<?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;
$data = array () ;
$id = $this->mTitle->getArticleID() ;
- # 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 ;
- }
+ # FIXME: add limits
+ $t = wfStrencode( $this->mTitle->getDBKey() );
+ $sql = "SELECT DISTINCT cur_title,cur_namespace FROM cur,categorylinks WHERE cl_to='$t' AND cl_from=cur_id ORDER BY cl_sortkey" ;
+ $res = wfQuery ( $sql, DB_READ ) ;
+ while ( $x = wfFetchObject ( $res ) ) $data[] = $x ;
# For all pages that link to this category
foreach ( $data AS $x )
wfFreeResult ( $res ) ;
# Showing subcategories
- if ( count ( $children ) > 0 )
- {
- asort ( $children ) ;
+ if ( count ( $children ) > 0 ) {
$r .= "<h2>".wfMsg("subcategories")."</h2>\n" ;
$r .= implode ( ", " , $children ) ;
}
# Showing pages in this category
- if ( count ( $articles ) > 0 )
- {
+ if ( count ( $articles ) > 0 ) {
$ti = $this->mTitle->getText() ;
- asort ( $articles ) ;
$h = wfMsg( "category_header", $ti );
$r .= "<h2>{$h}</h2>\n" ;
$r .= implode ( ", " , $articles ) ;
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() )
+ # Parses the text and adds the result to the strip state
+ # Returns the strip tag
+ function stripParse( $text, $linestart, $args )
+ {
+ $text = $this->strip( $text, $this->mStripState );
+ $text = $this->internalParse( $text, $linestart, $args, false );
+ if( $linestart ) {
+ $text = "\n" . $text;
+ }
+ return $this->insertStripItem( $text, $this->mStripState );
+ }
+
+ 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 );
if ( !isset ( $this->categoryMagicDone ) ) {
- $text .= $this->categoryMagic () ;
- $this->categoryMagicDone = true ;
- }
+ $text .= $this->categoryMagic () ;
+ $this->categoryMagicDone = true ;
+ }
wfProfileOut( $fname );
return $text;
return $s;
}
+ /* private */ function handle4Quotes( &$state, $token )
+ {
+ /* This one makes some assumptions.
+ * '''Caesar''''s army => <strong>Caesar</strong>'s army
+ * ''''Caesar'''' was a roman emperor => '<strong>Caesar</strong>' was a roman emperor
+ * These assumptions might be wrong, but any other assumption might be wrong, too.
+ * So here we go */
+ if ( $state["strong"] !== false ) {
+ return $this->handle3Quotes( $state, $token ) . "'";
+ } else {
+ return "'" . $this->handle3Quotes( $state, $token );
+ }
+ }
+
+
/* private */ function handle3Quotes( &$state, $token )
{
if ( $state["strong"] !== false ) {
$state["strong"] = FALSE;
} else {
$s = "<strong>";
- $state["strong"] = isset($token["pos"]) ? $token["pos"] : true;
+ $state["strong"] = $token["pos"];
}
return $s;
}
$state["em"] = FALSE;
} else {
$s = "<em>";
- $state["em"] = isset($token["pos"]) ? $token["pos"] : true;
+ $state["em"] = $token["pos"];
}
return $s;
$state["em"] = $token["pos"];
} else { # not $em and not $strong
$s .= "<strong><em>";
- $state["strong"] = $state["em"] = isset($token["pos"]) ? $token["pos"] : true;
+ $state["strong"] = $state["em"] = $token["pos"];
}
return $s;
}
/* private */ function doTokenizedParser( $str )
{
global $wgLang; # for language specific parser hook
+ global $wgUploadDirectory, $wgUseTimeline;
$tokenizer=Tokenizer::newFromString( $str );
$tokenStack = array();
$txt = "\n<hr />\n";
break;
case "'''":
- # This and the three next ones handle quotes
+ # This and the four next ones handle quotes
$txt = $this->handle3Quotes( $state, $token );
break;
case "''":
case "'''''":
$txt = $this->handle5Quotes( $state, $token );
break;
+ case "''''":
+ $txt = $this->handle4Quotes( $state, $token );
+ break;
case "":
# empty token
$txt="";
break;
+ case "h":
+ #heading- used to close all unbalanced bold or em tags in this section
+ $txt = '';
+ if( $state['em'] !== false and
+ ( $state['strong'] === false or $state['em'] > $state['strong'] ) )
+ {
+ $s .= '</em>';
+ $state['em'] = false;
+ }
+ if ( $state['strong'] !== false ) $txt .= '</strong>';
+ if ( $state['em'] !== false ) $txt .= '</em>';
+ $state['strong'] = $state['em'] = false;
+ break;
case "RFC ":
if ( $tagIsOpen ) {
$txt = "RFC ";
$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 );
$s .= $txt;
}
} #end while
+
+ # make 100% sure all strong and em tags are closed
+ # doBlockLevels often messes the last bit up though, but invalid nesting is better than unclosed tags
+ # tidy solves this though
+ if( $state['em'] !== false and
+ ( $state['strong'] === false or $state['em'] > $state['strong'] ) )
+ {
+ $s .= '</em>';
+ $state['em'] = false;
+ }
+ if ( $state['strong'] !== false ) $s .= '</strong>';
+ if ( $state['em'] !== false ) $s .= '</em>';
+
if ( count( $tokenStack ) != 0 )
