the tag name "table" | |
2096 | had been seen, then, if that token wasn't ignored, reprocess the | |
2097 | current token. */ | |
2098 | $this->emitToken(array( | |
2099 | 'name' => 'table', | |
2100 | 'type' => HTML5_Tokenizer::ENDTAG | |
2101 | )); | |
2102 | ||
2103 | if (!$this->ignored) $this->emitToken($token); | |
2104 | ||
2105 | /* An end tag whose tag name is "table" */ | |
2106 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2107 | $token['name'] === 'table') { | |
2108 | /* If the stack of open elements does not have an element in table | |
2109 | scope with the same tag name as the token, this is a parse error. | |
2110 | Ignore the token. (fragment case) */ | |
2111 | if(!$this->elementInScope($token['name'], self::SCOPE_TABLE)) { | |
2112 | $this->ignored = true; | |
2113 | ||
2114 | /* Otherwise: */ | |
2115 | } else { | |
2116 | do { | |
2117 | $node = array_pop($this->stack); | |
2118 | } while ($node->tagName !== 'table'); | |
2119 | ||
2120 | /* Reset the insertion mode appropriately. */ | |
2121 | $this->resetInsertionMode(); | |
2122 | } | |
2123 | ||
2124 | /* An end tag whose tag name is one of: "body", "caption", "col", | |
2125 | "colgroup", "html", "tbody", "td", "tfoot", "th", "thead", "tr" */ | |
2126 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && in_array($token['name'], | |
2127 | array('body', 'caption', 'col', 'colgroup', 'html', 'tbody', 'td', | |
2128 | 'tfoot', 'th', 'thead', 'tr'))) { | |
2129 | // Parse error. Ignore the token. | |
2130 | ||
2131 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2132 | ($token['name'] === 'style' || $token['name'] === 'script')) { | |
2133 | $this->processWithRulesFor($token, self::IN_HEAD); | |
2134 | ||
2135 | } elseif ($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'input' && | |
2136 | // assignment is intentional | |
2137 | /* If the token does not have an attribute with the name "type", or | |
2138 | * if it does, but that attribute's value is not an ASCII | |
2139 | * case-insensitive match for the string "hidden", then: act as | |
2140 | * described in the "anything else" entry below. */ | |
2141 | ($type = $this->getAttr($token, 'type')) && strtolower($type) === 'hidden') { | |
2142 | // I.e., if its an input with the type attribute == 'hidden' | |
2143 | /* Otherwise */ | |
2144 | // parse error | |
2145 | $this->insertElement($token); | |
2146 | array_pop($this->stack); | |
2147 | } elseif ($token['type'] === HTML5_Tokenizer::EOF) { | |
2148 | /* If the current node is not the root html element, then this is a parse error. */ | |
2149 | if (end($this->stack)->tagName !== 'html') { | |
2150 | // Note: It can only be the current node in the fragment case. | |
2151 | // parse error | |
2152 | } | |
2153 | /* Stop parsing. */ | |
2154 | /* Anything else */ | |
2155 | } else { | |
2156 | /* Parse error. Process the token as if the insertion mode was "in | |
2157 | body", with the following exception: */ | |
2158 | ||
2159 | $old = $this->foster_parent; | |
2160 | $this->foster_parent = true; | |
2161 | $this->processWithRulesFor($token, self::IN_BODY); | |
2162 | $this->foster_parent = $old; | |
2163 | } | |
2164 | break; | |
2165 | ||
2166 | case self::IN_TABLE_TEXT: | |
2167 | /* A character token */ | |
2168 | if($token['type'] === HTML5_Tokenizer::CHARACTER) { | |
2169 | /* Append the character token to the pending table | |
2170 | * character tokens list. */ | |
2171 | $this->pendingTableCharacters .= $token['data']; | |
2172 | $this->pendingTableCharactersDirty = true; | |
2173 | } elseif ($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
2174 | $this->pendingTableCharacters .= $token['data']; | |
2175 | /* Anything else */ | |
2176 | } else { | |
2177 | if ($this->pendingTableCharacters !== '' && is_string($this->pendingTableCharacters)) { | |
2178 | /* If any of the tokens in the pending table character tokens list | |
2179 | * are character tokens that are not one of U+0009 CHARACTER | |
2180 | * TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF), or | |
2181 | * U+0020 SPACE, then reprocess those character tokens using the | |
2182 | * rules given in the "anything else" entry in the in table" | |
2183 | * insertion mode.*/ | |
2184 | if ($this->pendingTableCharactersDirty) { | |
2185 | /* Parse error. Process the token using the rules for the | |
2186 | * "in body" insertion mode, except that if the current | |
2187 | * node is a table, tbody, tfoot, thead, or tr element, | |
2188 | * then, whenever a node would be inserted into the current | |
2189 | * node, it must instead be foster parented. */ | |
2190 | // XERROR | |
2191 | $old = $this->foster_parent; | |
2192 | $this->foster_parent = true; | |
2193 | $text_token = array( | |
2194 | 'type' => HTML5_Tokenizer::CHARACTER, | |
2195 | 'data' => $this->pendingTableCharacters, | |
2196 | ); | |
2197 | $this->processWithRulesFor($text_token, self::IN_BODY); | |
2198 | $this->foster_parent = $old; | |
2199 | ||
2200 | /* Otherwise, insert the characters given by the pending table | |
2201 | * character tokens list into the current node. */ | |
2202 | } else { | |
2203 | $this->insertText($this->pendingTableCharacters); | |
2204 | } | |
2205 | $this->pendingTableCharacters = null; | |
2206 | $this->pendingTableCharactersNull = null; | |
2207 | } | |
2208 | ||
2209 | /* Switch the insertion mode to the original insertion mode and | |
2210 | * reprocess the token. | |
2211 | */ | |
2212 | $this->mode = $this->original_mode; | |
2213 | $this->emitToken($token); | |
2214 | } | |
2215 | break; | |
2216 | ||
2217 | case self::IN_CAPTION: | |
2218 | /* An end tag whose tag name is "caption" */ | |
2219 | if($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'caption') { | |
2220 | /* If the stack of open elements does not have an element in table | |
2221 | scope with the same tag name as the token, this is a parse error. | |
2222 | Ignore the token. (fragment case) */ | |
2223 | if(!$this->elementInScope($token['name'], self::SCOPE_TABLE)) { | |
2224 | $this->ignored = true; | |
2225 | // Ignore | |
2226 | ||
2227 | /* Otherwise: */ | |
2228 | } else { | |
2229 | /* Generate implied end tags. */ | |
2230 | $this->generateImpliedEndTags(); | |
2231 | ||
2232 | /* Now, if the current node is not a caption element, then this | |
2233 | is a parse error. */ | |
2234 | // XERROR: implement | |
2235 | ||
2236 | /* Pop elements from this stack until a caption element has | |
2237 | been popped from the stack. */ | |
2238 | do { | |
2239 | $node = array_pop($this->stack); | |
2240 | } while ($node->tagName !== 'caption'); | |
2241 | ||
2242 | /* Clear the list of active formatting elements up to the last | |
2243 | marker. */ | |
2244 | $this->clearTheActiveFormattingElementsUpToTheLastMarker(); | |
2245 | ||
2246 | /* Switch the insertion mode to "in table". */ | |
2247 | $this->mode = self::IN_TABLE; | |
2248 | } | |
2249 | ||
2250 | /* A start tag whose tag name is one of: "caption", "col", "colgroup", | |
2251 | "tbody", "td", "tfoot", "th", "thead", "tr", or an end tag whose tag | |
2252 | name is "table" */ | |
2253 | } elseif(($token['type'] === HTML5_Tokenizer::STARTTAG && in_array($token['name'], | |
2254 | array('caption', 'col', 'colgroup', 'tbody', 'td', 'tfoot', 'th', | |
2255 | 'thead', 'tr'))) || ($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2256 | $token['name'] === 'table')) { | |
2257 | /* Parse error. Act as if an end tag with the tag name "caption" | |
2258 | had been seen, then, if that token wasn't ignored, reprocess the | |
2259 | current token. */ | |
2260 | $this->emitToken(array( | |
2261 | 'name' => 'caption', | |
2262 | 'type' => HTML5_Tokenizer::ENDTAG | |
2263 | )); | |
2264 | ||
2265 | if (!$this->ignored) $this->emitToken($token); | |
2266 | ||
2267 | /* An end tag whose tag name is one of: "body", "col", "colgroup", | |
2268 | "html", "tbody", "td", "tfoot", "th", "thead", "tr" */ | |
2269 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && in_array($token['name'], | |
2270 | array('body', 'col', 'colgroup', 'html', 'tbody', 'tfoot', 'th', | |
2271 | 'thead', 'tr'))) { | |
2272 | // Parse error. Ignore the token. | |
2273 | $this->ignored = true; | |
2274 | ||
2275 | /* Anything else */ | |
2276 | } else { | |
2277 | /* Process the token as if the insertion mode was "in body". */ | |
2278 | $this->processWithRulesFor($token, self::IN_BODY); | |
2279 | } | |
2280 | break; | |
2281 | ||
2282 | case self::IN_COLUMN_GROUP: | |
2283 | /* A character token that is one of one of U+0009 CHARACTER TABULATION, | |
2284 | U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), | |
2285 | or U+0020 SPACE */ | |
2286 | if($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
2287 | /* Append the character to the current node. */ | |
2288 | $this->insertText($token['data']); | |
2289 | ||
2290 | /* A comment token */ | |
2291 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
2292 | /* Append a Comment node to the current node with the data | |
2293 | attribute set to the data given in the comment token. */ | |
2294 | $this->insertToken($token['data']); | |
2295 | ||
2296 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
2297 | // parse error | |
2298 | ||
2299 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html') { | |
2300 | $this->processWithRulesFor($token, self::IN_BODY); | |
2301 | ||
2302 | /* A start tag whose tag name is "col" */ | |
