Initial commit

content/inc/Parsing/Handler/AbstractRewriter.php

@@ -0,0 +1,39 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+/**
+ * Basic implementation of the rewriter interface to be specialized by children
+ */
+abstract class AbstractRewriter implements ReWriterInterface
+{
+    /** @var CallWriterInterface original CallWriter */
+    protected $callWriter;
+
+    /** @var array[] list of calls */
+    public $calls = array();
+
+    /** @inheritdoc */
+    public function __construct(CallWriterInterface $callWriter)
+    {
+        $this->callWriter = $callWriter;
+    }
+
+    /** @inheritdoc */
+    public function writeCall($call)
+    {
+        $this->calls[] = $call;
+    }
+
+    /** * @inheritdoc */
+    public function writeCalls($calls)
+    {
+        $this->calls = array_merge($this->calls, $calls);
+    }
+
+    /** @inheritDoc */
+    public function getCallWriter()
+    {
+        return $this->callWriter;
+    }
+}

content/inc/Parsing/Handler/Block.php

@@ -0,0 +1,211 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+/**
+ * Handler for paragraphs
+ *
+ * @author Harry Fuecks <hfuecks@gmail.com>
+ */
+class Block
+{
+    protected $calls = array();
+    protected $skipEol = false;
+    protected $inParagraph = false;
+
+    // Blocks these should not be inside paragraphs
+    protected $blockOpen = array(
+        'header',
+        'listu_open','listo_open','listitem_open','listcontent_open',
+        'table_open','tablerow_open','tablecell_open','tableheader_open','tablethead_open',
+        'quote_open',
+        'code','file','hr','preformatted','rss',
+        'htmlblock','phpblock',
+        'footnote_open',
+    );
+
+    protected $blockClose = array(
+        'header',
+        'listu_close','listo_close','listitem_close','listcontent_close',
+        'table_close','tablerow_close','tablecell_close','tableheader_close','tablethead_close',
+        'quote_close',
+        'code','file','hr','preformatted','rss',
+        'htmlblock','phpblock',
+        'footnote_close',
+    );
+
+    // Stacks can contain paragraphs
+    protected $stackOpen = array(
+        'section_open',
+    );
+
+    protected $stackClose = array(
+        'section_close',
+    );
+
+
+    /**
+     * Constructor. Adds loaded syntax plugins to the block and stack
+     * arrays
+     *
+     * @author Andreas Gohr <andi@splitbrain.org>
+     */
+    public function __construct()
+    {
+        global $DOKU_PLUGINS;
+        //check if syntax plugins were loaded
+        if (empty($DOKU_PLUGINS['syntax'])) return;
+        foreach ($DOKU_PLUGINS['syntax'] as $n => $p) {
+            $ptype = $p->getPType();
+            if ($ptype == 'block') {
+                $this->blockOpen[]  = 'plugin_'.$n;
+                $this->blockClose[] = 'plugin_'.$n;
+            } elseif ($ptype == 'stack') {
+                $this->stackOpen[]  = 'plugin_'.$n;
+                $this->stackClose[] = 'plugin_'.$n;
+            }
+        }
+    }
+
+    protected function openParagraph($pos)
+    {
+        if ($this->inParagraph) return;
+        $this->calls[] = array('p_open',array(), $pos);
+        $this->inParagraph = true;
+        $this->skipEol = true;
+    }
+
+    /**
+     * Close a paragraph if needed
+     *
+     * This function makes sure there are no empty paragraphs on the stack
+     *
+     * @author Andreas Gohr <andi@splitbrain.org>
+     *
+     * @param string|integer $pos
+     */
+    protected function closeParagraph($pos)
+    {
+        if (!$this->inParagraph) return;
+        // look back if there was any content - we don't want empty paragraphs
+        $content = '';
+        $ccount = count($this->calls);
+        for ($i=$ccount-1; $i>=0; $i--) {
+            if ($this->calls[$i][0] == 'p_open') {
+                break;
+            } elseif ($this->calls[$i][0] == 'cdata') {
+                $content .= $this->calls[$i][1][0];
+            } else {
+                $content = 'found markup';
+                break;
+            }
+        }
+
+        if (trim($content)=='') {
+            //remove the whole paragraph
+            //array_splice($this->calls,$i); // <- this is much slower than the loop below
+            for ($x=$ccount; $x>$i;
+            $x--) array_pop($this->calls);
+        } else {
+            // remove ending linebreaks in the paragraph
+            $i=count($this->calls)-1;
+            if ($this->calls[$i][0] == 'cdata') $this->calls[$i][1][0] = rtrim($this->calls[$i][1][0], "\n");
+            $this->calls[] = array('p_close',array(), $pos);
+        }
+
+        $this->inParagraph = false;
+        $this->skipEol = true;
+    }
+
+    protected function addCall($call)
+    {
+        $key = count($this->calls);
+        if ($key and ($call[0] == 'cdata') and ($this->calls[$key-1][0] == 'cdata')) {
+            $this->calls[$key-1][1][0] .= $call[1][0];
+        } else {
+            $this->calls[] = $call;
+        }
+    }
+
+    // simple version of addCall, without checking cdata
+    protected function storeCall($call)
+    {
+        $this->calls[] = $call;
+    }
+
+    /**
+     * Processes the whole instruction stack to open and close paragraphs
+     *
+     * @author Harry Fuecks <hfuecks@gmail.com>
+     * @author Andreas Gohr <andi@splitbrain.org>
+     *
+     * @param array $calls
+     *
+     * @return array
+     */
+    public function process($calls)
+    {
+        // open first paragraph
+        $this->openParagraph(0);
+        foreach ($calls as $key => $call) {
+            $cname = $call[0];
+            if ($cname == 'plugin') {
+                $cname='plugin_'.$call[1][0];
+                $plugin = true;
+                $plugin_open = (($call[1][2] == DOKU_LEXER_ENTER) || ($call[1][2] == DOKU_LEXER_SPECIAL));
+                $plugin_close = (($call[1][2] == DOKU_LEXER_EXIT) || ($call[1][2] == DOKU_LEXER_SPECIAL));
+            } else {
+                $plugin = false;
+            }
+            /* stack */
+            if (in_array($cname, $this->stackClose) && (!$plugin || $plugin_close)) {
+                $this->closeParagraph($call[2]);
+                $this->storeCall($call);
+                $this->openParagraph($call[2]);
+                continue;
+            }
+            if (in_array($cname, $this->stackOpen) && (!$plugin || $plugin_open)) {
+                $this->closeParagraph($call[2]);
+                $this->storeCall($call);
+                $this->openParagraph($call[2]);
+                continue;
+            }
+            /* block */
+            // If it's a substition it opens and closes at the same call.
+            // To make sure next paragraph is correctly started, let close go first.
+            if (in_array($cname, $this->blockClose) && (!$plugin || $plugin_close)) {
+                $this->closeParagraph($call[2]);
+                $this->storeCall($call);
+                $this->openParagraph($call[2]);
+                continue;
+            }
+            if (in_array($cname, $this->blockOpen) && (!$plugin || $plugin_open)) {
+                $this->closeParagraph($call[2]);
+                $this->storeCall($call);
+                continue;
+            }
+            /* eol */
+            if ($cname == 'eol') {
+                // Check this isn't an eol instruction to skip...
+                if (!$this->skipEol) {
+                    // Next is EOL => double eol => mark as paragraph
+                    if (isset($calls[$key+1]) && $calls[$key+1][0] == 'eol') {
+                        $this->closeParagraph($call[2]);
+                        $this->openParagraph($call[2]);
+                    } else {
+                        //if this is just a single eol make a space from it
+                        $this->addCall(array('cdata',array("\n"), $call[2]));
+                    }
+                }
+                continue;
+            }
+            /* normal */
+            $this->addCall($call);
+            $this->skipEol = false;
+        }
+        // close last paragraph
+        $call = end($this->calls);
+        $this->closeParagraph($call[2]);
+        return $this->calls;
+    }
+}

content/inc/Parsing/Handler/CallWriter.php

@@ -0,0 +1,40 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+class CallWriter implements CallWriterInterface
+{
+
+    /** @var \Doku_Handler $Handler */
+    protected $Handler;
+
+    /**
+     * @param \Doku_Handler $Handler
+     */
+    public function __construct(\Doku_Handler $Handler)
+    {
+        $this->Handler = $Handler;
+    }
+
+    /** @inheritdoc */
+    public function writeCall($call)
+    {
+        $this->Handler->calls[] = $call;
+    }
+
+    /** @inheritdoc */
+    public function writeCalls($calls)
+    {
+        $this->Handler->calls = array_merge($this->Handler->calls, $calls);
+    }
+
+    /**
+     * @inheritdoc
+     * function is required, but since this call writer is first/highest in
+     * the chain it is not required to do anything
+     */
+    public function finalise()
+    {
+        unset($this->Handler);
+    }
+}

content/inc/Parsing/Handler/CallWriterInterface.php

@@ -0,0 +1,30 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+interface CallWriterInterface
+{
+    /**
+     * Add a call to our call list
+     *
+     * @param array $call the call to be added
+     */
+    public function writeCall($call);
+
+    /**
+     * Append a list of calls to our call list
+     *
+     * @param array[] $calls list of calls to be appended
+     */
+    public function writeCalls($calls);
+
+    /**
+     * Explicit request to finish up and clean up NOW!
+     * (probably because document end has been reached)
+     *
+     * If part of a CallWriter chain, call finalise on
+     * the original call writer
+     *
+     */
+    public function finalise();
+}

