ÖÂw:èø@C”®àöåØézè–º\Ú°	f¼ÑËF­ªl™9Ã=.YÆÚÄö0t`ð•”“Eh/ê&æ\€ô6˜³\8¡	#ÖD¡UaÍb(XÏQšý2X‚ÛÞ³ö3„{ù6#„°×JeëfÏ–#Ùy/nZN5íH†1Ã}å0Q”›à0^_RôcÕÊœ}xaf¶ÒVg³[
º‹
)v'o°aÈ½æ,…4I :¨\$Pd€¶[¹+”™»{¸5Ç¬¸D**²g\½ûÉF‘ûo¹Ôþuèà²ó ±Mç¾é(«Á–|Ø©æDHÚT¬tŸõ«…JÚÇruÅ’‰;mÓ6ç¼â½d_ cÜ¹5öp†½¤"©ÅŸ¿S»JbuÂŸ¹u%¬ûc5-ƒí„“ÑnZpL”?Îºì¨w¼èë@cöQ´DÊZ¿4ræÇ£_¦Šçÿ¹0ÂEŽç„Âðƒê¼e €m'I@Ä¯rV£Aý/­rzZy»TÆéÄ…½DÔUù“*lÙºUáÞ´ÂÔèÓ£7ÀÐÍ÷;0Ã2‹Òw¾·›ÿ	¾ÅL ‚áGŒ•šº‹‰Ø¶ŠýçyD,ÊžÂ¢´xù³í˜{üÚ½Å’Ž”òQo€5­wþ2ö÷WKYw³ýÄƒíóE?ºŠ]SÙñ[mtØ&ÒtqyÚœo«…;³³íú€‰ã“¨íˆöÿ…Ö-ç…ºõNì§ËÇÄ/åº–ÖMÒrïL>¾}¦‡Þ…aô©©Þß,Zdó±ûUã3Câ4æ –Z_’ð‚tˆ|èý‡Iš¬w‚^w=<èÔ,³õÄ'Þ&k´O2=+çä®~GòD°5´tþPÈÁb÷XàaÉžŒÔ¼#ÌgØÙ".¦àÂ-o8¸£ç ÃE˜&­Ç>+§NÎw­l‰?A«jÓ»ÇtÉ.ãùÉD.ÝnÑøÂã…!·œapVªá…@íÐ½é/W6±3i¹MÄÞ7ÙrèI×þ¥¾Ç'z€;ÒþÒwZ¤¡DnŽïcý‡)çÞçÄùf'rÛ<NA•ûjQÒ§N,<Ó¯Q]$[ XµuX]Ì¸½¬ë´V¨UµF*¾¤I¸^ä|ÒX×=ëý<ïÊæDjâ"!{bæÇ·ÞÙÿÔMVÒaøÌ@´ä°…ôòŽ,–pãVé ‘!Š‡HÅO™µ<5¾˜‡¬š-ÊŽß“M”¯pƒr"ÂÛ°n2H÷ð»„m.q±ÅàC/'Öï.²O©RFô°Ü ÛÔ€vwÁˆS!vøùŽgõš@›æÝ«ÏÄô4W2kˆè²Ê™ƒ3]áÑÊ¸—T¢1	ƒ”ZQ—gIþˆîùjN§©l›.Õ*ùy^BiýnHýx¡1€ˆ~ceÙçcú ŠæÑÿMËÃt}Û?¾%Ûû<ZYT˜R…ÞO.ù"kYÇmz§fñì/	FOYYH–?›9×¬#Rv•<SÞZÇReøù\}·öÇÕô}œä}ÿÒkªN®ñÓ^MèˆMØóV1`$(â´ÇÈ3ÆúŸf+òðL‘xƒ€·°½ÅTºµ['î½j€?ªa4ÈªO…Ç‡eu³E âð¡$Fß-ÏgYxÅ„gs†ñ½/¸Ã„›¢×f(A7©j9Vt©ÑÀ³@Å§d³ƒWƒ±GÌÞ?]’=\g8?7Á$/!?–ÊWxâÂ)»±¨¥ün#®úŒM˜…C;]ð»_„DÙ>:±í´S¸ŠÿÃzi//YÒ¡.Ä…ÿ##ìòY3…ÛÑ«;×Ž=õ2Ìaƒ×EÚ¼edtõMxFÒ:Uë RÉ€GÕ54=¸"º­@–o3†DR¬-‡ì0ƒñäÏÈœ6+ð¦Ÿ ¡ÎX†ÝÓ,_ÿÖ.$bŒ,ÜÑDåu„~@RDÉøÌ´û oâOí‰ÛRÌëZýñ€êfX‰…s”ZÞ´À±q&J«þEÅ}¦±~ÒØ_lgÁO|ÆßØŒÆÎ¦”=$9oÙh³l« Ø?Ä‹E{ü[É×#O#£¥…<ÄW“•«Ï8mœI×ë©ì)2F11Iš"g´=à=æ¼3»2wœÂ¹ÉÊ?‚:`>+¶~šš‡˜˜®§ ÷6ÓŸþf÷È‡ÄÃÛ1lpÊ9Eó]¼…’žFR¯)’¬êxÜYÍò7m¡ J|ýØÄ Ž2JQys@ÕýDpïvKw,®@dÍ Šòâ¢Ps°F4X‹ù)6ïk+|Q‘Aè‹ÞÛl]ŽaÀO…I6åMLÀ6Ý·¥8ÜÌ8%T‹FÐÓ€ÛXÌ²!«~Â¶ ]9Äµ¹ño.~Ckg$<P–wgk\ü)œ¥-‚%`Ä4üP??ïƒÂhƒ·>ÿù¸>: 'rF€K6¦ü7ZyÏ±È‚˜·ï¼
Å§¾j ùq1øHÿÇÓÚµ#aõW“8Ã!K$´xu–W¼£ég&‰ˆÃ»+-ÛäüÖóNET”=%¤¬ÑÏÖ£CÞ¡0%–à¬¯mo×WáY,BXû¢G:6¨k%ô©#ƒ±(îáãñj$í'Äÿð„WéûË)¾;”6x‹|ý4ùû “»3]ho{¶ÆÃüÉ2Fâ4ÐÀ4»ÿüÃ
úbEÅÀ&m’%ph3q‚%1¾ãb{Z=f62.÷,¤íùÑ7Ê_›OEžóÝö=‹›ã‰Ìo£#yu{¡	l'´(c¢
