Balancer.php

Go to the documentation of this file.

<?php
namespace MediaWiki\Tidy;

use Wikimedia\Assert\Assert;
use Wikimedia\Assert\ParameterAssertionException;
use \ExplodeIterator;
use \IteratorAggregate;
use \ReverseArrayIterator;
use \Sanitizer;

# A note for future librarization[1] -- this file is a good candidate
# for splitting into an independent library, except that it is currently
# highly optimized for MediaWiki use.  It only implements the portions
# of the HTML5 tree builder used by tags supported by MediaWiki, and
# does not contain a true tokenizer pass, instead relying on
# comment stripping, attribute normalization, and escaping done by
# the MediaWiki Sanitizer.  It also deliberately avoids building
# a true DOM in memory, instead serializing elements to an output string
# as soon as possible (usually as soon as the tag is closed) to reduce
# its memory footprint.

# We've been gradually lifting some of these restrictions to handle
# non-sanitized output generated by extensions, but we shortcut the tokenizer
# for speed (primarily by splitting on `<`) and so rely on syntactic
# well-formedness.

# On the other hand, I've been pretty careful to note with comments in the
# code the places where this implementation omits features of the spec or
# depends on the MediaWiki Sanitizer.  Perhaps in the future we'll want to
# implement the missing pieces and make this a standalone PHP HTML5 parser.
# In order to do so, some sort of MediaWiki-specific API will need
# to be added to (a) allow the Balancer to bypass the tokenizer,
# and (b) support on-the-fly flattening instead of DOM node creation.

# [1]: https://www.mediawiki.org/wiki/Library_infrastructure_for_MediaWiki

class BalanceSets {
    const HTML_NAMESPACE = 'http://www.w3.org/1999/xhtml';
    const MATHML_NAMESPACE = 'http://www.w3.org/1998/Math/MathML';
    const SVG_NAMESPACE = 'http://www.w3.org/2000/svg';

    public static $unsupportedSet = [
        self::HTML_NAMESPACE => [
            'html' => true, 'head' => true, 'body' => true, 'frameset' => true,
            'frame' => true,
            'plaintext' => true, 'isindex' => true,
            'xmp' => true, 'iframe' => true, 'noembed' => true,
            'noscript' => true, 'script' => true,
            'title' => true
        ]
    ];

    public static $emptyElementSet = [
        self::HTML_NAMESPACE => [
            'area' => true, 'base' => true, 'basefont' => true,
            'bgsound' => true, 'br' => true, 'col' => true, 'command' => true,
            'embed' => true, 'frame' => true, 'hr' => true, 'img' => true,
            'input' => true, 'keygen' => true, 'link' => true, 'meta' => true,
            'param' => true, 'source' => true, 'track' => true, 'wbr' => true
        ]
    ];

    public static $extraLinefeedSet = [
        self::HTML_NAMESPACE => [
            'pre' => true, 'textarea' => true, 'listing' => true,
        ]
    ];

    public static $headingSet = [
        self::HTML_NAMESPACE => [
            'h1' => true, 'h2' => true, 'h3' => true,
            'h4' => true, 'h5' => true, 'h6' => true
        ]
    ];

    public static $specialSet = [
        self::HTML_NAMESPACE => [
            'address' => true, 'applet' => true, 'area' => true,
            'article' => true, 'aside' => true, 'base' => true,
            'basefont' => true, 'bgsound' => true, 'blockquote' => true,
            'body' => true, 'br' => true, 'button' => true, 'caption' => true,
            'center' => true, 'col' => true, 'colgroup' => true, 'dd' => true,
            'details' => true, 'dir' => true, 'div' => true, 'dl' => true,
            'dt' => true, 'embed' => true, 'fieldset' => true,
            'figcaption' => true, 'figure' => true, 'footer' => true,
            'form' => true, 'frame' => true, 'frameset' => true, 'h1' => true,
            'h2' => true, 'h3' => true, 'h4' => true, 'h5' => true,
            'h6' => true, 'head' => true, 'header' => true, 'hgroup' => true,
            'hr' => true, 'html' => true, 'iframe' => true, 'img' => true,
            'input' => true, 'isindex' => true, 'li' => true, 'link' => true,
            'listing' => true, 'main' => true, 'marquee' => true,
            'menu' => true, 'menuitem' => true, 'meta' => true, 'nav' => true,
            'noembed' => true, 'noframes' => true, 'noscript' => true,
            'object' => true, 'ol' => true, 'p' => true, 'param' => true,
            'plaintext' => true, 'pre' => true, 'script' => true,
            'section' => true, 'select' => true, 'source' => true,
            'style' => true, 'summary' => true, 'table' => true,
            'tbody' => true, 'td' => true, 'template' => true,
            'textarea' => true, 'tfoot' => true, 'th' => true, 'thead' => true,
            'title' => true, 'tr' => true, 'track' => true, 'ul' => true,
            'wbr' => true, 'xmp' => true
        ],
        self::SVG_NAMESPACE => [
            'foreignobject' => true, 'desc' => true, 'title' => true
        ],
        self::MATHML_NAMESPACE => [
            'mi' => true, 'mo' => true, 'mn' => true, 'ms' => true,
            'mtext' => true, 'annotation-xml' => true
        ]
    ];

    public static $addressDivPSet = [
        self::HTML_NAMESPACE => [
            'address' => true, 'div' => true, 'p' => true
        ]
    ];

    public static $tableSectionRowSet = [
        self::HTML_NAMESPACE => [
            'table' => true, 'thead' => true, 'tbody' => true,
            'tfoot' => true, 'tr' => true
        ]
    ];

    public static $impliedEndTagsSet = [
        self::HTML_NAMESPACE => [
            'dd' => true, 'dt' => true, 'li' => true, 'optgroup' => true,
            'option' => true, 'p' => true, 'rb' => true, 'rp' => true,
            'rt' => true, 'rtc' => true
        ]
    ];

    public static $thoroughImpliedEndTagsSet = [
        self::HTML_NAMESPACE => [
            'caption' => true, 'colgroup' => true, 'dd' => true, 'dt' => true,
            'li' => true, 'optgroup' => true, 'option' => true, 'p' => true,
            'rb' => true, 'rp' => true, 'rt' => true, 'rtc' => true,
            'tbody' => true, 'td' => true, 'tfoot' => true, 'th' => true,
            'thead' => true, 'tr' => true
        ]
    ];

    public static $tableCellSet = [
        self::HTML_NAMESPACE => [
            'td' => true, 'th' => true
        ]
    ];
    public static $tableContextSet = [
        self::HTML_NAMESPACE => [
            'table' => true, 'template' => true, 'html' => true
        ]
    ];

    public static $tableBodyContextSet = [
        self::HTML_NAMESPACE => [
            'tbody' => true, 'tfoot' => true, 'thead' => true,
            'template' => true, 'html' => true
        ]
    ];

    public static $tableRowContextSet = [
        self::HTML_NAMESPACE => [
            'tr' => true, 'template' => true, 'html' => true
        ]
    ];

    // See https://html.spec.whatwg.org/multipage/forms.html#form-associated-element
    public static $formAssociatedSet = [
        self::HTML_NAMESPACE => [
            'button' => true, 'fieldset' => true, 'input' => true,
            'keygen' => true, 'object' => true, 'output' => true,
            'select' => true, 'textarea' => true, 'img' => true
        ]
    ];

    public static $inScopeSet = [
        self::HTML_NAMESPACE => [
            'applet' => true, 'caption' => true, 'html' => true,
            'marquee' => true, 'object' => true,
            'table' => true, 'td' => true, 'template' => true,
            'th' => true
        ],
        self::SVG_NAMESPACE => [
            'foreignobject' => true, 'desc' => true, 'title' => true
        ],
        self::MATHML_NAMESPACE => [
            'mi' => true, 'mo' => true, 'mn' => true, 'ms' => true,
            'mtext' => true, 'annotation-xml' => true
        ]
    ];

    private static $inListItemScopeSet = null;
    public static function inListItemScopeSet() {
        if ( self::$inListItemScopeSet === null ) {
            self::$inListItemScopeSet = self::$inScopeSet;
            self::$inListItemScopeSet[self::HTML_NAMESPACE]['ol'] = true;
            self::$inListItemScopeSet[self::HTML_NAMESPACE]['ul'] = true;
        }
        return self::$inListItemScopeSet;
    }

    private static $inButtonScopeSet = null;
    public static function inButtonScopeSet() {
        if ( self::$inButtonScopeSet === null ) {
            self::$inButtonScopeSet = self::$inScopeSet;
            self::$inButtonScopeSet[self::HTML_NAMESPACE]['button'] = true;
        }
        return self::$inButtonScopeSet;
    }

    public static $inTableScopeSet = [
        self::HTML_NAMESPACE => [
            'html' => true, 'table' => true, 'template' => true
        ]
    ];

    public static $inInvertedSelectScopeSet = [
        self::HTML_NAMESPACE => [
            'option' => true, 'optgroup' => true
        ]
    ];

    public static $mathmlTextIntegrationPointSet = [
        self::MATHML_NAMESPACE => [
            'mi' => true, 'mo' => true, 'mn' => true, 'ms' => true,
            'mtext' => true
        ]
    ];

    public static $htmlIntegrationPointSet = [
        self::SVG_NAMESPACE => [
            'foreignobject' => true,
            'desc' => true,
            'title' => true
        ]
    ];

    // For tidy compatibility.
    public static $tidyPWrapSet = [
        self::HTML_NAMESPACE => [
            'body' => true, 'blockquote' => true,
            // We parse with <body> as the fragment context, but the top-level
            // element on the stack is actually <html>.  We could use the
            // "adjusted current node" everywhere to work around this, but it's
            // easier just to add <html> to the p-wrap set.
            'html' => true,
        ],
    ];
    public static $tidyInlineSet = [
        self::HTML_NAMESPACE => [
            'a' => true, 'abbr' => true, 'acronym' => true, 'applet' => true,
            'b' => true, 'basefont' => true, 'bdo' => true, 'big' => true,
            'br' => true, 'button' => true, 'cite' => true, 'code' => true,
            'dfn' => true, 'em' => true, 'font' => true, 'i' => true,
            'iframe' => true, 'img' => true, 'input' => true, 'kbd' => true,
            'label' => true, 'legend' => true, 'map' => true, 'object' => true,
            'param' => true, 'q' => true, 'rb' => true, 'rbc' => true,
            'rp' => true, 'rt' => true, 'rtc' => true, 'ruby' => true,
            's' => true, 'samp' => true, 'select' => true, 'small' => true,
            'span' => true, 'strike' => true, 'strong' => true, 'sub' => true,
            'sup' => true, 'textarea' => true, 'tt' => true, 'u' => true,
            'var' => true,
        ],
    ];
}

class BalanceElement {
    public $namespaceURI;
    public $localName;
    public $attribs;

    public $parent;

    public $children;

    private $noahKey;

    public $nextAFE;

    public $prevAFE;

    public $nextNoah;

    public function __construct( $namespaceURI, $localName, array $attribs ) {
        $this->localName = $localName;
        $this->namespaceURI = $namespaceURI;
        $this->attribs = $attribs;
        $this->contents = '';
        $this->parent = null;
        $this->children = [];
    }

