<?php
-include_once('Tokenizer.php');
+require_once('Tokenizer.php');
if( $GLOBALS['wgUseWikiHiero'] ){
- include_once('wikihiero.php');
+ require_once('extensions/wikihiero/wikihiero.php');
+}
+if( $GLOBALS['wgUseTimeline'] ){
+ require_once('extensions/timeline/Timeline.php');
}
# PHP Parser
define( "OT_WIKI", 2 );
define( "OT_MSG", 3 );
+# string parameter for extractTags which will cause it
+# to strip HTML comments in addition to regular
+# <XML>-style tags. This should not be anything we
+# may want to use in wikisyntax
+define( "STRIP_COMMENTS", "HTMLCommentStrip" );
+
# prefix for escaping, used in two functions at least
define( "UNIQ_PREFIX", "NaodW29");
#
function parse( $text, &$title, $options, $linestart = true, $clearState = true )
{
+ global $wgUseTidy;
$fname = "Parser::parse";
wfProfileIn( $fname );
$text = $this->internalParse( $text, $linestart );
$text = $this->unstrip( $text, $this->mStripState );
# Clean up special characters, only run once, next-to-last before doBlockLevels
- $fixtags = array(
- "/<hr *>/i" => '<hr/>',
- "/<br *>/i" => '<br/>',
- "/<center *>/i"=>'<div class="center">',
- "/<\\/center *>/i" => '</div>',
- # Clean up spare ampersands; note that we probably ought to be
- # more careful about named entities.
- '/&(?!:amp;|#[Xx][0-9A-fa-f]+;|#[0-9]+;|[a-zA-Z0-9]+;)/' => '&'
- );
- $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
-
+ if(!$wgUseTidy) {
+ $fixtags = array(
+ "/<hr *>/i" => '<hr/>',
+ "/<br *>/i" => '<br/>',
+ "/<center *>/i"=>'<div class="center">',
+ "/<\\/center *>/i" => '</div>',
+ # Clean up spare ampersands; note that we probably ought to be
+ # more careful about named entities.
+ '/&(?!:amp;|#[Xx][0-9A-fa-f]+;|#[0-9]+;|[a-zA-Z0-9]+;)/' => '&'
+ );
+ $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
+ } else {
+ $fixtags = array(
+ "/<center *>/i"=>'<div class="center">',
+ "/<\\/center *>/i" => '</div>'
+ );
+ $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
+ }
# only once and last
$text = $this->doBlockLevels( $text, $linestart );
-
+ if($wgUseTidy) {
+ $text = $this->tidy($text);
+ }
$this->mOutput->setText( $text );
wfProfileOut( $fname );
return $this->mOutput;
# If $content is already set, the additional entries will be appended
+ # If $tag is set to STRIP_COMMENTS, the function will extract
+ # <!-- HTML comments -->
+
/* static */ function extractTags($tag, $text, &$content, $uniq_prefix = ""){
$rnd = $uniq_prefix . '-' . $tag . Parser::getRandomString();
if ( !$content ) {
$stripped = "";
while ( "" != $text ) {
- $p = preg_split( "/<\\s*$tag\\s*>/i", $text, 2 );
+ if($tag==STRIP_COMMENTS) {
+ $p = preg_split( "/<!--/i", $text, 2 );
+ } else {
+ $p = preg_split( "/<\\s*$tag\\s*>/i", $text, 2 );
+ }
$stripped .= $p[0];
if ( ( count( $p ) < 2 ) || ( "" == $p[1] ) ) {
$text = "";
} else {
- $q = preg_split( "/<\\/\\s*$tag\\s*>/i", $p[1], 2 );
+ if($tag==STRIP_COMMENTS) {
+ $q = preg_split( "/-->/i", $p[1], 2 );
+ } else {
+ $q = preg_split( "/<\\/\\s*$tag\\s*>/i", $p[1], 2 );
+ }
$marker = $rnd . sprintf("%08X", $n++);
$content[$marker] = $q[0];
$stripped .= $marker;
return $stripped;
}
- # Strips <nowiki>, <pre> and <math>
+ # 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()
# If the $state is already a valid strip state, it adds to the state
- #
- function strip( $text, &$state )
+
+ # When $stripcomments is 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
+ function strip( $text, &$state, $stripcomments = false )
{
$render = ($this->mOutputType == OT_HTML);
$nowiki_content = array();
$hiero_content = array();
$math_content = array();
$pre_content = array();
- $item_content = array();
