Verzeichnisstruktur phpBB-3.3.15
- Veröffentlicht
- 28.08.2024
So funktioniert es
|
|
Auf das letzte Element klicken. Dies geht jeweils ein Schritt zurück |
Auf das Icon klicken, dies öffnet das Verzeichnis. Nochmal klicken schließt das Verzeichnis. |
|
|
(Beispiel Datei-Icons)
|
Auf das Icon klicken um den Quellcode anzuzeigen |
ElementInspector.php
001 <?php
002
003 /**
004 * @package s9e\TextFormatter
005 * @copyright Copyright (c) 2010-2022 The s9e authors
006 * @license http://www.opensource.org/licenses/mit-license.php The MIT License
007 */
008 namespace s9e\TextFormatter\Configurator\Helpers;
009
010 use DOMElement;
011 use DOMXPath;
012
013 class ElementInspector
014 {
015 /**
016 * This is an abridged version of the HTML5 content models and rules, with some liberties taken.
017 *
018 * For each element, up to three bitfields are defined: "c", "ac" and "dd". Bitfields are stored
019 * as raw bytes, formatted using the octal notation to keep the sources ASCII.
020 *
021 * "c" represents the categories the element belongs to. The categories are comprised of HTML5
022 * content models (such as "phrasing content" or "interactive content") plus a few special
023 * categories created to cover the parts of the specs that refer to "a group of X and Y
024 * elements" rather than a specific content model.
025 *
026 * "ac" represents the categories that are allowed as children of given element.
027 *
028 * "dd" represents the categories that must not appear as a descendant of given element.
029 *
030 * Sometimes, HTML5 specifies some restrictions on when an element can accept certain children,
031 * or what categories the element belongs to. For example, an <img> element is only part of the
032 * "interactive content" category if it has a "usemap" attribute. Those restrictions are
033 * expressed as an XPath expression and stored using the concatenation of the key of the bitfield
034 * plus the bit number of the category. For instance, if "interactive content" got assigned to
035 * bit 2, the definition of the <img> element will contain a key "c2" with value "@usemap".
036 *
037 * Additionally, other flags are set:
038 *
039 * "t" indicates that the element uses the "transparent" content model.
040 * "e" indicates that the element uses the "empty" content model.
041 * "v" indicates that the element is a void element.
042 * "nt" indicates that the element does not accept text nodes. (no text)
043 * "to" indicates that the element should only contain text. (text-only)
044 * "fe" indicates that the element is a formatting element. It will automatically be reopened
045 * when closed by an end tag of a different name.
046 * "b" indicates that the element is not phrasing content, which makes it likely to act like
047 * a block element.
048 *
049 * Finally, HTML5 defines "optional end tag" rules, where one element automatically closes its
050 * predecessor. Those are used to generate closeParent rules and are stored in the "cp" key.
051 *
052 * @var array
053 * @see /scripts/patchElementInspector.php
054 */
055 protected static $htmlElements = [
056 'a'=>['c'=>"\17\0\0\0\200",'c3'=>'@href','ac'=>"\0",'dd'=>"\10\0\0\0\200",'t'=>1,'fe'=>1],
057 'abbr'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
058 'address'=>['c'=>"\3\20",'ac'=>"\1",'dd'=>"\100\24",'b'=>1,'cp'=>['p']],
059 'area'=>['c'=>"\5",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1],
060 'article'=>['c'=>"\3\4",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['p']],
061 'aside'=>['c'=>"\3\4",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['p']],