    private function removeChild( BalanceElement $elt ) {
        Assert::precondition(
            $this->parent !== 'flat', "Can't removeChild after flattening $this"
        );
        Assert::parameter(
            $elt->parent === $this, 'elt', 'must have $this as a parent'
        );
        $idx = array_search( $elt, $this->children, true );
        Assert::parameter( $idx !== false, '$elt', 'must be a child of $this' );
        $elt->parent = null;
        array_splice( $this->children, $idx, 1 );
    }

    public function insertBefore( BalanceElement $a, $b ) {
        Assert::precondition(
            $this->parent !== 'flat', "Can't insertBefore after flattening."
        );
        $idx = array_search( $a, $this->children, true );
        Assert::parameter( $idx !== false, '$a', 'must be a child of $this' );
        if ( is_string( $b ) ) {
            array_splice( $this->children, $idx, 0, [ $b ] );
        } else {
            Assert::parameter( $b->parent !== 'flat', '$b', "Can't be flat" );
            if ( $b->parent !== null ) {
                $b->parent->removeChild( $b );
            }
            array_splice( $this->children, $idx, 0, [ $b ] );
            $b->parent = $this;
        }
    }

    public function appendChild( $elt ) {
        Assert::precondition(
            $this->parent !== 'flat', "Can't appendChild after flattening."
        );
        if ( is_string( $elt ) ) {
            array_push( $this->children, $elt );
            return;
        }
        // Remove $elt from parent, if it had one.
        if ( $elt->parent !== null ) {
            $elt->parent->removeChild( $elt );
        }
        array_push( $this->children, $elt );
        $elt->parent = $this;
    }

    public function adoptChildren( BalanceElement $elt ) {
        Assert::precondition(
            $elt->parent !== 'flat', "Can't adoptChildren after flattening."
        );
        foreach ( $elt->children as $child ) {
            if ( !is_string( $child ) ) {
                // This is an optimization which avoids an O(n^2) set of
                // array_splice operations.
                $child->parent = null;
            }
            $this->appendChild( $child );
        }
        $elt->children = [];
    }

    public function flatten( $tidyCompat = false ) {
        Assert::parameter( $this->parent !== null, '$this', 'must be a child' );
        Assert::parameter( $this->parent !== 'flat', '$this', 'already flat' );
        $idx = array_search( $this, $this->parent->children, true );
        Assert::parameter(
            $idx !== false, '$this', 'must be a child of its parent'
        );
        if ( $tidyCompat ) {
            $blank = true;
            foreach ( $this->children as $elt ) {
                if ( !is_string( $elt ) ) {
                    $elt = $elt->flatten( $tidyCompat );
                }
                if ( $blank && preg_match( '/[^\t\n\f\r ]/', $elt ) ) {
                    $blank = false;
                }
            }
            if ( $this->isHtmlNamed( 'mw:p-wrap' ) ) {
                $this->localName = 'p';
            } elseif ( $blank ) {
                // Add 'mw-empty-elt' class so elements can be hidden via CSS
                // for compatibility with legacy tidy.
                if ( !count( $this->attribs ) &&
                    ( $this->localName === 'tr' || $this->localName === 'li' )
                ) {
                    $this->attribs = [ 'class' => "mw-empty-elt" ];
                }
                $blank = false;
            }
            $flat = $blank ? '' : "{$this}";
        } else {
            $flat = "{$this}";
        }
        $this->parent->children[$idx] = $flat;
        $this->parent = 'flat'; # for assertion checking
        return $flat;
    }

    public function __toString() {
        $encAttribs = '';
        foreach ( $this->attribs as $name => $value ) {
            $encValue = Sanitizer::encodeAttribute( $value );
            $encAttribs .= " $name=\"$encValue\"";
        }
        if ( !$this->isA( BalanceSets::$emptyElementSet ) ) {
            $out = "<{$this->localName}{$encAttribs}>";
            $len = strlen( $out );
            // flatten children
            foreach ( $this->children as $elt ) {
                $out .= "{$elt}";
            }
            $out .= "</{$this->localName}>";
            if (
                $this->isA( BalanceSets::$extraLinefeedSet ) &&
                $out[$len] === "\n"
            ) {
                // Double the linefeed after pre/listing/textarea
                // according to the HTML5 fragment serialization algorithm.
                $out = substr( $out, 0, $len + 1 ) .
                    substr( $out, $len );
            }
        } else {
            $out = "<{$this->localName}{$encAttribs} />";
            Assert::invariant(
                count( $this->children ) === 0,
                "Empty elements shouldn't have children."
            );
        }
        return $out;
    }

    # Utility functions on BalanceElements.

    public function isA( $set ) {
        if ( $set instanceof BalanceElement ) {
            return $this === $set;
        } elseif ( is_array( $set ) ) {
            return isset( $set[$this->namespaceURI] ) &&
                isset( $set[$this->namespaceURI][$this->localName] );
        } else {
            # assume this is an HTML element name.
            return $this->isHtml() && $this->localName === $set;
        }
    }

    public function isHtmlNamed( $tagName ) {
        return $this->namespaceURI === BalanceSets::HTML_NAMESPACE
            && $this->localName === $tagName;
    }

    public function isHtml() {
        return $this->namespaceURI === BalanceSets::HTML_NAMESPACE;
    }

    public function isMathmlTextIntegrationPoint() {
        return $this->isA( BalanceSets::$mathmlTextIntegrationPointSet );
    }

    public function isHtmlIntegrationPoint() {
        if ( $this->isA( BalanceSets::$htmlIntegrationPointSet ) ) {
            return true;
        }
        if (
            $this->namespaceURI === BalanceSets::MATHML_NAMESPACE &&
            $this->localName === 'annotation-xml' &&
            isset( $this->attribs['encoding'] ) &&
            ( strcasecmp( $this->attribs['encoding'], 'text/html' ) == 0 ||
            strcasecmp( $this->attribs['encoding'], 'application/xhtml+xml' ) == 0 )
        ) {
            return true;
        }
        return false;
    }

    public function getNoahKey() {
        if ( $this->noahKey === null ) {
            $attribs = $this->attribs;
            ksort( $attribs );
            $this->noahKey = serialize( [ $this->namespaceURI, $this->localName, $attribs ] );
        }
        return $this->noahKey;
    }
}

class BalanceStack implements IteratorAggregate {
    private $elements = [];
    public $fosterParentMode = false;
    public $tidyCompat = false;
    public $currentNode;

    public function __construct() {
        # always a root <html> element on the stack
        array_push(
            $this->elements,
            new BalanceElement( BalanceSets::HTML_NAMESPACE, 'html', [] )
        );
        $this->currentNode = $this->elements[0];
    }

    public function getOutput() {
        // Don't include the outer '<html>....</html>'
        $out = '';
        foreach ( $this->elements[0]->children as $elt ) {
            $out .= is_string( $elt ) ? $elt :
                $elt->flatten( $this->tidyCompat );
        }
        return $out;
    }

    public function insertComment( $value ) {
        // Just another type of text node, except for tidy p-wrapping.
        return $this->insertText( '<!--' . $value . '-->', true );
    }

    public function insertText( $value, $isComment = false ) {
        if (
            $this->fosterParentMode &&
            $this->currentNode->isA( BalanceSets::$tableSectionRowSet )
        ) {
            $this->fosterParent( $value );
        } elseif (
            $this->tidyCompat && !$isComment &&
            $this->currentNode->isA( BalanceSets::$tidyPWrapSet )
        ) {
            $this->insertHTMLELement( 'mw:p-wrap', [] );
            return $this->insertText( $value );
        } else {
            $this->currentNode->appendChild( $value );
        }
    }

    public function insertForeignElement( $namespaceURI, $tag, $attribs ) {
        return $this->insertElement(
            new BalanceElement( $namespaceURI, $tag, $attribs )
        );
    }

    public function insertHTMLElement( $tag, $attribs ) {
        return $this->insertForeignElement(
            BalanceSets::HTML_NAMESPACE, $tag, $attribs
        );
    }

    public function insertElement( BalanceElement $elt ) {
        if (
            $this->currentNode->isHtmlNamed( 'mw:p-wrap' ) &&
            !$elt->isA( BalanceSets::$tidyInlineSet )
        ) {
            // Tidy compatibility.
            $this->pop();
        }
        if (
            $this->fosterParentMode &&
            $this->currentNode->isA( BalanceSets::$tableSectionRowSet )
        ) {
            $elt = $this->fosterParent( $elt );
        } else {
            $this->currentNode->appendChild( $elt );
        }
        Assert::invariant( $elt->parent !== null, "$elt must be in tree" );
        Assert::invariant( $elt->parent !== 'flat', "$elt must not have been previous flattened" );
        array_push( $this->elements, $elt );
        $this->currentNode = $elt;
        return $elt;
    }

    public function inScope( $tag ) {
        return $this->inSpecificScope( $tag, BalanceSets::$inScopeSet );
    }

    public function inButtonScope( $tag ) {
        return $this->inSpecificScope( $tag, BalanceSets::inButtonScopeSet() );
    }

    public function inListItemScope( $tag ) {
        return $this->inSpecificScope( $tag, BalanceSets::inListItemScopeSet() );
    }

    public function inTableScope( $tag ) {
        return $this->inSpecificScope( $tag, BalanceSets::$inTableScopeSet );
    }

    public function inSelectScope( $tag ) {
        // Can't use inSpecificScope to implement this, since it involves
        // *inverting* a set of tags.  Implement manually.
        foreach ( $this as $elt ) {
            if ( $elt->isA( $tag ) ) {
                return true;
            }
            if ( !$elt->isA( BalanceSets::$inInvertedSelectScopeSet ) ) {
                return false;
            }
        }
        return false;
    }

    public function inSpecificScope( $tag, $set ) {
        foreach ( $this as $elt ) {
            if ( $elt->isA( $tag ) ) {
                return true;
            }
            if ( $elt->isA( $set ) ) {
                return false;
            }
        }
        return false;
    }

    public function generateImpliedEndTags( $butnot = null, $thorough = false ) {
        $endTagSet = $thorough ?
            BalanceSets::$thoroughImpliedEndTagsSet :
            BalanceSets::$impliedEndTagsSet;
        while ( $this->currentNode ) {
            if ( $butnot !== null && $this->currentNode->isHtmlNamed( $butnot ) ) {
                break;
            }
            if ( !$this->currentNode->isA( $endTagSet ) ) {
                break;
            }
            $this->pop();
        }
    }

    public function adjustedCurrentNode( $fragmentContext ) {
        return ( $fragmentContext && count( $this->elements ) === 1 ) ?
            $fragmentContext : $this->currentNode;
    }

    public function getIterator() {
        return new ReverseArrayIterator( $this->elements );
    }

    public function node( $idx ) {
        return $this->elements[ $idx ];
    }

    public function replaceAt( $idx, BalanceElement $elt ) {
        Assert::precondition(
            $this->elements[$idx]->parent !== 'flat',
            'Replaced element should not have already been flattened.'
        );
        Assert::precondition(
            $elt->parent !== 'flat',
            'New element should not have already been flattened.'
        );
        $this->elements[$idx] = $elt;
        if ( $idx === count( $this->elements ) - 1 ) {
            $this->currentNode = $elt;
        }
    }

    public function indexOf( $tag ) {
        for ( $i = count( $this->elements ) - 1; $i >= 0; $i-- ) {
            if ( $this->elements[$i]->isA( $tag ) ) {
                return $i;
            }
        }
        return -1;
    }

