<?php
-
/**
* File for Parser and related classes
*
* @package MediaWiki
+ * @subpackage Parser
*/
+/** */
+require_once( 'Sanitizer.php' );
+
/**
* Update this version number when the ParserOutput format
* changes in an incompatible way, so the parser cache
* can automatically discard old data.
*/
-define( 'MW_PARSER_VERSION', '1.4.0' );
+define( 'MW_PARSER_VERSION', '1.5.0' );
/**
* Variable substitution O(N^2) attack
function clearState() {
$this->mOutput = new ParserOutput;
$this->mAutonumber = 0;
- $this->mLastSection = "";
+ $this->mLastSection = '';
$this->mDTopen = false;
$this->mVariables = false;
$this->mIncludeCount = array();
* to internalParse() which does all the real work.
*
* @access private
+ * @param string $text Text we want to parse
+ * @param Title &$title A title object
+ * @param array $options
+ * @param boolean $linestart
+ * @param boolean $clearState
* @return ParserOutput a ParserOutput
*/
function parse( $text, &$title, $options, $linestart = true, $clearState = true ) {
$text = $this->strip( $text, $x );
$text = $this->internalParse( $text, $linestart );
+
+ $dashReplace = array(
+ '/ - /' => " – ", # N dash
+ '/(?<=[0-9])-(?=[0-9])/' => "–", # N dash between numbers
+ '/ -- /' => " — " # M dash
+ );
+ $text = preg_replace( array_keys($dashReplace), array_values($dashReplace), $text );
+
+
$text = $this->unstrip( $text, $this->mStripState );
# Clean up special characters, only run once, next-to-last before doBlockLevels
+ global $wgUseTidy;
if(!$wgUseTidy) {
$fixtags = array(
# french spaces, last one Guillemet-left
# only if there is something before the space
- '/(.) (?=\\?|:|;|!|\\302\\273)/i' => '\\1 \\2',
+ '/(.) (?=\\?|:|;|!|\\302\\273)/' => '\\1 \\2',
# french spaces, Guillemet-right
- "/(\\302\\253) /i"=>"\\1 ",
+ '/(\\302\\253) /' => '\\1 ',
'/<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 );
+ $text = Sanitizer::normalizeCharReferences( $text );
} else {
$fixtags = array(
# french spaces, last one Guillemet-left
- '/ (\\?|:|;|!|\\302\\273)/i' => ' \\1',
+ '/ (\\?|:|;|!|\\302\\273)/' => ' \\1',
# french spaces, Guillemet-right
- '/(\\302\\253) /i' => '\\1 ',
+ '/(\\302\\253) /' => '\\1 ',
'/<center *>/i' => '<div class="center">',
'/<\\/center *>/i' => '</div>'
);
$this->replaceLinkHolders( $text );
$text = $wgContLang->convert($text);
-
+ $this->mOutput->setTitleText($wgContLang->getParsedTitle());
$text = $this->unstripNoWiki( $text, $this->mStripState );
- global $wgUseTidy;
if ($wgUseTidy) {
$text = Parser::tidy($text);
}
while ( '' != $text ) {
if($tag==STRIP_COMMENTS) {
- $p = preg_split( '/<!--/i', $text, 2 );
+ $p = preg_split( '/<!--/', $text, 2 );
} else {
$p = preg_split( "/<\\s*$tag\\s*>/i", $text, 2 );
}
return $rnd;
}
- /**
- * Return allowed HTML attributes
- *
- * @access private
- */
- function getHTMLattrs () {
- $htmlattrs = array( # Allowed attributes--no scripting, etc.