{
# still objects on stack. opened [[ tag without closing ]] tag.
if ( !$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() );
} else {
$noslash=substr($m[1],1);
}
- if($wgNamespacesWithSubpages[$this->mTitle->getNamespace()]) { # subpages allowed here
+ if(!empty($wgNamespacesWithSubpages[$this->mTitle->getNamespace()])) { # subpages allowed here
$link = $this->mTitle->getPrefixedText(). "/" . trim($noslash);
if( "" == $text ) {
$text= $m[1];
} else {
$link = substr( $m[1], 1 );
}
- if( "" == $text )
+ $wasblank = ( "" == $text );
+ if( $wasblank )
$text = $link;
$nt = Title::newFromText( $link );
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 ( $ns == $category ) {
$t = $nt->getText() ;
$nnt = Title::newFromText ( Namespace::getCanonicalName($category).":".$t ) ;
+
+ $wgLinkCache->suspend(); # Don't save in links/brokenlinks
$t = $sk->makeLinkObj( $nnt, $t, "", "" , $prefix );
+ $wgLinkCache->resume();
+
+ $sortkey = $wasblank ? $this->mTitle->getPrefixedText() : $text;
+ $wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
$this->mOutput->mCategoryLinks[] = $t ;
$s .= $prefix . $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 );
+ $openmatch = preg_match("/(<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<td|<p|<ul|<li)/i", $t );
$closematch = preg_match(
"/(<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|".
- "<\\/div|<hr|<\\/td|<\\/pre|<\\/p|".$uniq_prefix."-pre|<\\/li|<\\/ul)/i", $t );
+ "<div|<\\/div|<hr|<\\/td|<\\/pre|<\\/p|".$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;
}
$inBlockElem = true;
}
} else if ( !$inBlockElem ) {
- if ( " " == $t{0} ) {
+ if ( " " == $t{0} and trim($t) != '' ) {
// 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 = '';
+ $p
aragraphStack = "<p>";
} else {
+ $p
aragraphStack = '</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_CURRENTDAYNAME:
case MAG_PAGENAME:
return $this->mTitle->getText();
case MAG_NAMESPACE:
- return Namespace::getCanonicalName($this->mTitle->getNamespace());
+ # 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:
return date( "Y" );
$this->initialiseVariables();
}
$titleChars = Title::legalChars();
- $regex = "/(\\n?){{([$titleChars]*?)(\\|.*?|)}}/s";
# This function is called recursively. To keep track of arguments we need a stack:
array_push( $this->mArgStack, $args );
# PHP global rebinding syntax is a bit weird, need to use the GLOBALS array
$GLOBALS['wgCurParser'] =& $this;
+
+ # Argument substitution
+ if ( $this->mOutputType == OT_HTML ) {
+ $text = preg_replace_callback( "/(\\n?){{{([$titleChars]*?)}}}/", "wfArgSubstitution", $text );
+ }
+
+ # Double brace substitution
+ $regex = "/(\\n?){{([$titleChars]*?)(\\|.*?|)}}/s";
$text = preg_replace_callback( $regex, "wfBraceSubstitution", $text );
array_pop( $this->mArgStack );
$fname = "Parser::braceSubstitution";
$found = false;
$nowiki = false;
+ $noparse = false;
+
$title = NULL;
# $newline is an optional newline character before the braces
$args = array();
}
$argc = count( $args );
+
+ # {{{}}}
+ if ( strpos( $matches[0], "{{{" ) !== false ) {
+ $text = $matches[0];
+ $found = true;
+ $noparse = true;
+ }
# SUBST
- $mwSubst =& MagicWord::get( MAG_SUBST );