    public function length() {
        return count( $this->elements );
    }

    public function pop() {
        $elt = array_pop( $this->elements );
        if ( count( $this->elements ) ) {
            $this->currentNode = $this->elements[ count( $this->elements ) - 1 ];
        } else {
            $this->currentNode = null;
        }
        if ( !$elt->isHtmlNamed( 'mw:p-wrap' ) ) {
            $elt->flatten( $this->tidyCompat );
        }
    }

    public function popTo( $idx ) {
        $length = count( $this->elements );
        for ( $length = count( $this->elements ); $length > $idx; $length-- ) {
            $this->pop();
        }
    }

    public function popTag( $tag ) {
        while ( $this->currentNode ) {
            if ( $this->currentNode->isA( $tag ) ) {
                $this->pop();
                break;
            }
            $this->pop();
        }
    }

    public function clearToContext( $set ) {
        // Note that we don't loop to 0. Never pop the <html> elt off.
        for ( $length = count( $this->elements ); $length > 1; $length-- ) {
            if ( $this->currentNode->isA( $set ) ) {
                break;
            }
            $this->pop();
        }
    }

    public function removeElement( BalanceElement $elt, $flatten = true ) {
        Assert::parameter(
            $elt->parent !== 'flat',
            '$elt',
            '$elt should not already have been flattened.'
        );
        Assert::parameter(
            $elt->parent->parent !== 'flat',
            '$elt',
            'The parent of $elt should not already have been flattened.'
        );
        $idx = array_search( $elt, $this->elements, true );
        Assert::parameter( $idx !== false, '$elt', 'must be in stack' );
        array_splice( $this->elements, $idx, 1 );
        if ( $idx === count( $this->elements ) ) {
            $this->currentNode = $this->elements[$idx - 1];
        }
        if ( $flatten ) {
            // serialize $elt into its parent
            // otherwise, it will eventually serialize when the parent
            // is serialized, we just hold onto the memory for its
            // tree of objects a little longer.
            $elt->flatten( $this->tidyCompat );
        }
        Assert::postcondition(
            array_search( $elt, $this->elements, true ) === false,
            '$elt should no longer be in open elements stack'
        );
    }

    public function insertAfter( BalanceElement $a, BalanceElement $b ) {
        $idx = $this->indexOf( $a );
        Assert::parameter( $idx !== false, '$a', 'must be in stack' );
        if ( $idx === count( $this->elements ) - 1 ) {
            array_push( $this->elements, $b );
            $this->currentNode = $b;
        } else {
            array_splice( $this->elements, $idx + 1, 0, [ $b ] );
        }
    }

    # Fostering and adoption.

    private function fosterParent( $elt ) {
        $lastTable = $this->indexOf( 'table' );
        $lastTemplate = $this->indexOf( 'template' );
        $parent = null;
        $before = null;

        if ( $lastTemplate >= 0 && ( $lastTable < 0 || $lastTemplate > $lastTable ) ) {
            $parent = $this->elements[$lastTemplate];
        } elseif ( $lastTable >= 0 ) {
            $parent = $this->elements[$lastTable]->parent;
            # Assume all tables have parents, since we're not running scripts!
            Assert::invariant(
                $parent !== null, "All tables should have parents"
            );
            $before = $this->elements[$lastTable];
        } else {
            $parent = $this->elements[0]; // the `html` element.
        }

        if ( $this->tidyCompat ) {
            if ( is_string( $elt ) ) {
                // We're fostering text: do we need a p-wrapper?
                if ( $parent->isA( BalanceSets::$tidyPWrapSet ) ) {
                    $this->insertHTMLElement( 'mw:p-wrap', [] );
                    $this->insertText( $elt );
                    return $elt;
                }
            } else {
                // We're fostering an element; do we need to merge p-wrappers?
                if ( $elt->isHtmlNamed( 'mw:p-wrap' ) ) {
                    $idx = $before ?
                        array_search( $before, $parent->children, true ) :
                        count( $parent->children );
                    $after = $idx > 0 ? $parent->children[$idx - 1] : '';
                    if (
                        $after instanceof BalanceElement &&
                        $after->isHtmlNamed( 'mw:p-wrap' )
                    ) {
                        return $after; // Re-use existing p-wrapper.
                    }
                }
            }
        }

        if ( $before ) {
            $parent->insertBefore( $before, $elt );
        } else {
            $parent->appendChild( $elt );
        }
        return $elt;
    }

    public function adoptionAgency( $tag, $afe ) {
        // If the current node is an HTML element whose tag name is subject,
        // and the current node is not in the list of active formatting
        // elements, then pop the current node off the stack of open
        // elements and abort these steps.
        if (
            $this->currentNode->isHtmlNamed( $tag ) &&
            !$afe->isInList( $this->currentNode )
        ) {
            $this->pop();
            return true; // no more handling required
        }

        // Let outer loop counter be zero.
        $outer = 0;

        // Outer loop: If outer loop counter is greater than or
        // equal to eight, then abort these steps.
        while ( $outer < 8 ) {
            // Increment outer loop counter by one.
            $outer++;

            // Let the formatting element be the last element in the list
            // of active formatting elements that: is between the end of
            // the list and the last scope marker in the list, if any, or
            // the start of the list otherwise, and has the same tag name
            // as the token.
            $fmtelt = $afe->findElementByTag( $tag );

            // If there is no such node, then abort these steps and instead
            // act as described in the "any other end tag" entry below.
            if ( !$fmtelt ) {
                return false; // false means handle by the default case
            }

            // Otherwise, if there is such a node, but that node is not in
            // the stack of open elements, then this is a parse error;
            // remove the element from the list, and abort these steps.
            $index = $this->indexOf( $fmtelt );
            if ( $index < 0 ) {
                $afe->remove( $fmtelt );
                return true;   // true means no more handling required
            }

            // Otherwise, if there is such a node, and that node is also in
            // the stack of open elements, but the element is not in scope,
            // then this is a parse error; ignore the token, and abort
            // these steps.
            if ( !$this->inScope( $fmtelt ) ) {
                return true;
            }

            // Let the furthest block be the topmost node in the stack of
            // open elements that is lower in the stack than the formatting
            // element, and is an element in the special category. There
            // might not be one.
            $furthestblock = null;
            $furthestblockindex = -1;
            $stacklen = $this->length();
            for ( $i = $index+1; $i < $stacklen; $i++ ) {
                if ( $this->node( $i )->isA( BalanceSets::$specialSet ) ) {
                    $furthestblock = $this->node( $i );
                    $furthestblockindex = $i;
                    break;
                }
            }

            // If there is no furthest block, then the UA must skip the
            // subsequent steps and instead just pop all the nodes from the
            // bottom of the stack of open elements, from the current node
            // up to and including the formatting element, and remove the
            // formatting element from the list of active formatting
            // elements.
            if ( !$furthestblock ) {
                $this->popTag( $fmtelt );
                $afe->remove( $fmtelt );
                return true;
            } else {
                // Let the common ancestor be the element immediately above
                // the formatting element in the stack of open elements.
                $ancestor = $this->node( $index-1 );

                // Let a bookmark note the position of the formatting
                // element in the list of active formatting elements
                // relative to the elements on either side of it in the
                // list.
                $BOOKMARK = new BalanceElement( '[bookmark]', '[bookmark]', [] );
                $afe->insertAfter( $fmtelt, $BOOKMARK );

                // Let node and last node be the furthest block.
                $node = $furthestblock;
                $lastnode = $furthestblock;
                $nodeindex = $furthestblockindex;
                $isAFE = false;

                // Let inner loop counter be zero.
                $inner = 0;

                while ( true ) {

                    // Increment inner loop counter by one.
                    $inner++;

                    // Let node be the element immediately above node in
                    // the stack of open elements, or if node is no longer
                    // in the stack of open elements (e.g. because it got
                    // removed by this algorithm), the element that was
                    // immediately above node in the stack of open elements
                    // before node was removed.
                    $node = $this->node( --$nodeindex );

                    // If node is the formatting element, then go
                    // to the next step in the overall algorithm.
                    if ( $node === $fmtelt ) break;

                    // If the inner loop counter is greater than three and node
                    // is in the list of active formatting elements, then remove
                    // node from the list of active formatting elements.
                    $isAFE = $afe->isInList( $node );
                    if ( $inner > 3 && $isAFE ) {
                        $afe->remove( $node );
                        $isAFE = false;
                    }

                    // If node is not in the list of active formatting
                    // elements, then remove node from the stack of open
                    // elements and then go back to the step labeled inner
                    // loop.
                    if ( !$isAFE ) {
                        // Don't flatten here, since we're about to relocate
                        // parts of this $node.
                        $this->removeElement( $node, false );
                        continue;
                    }

                    // Create an element for the token for which the
                    // element node was created with common ancestor as
                    // the intended parent, replace the entry for node
                    // in the list of active formatting elements with an
                    // entry for the new element, replace the entry for
                    // node in the stack of open elements with an entry for
                    // the new element, and let node be the new element.
                    $newelt = new BalanceElement(
                        $node->namespaceURI, $node->localName, $node->attribs );
                    $afe->replace( $node, $newelt );
                    $this->replaceAt( $nodeindex, $newelt );
                    $node = $newelt;

                    // If last node is the furthest block, then move the
                    // aforementioned bookmark to be immediately after the
                    // new node in the list of active formatting elements.
                    if ( $lastnode === $furthestblock ) {
                        $afe->remove( $BOOKMARK );
                        $afe->insertAfter( $newelt, $BOOKMARK );
                    }

                    // Insert last node into node, first removing it from
                    // its previous parent node if any.
                    $node->appendChild( $lastnode );

                    // Let last node be node.
                    $lastnode = $node;
                }

                // If the common ancestor node is a table, tbody, tfoot,
                // thead, or tr element, then, foster parent whatever last
                // node ended up being in the previous step, first removing
                // it from its previous parent node if any.
                if (
                    $this->fosterParentMode &&
                    $ancestor->isA( BalanceSets::$tableSectionRowSet )
                ) {
                    $this->fosterParent( $lastnode );
                } else {
                    // Otherwise, append whatever last node ended up being in
                    // the previous step to the common ancestor node, first
                    // removing it from its previous parent node if any.
                    $ancestor->appendChild( $lastnode );
                }

                // Create an element for the token for which the
                // formatting element was created, with furthest block
                // as the intended parent.
                $newelt2 = new BalanceElement(
                    $fmtelt->namespaceURI, $fmtelt->localName, $fmtelt->attribs );

                // Take all of the child nodes of the furthest block and
                // append them to the element created in the last step.
                $newelt2->adoptChildren( $furthestblock );

                // Append that new element to the furthest block.
                $furthestblock->appendChild( $newelt2 );

                // Remove the formatting element from the list of active
                // formatting elements, and insert the new element into the
                // list of active formatting elements at the position of
                // the aforementioned bookmark.
                $afe->remove( $fmtelt );
                $afe->replace( $BOOKMARK, $newelt2 );