+ $comment_content = array();
# Replace any instances of the placeholders
$uniq_prefix = UNIQ_PREFIX;
}
}
- if( $GLOBALS['wgUseWikiHiero'] ){
- $text = Parser::extractTags("hiero", $text, $hiero_content, $uniq_prefix);
- foreach( $hiero_content as $marker => $content ){
- if( $render ){
- $hiero_content[$marker] = WikiHiero( $content, WH_MODE_HTML);
- } else {
- $hiero_content[$marker] = "<hiero>$content</hiero>";
- }
+ $text = Parser::extractTags("hiero", $text, $hiero_content, $uniq_prefix);
+ foreach( $hiero_content as $marker => $content ){
+ if( $render && $GLOBALS['wgUseWikiHiero']){
+ $hiero_content[$marker] = WikiHiero( $content, WH_MODE_HTML);
+ } else {
+ $hiero_content[$marker] = "<hiero>$content</hiero>";
}
}
- if( $this->mOptions->getUseTeX() ){
- $text = Parser::extractTags("math", $text, $math_content, $uniq_prefix);
- foreach( $math_content as $marker => $content ){
- if( $render ){
+ $text = Parser::extractTags("math", $text, $math_content, $uniq_prefix);
+ foreach( $math_content as $marker => $content ){
+ if( $render ) {
+ if( $this->mOptions->getUseTeX() ) {
$math_content[$marker] = renderMath( $content );
} else {
- $math_content[$marker] = "<math>$content</math>";
+ $math_content[$marker] = "<math>$content<math>";
}
+ } else {
+ $math_content[$marker] = "<math>$content</math>";
}
}
$pre_content[$marker] = "<pre>$content</pre>";
}
}
+ if($stripcomments) {
+ $text = Parser::extractTags(STRIP_COMMENTS, $text, $comment_content, $uniq_prefix);
+ foreach( $comment_content as $marker => $content ){
+ $comment_content[$marker] = "<!--$content-->";
+ }
+ }
# Merge state with the pre-existing state, if there is one
if ( $state ) {
$state['hiero'] = $state['hiero'] + $hiero_content;
$state['math'] = $state['math'] + $math_content;
$state['pre'] = $state['pre'] + $pre_content;
+ $state['comment'] = $state['comment'] + $comment_content;
} else {
$state = array(
'nowiki' => $nowiki_content,
'hiero' => $hiero_content,
'math' => $math_content,
'pre' => $pre_content,
- 'item' => $item_content
+ 'comment' => $comment_content
);
}
return $text;
'nowiki' => array(),
'hiero' => array(),
'math' => array(),
- 'pre' => array(),
- 'item' => array()
+ 'pre' => array()
);
}
$state['item'][$rnd] = $text;
return $rnd;
}
+ # This method generates the list of subcategories and pages for a category
function categoryMagic ()
{
global $wgLang , $wgUser ;
- if ( !$this->mOptions->getUseCategoryMagic() ) return ;
- $id = $this->mTitle->getArticleID() ;
+ if ( !$this->mOptions->getUseCategoryMagic() ) return ; # Doesn't use categories at all
+
$cns = Namespace::getCategory() ;
- if ( $this->mTitle->getNamespace() != $cns ) return "" ;
- $ti = $this->mTitle->getText() ;
+ if ( $this->mTitle->getNamespace() != $cns ) return "" ; # This ain't a category page
+
$r = "<br style=\"clear:both;\"/>\n";
- $articles = array() ;
- $parents = array () ;
- $children = array() ;
$sk =& $wgUser->getSkin() ;
+ $articles = array() ;
+ $children = array() ;
$data = array () ;
- $sql1 = "SELECT DISTINCT cur_title,cur_namespace FROM cur,links WHERE l_to={$id} AND l_from=cur_id";
- $sql2 = "SELECT DISTINCT cur_title,cur_namespace FROM cur,brokenlinks WHERE bl_to={$id} AND bl_from=cur_id" ;
-
- $res = wfQuery ( $sql1, DB_READ ) ;
- while ( $x = wfFetchObject ( $res ) ) $data[] = $x ;
+ $id = $this->mTitle->getArticleID() ;
- $res = wfQuery ( $sql2, DB_READ ) ;
- while ( $x = wfFetchObject ( $res ) ) $data[] = $x ;
+ # For existing categories
+ if( $id ) {
+ $sql = "SELECT DISTINCT cur_title,cur_namespace FROM cur,links WHERE l_to={$id} AND l_from=cur_id";
+ $res = wfQuery ( $sql, DB_READ ) ;
+ while ( $x = wfFetchObject ( $res ) ) $data[] = $x ;