- if ( $mwSubst->matchStartAndRemove( $part1 ) ) {
- if ( $this->mOutputType != OT_WIKI ) {
- # Invalid SUBST not replaced at PST time
- # Return without further processing
+ if ( !$found ) {
+ $mwSubst =& MagicWord::get( MAG_SUBST );
+ if ( $mwSubst->matchStartAndRemove( $part1 ) ) {
+ if ( $this->mOutputType != OT_WIKI ) {
+ # Invalid SUBST not replaced at PST time
+ # Return without further processing
+ $text = $matches[0];
+ $found = true;
+ $noparse= true;
+ }
+ } elseif ( $this->mOutputType == OT_WIKI ) {
+ # SUBST not found in PST pass, do nothing
$text = $matches[0];
$found = true;
}
- } elseif ( $this->mOutputType == OT_WIKI ) {
- # SUBST not found in PST pass, do nothing
- $text = $matches[0];
- $found = true;
}
# MSG, MSGNW and INT
$found = true;
$this->mOutput->mContainsOldMagic = true;
}
-
+/*
# Arguments input from the caller
$inputArgs = end( $this->mArgStack );
if ( !$found && array_key_exists( $part1, $inputArgs ) ) {
$text = $inputArgs[$part1];
$found = true;
}
-
+*/
# Load from database
if ( !$found ) {
$title = Title::newFromText( $part1, NS_TEMPLATE );
# Only for HTML output
if ( $nowiki && $found && $this->mOutputType == OT_HTML ) {
$text = wfEscapeWikiText( $text );
- } elseif ( $this->mOutputType == OT_HTML && $found ) {
+ } elseif ( $this->mOutputType == OT_HTML && $found && !$noparse) {
# Clean up argument array
$assocArgs = array();
$index = 1;
}
# Run full parser on the included text
- $text = $this->strip( $text, $this->mStripState );
- $text = $this->internalParse( $text, (bool)$newline, $assocArgs );
-
- # Add the result to the strip state for re-inclusion after
- # the rest of the processing
- $text = $this->insertStripItem( $text, $this->mStripState );
+ $text = $this->stripParse( $text, (bool)$newline, $assocArgs );
# Resume the link cache and register the inclusion as a link
if ( !is_null( $title ) ) {
if ( !$found ) {
return $matches[0];
} else {
- return $newline . $text;
+ return $text;
}
}
+ # Triple brace replacement -- used for template arguments
+ function argSubstitution( $matches )
+ {
+ $newline = $matches[1];
+ $arg = trim( $matches[2] );
+ $text = $matches[0];
+ $inputArgs = end( $this->mArgStack );
+
+ if ( array_key_exists( $arg, $inputArgs ) ) {
+ $text = $this->stripParse( $inputArgs[$arg], (bool)$newline, array() );
+ }
+
+ return $text;
+ }
+
# Returns true if the function is allowed to include this entity
function incrementIncludeCount( $dbk )
{
# 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;
+
$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 ) ) );
+ # strip out urlencoded (inserted for french spaces, e.g. first space in 'something : something')
+ $canonized_headline = str_replace('%C2%A0','_', $canonized_headline);
$refer[$headlineCount] = $canonized_headline;
# count how many in assoc. array so we can track dupes in anchors
# $full .= $sk->editSectionLink(0);
}
$full .= $block;
- if( $doShowToc && !$i) {
+ if( $doShowToc && !$i && $isMain) {
# Top anchor now in skin
$full = $full.$toc;
}
return $wgCurParser->braceSubstitution( $matches );
}
+function wfArgSubstitution( $matches )
+{
+ global $wgCurParser;
+ return $wgCurParser->argSubstitution( $matches );
+}
+
?>
dépôts / lhc / web / wiklou.git / blobdiff