                // Remove the formatting element from the stack of open
                // elements, and insert the new element into the stack of
                // open elements immediately below the position of the
                // furthest block in that stack.
                $this->removeElement( $fmtelt );
                $this->insertAfter( $furthestblock, $newelt2 );
            }
        }

        return true;
    }

    public function __toString() {
        $r = [];
        foreach ( $this->elements as $elt ) {
            array_push( $r, $elt->localName );
        }
        return implode( $r, ' ' );
    }
}

class BalanceMarker {
    public $nextAFE;
    public $prevAFE;
}

class BalanceActiveFormattingElements {
    private $tail;

    private $head;

    private $noahTableStack = [ [] ];

    public function __destruct() {
        for ( $node = $this->head; $node; $node = $next ) {
            $next = $node->nextAFE;
            $node->prevAFE = $node->nextAFE = $node->nextNoah = null;
        }
        $this->head = $this->tail = $this->noahTableStack = null;
    }

    public function insertMarker() {
        $elt = new BalanceMarker;
        if ( $this->tail ) {
            $this->tail->nextAFE = $elt;
            $elt->prevAFE = $this->tail;
        } else {
            $this->head = $elt;
        }
        $this->tail = $elt;
        $this->noahTableStack[] = [];
    }

    public function push( BalanceElement $elt ) {
        // Must not be in the list already
        if ( $elt->prevAFE !== null || $this->head === $elt ) {
            throw new ParameterAssertionException( '$elt',
                'Cannot insert a node into the AFE list twice' );
        }

        // "Noah's Ark clause" -- if there are already three copies of
        // this element before we encounter a marker, then drop the last
        // one.
        $noahKey = $elt->getNoahKey();
        $table =& $this->noahTableStack[ count( $this->noahTableStack ) - 1 ];
        if ( !isset( $table[$noahKey] ) ) {
            $table[$noahKey] = $elt;
        } else {
            $count = 1;
            $head = $tail = $table[$noahKey];
            while ( $tail->nextNoah ) {
                $tail = $tail->nextNoah;
                $count++;
            }
            if ( $count >= 3 ) {
                $this->remove( $head );
            }
            $tail->nextNoah = $elt;
        }
        // Add to the main AFE list
        if ( $this->tail ) {
            $this->tail->nextAFE = $elt;
            $elt->prevAFE = $this->tail;
        } else {
            $this->head = $elt;
        }
        $this->tail = $elt;
    }

    public function clearToMarker() {
        // Iterate back through the list starting from the tail
        $tail = $this->tail;
        while ( $tail && !( $tail instanceof BalanceMarker ) ) {
            // Unlink the element
            $prev = $tail->prevAFE;
            $tail->prevAFE = null;
            if ( $prev ) {
                $prev->nextAFE = null;
            }
            $tail->nextNoah = null;
            $tail = $prev;
        }
        // If we finished on a marker, unlink it and pop it off the Noah table stack
        if ( $tail ) {
            $prev = $tail->prevAFE;
            if ( $prev ) {
                $prev->nextAFE = null;
            }
            $tail = $prev;
            array_pop( $this->noahTableStack );
        } else {
            // No marker: wipe the top-level Noah table (which is the only one)
            $this->noahTableStack[0] = [];
        }
        // If we removed all the elements, clear the head pointer
        if ( !$tail ) {
            $this->head = null;
        }
        $this->tail = $tail;
    }

    public function findElementByTag( $tag ) {
        $elt = $this->tail;
        while ( $elt && !( $elt instanceof BalanceMarker ) ) {
            if ( $elt->localName === $tag ) {
                return $elt;
            }
            $elt = $elt->prevAFE;
        }
        return null;
    }

    public function isInList( BalanceElement $elt ) {
        return $this->head === $elt || $elt->prevAFE;
    }

    public function remove( BalanceElement $elt ) {
        if ( $this->head !== $elt && !$elt->prevAFE ) {
            throw new ParameterAssertionException( '$elt',
                "Attempted to remove an element which is not in the AFE list" );
        }
        // Update head and tail pointers
        if ( $this->head === $elt ) {
            $this->head = $elt->nextAFE;
        }
        if ( $this->tail === $elt ) {
            $this->tail = $elt->prevAFE;
        }
        // Update previous element
        if ( $elt->prevAFE ) {
            $elt->prevAFE->nextAFE = $elt->nextAFE;
        }
        // Update next element
        if ( $elt->nextAFE ) {
            $elt->nextAFE->prevAFE = $elt->prevAFE;
        }
        // Clear pointers so that isInList() etc. will work
        $elt->prevAFE = $elt->nextAFE = null;
        // Update Noah list
        $this->removeFromNoahList( $elt );
    }

    private function addToNoahList( BalanceElement $elt ) {
        $noahKey = $elt->getNoahKey();
        $table =& $this->noahTableStack[ count( $this->noahTableStack ) - 1 ];
        if ( !isset( $table[$noahKey] ) ) {
            $table[$noahKey] = $elt;
        } else {
            $tail = $table[$noahKey];
            while ( $tail->nextNoah ) {
                $tail = $tail->nextNoah;
            }
            $tail->nextNoah = $elt;
        }
    }

    private function removeFromNoahList( BalanceElement $elt ) {
        $table =& $this->noahTableStack[ count( $this->noahTableStack ) - 1 ];
        $key = $elt->getNoahKey();
        $noahElt = $table[$key];
        if ( $noahElt === $elt ) {
            if ( $noahElt->nextNoah ) {
                $table[$key] = $noahElt->nextNoah;
                $noahElt->nextNoah = null;
            } else {
                unset( $table[$key] );
            }
        } else {
            do {
                $prevNoahElt = $noahElt;
                $noahElt = $prevNoahElt->nextNoah;
                if ( $noahElt === $elt ) {
                    // Found it, unlink
                    $prevNoahElt->nextNoah = $elt->nextNoah;
                    $elt->nextNoah = null;
                    break;
                }
            } while ( $noahElt );
        }
    }

    public function replace( BalanceElement $a, BalanceElement $b ) {
        if ( $this->head !== $a && !$a->prevAFE ) {
            throw new ParameterAssertionException( '$a',
                "Attempted to replace an element which is not in the AFE list" );
        }
        // Update head and tail pointers
        if ( $this->head === $a ) {
            $this->head = $b;
        }
        if ( $this->tail === $a ) {
            $this->tail = $b;
        }
        // Update previous element
        if ( $a->prevAFE ) {
            $a->prevAFE->nextAFE = $b;
        }
        // Update next element
        if ( $a->nextAFE ) {
            $a->nextAFE->prevAFE = $b;
        }
        $b->prevAFE = $a->prevAFE;
        $b->nextAFE = $a->nextAFE;
        $a->nextAFE = $a->prevAFE = null;
        // Update Noah list
        $this->removeFromNoahList( $a );
        $this->addToNoahList( $b );
    }

    public function insertAfter( BalanceElement $a, BalanceElement $b ) {
        if ( $this->head !== $a && !$a->prevAFE ) {
            throw new ParameterAssertionException( '$a',
                "Attempted to insert after an element which is not in the AFE list" );
        }
        if ( $this->tail === $a ) {
            $this->tail = $b;
        }
        if ( $a->nextAFE ) {
            $a->nextAFE->prevAFE = $b;
        }
        $b->nextAFE = $a->nextAFE;
        $b->prevAFE = $a;
        $a->nextAFE = $b;
        $this->addToNoahList( $b );
    }

    // @codingStandardsIgnoreStart Generic.Files.LineLength.TooLong
    // @codingStandardsIgnoreEnd
    public function reconstruct( $stack ) {
        $entry = $this->tail;
        // If there are no entries in the list of active formatting elements,
        // then there is nothing to reconstruct
        if ( !$entry ) {
            return;
        }
        // If the last is a marker, do nothing.
        if ( $entry instanceof BalanceMarker ) {
            return;
        }
        // Or if it is an open element, do nothing.
        if ( $stack->indexOf( $entry ) >= 0 ) {
            return;
        }

        // Loop backward through the list until we find a marker or an
        // open element
        $foundit = false;
        while ( $entry->prevAFE ) {
            $entry = $entry->prevAFE;
            if ( $entry instanceof BalanceMarker || $stack->indexOf( $entry ) >= 0 ) {
                $foundit = true;
                break;
            }
        }

        // Now loop forward, starting from the element after the current one (or
        // the first element if we didn't find a marker or open element),
        // recreating formatting elements and pushing them back onto the list
        // of open elements.
        if ( $foundit ) {
            $entry = $entry->nextAFE;
        }
        do {
            $newElement = $stack->insertHTMLElement(
                $entry->localName,
                $entry->attribs );
            $this->replace( $entry, $newElement );
            $entry = $newElement->nextAFE;
        } while ( $entry );
    }

    public function __toString() {
        $prev = null;
        $s = '';
        for ( $node = $this->head; $node; $prev = $node, $node = $node->nextAFE ) {
            if ( $node instanceof BalanceMarker ) {
                $s .= "MARKER\n";
                continue;
            }
            $s .= $node->localName . '#' . substr( md5( spl_object_hash( $node ) ), 0, 8 );
            if ( $node->nextNoah ) {
                $s .= " (noah sibling: {$node->nextNoah->localName}#" .
                    substr( md5( spl_object_hash( $node->nextNoah ) ), 0, 8 ) .
                    ')';
            }
            if ( $node->nextAFE && $node->nextAFE->prevAFE !== $node ) {
                $s .= " (reverse link is wrong!)";
            }
            $s .= "\n";
        }
        if ( $prev !== $this->tail ) {
            $s .= "(tail pointer is wrong!)\n";
        }
        return $s;
    }
}

class Balancer {
    private $parseMode;
    private $bitsIterator;
    private $allowedHtmlElements;
    private $afe;
    private $stack;
    private $strict;
    private $tidyCompat;
    private $allowComments;

    private $textIntegrationMode;
    private $pendingTableText;
    private $originalInsertionMode;
    private $fragmentContext;
    private $formElementPointer;
    private $ignoreLinefeed;
    private $inRCDATA;
    private $inRAWTEXT;

    const VALID_COMMENT_REGEX = "~ !--
        (                             # 1. Comment match detector
            > | -> | # Invalid short close
            (                         # 2. Comment contents
                (?:
                    (?! --> )
                    (?! --!> )
                    (?! --! \z )
                    (?! -- \z )
                    (?! - \z )
                    .
                )*+
            )
            (                         # 3. Comment close
                --> |   # Normal close
                --!> |  # Comment end bang
                (                     # 4. Indicate matches requiring EOF
                    --! |   # EOF in comment end bang state
                    -- |    # EOF in comment end state
                    -  |    # EOF in comment end dash state
                            # EOF in comment state
                )
            )
        )
        ([^<]*) \z                    # 5. Non-tag text after the comment
        ~xs";

    public function __construct( array $config = [] ) {
        $config = $config + [
            'strict' => false,
            'allowedHtmlElements' => null,
            'tidyCompat' => false,
            'allowComments' => true,
        ];
        $this->allowedHtmlElements = $config['allowedHtmlElements'];
        $this->strict = $config['strict'];
        $this->tidyCompat = $config['tidyCompat'];
        $this->allowComments = $config['allowComments'];
        if ( $this->allowedHtmlElements !== null ) {
            # Sanity check!
            $bad = array_uintersect_assoc(
                $this->allowedHtmlElements,
                BalanceSets::$unsupportedSet[BalanceSets::HTML_NAMESPACE],
                function( $a, $b ) {
                    // Ignore the values (just intersect the keys) by saying
                    // all values are equal to each other.
                    return 0;
                }
            );
            if ( count( $bad ) > 0 ) {
                $badstr = implode( array_keys( $bad ), ',' );
                throw new ParameterAssertionException(
                    '$config',
                    'Balance attempted with sanitization including ' .
                    "unsupported elements: {$badstr}"
                );
            }
        }
    }

    public function balance( $text, $processingCallback = null, $processingArgs = [] ) {
        $this->parseMode = 'inBodyMode';
        $this->bitsIterator = new ExplodeIterator( '<', $text );
        $this->afe = new BalanceActiveFormattingElements();
        $this->stack = new BalanceStack();
        $this->stack->tidyCompat = $this->tidyCompat;
        $this->processingCallback = $processingCallback;
        $this->processingArgs = $processingArgs;