+ } else {
+ # For non-existing categories
+ $t = wfStrencode( $this->mTitle->getPrefixedDBKey() );
+ $sql = "SELECT DISTINCT cur_title,cur_namespace FROM cur,brokenlinks WHERE bl_to='$t' AND bl_from=cur_id" ;
+ $res = wfQuery ( $sql, DB_READ ) ;
+ while ( $x = wfFetchObject ( $res ) ) $data[] = $x ;
+ }
+ # For all pages that link to this category
foreach ( $data AS $x )
{
$t = $wgLang->getNsText ( $x->cur_namespace ) ;
$t .= $x->cur_title ;
if ( $x->cur_namespace == $cns ) {
- array_push ( $children , $sk->makeLink ( $t ) ) ;
+ array_push ( $children , $sk->makeLink ( $t ) ) ; # Subcategory
} else {
- array_push ( $articles , $sk->makeLink ( $t ) ) ;
+ array_push ( $articles , $sk->makeLink ( $t ) ) ; # Page in this category
}
}
wfFreeResult ( $res ) ;
- # Children
+ # Showing subcategories
if ( count ( $children ) > 0 )
{
asort ( $children ) ;
$r .= implode ( ", " , $children ) ;
}
- # Articles
+ # Showing pages in this category
if ( count ( $articles ) > 0 )
{
+ $ti = $this->mTitle->getText() ;
asort ( $articles ) ;
$h = wfMsg( "category_header", $ti );
$r .= "<h2>{$h}</h2>\n" ;
return trim ( $t ) ;
}
+ /* interface with html tidy, used if $wgUseTidy = true */
+ function tidy ( $text ) {
+ global $wgTidyConf, $wgTidyBin, $wgTidyOpts;
+ global $wgInputEncoding, $wgOutputEncoding;
+ $cleansource = '';
+ switch(strtoupper($wgOutputEncoding)) {
+ case 'ISO-8859-1':
+ $wgTidyOpts .= ($wgInputEncoding == $wgOutputEncoding)? ' -latin1':' -raw';
+ break;
+ case 'UTF-8':
+ $wgTidyOpts .= ($wgInputEncoding == $wgOutputEncoding)? ' -utf8':' -raw';
+ break;
+ default:
+ $wgTidyOpts .= ' -raw';
+ }
+
+ $text = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'.
+' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>'.
+'<head><title>test</title></head><body>'.$text.'</body></html>';
+ $descriptorspec = array(
+ 0 => array("pipe", "r"),
+ 1 => array("pipe", "w"),
+ 2 => array("file", "/dev/null", "a")
+ );
+ $process = proc_open("$wgTidyBin -config $wgTidyConf $wgTidyOpts", $descriptorspec, $pipes);
+ if (is_resource($process)) {
+ fwrite($pipes[0], $text);
+ fclose($pipes[0]);
+ while (!feof($pipes[1])) {
+ $cleansource .= fgets($pipes[1], 1024);
+ }
+ fclose($pipes[1]);
+ $return_value = proc_close($process);
+ }
+ if( $cleansource == '' && $text != '') {
+ return '<h2>'.wfMsg('seriousxhtmlerrors').'</h2><pre>'.htmlspecialchars($text).'</pre>';
+ } else {
+ return $cleansource;
+ }
+ }
+
function doTableStuff ( $t )
{
$t = explode ( "\n" , $t ) ;
return $t ;
}
- function internalParse( $text, $linestart, $args = array() )
+ // set isMain=false if you call from a template etc. and don't want to do stuff
+ // like TOC insertion for that content
+ function internalParse( $text, $linestart, $args = array(), $isMain=true )
{
$fname = "Parser::internalParse";
wfProfileIn( $fname );
$text = $this->replaceExternalLinks( $text );
$text = $this->doTokenizedParser ( $text );
$text = $this->doTableStuff ( $text ) ;
- $text = $this->formatHeadings( $text );
+ $text = $this->formatHeadings( $text, $isMain );
$sk =& $this->mOptions->getSkin();
$text = $sk->transformContent( $text );
/* private */ function doTokenizedParser( $str )
{
global $wgLang; # for language specific parser hook
+ global $wgUploadDirectory, $wgUseTimeline;
$tokenizer=Tokenizer::newFromString( $str );
$tokenStack = array();
# simple text with no further markup
$txt = $token["text"];
break;
+ case "blank":
+ # Text that contains blanks that have to be converted to
+ # non-breakable spaces for French.
+ # U+202F NARROW NO-BREAK SPACE might be a better choice, but
+ # browser support for Unicode spacing is poor.