2303 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'col') { | |
2304 | /* Insert a col element for the token. Immediately pop the current | |
2305 | node off the stack of open elements. */ | |
2306 | $this->insertElement($token); | |
2307 | array_pop($this->stack); | |
2308 | // XERROR: Acknowledge the token's self-closing flag, if it is set. | |
2309 | ||
2310 | /* An end tag whose tag name is "colgroup" */ | |
2311 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2312 | $token['name'] === 'colgroup') { | |
2313 | /* If the current node is the root html element, then this is a | |
2314 | parse error, ignore the token. (fragment case) */ | |
2315 | if(end($this->stack)->tagName === 'html') { | |
2316 | $this->ignored = true; | |
2317 | ||
2318 | /* Otherwise, pop the current node (which will be a colgroup | |
2319 | element) from the stack of open elements. Switch the insertion | |
2320 | mode to "in table". */ | |
2321 | } else { | |
2322 | array_pop($this->stack); | |
2323 | $this->mode = self::IN_TABLE; | |
2324 | } | |
2325 | ||
2326 | /* An end tag whose tag name is "col" */ | |
2327 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'col') { | |
2328 | /* Parse error. Ignore the token. */ | |
2329 | $this->ignored = true; | |
2330 | ||
2331 | /* An end-of-file token */ | |
2332 | /* If the current node is the root html element */ | |
2333 | } elseif($token['type'] === HTML5_Tokenizer::EOF && end($this->stack)->tagName === 'html') { | |
2334 | /* Stop parsing */ | |
2335 | ||
2336 | /* Anything else */ | |
2337 | } else { | |
2338 | /* Act as if an end tag with the tag name "colgroup" had been seen, | |
2339 | and then, if that token wasn't ignored, reprocess the current token. */ | |
2340 | $this->emitToken(array( | |
2341 | 'name' => 'colgroup', | |
2342 | 'type' => HTML5_Tokenizer::ENDTAG | |
2343 | )); | |
2344 | ||
2345 | if (!$this->ignored) $this->emitToken($token); | |
2346 | } | |
2347 | break; | |
2348 | ||
2349 | case self::IN_TABLE_BODY: | |
2350 | $clear = array('tbody', 'tfoot', 'thead', 'html'); | |
2351 | ||
2352 | /* A start tag whose tag name is "tr" */ | |
2353 | if($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'tr') { | |
2354 | /* Clear the stack back to a table body context. */ | |
2355 | $this->clearStackToTableContext($clear); | |
2356 | ||
2357 | /* Insert a tr element for the token, then switch the insertion | |
2358 | mode to "in row". */ | |
2359 | $this->insertElement($token); | |
2360 | $this->mode = self::IN_ROW; | |
2361 | ||
2362 | /* A start tag whose tag name is one of: "th", "td" */ | |
2363 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2364 | ($token['name'] === 'th' || $token['name'] === 'td')) { | |
2365 | /* Parse error. Act as if a start tag with the tag name "tr" had | |
2366 | been seen, then reprocess the current token. */ | |
2367 | $this->emitToken(array( | |
2368 | 'name' => 'tr', | |
2369 | 'type' => HTML5_Tokenizer::STARTTAG, | |
2370 | 'attr' => array() | |
2371 | )); | |
2372 | ||
2373 | $this->emitToken($token); | |
2374 | ||
2375 | /* An end tag whose tag name is one of: "tbody", "tfoot", "thead" */ | |
2376 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2377 | in_array($token['name'], array('tbody', 'tfoot', 'thead'))) { | |
2378 | /* If the stack of open elements does not have an element in table | |
2379 | scope with the same tag name as the token, this is a parse error. | |
2380 | Ignore the token. */ | |
2381 | if(!$this->elementInScope($token['name'], self::SCOPE_TABLE)) { | |
2382 | // Parse error | |
2383 | $this->ignored = true; | |
2384 | ||
2385 | /* Otherwise: */ | |
2386 | } else { | |
2387 | /* Clear the stack back to a table body context. */ | |
2388 | $this->clearStackToTableContext($clear); | |
2389 | ||
2390 | /* Pop the current node from the stack of open elements. Switch | |
2391 | the insertion mode to "in table". */ | |
2392 | array_pop($this->stack); | |
2393 | $this->mode = self::IN_TABLE; | |
2394 | } | |
2395 | ||
2396 | /* A start tag whose tag name is one of: "caption", "col", "colgroup", | |
2397 | "tbody", "tfoot", "thead", or an end tag whose tag name is "table" */ | |
2398 | } elseif(($token['type'] === HTML5_Tokenizer::STARTTAG && in_array($token['name'], | |
2399 | array('caption', 'col', 'colgroup', 'tbody', 'tfoot', 'thead'))) || | |
2400 | ($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'table')) { | |
2401 | /* If the stack of open elements does not have a tbody, thead, or | |
2402 | tfoot element in table scope, this is a parse error. Ignore the | |
2403 | token. (fragment case) */ | |
2404 | if(!$this->elementInScope(array('tbody', 'thead', 'tfoot'), self::SCOPE_TABLE)) { | |
2405 | // parse error | |
2406 | $this->ignored = true; | |
2407 | ||
2408 | /* Otherwise: */ | |
2409 | } else { | |
2410 | /* Clear the stack back to a table body context. */ | |
2411 | $this->clearStackToTableContext($clear); | |
2412 | ||
2413 | /* Act as if an end tag with the same tag name as the current | |
2414 | node ("tbody", "tfoot", or "thead") had been seen, then | |
2415 | reprocess the current token. */ | |
2416 | $this->emitToken(array( | |
2417 | 'name' => end($this->stack)->tagName, | |
2418 | 'type' => HTML5_Tokenizer::ENDTAG | |
2419 | )); | |
2420 | ||
2421 | $this->emitToken($token); | |
2422 | } | |
2423 | ||
2424 | /* An end tag whose tag name is one of: "body", "caption", "col", | |
2425 | "colgroup", "html", "td", "th", "tr" */ | |
2426 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && in_array($token['name'], | |
2427 | array('body', 'caption', 'col', 'colgroup', 'html', 'td', 'th', 'tr'))) { | |
2428 | /* Parse error. Ignore the token. */ | |
2429 | $this->ignored = true; | |
2430 | ||
2431 | /* Anything else */ | |
2432 | } else { | |
2433 | /* Process the token as if the insertion mode was "in table". */ | |
2434 | $this->processWithRulesFor($token, self::IN_TABLE); | |
2435 | } | |
2436 | break; | |
2437 | ||
2438 | case self::IN_ROW: | |
2439 | $clear = array('tr', 'html'); | |
2440 | ||
2441 | /* A start tag whose tag name is one of: "th", "td" */ | |
2442 | if($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2443 | ($token['name'] === 'th' || $token['name'] === 'td')) { | |
2444 | /* Clear the stack back to a table row context. */ | |
2445 | $this->clearStackToTableContext($clear); | |
2446 | ||
2447 | /* Insert an HTML element for the token, then switch the insertion | |
2448 | mode to "in cell". */ | |
2449 | $this->insertElement($token); | |
2450 | $this->mode = self::IN_CELL; | |
2451 | ||
2452 | /* Insert a marker at the end of the list of active formatting | |
2453 | elements. */ | |
2454 | $this->a_formatting[] = self::MARKER; | |
2455 | ||
2456 | /* An end tag whose tag name is "tr" */ | |
2457 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'tr') { | |
2458 | /* If the stack of open elements does not have an element in table | |
2459 | scope with the same tag name as the token, this is a parse error. | |
2460 | Ignore the token. (fragment case) */ | |
2461 | if(!$this->elementInScope($token['name'], self::SCOPE_TABLE)) { | |
2462 | // Ignore. | |
2463 | $this->ignored = true; | |
2464 | ||
2465 | /* Otherwise: */ | |
2466 | } else { | |
2467 | /* Clear the stack back to a table row context. */ | |
2468 | $this->clearStackToTableContext($clear); | |
2469 | ||
2470 | /* Pop the current node (which will be a tr element) from the | |
2471 | stack of open elements. Switch the insertion mode to "in table | |
2472 | body". */ | |
2473 | array_pop($this->stack); | |
2474 | $this->mode = self::IN_TABLE_BODY; | |
2475 | } | |
2476 | ||
2477 | /* A start tag whose tag name is one of: "caption", "col", "colgroup", | |
2478 | "tbody", "tfoot", "thead", "tr" or an end tag whose tag name is "table" */ | |
2479 | } elseif(($token['type'] === HTML5_Tokenizer::STARTTAG && in_array($token['name'], | |
2480 | array('caption', 'col', 'colgroup', 'tbody', 'tfoot', 'thead', 'tr'))) || | |
2481 | ($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'table')) { | |
2482 | /* Act as if an end tag with the tag name "tr" had been seen, then, | |
2483 | if that token wasn't ignored, reprocess the current token. */ | |
2484 | $this->emitToken(array( | |
2485 | 'name' => 'tr', | |
2486 | 'type' => HTML5_Tokenizer::ENDTAG | |
2487 | )); | |
2488 | if (!$this->ignored) $this->emitToken($token); | |
2489 | ||
2490 | /* An end tag whose tag name is one of: "tbody", "tfoot", "thead" */ | |
2491 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2492 | in_array($token['name'], array('tbody', 'tfoot', 'thead'))) { | |
2493 | /* If the stack of open elements does not have an element in table | |
2494 | scope with the same tag name as the token, this is a parse error. | |
2495 | Ignore the token. */ | |
2496 | if(!$this->elementInScope($token['name'], self::SCOPE_TABLE)) { | |
2497 | $this->ignored = true; | |
2498 | ||
2499 | /* Otherwise: */ | |
2500 | } else { | |
2501 | /* Otherwise, act as if an end tag with the tag name "tr" had | |
2502 | been seen, then reprocess the current token. */ | |
2503 | $this->emitToken(array( | |
2504 | 'name' => 'tr', | |
2505 | 'type' => HTML5_Tokenizer::ENDTAG | |
2506 | )); | |
2507 | ||
2508 | $this->emitToken($token); | |
2509 | } | |
2510 | ||
2511 | /* An end tag whose tag name is one of: "body", "caption", "col", | |
2512 | "colgroup", "html", "td", "th" */ | |