- 'title', 'align', 'lang', 'dir', 'width', 'height',
- 'bgcolor', 'clear', /* BR */ 'noshade', /* HR */
- 'cite', /* BLOCKQUOTE, Q */ 'size', 'face', 'color',
- /* FONT */ 'type', 'start', 'value', 'compact',
- /* For various lists, mostly deprecated but safe */
- 'summary', 'width', 'border', 'frame', 'rules',
- 'cellspacing', 'cellpadding', 'valign', 'char',
- 'charoff', 'colgroup', 'col', 'span', 'abbr', 'axis',
- 'headers', 'scope', 'rowspan', 'colspan', /* Tables */
- 'id', 'class', 'name', 'style' /* For CSS */
- );
- return $htmlattrs ;
- }
-
- /**
- * Remove non approved attributes and javascript in css
- *
- * @access private
- */
- function fixTagAttributes ( $t ) {
- if ( trim ( $t ) == '' ) return '' ; # Saves runtime ;-)
- $htmlattrs = $this->getHTMLattrs() ;
-
- # Strip non-approved attributes from the tag
- $t = preg_replace(
- '/(\\w+)(\\s*=\\s*([^\\s\">]+|\"[^\">]*\"))?/e',
- "(in_array(strtolower(\"\$1\"),\$htmlattrs)?(\"\$1\".((\"x\$3\" != \"x\")?\"=\$3\":'')):'')",
- $t);
-
- $t = str_replace ( '<></>' , '' , $t ) ; # This should fix bug 980557
-
- # Strip javascript "expression" from stylesheets. Brute force approach:
- # If anythin offensive is found, all attributes of the HTML tag are dropped
-
- if( preg_match(
- '/style\\s*=.*(expression|tps*:\/\/|url\\s*\().*/is',
- wfMungeToUtf8( $t ) ) )
- {
- $t='';
- }
-
- return trim ( $t ) ;
- }
-
/**
* interface with html tidy, used if $wgUseTidy = true
*
*/
function tidy ( $text ) {
global $wgTidyConf, $wgTidyBin, $wgTidyOpts;
- global $wgInputEncoding, $wgOutputEncoding;
$fname = 'Parser::tidy';
wfProfileIn( $fname );
$cleansource = '';
- $opts = '';
- switch(strtoupper($wgOutputEncoding)) {
- case 'ISO-8859-1':
- $opts .= ($wgInputEncoding == $wgOutputEncoding)? ' -latin1':' -raw';
- break;
- case 'UTF-8':
- $opts .= ($wgInputEncoding == $wgOutputEncoding)? ' -utf8':' -raw';
- break;
- default:
- $opts .= ' -raw';
- }
+ $opts = ' -utf8';
$wrappedtext = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'.
' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>'.
$fc = substr ( $x , 0 , 1 ) ;
if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
$indent_level = strlen( $matches[1] );
- $t[$k] = "\n" .
- str_repeat( '<dl><dd>', $indent_level ) .
- '<table ' . $this->fixTagAttributes ( $matches[2] ) . '>' ;
+ $t[$k] = str_repeat( '<dl><dd>', $indent_level ) .
+ '<table' . Sanitizer::fixTagAttributes ( $matches[2], 'table' ) . '>' ;
array_push ( $td , false ) ;
array_push ( $ltd , '' ) ;
array_push ( $tr , false ) ;
}
else if ( count ( $td ) == 0 ) { } # Don't do any of the following
else if ( '|}' == substr ( $x , 0 , 2 ) ) {
- $z = "</table>\n" ;
+ $z = "</table>" . substr ( $x , 2);
$l = array_pop ( $ltd ) ;
if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
array_push ( $tr , false ) ;
array_push ( $td , false ) ;
array_push ( $ltd , '' ) ;
- array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
+ array_push ( $ltr , Sanitizer::fixTagAttributes ( $x, 'tr' ) ) ;
}
else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) { # Caption
# $x is a table row
if ( $fc != '+' )
{
$tra = array_pop ( $ltr ) ;
- if ( !array_pop ( $tr ) ) $z = '<tr '.$tra.">\n" ;
+ if ( !array_pop ( $tr ) ) $z = '<tr'.$tra.">\n" ;
array_push ( $tr , true ) ;
array_push ( $ltr , '' ) ;
}
}
if ( count ( $y ) == 1 )
$y = "{$z}<{$l}>{$y[0]}" ;
- else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
+ else $y = $y = "{$z}<{$l}".Sanitizer::fixTagAttributes($y[0], $l).">{$y[1]}" ;
$t[$k] .= $y ;
array_push ( $td , true ) ;
}
}
$t = implode ( "\n" , $t ) ;