+ $txt = str_replace( " ", " ", $token["text"] );
+ break;
case "[[[":
# remember the tag opened with 3 [
$threeopen = true;
$txt = $this->doMagicISBN( $tokenizer );
}
break;
+ case "<timeline>":
+ if ( $wgUseTimeline &&
+ "" != ( $timelinesrc = $tokenizer->readAllUntil("</timeline>") ) )
+ {
+ $txt = renderTimeline( $timelinesrc );
+ } else {
+ $txt=$token["text"];
+ }
+ break;
default:
# Call language specific Hook.
$txt = $wgLang->processToken( $token, $tokenStack );
if ( !$image ) { $image = Namespace::getImage(); }
if ( !$special ) { $special = Namespace::getSpecial(); }
if ( !$media ) { $media = Namespace::getMedia(); }
- if ( !$category ) { $category = Namespace::getCategory(); ; }
+ if ( !$category ) { $category = Namespace::getCategory(); }
$nottalk = !Namespace::isTalk( $this->mTitle->getNamespace() );
if( $noforce ) {
if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgLang->getLanguageName( $iw ) ) {
array_push( $this->mOutput->mLanguageLinks, $nt->getPrefixedText() );
+ $s .= $prefix . $trail ;
return (trim($s) == '')? '': $s;
}
if( $ns == $image ) {
}
if( ( $nt->getPrefixedText() == $this->mTitle->getPrefixedText() ) &&
( strpos( $link, "#" ) == FALSE ) ) {
- $s .= $prefix . "<strong>" . $text . "</strong>" . $trail;
+ # Self-links are handled specially; generally de-link and change to bold.
+ $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, "", $trail );
return $s;
}
return $text."\n";
}
- /* private */ function doBlockLevels( $text, $linestart )
- {
+ /* private */ function doBlockLevels( $text, $linestart ) {
$fname = "Parser::doBlockLevels";
wfProfileIn( $fname );
+
# 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.
#
- $a = explode( "\n", $text );
+ $textLines = explode( "\n", $text );
- $lastPref = $text = $lastLine = '';
+ $lastPrefix = $output = $lastLine = '';
$this->mDTopen = $inBlockElem = false;
- $npl = 0;
- $pstack = false;
-
- if ( ! $linestart ) { $text .= array_shift( $a ); }
- foreach ( $a as $t ) {
- $oLine = $t;
- $opl = strlen( $lastPref );
- $preCloseMatch = preg_match("/<\\/pre/i", $t );
- $preOpenMatch = preg_match("/<pre/i", $t );
+ $prefixLength = 0;
+ $paragraphStack = false;
+
+ if ( !$linestart ) {
+ $output .= array_shift( $textLines );
+ }
+ foreach ( $textLines as $oLine ) {
+ $lastPrefixLength = strlen( $lastPrefix );
+ $preCloseMatch = preg_match("/<\\/pre/i", $oLine );
+ $preOpenMatch = preg_match("/<pre/i", $oLine );
if (!$this->mInPre) {
$this->mInPre = !empty($preOpenMatch);
}
if ( !$this->mInPre ) {
- $npl = strspn( $t, "*#:;" );
- $pref = substr( $t, 0, $npl );
+ # Multiple prefixes may abut each other for nested lists.
+ $prefixLength = strspn( $oLine, "*#:;" );
+ $pref = substr( $oLine, 0, $prefixLength );
+
+ # eh?
$pref2 = str_replace( ";", ":", $pref );
- $t = substr( $t, $npl );
+ $t = substr( $oLine, $prefixLength );
} else {
- $npl = 0;
+ # Don't interpret any other prefixes in preformatted text
+ $prefixLength = 0;
$pref = $pref2 = '';
+ $t = $oLine;
}
- // list generation
- if ( 0 != $npl && 0 == strcmp( $lastPref, $pref2 ) ) {
- $text .= $this->nextItem( substr( $pref, -1 ) );
- if ( $pstack ) { $pstack = false; }
+ # List generation
+ if( $prefixLength && 0 == strcmp( $lastPrefix, $pref2 ) ) {
+ # Same as the last item, so no need to deal with nesting or opening stuff
+ $output .= $this->nextItem( substr( $pref, -1 ) );
+ $paragraphStack = false;
if ( ";" == substr( $pref, -1 ) ) {
- $cpos = strpos( $t, ":" );
- if ( false !== $cpos ) {
- $term = substr( $t, 0, $cpos );
- $text .= $term . $this->nextItem( ":" );
- $t = substr( $t, $cpos + 1 );
+ # The one nasty exception: definition lists work like this:
+ # ; title : definition text
+ # So we check for : in the remainder text to split up the
+ # title and definition, without b0rking links.
+ # FIXME: This is not foolproof. Something better in Tokenizer might help.
+ if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
+ $term = $match[1];
+ $output .= $term . $this->nextItem( ":" );
+ $t = $match[2];
}
}
- } else if (0 != $npl || 0 != $opl) {
- $cpl = $this->getCommon( $pref, $lastPref );
- if ( $pstack ) { $pstack = false; }
-
- while ( $cpl < $opl ) {
- $text .= $this->closeList( $lastPref{$opl-1} );
- --$opl;
+ } elseif( $prefixLength || $lastPrefixLength ) {
+ # Either open or close a level...