062 'audio'=>['c'=>"\57\0\10",'c3'=>'@controls','c1'=>'@controls','ac'=>"\0\0\0\220",'ac28'=>'not(@src)','dd'=>"\0\0\10",'t'=>1],
063 'b'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0",'fe'=>1],
064 'base'=>['c'=>"\20",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1,'b'=>1],
065 'bdi'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
066 'bdo'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
067 'blockquote'=>['c'=>"\3",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['p']],
068 'body'=>['c'=>"\0\0\20",'ac'=>"\1",'dd'=>"\0",'b'=>1],
069 'br'=>['c'=>"\5",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1],
070 'button'=>['c'=>"\217",'ac'=>"\4",'dd'=>"\10"],
071 'canvas'=>['c'=>"\47",'ac'=>"\0",'dd'=>"\10",'t'=>1],
072 'caption'=>['c'=>"\0\2",'ac'=>"\1",'dd'=>"\0\0\0\0\1",'b'=>1],
073 'cite'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
074 'code'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0",'fe'=>1],
075 'col'=>['c'=>"\0\0\100",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1,'b'=>1],
076 'colgroup'=>['c'=>"\0\2",'ac'=>"\0\0\100",'ac22'=>'not(@span)','dd'=>"\0",'nt'=>1,'e'=>1,'e?'=>'@span','b'=>1],
077 'data'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
078 'datalist'=>['c'=>"\5",'ac'=>"\4\200\0\40",'dd'=>"\0"],
079 'dd'=>['c'=>"\0\100\0\1",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['dd','dt']],
080 'del'=>['c'=>"\5",'ac'=>"\0",'dd'=>"\0",'t'=>1],
081 'details'=>['c'=>"\13",'ac'=>"\1\0\0\10",'dd'=>"\0",'b'=>1,'cp'=>['p']],
082 'dfn'=>['c'=>"\7\0\0\0\20",'ac'=>"\4",'dd'=>"\0\0\0\0\20"],
083 'dialog'=>['c'=>"\1",'ac'=>"\1",'dd'=>"\0",'b'=>1],
084 'div'=>['c'=>"\3\100",'ac'=>"\1\200\0\1",'ac0'=>'not(ancestor::dl)','dd'=>"\0",'b'=>1,'cp'=>['p']],
085 'dl'=>['c'=>"\3",'c1'=>'dt and dd','ac'=>"\0\300",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['p']],
086 'dt'=>['c'=>"\0\100\0\1",'ac'=>"\1",'dd'=>"\100\4\4",'b'=>1,'cp'=>['dd','dt']],
087 'em'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0",'fe'=>1],
088 'embed'=>['c'=>"\57",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1],
089 'fieldset'=>['c'=>"\203",'ac'=>"\1\0\0\100",'dd'=>"\0",'b'=>1,'cp'=>['p']],
090 'figcaption'=>['c'=>"\0\0\0\0\0\1",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['p']],
091 'figure'=>['c'=>"\3",'ac'=>"\1\0\0\0\0\1",'dd'=>"\0",'b'=>1,'cp'=>['p']],
092 'footer'=>['c'=>"\3\20\4",'ac'=>"\1",'dd'=>"\0\0\4",'b'=>1,'cp'=>['p']],
093 'form'=>['c'=>"\3\0\0\0\10",'ac'=>"\1",'dd'=>"\0\0\0\0\10",'b'=>1,'cp'=>['p']],
094 'h1'=>['c'=>"\103\1",'ac'=>"\4",'dd'=>"\0",'b'=>1,'cp'=>['p']],
095 'h2'=>['c'=>"\103\1",'ac'=>"\4",'dd'=>"\0",'b'=>1,'cp'=>['p']],
096 'h3'=>['c'=>"\103\1",'ac'=>"\4",'dd'=>"\0",'b'=>1,'cp'=>['p']],
097 'h4'=>['c'=>"\103\1",'ac'=>"\4",'dd'=>"\0",'b'=>1,'cp'=>['p']],
098 'h5'=>['c'=>"\103\1",'ac'=>"\4",'dd'=>"\0",'b'=>1,'cp'=>['p']],
099 'h6'=>['c'=>"\103\1",'ac'=>"\4",'dd'=>"\0",'b'=>1,'cp'=>['p']],
100 'head'=>['c'=>"\0\0\20",'ac'=>"\20",'dd'=>"\0",'nt'=>1,'b'=>1],
101 'header'=>['c'=>"\3\20\4",'ac'=>"\1",'dd'=>"\0\0\4",'b'=>1,'cp'=>['p']],
102 'hgroup'=>['c'=>"\103",'ac'=>"\0\201",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['p']],
103 'hr'=>['c'=>"\1\40",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1,'b'=>1,'cp'=>['p']],
104 'html'=>['c'=>"\0",'ac'=>"\0\0\20",'dd'=>"\0",'nt'=>1,'b'=>1],
105 'i'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0",'fe'=>1],
106 'iframe'=>['c'=>"\57",'ac'=>"\4",'dd'=>"\0"],
107 'img'=>['c'=>"\257\0\40",'c3'=>'@usemap','ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1],
108 'input'=>['c'=>"\217",'c3'=>'@type!="hidden"','c7'=>'@type!="hidden" or @type="hidden"','c1'=>'@type!="hidden"','ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1],