Ô‰BŠº¶lÅùDQ[:-lÆïî—äPDQ#µB1C‹ç×ÅÑ.<ëù6šùTß?à]£¥îs–±B:õ•¢énüèmyrk{OOìîª=CD¬D‰%èVH=@5.oø~Óý¹–¤."bj®ø¬/ëru½ìBxš¹ÔýV¦ô<oØšHíÜ˜]ˆåE²ÌÏbÑš4Š¨2¥“`q#µB1C‹ç×ÅÑ.<ëù6ÄBNÑ¡@@Ë ƒœéi8Áÿ+×I4¿Ãƒ:z%èL	µ]p:Çf‚5#,’µÚ°4 õ.~€ÍÀÉÔn
OýHjê"Î1”¨î}ïz˜õ´—uy•Ë…úÁÄû~rCÐc)}w5Š¦„î›ô4uÃ'QšïÛÐ­ôÂmf–Êq–©©.™ÌQ¼ÀÖL‚ŸƒùÇ@›ëˆ”¶íAÜ*]	p)”°ºàiÙãÿF¡SÏ‡?ã-3Œ§¸}(b°ÈFÝ aòú*]$üëý"µúÐIôösÿÎ¸µíªð¢¾¬|$D³‚Žü”"T\Ë¯c}u¥ŸE†Êì8Cß U›eÔ±
I•ð„LüÏÄ+çïõ2|ƒGÓKaÙ‹Iá«XÌ“8œØi=a{y.èÃÞÈKf\y×î§d[Mö‰j`ê~ð8#üz™oðvŸvõp×¼p8xâ[iïo#Ò!8‹iz˜UsöN¡ó5ÛÇàjÎ”“¾5—Œ§ZÅu[N‘¨æÇ-¶“°KY°-r‰Zô5JÉ(í–*¦í3#YWÌý¥þ(wG"XŠ|^
}~±‚Â[$ÎÿjU1°VïsN?Èæ˜ëñ £·Xâ‡5OË¬g¾-ãÆ»Â0)èOEiý|Æ³ê@§lÆÞ¬\êeyCÂqYãyÖ9THWdÓÈGª·ôÜêÐø$kª4ð7ƒòöFq[«~f¦ê²abÔ›ÃpÆh÷DçÂÚºŠ-w°†ù
%ÕÀòýp¹Ö¿à¢MÀ¡÷<a½»RXê©s²'fo(ÙgSGnØøÅA]|ZR£¦fšõÄ®õYëaÑYãlä<õœâÍB·?9¯ù’‚.ÆÓ49œŒ»Zìñ¡Âl@>žJ©UçM¹DQý¥	Çª‰Bgt<÷¾ù·Ç<ieÛEà2 Þÿ—!¸R|;à˜jA)ZåD‰*¾)ÞÃž˜šƒÀÎ{Ÿ({È’¡À,Ævóÿy&+`ãçPjh`²eaÓ¿ì†®ufÂMÄ$¶Sëùc$ôÜ…9æ4SŠhQô$r$UÑ£B³ŸçGÃW’;îF¨Þn«Îpo7’óüô ¶”'¢2QŸD^å
(ô&¨jLâ6b8˜.àÌ•‹¿?P†¾O9.ÜYÃÆFFpŸ½l|K1c«˜ŒÑ:â“˜iÇ‚K‘Ñï­zÁ–õ³…ôM¶ù°ÄA˜	¤¾h$Í^”ö¢¥“,jØ‹¿i¯=ôåÍSÔëÉð¿»ë´ÿ*wÌ½ÓÆŸÛ=çnø;õ:“ƒê‚SžPáóí„g™!´pšâaÜáèQÖ‘Áˆl.¬¡`ÍC?`&êñ­ùþæíècrRÐSä°?úw½ÊgÖFœN5¶·Ký]O°ÚïÜ™§DÎ­`b?ù'¿8íkÙÆf^Ø¯úÊµu¹—œ)c|(qs ¥œîGq)KfGÚ
)
ÆN;îŠaÓøõ5,¶Ph¹-|Laûß¶U¢Ò‰H\­oTWàt÷mý×‹Aö›XÒ¥øÏ‰ëNãÿßI7ˆyÞWd„×¾5£n'÷h—¢è*~ÖÍÓÆô©ÏûZ1ýó™zjñQ‰W>/ä6žþŠ–Éà÷Üµ$z«¬ÄfS¨‘§àoSé“˜Þ{â*â<w#!ÃÝ=AôÕKD\«a[Y¼»%‘p(“°8:‡E4ë~¸qI*è ³P/4 ¹÷I ×Ìftµ®šezKícDHC2@Ñj¯!¢Ál5bk½Ã”‡u~ÿ’íohæ#n]áÑÊ¸—T¢1	ƒ”ZQ—Ÿá„díU‰ùå¯	cŽ
ñÀKd¤CÁ¢>°½0AîÅbˆùN¼%sá,	¶sMô8é+-2‘ýŠßºä*Ùs¹±éõd¢ƒâ€ËuÛ?a3Æ|bŒgoN…¼(ÒâÑC8DékT2ø£Î×ð7a¬pç»ÜXù˜þâ]{µÞÑºù}?ƒ±³K†T˜T–/A„'O`0`Sƒw·á‚FÕxñÏb4Â„ÓùQ»­Ç±*È>U_te€.ý\ãçÍà	ô’sÈd1à*!>z€ß¤«<ð·Öl©sÚî&Ä¦ÒœÄw?þ2LAËÙô8¶MÌ©Å@'± \3UHï=pÞ1Ä±?jôhr&Ow»NE4ç&Íæ8¬hãqàÀµex6äÿ6Û.\58úÉDº¤&À»”#E,íÚLnW¡gäºâÅCýÙÛÞYFŸÇ@“eú’3¬Ïíï¡Pñ¨â5ÿ—Jt£Dôô~'¿T—ÚGkÖÎNzAß÷5”£="¢¸èûõ%ú õ´g³òÞZCêœ¼ÛtäŸ·—T	á*ˆí5Ý8  $tok = $this->scanner->next();