+ $commonPrefixLength = $this->getCommon( $pref, $lastPrefix );
+ $paragraphStack = false;
+
+ while( $commonPrefixLength < $lastPrefixLength ) {
+ $output .= $this->closeList( $lastPrefix{$lastPrefixLength-1} );
+ --$lastPrefixLength;
}
- if ( $npl <= $cpl && $cpl > 0 ) {
- $text .= $this->nextItem( $pref{$cpl-1} );
+ if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) {
+ $output .= $this->nextItem( $pref{$commonPrefixLength-1} );
}
- while ( $npl > $cpl ) {
- $char = substr( $pref, $cpl, 1 );
- $text .= $this->openList( $char );
+ while ( $prefixLength > $commonPrefixLength ) {
+ $char = substr( $pref, $commonPrefixLength, 1 );
+ $output .= $this->openList( $char );
if ( ";" == $char ) {
- $cpos = strpos( $t, ":" );
- if ( ! ( false === $cpos ) ) {
- $term = substr( $t, 0, $cpos );
- $text .= $term . $this->nextItem( ":" );
- $t = substr( $t, $cpos + 1 );
+ # FIXME: This is dupe of code above
+ if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
+ $term = $match[1];
+ $output .= $term . $this->nextItem( ":" );
+ $t = $match[2];
}
}
- ++$cpl;
+ ++$commonPrefixLength;
}
- $lastPref = $pref2;
+ $lastPrefix = $pref2;
}
- if ( 0 == $npl ) { # No prefix (not in list)--go to paragraph mode
+ if( 0 == $prefixLength ) {
+ # No prefix (not in list)--go to paragraph mode
$uniq_prefix = UNIQ_PREFIX;
// XXX: use a stack for nestable elements like span, table and div
$openmatch = preg_match("/(<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<div|<pre|<tr|<td|<p|<ul|<li)/i", $t );
"/(<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|".
"<\\/div|<hr|<\\/td|<\\/pre|<\\/p|".$uniq_prefix."-pre|<\\/li|<\\/ul)/i", $t );
if ( $openmatch or $closematch ) {
- if ( $pstack ) { $pstack = false; }
- $text .= $this->closeParagraph();
+ $paragraphStack = false;
+ $output .= $this->closeParagraph();
if($preOpenMatch and !$preCloseMatch) {
$this->mInPre = true;
}
if ( " " == $t{0} ) {
// pre
if ($this->mLastSection != 'pre') {
- $pstack = false;
- $text .= $this->closeParagraph().'<pre>';
+ $paragraphStack = false;
+ $output .= $this->closeParagraph().'<pre>';
$this->mLastSection = 'pre';
}
} else {
// paragraph
if ( '' == trim($t) ) {
- if ( $pstack ) {
- $text .= $pstack.'<br/>';
- $pstack = false;
+ if ( $paragraphStack ) {
+ $output .= $paragraphStack.'<br/>';
+ $paragraphStack = false;
$this->mLastSection = 'p';
} else {
if ($this->mLastSection != 'p' ) {
- $text .= $this->closeParagraph();
+ $output .= $this->closeParagraph();
$this->mLastSection = '';
- $pstack = "<p>";
+ $paragraphStack = "<p>";
} else {
- $pstack = '</p><p>';
+ $paragraphStack = '</p><p>';
}
}
} else {
- if ( $pstack ) {
- $text .= $pstack;
- $pstack = false;
+ if ( $paragraphStack ) {
+ $output .= $paragraphStack;
+ $paragraphStack = false;
$this->mLastSection = 'p';
} else if ($this->mLastSection != 'p') {
- $text .= $this->closeParagraph().'<p>';
+ $output .= $this->closeParagraph().'<p>';
$this->mLastSection = 'p';
}
}
}
}
}
- if ($pstack === false) {
- $text .= $t."\n";
+ if ($paragraphStack === false) {
+ $output .= $t."\n";
}
}
- while ( $npl ) {
- $text .= $this->closeList( $pref2{$npl-1} );
- --$npl;
+ while ( $prefixLength ) {
+ $output .= $this->closeList( $pref2{$prefixLength-1} );
+ --$prefixLength;
}
if ( "" != $this->mLastSection ) {
- $text .= "</" . $this->mLastSection . ">";
+ $output .= "</" . $this->mLastSection . ">";
$this->mLastSection = "";
}
wfProfileOut( $fname );
- return $text;
+ return $output;
}
function getVariableValue( $index ) {
return $wgLang->getMonthNameGen( date("n") );
case MAG_CURRENTDAY:
return date("j");
+ case MAG_PAGENAME:
+ return $this->mTitle->getText();
+ case MAG_NAMESPACE:
