/**
* @param string $text
* @return string
*/
public function postprocess($text)
{
// Revert <html-{link,meta}> back to <{link,meta}>
$text = preg_replace('!<html-(link|meta)([^>]*?)(/{0,1}>)!', '<$1$2$3', $text);
// Restore the contents of placeholder tokens
$text = $this->mTokens->replace($text);
return $text;
}
/**
* libxml in its usual pointlessness converts many chars to entities - this function
* perfoms a reverse conversion
* @param string $html
* @return string
*/
private function fixLibXML( $html ) {
wfProfileIn( __METHOD__ );
static $replacements;
if ( ! $replacements ) {
// We don't include rules like '"' => '&quot;' because entities had already been
// normalized by libxml. Using this function with input not sanitized by libxml is UNSAFE!
$replacements = new ReplacementArray( array(
'"' => '&quot;',
'&' => '&amp;',
'<' => '&lt;',
'>' => '&gt;',
) );
}
$html = $replacements->replace( $html );
$html = mb_convert_encoding( $html, 'UTF-8', 'HTML-ENTITIES' );
wfProfileOut( __METHOD__ );
return $html;
}
/**
* @param $text string
* @return string
*/
public function postprocess($text)
{
return $this->mTokens->replace($text);
}
static function onAddHTML(&$out, &$text)
{
if (!isset($out->mLqtReplacements) || !count($out->mLqtReplacements)) {
return true;
}
$replacements = $out->mLqtReplacements;
$replacer = new ReplacementArray($replacements);
$text = $replacer->replace($text);
return true;
}
protected function ccnorm($s)
{
static $equivset = null;
static $replacementArray = null;
if (is_null($equivset) || is_null($replacementArray)) {
global $IP;
require_once "{$IP}/extensions/AntiSpoof/equivset.php";
$replacementArray = new ReplacementArray($equivset);
}
return $replacementArray->replace($s);
}
/**
* Strips and renders nowiki, pre, math, hiero
* If $render is set, performs necessary rendering operations on plugins
* Returns the text, and fills an array with data needed in unstrip()
*
* @param StripState $state
*
* @param bool $stripcomments when set, HTML comments <!-- like this -->
* will be stripped in addition to other tags. This is important
* for section editing, where these comments cause confusion when
* counting the sections in the wikisource
*
* @param array dontstrip contains tags which should not be stripped;
* used to prevent stipping of <gallery> when saving (fixes bug 2700)
*
* @private
*/
function strip($text, $state, $stripcomments = false, $dontstrip = array())
{
global $wgContLang, $wgUseTeX, $wgScriptPath, $wgVersion, $wgHooks, $wgExtensionFunctions;
wfProfileIn(__METHOD__);
$render = $this->mOutputType == OT_HTML;
$uniq_prefix = $this->mUniqPrefix;
$commentState = new ReplacementArray();
$nowikiItems = array();
$generalItems = array();
$elements = array_merge(array('nowiki', 'gallery', 'math'), array_keys($this->mTagHooks));
if (isset($wgHooks['ParserFirstCallInit']) && in_array('efSyntaxHighlight_GeSHiSetup', $wgHooks['ParserFirstCallInit']) || isset($wgExtensionFunctions) && in_array('efSyntaxHighlight_GeSHiSetup', $wgExtensionFunctions)) {
$elements = array_merge($elements, array('source'));
}
if (isset($wgHooks['ParserFirstCallInit']) && in_array('wfCite', $wgHooks['ParserFirstCallInit']) || isset($wgExtensionFunctions) && in_array('wfCite', $wgExtensionFunctions)) {
$elements = array_merge($elements, array('ref', 'references'));
}
global $wgRawHtml;
if ($wgRawHtml) {
$elements[] = 'html';
}
# Removing $dontstrip tags from $elements list (currently only 'gallery', fixing bug 2700)
foreach ($elements as $k => $v) {
if (!in_array($v, $dontstrip)) {
continue;
}
unset($elements[$k]);
}
$elements = array_unique($elements);
$matches = array();
if (version_compare("1.12", $wgVersion, ">")) {
$text = Parser::extractTagsAndParams($elements, $text, $matches, $uniq_prefix);
} else {
$text = self::extractTagsAndParams($elements, $text, $matches, $uniq_prefix);
}
foreach ($matches as $marker => $data) {
list($element, $content, $params, $tag) = $data;
if ($render) {
$tagName = strtolower($element);
wfProfileIn(__METHOD__ . "-render-{$tagName}");
switch ($tagName) {
case '!--':
// Comment
if (substr($tag, -3) == '-->') {
$output = $tag;
} else {
// Unclosed comment in input.