2513 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && in_array($token['name'], | |
2514 | array('body', 'caption', 'col', 'colgroup', 'html', 'td', 'th'))) { | |
2515 | /* Parse error. Ignore the token. */ | |
2516 | $this->ignored = true; | |
2517 | ||
2518 | /* Anything else */ | |
2519 | } else { | |
2520 | /* Process the token as if the insertion mode was "in table". */ | |
2521 | $this->processWithRulesFor($token, self::IN_TABLE); | |
2522 | } | |
2523 | break; | |
2524 | ||
2525 | case self::IN_CELL: | |
2526 | /* An end tag whose tag name is one of: "td", "th" */ | |
2527 | if($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2528 | ($token['name'] === 'td' || $token['name'] === 'th')) { | |
2529 | /* If the stack of open elements does not have an element in table | |
2530 | scope with the same tag name as that of the token, then this is a | |
2531 | parse error and the token must be ignored. */ | |
2532 | if(!$this->elementInScope($token['name'], self::SCOPE_TABLE)) { | |
2533 | $this->ignored = true; | |
2534 | ||
2535 | /* Otherwise: */ | |
2536 | } else { | |
2537 | /* Generate implied end tags, except for elements with the same | |
2538 | tag name as the token. */ | |
2539 | $this->generateImpliedEndTags(array($token['name'])); | |
2540 | ||
2541 | /* Now, if the current node is not an element with the same tag | |
2542 | name as the token, then this is a parse error. */ | |
2543 | // XERROR: Implement parse error code | |
2544 | ||
2545 | /* Pop elements from this stack until an element with the same | |
2546 | tag name as the token has been popped from the stack. */ | |
2547 | do { | |
2548 | $node = array_pop($this->stack); | |
2549 | } while ($node->tagName !== $token['name']); | |
2550 | ||
2551 | /* Clear the list of active formatting elements up to the last | |
2552 | marker. */ | |
2553 | $this->clearTheActiveFormattingElementsUpToTheLastMarker(); | |
2554 | ||
2555 | /* Switch the insertion mode to "in row". (The current node | |
2556 | will be a tr element at this point.) */ | |
2557 | $this->mode = self::IN_ROW; | |
2558 | } | |
2559 | ||
2560 | /* A start tag whose tag name is one of: "caption", "col", "colgroup", | |
2561 | "tbody", "td", "tfoot", "th", "thead", "tr" */ | |
2562 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && in_array($token['name'], | |
2563 | array('caption', 'col', 'colgroup', 'tbody', 'td', 'tfoot', 'th', | |
2564 | 'thead', 'tr'))) { | |
2565 | /* If the stack of open elements does not have a td or th element | |
2566 | in table scope, then this is a parse error; ignore the token. | |
2567 | (fragment case) */ | |
2568 | if(!$this->elementInScope(array('td', 'th'), self::SCOPE_TABLE)) { | |
2569 | // parse error | |
2570 | $this->ignored = true; | |
2571 | ||
2572 | /* Otherwise, close the cell (see below) and reprocess the current | |
2573 | token. */ | |
2574 | } else { | |
2575 | $this->closeCell(); | |
2576 | $this->emitToken($token); | |
2577 | } | |
2578 | ||
2579 | /* An end tag whose tag name is one of: "body", "caption", "col", | |
2580 | "colgroup", "html" */ | |
2581 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && in_array($token['name'], | |
2582 | array('body', 'caption', 'col', 'colgroup', 'html'))) { | |
2583 | /* Parse error. Ignore the token. */ | |
2584 | $this->ignored = true; | |
2585 | ||
2586 | /* An end tag whose tag name is one of: "table", "tbody", "tfoot", | |
2587 | "thead", "tr" */ | |
2588 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && in_array($token['name'], | |
2589 | array('table', 'tbody', 'tfoot', 'thead', 'tr'))) { | |
2590 | /* If the stack of open elements does not have a td or th element | |
2591 | in table scope, then this is a parse error; ignore the token. | |
2592 | (innerHTML case) */ | |
2593 | if(!$this->elementInScope(array('td', 'th'), self::SCOPE_TABLE)) { | |
2594 | // Parse error | |
2595 | $this->ignored = true; | |
2596 | ||
2597 | /* Otherwise, close the cell (see below) and reprocess the current | |
2598 | token. */ | |
2599 | } else { | |
2600 | $this->closeCell(); | |
2601 | $this->emitToken($token); | |
2602 | } | |
2603 | ||
2604 | /* Anything else */ | |
2605 | } else { | |
2606 | /* Process the token as if the insertion mode was "in body". */ | |
2607 | $this->processWithRulesFor($token, self::IN_BODY); | |
2608 | } | |
2609 | break; | |
2610 | ||
2611 | case self::IN_SELECT: | |
2612 | /* Handle the token as follows: */ | |
2613 | ||
2614 | /* A character token */ | |
2615 | if( | |
2616 | $token['type'] === HTML5_Tokenizer::CHARACTER || | |
2617 | $token['type'] === HTML5_Tokenizer::SPACECHARACTER | |
2618 | ) { | |
2619 | /* Append the token's character to the current node. */ | |
2620 | $this->insertText($token['data']); | |
2621 | ||
2622 | /* A comment token */ | |
2623 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
2624 | /* Append a Comment node to the current node with the data | |
2625 | attribute set to the data given in the comment token. */ | |
2626 | $this->insertComment($token['data']); | |
2627 | ||
2628 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
2629 | // parse error | |
2630 | ||
2631 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html') { | |
2632 | $this->processWithRulesFor($token, self::INBODY); | |
2633 | ||
2634 | /* A start tag token whose tag name is "option" */ | |
2635 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2636 | $token['name'] === 'option') { | |
2637 | /* If the current node is an option element, act as if an end tag | |
2638 | with the tag name "option" had been seen. */ | |
2639 | if(end($this->stack)->tagName === 'option') { | |
2640 | $this->emitToken(array( | |
2641 | 'name' => 'option', | |
2642 | 'type' => HTML5_Tokenizer::ENDTAG | |
2643 | )); | |
2644 | } | |
2645 | ||
2646 | /* Insert an HTML element for the token. */ | |
2647 | $this->insertElement($token); | |
2648 | ||
2649 | /* A start tag token whose tag name is "optgroup" */ | |
2650 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2651 | $token['name'] === 'optgroup') { | |
2652 | /* If the current node is an option element, act as if an end tag | |
2653 | with the tag name "option" had been seen. */ | |
2654 | if(end($this->stack)->tagName === 'option') { | |
2655 | $this->emitToken(array( | |
2656 | 'name' => 'option', | |
2657 | 'type' => HTML5_Tokenizer::ENDTAG | |
2658 | )); | |
2659 | } | |
2660 | ||
2661 | /* If the current node is an optgroup element, act as if an end tag | |
2662 | with the tag name "optgroup" had been seen. */ | |
2663 | if(end($this->stack)->tagName === 'optgroup') { | |
2664 | $this->emitToken(array( | |
2665 | 'name' => 'optgroup', | |
2666 | 'type' => HTML5_Tokenizer::ENDTAG | |
2667 | )); | |
2668 | } | |
2669 | ||
2670 | /* Insert an HTML element for the token. */ | |
2671 | $this->insertElement($token); | |
2672 | ||
2673 | /* An end tag token whose tag name is "optgroup" */ | |
2674 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2675 | $token['name'] === 'optgroup') { | |
2676 | /* First, if the current node is an option element, and the node | |
2677 | immediately before it in the stack of open elements is an optgroup | |
2678 | element, then act as if an end tag with the tag name "option" had | |
2679 | been seen. */ | |
2680 | $elements_in_stack = count($this->stack); | |
2681 | ||
2682 | if($this->stack[$elements_in_stack - 1]->tagName === 'option' && | |
2683 | $this->stack[$elements_in_stack - 2]->tagName === 'optgroup') { | |
2684 | $this->emitToken(array( | |
2685 | 'name' => 'option', | |
2686 | 'type' => HTML5_Tokenizer::ENDTAG | |
2687 | )); | |
2688 | } | |
2689 | ||
2690 | /* If the current node is an optgroup element, then pop that node | |
2691 | from the stack of open elements. Otherwise, this is a parse error, | |
2692 | ignore the token. */ | |
2693 | if(end($this->stack)->tagName === 'optgroup') { | |
2694 | array_pop($this->stack); | |
2695 | } else { | |
2696 | // parse error | |
2697 | $this->ignored = true; | |
2698 | } | |
2699 | ||
2700 | /* An end tag token whose tag name is "option" */ | |
2701 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2702 | $token['name'] === 'option') { | |
2703 | /* If the current node is an option element, then pop that node | |
2704 | from the stack of open elements. Otherwise, this is a parse error, | |
2705 | ignore the token. */ | |
2706 | if(end($this->stack)->tagName === 'option') { | |
2707 | array_pop($this->stack); | |
2708 | } else { | |
2709 | // parse error | |
2710 | $this->ignored = true; | |
2711 | } | |
2712 | ||
2713 | /* An end tag whose tag name is "select" */ | |
2714 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2715 | $token['name'] === 'select') { | |
2716 | /* If the stack of open elements does not have an element in table | |
2717 | scope with the same tag name as the token, this is a parse error. | |
2718 | Ignore the token. (fragment case) */ | |
2719 | if(!$this->elementInScope($token['name'], self::SCOPE_TABLE)) { | |
2720 | $this->ignored = true; | |
2721 | // parse error | |
2722 | ||
2723 | /* Otherwise: */ | |
2724 | } else { | |
2725 | /* Pop elements from the stack of open elements until a select | |
2726 | element has been popped from the stack. */ | |
2727 | do { | |
2728 | $node = array_pop($this->stack); | |
2729 | } while ($node->tagName !== 'select'); | |
2730 | ||