+ # return Namespace::getCanonicalName($this->mTitle->getNamespace());
+ return $wgLang->getNsText($this->mTitle->getNamespace()); // Patch by Dori
case MAG_CURRENTDAYNAME:
return $wgLang->getWeekdayName( date("w")+1 );
case MAG_CURRENTYEAR:
# Run full parser on the included text
$text = $this->strip( $text, $this->mStripState );
- $text = $this->internalParse( $text, (bool)$newline, $assocArgs );
+ $text = $this->internalParse( $text, (bool)$newline, $assocArgs, false );
+ if(!empty($newline)) $text = "\n".$text;
# Add the result to the strip state for re-inclusion after
# the rest of the processing
if ( !$found ) {
return $matches[0];
} else {
- return $newline . $text;
+ return $text;
}
}
# Cleans up HTML, removes dangerous tags and attributes
/* private */ function removeHTMLtags( $text )
{
+ global $wgUseTidy, $wgUserHtml;
$fname = "Parser::removeHTMLtags";
wfProfileIn( $fname );
- $htmlpairs = array( # Tags that must be closed
- "b", "del", "i", "ins", "u", "font", "big", "small", "sub", "sup", "h1",
- "h2", "h3", "h4", "h5", "h6", "cite", "code", "em", "s",
- "strike", "strong", "tt", "var", "div", "center",
- "blockquote", "ol", "ul", "dl", "table", "caption", "pre",
- "ruby", "rt" , "rb" , "rp", "p"
- );
- $htmlsingle = array(
- "br", "hr", "li", "dt", "dd"
- );
- $htmlnest = array( # Tags that can be nested--??
- "table", "tr", "td", "th", "div", "blockquote", "ol", "ul",
- "dl", "font", "big", "small", "sub", "sup"
- );
- $tabletags = array( # Can only appear inside table
- "td", "th", "tr"
- );
+
+ if( $wgUserHtml ) {
+ $htmlpairs = array( # Tags that must be closed
+ "b", "del", "i", "ins", "u", "font", "big", "small", "sub", "sup", "h1",
+ "h2", "h3", "h4", "h5", "h6", "cite", "code", "em", "s",
+ "strike", "strong", "tt", "var", "div", "center",
+ "blockquote", "ol", "ul", "dl", "table", "caption", "pre",
+ "ruby", "rt" , "rb" , "rp", "p"
+ );
+ $htmlsingle = array(
+ "br", "hr", "li", "dt", "dd"
+ );
+ $htmlnest = array( # Tags that can be nested--??
+ "table", "tr", "td", "th", "div", "blockquote", "ol", "ul",
+ "dl", "font", "big", "small", "sub", "sup"
+ );
+ $tabletags = array( # Can only appear inside table
+ "td", "th", "tr"
+ );
+ } else {
+ $htmlpairs = array();
+ $htmlsingle = array();
+ $htmlnest = array();
+ $tabletags = array();
+ }
$htmlsingle = array_merge( $tabletags, $htmlsingle );
$htmlelements = array_merge( $htmlsingle, $htmlpairs );
$htmlattrs = $this->getHTMLattrs () ;
# Remove HTML comments
- $text = preg_replace( "/<!--.*-->/sU", "", $text );
+ $text = preg_replace( "/(\\n *<!--.*--> *(?=\\n)|<!--.*-->)/sU", "$2", $text );
$bits = explode( "<", $text );
$text = array_shift( $bits );
- $tagstack = array(); $tablestack = array();
-
- foreach ( $bits as $x ) {
- $prev = error_reporting( E_ALL & ~( E_NOTICE | E_WARNING ) );
- preg_match( "/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/",
- $x, $regs );
- list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
- error_reporting( $prev );
-
- $badtag = 0 ;
- if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
- # Check our stack
- if ( $slash ) {
- # Closing a tag...
- if ( ! in_array( $t, $htmlsingle ) &&
- ( $ot = array_pop( $tagstack ) ) != $t ) {
- array_push( $tagstack, $ot );
- $badtag = 1;
+ if(!$wgUseTidy) {
+ $tagstack = array(); $tablestack = array();
+ foreach ( $bits as $x ) {
+ $prev = error_reporting( E_ALL & ~( E_NOTICE | E_WARNING ) );
+ preg_match( "/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/",
+ $x, $regs );
+ list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
+ error_reporting( $prev );
+
+ $badtag = 0 ;
+ if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
+ # Check our stack
+ if ( $slash ) {
+ # Closing a tag...