        $this->textIntegrationMode =
            $this->ignoreLinefeed =
            $this->inRCDATA =
            $this->inRAWTEXT = false;

        # The stack is constructed with an <html> element already on it.
        # Set this up as a fragment parsed with <body> as the context.
        $this->fragmentContext =
            new BalanceElement( BalanceSets::HTML_NAMESPACE, 'body', [] );
        $this->resetInsertionMode();
        $this->formElementPointer = null;
        for ( $e = $this->fragmentContext; $e != null; $e = $e->parent ) {
            if ( $e->isHtmlNamed( 'form' ) ) {
                $this->formElementPointer = $e;
                break;
            }
        }

        // First element is text not tag
        $x = $this->bitsIterator->current();
        $this->bitsIterator->next();
        $this->insertToken( 'text', str_replace( '>', '&gt;', $x ) );
        // Now process each tag.
        while ( $this->bitsIterator->valid() ) {
            $this->advance();
        }
        $this->insertToken( 'eof', null );
        $result = $this->stack->getOutput();
        // Free memory before returning.
        $this->bitsIterator = null;
        $this->afe = null;
        $this->stack = null;
        $this->fragmentContext = null;
        $this->formElementPointer = null;
        return $result;
    }

    private function insertToken( $token, $value, $attribs = null, $selfclose = false ) {
        // validate tags against $unsupportedSet
        if ( $token === 'tag' || $token === 'endtag' ) {
            if ( isset( BalanceSets::$unsupportedSet[BalanceSets::HTML_NAMESPACE][$value] ) ) {
                # As described in "simplifications" above, these tags are
                # not supported in the balancer.
                Assert::invariant(
                    !$this->strict,
                    "Unsupported $token <$value> found."
                );
                return false;
            }
        } elseif ( $token === 'text' && $value === '' ) {
            # Don't actually inject the empty string as a text token.
            return true;
        }
        // Support pre/listing/textarea by suppressing initial linefeed
        if ( $this->ignoreLinefeed ) {
            $this->ignoreLinefeed = false;
            if ( $token === 'text' ) {
                if ( $value[0] === "\n" ) {
                    if ( $value === "\n" ) {
                        # Nothing would be left, don't inject the empty string.
                        return true;
                    }
                    $value = substr( $value, 1 );
                }
            }
        }
        // Some hoops we have to jump through
        $adjusted = $this->stack->adjustedCurrentNode( $this->fragmentContext );

        $isForeign = true;
        if (
            $this->stack->length() === 0 ||
            $adjusted->isHtml() ||
            $token === 'eof'
        ) {
            $isForeign = false;
        } elseif ( $adjusted->isMathmlTextIntegrationPoint() ) {
            if ( $token === 'text' ) {
                $isForeign = false;
            } elseif (
                $token === 'tag' &&
                $value !== 'mglyph' && $value !== 'malignmark'
            ) {
                $isForeign = false;
            }
        } elseif (
            $adjusted->namespaceURI === BalanceSets::MATHML_NAMESPACE &&
            $adjusted->localName === 'annotation-xml' &&
            $token === 'tag' && $value === 'svg'
        ) {
            $isForeign = false;
        } elseif (
            $adjusted->isHtmlIntegrationPoint() &&
            ( $token === 'tag' || $token === 'text' )
        ) {
            $isForeign = false;
        }
        if ( $isForeign ) {
            return $this->insertForeignToken( $token, $value, $attribs, $selfclose );
        } else {
            $func = $this->parseMode;
            return $this->$func( $token, $value, $attribs, $selfclose );
        }
    }

    private function insertForeignToken( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' ) {
            $this->stack->insertText( $value );
            return true;
        } elseif ( $token === 'tag' ) {
            switch ( $value ) {
            case 'font':
                if ( isset( $attribs['color'] )
                    || isset( $attribs['face'] )
                    || isset( $attribs['size'] )
                ) {
                    break;
                }
                /* otherwise, fall through */
            case 'b':
            case 'big':
            case 'blockquote':
            case 'body':
            case 'br':
            case 'center':
            case 'code':
            case 'dd':
            case 'div':
            case 'dl':
            case 'dt':
            case 'em':
            case 'embed':
            case 'h1':
            case 'h2':
            case 'h3':
            case 'h4':
            case 'h5':
            case 'h6':
            case 'head':
            case 'hr':
            case 'i':
            case 'img':
            case 'li':
            case 'listing':
            case 'menu':
            case 'meta':
            case 'nobr':
            case 'ol':
            case 'p':
            case 'pre':
            case 'ruby':
            case 's':
            case 'small':
            case 'span':
            case 'strong':
            case 'strike':
            case 'sub':
            case 'sup':
            case 'table':
            case 'tt':
            case 'u':
            case 'ul':
            case 'var':
                if ( $this->fragmentContext ) {
                    break;
                }
                while ( true ) {
                    $this->stack->pop();
                    $node = $this->stack->currentNode;
                    if (
                        $node->isMathmlTextIntegrationPoint() ||
                        $node->isHtmlIntegrationPoint() ||
                        $node->isHtml()
                    ) {
                        break;
                    }
                }
                return $this->insertToken( $token, $value, $attribs, $selfclose );
            }
            // "Any other start tag"
            $adjusted = ( $this->fragmentContext && $this->stack->length()===1 ) ?
                $this->fragmentContext : $this->stack->currentNode;
            $this->stack->insertForeignElement(
                $adjusted->namespaceURI, $value, $attribs
            );
            if ( $selfclose ) {
                $this->stack->pop();
            }
            return true;
        } elseif ( $token === 'endtag' ) {
            $first = true;
            foreach ( $this->stack as $i => $node ) {
                if ( $node->isHtml() && !$first ) {
                    // process the end tag as HTML
                    $func = $this->parseMode;
                    return $this->$func( $token, $value, $attribs, $selfclose );
                } elseif ( $i === 0 ) {
                    return true;
                } elseif ( $node->localName === $value ) {
                    $this->stack->popTag( $node );
                    return true;
                }
                $first = false;
            }
        }
    }

    private function advance() {
        $x = $this->bitsIterator->current();
        $this->bitsIterator->next();
        $regs = [];
        # Handle comments.  These won't be generated by mediawiki (they
        # are stripped in the Sanitizer) but may be generated by extensions.
        if (
            $this->allowComments &&
            !( $this->inRCDATA || $this->inRAWTEXT ) &&
            preg_match( Balancer::VALID_COMMENT_REGEX, $x, $regs, PREG_OFFSET_CAPTURE ) &&
            /* verify EOF condition where necessary */
            ( $regs[4][1] < 0 || !$this->bitsIterator->valid() )
        ) {
            $contents = $regs[2][0];
            $rest = $regs[5][0];
            $this->insertToken( 'comment', $contents );
            $this->insertToken( 'text', str_replace( '>', '&gt;', $rest ) );
            return;
        }
        # $slash: Does the current element start with a '/'?
        # $t: Current element name
        # $attribStr: String between element name and >
        # $brace: Ending '>' or '/>'
        # $rest: Everything until the next element from the $bitsIterator
        if ( preg_match( Sanitizer::ELEMENT_BITS_REGEX, $x, $regs ) ) {
            list( /* $qbar */, $slash, $t, $attribStr, $brace, $rest ) = $regs;
            $t = strtolower( $t );
            if ( $this->strict ) {
                /* Verify that attributes are all properly double-quoted */
                Assert::invariant(
                    preg_match(
                        '/^( [:_A-Z0-9][-.:_A-Z0-9]*="[^"]*")*[ ]*$/i', $attribStr
                    ),
                    "Bad attribute string found"
                );
            }
        } else {
            Assert::invariant(
                !$this->strict, "< found which does not start a valid tag"
            );
            $slash = $t = $attribStr = $brace = $rest = null;
        }
        $goodtag = $t;
        if ( $this->inRCDATA ) {
            if ( $slash && $t === $this->inRCDATA ) {
                $this->inRCDATA = false;
            } else {
                // No tags allowed; this emulates the "rcdata" tokenizer mode.
                $goodtag = false;
            }
        }
        if ( $this->inRAWTEXT ) {
            if ( $slash && $t === $this->inRAWTEXT ) {
                $this->inRAWTEXT = false;
            } else {
                // No tags allowed, no entity-escaping done.
                $goodtag = false;
            }
        }
        $sanitize = $this->allowedHtmlElements !== null;
        if ( $sanitize ) {
            $goodtag = $t && isset( $this->allowedHtmlElements[$t] );
        }
        if ( $goodtag ) {
            if ( is_callable( $this->processingCallback ) ) {
                call_user_func_array( $this->processingCallback, [ &$attribStr, $this->processingArgs ] );
            }
            if ( $sanitize ) {
                $goodtag = Sanitizer::validateTag( $attribStr, $t );
            }
        }
        if ( $goodtag ) {
            if ( $sanitize ) {
                $attribs = Sanitizer::decodeTagAttributes( $attribStr );
                $attribs = Sanitizer::validateTagAttributes( $attribs, $t );
            } else {
                $attribs = Sanitizer::decodeTagAttributes( $attribStr );
            }
            $goodtag = $this->insertToken(
                $slash ? 'endtag' : 'tag', $t, $attribs, $brace === '/>'
            );
        }
        if ( $goodtag ) {
            $rest = str_replace( '>', '&gt;', $rest );
            $this->insertToken( 'text', str_replace( '>', '&gt;', $rest ) );
        } elseif ( $this->inRAWTEXT ) {
            $this->insertToken( 'text', "<$x" );
        } else {
            # bad tag; serialize entire thing as text.
            $this->insertToken( 'text', '&lt;' . str_replace( '>', '&gt;', $x ) );
        }
    }

    private function switchMode( $mode ) {
        Assert::parameter(
            substr( $mode, -4 )==='Mode', '$mode', 'should end in Mode'
        );
        $oldMode = $this->parseMode;
        $this->parseMode = $mode;
        return $oldMode;
    }

    private function switchModeAndReprocess( $mode, $token, $value, $attribs, $selfclose ) {
        $this->switchMode( $mode );
        return $this->insertToken( $token, $value, $attribs, $selfclose );
    }

    private function resetInsertionMode() {
        $last = false;
        foreach ( $this->stack as $i => $node ) {
            if ( $i === 0 ) {
                $last = true;
                if ( $this->fragmentContext ) {
                    $node = $this->fragmentContext;
                }
            }
            if ( $node->isHtml() ) {
                switch ( $node->localName ) {
                case 'select':
                    $stacklen = $this->stack->length();
                    for ( $j = $i + 1; $j < $stacklen-1; $j++ ) {
                        $ancestor = $this->stack->node( $stacklen-$j-1 );
                        if ( $ancestor->isHtmlNamed( 'template' ) ) {
                            break;
                        }
                        if ( $ancestor->isHtmlNamed( 'table' ) ) {
                            $this->switchMode( 'inSelectInTableMode' );
                            return;
                        }
                    }
                    $this->switchMode( 'inSelectMode' );
                    return;
                case 'tr':
                    $this->switchMode( 'inRowMode' );
                    return;
                case 'tbody':
                case 'tfoot':
                case 'thead':
                    $this->switchMode( 'inTableBodyMode' );
                    return;
                case 'caption':
                    $this->switchMode( 'inCaptionMode' );
                    return;
                case 'colgroup':
                    $this->switchMode( 'inColumnGroupMode' );
                    return;
                case 'table':
                    $this->switchMode( 'inTableMode' );
                    return;
                case 'template':
                    $this->switchMode(
                        array_slice( $this->templateInsertionModes, -1 )[0]
                    );
                    return;
                case 'body':
                    $this->switchMode( 'inBodyMode' );
                    return;
                # OMITTED: <frameset>
                # OMITTED: <html>
                # OMITTED: <head>
                default:
                    if ( !$last ) {
                        # OMITTED: <head>
                        if ( $node->isA( BalanceSets::$tableCellSet ) ) {
                            $this->switchMode( 'inCellMode' );
                            return;
                        }
                    }
                }
            }
            if ( $last ) {
                $this->switchMode( 'inBodyMode' );
                return;
            }
        }
    }

    private function stopParsing() {
        # Most of the spec methods are inapplicable, other than step 2:
        # "pop all the nodes off the stack of open elements".
        # We're going to keep the top-most <html> element on the stack, though.