            if ('!' === $tok) {
                $this->markupDeclaration();
            } elseif ('/' === $tok) {
                $this->endTag();
            } elseif ('?' === $tok) {
                $this->processingInstruction();
            } elseif (ctype_alpha($tok)) {
                $this->tagName();
            } else {
                $this->parseError('Illegal tag opening');
                // TODO is this necessary ?
                $this->characterData();
            }

            $tok = $this->scanner->current();
        }

        if (false === $tok) {
            // Handle end of document
            $this->eof();
        } else {
            // Parse character
            switch ($this->textMode) {
                case Elements::TEXT_RAW:
                    $this->rawText($tok);
                    break;

                case Elements::TEXT_RCDATA:
                    $this->rcdata($tok);
                    break;

                default:
                    if ('<' === $tok || '&' === $tok) {
                        break;
                    }

                    // NULL character
                    if ("\00" === $tok) {
                        $this->parseError('Received null character.');

                        $this->text .= $tok;
                        $this->scanner->consume();

                        break;
                    }

                    $this->text .= $this->scanner->charsUntil("<&\0");
            }
        }

        return $this->carryOn;
    }

    /**
     * Parse anything that looks like character data.
     *
     * Different rules apply based on the current text mode.
     *
     * @see Elements::TEXT_RAW Elements::TEXT_RCDATA.
     */
    protected function characterData()
    {
        $tok = $this->scanner->current();
        if (false === $tok) {
            return false;
        }
        switch ($this->textMode) {
            case Elements::TEXT_RAW:
                return $this->rawText($tok);
            case Elements::TEXT_RCDATA:
                return $this->rcdata($tok);
            default:
                if ('<' === $tok || '&' === $tok) {
                    return false;
                }

                return $this->text($tok);
        }
    }

    /**
     * This buffers the current token as character data.
     *
     * @param string $tok The current token.
     *
     * @return bool
     */
    protected function text($tok)
    {
        // This should never happen...
        if (false === $tok) {
            return false;
        }

        // NULL character
        if ("\00" === $tok) {
            $this->parseError('Received null character.');
        }

        $this->buffer($tok);
        $this->scanner->consume();

        return true;
    }

    /**
     * Read text in RAW mode.
     *
     * @param string $tok The current token.
     *
     * @return bool
     */
    protected function rawText($tok)
    {
        if (is_null($this->untilTag)) {
            return $this->text($tok);
        }