// Close it so later stripping can remove it
$output = "{$tag}-->";
}
break;
case 'references':
$output = $this->fck_wikiTag('references', $content, $params);
break;
case 'ref':
$output = $this->fck_wikiTag('ref', $content, $params);
break;
case 'source':
$output = $this->fck_wikiTag('source', $content, $params);
break;
case 'html':
if ($wgRawHtml) {
$output = $this->fck_wikiTag('html', $content, $params);
}
break;
case 'nowiki':
$output = $this->fck_wikiTag('nowiki', $content, $params);
// required by FCKeditor
break;
case 'math':
if ($wgUseTeX) {
//normal render
$output = $wgContLang->armourMath(MathRenderer::renderMath($content));
} else {
// show fakeimage
$output = '<img _fckfakelement="true" class="FCK__MWMath" _fck_mw_math="' . $content . '" src="' . $wgScriptPath . '/skins/common/images/button_math.png" />';
}
break;
case 'gallery':
$output = $this->fck_wikiTag('gallery', $content, $params);
// required by FCKeditor
//$output = $this->renderImageGallery( $content, $params );
break;
default:
if (isset($this->mTagHooks[$tagName])) {
$this->fck_mw_taghook = $tagName;
// required by FCKeditor
$output = call_user_func_array($this->mTagHooks[$tagName], array($content, $params, $this));
} else {
throw new MWException("Invalid call hook {$element}");
}
}
wfProfileOut(__METHOD__ . "-render-{$tagName}");
} else {
// Just stripping tags; keep the source
$output = $tag;
}
// Unstrip the output, to support recursive strip() calls
$output = $state->unstripBoth($output);
if (!$stripcomments && $element == '!--') {
$commentState->setPair($marker, $output);
} elseif ($element == 'html' || $element == 'nowiki') {
# $nowikiItems[$marker] = $output;
$state->addNoWiki($marker, $output);
} else {
# $generalItems[$marker] = $output;
$state->addGeneral($marker, $output);
}
}
# Unstrip comments unless explicitly told otherwise.
# (The comments are always stripped prior to this point, so as to
# not invoke any extension tags / parser hooks contained within
# a comment.)
if (!$stripcomments) {
// Put them all back and forget them
$text = $commentState->replace($text);
}
$this->fck_matches = $matches;
wfProfileOut(__METHOD__);
return $text;
}
/**
* Strips and renders nowiki, pre, math, hiero
* If $render is set, performs necessary rendering operations on plugins
* Returns the text, and fills an array with data needed in unstrip()
*
* @param StripState $state
*
* @param bool $stripcomments when set, HTML comments <!-- like this -->
* will be stripped in addition to other tags. This is important
* for section editing, where these comments cause confusion when
* counting the sections in the wikisource
*
* @param array dontstrip contains tags which should not be stripped;
* used to prevent stipping of <gallery> when saving (fixes bug 2700)
*
* @private
*/
function strip($text, $state, $stripcomments = false, $dontstrip = array())
{
global $wgContLang;
wfProfileIn(__METHOD__);
$render = $this->mOutputType == self::OT_HTML;
$uniq_prefix = $this->mUniqPrefix;
$commentState = new ReplacementArray();
$nowikiItems = array();
$generalItems = array();
$elements = array_merge(array('nowiki', 'gallery'), array_keys($this->mTagHooks));
global $wgRawHtml;
if ($wgRawHtml) {
$elements[] = 'html';
}
if ($this->mOptions->getUseTeX()) {
$elements[] = 'math';
}
# Removing $dontstrip tags from $elements list (currently only 'gallery', fixing bug 2700)
foreach ($elements as $k => $v) {
if (!in_array($v, $dontstrip)) {
continue;
}
unset($elements[$k]);
}
$matches = array();
$text = self::extractTagsAndParams($elements, $text, $matches, $uniq_prefix);
foreach ($matches as $marker => $data) {
list($element, $content, $params, $tag) = $data;
if ($render) {
$tagName = strtolower($element);
wfProfileIn(__METHOD__ . "-render-{$tagName}");
switch ($tagName) {
case '!--':
// Comment
if (substr($tag, -3) == '-->') {
$output = $tag;
} else {
// Unclosed comment in input.