        # Clear the AFE list first, otherwise the element objects will stay live
        # during serialization, potentially using O(N^2) memory. Note that
        # popping the stack will never result in reconstructing the active
        # formatting elements.
        $this->afe = null;
        $this->stack->popTo( 1 );
    }

    private function parseRawText( $value, $attribs = null ) {
        $this->stack->insertHTMLElement( $value, $attribs );
        $this->inRAWTEXT = $value;
        $this->originalInsertionMode = $this->switchMode( 'inTextMode' );
        return true;
    }

    private function inTextMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' ) {
            $this->stack->insertText( $value );
            return true;
        } elseif ( $token === 'eof' ) {
            $this->stack->pop();
            return $this->switchModeAndReprocess(
                $this->originalInsertionMode, $token, $value, $attribs, $selfclose
            );
        } elseif ( $token === 'endtag' ) {
            $this->stack->pop();
            $this->switchMode( $this->originalInsertionMode );
            return true;
        }
        return true;
    }

    private function inHeadMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' ) {
            if ( preg_match( '/^[\x09\x0A\x0C\x0D\x20]+/', $value, $matches ) ) {
                $this->stack->insertText( $matches[0] );
                $value = substr( $value, strlen( $matches[0] ) );
            }
            if ( strlen( $value ) === 0 ) {
                return true; // All text handled.
            }
            // Fall through to handle non-whitespace below.
        } elseif ( $token === 'tag' ) {
            switch ( $value ) {
            case 'meta':
                # OMITTED: in a full HTML parser, this might change the encoding.
                /* falls through */
            # OMITTED: <html>
            case 'base':
            case 'basefont':
            case 'bgsound':
            case 'link':
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->stack->pop();
                return true;
            # OMITTED: <title>
            # OMITTED: <noscript>
            case 'noframes':
            case 'style':
                return $this->parseRawText( $value, $attribs );
            # OMITTED: <script>
            case 'template':
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->afe->insertMarker();
                # OMITTED: frameset_ok
                $this->switchMode( 'inTemplateMode' );
                $this->templateInsertionModes[] = $this->parseMode;
                return true;
            # OMITTED: <head>
            }
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            # OMITTED: <head>
            # OMITTED: <body>
            # OMITTED: <html>
            case 'br':
                break; // handle at the bottom of the function
            case 'template':
                if ( $this->stack->indexOf( $value ) < 0 ) {
                    return true; // Ignore the token.
                }
                $this->stack->generateImpliedEndTags( null, true /* thorough */ );
                $this->stack->popTag( $value );
                $this->afe->clearToMarker();
                array_pop( $this->templateInsertionModes );
                $this->resetInsertionMode();
                return true;
            default:
                // ignore any other end tag
                return true;
            }
        } elseif ( $token === 'comment' ) {
            $this->stack->insertComment( $value );
            return true;
        }

        // If not handled above
        $this->inHeadMode( 'endtag', 'head' ); // synthetic </head>
        // Then redo this one
        return $this->insertToken( $token, $value, $attribs, $selfclose );
    }

    private function inBodyMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' ) {
            $this->afe->reconstruct( $this->stack );
            $this->stack->insertText( $value );
            return true;
        } elseif ( $token === 'eof' ) {
            if ( !empty( $this->templateInsertionModes ) ) {
                return $this->inTemplateMode( $token, $value, $attribs, $selfclose );
            }
            $this->stopParsing();
            return true;
        } elseif ( $token === 'tag' ) {
            switch ( $value ) {
            # OMITTED: <html>
            case 'base':
            case 'basefont':
            case 'bgsound':
            case 'link':
            case 'meta':
            case 'noframes':
            # OMITTED: <script>
            case 'style':
            case 'template':
            # OMITTED: <title>
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );
            # OMITTED: <body>
            # OMITTED: <frameset>

            case 'address':
            case 'article':
            case 'aside':
            case 'blockquote':
            case 'center':
            case 'details':
            case 'dialog':
            case 'dir':
            case 'div':
            case 'dl':
            case 'fieldset':
            case 'figcaption':
            case 'figure':
            case 'footer':
            case 'header':
            case 'hgroup':
            case 'main':
            case 'menu':
            case 'nav':
            case 'ol':
            case 'p':
            case 'section':
            case 'summary':
            case 'ul':
                if ( $this->stack->inButtonScope( 'p' ) ) {
                    $this->inBodyMode( 'endtag', 'p' );
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;

            case 'h1':
            case 'h2':
            case 'h3':
            case 'h4':
            case 'h5':
            case 'h6':
                if ( $this->stack->inButtonScope( 'p' ) ) {
                    $this->inBodyMode( 'endtag', 'p' );
                }
                if ( $this->stack->currentNode->isA( BalanceSets::$headingSet ) ) {
                    $this->stack->pop();
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;

            case 'pre':
            case 'listing':
                if ( $this->stack->inButtonScope( 'p' ) ) {
                    $this->inBodyMode( 'endtag', 'p' );
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->ignoreLinefeed = true;
                # OMITTED: frameset_ok
                return true;

            case 'form':
                if (
                    $this->formElementPointer &&
                    $this->stack->indexOf( 'template' ) < 0
                ) {
                    return true; // in a form, not in a template.
                }
                if ( $this->stack->inButtonScope( "p" ) ) {
                    $this->inBodyMode( 'endtag', 'p' );
                }
                $elt = $this->stack->insertHTMLElement( $value, $attribs );
                if ( $this->stack->indexOf( 'template' ) < 0 ) {
                    $this->formElementPointer = $elt;
                }
                return true;

            case 'li':
                # OMITTED: frameset_ok
                foreach ( $this->stack as $node ) {
                    if ( $node->isHtmlNamed( 'li' ) ) {
                        $this->inBodyMode( 'endtag', 'li' );
                        break;
                    }
                    if (
                        $node->isA( BalanceSets::$specialSet ) &&
                        !$node->isA( BalanceSets::$addressDivPSet )
                    ) {
                        break;
                    }
                }
                if ( $this->stack->inButtonScope( 'p' ) ) {
                    $this->inBodyMode( 'endtag', 'p' );
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;

            case 'dd':
            case 'dt':
                # OMITTED: frameset_ok
                foreach ( $this->stack as $node ) {
                    if ( $node->isHtmlNamed( 'dd' ) ) {
                        $this->inBodyMode( 'endtag', 'dd' );
                        break;
                    }
                    if ( $node->isHtmlNamed( 'dt' ) ) {
                        $this->inBodyMode( 'endtag', 'dt' );
                        break;
                    }
                    if (
                        $node->isA( BalanceSets::$specialSet ) &&
                        !$node->isA( BalanceSets::$addressDivPSet )
                    ) {
                        break;
                    }
                }
                if ( $this->stack->inButtonScope( 'p' ) ) {
                    $this->inBodyMode( 'endtag', 'p' );
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;

            # OMITTED: <plaintext>

            case 'button':
                if ( $this->stack->inScope( 'button' ) ) {
                    $this->inBodyMode( 'endtag', 'button' );
                    return $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                $this->afe->reconstruct( $this->stack );
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;

            case 'a':
                $activeElement = $this->afe->findElementByTag( 'a' );
                if ( $activeElement ) {
                    $this->inBodyMode( 'endtag', 'a' );
                    if ( $this->afe->isInList( $activeElement ) ) {
                        $this->afe->remove( $activeElement );
                        // Don't flatten here, since when we fall
                        // through below we might foster parent
                        // the new <a> tag inside this one.
                        $this->stack->removeElement( $activeElement, false );
                    }
                }
                /* Falls through */
            case 'b':
            case 'big':
            case 'code':
            case 'em':
            case 'font':
            case 'i':
            case 's':
            case 'small':
            case 'strike':
            case 'strong':
            case 'tt':
            case 'u':
                $this->afe->reconstruct( $this->stack );
                $this->afe->push( $this->stack->insertHTMLElement( $value, $attribs ), $attribs );
                return true;

            case 'nobr':
                $this->afe->reconstruct( $this->stack );
                if ( $this->stack->inScope( 'nobr' ) ) {
                    $this->inBodyMode( 'endtag', 'nobr' );
                    $this->afe->reconstruct( $this->stack );
                }
                $this->afe->push( $this->stack->insertHTMLElement( $value, $attribs ), $attribs );
                return true;

            case 'applet':
            case 'marquee':
            case 'object':
                $this->afe->reconstruct( $this->stack );
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->afe->insertMarker();
                # OMITTED: frameset_ok
                return true;

            case 'table':
                # The document is never in "quirks mode"; see simplifications
                # above.
                if ( $this->stack->inButtonScope( 'p' ) ) {
                    $this->inBodyMode( 'endtag', 'p' );
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                # OMITTED: frameset_ok
                $this->switchMode( 'inTableMode' );
                return true;

            case 'area':
            case 'br':
            case 'embed':
            case 'img':
            case 'keygen':
            case 'wbr':
                $this->afe->reconstruct( $this->stack );
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->stack->pop();
                # OMITTED: frameset_ok
                return true;

            case 'input':
                $this->afe->reconstruct( $this->stack );
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->stack->pop();
                # OMITTED: frameset_ok
                # (hence we don't need to examine the tag's "type" attribute)
                return true;

            case 'menuitem':
            case 'param':
            case 'source':
            case 'track':
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->stack->pop();
                return true;

            case 'hr':
                if ( $this->stack->inButtonScope( 'p' ) ) {
                    $this->inBodyMode( 'endtag', 'p' );
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->stack->pop();
                return true;

            case 'image':
                # warts!
                return $this->inBodyMode( $token, 'img', $attribs, $selfclose );

            # OMITTED: <isindex>

            case 'textarea':
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->ignoreLinefeed = true;
                $this->inRCDATA = $value; // emulate rcdata tokenizer mode
                # OMITTED: frameset_ok
                return true;