        $sequence = '</' . $this->untilTag . '>';
        $txt = $this->readUntilSequence($sequence);
        $this->events->text($txt);
        $this->setTextMode(0);

        return $this->endTag();
    }

    /**
     * Read text in RCDATA mode.
     *
     * @param string $tok The current token.
     *
     * @return bool
     */
    protected function rcdata($tok)
    {
        if (is_null($this->untilTag)) {
            return $this->text($tok);
        }

        $sequence = '</' . $this->untilTag;
        $txt = '';

        $caseSensitive = !Elements::isHtml5Element($this->untilTag);
        while (false !== $tok && !('<' == $tok && ($this->scanner->sequenceMatches($sequence, $caseSensitive)))) {
            if ('&' == $tok) {
                $txt .= $this->decodeCharacterReference();
                $tok = $this->scanner->current();
            } else {
                $txt .= $tok;
                $tok = $this->scanner->next();
            }
        }
        $len = strlen($sequence);
        $this->scanner->consume($len);
        $len += $this->scanner->whitespace();
        if ('>' !== $this->scanner->current()) {
            $this->parseError('Unclosed RCDATA end tag');
        }

        $this->scanner->unconsume($len);
        $this->events->text($txt);
        $this->setTextMode(0);

        return $this->endTag();
    }

    /**
     * If the document is read, emit an EOF event.
     */
    protected function eof()
    {
        // fprintf(STDOUT, "EOF");
        $this->flushBuffer();
        $this->events->eof();
        $this->carryOn = false;
    }

    /**
     * Look for markup.
     */
    protected function markupDeclaration()
    {
        $tok = $this->scanner->next();

        // Comment:
        if ('-' == $tok && '-' == $this->scanner->peek()) {
            $this->scanner->consume(2);

            return $this->comment();
        } elseif ('D' == $tok || 'd' == $tok) { // Doctype
            return $this->doctype();
        } elseif ('[' == $tok) { // CDATA section
            return $this->cdataSection();
        }

        // FINISH
        $this->parseError('Expected <!--, <![CDATA[, or <!DOCTYPE. Got <!%s', $tok);
        $this->bogusComment('<!');

        return true;
    }

    /**
     * Consume an end tag. See section 8.2.4.9.
     */
    protected function endTag()
    {
        if ('/' != $this->scanner->current()) {
            return false;
        }
        $tok = $this->scanner->next();

        // a-zA-Z -> tagname
        // > -> parse error
        // EOF -> parse error
        // -> parse error
        if (!ctype_alpha($tok)) {
            $this->parseError("Expected tag name, got '%s'", $tok);
            if ("\0" == $tok || false === $tok) {
                return false;
            }

            return $this->bogusComment('</');
        }

        $name = $this->scanner->charsUntil("\n\f \t>");
        $name = self::CONFORMANT_XML === $this->mode ? $name : strtolower($name);
        // Trash whitespace.
        $this->scanner->whitespace();

        $tok = $this->scanner->current();
        if ('>' != $tok) {
            $this->parseError("Expected >, got '%s'", $tok);
            // We just trash stuff until we get to the next tag close.
            $this->scanner->charsUntil('>');
        }

        $this->events->endTag($name);
        $this->scanner->consume();

        return true;
    }

    /**
     * Consume a tag name and body. See section 8.2.4.10.
     */
    protected function tagName()
    {
        // We know this is at least one char.
        $name = $this->scanner->charsWhile(':_-0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz');
        $name = self::CONFORMANT_XML === $this->mode ? $name : strtolower($name);
        $attributes = array();
        $selfClose = false;

        // Handle attribute parse exceptions here so that we can
        // react by trying to build a sensible parse tree.
        try {
            do {
                $this->scanner->whitespace();
                $this->attribute($attributes);
            } while (!$this->isTagEnd($selfClose));
        } catch (ParseError $e) {
            $selfClose = false;
        }

        $mode = $this->events->startTag($name, $attributes, $selfClose);

        if (is_int($mode)) {
            $this->setTextMode($mode, $name);
        }

        $this->scanner->consume();

        return true;
    }

    /**
     * Check if the scanner has reached the end of a tag.
     */
    protected function isTagEnd(&$selfClose)
    {
        $tok = $this->scanner->current();
        if ('/' == $tok) {
            $this->scanner->consume();
            $this->scanner->whitespace();
            $tok = $this->scanner->current();

            if ('>' == $tok) {
                $selfClose = true;

                return true;
            }
            if (false === $tok) {
                $this->parseError('Unexpected EOF inside of tag.');

                return true;
            }
            // Basically, we skip the / token and go on.
            // See 8.2.4.43.
            $this->parseError("Unexpected '%s' inside of a tag.", $tok);

            return false;
        }

        if ('>' == $tok) {
            return true;
        }
        if (false === $tok) {
            $this->parseError('Unexpected EOF inside of tag.');

            return true;
        }

        return false;
    }

    /**
     * Parse attributes from inside of a tag.
     *
     * @param string[] $attributes
     *
     * @return bool
     *
     * @throws ParseError
     */
    protected function attribute(&$attributes)
    {
        $tok = $this->scanner->current();
        if ('/' == $tok || '>' == $tok || false === $tok) {
            return false;
        }

        if ('<' == $tok) {
            $this->parseError("Unexpected '<' inside of attributes list.");
            // Push the < back onto the stack.
            $this->scanner->unconsume();
            // Let the caller figure out how to handle this.
            throw new ParseError('Start tag inside of attribute.');
        }