content/inc/Parsing/Handler/Lists.php

@@ -0,0 +1,186 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+class Lists extends AbstractRewriter
+{
+    protected $listCalls = array();
+    protected $listStack = array();
+
+    protected $initialDepth = 0;
+
+    const NODE = 1;
+
+    /** @inheritdoc */
+    public function finalise()
+    {
+        $last_call = end($this->calls);
+        $this->writeCall(array('list_close',array(), $last_call[2]));
+
+        $this->process();
+        $this->callWriter->finalise();
+        unset($this->callWriter);
+    }
+
+    /** @inheritdoc */
+    public function process()
+    {
+
+        foreach ($this->calls as $call) {
+            switch ($call[0]) {
+                case 'list_item':
+                    $this->listOpen($call);
+                    break;
+                case 'list_open':
+                    $this->listStart($call);
+                    break;
+                case 'list_close':
+                    $this->listEnd($call);
+                    break;
+                default:
+                    $this->listContent($call);
+                    break;
+            }
+        }
+
+        $this->callWriter->writeCalls($this->listCalls);
+        return $this->callWriter;
+    }
+
+    protected function listStart($call)
+    {
+        $depth = $this->interpretSyntax($call[1][0], $listType);
+
+        $this->initialDepth = $depth;
+        //                   array(list type, current depth, index of current listitem_open)
+        $this->listStack[] = array($listType, $depth, 1);
+
+        $this->listCalls[] = array('list'.$listType.'_open',array(),$call[2]);
+        $this->listCalls[] = array('listitem_open',array(1),$call[2]);
+        $this->listCalls[] = array('listcontent_open',array(),$call[2]);
+    }
+
+
+    protected function listEnd($call)
+    {
+        $closeContent = true;
+
+        while ($list = array_pop($this->listStack)) {
+            if ($closeContent) {
+                $this->listCalls[] = array('listcontent_close',array(),$call[2]);
+                $closeContent = false;
+            }
+            $this->listCalls[] = array('listitem_close',array(),$call[2]);
+            $this->listCalls[] = array('list'.$list[0].'_close', array(), $call[2]);
+        }
+    }
+
+    protected function listOpen($call)
+    {
+        $depth = $this->interpretSyntax($call[1][0], $listType);
+        $end = end($this->listStack);
+        $key = key($this->listStack);
+
+        // Not allowed to be shallower than initialDepth
+        if ($depth < $this->initialDepth) {
+            $depth = $this->initialDepth;
+        }
+
+        if ($depth == $end[1]) {
+            // Just another item in the list...
+            if ($listType == $end[0]) {
+                $this->listCalls[] = array('listcontent_close',array(),$call[2]);
+                $this->listCalls[] = array('listitem_close',array(),$call[2]);
+                $this->listCalls[] = array('listitem_open',array($depth-1),$call[2]);
+                $this->listCalls[] = array('listcontent_open',array(),$call[2]);
+
+                // new list item, update list stack's index into current listitem_open
+                $this->listStack[$key][2] = count($this->listCalls) - 2;
+
+                // Switched list type...
+            } else {
+                $this->listCalls[] = array('listcontent_close',array(),$call[2]);
+                $this->listCalls[] = array('listitem_close',array(),$call[2]);
+                $this->listCalls[] = array('list'.$end[0].'_close', array(), $call[2]);
+                $this->listCalls[] = array('list'.$listType.'_open', array(), $call[2]);
+                $this->listCalls[] = array('listitem_open', array($depth-1), $call[2]);
+                $this->listCalls[] = array('listcontent_open',array(),$call[2]);
+
+                array_pop($this->listStack);
+                $this->listStack[] = array($listType, $depth, count($this->listCalls) - 2);
+            }
+        } elseif ($depth > $end[1]) { // Getting deeper...
+            $this->listCalls[] = array('listcontent_close',array(),$call[2]);
+            $this->listCalls[] = array('list'.$listType.'_open', array(), $call[2]);
+            $this->listCalls[] = array('listitem_open', array($depth-1), $call[2]);
+            $this->listCalls[] = array('listcontent_open',array(),$call[2]);
+
+            // set the node/leaf state of this item's parent listitem_open to NODE
+            $this->listCalls[$this->listStack[$key][2]][1][1] = self::NODE;
+
+            $this->listStack[] = array($listType, $depth, count($this->listCalls) - 2);
+        } else { // Getting shallower ( $depth < $end[1] )
+            $this->listCalls[] = array('listcontent_close',array(),$call[2]);
+            $this->listCalls[] = array('listitem_close',array(),$call[2]);
+            $this->listCalls[] = array('list'.$end[0].'_close',array(),$call[2]);
+
+            // Throw away the end - done
+            array_pop($this->listStack);
+
+            while (1) {
+                $end = end($this->listStack);
+                $key = key($this->listStack);
+
+                if ($end[1] <= $depth) {
+                    // Normalize depths
+                    $depth = $end[1];
+
+                    $this->listCalls[] = array('listitem_close',array(),$call[2]);
+
+                    if ($end[0] == $listType) {
+                        $this->listCalls[] = array('listitem_open',array($depth-1),$call[2]);
+                        $this->listCalls[] = array('listcontent_open',array(),$call[2]);
+
+                        // new list item, update list stack's index into current listitem_open
+                        $this->listStack[$key][2] = count($this->listCalls) - 2;
+                    } else {
+                        // Switching list type...
+                        $this->listCalls[] = array('list'.$end[0].'_close', array(), $call[2]);
+                        $this->listCalls[] = array('list'.$listType.'_open', array(), $call[2]);
+                        $this->listCalls[] = array('listitem_open', array($depth-1), $call[2]);
+                        $this->listCalls[] = array('listcontent_open',array(),$call[2]);
+
+                        array_pop($this->listStack);
+                        $this->listStack[] = array($listType, $depth, count($this->listCalls) - 2);
+                    }
+
+                    break;
+
+                    // Haven't dropped down far enough yet.... ( $end[1] > $depth )
+                } else {
+                    $this->listCalls[] = array('listitem_close',array(),$call[2]);
+                    $this->listCalls[] = array('list'.$end[0].'_close',array(),$call[2]);
+
+                    array_pop($this->listStack);
+                }
+            }
+        }
+    }
+
+    protected function listContent($call)
+    {
+        $this->listCalls[] = $call;
+    }
+
+    protected function interpretSyntax($match, & $type)
+    {
+        if (substr($match, -1) == '*') {
+            $type = 'u';
+        } else {
+            $type = 'o';
+        }
+        // Is the +1 needed? It used to be count(explode(...))
+        // but I don't think the number is seen outside this handler
+        return substr_count(str_replace("\t", '  ', $match), '  ') + 1;
+    }
+}

content/inc/Parsing/Handler/Nest.php

@@ -0,0 +1,82 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+/**
+ * Generic call writer class to handle nesting of rendering instructions
+ * within a render instruction. Also see nest() method of renderer base class
+ *
+ * @author    Chris Smith <chris@jalakai.co.uk>
+ */
+class Nest extends AbstractRewriter
+{
+    protected $closingInstruction;
+
+    /**
+     * @inheritdoc
+     *
+     * @param  CallWriterInterface $CallWriter     the parser's current call writer, i.e. the one above us in the chain
+     * @param  string     $close          closing instruction name, this is required to properly terminate the
+     *                                    syntax mode if the document ends without a closing pattern
+     */
+    public function __construct(CallWriterInterface $CallWriter, $close = "nest_close")
+    {
+        parent::__construct($CallWriter);
+        $this->closingInstruction = $close;
+    }
+
+    /** @inheritdoc */
+    public function writeCall($call)
+    {
+        $this->calls[] = $call;
+    }
+
+    /** @inheritdoc */
+    public function writeCalls($calls)
+    {
+        $this->calls = array_merge($this->calls, $calls);
+    }
+
+    /** @inheritdoc */
+    public function finalise()
+    {
+        $last_call = end($this->calls);
+        $this->writeCall(array($this->closingInstruction,array(), $last_call[2]));
+
+        $this->process();
+        $this->callWriter->finalise();
+        unset($this->callWriter);
+    }
+
+    /** @inheritdoc */
+    public function process()
+    {
+        // merge consecutive cdata
+        $unmerged_calls = $this->calls;
+        $this->calls = array();
+
+        foreach ($unmerged_calls as $call) $this->addCall($call);
+
+        $first_call = reset($this->calls);
+        $this->callWriter->writeCall(array("nest", array($this->calls), $first_call[2]));
+
+        return $this->callWriter;
+    }
+
+    /**
+     * @param array $call
+     */
+    protected function addCall($call)
+    {
+        $key = count($this->calls);
+        if ($key and ($call[0] == 'cdata') and ($this->calls[$key-1][0] == 'cdata')) {
+            $this->calls[$key-1][1][0] .= $call[1][0];
+        } elseif ($call[0] == 'eol') {
+            // do nothing (eol shouldn't be allowed, to counter preformatted fix in #1652 & #1699)
+        } else {
+            $this->calls[] = $call;
+        }
+    }
+
+
+}

content/inc/Parsing/Handler/Preformatted.php

@@ -0,0 +1,49 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+class Preformatted extends AbstractRewriter
+{
+
+    protected $pos;
+    protected $text ='';
+
+    /** @inheritdoc */
+    public function finalise()
+    {
+        $last_call = end($this->calls);
+        $this->writeCall(array('preformatted_end',array(), $last_call[2]));
+
+        $this->process();
+        $this->callWriter->finalise();
+        unset($this->callWriter);
+    }
+
+    /** @inheritdoc */
+    public function process()
+    {
+        foreach ($this->calls as $call) {
+            switch ($call[0]) {
+                case 'preformatted_start':
+                    $this->pos = $call[2];
+                    break;
+                case 'preformatted_newline':
+                    $this->text .= "\n";
+                    break;
+                case 'preformatted_content':
+                    $this->text .= $call[1][0];
+                    break;
+                case 'preformatted_end':
+                    if (trim($this->text)) {
+                        $this->callWriter->writeCall(array('preformatted', array($this->text), $this->pos));
+                    }
+                    // see FS#1699 & FS#1652, add 'eol' instructions to ensure proper triggering of following p_open
+                    $this->callWriter->writeCall(array('eol', array(), $this->pos));
+                    $this->callWriter->writeCall(array('eol', array(), $this->pos));
+                    break;
+            }
+        }
+
+        return $this->callWriter;
+    }
+}