+ if ( ! in_array( $t, $htmlsingle ) &&
+ ( count($tagstack) && $ot = array_pop( $tagstack ) ) != $t ) {
+ if(!empty($ot)) array_push( $tagstack, $ot );
+ $badtag = 1;
+ } else {
+ if ( $t == "table" ) {
+ $tagstack = array_pop( $tablestack );
+ }
+ $newparams = "";
+ }
} else {
- if ( $t == "table" ) {
- $tagstack = array_pop( $tablestack );
+ # Keep track for later
+ if ( in_array( $t, $tabletags ) &&
+ ! in_array( "table", $tagstack ) ) {
+ $badtag = 1;
+ } else if ( in_array( $t, $tagstack ) &&
+ ! in_array ( $t , $htmlnest ) ) {
+ $badtag = 1 ;
+ } else if ( ! in_array( $t, $htmlsingle ) ) {
+ if ( $t == "table" ) {
+ array_push( $tablestack, $tagstack );
+ $tagstack = array();
+ }
+ array_push( $tagstack, $t );
}
- $newparams = "";
+ # Strip non-approved attributes from the tag
+ $newparams = $this->fixTagAttributes($params);
+
}
- } else {
- # Keep track for later
- if ( in_array( $t, $tabletags ) &&
- ! in_array( "table", $tagstack ) ) {
- $badtag = 1;
- } else if ( in_array( $t, $tagstack ) &&
- ! in_array ( $t , $htmlnest ) ) {
- $badtag = 1 ;
- } else if ( ! in_array( $t, $htmlsingle ) ) {
- if ( $t == "table" ) {
- array_push( $tablestack, $tagstack );
- $tagstack = array();
- }
- array_push( $tagstack, $t );
+ if ( ! $badtag ) {
+ $rest = str_replace( ">", ">", $rest );
+ $text .= "<$slash$t $newparams$brace$rest";
+ continue;
}
- # Strip non-approved attributes from the tag
- $newparams = $this->fixTagAttributes($params);
-
}
- if ( ! $badtag ) {
+ $text .= "<" . str_replace( ">", ">", $x);
+ }
+ # Close off any remaining tags
+ while ( $t = array_pop( $tagstack ) ) {
+ $text .= "</$t>\n";
+ if ( $t == "table" ) { $tagstack = array_pop( $tablestack ); }
+ }
+ } else {
+ # this might be possible using tidy itself
+ foreach ( $bits as $x ) {
+ preg_match( "/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/",
+ $x, $regs );
+ @list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
+ if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
+ $newparams = $this->fixTagAttributes($params);
$rest = str_replace( ">", ">", $rest );
$text .= "<$slash$t $newparams$brace$rest";
- continue;
+ } else {
+ $text .= "<" . str_replace( ">", ">", $x);
}
- }
- $text .= "<" . str_replace( ">", ">", $x);
- }
- # Close off any remaining tags
- while ( $t = array_pop( $tagstack ) ) {
- $text .= "</$t>\n";
- if ( $t == "table" ) { $tagstack = array_pop( $tablestack ); }
+ }
}
wfProfileOut( $fname );
return $text;
}
+
/*
*
* This function accomplishes several tasks:
*
*/
- /* private */ function formatHeadings( $text )
+ /* private */ function formatHeadings( $text, $isMain=true )
{
+ global $wgInputEncoding,$wgRequest,$wgOut;
+
+ $startsection=$wgRequest->getVal('section');
+ if($startsection) { $startsection--;}
$doNumberHeadings = $this->mOptions->getNumberHeadings();
$doShowToc = $this->mOptions->getShowToc();
if( !$this->mTitle->userCanEdit() ) {
# strip out HTML
$canonized_headline = preg_replace( "/<.*?" . ">/","",$canonized_headline );
$tocline = trim( $canonized_headline );
- $canonized_headline = preg_replace("/[ \\?&\\/<>\\(\\)\\[\\]=,+']+/", '_', html_entity_decode( $tocline));
+ $canonized_headline = preg_replace("/[ \\?&\\/<>\\(\\)\\[\\]=,+']+/", '_', urlencode( do_html_entity_decode( $tocline, ENT_COMPAT, $wgInputEncoding ) ) );
$refer[$headlineCount] = $canonized_headline;
# count how many in assoc. array so we can track dupes in anchors
if ( empty( $head[$headlineCount] ) ) {
$head[$headlineCount] = "";
}
- $head[$headlineCount] .= $sk->editSectionLink($headlineCount+1);
+ $head[$headlineCount] .= $sk->editSectionLink($startsection+$headlineCount+1);
}
# Add the edit section span
if( $rightClickHack ) {
- $headline = $sk->editSectionScript($headlineCount+1,$headline);
+ $headline = $sk->editSectionScript($startsection+$headlineCount+1,$headline);
}
# give headline the correct <h#> tag
# Top anchor now in skin
$full = $full.$toc;
}
+
+ # If a page is viewed in collapsed mode, a TOC generated
+ # from the wikisource is stored in the title object.
+ # This TOC is now fetched and inserted here if it exists.