        $name = strtolower($this->scanner->charsUntil("/>=\n\f\t "));

        if (0 == strlen($name)) {
            $tok = $this->scanner->current();
            $this->parseError('Expected an attribute name, got %s.', $tok);
            // Really, only '=' can be the char here. Everything else gets absorbed
            // under one rule or another.
            $name = $tok;
            $this->scanner->consume();
        }

        $isValidAttribute = true;
        // Attribute names can contain most Unicode characters for HTML5.
        // But method "DOMElement::setAttribute" is throwing exception
        // because of it's own internal restriction so these have to be filtered.
        // see issue #23: https://github.com/Masterminds/html5-php/issues/23
        // and http://www.w3.org/TR/2011/WD-html5-20110525/syntax.html#syntax-attribute-name
        if (preg_match("/[\x1-\x2C\\/\x3B-\x40\x5B-\x5E\x60\x7B-\x7F]/u", $name)) {
            $this->parseError('Unexpected characters in attribute name: %s', $name);
            $isValidAttribute = false;
        }         // There is no limitation for 1st character in HTML5.
        // But method "DOMElement::setAttribute" is throwing exception for the
        // characters below so they have to be filtered.
        // see issue #23: https://github.com/Masterminds/html5-php/issues/23
        // and http://www.w3.org/TR/2011/WD-html5-20110525/syntax.html#syntax-attribute-name
        elseif (preg_match('/^[0-9.-]/u', $name)) {
            $this->parseError('Unexpected character at the begining of attribute name: %s', $name);
            $isValidAttribute = false;
        }
        // 8.1.2.3
        $this->scanner->whitespace();

        $val = $this->attributeValue();
        if ($isValidAttribute) {
            $attributes[$name] = $val;
        }

        return true;
    }

    /**
     * Consume an attribute value. See section 8.2.4.37 and after.
     *
     * @return string|null
     */
    protected function attributeValue()
    {
        if ('=' != $this->scanner->current()) {
            return null;
        }
        $this->scanner->consume();
        // 8.1.2.3
        $this->scanner->whitespace();

        $tok = $this->scanner->current();
        switch ($tok) {
            case "\n":
            case "\f":
            case ' ':
            case "\t":
                // Whitespace here indicates an empty value.
                return null;
            case '"':
            case "'":
                $this->scanner->consume();

                return $this->quotedAttributeValue($tok);
            case '>':
                // case '/': // 8.2.4.37 seems to allow foo=/ as a valid attr.
                $this->parseError('Expected attribute value, got tag end.');

                return null;
            case '=':
            case '`':
                $this->parseError('Expecting quotes, got %s.', $tok);

                return $this->unquotedAttributeValue();
            default:
                return $this->unquotedAttributeValue();
        }
    }

    /**
     * Get an attribute value string.
     *
     * @param string $quote IMPORTANT: This is a series of chars! Any one of which will be considered
     *                      termination of an attribute's value. E.g. "\"'" will stop at either
     *                      ' or ".
     *
     * @return string The attribute value.
     */
    protected function quotedAttributeValue($quote)
    {
        $stoplist = "\f" . $quote;
        $val = '';

        while (true) {
            $tokens = $this->scanner->charsUntil($stoplist . '&');
            if (false !== $tokens) {
                $val .= $tokens;
            } else {
                break;
            }

            $tok = $this->scanner->current();
            if ('&' == $tok) {
                $val .= $this->decodeCharacterReference(true);
                continue;
            }
            break;
        }
        $this->scanner->consume();

        return $val;
    }

    protected function unquotedAttributeValue()
    {
        $val = '';
        $tok = $this->scanner->current();
        while (false !== $tok) {
            switch ($tok) {
                case "\n":
                case "\f":
                case ' ':
                case "\t":
                case '>':
                    break 2;

                case '&':
                    $val .= $this->decodeCharacterReference(true);
                    $tok = $this->scanner->current();

                    break;

                case "'":
                case '"':
                case '<':
                case '=':
                case '`':
                    $this->parseError('Unexpected chars in unquoted attribute value %s', $tok);
                    $val .= $tok;
                    $tok = $this->scanner->next();
                    break;

                default:
                    $val .= $this->scanner->charsUntil("\t\n\f >&\"'<=`");

                    $tok = $this->scanner->current();
            }
        }

        return $val;
    }

    /**
     * Consume malformed markup as if it were a comment.
     * 8.2.4.44.
     *
     * The spec requires that the ENTIRE tag-like thing be enclosed inside of
     * the comment. So this will generate comments like:
     *
     * &lt;!--&lt/+foo&gt;--&gt;
     *
     * @param string $leading Prepend any leading characters. This essentially
     *                        negates the need to backtrack, but it's sort of a hack.
     *
     * @return bool
     */
    protected function bogusComment($leading = '')
    {
        $comment = $leading;
        $tokens = $this->scanner->charsUntil('>');
        if (false !== $tokens) {
            $comment .= $tokens;
        }
        $tok = $this->scanner->current();
        if (false !== $tok) {
            $comment .= $tok;
        }

