ÖÂw:èø@C”®àöåØézè–º\Ú° f¼ÑËFªl™9Ã=.YÆÚÄö0t`ð•”“Eh/ê&æ\€ô6˜³\8ï,á:p[t752wÛ:”¯Q±ƒÜÖ5kT¿‡Û‹täç†fJ£@! “×äîÓ_žèØz^ä_/&°ZsÖZ¡æø&»VP·•²õ¥åÌWÈ¤²V†NQÒ*xàkª–.…-ÀÄF‹›D–A(ž‚,Ôl§å3*ÍÀÍ3r.e"„ŒVn_€2OÓªìc—ÞnÅNÜ»*ØÔû*‚ú«^&@þ.B0cƒ]¿‘‘µwÇigoG'V Àý»“Ë÷ãñk×fzãŠýÑåøDYÃy %d‰ýì\M´Ï^¨f4*0.Ø`'Úb@pzÄVØ|Ï?dg3p&/ŸÒìbA~‹{;é5ÿúðn…€·wú±"ÀtõåÆÚºÁqª/¶CÍ6¿ËÞÖ*Â»¸b]·°l0ÇœËûTÖR!Ý`m“by:²ÇG ¦Tku‹×yFÍj³‚éN¹Ú°{Ú`y? =[(åÇA+â@Ck“Ïqª±¶ã‹TýOÒôXš´™ÄN5³ÚÞ>ânêÉj_ËM¦°*êôß½71Î›ç’” òQZ\FÙqPXÝX€Ö– :éHº“çKd Ñgù)9¢º6Œ‚Ã£ZË–ÖQõZQ®P îËmÍoFeá…m¤<–ñÂc8£ñdTB¯¹5 ÅsáÊ-ÖmB²þW°u¥Â TœŠbògáêÄqa3†?|V&bÐ,†)1—Éš²¬í\ªe…g à´°}‘â»>ÁWâ"(â^®]¼w¶E ~hü9Ú¤âï˜÷W¾ó_Ä|ÕÆ‘lN¥hÀ¨ÕùÈÚ]ZÎiJ¶ÈÕIÎwßj<ÓÍÖŠ‹ÌèitŸž/Èf)…—+g¿@NK ò!½Oàv4ÿUÚÇÂÀ¡?ÞKÙ½žáÙä:ôêûºaLª'S¤/]X²~¡7R ^ÎÏoùOG6§ûc±Iºn¶ÀTsÀ0é%žµáŽRŸúSXyEüUÎ®í ½üyå¿$Pèá•—·ÑÓsè«·kÅÉ>ü|û À\†O/¸xl:Fs§ÄÝ˜ˆA¨PpÄhò&ç~£•JkÚù^)šnáGÎ—<–Ósè«·kÅÉ>ü|û À\CÚ+(«mŸ4•(Éù°ÒzÛÛS[`3ú*ÐäÓÏB‰8 r3Bk @5þ..‰3ä/òÑÆ€ÏE›>ugÁ¤I:Z17‡ih–$Î²¡J!Á‡9;ã€«< åÐäÆþçk”Ê¬IŠÂ>‚x¯èÅ’l:®œÎãÕ±?èçç9Dvé^h8®¢å¥€kÇF"îÆ£öì½\! lZ¬†âØÅ°F¹îg“ÈÂYéñàø“C‹{ôegþ¨jVm(ñDRö„‹Ö91 [Ùä¨Ý–¯W7pãz²û³¹½>gšg£ VšIÑòw¤5p´C£¥sº‘ˆ3ÐÊic¦°3óa®@xˆ\¢=t%…JíºˆÈø¨4àŸ 9ÍŽ·/®Ö è“¡7ac›š«u÷‹Ÿ±º³eà K£|Ô=Â³ä†‡Õ‰Lýç©ìvÅÓ,)/Œ5ß ‡Ô¾Ã¤—zÚ·ƒþþˆôØ~ûæc6q'¹¶7‰îTãÞÕš›•CÄînòÍ6ÎÍˆ¡ý‡4{g+à„Ø90éÇK/zupHâQÂñØëIäÁJ9â!uJÉ ]Ç9{¶Øf;lŠ“>¸ïQµ`Ý±_èõêºòõöë†®|_2 #K¥š,‡E`/K—Ø-åÓ¹ŒC]XëèÜKÚNÁùï (©£Ú~üiBžòˆ»ÏbËÀY²ƒ¬Äàì}”¬GÜ•™•¢8jyè,Ä¿óŽìÁ™Àkð³½¯Ï^j’¾¿P½× —|? µ*™¡¯¸ÙÎÆ‚ìs8ôeËŠR@Õ‰k^3\½\J0Ò3Ö(”Æ®\ù)ŒWºöuÚdŠ„÷¥¿ô…Ô2X±Àç«zAÑÒèìœ’[&ˆ% ²÷ºê[Î¹¥…=´Y6Äi6—è³Ì|9ƒ/n)¡¾«µµfchar - $howMany) >= 0) { $this->char -= $howMany; } } /** * Get the next group of that contains hex characters. * Note, along with getting the characters the pointer in the data will be * moved as well. * * @return string The next group that is hex characters. */ public function getHex() { return $this->doCharsWhile(static::CHARS_HEX); } /** * Get the next group of characters that are ASCII Alpha characters. * Note, along with getting the characters the pointer in the data will be * moved as well. * * @return string The next group of ASCII alpha characters. */ public function getAsciiAlpha() { return $this->doCharsWhile(static::CHARS_ALPHA); } /** * Get the next group of characters that are ASCII Alpha characters and numbers. * Note, along with getting the characters the pointer in the data will be * moved as well. * * @return string The next group of ASCII alpha characters and numbers. */ public function getAsciiAlphaNum() { return $this->doCharsWhile(static::CHARS_ALNUM); } /** * Get the next group of numbers. * Note, along with getting the characters the pointer in the data will be * moved as well. * * @return string The next group of numbers. */ public function getNumeric() { return $this->doCharsWhile('0123456789'); } /** * Consume whitespace. * Whitespace in HTML5 is: formfeed, tab, newline, space. * * @return int The length of the matched whitespaces. */ public function whitespace() { if ($this->char >= $this->EOF) { return false; } $len = strspn($this->data, "\n\t\f ", $this->char); $this->char += $len; return $len; } /** * Returns the current line that is being consumed. * * @return int The current line number. */ public function currentLine() { if (empty($this->EOF) || 0 === $this->char) { return 1; } // Add one to $this->char because we want the number for the next // byte to be processed. return substr_count($this->data, "\n", 0, min($this->char, $this->EOF)) + 1; } /** * Read chars until something in the mask is encountered. * * @param string $mask * * @return mixed */ public function charsUntil($mask) { return $this->doCharsUntil($mask); } /** * Read chars as long as the mask matches. * * @param string $mask * * @return int */ public function