public static function convertDataToAttributes($data)
{
// get ID of current CK instance
$instance = RTE::getInstanceId();
// properly encode JSON
$encoded = RTEReverseParser::encodeRTEData($data);
$encoded = Sanitizer::encodeAttribute($encoded);
return " data-rte-meta=\"{$encoded}\" data-rte-instance=\"{$instance}\" ";
}
/**
* Make lists from lines starting with ':', '*', '#', etc. (DBL)
*
* @param $text String
* @param $linestart Boolean: whether or not this is at the start of a line.
* @private
* @return string the lists rendered as HTML
*/
function doBlockLevels($text, $linestart)
{
wfProfileIn(__METHOD__);
# Parsing through the text line by line. The main thing
# happening here is handling of block-level elements p, pre,
# and making lists from lines starting with * # : etc.
#
$textLines = StringUtils::explode("\n", $text);
$lastPrefix = $output = '';
$this->mDTopen = $inBlockElem = false;
$prefixLength = 0;
$paragraphStack = false;
$comment = $prefixType = false;
# wikia change
foreach ($textLines as $oLine) {
# Fix up $linestart
if (!$linestart) {
$output .= $oLine;
$linestart = true;
continue;
}
# * = ul
# # = ol
# ; = dt
# : = dd
# RTE (Rich Text Editor) - begin
# @author: Macbre
$this->doBlockLevelsLineStart($oLine, $output);
# RTE - end
$lastPrefixLength = strlen($lastPrefix);
$preCloseMatch = preg_match('/<\\/pre/i', $oLine);
$preOpenMatch = preg_match('/<pre/i', $oLine);
# If not in a <pre> element, scan for and figure out what prefixes are there.
if (!$this->mInPre) {
# Multiple prefixes may abut each other for nested lists.
$prefixLength = strspn($oLine, '*#:;');
$prefix = substr($oLine, 0, $prefixLength);
# eh?
# ; and : are both from definition-lists, so they're equivalent
# for the purposes of determining whether or not we need to open/close
# elements.
$prefix2 = str_replace(';', ':', $prefix);
$t = substr($oLine, $prefixLength);
$this->mInPre = (bool) $preOpenMatch;
} else {
# Don't interpret any other prefixes in preformatted text
$prefixLength = 0;
$prefix = $prefix2 = '';
$t = $oLine;
}
# RTE (Rich Text Editor) - begin
# @author: Macbre
$this->mCurrentPrefix = $prefix;
# RTE - end
# List generation
if ($prefixLength && $lastPrefix === $prefix2) {
# Same as the last item, so no need to deal with nesting or opening stuff
$output .= $this->nextItem(substr($prefix, -1));
$paragraphStack = false;
if (substr($prefix, -1) === ';') {
# The one nasty exception: definition lists work like this:
# ; title : definition text
# So we check for : in the remainder text to split up the
# title and definition, without b0rking links.
$term = $t2 = '';
if ($this->findColonNoLinks($t, $term, $t2) !== false) {
$t = $t2;
$output .= $term . $this->nextItem(':');
}
}
} elseif ($prefixLength || $lastPrefixLength) {
# We need to open or close prefixes, or both.
# Either open or close a level...
$commonPrefixLength = $this->getCommon($prefix, $lastPrefix);
$paragraphStack = false;
# Close all the prefixes which aren't shared.
while ($commonPrefixLength < $lastPrefixLength) {
$output .= $this->closeList($lastPrefix[$lastPrefixLength - 1]);
--$lastPrefixLength;
}
# Continue the current prefix if appropriate.
if ($prefixLength <= $commonPrefixLength && $commonPrefixLength > 0) {
$output .= $this->nextItem($prefix[$commonPrefixLength - 1]);
}
# Open prefixes where appropriate.
while ($prefixLength > $commonPrefixLength) {
$char = substr($prefix, $commonPrefixLength, 1);
/* Wikia change begin - @author: Macbre */
$this->mLastCommonPrefix = $prefixLength == $commonPrefixLength + 1;
/* Wikia change end */
$output .= $this->openList($char);
if (';' === $char) {
# @todo FIXME: This is dupe of code above
if ($this->findColonNoLinks($t, $term, $t2) !== false) {
$t = $t2;
$output .= $term . $this->nextItem(':');
}
}
++$commonPrefixLength;
}
$lastPrefix = $prefix2;