- # $t = $this->removeHTMLtags( $t );
+ # $t = Sanitizer::removeHTMLtags( $t );
wfProfileOut( $fname );
return $t ;
}
$fname = 'Parser::internalParse';
wfProfileIn( $fname );
- $text = $this->removeHTMLtags( $text );
+ $text = Sanitizer::removeHTMLtags( $text );
$text = $this->replaceVariables( $text, $args );
$text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
* @access private
*/
function &doMagicLinks( &$text ) {
- global $wgUseGeoMode;
$text = $this->magicISBN( $text );
- if ( isset( $wgUseGeoMode ) && $wgUseGeoMode ) {
- $text = $this->magicGEO( $text );
- }
$text = $this->magicRFC( $text, 'RFC ', 'rfcurl' );
$text = $this->magicRFC( $text, 'PMID ', 'pubmedurl' );
return $text;
$dtrail = '';
+ # Set linktype for CSS - if URL==text, link is essentially free
+ $linktype = ($text == $url) ? 'free' : 'text';
+
# No link text, e.g. [http://domain.tld/some.link]
if ( $text == '' ) {
# Autonumber if allowed
if ( strpos( HTTP_PROTOCOLS, $protocol ) !== false ) {
$text = '[' . ++$this->mAutonumber . ']';
+ $linktype = 'autonumber';
} else {
# Otherwise just use the URL
$text = htmlspecialchars( $url );
+ $linktype = 'free';
}
} else {
# Have link text, e.g. [http://domain.tld/some.link text]s
}
}
- $encUrl = htmlspecialchars( $url );
- # Bit in parentheses showing the URL for the printable version
- if( $url == $text || preg_match( "!$protocol://" . preg_quote( $text, '/' ) . "/?$!", $url ) ) {
- $paren = '';
- } else {
- # Expand the URL for printable version
- if ( ! $sk->suppressUrlExpansion() ) {
- $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $encUrl ) . "</i>)</span>";
- } else {
- $paren = '';
- }
- }
+ # Replace & from obsolete syntax with &.
+ # All HTML entities will be escaped by makeExternalLink()
+ # or maybeMakeImageLink()
+ $url = str_replace( '&', '&', $url );
# Process the trail (i.e. everything after this link up until start of the next link),
# replacing any non-bracketed links
# This means that users can paste URLs directly into the text
# Funny characters like ö aren't valid in URLs anyway
# This was changed in August 2004
- $s .= $sk->makeExternalLink( $url, $text, false ) . $dtrail. $paren . $trail;
+ $s .= $sk->makeExternalLink( $url, $text, false, $linktype ) . $dtrail . $trail;
}
wfProfileOut( $fname );
$text = $this->maybeMakeImageLink( $url );
if ( $text === false ) {
# Not an image, make a link
- $text = $sk->makeExternalLink( $url, $url );
+ $text = $sk->makeExternalLink( $url, $url, true, 'free' );
}
$s .= $text . $trail;
} else {
*
* @access private
*/
-
function replaceInternalLinks( $s ) {
global $wgLang, $wgContLang, $wgLinkCache;
global $wgDisableLangConversion;
# Match a link having the form [[namespace:link|alternate]]trail
static $e1 = FALSE;
- if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|([^]]+))?]](.*)\$/sD"; }
+ if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|(.+?))?]](.*)\$/sD"; }
# Match cases where there is no "]]", which might still be images
static $e1_img = FALSE;
if ( !$e1_img ) { $e1_img = "/^([{$tc}]+)\\|(.*)\$/sD"; }
$useLinkPrefixExtension = $wgContLang->linkPrefixExtension();
- $nottalk = !Namespace::isTalk( $this->mTitle->getNamespace() );
+ if( is_null( $this->mTitle ) ) {
+ wfDebugDieBacktrace( 'nooo' );
+ }
+ $nottalk = !$this->mTitle->isTalkPage();
if ( $useLinkPrefixExtension ) {
if ( preg_match( $e2, $s, $m ) ) {
if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
$text = $m[2];
+ # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
+ # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
+ # the real problem is with the $e1 regex
+ # See bug 1300.