109 'ins'=>['c'=>"\7",'ac'=>"\0",'dd'=>"\0",'t'=>1],
110 'kbd'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
111 'label'=>['c'=>"\17\0\0\0\4",'ac'=>"\4",'dd'=>"\0\0\1\0\4"],
112 'legend'=>['c'=>"\0\0\0\100",'ac'=>"\104",'dd'=>"\0",'b'=>1],
113 'li'=>['c'=>"\0\0\0\0\100",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['li']],
114 'link'=>['c'=>"\25",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1],
115 'main'=>['c'=>"\3",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['p']],
116 'map'=>['c'=>"\7",'ac'=>"\0",'dd'=>"\0",'t'=>1],
117 'mark'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
118 'menu'=>['c'=>"\3\40",'c1'=>'li','ac'=>"\0\240",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['p']],
119 'menuitem'=>['c'=>"\0\40",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'b'=>1],
120 'meta'=>['c'=>"\25",'c0'=>'@itemprop','c2'=>'@itemprop','ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1],
121 'meter'=>['c'=>"\7\0\1\0\2",'ac'=>"\4",'dd'=>"\0\0\0\0\2"],
122 'nav'=>['c'=>"\3\4",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['p']],
123 'noscript'=>['c'=>"\25",'ac'=>"\0",'dd'=>"\0",'nt'=>1],
124 'object'=>['c'=>"\47\0\0\2",'ac'=>"\0",'dd'=>"\0",'t'=>1],
125 'ol'=>['c'=>"\3",'c1'=>'li','ac'=>"\0\200\0\0\100",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['p']],
126 'optgroup'=>['c'=>"\0\0\2",'ac'=>"\0\200\0\40",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['optgroup','option']],
127 'option'=>['c'=>"\0\0\2\40",'ac'=>"\0",'dd'=>"\0",'b'=>1,'cp'=>['option']],
128 'output'=>['c'=>"\207",'ac'=>"\4",'dd'=>"\0"],
129 'p'=>['c'=>"\3\1",'ac'=>"\4",'dd'=>"\0",'b'=>1,'cp'=>['p']],
130 'picture'=>['c'=>"\45",'ac'=>"\0\200\40",'dd'=>"\0",'nt'=>1],
131 'pre'=>['c'=>"\3",'ac'=>"\4",'dd'=>"\0",'pre'=>1,'b'=>1,'cp'=>['p']],
132 'progress'=>['c'=>"\7\0\1\4",'ac'=>"\4",'dd'=>"\0\0\0\4"],
133 'q'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
134 'rb'=>['c'=>"\0\10",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'b'=>1],
135 'rp'=>['c'=>"\0\10",'ac'=>"\0",'dd'=>"\0",'to'=>1,'b'=>1,'cp'=>['rp','rt']],
136 'rt'=>['c'=>"\0\10",'ac'=>"\4",'dd'=>"\0",'b'=>1,'cp'=>['rp','rt']],
137 'rtc'=>['c'=>"\0\10",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'b'=>1],
138 'ruby'=>['c'=>"\7",'ac'=>"\4\10",'dd'=>"\0"],
139 's'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0",'fe'=>1],
140 'samp'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
141 'script'=>['c'=>"\25\200",'ac'=>"\0",'dd'=>"\0",'to'=>1],
142 'section'=>['c'=>"\3\4",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['p']],
143 'select'=>['c'=>"\217",'ac'=>"\0\200\2",'dd'=>"\0",'nt'=>1],
144 'slot'=>['c'=>"\5",'ac'=>"\0",'dd'=>"\0",'t'=>1],
145 'small'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0",'fe'=>1],
146 'source'=>['c'=>"\0\0\40\20",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1,'b'=>1],
147 'span'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
148 'strong'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0",'fe'=>1],
149 'style'=>['c'=>"\20",'ac'=>"\0",'dd'=>"\0",'to'=>1,'b'=>1],
150 'sub'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
151 'summary'=>['c'=>"\0\0\0\10",'ac'=>"\104",'dd'=>"\0",'b'=>1],
152 'sup'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
153 'table'=>['c'=>"\3\0\0\0\1",'ac'=>"\0\202",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['p']],
154 'tbody'=>['c'=>"\0\2",'ac'=>"\0\200\0\0\40",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['tbody','td','th','thead','tr']],