}
# If we have no prefixes, go to paragraph mode.
if (0 == $prefixLength) {
wfProfileIn(__METHOD__ . "-paragraph");
# No prefix (not in list)--go to paragraph mode
# XXX: use a stack for nestable elements like span, table and div
$openmatch = preg_match('/(?:<aside|<table|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<li|<\\/tr|<\\/td|<\\/th)/iS', $t);
/**
* Wikia change start
* @author Federico "Lox" Lucignano
*
* Stop the parser from wrapping figure tags in paragraphs
*/
$closematch = preg_match('/(?:<\\/?aside|<\\/table|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|' . '<td|<th|<\\/?div|<\\/?figure|<hr|<\\/pre|<\\/p|' . $this->mUniqPrefix . '-pre|' . $this->mUniqPrefix . '-bloglist|' . $this->mUniqPrefix . '-infobox|<\\/li|<\\/ul|<\\/ol|<\\/?center)/iS', $t);
/**
* Wikia change end
*/
# RTE (Rich Text Editor) - begin
# @author: Macbre
$this->doOpenCloseMatch($t, $openmatch, $closematch);
# RTE - end
if ($openmatch or $closematch) {
$paragraphStack = false;
# TODO bug 5718: paragraph closed
$output .= $this->closeParagraph();
if ($preOpenMatch and !$preCloseMatch) {
$this->mInPre = true;
}
$inBlockElem = !$closematch;
} elseif (!$inBlockElem && !$this->mInPre) {
if (' ' == substr($t, 0, 1) and $this->mLastSection === 'pre' || trim($t) != '') {
# pre
if ($this->mLastSection !== 'pre') {
$paragraphStack = false;
$output .= $this->closeParagraph() . '<pre>';
$this->mLastSection = 'pre';
}
$t = substr($t, 1);
} else {
# paragraph
if (trim($t) === '') {
if ($paragraphStack) {
$output .= $paragraphStack . '<br />';
$paragraphStack = false;
$this->mLastSection = 'p';
} else {
if ($this->mLastSection !== 'p') {
$output .= $this->closeParagraph();
$this->mLastSection = '';
$paragraphStack = '<p>';
} else {
$paragraphStack = '</p><p>';
}
}
} else {
if ($paragraphStack) {
$output .= $paragraphStack;
$paragraphStack = false;
$this->mLastSection = 'p';
} elseif ($this->mLastSection !== 'p') {
$output .= $this->closeParagraph() . '<p>';
$this->mLastSection = 'p';
} else {
# RTE (Rich Text Editor) - begin
# @author: Macbre
global $wgRTEParserEnabled;
if (!empty($wgRTEParserEnabled)) {
// count trailing spaces
$text = ltrim(strrev($output), "\n");
$spaces = strspn($text, ' ');
$output .= RTEReverseParser::buildComment('LINE_BREAK', array('spaces' => $spaces));
}
# RTE - end
}
}
}
}
wfProfileOut(__METHOD__ . "-paragraph");
}
# somewhere above we forget to get out of pre block (bug 785)
if ($preCloseMatch && $this->mInPre) {
$this->mInPre = false;
}
if ($paragraphStack === false) {
$output .= $t . "\n";
}
# RTE (Rich Text Editor) - begin
# @author: Macbre
$this->doBlockLevelsLineEnd($oLine, $output);
# RTE - end
}
while ($prefixLength) {
$output .= $this->closeList($prefix2[$prefixLength - 1]);
--$prefixLength;
}
if ($this->mLastSection != '') {
$output .= '</' . $this->mLastSection . '>';
$this->mLastSection = '';
}
wfProfileOut(__METHOD__);
return $output;
}
/**
* Convert wikitext to HTML and add extra HTML attributes for RTE
*
* @param $text String: text we want to parse
* @param $title A title object
* @param $options ParserOptions
* @param $linestart boolean
* @param $clearState boolean
* @param $revid Int: number to pass in {{REVISIONID}}
* @return ParserOutput a ParserOutput
*/
public function parse($text, Title $title, ParserOptions $options, $linestart = true, $clearState = true, $revid = null)
{
wfProfileIn(__METHOD__);
// get rid of all \r in wikitext
$text = str_replace("\r\n", "\n", $text);