content/inc/Parsing/Handler/Quote.php

@@ -0,0 +1,86 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+class Quote extends AbstractRewriter
+{
+    protected $quoteCalls = array();
+
+    /** @inheritdoc */
+    public function finalise()
+    {
+        $last_call = end($this->calls);
+        $this->writeCall(array('quote_end',array(), $last_call[2]));
+
+        $this->process();
+        $this->callWriter->finalise();
+        unset($this->callWriter);
+    }
+
+    /** @inheritdoc */
+    public function process()
+    {
+
+        $quoteDepth = 1;
+
+        foreach ($this->calls as $call) {
+            switch ($call[0]) {
+
+                /** @noinspection PhpMissingBreakStatementInspection */
+                case 'quote_start':
+                    $this->quoteCalls[] = array('quote_open',array(),$call[2]);
+                    // fallthrough
+                case 'quote_newline':
+                    $quoteLength = $this->getDepth($call[1][0]);
+
+                    if ($quoteLength > $quoteDepth) {
+                        $quoteDiff = $quoteLength - $quoteDepth;
+                        for ($i = 1; $i <= $quoteDiff; $i++) {
+                            $this->quoteCalls[] = array('quote_open',array(),$call[2]);
+                        }
+                    } elseif ($quoteLength < $quoteDepth) {
+                        $quoteDiff = $quoteDepth - $quoteLength;
+                        for ($i = 1; $i <= $quoteDiff; $i++) {
+                            $this->quoteCalls[] = array('quote_close',array(),$call[2]);
+                        }
+                    } else {
+                        if ($call[0] != 'quote_start') $this->quoteCalls[] = array('linebreak',array(),$call[2]);
+                    }
+
+                    $quoteDepth = $quoteLength;
+
+                    break;
+
+                case 'quote_end':
+                    if ($quoteDepth > 1) {
+                        $quoteDiff = $quoteDepth - 1;
+                        for ($i = 1; $i <= $quoteDiff; $i++) {
+                            $this->quoteCalls[] = array('quote_close',array(),$call[2]);
+                        }
+                    }
+
+                    $this->quoteCalls[] = array('quote_close',array(),$call[2]);
+
+                    $this->callWriter->writeCalls($this->quoteCalls);
+                    break;
+
+                default:
+                    $this->quoteCalls[] = $call;
+                    break;
+            }
+        }
+
+        return $this->callWriter;
+    }
+
+    /**
+     * @param string $marker
+     * @return int
+     */
+    protected function getDepth($marker)
+    {
+        preg_match('/>{1,}/', $marker, $matches);
+        $quoteLength = strlen($matches[0]);
+        return $quoteLength;
+    }
+}

content/inc/Parsing/Handler/ReWriterInterface.php

@@ -0,0 +1,37 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+/**
+ * A ReWriter takes over from the orignal call writer and handles all new calls itself until
+ * the process method is called and control is given back to the original writer.
+ *
+ * @property array[] $calls The list of current calls
+ */
+interface ReWriterInterface extends CallWriterInterface
+{
+    /**
+     * ReWriterInterface constructor.
+     *
+     * This rewriter will be registered as the new call writer in the Handler.
+     * The original is passed as parameter
+     *
+     * @param CallWriterInterface $callWriter the original callwriter
+     */
+    public function __construct(CallWriterInterface $callWriter);
+
+    /**
+     * Process any calls that have been added and add them to the
+     * original call writer
+     *
+     * @return CallWriterInterface the orignal call writer
+     */
+    public function process();
+
+    /**
+     * Accessor for this rewriter's original CallWriter
+     *
+     * @return CallWriterInterface
+     */
+    public function getCallWriter();
+}

content/inc/Parsing/Handler/Table.php

@@ -0,0 +1,320 @@
+<?php
+
+namespace dokuwiki\Parsing\Handler;
+
+class Table extends AbstractRewriter
+{
+
+    protected $tableCalls = array();
+    protected $maxCols = 0;
+    protected $maxRows = 1;
+    protected $currentCols = 0;
+    protected $firstCell = false;
+    protected $lastCellType = 'tablecell';
+    protected $inTableHead = true;
+    protected $currentRow = array('tableheader' => 0, 'tablecell' => 0);
+    protected $countTableHeadRows = 0;
+
+    /** @inheritdoc */
+    public function finalise()
+    {
+        $last_call = end($this->calls);
+        $this->writeCall(array('table_end',array(), $last_call[2]));
+
+        $this->process();
+        $this->callWriter->finalise();
+        unset($this->callWriter);
+    }
+
+    /** @inheritdoc */
+    public function process()
+    {
+        foreach ($this->calls as $call) {
+            switch ($call[0]) {
+                case 'table_start':
+                    $this->tableStart($call);
+                    break;
+                case 'table_row':
+                    $this->tableRowClose($call);
+                    $this->tableRowOpen(array('tablerow_open',$call[1],$call[2]));
+                    break;
+                case 'tableheader':
+                case 'tablecell':
+                    $this->tableCell($call);
+                    break;
+                case 'table_end':
+                    $this->tableRowClose($call);
+                    $this->tableEnd($call);
+                    break;
+                default:
+                    $this->tableDefault($call);
+                    break;
+            }
+        }
+        $this->callWriter->writeCalls($this->tableCalls);
+
+        return $this->callWriter;
+    }
+
+    protected function tableStart($call)
+    {
+        $this->tableCalls[] = array('table_open',$call[1],$call[2]);
+        $this->tableCalls[] = array('tablerow_open',array(),$call[2]);
+        $this->firstCell = true;
+    }
+
+    protected function tableEnd($call)
+    {
+        $this->tableCalls[] = array('table_close',$call[1],$call[2]);
+        $this->finalizeTable();
+    }
+
+    protected function tableRowOpen($call)
+    {
+        $this->tableCalls[] = $call;
+        $this->currentCols = 0;
+        $this->firstCell = true;
+        $this->lastCellType = 'tablecell';
+        $this->maxRows++;
+        if ($this->inTableHead) {
+            $this->currentRow = array('tablecell' => 0, 'tableheader' => 0);
+        }
+    }
+
+    protected function tableRowClose($call)
+    {
+        if ($this->inTableHead && ($this->inTableHead = $this->isTableHeadRow())) {
+            $this->countTableHeadRows++;
+        }
+        // Strip off final cell opening and anything after it
+        while ($discard = array_pop($this->tableCalls)) {
+            if ($discard[0] == 'tablecell_open' || $discard[0] == 'tableheader_open') {
+                break;
+            }
+            if (!empty($this->currentRow[$discard[0]])) {
+                $this->currentRow[$discard[0]]--;
+            }
+        }
+        $this->tableCalls[] = array('tablerow_close', array(), $call[2]);
+
+        if ($this->currentCols > $this->maxCols) {
+            $this->maxCols = $this->currentCols;
+        }
+    }
+
+    protected function isTableHeadRow()
+    {
+        $td = $this->currentRow['tablecell'];
+        $th = $this->currentRow['tableheader'];
+
+        if (!$th || $td > 2) return false;
+        if (2*$td > $th) return false;
+
+        return true;
+    }
+
+    protected function tableCell($call)
+    {
+        if ($this->inTableHead) {
+            $this->currentRow[$call[0]]++;
+        }
+        if (!$this->firstCell) {
+            // Increase the span
+            $lastCall = end($this->tableCalls);
+
+            // A cell call which follows an open cell means an empty cell so span
+            if ($lastCall[0] == 'tablecell_open' || $lastCall[0] == 'tableheader_open') {
+                $this->tableCalls[] = array('colspan',array(),$call[2]);
+            }
+
+            $this->tableCalls[] = array($this->lastCellType.'_close',array(),$call[2]);
+            $this->tableCalls[] = array($call[0].'_open',array(1,null,1),$call[2]);
+            $this->lastCellType = $call[0];
+        } else {
+            $this->tableCalls[] = array($call[0].'_open',array(1,null,1),$call[2]);
+            $this->lastCellType = $call[0];
+            $this->firstCell = false;
+        }
+
+        $this->currentCols++;
+    }
+
+    protected function tableDefault($call)
+    {
+        $this->tableCalls[] = $call;
+    }
+
+    protected function finalizeTable()
+    {
+
+        // Add the max cols and rows to the table opening
+        if ($this->tableCalls[0][0] == 'table_open') {
+            // Adjust to num cols not num col delimeters
+            $this->tableCalls[0][1][] = $this->maxCols - 1;
+            $this->tableCalls[0][1][] = $this->maxRows;
+            $this->tableCalls[0][1][] = array_shift($this->tableCalls[0][1]);
+        } else {
+            trigger_error('First element in table call list is not table_open');
+        }
+
+        $lastRow = 0;
+        $lastCell = 0;
+        $cellKey = array();
+        $toDelete = array();
+
+        // if still in tableheader, then there can be no table header
+        // as all rows can't be within <THEAD>
+        if ($this->inTableHead) {
+            $this->inTableHead = false;
+            $this->countTableHeadRows = 0;
+        }
+
+        // Look for the colspan elements and increment the colspan on the
+        // previous non-empty opening cell. Once done, delete all the cells
+        // that contain colspans
+        for ($key = 0; $key < count($this->tableCalls); ++$key) {
+            $call = $this->tableCalls[$key];
+
+            switch ($call[0]) {
+                case 'table_open':
+                    if ($this->countTableHeadRows) {
+                        array_splice($this->tableCalls, $key+1, 0, array(
+                                                          array('tablethead_open', array(), $call[2])));
+                    }
+                    break;
+
+                case 'tablerow_open':
+                    $lastRow++;
+                    $lastCell = 0;
+                    break;
+
+                case 'tablecell_open':
+                case 'tableheader_open':
+                    $lastCell++;
+                    $cellKey[$lastRow][$lastCell] = $key;
+                    break;
+
+                case 'table_align':
+                    $prev = in_array($this->tableCalls[$key-1][0], array('tablecell_open', 'tableheader_open'));
+                    $next = in_array($this->tableCalls[$key+1][0], array('tablecell_close', 'tableheader_close'));
+                    // If the cell is empty, align left
+                    if ($prev && $next) {
+                        $this->tableCalls[$key-1][1][1] = 'left';
+
+                        // If the previous element was a cell open, align right
+                    } elseif ($prev) {
+                        $this->tableCalls[$key-1][1][1] = 'right';
+
+                        // If the next element is the close of an element, align either center or left
+                    } elseif ($next) {
+                        if ($this->tableCalls[$cellKey[$lastRow][$lastCell]][1][1] == 'right') {
+                            $this->tableCalls[$cellKey[$lastRow][$lastCell]][1][1] = 'center';
+                        } else {
+                            $this->tableCalls[$cellKey[$lastRow][$lastCell]][1][1] = 'left';
+                        }
+                    }
+
+                    // Now convert the whitespace back to cdata
+                    $this->tableCalls[$key][0] = 'cdata';
+                    break;
+
+                case 'colspan':
+                    $this->tableCalls[$key-1][1][0] = false;
+
+                    for ($i = $key-2; $i >= $cellKey[$lastRow][1]; $i--) {
+                        if ($this->tableCalls[$i][0] == 'tablecell_open' ||
+                            $this->tableCalls[$i][0] == 'tableheader_open'
+                        ) {
+                            if (false !== $this->tableCalls[$i][1][0]) {
+                                $this->tableCalls[$i][1][0]++;
+                                break;
+                            }
+                        }
+                    }
+
+                    $toDelete[] = $key-1;
+                    $toDelete[] = $key;
+                    $toDelete[] = $key+1;
+                    break;
+
+                case 'rowspan':
+                    if ($this->tableCalls[$key-1][0] == 'cdata') {
+                        // ignore rowspan if previous call was cdata (text mixed with :::)
+                        // we don't have to check next call as that wont match regex
+                        $this->tableCalls[$key][0] = 'cdata';
+                    } else {
+                        $spanning_cell = null;
+
+                        // can't cross thead/tbody boundary
+                        if (!$this->countTableHeadRows || ($lastRow-1 != $this->countTableHeadRows)) {
+                            for ($i = $lastRow-1; $i > 0; $i--) {
+                                if ($this->tableCalls[$cellKey[$i][$lastCell]][0] == 'tablecell_open' ||
+                                    $this->tableCalls[$cellKey[$i][$lastCell]][0] == 'tableheader_open'
+                                ) {
+                                    if ($this->tableCalls[$cellKey[$i][$lastCell]][1][2] >= $lastRow - $i) {
+                                        $spanning_cell = $i;
+                                        break;
+                                    }
+                                }
+                            }
+                        }
+                        if (is_null($spanning_cell)) {
+                            // No spanning cell found, so convert this cell to
+                            // an empty one to avoid broken tables
+                            $this->tableCalls[$key][0] = 'cdata';
+                            $this->tableCalls[$key][1][0] = '';
+                            break;
+                        }
+                        $this->tableCalls[$cellKey[$spanning_cell][$lastCell]][1][2]++;
+
+                        $this->tableCalls[$key-1][1][2] = false;
+
+                        $toDelete[] = $key-1;
+                        $toDelete[] = $key;
+                        $toDelete[] = $key+1;
+                    }
+                    break;
+
+                case 'tablerow_close':
+                    // Fix broken tables by adding missing cells
+                    $moreCalls = array();
+                    while (++$lastCell < $this->maxCols) {
+                        $moreCalls[] = array('tablecell_open', array(1, null, 1), $call[2]);
+                        $moreCalls[] = array('cdata', array(''), $call[2]);
+                        $moreCalls[] = array('tablecell_close', array(), $call[2]);
+                    }
+                    $moreCallsLength = count($moreCalls);
+                    if ($moreCallsLength) {
+                        array_splice($this->tableCalls, $key, 0, $moreCalls);
+                        $key += $moreCallsLength;
+                    }
+
+                    if ($this->countTableHeadRows == $lastRow) {
+                        array_splice($this->tableCalls, $key+1, 0, array(
+                            array('tablethead_close', array(), $call[2])));
+                    }
+                    break;
+            }
+        }
+
+        // condense cdata
+        $cnt = count($this->tableCalls);
+        for ($key = 0; $key < $cnt; $key++) {
+            if ($this->tableCalls[$key][0] == 'cdata') {
+                $ckey = $key;
+                $key++;
+                while ($this->tableCalls[$key][0] == 'cdata') {
+                    $this->tableCalls[$ckey][1][0] .= $this->tableCalls[$key][1][0];
+                    $toDelete[] = $key;
+                    $key++;
+                }
+                continue;
+            }
+        }
+
+        foreach ($toDelete as $delete) {
+            unset($this->tableCalls[$delete]);
+        }
+        $this->tableCalls = array_values($this->tableCalls);
+    }
+}