+ if (preg_match( "/^\](.*)/", $m[3], $n ) ) {
+ $text .= ']'; # so that replaceExternalLinks($text) works later
+ $m[3] = $n[1];
+ }
# fix up urlencoded title texts
if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
$trail = $m[3];
# Interwikis
if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) {
array_push( $this->mOutput->mLanguageLinks, $nt->getFullText() );
- $tmp = $prefix . $trail ;
- $s .= (trim($tmp) == '')? '': $tmp;
+ $s = rtrim($s . "\n");
+ $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail;
continue;
}
if ( $ns == NS_IMAGE ) {
wfProfileIn( "$fname-image" );
-
- # recursively parse links inside the image caption
- # actually, this will parse them in any other parameters, too,
- # but it might be hard to fix that, and it doesn't matter ATM
- $text = $this->replaceExternalLinks($text);
- $text = $this->replaceInternalLinks($text);
-
- # replace the image with a link-holder so that replaceExternalLinks() can't mess with it
- $s .= $prefix . $this->insertStripItem( $sk->makeImageLinkObj( $nt, $text ), $this->mStripState ) . $trail;
- $wgLinkCache->addImageLinkObj( $nt );
-
+ if ( !wfIsBadImage( $nt->getDBkey() ) ) {
+ # recursively parse links inside the image caption
+ # actually, this will parse them in any other parameters, too,
+ # but it might be hard to fix that, and it doesn't matter ATM
+ $text = $this->replaceExternalLinks($text);
+ $text = $this->replaceInternalLinks($text);
+
+ # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
+ $s .= $prefix . str_replace('http://', 'http-noparse://', $sk->makeImageLinkObj( $nt, $text ) ) . $trail;
+ $wgLinkCache->addImageLinkObj( $nt );
+
+ wfProfileOut( "$fname-image" );
+ continue;
+ }
wfProfileOut( "$fname-image" );
- continue;
+
}
if ( $ns == NS_CATEGORY ) {
wfProfileIn( "$fname-category" );
$t = $nt->getText();
+ $s = rtrim($s . "\n"); # bug 87
$wgLinkCache->suspend(); # Don't save in links/brokenlinks
$pPLC=$sk->postParseLinkColour();
}
$wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
$this->mOutput->addCategoryLink( $t );
- $s .= $prefix . $trail ;
+
+ /**
+ * Strip the whitespace Category links produce, see bug 87
+ * @todo We might want to use trim($tmp, "\n") here.
+ */
+ $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail;
wfProfileOut( "$fname-category" );
continue;
* @access private
*/
function getVariableValue( $index ) {
- global $wgContLang, $wgSitename, $wgServer;
+ global $wgContLang, $wgSitename, $wgServer, $wgArticle;
/**
* Some of these require message or data lookups and can be
return $varCache[$index] = $wgContLang->getMonthName( date('n') );
case MAG_CURRENTMONTHNAMEGEN:
return $varCache[$index] = $wgContLang->getMonthNameGen( date('n') );
+ case MAG_CURRENTMONTHABBREV:
+ return $varCache[$index] = $wgContLang->getMonthAbbreviation( date('n') );
case MAG_CURRENTDAY:
return $varCache[$index] = $wgContLang->formatNum( date('j') );
case MAG_PAGENAME:
return $this->mTitle->getText();
case MAG_PAGENAMEE:
return $this->mTitle->getPartialURL();
+ case MAG_REVISIONID:
+ return $wgArticle->getRevIdFetched();
case MAG_NAMESPACE:
# return Namespace::getCanonicalName($this->mTitle->getNamespace());
return $wgContLang->getNsText($this->mTitle->getNamespace()); # Patch by Dori
case MAG_CURRENTDAYNAME:
return $varCache[$index] = $wgContLang->getWeekdayName( date('w')+1 );
case MAG_CURRENTYEAR:
- return $varCache[$index] = $wgContLang->formatNum( date( 'Y' ) );
+ return $varCache[$index] = $wgContLang->formatNum( date( 'Y' ), true );
case MAG_CURRENTTIME:
return $varCache[$index] = $wgContLang->time( wfTimestampNow(), false );
+ case MAG_CURRENTWEEK:
+ return $varCache[$index] = $wgContLang->formatNum( date('W') );
+ case MAG_CURRENTDOW:
+ return $varCache[$index] = $wgContLang->formatNum( date('w') );
case MAG_NUMBEROFARTICLES:
return $varCache[$index] = $wgContLang->formatNum( wfNumberOfArticles() );
case MAG_SITENAME:
$this->mVariables = array();
foreach ( $wgVariableIDs as $id ) {
$mw =& MagicWord::get( $id );
- $mw->addToArray( $this->mVariables, $this->getVariableValue( $id ) );
+ $mw->addToArray( $this->mVariables, $id );
}
wfProfileOut( $fname );
}
*/
function variableSubstitution( $matches ) {
$fname = 'parser::variableSubstitution';
+ $varname = $matches[1];
wfProfileIn( $fname );
if ( !$this->mVariables ) {
$this->initialiseVariables();
if ( $this->mOutputType == OT_WIKI ) {
# Do only magic variables prefixed by SUBST
$mwSubst =& MagicWord::get( MAG_SUBST );
- if (!$mwSubst->matchStartAndRemove( $matches[1] ))
+ if (!$mwSubst->matchStartAndRemove( $varname ))
$skip = true;