2731 | /* Reset the insertion mode appropriately. */ | |
2732 | $this->resetInsertionMode(); | |
2733 | } | |
2734 | ||
2735 | /* A start tag whose tag name is "select" */ | |
2736 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'select') { | |
2737 | /* Parse error. Act as if the token had been an end tag with the | |
2738 | tag name "select" instead. */ | |
2739 | $this->emitToken(array( | |
2740 | 'name' => 'select', | |
2741 | 'type' => HTML5_Tokenizer::ENDTAG | |
2742 | )); | |
2743 | ||
2744 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2745 | ($token['name'] === 'input' || $token['name'] === 'keygen' || $token['name'] === 'textarea')) { | |
2746 | // parse error | |
2747 | $this->emitToken(array( | |
2748 | 'name' => 'select', | |
2749 | 'type' => HTML5_Tokenizer::ENDTAG | |
2750 | )); | |
2751 | $this->emitToken($token); | |
2752 | ||
2753 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'script') { | |
2754 | $this->processWithRulesFor($token, self::IN_HEAD); | |
2755 | ||
2756 | } elseif($token['type'] === HTML5_Tokenizer::EOF) { | |
2757 | // XERROR: If the current node is not the root html element, then this is a parse error. | |
2758 | /* Stop parsing */ | |
2759 | ||
2760 | /* Anything else */ | |
2761 | } else { | |
2762 | /* Parse error. Ignore the token. */ | |
2763 | $this->ignored = true; | |
2764 | } | |
2765 | break; | |
2766 | ||
2767 | case self::IN_SELECT_IN_TABLE: | |
2768 | ||
2769 | if($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2770 | in_array($token['name'], array('caption', 'table', 'tbody', | |
2771 | 'tfoot', 'thead', 'tr', 'td', 'th'))) { | |
2772 | // parse error | |
2773 | $this->emitToken(array( | |
2774 | 'name' => 'select', | |
2775 | 'type' => HTML5_Tokenizer::ENDTAG, | |
2776 | )); | |
2777 | $this->emitToken($token); | |
2778 | ||
2779 | /* An end tag whose tag name is one of: "caption", "table", "tbody", | |
2780 | "tfoot", "thead", "tr", "td", "th" */ | |
2781 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2782 | in_array($token['name'], array('caption', 'table', 'tbody', 'tfoot', 'thead', 'tr', 'td', 'th'))) { | |
2783 | /* Parse error. */ | |
2784 | // parse error | |
2785 | ||
2786 | /* If the stack of open elements has an element in table scope with | |
2787 | the same tag name as that of the token, then act as if an end tag | |
2788 | with the tag name "select" had been seen, and reprocess the token. | |
2789 | Otherwise, ignore the token. */ | |
2790 | if($this->elementInScope($token['name'], self::SCOPE_TABLE)) { | |
2791 | $this->emitToken(array( | |
2792 | 'name' => 'select', | |
2793 | 'type' => HTML5_Tokenizer::ENDTAG | |
2794 | )); | |
2795 | ||
2796 | $this->emitToken($token); | |
2797 | } else { | |
2798 | $this->ignored = true; | |
2799 | } | |
2800 | } else { | |
2801 | $this->processWithRulesFor($token, self::IN_SELECT); | |
2802 | } | |
2803 | break; | |
2804 | ||
2805 | case self::IN_FOREIGN_CONTENT: | |
2806 | if ($token['type'] === HTML5_Tokenizer::CHARACTER || | |
2807 | $token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
2808 | $this->insertText($token['data']); | |
2809 | } elseif ($token['type'] === HTML5_Tokenizer::COMMENT) { | |
2810 | $this->insertComment($token['data']); | |
2811 | } elseif ($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
2812 | // XERROR: parse error | |
2813 | } elseif ($token['type'] === HTML5_Tokenizer::ENDTAG && | |
2814 | $token['name'] === 'script' && end($this->stack)->tagName === 'script' && | |
2815 | // XDOM | |
2816 | end($this->stack)->namespaceURI === self::NS_SVG) { | |
2817 | array_pop($this->stack); | |
2818 | // a bunch of script running mumbo jumbo | |
2819 | } elseif ( | |
2820 | ($token['type'] === HTML5_Tokenizer::STARTTAG && | |
2821 | (( | |
2822 | $token['name'] !== 'mglyph' && | |
2823 | $token['name'] !== 'malignmark' && | |
2824 | // XDOM | |
2825 | end($this->stack)->namespaceURI === self::NS_MATHML && | |
2826 | in_array(end($this->stack)->tagName, array('mi', 'mo', 'mn', 'ms', 'mtext')) | |
2827 | ) || | |
2828 | ( | |
2829 | $token['name'] === 'svg' && | |
2830 | // XDOM | |
2831 | end($this->stack)->namespaceURI === self::NS_MATHML && | |
2832 | end($this->stack)->tagName === 'annotation-xml' | |
2833 | ) || | |
2834 | ( | |
2835 | // XDOM | |
2836 | end($this->stack)->namespaceURI === self::NS_SVG && | |
2837 | in_array(end($this->stack)->tagName, array('foreignObject', 'desc', 'title')) | |
2838 | ) || | |
2839 | ( | |
2840 | // XSKETCHY && XDOM | |
2841 | end($this->stack)->namespaceURI === self::NS_HTML | |
2842 | )) | |
2843 | ) || $token['type'] === HTML5_Tokenizer::ENDTAG | |
2844 | ) { | |
2845 | $this->processWithRulesFor($token, $this->secondary_mode); | |
2846 | /* If, after doing so, the insertion mode is still "in foreign | |
2847 | * content", but there is no element in scope that has a namespace | |
2848 | * other than the HTML namespace, switch the insertion mode to the | |
2849 | * secondary insertion mode. */ | |
2850 | if ($this->mode === self::IN_FOREIGN_CONTENT) { | |
2851 | $found = false; | |
2852 | // this basically duplicates elementInScope() | |
2853 | for ($i = count($this->stack) - 1; $i >= 0; $i--) { | |
2854 | // XDOM | |
2855 | $node = $this->stack[$i]; | |
2856 | if ($node->namespaceURI !== self::NS_HTML) { | |
2857 | $found = true; | |
2858 | break; | |
2859 | } elseif (in_array($node->tagName, array('table', 'html', | |
2860 | 'applet', 'caption', 'td', 'th', 'button', 'marquee', | |
2861 | 'object')) || ($node->tagName === 'foreignObject' && | |
2862 | $node->namespaceURI === self::NS_SVG)) { | |
2863 | break; | |
2864 | } | |
2865 | } | |
2866 | if (!$found) { | |
2867 | $this->mode = $this->secondary_mode; | |
2868 | } | |
2869 | } | |
2870 | } elseif ($token['type'] === HTML5_Tokenizer::EOF || ( | |
2871 | $token['type'] === HTML5_Tokenizer::STARTTAG && | |
2872 | (in_array($token['name'], array('b', "big", "blockquote", "body", "br", | |
2873 | "center", "code", "dc", "dd", "div", "dl", "ds", "dt", "em", "embed", "h1", "h2", | |
2874 | "h3", "h4", "h5", "h6", "head", "hr", "i", "img", "li", "listing", | |
2875 | "menu", "meta", "nobr", "ol", "p", "pre", "ruby", "s", "small", | |
2876 | "span", "strong", "strike", "sub", "sup", "table", "tt", "u", "ul", | |
2877 | "var")) || ($token['name'] === 'font' && ($this->getAttr($token, 'color') || | |
2878 | $this->getAttr($token, 'face') || $this->getAttr($token, 'size')))))) { | |
2879 | // XERROR: parse error | |
2880 | do { | |
2881 | $node = array_pop($this->stack); | |
2882 | // XDOM | |
2883 | } while ($node->namespaceURI !== self::NS_HTML); | |
2884 | $this->stack[] = $node; | |
2885 | $this->mode = $this->secondary_mode; | |
2886 | $this->emitToken($token); | |
2887 | } elseif ($token['type'] === HTML5_Tokenizer::STARTTAG) { | |
2888 | static $svg_lookup = array( | |
2889 | 'altglyph' => 'altGlyph', | |
2890 | 'altglyphdef' => 'altGlyphDef', | |
2891 | 'altglyphitem' => 'altGlyphItem', | |
2892 | 'animatecolor' => 'animateColor', | |
2893 | 'animatemotion' => 'animateMotion', | |
2894 | 'animatetransform' => 'animateTransform', | |
2895 | 'clippath' => 'clipPath', | |
2896 | 'feblend' => 'feBlend', | |
2897 | 'fecolormatrix' => 'feColorMatrix', | |
2898 | 'fecomponenttransfer' => 'feComponentTransfer', | |
2899 | 'fecomposite' => 'feComposite', | |
2900 | 'feconvolvematrix' => 'feConvolveMatrix', | |
2901 | 'fediffuselighting' => 'feDiffuseLighting', | |
2902 | 'fedisplacementmap' => 'feDisplacementMap', | |
2903 | 'fedistantlight' => 'feDistantLight', | |
2904 | 'feflood' => 'feFlood', | |
2905 | 'fefunca' => 'feFuncA', | |
2906 | 'fefuncb' => 'feFuncB', | |
2907 | 'fefuncg' => 'feFuncG', | |
2908 | 'fefuncr' => 'feFuncR', | |
2909 | 'fegaussianblur' => 'feGaussianBlur', | |
2910 | 'feimage' => 'feImage', | |
2911 | 'femerge' => 'feMerge', | |
2912 | 'femergenode' => 'feMergeNode', | |
2913 | 'femorphology' => 'feMorphology', | |
2914 | 'feoffset' => 'feOffset', | |
2915 | 'fepointlight' => 'fePointLight', | |
2916 | 'fespecularlighting' => 'feSpecularLighting', | |
2917 | 'fespotlight' => 'feSpotLight', | |
2918 | 'fetile' => 'feTile', | |
2919 | 'feturbulence' => 'feTurbulence', | |
2920 | 'foreignobject' => 'foreignObject', | |
2921 | 'glyphref' => 'glyphRef', | |
2922 | 'lineargradient' => 'linearGradient', | |
2923 | 'radialgradient' => 'radialGradient', | |
2924 | 'textpath' => 'textPath', | |
2925 | ); | |
2926 | // XDOM | |
2927 | $current = end($this->stack); | |
2928 | if ($current->namespaceURI === self::NS_MATHML) { | |
2929 | $token = $this->adjustMathMLAttributes($token); | |
2930 | } | |
2931 | if ($current->namespaceURI === self::NS_SVG && | |
2932 | isset($svg_lookup[$token['name']])) { | |
2933 | $token['name'] = $svg_lookup[$token['name']]; | |
2934 | } | |
2935 | if ($current->namespaceURI === self::NS_SVG) { | |
2936 | $token = $this->adjustSVGAttributes($token); | |
2937 | } | |
2938 | $token = $this->adjustForeignAttributes($token); | |
2939 | $this->insertForeignElement($token, $current->namespaceURI); | |
2940 | if (isset($token['self-closing'])) { | |
2941 | array_pop($this->stack); | |
2942 | // XERROR: acknowledge self-closing flag | |
2943 | } | |
2944 | } | |
2945 | break; | |
2946 | ||
2947 | case self::AFTER_BODY: | |
2948 | /* Handle the token as follows: */ | |
2949 | ||
2950 | /* A character token that is one of one of U+0009 CHARACTER TABULATION, | |