content/inc/Parsing/Lexer/Lexer.php

@@ -0,0 +1,349 @@
+<?php
+/**
+ * Lexer adapted from Simple Test: http://sourceforge.net/projects/simpletest/
+ * For an intro to the Lexer see:
+ * https://web.archive.org/web/20120125041816/http://www.phppatterns.com/docs/develop/simple_test_lexer_notes
+ *
+ * @author Marcus Baker http://www.lastcraft.com
+ */
+
+namespace dokuwiki\Parsing\Lexer;
+
+/**
+ * Accepts text and breaks it into tokens.
+ *
+ * Some optimisation to make the sure the content is only scanned by the PHP regex
+ * parser once. Lexer modes must not start with leading underscores.
+ */
+class Lexer
+{
+    /** @var ParallelRegex[] */
+    protected $regexes;
+    /** @var \Doku_Handler */
+    protected $handler;
+    /** @var StateStack */
+    protected $modeStack;
+    /** @var array mode "rewrites" */
+    protected $mode_handlers;
+    /** @var bool case sensitive? */
+    protected $case;
+
+    /**
+     * Sets up the lexer in case insensitive matching by default.
+     *
+     * @param \Doku_Handler $handler  Handling strategy by reference.
+     * @param string $start            Starting handler.
+     * @param boolean $case            True for case sensitive.
+     */
+    public function __construct($handler, $start = "accept", $case = false)
+    {
+        $this->case = $case;
+        $this->regexes = array();
+        $this->handler = $handler;
+        $this->modeStack = new StateStack($start);
+        $this->mode_handlers = array();
+    }
+
+    /**
+     * Adds a token search pattern for a particular parsing mode.
+     *
+     * The pattern does not change the current mode.
+     *
+     * @param string $pattern      Perl style regex, but ( and )
+     *                             lose the usual meaning.
+     * @param string $mode         Should only apply this
+     *                             pattern when dealing with
+     *                             this type of input.
+     */
+    public function addPattern($pattern, $mode = "accept")
+    {
+        if (! isset($this->regexes[$mode])) {
+            $this->regexes[$mode] = new ParallelRegex($this->case);
+        }
+        $this->regexes[$mode]->addPattern($pattern);
+    }
+
+    /**
+     * Adds a pattern that will enter a new parsing mode.
+     *
+     * Useful for entering parenthesis, strings, tags, etc.
+     *
+     * @param string $pattern      Perl style regex, but ( and ) lose the usual meaning.
+     * @param string $mode         Should only apply this pattern when dealing with this type of input.
+     * @param string $new_mode     Change parsing to this new nested mode.
+     */
+    public function addEntryPattern($pattern, $mode, $new_mode)
+    {
+        if (! isset($this->regexes[$mode])) {
+            $this->regexes[$mode] = new ParallelRegex($this->case);
+        }
+        $this->regexes[$mode]->addPattern($pattern, $new_mode);
+    }
+
+    /**
+     * Adds a pattern that will exit the current mode and re-enter the previous one.
+     *
+     * @param string $pattern      Perl style regex, but ( and ) lose the usual meaning.
+     * @param string $mode         Mode to leave.
+     */
+    public function addExitPattern($pattern, $mode)
+    {
+        if (! isset($this->regexes[$mode])) {
+            $this->regexes[$mode] = new ParallelRegex($this->case);
+        }
+        $this->regexes[$mode]->addPattern($pattern, "__exit");
+    }
+
+    /**
+     * Adds a pattern that has a special mode.
+     *
+     * Acts as an entry and exit pattern in one go, effectively calling a special
+     * parser handler for this token only.
+     *
+     * @param string $pattern      Perl style regex, but ( and ) lose the usual meaning.
+     * @param string $mode         Should only apply this pattern when dealing with this type of input.
+     * @param string $special      Use this mode for this one token.
+     */
+    public function addSpecialPattern($pattern, $mode, $special)
+    {
+        if (! isset($this->regexes[$mode])) {
+            $this->regexes[$mode] = new ParallelRegex($this->case);
+        }
+        $this->regexes[$mode]->addPattern($pattern, "_$special");
+    }
+
+    /**
+     * Adds a mapping from a mode to another handler.
+     *
+     * @param string $mode        Mode to be remapped.
+     * @param string $handler     New target handler.
+     */
+    public function mapHandler($mode, $handler)
+    {
+        $this->mode_handlers[$mode] = $handler;
+    }
+
+    /**
+     * Splits the page text into tokens.
+     *
+     * Will fail if the handlers report an error or if no content is consumed. If successful then each
+     * unparsed and parsed token invokes a call to the held listener.
+     *
+     * @param string $raw        Raw HTML text.
+     * @return boolean           True on success, else false.
+     */
+    public function parse($raw)
+    {
+        if (! isset($this->handler)) {
+            return false;
+        }
+        $initialLength = strlen($raw);
+        $length = $initialLength;
+        $pos = 0;
+        while (is_array($parsed = $this->reduce($raw))) {
+            list($unmatched, $matched, $mode) = $parsed;
+            $currentLength = strlen($raw);
+            $matchPos = $initialLength - $currentLength - strlen($matched);
+            if (! $this->dispatchTokens($unmatched, $matched, $mode, $pos, $matchPos)) {
+                return false;
+            }
+            if ($currentLength == $length) {
+                return false;
+            }
+            $length = $currentLength;
+            $pos = $initialLength - $currentLength;
+        }
+        if (!$parsed) {
+            return false;
+        }
+        return $this->invokeHandler($raw, DOKU_LEXER_UNMATCHED, $pos);
+    }
+
+    /**
+     * Gives plugins access to the mode stack
+     *
+     * @return StateStack
+     */
+    public function getModeStack()
+    {
+        return $this->modeStack;
+    }
+
+    /**
+     * Sends the matched token and any leading unmatched
+     * text to the parser changing the lexer to a new
+     * mode if one is listed.
+     *
+     * @param string $unmatched Unmatched leading portion.
+     * @param string $matched Actual token match.
+     * @param bool|string $mode Mode after match. A boolean false mode causes no change.
+     * @param int $initialPos
+     * @param int $matchPos Current byte index location in raw doc thats being parsed
+     * @return boolean             False if there was any error from the parser.
+     */
+    protected function dispatchTokens($unmatched, $matched, $mode, $initialPos, $matchPos)
+    {
+        if (! $this->invokeHandler($unmatched, DOKU_LEXER_UNMATCHED, $initialPos)) {
+            return false;
+        }
+        if ($this->isModeEnd($mode)) {
+            if (! $this->invokeHandler($matched, DOKU_LEXER_EXIT, $matchPos)) {
+                return false;
+            }
+            return $this->modeStack->leave();
+        }
+        if ($this->isSpecialMode($mode)) {
+            $this->modeStack->enter($this->decodeSpecial($mode));
+            if (! $this->invokeHandler($matched, DOKU_LEXER_SPECIAL, $matchPos)) {
+                return false;
+            }
+            return $this->modeStack->leave();
+        }
+        if (is_string($mode)) {
+            $this->modeStack->enter($mode);
+            return $this->invokeHandler($matched, DOKU_LEXER_ENTER, $matchPos);
+        }
+        return $this->invokeHandler($matched, DOKU_LEXER_MATCHED, $matchPos);
+    }
+
+    /**
+     * Tests to see if the new mode is actually to leave the current mode and pop an item from the matching
+     * mode stack.
+     *
+     * @param string $mode    Mode to test.
+     * @return boolean        True if this is the exit mode.
+     */
+    protected function isModeEnd($mode)
+    {
+        return ($mode === "__exit");
+    }
+
+    /**
+     * Test to see if the mode is one where this mode is entered for this token only and automatically
+     * leaves immediately afterwoods.
+     *
+     * @param string $mode    Mode to test.
+     * @return boolean        True if this is the exit mode.
+     */
+    protected function isSpecialMode($mode)
+    {
+        return (strncmp($mode, "_", 1) == 0);
+    }
+
+    /**
+     * Strips the magic underscore marking single token modes.
+     *
+     * @param string $mode    Mode to decode.
+     * @return string         Underlying mode name.
+     */
+    protected function decodeSpecial($mode)
+    {
+        return substr($mode, 1);
+    }
+
+    /**
+     * Calls the parser method named after the current mode.
+     *
+     * Empty content will be ignored. The lexer has a parser handler for each mode in the lexer.
+     *
+     * @param string $content Text parsed.
+     * @param boolean $is_match Token is recognised rather
+     *                               than unparsed data.
+     * @param int $pos Current byte index location in raw doc
+     *                             thats being parsed
+     * @return bool
+     */
+    protected function invokeHandler($content, $is_match, $pos)
+    {
+        if (($content === "") || ($content === false)) {
+            return true;
+        }
+        $handler = $this->modeStack->getCurrent();
+        if (isset($this->mode_handlers[$handler])) {
+            $handler = $this->mode_handlers[$handler];
+        }
+
+        // modes starting with plugin_ are all handled by the same
+        // handler but with an additional parameter
+        if (substr($handler, 0, 7)=='plugin_') {
+            list($handler,$plugin) = explode('_', $handler, 2);
+            return $this->handler->$handler($content, $is_match, $pos, $plugin);
+        }
+
+        return $this->handler->$handler($content, $is_match, $pos);
+    }
+
+    /**
+     * Tries to match a chunk of text and if successful removes the recognised chunk and any leading
+     * unparsed data. Empty strings will not be matched.
+     *
+     * @param string $raw         The subject to parse. This is the content that will be eaten.
+     * @return array|bool         Three item list of unparsed content followed by the
+     *                            recognised token and finally the action the parser is to take.
+     *                            True if no match, false if there is a parsing error.
+     */
+    protected function reduce(&$raw)
+    {
+        if (! isset($this->regexes[$this->modeStack->getCurrent()])) {
+            return false;
+        }
+        if ($raw === "") {
+            return true;
+        }
+        if ($action = $this->regexes[$this->modeStack->getCurrent()]->split($raw, $split)) {
+            list($unparsed, $match, $raw) = $split;
+            return array($unparsed, $match, $action);
+        }
+        return true;
+    }
+
+    /**
+     * Escapes regex characters other than (, ) and /
+     *
+     * @param string $str
+     * @return string
+     */
+    public static function escape($str)
+    {
+        $chars = array(
+            '/\\\\/',
+            '/\./',
+            '/\+/',
+            '/\*/',
+            '/\?/',
+            '/\[/',
+            '/\^/',
+            '/\]/',
+            '/\$/',
+            '/\{/',
+            '/\}/',
+            '/\=/',
+            '/\!/',
+            '/\</',
+            '/\>/',
+            '/\|/',
+            '/\:/'
+        );
+
+        $escaped = array(
+            '\\\\\\\\',
+            '\.',
+            '\+',
+            '\*',
+            '\?',
+            '\[',
+            '\^',
+            '\]',
+            '\$',
+            '\{',
+            '\}',
+            '\=',
+            '\!',
+            '\<',
+            '\>',
+            '\|',
+            '\:'
+        );
+        return preg_replace($chars, $escaped, $str);
+    }
+}