        $this->flushBuffer();
        $this->events->comment($comment);
        $this->scanner->consume();

        return true;
    }

    /**
     * Read a comment.
     * Expects the first tok to be inside of the comment.
     *
     * @return bool
     */
    protected function comment()
    {
        $tok = $this->scanner->current();
        $comment = '';

        // <!-->. Emit an empty comment because 8.2.4.46 says to.
        if ('>' == $tok) {
            // Parse error. Emit the comment token.
            $this->parseError("Expected comment data, got '>'");
            $this->events->comment('');
            $this->scanner->consume();

            return true;
        }

        // Replace NULL with the replacement char.
        if ("\0" == $tok) {
            $tok = UTF8Utils::FFFD;
        }
        while (!$this->isCommentEnd()) {
            $comment .= $tok;
            $tok = $this->scanner->next();
        }

        $this->events->comment($comment);
        $this->scanner->consume();

        return true;
    }

    /**
     * Check if the scanner has reached the end of a comment.
     *
     * @return bool
     */
    protected function isCommentEnd()
    {
        $tok = $this->scanner->current();

        // EOF
        if (false === $tok) {
            // Hit the end.
            $this->parseError('Unexpected EOF in a comment.');

            return true;
        }

        // If next two tokens are not '--', not the end.
        if ('-' != $tok || '-' != $this->scanner->peek()) {
            return false;
        }

        $this->scanner->consume(2); // Consume '-' and one of '!' or '>'

        // Test for '>'
        if ('>' == $this->scanner->current()) {
            return true;
        }
        // Test for '!>'
        if ('!' == $this->scanner->current() && '>' == $this->scanner->peek()) {
            $this->scanner->consume(); // Consume the last '>'
            return true;
        }
        // Unread '-' and one of '!' or '>';
        $this->scanner->unconsume(2);

        return false;
    }

    /**
     * Parse a DOCTYPE.
     *
     * Parse a DOCTYPE declaration. This method has strong bearing on whether or
     * not Quirksmode is enabled on the event handler.
     *
     * @todo This method is a little long. Should probably refactor.
     *
     * @return bool
     */
    protected function doctype()
    {
        // Check that string is DOCTYPE.
        if ($this->scanner->sequenceMatches('DOCTYPE', false)) {
            $this->scanner->consume(7);
        } else {
            $chars = $this->scanner->charsWhile('DOCTYPEdoctype');
            $this->parseError('Expected DOCTYPE, got %s', $chars);

            return $this->bogusComment('<!' . $chars);
        }

        $this->scanner->whitespace();
        $tok = $this->scanner->current();

        // EOF: die.
        if (false === $tok) {
            $this->events->doctype('html5', EventHandler::DOCTYPE_NONE, '', true);
            $this->eof();

            return true;
        }

        // NULL char: convert.
        if ("\0" === $tok) {
            $this->parseError('Unexpected null character in DOCTYPE.');
        }

        $stop = " \n\f>";
        $doctypeName = $this->scanner->charsUntil($stop);
        // Lowercase ASCII, replace \0 with FFFD
        $doctypeName = strtolower(strtr($doctypeName, "\0", UTF8Utils::FFFD));

        $tok = $this->scanner->current();

        // If false, emit a parse error, DOCTYPE, and return.
        if (false === $tok) {
            $this->parseError('Unexpected EOF in DOCTYPE declaration.');
            $this->events->doctype($doctypeName, EventHandler::DOCTYPE_NONE, null, true);

            return true;
        }

        // Short DOCTYPE, like <!DOCTYPE html>
        if ('>' == $tok) {
            // DOCTYPE without a name.
            if (0 == strlen($doctypeName)) {
                $this->parseError('Expected a DOCTYPE name. Got nothing.');
                $this->events->doctype($doctypeName, 0, null, true);
                $this->scanner->consume();

                return true;
            }
            $this->events->doctype($doctypeName);
            $this->scanner->consume();

            return true;
        }
        $this->scanner->whitespace();

        $pub = strtoupper($this->scanner->getAsciiAlpha());
        $white = $this->scanner->whitespace();