// Close it so later stripping can remove it
$output = "{$tag}-->";
}
break;
case 'html':
if ($wgRawHtml) {
$output = $content;
break;
}
// Shouldn't happen otherwise. :)
// Shouldn't happen otherwise. :)
case 'nowiki':
$output = Xml::escapeTagsOnly($content);
break;
case 'math':
$output = $wgContLang->armourMath(MathRenderer::renderMath($content, $params));
break;
case 'gallery':
$output = $this->renderImageGallery($content, $params);
break;
default:
if (isset($this->mTagHooks[$tagName])) {
$output = call_user_func_array($this->mTagHooks[$tagName], array($content, $params, $this));
} else {
throw new MWException("Invalid call hook {$element}");
}
}
wfProfileOut(__METHOD__ . "-render-{$tagName}");
} else {
// Just stripping tags; keep the source
$output = $tag;
}
// Unstrip the output, to support recursive strip() calls
$output = $state->unstripBoth($output);
if (!$stripcomments && $element == '!--') {
$commentState->setPair($marker, $output);
} elseif ($element == 'html' || $element == 'nowiki') {
$nowikiItems[$marker] = $output;
} else {
$generalItems[$marker] = $output;
}
}
# Add the new items to the state
# We do this after the loop instead of during it to avoid slowing
# down the recursive unstrip
$state->nowiki->mergeArray($nowikiItems);
$state->general->mergeArray($generalItems);
# Unstrip comments unless explicitly told otherwise.
# (The comments are always stripped prior to this point, so as to
# not invoke any extension tags / parser hooks contained within
# a comment.)
if (!$stripcomments) {
// Put them all back and forget them
$text = $commentState->replace($text);
}
wfProfileOut(__METHOD__);
return $text;
}
/**
* libxml in its usual pointlessness converts many chars to entities - this function
* perfoms a reverse conversion
* @param string $html
* @return string
*/
private function fixLibXML($html)
{
static $replacements;
if (!$replacements) {
// We don't include rules like '"' => '&quot;' because entities had already been
// normalized by libxml. Using this function with input not sanitized by libxml is UNSAFE!
$replacements = new ReplacementArray(array('"' => '&quot;', '&' => '&amp;', '<' => '&lt;', '>' => '&gt;'));
}
$html = $replacements->replace($html);
if (function_exists('mb_convert_encoding')) {
// Just in case the conversion in getDoc() above used named
// entities that aren't known to html_entity_decode().
$html = mb_convert_encoding($html, 'UTF-8', 'HTML-ENTITIES');
} else {
$html = html_entity_decode($html, ENT_COMPAT, 'utf-8');
}
return $html;
}
/**
* @param $s
* @return mixed
*/
protected function ccnorm($s)
{
static $replacementArray = null;
if (is_null($replacementArray)) {
global $IP;
if (is_readable("{$IP}/extensions/AntiSpoof/equivset.php")) {
// Satisfy analyzer.
$equivset = null;
// Contains a map of characters in $equivset.
require "{$IP}/extensions/AntiSpoof/equivset.php";
$replacementArray = new ReplacementArray($equivset);
} else {
// AntiSpoof isn't available, so just create a dummy
wfDebugLog('AbuseFilter', "Can't compute normalized string (ccnorm) as the AntiSpoof Extension isn't isntalled.");
$replacementArray = new ReplacementArray(array());
}
}
return $replacementArray->replace($s);
}