content/inc/Parsing/Lexer/ParallelRegex.php

@@ -0,0 +1,203 @@
+<?php
+/**
+ * Lexer adapted from Simple Test: http://sourceforge.net/projects/simpletest/
+ * For an intro to the Lexer see:
+ * https://web.archive.org/web/20120125041816/http://www.phppatterns.com/docs/develop/simple_test_lexer_notes
+ *
+ * @author Marcus Baker http://www.lastcraft.com
+ */
+
+namespace dokuwiki\Parsing\Lexer;
+
+/**
+ * Compounded regular expression.
+ *
+ * Any of the contained patterns could match and when one does it's label is returned.
+ */
+class ParallelRegex
+{
+    /** @var string[] patterns to match */
+    protected $patterns;
+    /** @var string[] labels for above patterns */
+    protected $labels;
+    /** @var string the compound regex matching all patterns */
+    protected $regex;
+    /** @var bool case sensitive matching? */
+    protected $case;
+
+    /**
+     * Constructor. Starts with no patterns.
+     *
+     * @param boolean $case    True for case sensitive, false
+     *                         for insensitive.
+     */
+    public function __construct($case)
+    {
+        $this->case = $case;
+        $this->patterns = array();
+        $this->labels = array();
+        $this->regex = null;
+    }
+
+    /**
+     * Adds a pattern with an optional label.
+     *
+     * @param mixed       $pattern Perl style regex. Must be UTF-8
+     *                             encoded. If its a string, the (, )
+     *                             lose their meaning unless they
+     *                             form part of a lookahead or
+     *                             lookbehind assertation.
+     * @param bool|string $label   Label of regex to be returned
+     *                             on a match. Label must be ASCII
+     */
+    public function addPattern($pattern, $label = true)
+    {
+        $count = count($this->patterns);
+        $this->patterns[$count] = $pattern;
+        $this->labels[$count] = $label;
+        $this->regex = null;
+    }
+
+    /**
+     * Attempts to match all patterns at once against a string.
+     *
+     * @param string $subject      String to match against.
+     * @param string $match        First matched portion of
+     *                             subject.
+     * @return bool|string         False if no match found, label if label exists, true if not
+     */
+    public function match($subject, &$match)
+    {
+        if (count($this->patterns) == 0) {
+            return false;
+        }
+        if (! preg_match($this->getCompoundedRegex(), $subject, $matches)) {
+            $match = "";
+            return false;
+        }
+
+        $match = $matches[0];
+        $size = count($matches);
+        // FIXME this could be made faster by storing the labels as keys in a hashmap
+        for ($i = 1; $i < $size; $i++) {
+            if ($matches[$i] && isset($this->labels[$i - 1])) {
+                return $this->labels[$i - 1];
+            }
+        }
+        return true;
+    }
+
+    /**
+     * Attempts to split the string against all patterns at once
+     *
+     * @param string $subject      String to match against.
+     * @param array $split         The split result: array containing, pre-match, match & post-match strings
+     * @return boolean             True on success.
+     *
+     * @author Christopher Smith <chris@jalakai.co.uk>
+     */
+    public function split($subject, &$split)
+    {
+        if (count($this->patterns) == 0) {
+            return false;
+        }
+
+        if (! preg_match($this->getCompoundedRegex(), $subject, $matches)) {
+            if (function_exists('preg_last_error')) {
+                $err = preg_last_error();
+                switch ($err) {
+                    case PREG_BACKTRACK_LIMIT_ERROR:
+                        msg('A PCRE backtrack error occured. Try to increase the pcre.backtrack_limit in php.ini', -1);
+                        break;
+                    case PREG_RECURSION_LIMIT_ERROR:
+                        msg('A PCRE recursion error occured. Try to increase the pcre.recursion_limit in php.ini', -1);
+                        break;
+                    case PREG_BAD_UTF8_ERROR:
+                        msg('A PCRE UTF-8 error occured. This might be caused by a faulty plugin', -1);
+                        break;
+                    case PREG_INTERNAL_ERROR:
+                        msg('A PCRE internal error occured. This might be caused by a faulty plugin', -1);
+                        break;
+                }
+            }
+
+            $split = array($subject, "", "");
+            return false;
+        }
+
+        $idx = count($matches)-2;
+        list($pre, $post) = preg_split($this->patterns[$idx].$this->getPerlMatchingFlags(), $subject, 2);
+        $split = array($pre, $matches[0], $post);
+
+        return isset($this->labels[$idx]) ? $this->labels[$idx] : true;
+    }
+
+    /**
+     * Compounds the patterns into a single
+     * regular expression separated with the
+     * "or" operator. Caches the regex.
+     * Will automatically escape (, ) and / tokens.
+     *
+     * @return null|string
+     */
+    protected function getCompoundedRegex()
+    {
+        if ($this->regex == null) {
+            $cnt = count($this->patterns);
+            for ($i = 0; $i < $cnt; $i++) {
+                /*
+                 * decompose the input pattern into "(", "(?", ")",
+                 * "[...]", "[]..]", "[^]..]", "[...[:...:]..]", "\x"...
+                 * elements.
+                 */
+                preg_match_all('/\\\\.|' .
+                               '\(\?|' .
+                               '[()]|' .
+                               '\[\^?\]?(?:\\\\.|\[:[^]]*:\]|[^]\\\\])*\]|' .
+                               '[^[()\\\\]+/', $this->patterns[$i], $elts);
+
+                $pattern = "";
+                $level = 0;
+
+                foreach ($elts[0] as $elt) {
+                    /*
+                     * for "(", ")" remember the nesting level, add "\"
+                     * only to the non-"(?" ones.
+                     */
+
+                    switch ($elt) {
+                        case '(':
+                            $pattern .= '\(';
+                            break;
+                        case ')':
+                            if ($level > 0)
+                                $level--; /* closing (? */
+                            else $pattern .= '\\';
+                            $pattern .= ')';
+                            break;
+                        case '(?':
+                            $level++;
+                            $pattern .= '(?';
+                            break;
+                        default:
+                            if (substr($elt, 0, 1) == '\\')
+                                $pattern .= $elt;
+                            else $pattern .= str_replace('/', '\/', $elt);
+                    }
+                }
+                $this->patterns[$i] = "($pattern)";
+            }
+            $this->regex = "/" . implode("|", $this->patterns) . "/" . $this->getPerlMatchingFlags();
+        }
+        return $this->regex;
+    }
+
+    /**
+     * Accessor for perl regex mode flags to use.
+     * @return string       Perl regex flags.
+     */
+    protected function getPerlMatchingFlags()
+    {
+        return ($this->case ? "msS" : "msSi");
+    }
+}