        // Get ID, and flag it as pub or system.
        if (('PUBLIC' == $pub || 'SYSTEM' == $pub) && $white > 0) {
            // Get the sys ID.
            $type = 'PUBLIC' == $pub ? EventHandler::DOCTYPE_PUBLIC : EventHandler::DOCTYPE_SYSTEM;
            $id = $this->quotedString("\0>");
            if (false === $id) {
                $this->events->doctype($doctypeName, $type, $pub, false);

                return true;
            }

            // Premature EOF.
            if (false === $this->scanner->current()) {
                $this->parseError('Unexpected EOF in DOCTYPE');
                $this->events->doctype($doctypeName, $type, $id, true);

                return true;
            }

            // Well-formed complete DOCTYPE.
            $this->scanner->whitespace();
            if ('>' == $this->scanner->current()) {
                $this->events->doctype($doctypeName, $type, $id, false);
                $this->scanner->consume();

                return true;
            }

            // If we get here, we have <!DOCTYPE foo PUBLIC "bar" SOME_JUNK
            // Throw away the junk, parse error, quirks mode, return true.
            $this->scanner->charsUntil('>');
            $this->parseError('Malformed DOCTYPE.');
            $this->events->doctype($doctypeName, $type, $id, true);
            $this->scanner->consume();

            return true;
        }

        // Else it's a bogus DOCTYPE.
        // Consume to > and trash.
        $this->scanner->charsUntil('>');

        $this->parseError('Expected PUBLIC or SYSTEM. Got %s.', $pub);
        $this->events->doctype($doctypeName, 0, null, true);
        $this->scanner->consume();

        return true;
    }

    /**
     * Utility for reading a quoted string.
     *
     * @param string $stopchars Characters (in addition to a close-quote) that should stop the string.
     *                          E.g. sometimes '>' is higher precedence than '"' or "'".
     *
     * @return mixed String if one is found (quotations omitted).
     */
    protected function quotedString($stopchars)
    {
        $tok = $this->scanner->current();
        if ('"' == $tok || "'" == $tok) {
            $this->scanner->consume();
            $ret = $this->scanner->charsUntil($tok . $stopchars);
            if ($this->scanner->current() == $tok) {
                $this->scanner->consume();
            } else {
                // Parse error because no close quote.
                $this->parseError('Expected %s, got %s', $tok, $this->scanner->current());
            }

            return $ret;
        }

        return false;
    }

    /**
     * Handle a CDATA section.
     *
     * @return bool
     */
    protected function cdataSection()
    {
        $cdata = '';
        $this->scanner->consume();

        $chars = $this->scanner->charsWhile('CDAT');
        if ('CDATA' != $chars || '[' != $this->scanner->current()) {
            $this->parseError('Expected [CDATA[, got %s', $chars);

            return $this->bogusComment('<![' . $chars);
        }

        $tok = $this->scanner->next();
        do {
            if (false === $tok) {
                $this->parseError('Unexpected EOF inside CDATA.');
                $this->bogusComment('<![CDATA[' . $cdata);

                return true;
            }
            $cdata .= $tok;
            $tok = $this->scanner->next();
        } while (!$this->scanner->sequenceMatches(']]>'));

        // Consume ]]>
        $this->scanner->consume(3);

        $this->events->cdata($cdata);

        return true;
    }

    // ================================================================
    // Non-HTML5
    // ================================================================

    /**
     * Handle a processing instruction.
     *
     * XML processing instructions are supposed to be ignored in HTML5,
     * treated as "bogus comments". However, since we're not a user
     * agent, we allow them. We consume until ?> and then issue a
     * EventListener::processingInstruction() event.
     *
     * @return bool
     */
    protected function processingInstruction()
    {
        if ('?' != $this->scanner->current()) {
            return false;
        }

        $tok = $this->scanner->next();
        $procName = $this->scanner->getAsciiAlpha();
        $white = $this->scanner->whitespace();

        // If not a PI, send to bogusComment.
        if (0 == strlen($procName) || 0 == $white || false == $this->scanner->current()) {
            $this->parseError("Expected processing instruction name, got $tok");
            $this->bogusComment('<?' . $tok . $procName);

            return true;
        }

        $data = '';
        // As long as it's not the case that the next two chars are ? and >.
        while (!('?' == $this->scanner->current() && '>' == $this->scanner->peek())) {
            $data .= $this->scanner->current();

            $tok = $this->scanner->next();
            if (false === $tok) {
                $this->parseError('Unexpected EOF in processing instruction.');
                $this->events->processingInstruction($procName, $data);

                return true;
            }
        }

        $this->scanner->consume(2); // Consume the closing tag
        $this->events->processingInstruction($procName, $data);

        return true;
    }

    // ================================================================
    // UTILITY FUNCTIONS
    // ================================================================

    /**
     * Read from the input stream until we get to the desired sequene
     * or hit the end of the input stream.
     *
     * @param string $sequence
     *
     * @return string
     */
    protected function readUntilSequence($sequence)
    {
        $buffer = '';

        // Optimization for reading larger blocks faster.
        $first = substr($sequence, 0, 1);
        while (false !== $this->scanner->current()) {
            $buffer .= $this->scanner->charsUntil($first);

            // Stop as soon as we hit the stopping condition.
            if ($this->scanner->sequenceMatches($sequence, false)) {
                return $buffer;
            }
            $buffer .= $this->scanner->current();
            $this->scanner->consume();
        }