// count newlines at the beginning of the wikitext
$emptyLinesAtStart = strspn($text, "\n");
if ($emptyLinesAtStart > 0) {
// don't double count empty lines at the beginning of wikitext
$this->emptyLinesBefore = $emptyLinesAtStart * -1;
}
// mark HTML entities from wikitext (& : _)
$text = self::markEntities($text);
// set flag indicating parsing for CK
global $wgRTEParserEnabled;
$wgRTEParserEnabled = true;
// don't use XML cache in preprocessor
global $wgPreprocessorCacheThreshold;
$wgPreprocessorCacheThreshold = 1000000;
//RTE::log(__METHOD__ . '::beforeParse', $text);
// parse to HTML
$output = parent::parse($text, $title, $options, $linestart, $clearState, $revid);
$wgRTEParserEnabled = false;
// add extra RTE attributes to HTML elements (for correct handling of spaces and newlines when parsing back to wikitext)
$html = $output->getText();
// add RTE_EMPTY_LINES_BEFORE comment
if ($emptyLinesAtStart > 0) {
$html = RTEReverseParser::addEmptyLinesBeforeComment($emptyLinesAtStart) . $html;
}
//RTE::log(__METHOD__ . '::beforeReplace', $html);
// wrap HTML entities inside span "placeholders" (& : _)
$html = self::wrapEntities($html);
wfProfileIn(__METHOD__ . '::regexp');
// remove EMPTY_LINES_BEFORE comments which are before closing tags - refs RT#38889
// <!-- RTE_EMPTY_LINES_BEFORE_1 --></td></tr></table> <= remove this one
// <!-- RTE_EMPTY_LINES_BEFORE_1 --><p>
$html = preg_replace('%<!-- RTE_EMPTY_LINES_BEFORE_(\\d+) -->(</[^>]+></)%s', '\\2', $html);
// move empty lines counter data from comment to next opening tag attribute (thx to Marooned)
$html = preg_replace('%<!-- RTE_EMPTY_LINES_BEFORE_(\\d+) -->(?!<!)(.*?)(<[^/][^>]*)>%s', '\\2\\3 data-rte-empty-lines-before="\\1">', $html);
// remove not replaced EMPTY_LINES_BEFORE comments
// <!-- RTE_EMPTY_LINES_BEFORE_1 -- data-rte-empty-lines-before="1">
$html = preg_replace('%<!-- RTE_EMPTY_LINES_BEFORE_(\\d+) [^>]+>%s', '', $html);
// add data-rte-spaces-before for list items and table cells
$html = preg_replace_callback("/<(li|dd|dt|td|th)([^>]*)>( +)/", 'RTEParser::spacesBeforeCallback', $html);
// replace placeholder markers with placeholders
$html = preg_replace_callback("/-01-(\\d{4})/", 'RTE::replacePlaceholder', $html);
// replace dataidx attribute with data-rte-meta attribute storing JSON encoded meta data
$html = preg_replace_callback('/ _rte_dataidx="(\\d{4})" /', 'RTEData::replaceIdxByData', $html);
$html = preg_replace("/-(?:" . RTEMarker::INTERNAL_WIKITEXT . "|" . RTEMarker::EXTERNAL_WIKITEXT . ")-\\d{4}/", '', $html);
// RT#40786: add empty paragraphs between headings (</h3>\n<h3 ...)
$html = preg_replace("%(</h\\d>\\s)(<h\\d)%s", '$1<p data-rte-filler="true"></p>$2', $html);
wfProfileOut(__METHOD__ . '::regexp');
// add extra attribute for p tags coming from parser
$html = strtr($html, array('<p>' => '<p data-rte-fromparser="true">', '<p ' => '<p data-rte-fromparser="true" '));
// add empty paragraph for new / empty pages
if ($html == '') {
$html = Xml::element('p');
}
// update parser output
RTE::log(__METHOD__, $html);
$output->setText($html);
wfProfileOut(__METHOD__);
return $output;
}
function parseInternal()
{
$parser = new RTEReverseParser();
return $parser->parse($this->mText);
}
/**
* Parse given HTML from CK back to wikitext
*/
public static function HtmlToWikitext($html)
{
wfProfileIn(__METHOD__);
$RTEReverseParser = new RTEReverseParser();
$wikitext = $RTEReverseParser->parse($html);
wfProfileOut(__METHOD__);
return $wikitext;
}