content/inc/Parsing/Lexer/StateStack.php

@@ -0,0 +1,60 @@
+<?php
+/**
+ * Lexer adapted from Simple Test: http://sourceforge.net/projects/simpletest/
+ * For an intro to the Lexer see:
+ * https://web.archive.org/web/20120125041816/http://www.phppatterns.com/docs/develop/simple_test_lexer_notes
+ *
+ * @author Marcus Baker http://www.lastcraft.com
+ */
+
+namespace dokuwiki\Parsing\Lexer;
+
+/**
+ * States for a stack machine.
+ */
+class StateStack
+{
+    protected $stack;
+
+    /**
+     * Constructor. Starts in named state.
+     * @param string $start        Starting state name.
+     */
+    public function __construct($start)
+    {
+        $this->stack = array($start);
+    }
+
+    /**
+     * Accessor for current state.
+     * @return string       State.
+     */
+    public function getCurrent()
+    {
+        return $this->stack[count($this->stack) - 1];
+    }
+
+    /**
+     * Adds a state to the stack and sets it to be the current state.
+     *
+     * @param string $state        New state.
+     */
+    public function enter($state)
+    {
+        array_push($this->stack, $state);
+    }
+
+    /**
+     * Leaves the current state and reverts
+     * to the previous one.
+     * @return boolean    false if we attempt to drop off the bottom of the list.
+     */
+    public function leave()
+    {
+        if (count($this->stack) == 1) {
+            return false;
+        }
+        array_pop($this->stack);
+        return true;
+    }
+}

content/inc/Parsing/Parser.php

@@ -0,0 +1,128 @@
+<?php
+
+namespace dokuwiki\Parsing;
+
+use Doku_Handler;
+use dokuwiki\Parsing\Lexer\Lexer;
+use dokuwiki\Parsing\ParserMode\Base;
+use dokuwiki\Parsing\ParserMode\ModeInterface;
+
+/**
+ * Sets up the Lexer with modes and points it to the Handler
+ * For an intro to the Lexer see: wiki:parser
+ */
+class Parser {
+
+    /** @var Doku_Handler */
+    protected $handler;
+
+    /** @var Lexer $lexer */
+    protected $lexer;
+
+    /** @var ModeInterface[] $modes */
+    protected $modes = array();
+
+    /** @var bool mode connections may only be set up once */
+    protected $connected = false;
+
+    /**
+     * dokuwiki\Parsing\Doku_Parser constructor.
+     *
+     * @param Doku_Handler $handler
+     */
+    public function __construct(Doku_Handler $handler) {
+        $this->handler = $handler;
+    }
+
+    /**
+     * Adds the base mode and initialized the lexer
+     *
+     * @param Base $BaseMode
+     */
+    protected function addBaseMode($BaseMode) {
+        $this->modes['base'] = $BaseMode;
+        if(!$this->lexer) {
+            $this->lexer = new Lexer($this->handler, 'base', true);
+        }
+        $this->modes['base']->Lexer = $this->lexer;
+    }
+
+    /**
+     * Add a new syntax element (mode) to the parser
+     *
+     * PHP preserves order of associative elements
+     * Mode sequence is important
+     *
+     * @param string $name
+     * @param ModeInterface $Mode
+     */
+    public function addMode($name, ModeInterface $Mode) {
+        if(!isset($this->modes['base'])) {
+            $this->addBaseMode(new Base());
+        }
+        $Mode->Lexer = $this->lexer; // FIXME should be done by setter
+        $this->modes[$name] = $Mode;
+    }
+
+    /**
+     * Connect all modes with each other
+     *
+     * This is the last step before actually parsing.
+     */
+    protected function connectModes() {
+
+        if($this->connected) {
+            return;
+        }
+
+        foreach(array_keys($this->modes) as $mode) {
+            // Base isn't connected to anything
+            if($mode == 'base') {
+                continue;
+            }
+            $this->modes[$mode]->preConnect();
+
+            foreach(array_keys($this->modes) as $cm) {
+
+                if($this->modes[$cm]->accepts($mode)) {
+                    $this->modes[$mode]->connectTo($cm);
+                }
+
+            }
+
+            $this->modes[$mode]->postConnect();
+        }
+
+        $this->connected = true;
+    }
+
+    /**
+     * Parses wiki syntax to instructions
+     *
+     * @param string $doc the wiki syntax text
+     * @return array instructions
+     */
+    public function parse($doc) {
+        $this->connectModes();
+        // Normalize CRs and pad doc
+        $doc = "\n" . str_replace("\r\n", "\n", $doc) . "\n";
+        $this->lexer->parse($doc);
+
+        if (!method_exists($this->handler, 'finalize')) {
+            /** @deprecated 2019-10 we have a legacy handler from a plugin, assume legacy _finalize exists */
+
+            \dokuwiki\Debug\DebugHelper::dbgCustomDeprecationEvent(
+                'finalize()',
+                get_class($this->handler) . '::_finalize()',
+                __METHOD__,
+                __FILE__,
+                __LINE__
+            );
+            $this->handler->_finalize();
+        } else {
+            $this->handler->finalize();
+        }
+        return $this->handler->calls;
+    }
+
+}

content/inc/Parsing/ParserMode/AbstractMode.php

@@ -0,0 +1,40 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+/**
+ * This class and all the subclasses below are used to reduce the effort required to register
+ * modes with the Lexer.
+ *
+ * @author Harry Fuecks <hfuecks@gmail.com>
+ */
+abstract class AbstractMode implements ModeInterface
+{
+    /** @var \dokuwiki\Parsing\Lexer\Lexer $Lexer will be injected on loading FIXME this should be done by setter */
+    public $Lexer;
+    protected $allowedModes = array();
+
+    /** @inheritdoc */
+    abstract public function getSort();
+
+    /** @inheritdoc */
+    public function preConnect()
+    {
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+    }
+
+    /** @inheritdoc */
+    public function accepts($mode)
+    {
+        return in_array($mode, (array) $this->allowedModes);
+    }
+}

content/inc/Parsing/ParserMode/Acronym.php

@@ -0,0 +1,68 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Acronym extends AbstractMode
+{
+    // A list
+    protected $acronyms = array();
+    protected $pattern = '';
+
+    /**
+     * Acronym constructor.
+     *
+     * @param string[] $acronyms
+     */
+    public function __construct($acronyms)
+    {
+        usort($acronyms, array($this,'compare'));
+        $this->acronyms = $acronyms;
+    }
+
+    /** @inheritdoc */
+    public function preConnect()
+    {
+        if (!count($this->acronyms)) return;
+
+        $bound = '[\x00-\x2f\x3a-\x40\x5b-\x60\x7b-\x7f]';
+        $acronyms = array_map(['\\dokuwiki\\Parsing\\Lexer\\Lexer', 'escape'], $this->acronyms);
+        $this->pattern = '(?<=^|'.$bound.')(?:'.join('|', $acronyms).')(?='.$bound.')';
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        if (!count($this->acronyms)) return;
+
+        if (strlen($this->pattern) > 0) {
+            $this->Lexer->addSpecialPattern($this->pattern, $mode, 'acronym');
+        }
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 240;
+    }
+
+    /**
+     * sort callback to order by string length descending
+     *
+     * @param string $a
+     * @param string $b
+     *
+     * @return int
+     */
+    protected function compare($a, $b)
+    {
+        $a_len = strlen($a);
+        $b_len = strlen($b);
+        if ($a_len > $b_len) {
+            return -1;
+        } elseif ($a_len < $b_len) {
+            return 1;
+        }
+
+        return 0;
+    }
+}

content/inc/Parsing/ParserMode/Base.php

@@ -0,0 +1,31 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Base extends AbstractMode
+{
+
+    /**
+     * Base constructor.
+     */
+    public function __construct()
+    {
+        global $PARSER_MODES;
+
+        $this->allowedModes = array_merge(
+            $PARSER_MODES['container'],
+            $PARSER_MODES['baseonly'],
+            $PARSER_MODES['paragraphs'],
+            $PARSER_MODES['formatting'],
+            $PARSER_MODES['substition'],
+            $PARSER_MODES['protected'],
+            $PARSER_MODES['disabled']
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 0;
+    }
+}

content/inc/Parsing/ParserMode/Camelcaselink.php

@@ -0,0 +1,23 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Camelcaselink extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addSpecialPattern(
+            '\b[A-Z]+[a-z]+[A-Z][A-Za-z]*\b',
+            $mode,
+            'camelcaselink'
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 290;
+    }
+}