2951 | U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), | |
2952 | or U+0020 SPACE */ | |
2953 | if($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
2954 | /* Process the token as it would be processed if the insertion mode | |
2955 | was "in body". */ | |
2956 | $this->processWithRulesFor($token, self::IN_BODY); | |
2957 | ||
2958 | /* A comment token */ | |
2959 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
2960 | /* Append a Comment node to the first element in the stack of open | |
2961 | elements (the html element), with the data attribute set to the | |
2962 | data given in the comment token. */ | |
2963 | // XDOM | |
2964 | $comment = $this->dom->createComment($token['data']); | |
2965 | $this->stack[0]->appendChild($comment); | |
2966 | ||
2967 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
2968 | // parse error | |
2969 | ||
2970 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html') { | |
2971 | $this->processWithRulesFor($token, self::IN_BODY); | |
2972 | ||
2973 | /* An end tag with the tag name "html" */ | |
2974 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && $token['name'] === 'html') { | |
2975 | /* If the parser was originally created as part of the HTML | |
2976 | * fragment parsing algorithm, this is a parse error; ignore | |
2977 | * the token. (fragment case) */ | |
2978 | $this->ignored = true; | |
2979 | // XERROR: implement this | |
2980 | ||
2981 | $this->mode = self::AFTER_AFTER_BODY; | |
2982 | ||
2983 | } elseif($token['type'] === HTML5_Tokenizer::EOF) { | |
2984 | /* Stop parsing */ | |
2985 | ||
2986 | /* Anything else */ | |
2987 | } else { | |
2988 | /* Parse error. Set the insertion mode to "in body" and reprocess | |
2989 | the token. */ | |
2990 | $this->mode = self::IN_BODY; | |
2991 | $this->emitToken($token); | |
2992 | } | |
2993 | break; | |
2994 | ||
2995 | case self::IN_FRAMESET: | |
2996 | /* Handle the token as follows: */ | |
2997 | ||
2998 | /* A character token that is one of one of U+0009 CHARACTER TABULATION, | |
2999 | U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), | |
3000 | U+000D CARRIAGE RETURN (CR), or U+0020 SPACE */ | |
3001 | if($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
3002 | /* Append the character to the current node. */ | |
3003 | $this->insertText($token['data']); | |
3004 | ||
3005 | /* A comment token */ | |
3006 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
3007 | /* Append a Comment node to the current node with the data | |
3008 | attribute set to the data given in the comment token. */ | |
3009 | $this->insertComment($token['data']); | |
3010 | ||
3011 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
3012 | // parse error | |
3013 | ||
3014 | /* A start tag with the tag name "frameset" */ | |
3015 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
3016 | $token['name'] === 'frameset') { | |
3017 | $this->insertElement($token); | |
3018 | ||
3019 | /* An end tag with the tag name "frameset" */ | |
3020 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
3021 | $token['name'] === 'frameset') { | |
3022 | /* If the current node is the root html element, then this is a | |
3023 | parse error; ignore the token. (fragment case) */ | |
3024 | if(end($this->stack)->tagName === 'html') { | |
3025 | $this->ignored = true; | |
3026 | // Parse error | |
3027 | ||
3028 | } else { | |
3029 | /* Otherwise, pop the current node from the stack of open | |
3030 | elements. */ | |
3031 | array_pop($this->stack); | |
3032 | ||
3033 | /* If the parser was not originally created as part of the HTML | |
3034 | * fragment parsing algorithm (fragment case), and the current | |
3035 | * node is no longer a frameset element, then switch the | |
3036 | * insertion mode to "after frameset". */ | |
3037 | $this->mode = self::AFTER_FRAMESET; | |
3038 | } | |
3039 | ||
3040 | /* A start tag with the tag name "frame" */ | |
3041 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
3042 | $token['name'] === 'frame') { | |
3043 | /* Insert an HTML element for the token. */ | |
3044 | $this->insertElement($token); | |
3045 | ||
3046 | /* Immediately pop the current node off the stack of open elements. */ | |
3047 | array_pop($this->stack); | |
3048 | ||
3049 | // XERROR: Acknowledge the token's self-closing flag, if it is set. | |
3050 | ||
3051 | /* A start tag with the tag name "noframes" */ | |
3052 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
3053 | $token['name'] === 'noframes') { | |
3054 | /* Process the token using the rules for the "in head" insertion mode. */ | |
3055 | $this->processwithRulesFor($token, self::IN_HEAD); | |
3056 | ||
3057 | } elseif($token['type'] === HTML5_Tokenizer::EOF) { | |
3058 | // XERROR: If the current node is not the root html element, then this is a parse error. | |
3059 | /* Stop parsing */ | |
3060 | /* Anything else */ | |
3061 | } else { | |
3062 | /* Parse error. Ignore the token. */ | |
3063 | $this->ignored = true; | |
3064 | } | |
3065 | break; | |
3066 | ||
3067 | case self::AFTER_FRAMESET: | |
3068 | /* Handle the token as follows: */ | |
3069 | ||
3070 | /* A character token that is one of one of U+0009 CHARACTER TABULATION, | |
3071 | U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), | |
3072 | U+000D CARRIAGE RETURN (CR), or U+0020 SPACE */ | |
3073 | if($token['type'] === HTML5_Tokenizer::SPACECHARACTER) { | |
3074 | /* Append the character to the current node. */ | |
3075 | $this->insertText($token['data']); | |
3076 | ||
3077 | /* A comment token */ | |
3078 | } elseif($token['type'] === HTML5_Tokenizer::COMMENT) { | |
3079 | /* Append a Comment node to the current node with the data | |
3080 | attribute set to the data given in the comment token. */ | |
3081 | $this->insertComment($token['data']); | |
3082 | ||
3083 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE) { | |
3084 | // parse error | |
3085 | ||
3086 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html') { | |
3087 | $this->processWithRulesFor($token, self::IN_BODY); | |
3088 | ||
3089 | /* An end tag with the tag name "html" */ | |
3090 | } elseif($token['type'] === HTML5_Tokenizer::ENDTAG && | |
3091 | $token['name'] === 'html') { | |
3092 | $this->mode = self::AFTER_AFTER_FRAMESET; | |
3093 | ||
3094 | /* A start tag with the tag name "noframes" */ | |
3095 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && | |
3096 | $token['name'] === 'noframes') { | |
3097 | $this->processWithRulesFor($token, self::IN_HEAD); | |
3098 | ||
3099 | } elseif($token['type'] === HTML5_Tokenizer::EOF) { | |
3100 | /* Stop parsing */ | |
3101 | ||
3102 | /* Anything else */ | |
3103 | } else { | |
3104 | /* Parse error. Ignore the token. */ | |
3105 | $this->ignored = true; | |
3106 | } | |
3107 | break; | |
3108 | ||
3109 | case self::AFTER_AFTER_BODY: | |
3110 | /* A comment token */ | |
3111 | if($token['type'] === HTML5_Tokenizer::COMMENT) { | |
3112 | /* Append a Comment node to the Document object with the data | |
3113 | attribute set to the data given in the comment token. */ | |
3114 | // XDOM | |
3115 | $comment = $this->dom->createComment($token['data']); | |
3116 | $this->dom->appendChild($comment); | |
3117 | ||
3118 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE || | |
3119 | $token['type'] === HTML5_Tokenizer::SPACECHARACTER || | |
3120 | ($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html')) { | |
3121 | $this->processWithRulesFor($token, self::IN_BODY); | |
3122 | ||
3123 | /* An end-of-file token */ | |
3124 | } elseif($token['type'] === HTML5_Tokenizer::EOF) { | |
3125 | /* OMG DONE!! */ | |
3126 | } else { | |
3127 | // parse error | |
3128 | $this->mode = self::IN_BODY; | |
3129 | $this->emitToken($token); | |
3130 | } | |
3131 | break; | |
3132 | ||
3133 | case self::AFTER_AFTER_FRAMESET: | |
3134 | /* A comment token */ | |
3135 | if($token['type'] === HTML5_Tokenizer::COMMENT) { | |
3136 | /* Append a Comment node to the Document object with the data | |
3137 | attribute set to the data given in the comment token. */ | |
3138 | // XDOM | |
3139 | $comment = $this->dom->createComment($token['data']); | |
3140 | $this->dom->appendChild($comment); | |
3141 | ||
3142 | } elseif($token['type'] === HTML5_Tokenizer::DOCTYPE || | |
3143 | $token['type'] === HTML5_Tokenizer::SPACECHARACTER || | |
3144 | ($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'html')) { | |
3145 | $this->processWithRulesFor($token, self::IN_BODY); | |
3146 | ||
3147 | /* An end-of-file token */ | |
3148 | } elseif($token['type'] === HTML5_Tokenizer::EOF) { | |
3149 | /* OMG DONE!! */ | |
3150 | } elseif($token['type'] === HTML5_Tokenizer::STARTTAG && $token['name'] === 'nofrmaes') { | |
3151 | $this->processWithRulesFor($token, self::IN_HEAD); | |
3152 | } else { | |
3153 | // parse error | |
3154 | } | |
3155 | break; | |
3156 | } | |
3157 | // end funky indenting | |
3158 | } | |
3159 | ||
3160 | private function insertElement($token, $append = true) { | |
3ec62cf9 MR |
3161 | //$el = $this->dom->createElementNS(self::NS_HTML, $token['name']); |
3162 | $namespaceURI = strpos($token['name'], ':') ? self::NS_XHTML : self::NS_HTML; | |
3163 | $el = $this->dom->createElementNS($namespaceURI, $token['name']); | |
42c80841 NL |
3164 | |
3165 | if (!empty($token['attr'])) { | |
3166 | foreach($token['attr'] as $attr) { | |
3ec62cf9 MR |
3167 | |
3168 | // mike@macgirvin.com 2011-11-17, check attribute name for | |