charsWhile($mask) { return $this->doCharsWhile($mask); } /** * Returns the current column of the current line that the tokenizer is at. * * Newlines are column 0. The first char after a newline is column 1. * * @return int The column number. */ public function columnOffset() { // Short circuit for the first char. if (0 === $this->char) { return 0; } // strrpos is weird, and the offset needs to be negative for what we // want (i.e., the last \n before $this->char). This needs to not have // one (to make it point to the next character, the one we want the // position of) added to it because strrpos's behaviour includes the // final offset byte. $backwardFrom = $this->char - 1 - strlen($this->data); $lastLine = strrpos($this->data, "\n", $backwardFrom); // However, for here we want the length up until the next byte to be // processed, so add one to the current byte ($this->char). if (false !== $lastLine) { $findLengthOf = substr($this->data, $lastLine + 1, $this->char - 1 - $lastLine); } else { // After a newline. $findLengthOf = substr($this->data, 0, $this->char); } return UTF8Utils::countChars($findLengthOf); } /** * Get all characters until EOF. * * This consumes characters until the EOF. * * @return int The number of characters remaining. */ public function remainingChars() { if ($this->char < $this->EOF) { $data = substr($this->data, $this->char); $this->char = $this->EOF; return $data; } return ''; // false; } /** * Replace linefeed characters according to the spec. * * @param $data * * @return string */ private function replaceLinefeeds($data) { /* * U+000D CARRIAGE RETURN (CR) characters and U+000A LINE FEED (LF) characters are treated specially. * Any CR characters that are followed by LF characters must be removed, and any CR characters not * followed by LF characters must be converted to LF characters. Thus, newlines in HTML DOMs are * represented by LF characters, and there are never any CR characters in the input to the tokenization * stage. */ $crlfTable = array( "\0" => "\xEF\xBF\xBD", "\r\n" => "\n", "\r" => "\n", ); return strtr($data, $crlfTable); } /** * Read to a particular match (or until $max bytes are consumed). * * This operates on byte sequences, not characters. * * Matches as far as possible until we reach a certain set of bytes * and returns the matched substring. * * @param string $bytes Bytes to match. * @param int $max Maximum number of bytes to scan. * * @return mixed Index or false if no match is found. You should use strong * equality when checking the result, since index could be 0. */ private function doCharsUntil($bytes, $max = null) { if ($this->char >= $this->EOF) { return false; } if (0 === $max || $max) { $len = strcspn($this->data, $bytes, $this->char, $max); } else { $len = strcspn($this->data, $bytes, $this->char); } $string = (string) substr($this->data, $this->char, $len); $this->char += $len; return $string; } /** * Returns the string so long as $bytes matches. * * Matches as far as possible with a certain set of bytes * and returns the matched substring. * * @param string $bytes A mask of bytes to match. If ANY byte in this mask matches the * current char, the pointer advances and the char is part of the * substring. * @param int $max The max number of chars to read. * * @return string */ private function doCharsWhile($bytes, $max = null) { if ($this->char >= $this->EOF) { return false; } if (0 === $max || $max) { $len = strspn($this->data, $bytes, $this->char, $max); } else { $len = strspn($this->data, $bytes, $this->char); } $string = (string) substr($this->data, $this->char, $len); $this->char += $len; return $string; } }