content/inc/Parsing/ParserMode/Code.php

@@ -0,0 +1,25 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Code extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern('<code\b(?=.*</code>)', $mode, 'code');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addExitPattern('</code>', 'code');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 200;
+    }
+}

content/inc/Parsing/ParserMode/Emaillink.php

@@ -0,0 +1,20 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Emaillink extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        // pattern below is defined in inc/mail.php
+        $this->Lexer->addSpecialPattern('<'.PREG_PATTERN_VALID_EMAIL.'>', $mode, 'emaillink');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 340;
+    }
+}

content/inc/Parsing/ParserMode/Entity.php

@@ -0,0 +1,50 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+use dokuwiki\Parsing\Lexer\Lexer;
+
+class Entity extends AbstractMode
+{
+
+    protected $entities = array();
+    protected $pattern = '';
+
+    /**
+     * Entity constructor.
+     * @param string[] $entities
+     */
+    public function __construct($entities)
+    {
+        $this->entities = $entities;
+    }
+
+
+    /** @inheritdoc */
+    public function preConnect()
+    {
+        if (!count($this->entities) || $this->pattern != '') return;
+
+        $sep = '';
+        foreach ($this->entities as $entity) {
+            $this->pattern .= $sep. Lexer::escape($entity);
+            $sep = '|';
+        }
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        if (!count($this->entities)) return;
+
+        if (strlen($this->pattern) > 0) {
+            $this->Lexer->addSpecialPattern($this->pattern, $mode, 'entity');
+        }
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 260;
+    }
+}

content/inc/Parsing/ParserMode/Eol.php

@@ -0,0 +1,25 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Eol extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $badModes = array('listblock','table');
+        if (in_array($mode, $badModes)) {
+            return;
+        }
+        // see FS#1652, pattern extended to swallow preceding whitespace to avoid
+        // issues with lines that only contain whitespace
+        $this->Lexer->addSpecialPattern('(?:^[ \t]*)?\n', $mode, 'eol');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 370;
+    }
+}

content/inc/Parsing/ParserMode/Externallink.php

@@ -0,0 +1,44 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Externallink extends AbstractMode
+{
+    protected $schemes = array();
+    protected $patterns = array();
+
+    /** @inheritdoc */
+    public function preConnect()
+    {
+        if (count($this->patterns)) return;
+
+        $ltrs = '\w';
+        $gunk = '/\#~:.?+=&%@!\-\[\]';
+        $punc = '.:?\-;,';
+        $host = $ltrs.$punc;
+        $any  = $ltrs.$gunk.$punc;
+
+        $this->schemes = getSchemes();
+        foreach ($this->schemes as $scheme) {
+            $this->patterns[] = '\b(?i)'.$scheme.'(?-i)://['.$any.']+?(?=['.$punc.']*[^'.$any.'])';
+        }
+
+        $this->patterns[] = '(?<=\s)(?i)www?(?-i)\.['.$host.']+?\.['.$host.']+?['.$any.']+?(?=['.$punc.']*[^'.$any.'])';
+        $this->patterns[] = '(?<=\s)(?i)ftp?(?-i)\.['.$host.']+?\.['.$host.']+?['.$any.']+?(?=['.$punc.']*[^'.$any.'])';
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+
+        foreach ($this->patterns as $pattern) {
+            $this->Lexer->addSpecialPattern($pattern, $mode, 'externallink');
+        }
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 330;
+    }
+}

content/inc/Parsing/ParserMode/File.php

@@ -0,0 +1,25 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class File extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern('<file\b(?=.*</file>)', $mode, 'file');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addExitPattern('</file>', 'file');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 210;
+    }
+}

content/inc/Parsing/ParserMode/Filelink.php

@@ -0,0 +1,39 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Filelink extends AbstractMode
+{
+
+    protected $pattern;
+
+    /** @inheritdoc */
+    public function preConnect()
+    {
+
+        $ltrs = '\w';
+        $gunk = '/\#~:.?+=&%@!\-';
+        $punc = '.:?\-;,';
+        $host = $ltrs.$punc;
+        $any  = $ltrs.$gunk.$punc;
+
+        $this->pattern = '\b(?i)file(?-i)://['.$any.']+?['.
+            $punc.']*[^'.$any.']';
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addSpecialPattern(
+            $this->pattern,
+            $mode,
+            'filelink'
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 360;
+    }
+}

content/inc/Parsing/ParserMode/Footnote.php

@@ -0,0 +1,50 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Footnote extends AbstractMode
+{
+
+    /**
+     * Footnote constructor.
+     */
+    public function __construct()
+    {
+        global $PARSER_MODES;
+
+        $this->allowedModes = array_merge(
+            $PARSER_MODES['container'],
+            $PARSER_MODES['formatting'],
+            $PARSER_MODES['substition'],
+            $PARSER_MODES['protected'],
+            $PARSER_MODES['disabled']
+        );
+
+        unset($this->allowedModes[array_search('footnote', $this->allowedModes)]);
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern(
+            '\x28\x28(?=.*\x29\x29)',
+            $mode,
+            'footnote'
+        );
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addExitPattern(
+            '\x29\x29',
+            'footnote'
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 150;
+    }
+}

content/inc/Parsing/ParserMode/Formatting.php

@@ -0,0 +1,115 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+/**
+ * This class sets the markup for bold (=strong),
+ * italic (=emphasis), underline etc.
+ */
+class Formatting extends AbstractMode
+{
+    protected $type;
+
+    protected $formatting = array(
+        'strong' => array(
+            'entry' => '\*\*(?=.*\*\*)',
+            'exit' => '\*\*',
+            'sort' => 70
+        ),
+
+        'emphasis' => array(
+            'entry' => '//(?=[^\x00]*[^:])', //hack for bugs #384 #763 #1468
+            'exit' => '//',
+            'sort' => 80
+        ),
+
+        'underline' => array(
+            'entry' => '__(?=.*__)',
+            'exit' => '__',
+            'sort' => 90
+        ),
+
+        'monospace' => array(
+            'entry' => '\x27\x27(?=.*\x27\x27)',
+            'exit' => '\x27\x27',
+            'sort' => 100
+        ),
+
+        'subscript' => array(
+            'entry' => '<sub>(?=.*</sub>)',
+            'exit' => '</sub>',
+            'sort' => 110
+        ),
+
+        'superscript' => array(
+            'entry' => '<sup>(?=.*</sup>)',
+            'exit' => '</sup>',
+            'sort' => 120
+        ),
+
+        'deleted' => array(
+            'entry' => '<del>(?=.*</del>)',
+            'exit' => '</del>',
+            'sort' => 130
+        ),
+    );
+
+    /**
+     * @param string $type
+     */
+    public function __construct($type)
+    {
+        global $PARSER_MODES;
+
+        if (!array_key_exists($type, $this->formatting)) {
+            trigger_error('Invalid formatting type ' . $type, E_USER_WARNING);
+        }
+
+        $this->type = $type;
+
+        // formatting may contain other formatting but not it self
+        $modes = $PARSER_MODES['formatting'];
+        $key = array_search($type, $modes);
+        if (is_int($key)) {
+            unset($modes[$key]);
+        }
+
+        $this->allowedModes = array_merge(
+            $modes,
+            $PARSER_MODES['substition'],
+            $PARSER_MODES['disabled']
+        );
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+
+        // Can't nest formatting in itself
+        if ($mode == $this->type) {
+            return;
+        }
+
+        $this->Lexer->addEntryPattern(
+            $this->formatting[$this->type]['entry'],
+            $mode,
+            $this->type
+        );
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+
+        $this->Lexer->addExitPattern(
+            $this->formatting[$this->type]['exit'],
+            $this->type
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return $this->formatting[$this->type]['sort'];
+    }
+}

content/inc/Parsing/ParserMode/Header.php

@@ -0,0 +1,24 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Header extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        //we're not picky about the closing ones, two are enough
+        $this->Lexer->addSpecialPattern(
+            '[ \t]*={2,}[^\n]+={2,}[ \t]*(?=\n)',
+            $mode,
+            'header'
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 50;
+    }
+}

content/inc/Parsing/ParserMode/Hr.php

@@ -0,0 +1,19 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Hr extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addSpecialPattern('\n[ \t]*-{4,}[ \t]*(?=\n)', $mode, 'hr');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 160;
+    }
+}

content/inc/Parsing/ParserMode/Html.php

@@ -0,0 +1,27 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Html extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern('<html>(?=.*</html>)', $mode, 'html');
+        $this->Lexer->addEntryPattern('<HTML>(?=.*</HTML>)', $mode, 'htmlblock');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addExitPattern('</html>', 'html');
+        $this->Lexer->addExitPattern('</HTML>', 'htmlblock');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 190;
+    }
+}

content/inc/Parsing/ParserMode/Internallink.php

@@ -0,0 +1,20 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Internallink extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        // Word boundaries?
+        $this->Lexer->addSpecialPattern("\[\[.*?\]\](?!\])", $mode, 'internallink');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 300;
+    }
+}

content/inc/Parsing/ParserMode/Linebreak.php

@@ -0,0 +1,19 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Linebreak extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addSpecialPattern('\x5C{2}(?:[ \t]|(?=\n))', $mode, 'linebreak');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 140;
+    }
+}

content/inc/Parsing/ParserMode/Listblock.php

@@ -0,0 +1,44 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Listblock extends AbstractMode
+{
+
+    /**
+     * Listblock constructor.
+     */
+    public function __construct()
+    {
+        global $PARSER_MODES;
+
+        $this->allowedModes = array_merge(
+            $PARSER_MODES['formatting'],
+            $PARSER_MODES['substition'],
+            $PARSER_MODES['disabled'],
+            $PARSER_MODES['protected']
+        );
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern('[ \t]*\n {2,}[\-\*]', $mode, 'listblock');
+        $this->Lexer->addEntryPattern('[ \t]*\n\t{1,}[\-\*]', $mode, 'listblock');
+
+        $this->Lexer->addPattern('\n {2,}[\-\*]', 'listblock');
+        $this->Lexer->addPattern('\n\t{1,}[\-\*]', 'listblock');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addExitPattern('\n', 'listblock');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 10;
+    }
+}