            # OMITTED: <xmp>
            # OMITTED: <iframe>
            # OMITTED: <noembed>
            # OMITTED: <noscript>

            case 'select':
                $this->afe->reconstruct( $this->stack );
                $this->stack->insertHTMLElement( $value, $attribs );
                switch ( $this->parseMode ) {
                case 'inTableMode':
                case 'inCaptionMode':
                case 'inTableBodyMode':
                case 'inRowMode':
                case 'inCellMode':
                    $this->switchMode( 'inSelectInTableMode' );
                    return true;
                default:
                    $this->switchMode( 'inSelectMode' );
                    return true;
                }

            case 'optgroup':
            case 'option':
                if ( $this->stack->currentNode->isHtmlNamed( 'option' ) ) {
                    $this->inBodyMode( 'endtag', 'option' );
                }
                $this->afe->reconstruct( $this->stack );
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;

            case 'rb':
            case 'rtc':
                if ( $this->stack->inScope( 'ruby' ) ) {
                    $this->stack->generateImpliedEndTags();
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;

            case 'rp':
            case 'rt':
                if ( $this->stack->inScope( 'ruby' ) ) {
                    $this->stack->generateImpliedEndTags( 'rtc' );
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;

            case 'math':
                $this->afe->reconstruct( $this->stack );
                # We skip the spec's "adjust MathML attributes" and
                # "adjust foreign attributes" steps, since the browser will
                # do this later when it parses the output and it doesn't affect
                # balancing.
                $this->stack->insertForeignElement(
                    BalanceSets::MATHML_NAMESPACE, $value, $attribs
                );
                if ( $selfclose ) {
                    # emit explicit </math> tag.
                    $this->stack->pop();
                }
                return true;

            case 'svg':
                $this->afe->reconstruct( $this->stack );
                # We skip the spec's "adjust SVG attributes" and
                # "adjust foreign attributes" steps, since the browser will
                # do this later when it parses the output and it doesn't affect
                # balancing.
                $this->stack->insertForeignElement(
                    BalanceSets::SVG_NAMESPACE, $value, $attribs
                );
                if ( $selfclose ) {
                    # emit explicit </svg> tag.
                    $this->stack->pop();
                }
                return true;

            case 'caption':
            case 'col':
            case 'colgroup':
            # OMITTED: <frame>
            case 'head':
            case 'tbody':
            case 'td':
            case 'tfoot':
            case 'th':
            case 'thead':
            case 'tr':
                // Ignore table tags if we're not inTableMode
                return true;
            }

            // Handle any other start tag here
            $this->afe->reconstruct( $this->stack );
            $this->stack->insertHTMLElement( $value, $attribs );
            return true;
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            # </body>,</html> are unsupported.

            case 'template':
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );

            case 'address':
            case 'article':
            case 'aside':
            case 'blockquote':
            case 'button':
            case 'center':
            case 'details':
            case 'dialog':
            case 'dir':
            case 'div':
            case 'dl':
            case 'fieldset':
            case 'figcaption':
            case 'figure':
            case 'footer':
            case 'header':
            case 'hgroup':
            case 'listing':
            case 'main':
            case 'menu':
            case 'nav':
            case 'ol':
            case 'pre':
            case 'section':
            case 'summary':
            case 'ul':
                // Ignore if there is not a matching open tag
                if ( !$this->stack->inScope( $value ) ) {
                    return true;
                }
                $this->stack->generateImpliedEndTags();
                $this->stack->popTag( $value );
                return true;

            case 'form':
                if ( $this->stack->indexOf( 'template' ) < 0 ) {
                    $openform = $this->formElementPointer;
                    $this->formElementPointer = null;
                    if ( !$openform || !$this->stack->inScope( $openform ) ) {
                        return true;
                    }
                    $this->stack->generateImpliedEndTags();
                    // Don't flatten yet if we're removing a <form> element
                    // out-of-order. (eg. `<form><div></form>`)
                    $flatten = ( $this->stack->currentNode === $openform );
                    $this->stack->removeElement( $openform, $flatten );
                } else {
                    if ( !$this->stack->inScope( 'form' ) ) {
                        return true;
                    }
                    $this->stack->generateImpliedEndTags();
                    $this->stack->popTag( 'form' );
                }
                return true;

            case 'p':
                if ( !$this->stack->inButtonScope( 'p' ) ) {
                    $this->inBodyMode( 'tag', 'p', [] );
                    return $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                $this->stack->generateImpliedEndTags( $value );
                $this->stack->popTag( $value );
                return true;

            case 'li':
                if ( !$this->stack->inListItemScope( $value ) ) {
                    return true; # ignore
                }
                $this->stack->generateImpliedEndTags( $value );
                $this->stack->popTag( $value );
                return true;

            case 'dd':
            case 'dt':
                if ( !$this->stack->inScope( $value ) ) {
                    return true; # ignore
                }
                $this->stack->generateImpliedEndTags( $value );
                $this->stack->popTag( $value );
                return true;

            case 'h1':
            case 'h2':
            case 'h3':
            case 'h4':
            case 'h5':
            case 'h6':
                if ( !$this->stack->inScope( BalanceSets::$headingSet ) ) {
                    return true; # ignore
                }
                $this->stack->generateImpliedEndTags();
                $this->stack->popTag( BalanceSets::$headingSet );
                return true;

            case 'sarcasm':
                # Take a deep breath, then:
                break;

            case 'a':
            case 'b':
            case 'big':
            case 'code':
            case 'em':
            case 'font':
            case 'i':
            case 'nobr':
            case 's':
            case 'small':
            case 'strike':
            case 'strong':
            case 'tt':
            case 'u':
                if ( $this->stack->adoptionAgency( $value, $this->afe ) ) {
                    return true; # If we did something, we're done.
                }
                break; # Go to the "any other end tag" case.

            case 'applet':
            case 'marquee':
            case 'object':
                if ( !$this->stack->inScope( $value ) ) {
                    return true; # ignore
                }
                $this->stack->generateImpliedEndTags();
                $this->stack->popTag( $value );
                $this->afe->clearToMarker();
                return true;

            case 'br':
                # Turn </br> into <br>
                return $this->inBodyMode( 'tag', $value, [] );
            }

            // Any other end tag goes here
            foreach ( $this->stack as $i => $node ) {
                if ( $node->isHtmlNamed( $value ) ) {
                    $this->stack->generateImpliedEndTags( $value );
                    $this->stack->popTo( $i ); # including $i
                    break;
                } elseif ( $node->isA( BalanceSets::$specialSet ) ) {
                    return true; // ignore this close token.
                }
            }
            return true;
        } elseif ( $token === 'comment' ) {
            $this->stack->insertComment( $value );
            return true;
        } else {
            Assert::invariant( false, "Bad token type: $token" );
        }
    }

    private function inTableMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' ) {
            if ( $this->textIntegrationMode ) {
                return $this->inBodyMode( $token, $value, $attribs, $selfclose );
            } elseif ( $this->stack->currentNode->isA( BalanceSets::$tableSectionRowSet ) ) {
                $this->pendingTableText = '';
                $this->originalInsertionMode = $this->parseMode;
                return $this->switchModeAndReprocess( 'inTableTextMode', $token, $value, $attribs, $selfclose );
            }
            // fall through to default case.
        } elseif ( $token === 'eof' ) {
            $this->stopParsing();
            return true;
        } elseif ( $token === 'tag' ) {
            switch ( $value ) {
            case 'caption':
                $this->afe->insertMarker();
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->switchMode( 'inCaptionMode' );
                return true;
            case 'colgroup':
                $this->stack->clearToContext( BalanceSets::$tableContextSet );
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->switchMode( 'inColumnGroupMode' );
                return true;
            case 'col':
                $this->inTableMode( 'tag', 'colgroup', [] );
                return $this->insertToken( $token, $value, $attribs, $selfclose );
            case 'tbody':
            case 'tfoot':
            case 'thead':
                $this->stack->clearToContext( BalanceSets::$tableContextSet );
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->switchMode( 'inTableBodyMode' );
                return true;
            case 'td':
            case 'th':
            case 'tr':
                $this->inTableMode( 'tag', 'tbody', [] );
                return $this->insertToken( $token, $value, $attribs, $selfclose );
            case 'table':
                if ( !$this->stack->inTableScope( $value ) ) {
                    return true; // Ignore this tag.
                }
                $this->inTableMode( 'endtag', $value );
                return $this->insertToken( $token, $value, $attribs, $selfclose );

            case 'style':
            # OMITTED: <script>
            case 'template':
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );

            case 'input':
                if ( !isset( $attribs['type'] ) || strcasecmp( $attribs['type'], 'hidden' ) !== 0 ) {
                    break; // Handle this as "everything else"
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->stack->pop();
                return true;

            case 'form':
                if (
                    $this->formElementPointer ||
                    $this->stack->indexOf( 'template' ) >= 0
                ) {
                    return true; // ignore this token
                }
                $this->formElementPointer =
                    $this->stack->insertHTMLElement( $value, $attribs );
                $this->stack->popTag( $this->formElementPointer );
                return true;
            }
            // Fall through for "anything else" clause.
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            case 'table':
                if ( !$this->stack->inTableScope( $value ) ) {
                    return true; // Ignore.
                }
                $this->stack->popTag( $value );
                $this->resetInsertionMode();
                return true;
            # OMITTED: <body>
            case 'caption':
            case 'col':
            case 'colgroup':
            # OMITTED: <html>
            case 'tbody':
            case 'td':
            case 'tfoot':
            case 'th':
            case 'thead':
            case 'tr':
                return true; // Ignore the token.
            case 'template':
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );
            }
            // Fall through for "anything else" clause.
        } elseif ( $token === 'comment' ) {
            $this->stack->insertComment( $value );
            return true;
        }
        // This is the "anything else" case:
        $this->stack->fosterParentMode = true;
        $this->inBodyMode( $token, $value, $attribs, $selfclose );
        $this->stack->fosterParentMode = false;
        return true;
    }

    private function inTableTextMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' ) {
            $this->pendingTableText .= $value;
            return true;
        }
        // Non-text token:
        $text = $this->pendingTableText;
        $this->pendingTableText = '';
        if ( preg_match( '/[^\x09\x0A\x0C\x0D\x20]/', $text ) ) {
            // This should match the "anything else" case inTableMode
            $this->stack->fosterParentMode = true;
            $this->inBodyMode( 'text', $text );
            $this->stack->fosterParentMode = false;
        } else {
            // Pending text is just whitespace.
            $this->stack->insertText( $text );
        }
        return $this->switchModeAndReprocess(
            $this->originalInsertionMode, $token, $value, $attribs, $selfclose
        );
    }