3169 | // validity (ignoring extenders and combiners) as illegal chars in names | |
3170 | // causes everything to abort | |
3171 | ||
3172 | $valid = preg_match('/^[a-zA-Z\_\:]([\-a-zA-Z0-9\_\:\.]+$)/',$attr['name']); | |
3173 | if($attr['name'] && (!$el->hasAttribute($attr['name'])) && ($valid)) { | |
42c80841 NL |
3174 | $el->setAttribute($attr['name'], $attr['value']); |
3175 | } | |
3176 | } | |
3177 | } | |
3178 | if ($append) { | |
3179 | $this->appendToRealParent($el); | |
3180 | $this->stack[] = $el; | |
3181 | } | |
3182 | ||
3183 | return $el; | |
3184 | } | |
3185 | ||
3186 | private function insertText($data) { | |
3187 | if ($data === '') return; | |
3188 | if ($this->ignore_lf_token) { | |
3189 | if ($data[0] === "\n") { | |
3190 | $data = substr($data, 1); | |
3191 | if ($data === false) return; | |
3192 | } | |
3193 | } | |
3194 | $text = $this->dom->createTextNode($data); | |
3195 | $this->appendToRealParent($text); | |
3196 | } | |
3197 | ||
3198 | private function insertComment($data) { | |
3199 | $comment = $this->dom->createComment($data); | |
3200 | $this->appendToRealParent($comment); | |
3201 | } | |
3202 | ||
3203 | private function appendToRealParent($node) { | |
3204 | // this is only for the foster_parent case | |
3205 | /* If the current node is a table, tbody, tfoot, thead, or tr | |
3206 | element, then, whenever a node would be inserted into the current | |
3207 | node, it must instead be inserted into the foster parent element. */ | |
3208 | if(!$this->foster_parent || !in_array(end($this->stack)->tagName, | |
3209 | array('table', 'tbody', 'tfoot', 'thead', 'tr'))) { | |
3210 | end($this->stack)->appendChild($node); | |
3211 | } else { | |
3212 | $this->fosterParent($node); | |
3213 | } | |
3214 | } | |
3215 | ||
3216 | private function elementInScope($el, $scope = self::SCOPE) { | |
3217 | if(is_array($el)) { | |
3218 | foreach($el as $element) { | |
3219 | if($this->elementInScope($element, $scope)) { | |
3220 | return true; | |
3221 | } | |
3222 | } | |
3223 | ||
3224 | return false; | |
3225 | } | |
3226 | ||
3227 | $leng = count($this->stack); | |
3228 | ||
3229 | for($n = 0; $n < $leng; $n++) { | |
3230 | /* 1. Initialise node to be the current node (the bottommost node of | |
3231 | the stack). */ | |
3232 | $node = $this->stack[$leng - 1 - $n]; | |
3233 | ||
3234 | if($node->tagName === $el) { | |
3235 | /* 2. If node is the target node, terminate in a match state. */ | |
3236 | return true; | |
3237 | ||
3238 | // We've expanded the logic for these states a little differently; | |
3239 | // Hixie's refactoring into "specific scope" is more general, but | |
3240 | // this "gets the job done" | |
3241 | ||
3242 | // these are the common states for all scopes | |
3243 | } elseif($node->tagName === 'table' || $node->tagName === 'html') { | |
3244 | return false; | |
3245 | ||
3246 | // these are valid for "in scope" and "in list item scope" | |
3247 | } elseif($scope !== self::SCOPE_TABLE && | |
3248 | (in_array($node->tagName, array('applet', 'caption', 'td', | |
3249 | 'th', 'button', 'marquee', 'object')) || | |
3250 | $node->tagName === 'foreignObject' && $node->namespaceURI === self::NS_SVG)) { | |
3251 | return false; | |
3252 | ||
3253 | ||
3254 | // these are valid for "in list item scope" | |
3255 | } elseif($scope === self::SCOPE_LISTITEM && in_array($node->tagName, array('ol', 'ul'))) { | |
3256 | return false; | |
3257 | } | |
3258 | ||
3259 | /* Otherwise, set node to the previous entry in the stack of open | |
3260 | elements and return to step 2. (This will never fail, since the loop | |
3261 | will always terminate in the previous step if the top of the stack | |
3262 | is reached.) */ | |
3263 | } | |
3264 | } | |
3265 | ||
3266 | private function reconstructActiveFormattingElements() { | |
3267 | /* 1. If there are no entries in the list of active formatting elements, | |
3268 | then there is nothing to reconstruct; stop this algorithm. */ | |
3269 | $formatting_elements = count($this->a_formatting); | |
3270 | ||
3271 | if($formatting_elements === 0) { | |
3272 | return false; | |
3273 | } | |
3274 | ||
3275 | /* 3. Let entry be the last (most recently added) element in the list | |
3276 | of active formatting elements. */ | |
3277 | $entry = end($this->a_formatting); | |
3278 | ||
3279 | /* 2. If the last (most recently added) entry in the list of active | |
3280 | formatting elements is a marker, or if it is an element that is in the | |
3281 | stack of open elements, then there is nothing to reconstruct; stop this | |
3282 | algorithm. */ | |
3283 | if($entry === self::MARKER || in_array($entry, $this->stack, true)) { | |
3284 | return false; | |
3285 | } | |
3286 | ||
3287 | for($a = $formatting_elements - 1; $a >= 0; true) { | |
3288 | /* 4. If there are no entries before entry in the list of active | |
3289 | formatting elements, then jump to step 8. */ | |
3290 | if($a === 0) { | |
3291 | $step_seven = false; | |
3292 | break; | |
3293 | } | |
3294 | ||
3295 | /* 5. Let entry be the entry one earlier than entry in the list of | |
3296 | active formatting elements. */ | |
3297 | $a--; | |
3298 | $entry = $this->a_formatting[$a]; | |
3299 | ||
3300 | /* 6. If entry is neither a marker nor an element that is also in | |
3301 | thetack of open elements, go to step 4. */ | |
3302 | if($entry === self::MARKER || in_array($entry, $this->stack, true)) { | |
3303 | break; | |
3304 | } | |
3305 | } | |
3306 | ||
3307 | while(true) { | |
3308 | /* 7. Let entry be the element one later than entry in the list of | |
3309 | active formatting elements. */ | |
3310 | if(isset($step_seven) && $step_seven === true) { | |
3311 | $a++; | |
3312 | $entry = $this->a_formatting[$a]; | |
3313 | } | |
3314 | ||
3315 | /* 8. Perform a shallow clone of the element entry to obtain clone. */ | |
3316 | $clone = $entry->cloneNode(); | |
3317 | ||
3318 | /* 9. Append clone to the current node and push it onto the stack | |
3319 | of open elements so that it is the new current node. */ | |
3320 | $this->appendToRealParent($clone); | |
3321 | $this->stack[] = $clone; | |
3322 | ||
3323 | /* 10. Replace the entry for entry in the list with an entry for | |
3324 | clone. */ | |
3325 | $this->a_formatting[$a] = $clone; | |
3326 | ||
3327 | /* 11. If the entry for clone in the list of active formatting | |
3328 | elements is not the last entry in the list, return to step 7. */ | |
3329 | if(end($this->a_formatting) !== $clone) { | |
3330 | $step_seven = true; | |
3331 | } else { | |
3332 | break; | |
3333 | } | |
3334 | } | |
3335 | } | |
3336 | ||
3337 | private function clearTheActiveFormattingElementsUpToTheLastMarker() { | |
3338 | /* When the steps below require the UA to clear the list of active | |
3339 | formatting elements up to the last marker, the UA must perform the | |
3340 | following steps: */ | |
3341 | ||
3342 | while(true) { | |
3343 | /* 1. Let entry be the last (most recently added) entry in the list | |
3344 | of active formatting elements. */ | |
3345 | $entry = end($this->a_formatting); | |
3346 | ||
3347 | /* 2. Remove entry from the list of active formatting elements. */ | |
3348 | array_pop($this->a_formatting); | |
3349 | ||
3350 | /* 3. If entry was a marker, then stop the algorithm at this point. | |
3351 | The list has been cleared up to the last marker. */ | |
3352 | if($entry === self::MARKER) { | |
3353 | break; | |
3354 | } | |
3355 | } | |
3356 | } | |
3357 | ||
3358 | private function generateImpliedEndTags($exclude = array()) { | |
3359 | /* When the steps below require the UA to generate implied end tags, | |
3360 | * then, while the current node is a dc element, a dd element, a ds | |
3361 | * element, a dt element, an li element, an option element, an optgroup | |
3362 | * element, a p element, an rp element, or an rt element, the UA must | |
3363 | * pop the current node off the stack of open elements. */ | |
3364 | $node = end($this->stack); | |
3365 | $elements = array_diff(array('dc', 'dd', 'ds', 'dt', 'li', 'p', 'td', 'th', 'tr'), $exclude); | |
3366 | ||
3367 | while(in_array(end($this->stack)->tagName, $elements)) { | |
3368 | array_pop($this->stack); | |
3369 | } | |
3370 | } | |
3371 | ||
3372 | private function getElementCategory($node) { | |
3373 | if (!is_object($node)) debug_print_backtrace(); | |
3374 | $name = $node->tagName; | |
3375 | if(in_array($name, $this->special)) | |
3376 | return self::SPECIAL; | |
3377 | ||
3378 | elseif(in_array($name, $this->scoping)) | |
3379 | return self::SCOPING; | |
3380 | ||
3381 | elseif(in_array($name, $this->formatting)) | |
3382 | return self::FORMATTING; | |
3383 | ||
3384 | else | |
3385 | return self::PHRASING; | |
3386 | } | |
3387 | ||
3388 | private function clearStackToTableContext($elements) { | |
3389 | /* When the steps above require the UA to clear the stack back to a | |
3390 | table context, it means that the UA must, while the current node is not | |
3391 | a table element or an html element, pop elements from the stack of open | |
3392 | elements. */ | |
3393 | while(true) { | |
3394 | $name = end($this->stack)->tagName; | |
3395 | ||
3396 | if(in_array($name, $elements)) { | |
3397 | break; | |
3398 | } else { | |
3399 | array_pop($this->stack); | |
3400 | } | |
3401 | } | |
3402 | } | |
3403 | ||
3404 | private function resetInsertionMode($context = null) { | |