155 'td'=>['c'=>"\0\0\200",'ac'=>"\1",'dd'=>"\0",'b'=>1,'cp'=>['td','th']],
156 'template'=>['c'=>"\25\200\100",'ac'=>"\0",'dd'=>"\0",'nt'=>1],
157 'textarea'=>['c'=>"\217",'ac'=>"\0",'dd'=>"\0",'pre'=>1,'to'=>1],
158 'tfoot'=>['c'=>"\0\2",'ac'=>"\0\200\0\0\40",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['tbody','td','th','thead','tr']],
159 'th'=>['c'=>"\0\0\200",'ac'=>"\1",'dd'=>"\100\4\4",'b'=>1,'cp'=>['td','th']],
160 'thead'=>['c'=>"\0\2",'ac'=>"\0\200\0\0\40",'dd'=>"\0",'nt'=>1,'b'=>1],
161 'time'=>['c'=>"\7",'ac'=>"\4",'ac2'=>'@datetime','dd'=>"\0"],
162 'title'=>['c'=>"\20",'ac'=>"\0",'dd'=>"\0",'to'=>1,'b'=>1],
163 'tr'=>['c'=>"\0\2\0\0\40",'ac'=>"\0\200\200",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['td','th','tr']],
164 'track'=>['c'=>"\0\0\0\200",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1,'b'=>1],
165 'u'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0",'fe'=>1],
166 'ul'=>['c'=>"\3",'c1'=>'li','ac'=>"\0\200\0\0\100",'dd'=>"\0",'nt'=>1,'b'=>1,'cp'=>['p']],
167 'var'=>['c'=>"\7",'ac'=>"\4",'dd'=>"\0"],
168 'video'=>['c'=>"\57\0\10",'c3'=>'@controls','ac'=>"\0\0\0\220",'ac28'=>'not(@src)','dd'=>"\0\0\10",'t'=>1],
169 'wbr'=>['c'=>"\5",'ac'=>"\0",'dd'=>"\0",'nt'=>1,'e'=>1,'v'=>1]
170 ];
171
172 /**
173 * Test whether given child element closes given parent element
174 *
175 * @param DOMElement $child
176 * @param DOMElement $parent
177 * @return bool
178 */
179 public static function closesParent(DOMElement $child, DOMElement $parent)
180 {
181 $parentName = $parent->nodeName;
182 $childName = $child->nodeName;
183
184 return !empty(self::$htmlElements[$childName]['cp']) && in_array($parentName, self::$htmlElements[$childName]['cp'], true);
185 }
186
187 /**
188 * Test whether given element disallows text nodes
189 *
190 * @param DOMElement $element
191 * @return bool
192 */
193 public static function disallowsText(DOMElement $element)
194 {
195 return self::hasProperty($element, 'nt');
196 }
197
198 /**
199 * Return the "allowChild" bitfield for given element
200 *
201 * @param DOMElement $element
202 * @return string
203 */
204 public static function getAllowChildBitfield(DOMElement $element)
205 {
206 return self::getBitfield($element, 'ac');
207 }
208
209 /**
210 * Return the "category" bitfield for given element
211 *
212 * @param DOMElement $element
213 * @return string
214 */
215 public static function getCategoryBitfield(DOMElement $element)
216 {
217 return self::getBitfield($element, 'c');
218 }
219
220 /**
221 * Return the "denyDescendant" bitfield for given element
222 *
223 * @param DOMElement $element
224 * @return string
225 */
226 public static function getDenyDescendantBitfield(DOMElement $element)
227 {
228 return self::getBitfield($element, 'dd');
229 }
230
231 /**
232 * Test whether given element is a block element
233 *
234 * @param DOMElement $element
235 * @return bool
236 */
237 public static function isBlock(DOMElement $element)
238 {
239 return self::hasProperty($element, 'b');
240 }
241
242 /**
243 * Test whether given element uses the empty content model
244 *
245 * @param DOMElement $element
246 * @return bool
247 */
248 public static function isEmpty(DOMElement $element)
249 {
250 return self::hasProperty($element, 'e');
251 }
252
253 /**
254 * Test whether given element is a formatting element
255 *
256 * @param DOMElement $element
257 * @return bool
258 */
259 public static function isFormattingElement(DOMElement $element)
260 {
261 return self::hasProperty($element, 'fe');
262 }
263
264 /**
265 * Test whether given element only accepts text nodes
266 *
267 * @param DOMElement $element
268 * @return bool