content/inc/Parsing/ParserMode/Media.php

@@ -0,0 +1,20 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Media extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        // Word boundaries?
+        $this->Lexer->addSpecialPattern("\{\{(?:[^\}]|(?:\}[^\}]))+\}\}", $mode, 'media');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 320;
+    }
+}

content/inc/Parsing/ParserMode/ModeInterface.php

@@ -0,0 +1,46 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+/**
+ * Defines a mode (syntax component) in the Parser
+ */
+interface ModeInterface
+{
+    /**
+     * returns a number used to determine in which order modes are added
+     *
+     * @return int;
+     */
+    public function getSort();
+
+    /**
+     * Called before any calls to connectTo
+     *
+     * @return void
+     */
+    public function preConnect();
+
+    /**
+     * Connects the mode
+     *
+     * @param string $mode
+     * @return void
+     */
+    public function connectTo($mode);
+
+    /**
+     * Called after all calls to connectTo
+     *
+     * @return void
+     */
+    public function postConnect();
+
+    /**
+     * Check if given mode is accepted inside this mode
+     *
+     * @param string $mode
+     * @return bool
+     */
+    public function accepts($mode);
+}

content/inc/Parsing/ParserMode/Multiplyentity.php

@@ -0,0 +1,27 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+/**
+ * Implements the 640x480 replacement
+ */
+class Multiplyentity extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+
+        $this->Lexer->addSpecialPattern(
+            '(?<=\b)(?:[1-9]|\d{2,})[xX]\d+(?=\b)',
+            $mode,
+            'multiplyentity'
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 270;
+    }
+}

content/inc/Parsing/ParserMode/Nocache.php

@@ -0,0 +1,19 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Nocache extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addSpecialPattern('~~NOCACHE~~', $mode, 'nocache');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 40;
+    }
+}

content/inc/Parsing/ParserMode/Notoc.php

@@ -0,0 +1,19 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Notoc extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addSpecialPattern('~~NOTOC~~', $mode, 'notoc');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 30;
+    }
+}

content/inc/Parsing/ParserMode/Php.php

@@ -0,0 +1,27 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Php extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern('<php>(?=.*</php>)', $mode, 'php');
+        $this->Lexer->addEntryPattern('<PHP>(?=.*</PHP>)', $mode, 'phpblock');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addExitPattern('</php>', 'php');
+        $this->Lexer->addExitPattern('</PHP>', 'phpblock');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 180;
+    }
+}

content/inc/Parsing/ParserMode/Plugin.php

@@ -0,0 +1,8 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+/**
+ * @fixme do we need this anymore or could the syntax plugin inherit directly from abstract mode?
+ */
+abstract class Plugin extends AbstractMode {}

content/inc/Parsing/ParserMode/Preformatted.php

@@ -0,0 +1,31 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Preformatted extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        // Has hard coded awareness of lists...
+        $this->Lexer->addEntryPattern('\n  (?![\*\-])', $mode, 'preformatted');
+        $this->Lexer->addEntryPattern('\n\t(?![\*\-])', $mode, 'preformatted');
+
+        // How to effect a sub pattern with the Lexer!
+        $this->Lexer->addPattern('\n  ', 'preformatted');
+        $this->Lexer->addPattern('\n\t', 'preformatted');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addExitPattern('\n', 'preformatted');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 20;
+    }
+}

content/inc/Parsing/ParserMode/Quote.php

@@ -0,0 +1,41 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Quote extends AbstractMode
+{
+
+    /**
+     * Quote constructor.
+     */
+    public function __construct()
+    {
+        global $PARSER_MODES;
+
+        $this->allowedModes = array_merge(
+            $PARSER_MODES['formatting'],
+            $PARSER_MODES['substition'],
+            $PARSER_MODES['disabled'],
+            $PARSER_MODES['protected']
+        );
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern('\n>{1,}', $mode, 'quote');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addPattern('\n>{1,}', 'quote');
+        $this->Lexer->addExitPattern('\n', 'quote');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 220;
+    }
+}

content/inc/Parsing/ParserMode/Quotes.php

@@ -0,0 +1,51 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Quotes extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        global $conf;
+
+        $ws   =  '\s/\#~:+=&%@\-\x28\x29\]\[{}><"\'';   // whitespace
+        $punc =  ';,\.?!';
+
+        if ($conf['typography'] == 2) {
+            $this->Lexer->addSpecialPattern(
+                "(?<=^|[$ws])'(?=[^$ws$punc])",
+                $mode,
+                'singlequoteopening'
+            );
+            $this->Lexer->addSpecialPattern(
+                "(?<=^|[^$ws]|[$punc])'(?=$|[$ws$punc])",
+                $mode,
+                'singlequoteclosing'
+            );
+            $this->Lexer->addSpecialPattern(
+                "(?<=^|[^$ws$punc])'(?=$|[^$ws$punc])",
+                $mode,
+                'apostrophe'
+            );
+        }
+
+        $this->Lexer->addSpecialPattern(
+            "(?<=^|[$ws])\"(?=[^$ws$punc])",
+            $mode,
+            'doublequoteopening'
+        );
+        $this->Lexer->addSpecialPattern(
+            "\"",
+            $mode,
+            'doublequoteclosing'
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 280;
+    }
+}

content/inc/Parsing/ParserMode/Rss.php

@@ -0,0 +1,19 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Rss extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addSpecialPattern("\{\{rss>[^\}]+\}\}", $mode, 'rss');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 310;
+    }
+}

content/inc/Parsing/ParserMode/Smiley.php

@@ -0,0 +1,48 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+use dokuwiki\Parsing\Lexer\Lexer;
+
+class Smiley extends AbstractMode
+{
+    protected $smileys = array();
+    protected $pattern = '';
+
+    /**
+     * Smiley constructor.
+     * @param string[] $smileys
+     */
+    public function __construct($smileys)
+    {
+        $this->smileys = $smileys;
+    }
+
+    /** @inheritdoc */
+    public function preConnect()
+    {
+        if (!count($this->smileys) || $this->pattern != '') return;
+
+        $sep = '';
+        foreach ($this->smileys as $smiley) {
+            $this->pattern .= $sep.'(?<=\W|^)'. Lexer::escape($smiley).'(?=\W|$)';
+            $sep = '|';
+        }
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        if (!count($this->smileys)) return;
+
+        if (strlen($this->pattern) > 0) {
+            $this->Lexer->addSpecialPattern($this->pattern, $mode, 'smiley');
+        }
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 230;
+    }
+}

content/inc/Parsing/ParserMode/Table.php

@@ -0,0 +1,47 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Table extends AbstractMode
+{
+
+    /**
+     * Table constructor.
+     */
+    public function __construct()
+    {
+        global $PARSER_MODES;
+
+        $this->allowedModes = array_merge(
+            $PARSER_MODES['formatting'],
+            $PARSER_MODES['substition'],
+            $PARSER_MODES['disabled'],
+            $PARSER_MODES['protected']
+        );
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern('[\t ]*\n\^', $mode, 'table');
+        $this->Lexer->addEntryPattern('[\t ]*\n\|', $mode, 'table');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addPattern('\n\^', 'table');
+        $this->Lexer->addPattern('\n\|', 'table');
+        $this->Lexer->addPattern('[\t ]*:::[\t ]*(?=[\|\^])', 'table');
+        $this->Lexer->addPattern('[\t ]+', 'table');
+        $this->Lexer->addPattern('\^', 'table');
+        $this->Lexer->addPattern('\|', 'table');
+        $this->Lexer->addExitPattern('\n', 'table');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 60;
+    }
+}

content/inc/Parsing/ParserMode/Unformatted.php

@@ -0,0 +1,28 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Unformatted extends AbstractMode
+{
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addEntryPattern('<nowiki>(?=.*</nowiki>)', $mode, 'unformatted');
+        $this->Lexer->addEntryPattern('%%(?=.*%%)', $mode, 'unformattedalt');
+    }
+
+    /** @inheritdoc */
+    public function postConnect()
+    {
+        $this->Lexer->addExitPattern('</nowiki>', 'unformatted');
+        $this->Lexer->addExitPattern('%%', 'unformattedalt');
+        $this->Lexer->mapHandler('unformattedalt', 'unformatted');
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 170;
+    }
+}

content/inc/Parsing/ParserMode/Windowssharelink.php

@@ -0,0 +1,31 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+class Windowssharelink extends AbstractMode
+{
+
+    protected $pattern;
+
+    /** @inheritdoc */
+    public function preConnect()
+    {
+        $this->pattern = "\\\\\\\\\w+?(?:\\\\[\w\-$]+)+";
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        $this->Lexer->addSpecialPattern(
+            $this->pattern,
+            $mode,
+            'windowssharelink'
+        );
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 350;
+    }
+}

content/inc/Parsing/ParserMode/Wordblock.php

@@ -0,0 +1,52 @@
+<?php
+
+namespace dokuwiki\Parsing\ParserMode;
+
+use dokuwiki\Parsing\Lexer\Lexer;
+
+/**
+ * @fixme is this actually used?
+ */
+class Wordblock extends AbstractMode
+{
+    protected $badwords = array();
+    protected $pattern = '';
+
+    /**
+     * Wordblock constructor.
+     * @param $badwords
+     */
+    public function __construct($badwords)
+    {
+        $this->badwords = $badwords;
+    }
+
+    /** @inheritdoc */
+    public function preConnect()
+    {
+
+        if (count($this->badwords) == 0 || $this->pattern != '') {
+            return;
+        }
+
+        $sep = '';
+        foreach ($this->badwords as $badword) {
+            $this->pattern .= $sep.'(?<=\b)(?i)'. Lexer::escape($badword).'(?-i)(?=\b)';
+            $sep = '|';
+        }
+    }
+
+    /** @inheritdoc */
+    public function connectTo($mode)
+    {
+        if (strlen($this->pattern) > 0) {
+            $this->Lexer->addSpecialPattern($this->pattern, $mode, 'wordblock');
+        }
+    }
+
+    /** @inheritdoc */
+    public function getSort()
+    {
+        return 250;
+    }
+}