        // If we get here, we hit the EOF.
        $this->parseError('Unexpected EOF during text read.');

        return $buffer;
    }

    /**
     * Check if upcomming chars match the given sequence.
     *
     * This will read the stream for the $sequence. If it's
     * found, this will return true. If not, return false.
     * Since this unconsumes any chars it reads, the caller
     * will still need to read the next sequence, even if
     * this returns true.
     *
     * Example: $this->scanner->sequenceMatches('</script>') will
     * see if the input stream is at the start of a
     * '</script>' string.
     *
     * @param string $sequence
     * @param bool   $caseSensitive
     *
     * @return bool
     */
    protected function sequenceMatches($sequence, $caseSensitive = true)
    {
        @trigger_error(__METHOD__ . ' method is deprecated since version 2.4 and will be removed in 3.0. Use Scanner::sequenceMatches() instead.', E_USER_DEPRECATED);

        return $this->scanner->sequenceMatches($sequence, $caseSensitive);
    }

    /**
     * Send a TEXT event with the contents of the text buffer.
     *
     * This emits an EventHandler::text() event with the current contents of the
     * temporary text buffer. (The buffer is used to group as much PCDATA
     * as we can instead of emitting lots and lots of TEXT events.)
     */
    protected function flushBuffer()
    {
        if ('' === $this->text) {
            return;
        }
        $this->events->text($this->text);
        $this->text = '';
    }

    /**
     * Add text to the temporary buffer.
     *
     * @see flushBuffer()
     *
     * @param string $str
     */
    protected function buffer($str)
    {
        $this->text .= $str;
    }

    /**
     * Emit a parse error.
     *
     * A parse error always returns false because it never consumes any
     * characters.
     *
     * @param string $msg
     *
     * @return string
     */
    protected function parseError($msg)
    {
        $args = func_get_args();

        if (count($args) > 1) {
            array_shift($args);
            $msg = vsprintf($msg, $args);
        }

        $line = $this->scanner->currentLine();
        $col = $this->scanner->columnOffset();
        $this->events->parseError($msg, $line, $col);

        return false;
    }

    /**
     * Decode a character reference and return the string.
     *
     * If $inAttribute is set to true, a bare & will be returned as-is.
     *
     * @param bool $inAttribute Set to true if the text is inside of an attribute value.
     *                          false otherwise.
     *
     * @return string
     */
    protected function decodeCharacterReference($inAttribute = false)
    {
        // Next char after &.
        $tok = $this->scanner->next();
        $start = $this->scanner->position();

        if (false === $tok) {
            return '&';
        }

        // These indicate not an entity. We return just
        // the &.
        if ("\t" === $tok || "\n" === $tok || "\f" === $tok || ' ' === $tok || '&' === $tok || '<' === $tok) {
            // $this->scanner->next();
            return '&';
        }

        // Numeric entity
        if ('#' === $tok) {
            $tok = $this->scanner->next();

            if (false === $tok) {
                $this->parseError('Expected &#DEC; &#HEX;, got EOF');
                $this->scanner->unconsume(1);

                return '&';
            }

            // Hexidecimal encoding.
            // X[0-9a-fA-F]+;
            // x[0-9a-fA-F]+;
            if ('x' === $tok || 'X' === $tok) {
                $tok = $this->scanner->next(); // Consume x

                // Convert from hex code to char.
                $hex = $this->scanner->getHex();
                if (empty($hex)) {
                    $this->parseError('Expected &#xHEX;, got &#x%s', $tok);
                    // We unconsume because we don't know what parser rules might
                    // be in effect for the remaining chars. For example. '&#>'
                    // might result in a specific parsing rule inside of tag
                    // contexts, while not inside of pcdata context.
                    $this->scanner->unconsume(2);

                    return '&';
                }
                $entity = CharacterReference::lookupHex($hex);
            }             // Decimal encoding.
            // [0-9]+;
            else {
                // Convert from decimal to char.
                $numeric = $this->scanner->getNumeric();
                if (false === $numeric) {
                    $this->parseError('Expected &#DIGITS;, got &#%s', $tok);
                    $this->scanner->unconsume(2);

                    return '&';
                }
                $entity = CharacterReference::lookupDecimal($numeric);
            }
        } elseif ('=' === $tok && $inAttribute) {
            return '&';
        } else { // String entity.
            // Attempt to consume a string up to a ';'.
            // [a-zA-Z0-9]+;
            $cname = $this->scanner->getAsciiAlphaNum();
            $entity = CharacterReference::lookupName($cname);

            // When no entity is found provide the name of the unmatched string
            // and continue on as the & is not part of an entity. The & will
            // be converted to &amp; elsewhere.
            if (null === $entity) {
                if (!$inAttribute || '' === $cname) {
                    $this->parseError("No match in entity table for '%s'", $cname);
                }
                $this->scanner->unconsume($this->scanner->position() - $start);

                return '&';
            }
        }

        // The scanner has advanced the cursor for us.
        $tok = $this->scanner->current();

        // We have an entity. We're done here.
        if (';' === $tok) {
            $this->scanner->consume();

            return $entity;
        }

        // Failing to match ; means unconsume the entire string.
        $this->scanner->unconsume($this->scanner->position() - $start);

        $this->parseError('Expected &ENTITY;, got &ENTITY%s (no trailing ;) ', $tok);

        return '&';
    }
}