+ $collapsedtoc=$wgOut->getToc();
+ if ($collapsedtoc && !$i && $isMain) {
+ $full = $full.$collapsedtoc;
+ }
if( !empty( $head[$i] ) ) {
$full .= $head[$i];
return $full;
}
+ /* Generates a HTML-formatted table of contents which links to individual sections
+ from the wikisource. Used for collapsing long pages.
+ */
+ /* static */ function getTocFromSource( $text ) {
+
+ global $wgUser,$wgInputEncoding,$wgTitle,$wgOut,$wgParser;
+ $sk = $wgUser->getSkin();
+
+ $striparray=array();
+ $oldtype=$wgParser->mOutputType;
+ $wgParser->mOutputType=OT_WIKI;
+ $text=$wgParser->strip($text, $striparray, true);
+ $wgParser->mOutputType=$oldtype;
+
+ $numMatches = preg_match_all( "/^(=+)(.*?)=+|^<h([1-6]).*?>(.*?)<\/h[1-6].*?>/mi",$text,$matches);
+
+ # no headings: text cannot be collapsed
+ if( $numMatches == 0 ) {
+ return "";
+ }
+
+ # We combine the headlines into a bundle and convert them to HTML
+ # in order to make stripping out the wikicrap easier.
+ $combined=implode("!@@@!",$matches[2]);
+ $myout=$wgParser->parse($combined,$wgTitle,$wgOut->mParserOptions);
+ $combined_html=$myout->getText();
+ $headlines=array();
+ $headlines=explode("!@@@!",$combined_html);
+
+ # headline counter
+ $headlineCount = 0;
+ $toclevel = 0;
+ $toc = "";
+ $full = "";
+ $head = array();
+ $sublevelCount = array();
+ $level = 0;
+ $prevlevel = 0;
+ foreach( $headlines as $headline ) {
+ $headline=trim($headline);
+ $numbering = "";
+ if( $level ) {
+ $prevlevel = $level;
+ }
+ $level = $matches[1][$headlineCount];
+
+ # wikisource headings need to be converted into numbers
+ # =foo= equals <h1>foo</h1>, ==foo== equals <h2>foo</h2> etc.
+ if(strpos($level,"=")!==false) {
+ $level=strlen($level);
+ }
+
+ if( $prevlevel && $level > $prevlevel ) {
+ # reset when we enter a new level
+ $sublevelCount[$level] = 0;
+ $toc .= $sk->tocIndent( $level - $prevlevel );
+ $toclevel += $level - $prevlevel;
+ }
+ if( $level < $prevlevel ) {
+ # reset when we step back a level
+ $sublevelCount[$level+1]=0;
+ $toc .= $sk->tocUnindent( $prevlevel - $level );
+ $toclevel -= $prevlevel - $level;
+ }
+ # count number of headlines for each level
+ @$sublevelCount[$level]++;
+ $dot = 0;
+ for( $i = 1; $i <= $level; $i++ ) {
+ if( !empty( $sublevelCount[$i] ) ) {
+ if( $dot ) {
+ $numbering .= ".";
+ }
+ $numbering .= $sublevelCount[$i];
+ $dot = 1;
+ }
+ }
+
+
+ # The canonized header is a version of the header text safe to use for links
+ # Avoid insertion of weird stuff like <math> by expanding the relevant sections
+ $state=array();
+ $canonized_headline = Parser::unstrip( $headline, $state);
+
+ # strip out HTML
+ $canonized_headline = preg_replace( "/<.*?" . ">/","",$canonized_headline );
+ $tocline = trim( $canonized_headline );
+ $canonized_headline = preg_replace("/[ \\?&\\/<>\\(\\)\\[\\]=,+']+/", '_', urlencode( do_html_entity_decode( $tocline, ENT_COMPAT, $wgInputEncoding ) ) );
+ $refer[$headlineCount] = $canonized_headline;
+
+ # count how many in assoc. array so we can track dupes in anchors
+ @$refers[$canonized_headline]++;
+ $refcount[$headlineCount]=$refers[$canonized_headline];
+ $tocline = $numbering . " " . $tocline;
+
+ # Create the anchor for linking from the TOC to the section
+ $anchor = trim($canonized_headline);
+
+ if($refcount[$headlineCount] > 1 ) {
+ $anchor .= "_" . $refcount[$headlineCount];
+ }
+ $headlineCount++;
+ $toc .= $sk->tocLine($anchor,$tocline,$toclevel,$headlineCount);
+ }
+ $toclines = $headlineCount;
+ $toc .= $sk->tocUnindent( $toclevel );
+ $toc = $sk->tocTable( $toc );
+ return $toc;
+
+ }
/* private */ function doMagicISBN( &$tokenizer )
{
global $wgLang;