269 */
270 public static function isTextOnly(DOMElement $element)
271 {
272 return self::hasProperty($element, 'to');
273 }
274
275 /**
276 * Test whether given element uses the transparent content model
277 *
278 * @param DOMElement $element
279 * @return bool
280 */
281 public static function isTransparent(DOMElement $element)
282 {
283 return self::hasProperty($element, 't');
284 }
285
286 /**
287 * Test whether given element uses the void content model
288 *
289 * @param DOMElement $element
290 * @return bool
291 */
292 public static function isVoid(DOMElement $element)
293 {
294 return self::hasProperty($element, 'v');
295 }
296
297 /**
298 * Test whether given element preserves whitespace in its content
299 *
300 * @param DOMElement $element
301 * @return bool
302 */
303 public static function preservesWhitespace(DOMElement $element)
304 {
305 return self::hasProperty($element, 'pre');
306 }
307
308 /**
309 * Evaluate an XPath query using given element as context node
310 *
311 * @param string $query XPath query
312 * @param DOMElement $element Context node
313 * @return bool
314 */
315 protected static function evaluate($query, DOMElement $element)
316 {
317 $xpath = new DOMXPath($element->ownerDocument);
318
319 return $xpath->evaluate('boolean(' . $query . ')', $element);
320 }
321
322 /**
323 * Get the bitfield value for a given element
324 *
325 * @param DOMElement $element Context node
326 * @param string $name Bitfield name: either 'c', 'ac' or 'dd'
327 * @return string
328 */
329 protected static function getBitfield(DOMElement $element, $name)
330 {
331 $props = self::getProperties($element);
332 $bitfield = self::toBin($props[$name]);
333
334 // For each bit set to 1, test whether there is an XPath condition to it and whether it is
335 // fulfilled. If not, turn the bit to 0
336 foreach (array_keys(array_filter(str_split($bitfield, 1))) as $bitNumber)
337 {
338 $conditionName = $name . $bitNumber;
339 if (isset($props[$conditionName]) && !self::evaluate($props[$conditionName], $element))
340 {
341 $bitfield[$bitNumber] = '0';
342 }
343 }
344
345 return self::toRaw($bitfield);
346 }
347
348 /**
349 * Return the properties associated with given element
350 *
351 * Returns span's properties if the element is not defined
352 *
353 * @param DOMElement $element
354 * @return array
355 */
356 protected static function getProperties(DOMElement $element)
357 {
358 return self::$htmlElements[$element->nodeName] ?? self::$htmlElements['span'];
359 }
360
361 /**
362 * Test whether given element has given property in context
363 *
364 * @param DOMElement $element Context node
365 * @param string $propName Property name, see self::$htmlElements
366 * @return bool
367 */
368 protected static function hasProperty(DOMElement $element, $propName)
369 {
370 $props = self::getProperties($element);
371
372 return !empty($props[$propName]) && (!isset($props[$propName . '?']) || self::evaluate($props[$propName . '?'], $element));
373 }
374
375 /**
376 * Convert a raw string to a series of 0 and 1 in LSB order
377 *
378 * @param string $raw
379 * @return string
380 */
381 protected static function toBin($raw)
382 {
383 $bin = '';
384 foreach (str_split($raw, 1) as $char)
385 {
386 $bin .= strrev(substr('0000000' . decbin(ord($char)), -8));
387 }
388
389 return $bin;
390 }
391
392 /**
393 * Convert a series of 0 and 1 in LSB order to a raw string
394 *
395 * @param string $bin
396 * @return string
397 */
398 protected static function toRaw($bin)
399 {
400 return implode('', array_map('chr', array_map('bindec', array_map('strrev', str_split($bin, 8)))));
401 }
402 }