3405 | /* 1. Let last be false. */ | |
3406 | $last = false; | |
3407 | $leng = count($this->stack); | |
3408 | ||
3409 | for($n = $leng - 1; $n >= 0; $n--) { | |
3410 | /* 2. Let node be the last node in the stack of open elements. */ | |
3411 | $node = $this->stack[$n]; | |
3412 | ||
3413 | /* 3. If node is the first node in the stack of open elements, then | |
3414 | * set last to true and set node to the context element. (fragment | |
3415 | * case) */ | |
3416 | if($this->stack[0]->isSameNode($node)) { | |
3417 | $last = true; | |
3418 | $node = $context; | |
3419 | } | |
3420 | ||
3421 | /* 4. If node is a select element, then switch the insertion mode to | |
3422 | "in select" and abort these steps. (fragment case) */ | |
3423 | if($node->tagName === 'select') { | |
3424 | $this->mode = self::IN_SELECT; | |
3425 | break; | |
3426 | ||
3427 | /* 5. If node is a td or th element, then switch the insertion mode | |
3428 | to "in cell" and abort these steps. */ | |
3429 | } elseif($node->tagName === 'td' || $node->nodeName === 'th') { | |
3430 | $this->mode = self::IN_CELL; | |
3431 | break; | |
3432 | ||
3433 | /* 6. If node is a tr element, then switch the insertion mode to | |
3434 | "in row" and abort these steps. */ | |
3435 | } elseif($node->tagName === 'tr') { | |
3436 | $this->mode = self::IN_ROW; | |
3437 | break; | |
3438 | ||
3439 | /* 7. If node is a tbody, thead, or tfoot element, then switch the | |
3440 | insertion mode to "in table body" and abort these steps. */ | |
3441 | } elseif(in_array($node->tagName, array('tbody', 'thead', 'tfoot'))) { | |
3442 | $this->mode = self::IN_TABLE_BODY; | |
3443 | break; | |
3444 | ||
3445 | /* 8. If node is a caption element, then switch the insertion mode | |
3446 | to "in caption" and abort these steps. */ | |
3447 | } elseif($node->tagName === 'caption') { | |
3448 | $this->mode = self::IN_CAPTION; | |
3449 | break; | |
3450 | ||
3451 | /* 9. If node is a colgroup element, then switch the insertion mode | |
3452 | to "in column group" and abort these steps. (innerHTML case) */ | |
3453 | } elseif($node->tagName === 'colgroup') { | |
3454 | $this->mode = self::IN_COLUMN_GROUP; | |
3455 | break; | |
3456 | ||
3457 | /* 10. If node is a table element, then switch the insertion mode | |
3458 | to "in table" and abort these steps. */ | |
3459 | } elseif($node->tagName === 'table') { | |
3460 | $this->mode = self::IN_TABLE; | |
3461 | break; | |
3462 | ||
3463 | /* 11. If node is an element from the MathML namespace or the SVG | |
3464 | * namespace, then switch the insertion mode to "in foreign | |
3465 | * content", let the secondary insertion mode be "in body", and | |
3466 | * abort these steps. */ | |
3467 | } elseif($node->namespaceURI === self::NS_SVG || | |
3468 | $node->namespaceURI === self::NS_MATHML) { | |
3469 | $this->mode = self::IN_FOREIGN_CONTENT; | |
3470 | $this->secondary_mode = self::IN_BODY; | |
3471 | break; | |
3472 | ||
3473 | /* 12. If node is a head element, then switch the insertion mode | |
3474 | to "in body" ("in body"! not "in head"!) and abort these steps. | |
3475 | (fragment case) */ | |
3476 | } elseif($node->tagName === 'head') { | |
3477 | $this->mode = self::IN_BODY; | |
3478 | break; | |
3479 | ||
3480 | /* 13. If node is a body element, then switch the insertion mode to | |
3481 | "in body" and abort these steps. */ | |
3482 | } elseif($node->tagName === 'body') { | |
3483 | $this->mode = self::IN_BODY; | |
3484 | break; | |
3485 | ||
3486 | /* 14. If node is a frameset element, then switch the insertion | |
3487 | mode to "in frameset" and abort these steps. (fragment case) */ | |
3488 | } elseif($node->tagName === 'frameset') { | |
3489 | $this->mode = self::IN_FRAMESET; | |
3490 | break; | |
3491 | ||
3492 | /* 15. If node is an html element, then: if the head element | |
3493 | pointer is null, switch the insertion mode to "before head", | |
3494 | otherwise, switch the insertion mode to "after head". In either | |
3495 | case, abort these steps. (fragment case) */ | |
3496 | } elseif($node->tagName === 'html') { | |
3497 | $this->mode = ($this->head_pointer === null) | |
3498 | ? self::BEFORE_HEAD | |
3499 | : self::AFTER_HEAD; | |
3500 | ||
3501 | break; | |
3502 | ||
3503 | /* 16. If last is true, then set the insertion mode to "in body" | |
3504 | and abort these steps. (fragment case) */ | |
3505 | } elseif($last) { | |
3506 | $this->mode = self::IN_BODY; | |
3507 | break; | |
3508 | } | |
3509 | } | |
3510 | } | |
3511 | ||
3512 | private function closeCell() { | |
3513 | /* If the stack of open elements has a td or th element in table scope, | |
3514 | then act as if an end tag token with that tag name had been seen. */ | |
3515 | foreach(array('td', 'th') as $cell) { | |
3516 | if($this->elementInScope($cell, self::SCOPE_TABLE)) { | |
3517 | $this->emitToken(array( | |
3518 | 'name' => $cell, | |
3519 | 'type' => HTML5_Tokenizer::ENDTAG | |
3520 | )); | |
3521 | ||
3522 | break; | |
3523 | } | |
3524 | } | |
3525 | } | |
3526 | ||
3527 | private function processWithRulesFor($token, $mode) { | |
3528 | /* "using the rules for the m insertion mode", where m is one of these | |
3529 | * modes, the user agent must use the rules described under the m | |
3530 | * insertion mode's section, but must leave the insertion mode | |
3531 | * unchanged unless the rules in m themselves switch the insertion mode | |
3532 | * to a new value. */ | |
3533 | return $this->emitToken($token, $mode); | |
3534 | } | |
3535 | ||
3536 | private function insertCDATAElement($token) { | |
3537 | $this->insertElement($token); | |
3538 | $this->original_mode = $this->mode; | |
3539 | $this->mode = self::IN_CDATA_RCDATA; | |
3540 | $this->content_model = HTML5_Tokenizer::CDATA; | |
3541 | } | |
3542 | ||
3543 | private function insertRCDATAElement($token) { | |
3544 | $this->insertElement($token); | |
3545 | $this->original_mode = $this->mode; | |
3546 | $this->mode = self::IN_CDATA_RCDATA; | |
3547 | $this->content_model = HTML5_Tokenizer::RCDATA; | |
3548 | } | |
3549 | ||
3550 | private function getAttr($token, $key) { | |
3551 | if (!isset($token['attr'])) return false; | |
3552 | $ret = false; | |
3553 | foreach ($token['attr'] as $keypair) { | |
3554 | if ($keypair['name'] === $key) $ret = $keypair['value']; | |
3555 | } | |
3556 | return $ret; | |
3557 | } | |
3558 | ||
3559 | private function getCurrentTable() { | |
3560 | /* The current table is the last table element in the stack of open | |
3561 | * elements, if there is one. If there is no table element in the stack | |
3562 | * of open elements (fragment case), then the current table is the | |
3563 | * first element in the stack of open elements (the html element). */ | |
3564 | for ($i = count($this->stack) - 1; $i >= 0; $i--) { | |
3565 | if ($this->stack[$i]->tagName === 'table') { | |
3566 | return $this->stack[$i]; | |
3567 | } | |
3568 | } | |
3569 | return $this->stack[0]; | |
3570 | } | |
3571 | ||
3572 | private function getFosterParent() { | |
3573 | /* The foster parent element is the parent element of the last | |
3574 | table element in the stack of open elements, if there is a | |
3575 | table element and it has such a parent element. If there is no | |
3576 | table element in the stack of open elements (innerHTML case), | |
3577 | then the foster parent element is the first element in the | |
3578 | stack of open elements (the html element). Otherwise, if there | |
3579 | is a table element in the stack of open elements, but the last | |
3580 | table element in the stack of open elements has no parent, or | |
3581 | its parent node is not an element, then the foster parent | |
3582 | element is the element before the last table element in the | |
3583 | stack of open elements. */ | |
3584 | for($n = count($this->stack) - 1; $n >= 0; $n--) { | |
3585 | if($this->stack[$n]->tagName === 'table') { | |
3586 | $table = $this->stack[$n]; | |
3587 | break; | |
3588 | } | |
3589 | } | |
3590 | ||
3591 | if(isset($table) && $table->parentNode !== null) { | |
3592 | return $table->parentNode; | |
3593 | ||
3594 | } elseif(!isset($table)) { | |
3595 | return $this->stack[0]; | |
3596 | ||
3597 | } elseif(isset($table) && ($table->parentNode === null || | |
3598 | $table->parentNode->nodeType !== XML_ELEMENT_NODE)) { | |
3599 | return $this->stack[$n - 1]; | |
3600 | } | |
3601 | } | |
3602 | ||
3603 | public function fosterParent($node) { | |
3604 | $foster_parent = $this->getFosterParent(); | |
3605 | $table = $this->getCurrentTable(); // almost equivalent to last table element, except it can be html | |
3606 | /* When a node node is to be foster parented, the node node must be | |
3607 | * be inserted into the foster parent element. */ | |
3608 | /* If the foster parent element is the parent element of the last table | |
3609 | * element in the stack of open elements, then node must be inserted | |
3610 | * immediately before the last table element in the stack of open | |
3611 | * elements in the foster parent element; otherwise, node must be | |
3612 | * appended to the foster parent element. */ | |
3613 | if ($table->tagName === 'table' && $table->parentNode->isSameNode($foster_parent)) { | |
3614 | $foster_parent->insertBefore($node, $table); | |
3615 | } else { | |
3616 | $foster_parent->appendChild($node); | |
3617 | } | |
3618 | } | |
3619 | ||
3620 | /** | |
3621 | * For debugging, prints the stack | |