    // helper for inCaptionMode
    private function endCaption() {
        if ( !$this->stack->inTableScope( 'caption' ) ) {
            return false;
        }
        $this->stack->generateImpliedEndTags();
        $this->stack->popTag( 'caption' );
        $this->afe->clearToMarker();
        $this->switchMode( 'inTableMode' );
        return true;
    }

    private function inCaptionMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'tag' ) {
            switch ( $value ) {
            case 'caption':
            case 'col':
            case 'colgroup':
            case 'tbody':
            case 'td':
            case 'tfoot':
            case 'th':
            case 'thead':
            case 'tr':
                if ( $this->endCaption() ) {
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            }
            // Fall through to "anything else" case.
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            case 'caption':
                $this->endCaption();
                return true;
            case 'table':
                if ( $this->endCaption() ) {
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            case 'body':
            case 'col':
            case 'colgroup':
            # OMITTED: <html>
            case 'tbody':
            case 'td':
            case 'tfoot':
            case 'th':
            case 'thead':
            case 'tr':
                // Ignore the token
                return true;
            }
            // Fall through to "anything else" case.
        }
        // The Anything Else case
        return $this->inBodyMode( $token, $value, $attribs, $selfclose );
    }

    private function inColumnGroupMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' ) {
            if ( preg_match( '/^[\x09\x0A\x0C\x0D\x20]+/', $value, $matches ) ) {
                $this->stack->insertText( $matches[0] );
                $value = substr( $value, strlen( $matches[0] ) );
            }
            if ( strlen( $value ) === 0 ) {
                return true; // All text handled.
            }
            // Fall through to handle non-whitespace below.
        } elseif ( $token === 'tag' ) {
            switch ( $value ) {
            # OMITTED: <html>
            case 'col':
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->stack->pop();
                return true;
            case 'template':
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );
            }
            // Fall through for "anything else".
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            case 'colgroup':
                if ( !$this->stack->currentNode->isHtmlNamed( 'colgroup' ) ) {
                    return true; // Ignore the token.
                }
                $this->stack->pop();
                $this->switchMode( 'inTableMode' );
                return true;
            case 'col':
                return true; // Ignore the token.
            case 'template':
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );
            }
            // Fall through for "anything else".
        } elseif ( $token === 'eof' ) {
            return $this->inBodyMode( $token, $value, $attribs, $selfclose );
        } elseif ( $token === 'comment' ) {
            $this->stack->insertComment( $value );
            return true;
        }

        // Anything else
        if ( !$this->stack->currentNode->isHtmlNamed( 'colgroup' ) ) {
            return true; // Ignore the token.
        }
        $this->inColumnGroupMode( 'endtag', 'colgroup' );
        return $this->insertToken( $token, $value, $attribs, $selfclose );
    }

    // Helper function for inTableBodyMode
    private function endSection() {
        if ( !(
            $this->stack->inTableScope( 'tbody' ) ||
            $this->stack->inTableScope( 'thead' ) ||
            $this->stack->inTableScope( 'tfoot' )
        ) ) {
            return false;
        }
        $this->stack->clearToContext( BalanceSets::$tableBodyContextSet );
        $this->stack->pop();
        $this->switchMode( 'inTableMode' );
        return true;
    }
    private function inTableBodyMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'tag' ) {
            switch ( $value ) {
            case 'tr':
                $this->stack->clearToContext( BalanceSets::$tableBodyContextSet );
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->switchMode( 'inRowMode' );
                return true;
            case 'th':
            case 'td':
                $this->inTableBodyMode( 'tag', 'tr', [] );
                $this->insertToken( $token, $value, $attribs, $selfclose );
                return true;
            case 'caption':
            case 'col':
            case 'colgroup':
            case 'tbody':
            case 'tfoot':
            case 'thead':
                if ( $this->endSection() ) {
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            }
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            case 'table':
                if ( $this->endSection() ) {
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            case 'tbody':
            case 'tfoot':
            case 'thead':
                if ( $this->stack->inTableScope( $value ) ) {
                    $this->endSection();
                }
                return true;
            # OMITTED: <body>
            case 'caption':
            case 'col':
            case 'colgroup':
            # OMITTED: <html>
            case 'td':
            case 'th':
            case 'tr':
                return true; // Ignore the token.
            }
        }
        // Anything else:
        return $this->inTableMode( $token, $value, $attribs, $selfclose );
    }

    // Helper function for inRowMode
    private function endRow() {
        if ( !$this->stack->inTableScope( 'tr' ) ) {
            return false;
        }
        $this->stack->clearToContext( BalanceSets::$tableRowContextSet );
        $this->stack->pop();
        $this->switchMode( 'inTableBodyMode' );
        return true;
    }
    private function inRowMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'tag' ) {
            switch ( $value ) {
            case 'th':
            case 'td':
                $this->stack->clearToContext( BalanceSets::$tableRowContextSet );
                $this->stack->insertHTMLElement( $value, $attribs );
                $this->switchMode( 'inCellMode' );
                $this->afe->insertMarker();
                return true;
            case 'caption':
            case 'col':
            case 'colgroup':
            case 'tbody':
            case 'tfoot':
            case 'thead':
            case 'tr':
                if ( $this->endRow() ) {
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            }
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            case 'tr':
                $this->endRow();
                return true;
            case 'table':
                if ( $this->endRow() ) {
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            case 'tbody':
            case 'tfoot':
            case 'thead':
                if (
                    $this->stack->inTableScope( $value ) &&
                    $this->endRow()
                ) {
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            # OMITTED: <body>
            case 'caption':
            case 'col':
            case 'colgroup':
            # OMITTED: <html>
            case 'td':
            case 'th':
                return true; // Ignore the token.
            }
        }
        // Anything else:
        return $this->inTableMode( $token, $value, $attribs, $selfclose );
    }

    // Helper for inCellMode
    private function endCell() {
        if ( $this->stack->inTableScope( 'td' ) ) {
            $this->inCellMode( 'endtag', 'td' );
            return true;
        } elseif ( $this->stack->inTableScope( 'th' ) ) {
            $this->inCellMode( 'endtag', 'th' );
            return true;
        } else {
            return false;
        }
    }
    private function inCellMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'tag' ) {
            switch ( $value ) {
            case 'caption':
            case 'col':
            case 'colgroup':
            case 'tbody':
            case 'td':
            case 'tfoot':
            case 'th':
            case 'thead':
            case 'tr':
                if ( $this->endCell() ) {
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            }
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            case 'td':
            case 'th':
                if ( $this->stack->inTableScope( $value ) ) {
                    $this->stack->generateImpliedEndTags();
                    $this->stack->popTag( $value );
                    $this->afe->clearToMarker();
                    $this->switchMode( 'inRowMode' );
                }
                return true;
            # OMITTED: <body>
            case 'caption':
            case 'col':
            case 'colgroup':
            # OMITTED: <html>
                return true;

            case 'table':
            case 'tbody':
            case 'tfoot':
            case 'thead':
            case 'tr':
                if ( $this->stack->inTableScope( $value ) ) {
                    $this->stack->generateImpliedEndTags();
                    $this->stack->popTag( BalanceSets::$tableCellSet );
                    $this->afe->clearToMarker();
                    $this->switchMode( 'inRowMode' );
                    $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            }
        }
        // Anything else:
        return $this->inBodyMode( $token, $value, $attribs, $selfclose );
    }

    private function inSelectMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' ) {
            $this->stack->insertText( $value );
            return true;
        } elseif ( $token === 'eof' ) {
            return $this->inBodyMode( $token, $value, $attribs, $selfclose );
        } elseif ( $token === 'tag' ) {
            switch ( $value ) {
            # OMITTED: <html>
            case 'option':
                if ( $this->stack->currentNode->isHtmlNamed( 'option' ) ) {
                    $this->stack->pop();
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;
            case 'optgroup':
                if ( $this->stack->currentNode->isHtmlNamed( 'option' ) ) {
                    $this->stack->pop();
                }
                if ( $this->stack->currentNode->isHtmlNamed( 'optgroup' ) ) {
                    $this->stack->pop();
                }
                $this->stack->insertHTMLElement( $value, $attribs );
                return true;
            case 'select':
                $this->inSelectMode( 'endtag', $value ); // treat it like endtag
                return true;
            case 'input':
            case 'keygen':
            case 'textarea':
                if ( !$this->stack->inSelectScope( 'select' ) ) {
                    return true; // ignore token (fragment case)
                }
                $this->inSelectMode( 'endtag', 'select' );
                return $this->insertToken( $token, $value, $attribs, $selfclose );
            case 'script':
            case 'template':
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );
            }
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            case 'optgroup':
                if (
                    $this->stack->currentNode->isHtmlNamed( 'option' ) &&
                    $this->stack->length() >= 2 &&
                    $this->stack->node( $this->stack->length() - 2 )->isHtmlNamed( 'optgroup' )
                ) {
                    $this->stack->pop();
                }
                if ( $this->stack->currentNode->isHtmlNamed( 'optgroup' ) ) {
                    $this->stack->pop();
                }
                return true;
            case 'option':
                if ( $this->stack->currentNode->isHtmlNamed( 'option' ) ) {
                    $this->stack->pop();
                }
                return true;
            case 'select':
                if ( !$this->stack->inSelectScope( $value ) ) {
                    return true; // fragment case
                }
                $this->stack->popTag( $value );
                $this->resetInsertionMode();
                return true;
            case 'template':
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );
            }
        } elseif ( $token === 'comment' ) {
            $this->stack->insertComment( $value );
            return true;
        }
        // anything else: just ignore the token
        return true;
    }

    private function inSelectInTableMode( $token, $value, $attribs = null, $selfclose = false ) {
        switch ( $value ) {
        case 'caption':
        case 'table':
        case 'tbody':
        case 'tfoot':
        case 'thead':
        case 'tr':
        case 'td':
        case 'th':
            if ( $token === 'tag' ) {
                $this->inSelectInTableMode( 'endtag', 'select' );
                return $this->insertToken( $token, $value, $attribs, $selfclose );
            } elseif ( $token === 'endtag' ) {
                if ( $this->stack->inTableScope( $value ) ) {
                    $this->inSelectInTableMode( 'endtag', 'select' );
                    return $this->insertToken( $token, $value, $attribs, $selfclose );
                }
                return true;
            }
        }
        // anything else
        return $this->inSelectMode( $token, $value, $attribs, $selfclose );
    }

    private function inTemplateMode( $token, $value, $attribs = null, $selfclose = false ) {
        if ( $token === 'text' || $token === 'comment' ) {
            return $this->inBodyMode( $token, $value, $attribs, $selfclose );
        } elseif ( $token === 'eof' ) {
            if ( $this->stack->indexOf( 'template' ) < 0 ) {
                $this->stopParsing();
            } else {
                $this->stack->popTag( 'template' );
                $this->afe->clearToMarker();
                array_pop( $this->templateInsertionModes );
                $this->resetInsertionMode();
                $this->insertToken( $token, $value, $attribs, $selfclose );
            }
            return true;
        } elseif ( $token === 'tag' ) {
            switch ( $value ) {
            case 'base':
            case 'basefont':
            case 'bgsound':
            case 'link':
            case 'meta':
            case 'noframes':
            # OMITTED: <script>
            case 'style':
            case 'template':
            # OMITTED: <title>
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );

            case 'caption':
            case 'colgroup':
            case 'tbody':
            case 'tfoot':
            case 'thead':
                return $this->switchModeAndReprocess(
                    'inTableMode', $token, $value, $attribs, $selfclose
                );

            case 'col':
                return $this->switchModeAndReprocess(
                    'inColumnGroupMode', $token, $value, $attribs, $selfclose
                );

            case 'tr':
                return $this->switchModeAndReprocess(
                    'inTableBodyMode', $token, $value, $attribs, $selfclose
                );

            case 'td':
            case 'th':
                return $this->switchModeAndReprocess(
                    'inRowMode', $token, $value, $attribs, $selfclose
                );
            }
            return $this->switchModeAndReprocess(
                'inBodyMode', $token, $value, $attribs, $selfclose
            );
        } elseif ( $token === 'endtag' ) {
            switch ( $value ) {
            case 'template':
                return $this->inHeadMode( $token, $value, $attribs, $selfclose );
            }
            return true;
        } else {
            Assert::invariant( false, "Bad token type: $token" );
        }
    }
}