3622 | */ | |
3623 | private function printStack() { | |
3624 | $names = array(); | |
3625 | foreach ($this->stack as $i => $element) { | |
3626 | $names[] = $element->tagName; | |
3627 | } | |
3628 | echo " -> stack [" . implode(', ', $names) . "]\n"; | |
3629 | } | |
3630 | ||
3631 | /** | |
3632 | * For debugging, prints active formatting elements | |
3633 | */ | |
3634 | private function printActiveFormattingElements() { | |
3635 | if (!$this->a_formatting) return; | |
3636 | $names = array(); | |
3637 | foreach ($this->a_formatting as $node) { | |
3638 | if ($node === self::MARKER) $names[] = 'MARKER'; | |
3639 | else $names[] = $node->tagName; | |
3640 | } | |
3641 | echo " -> active formatting [" . implode(', ', $names) . "]\n"; | |
3642 | } | |
3643 | ||
3644 | public function currentTableIsTainted() { | |
3645 | return !empty($this->getCurrentTable()->tainted); | |
3646 | } | |
3647 | ||
3648 | /** | |
3649 | * Sets up the tree constructor for building a fragment. | |
3650 | */ | |
3651 | public function setupContext($context = null) { | |
3652 | $this->fragment = true; | |
3653 | if ($context) { | |
3654 | $context = $this->dom->createElementNS(self::NS_HTML, $context); | |
3655 | /* 4.1. Set the HTML parser's tokenization stage's content model | |
3656 | * flag according to the context element, as follows: */ | |
3657 | switch ($context->tagName) { | |
3658 | case 'title': case 'textarea': | |
3659 | $this->content_model = HTML5_Tokenizer::RCDATA; | |
3660 | break; | |
3661 | case 'style': case 'script': case 'xmp': case 'iframe': | |
3662 | case 'noembed': case 'noframes': | |
3663 | $this->content_model = HTML5_Tokenizer::CDATA; | |
3664 | break; | |
3665 | case 'noscript': | |
3666 | // XSCRIPT: assuming scripting is enabled | |
3667 | $this->content_model = HTML5_Tokenizer::CDATA; | |
3668 | break; | |
3669 | case 'plaintext': | |
3670 | $this->content_model = HTML5_Tokenizer::PLAINTEXT; | |
3671 | break; | |
3672 | } | |
3673 | /* 4.2. Let root be a new html element with no attributes. */ | |
3674 | $root = $this->dom->createElementNS(self::NS_HTML, 'html'); | |
3675 | $this->root = $root; | |
3676 | /* 4.3 Append the element root to the Document node created above. */ | |
3677 | $this->dom->appendChild($root); | |
3678 | /* 4.4 Set up the parser's stack of open elements so that it | |
3679 | * contains just the single element root. */ | |
3680 | $this->stack = array($root); | |
3681 | /* 4.5 Reset the parser's insertion mode appropriately. */ | |
3682 | $this->resetInsertionMode($context); | |
3683 | /* 4.6 Set the parser's form element pointer to the nearest node | |
3684 | * to the context element that is a form element (going straight up | |
3685 | * the ancestor chain, and including the element itself, if it is a | |
3686 | * form element), or, if there is no such form element, to null. */ | |
3687 | $node = $context; | |
3688 | do { | |
3689 | if ($node->tagName === 'form') { | |
3690 | $this->form_pointer = $node; | |
3691 | break; | |
3692 | } | |
3693 | } while ($node = $node->parentNode); | |
3694 | } | |
3695 | } | |
3696 | ||
3697 | public function adjustMathMLAttributes($token) { | |
3698 | foreach ($token['attr'] as &$kp) { | |
3699 | if ($kp['name'] === 'definitionurl') { | |
3700 | $kp['name'] = 'definitionURL'; | |
3701 | } | |
3702 | } | |
3703 | return $token; | |
3704 | } | |
3705 | ||
3706 | public function adjustSVGAttributes($token) { | |
3707 | static $lookup = array( | |
3708 | 'attributename' => 'attributeName', | |
3709 | 'attributetype' => 'attributeType', | |
3710 | 'basefrequency' => 'baseFrequency', | |
3711 | 'baseprofile' => 'baseProfile', | |
3712 | 'calcmode' => 'calcMode', | |
3713 | 'clippathunits' => 'clipPathUnits', | |
3714 | 'contentscripttype' => 'contentScriptType', | |
3715 | 'contentstyletype' => 'contentStyleType', | |
3716 | 'diffuseconstant' => 'diffuseConstant', | |
3717 | 'edgemode' => 'edgeMode', | |
3718 | 'externalresourcesrequired' => 'externalResourcesRequired', | |
3719 | 'filterres' => 'filterRes', | |
3720 | 'filterunits' => 'filterUnits', | |
3721 | 'glyphref' => 'glyphRef', | |
3722 | 'gradienttransform' => 'gradientTransform', | |
3723 | 'gradientunits' => 'gradientUnits', | |
3724 | 'kernelmatrix' => 'kernelMatrix', | |
3725 | 'kernelunitlength' => 'kernelUnitLength', | |
3726 | 'keypoints' => 'keyPoints', | |
3727 | 'keysplines' => 'keySplines', | |
3728 | 'keytimes' => 'keyTimes', | |
3729 | 'lengthadjust' => 'lengthAdjust', | |
3730 | 'limitingconeangle' => 'limitingConeAngle', | |
3731 | 'markerheight' => 'markerHeight', | |
3732 | 'markerunits' => 'markerUnits', | |
3733 | 'markerwidth' => 'markerWidth', | |
3734 | 'maskcontentunits' => 'maskContentUnits', | |
3735 | 'maskunits' => 'maskUnits', | |
3736 | 'numoctaves' => 'numOctaves', | |
3737 | 'pathlength' => 'pathLength', | |
3738 | 'patterncontentunits' => 'patternContentUnits', | |
3739 | 'patterntransform' => 'patternTransform', | |
3740 | 'patternunits' => 'patternUnits', | |
3741 | 'pointsatx' => 'pointsAtX', | |
3742 | 'pointsaty' => 'pointsAtY', | |
3743 | 'pointsatz' => 'pointsAtZ', | |
3744 | 'preservealpha' => 'preserveAlpha', | |
3745 | 'preserveaspectratio' => 'preserveAspectRatio', | |
3746 | 'primitiveunits' => 'primitiveUnits', | |
3747 | 'refx' => 'refX', | |
3748 | 'refy' => 'refY', | |
3749 | 'repeatcount' => 'repeatCount', | |
3750 | 'repeatdur' => 'repeatDur', | |
3751 | 'requiredextensions' => 'requiredExtensions', | |
3752 | 'requiredfeatures' => 'requiredFeatures', | |
3753 | 'specularconstant' => 'specularConstant', | |
3754 | 'specularexponent' => 'specularExponent', | |
3755 | 'spreadmethod' => 'spreadMethod', | |
3756 | 'startoffset' => 'startOffset', | |
3757 | 'stddeviation' => 'stdDeviation', | |
3758 | 'stitchtiles' => 'stitchTiles', | |
3759 | 'surfacescale' => 'surfaceScale', | |
3760 | 'systemlanguage' => 'systemLanguage', | |
3761 | 'tablevalues' => 'tableValues', | |
3762 | 'targetx' => 'targetX', | |
3763 | 'targety' => 'targetY', | |
3764 | 'textlength' => 'textLength', | |
3765 | 'viewbox' => 'viewBox', | |
3766 | 'viewtarget' => 'viewTarget', | |
3767 | 'xchannelselector' => 'xChannelSelector', | |
3768 | 'ychannelselector' => 'yChannelSelector', | |
3769 | 'zoomandpan' => 'zoomAndPan', | |
3770 | ); | |
3771 | foreach ($token['attr'] as &$kp) { | |
3772 | if (isset($lookup[$kp['name']])) { | |
3773 | $kp['name'] = $lookup[$kp['name']]; | |
3774 | } | |
3775 | } | |
3776 | return $token; | |
3777 | } | |
3778 | ||
3779 | public function adjustForeignAttributes($token) { | |
3780 | static $lookup = array( | |
3781 | 'xlink:actuate' => array('xlink', 'actuate', self::NS_XLINK), | |
3782 | 'xlink:arcrole' => array('xlink', 'arcrole', self::NS_XLINK), | |
3783 | 'xlink:href' => array('xlink', 'href', self::NS_XLINK), | |
3784 | 'xlink:role' => array('xlink', 'role', self::NS_XLINK), | |
3785 | 'xlink:show' => array('xlink', 'show', self::NS_XLINK), | |
3786 | 'xlink:title' => array('xlink', 'title', self::NS_XLINK), | |
3787 | 'xlink:type' => array('xlink', 'type', self::NS_XLINK), | |
3788 | 'xml:base' => array('xml', 'base', self::NS_XML), | |
3789 | 'xml:lang' => array('xml', 'lang', self::NS_XML), | |
3790 | 'xml:space' => array('xml', 'space', self::NS_XML), | |
3791 | 'xmlns' => array(null, 'xmlns', self::NS_XMLNS), | |
3792 | 'xmlns:xlink' => array('xmlns', 'xlink', self::NS_XMLNS), | |
3793 | ); | |
3794 | foreach ($token['attr'] as &$kp) { | |
3795 | if (isset($lookup[$kp['name']])) { | |
3796 | $kp['name'] = $lookup[$kp['name']]; | |
3797 | } | |
3798 | } | |
3799 | return $token; | |
3800 | } | |
3801 | ||
3802 | public function insertForeignElement($token, $namespaceURI) { | |
3803 | $el = $this->dom->createElementNS($namespaceURI, $token['name']); | |
3804 | if (!empty($token['attr'])) { | |
3805 | foreach ($token['attr'] as $kp) { | |
3806 | $attr = $kp['name']; | |
3807 | if (is_array($attr)) { | |
3808 | $ns = $attr[2]; | |
3809 | $attr = $attr[1]; | |
3810 | } else { | |
3811 | $ns = self::NS_HTML; | |
3812 | } | |
3813 | if (!$el->hasAttributeNS($ns, $attr)) { | |
3814 | // XSKETCHY: work around godawful libxml bug | |
3815 | if ($ns === self::NS_XLINK) { | |
3816 | $el->setAttribute('xlink:'.$attr, $kp['value']); | |
3817 | } elseif ($ns === self::NS_HTML) { | |
3818 | // Another godawful libxml bug | |
3819 | $el->setAttribute($attr, $kp['value']); | |
3820 | } else { | |
3821 | $el->setAttributeNS($ns, $attr, $kp['value']); | |
3822 | } | |
3823 | } | |
3824 | } | |
3825 | } | |
3826 | $this->appendToRealParent($el); | |
3827 | $this->stack[] = $el; | |
3828 | // XERROR: see below | |
3829 | /* If the newly created element has an xmlns attribute in the XMLNS | |
3830 | * namespace whose value is not exactly the same as the element's | |
3831 | * namespace, that is a parse error. Similarly, if the newly created | |
3832 | * element has an xmlns:xlink attribute in the XMLNS namespace whose | |
3833 | * value is not the XLink Namespace, that is a parse error. */ | |
3834 | } | |
3835 | ||
3836 | public function save() { | |
3837 | $this->dom->normalize(); | |
3838 | if (!$this->fragment) { | |
3839 | return $this->dom; | |
3840 | } else { | |
3841 | if ($this->root) { | |
3842 | return $this->root->childNodes; | |
3843 | } else { | |
3844 | return $this->dom->childNodes; | |
3845 | } | |
3846 | } | |
3847 | } | |
3848 | } | |
3849 |