# Note that if we don't substitute the variable below,
# we don't remove the {{subst:}} magic word, in case
# it is a template rather than a magic variable.
}
- if ( !$skip && array_key_exists( $matches[1], $this->mVariables ) ) {
- $text = $this->mVariables[$matches[1]];
+ if ( !$skip && array_key_exists( $varname, $this->mVariables ) ) {
+ $id = $this->mVariables[$varname];
+ $text = $this->getVariableValue( $id );
$this->mOutput->mContainsOldMagic = true;
} else {
$text = $matches[0];
# merged with the next arg because the '|' character between belongs
# to the link syntax and not the template parameter syntax.
$argc = count($args);
- $i = 0;
+
for ( $i = 0; $i < $argc-1; $i++ ) {
if ( substr_count ( $args[$i], '[[' ) != substr_count ( $args[$i], ']]' ) ) {
$args[$i] .= '|'.$args[$i+1];
# Did we encounter this template already? If yes, it is in the cache
# and we need to check for loops.
if ( !$found && isset( $this->mTemplates[$part1] ) ) {
- # set $text to cached message.
- $text = $linestart . $this->mTemplates[$part1];
$found = true;
# Infinite loop test
if ( isset( $this->mTemplatePath[$part1] ) ) {
$noparse = true;
$found = true;
- $text .= '<!-- WARNING: template loop detected -->';
+ $text = $linestart .
+ "\{\{$part1}}" .
+ '<!-- WARNING: template loop detected -->';
+ wfDebug( "$fname: template loop broken at '$part1'\n" );
+ } else {
+ # set $text to cached message.
+ $text = $linestart . $this->mTemplates[$part1];
}
}
# Load from database
$itcamefromthedatabase = false;
+ $lastPathLevel = $this->mTemplatePath;
if ( !$found ) {
$ns = NS_TEMPLATE;
$part1 = $this->maybeDoSubpageLink( $part1, $subpage='' );
}
# Template cache array insertion
- $this->mTemplates[$part1] = $text;
+ if( $found ) {
+ $this->mTemplates[$part1] = $text;
+ }
}
}
$this->mTemplatePath[$part1] = 1;
$text = $this->strip( $text, $this->mStripState );
- $text = $this->removeHTMLtags( $text );
+ $text = Sanitizer::removeHTMLtags( $text );
$text = $this->replaceVariables( $text, $assocArgs );
# Resume the link cache and register the inclusion as a link
$text = "\n" . $text;
}
}
-
- # Empties the template path
- $this->mTemplatePath = array();
+ # Prune lower levels off the recursion check path
+ $this->mTemplatePath = $lastPathLevel;
+
if ( !$found ) {
wfProfileOut( $fname );
return $matches[0];
}
}
}
-
- # Empties the template path
- $this->mTemplatePath = array();
+ # Prune lower levels off the recursion check path
+ $this->mTemplatePath = $lastPathLevel;
if ( !$found ) {
wfProfileOut( $fname );
}
}
-
- /**
- * Cleans up HTML, removes dangerous tags and attributes, and
- * removes HTML comments
- * @access private
- */
- function removeHTMLtags( $text ) {
- global $wgUseTidy, $wgUserHtml;
- $fname = 'Parser::removeHTMLtags';
- wfProfileIn( $fname );
-
- 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', 'span'
- );
- $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', 'span'
- );
- $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 = $this->removeHTMLcomments( $text );
-
- $bits = explode( '<', $text );
- $text = array_shift( $bits );
- if(!$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 ) &&
- ( $ot = @array_pop( $tagstack ) ) != $t ) {
- @array_push( $tagstack, $ot );
- $badtag = 1;
- } else {
- if ( $t == 'table' ) {
- $tagstack = array_pop( $tablestack );
- }
- $newparams = '';
- }
- } 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 );
- }
- # Strip non-approved attributes from the tag
- $newparams = $this->fixTagAttributes($params);
-
- }
- if ( ! $badtag ) {
- $rest = str_replace( '>', '>', $rest );
- $text .= "<$slash$t $newparams$brace$rest";
- continue;
- }
- }
- $text .= '<' . str_replace( '>', '>', $x);
- }
- # Close off any remaining tags
- while ( is_array( $tagstack ) && ($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";
- } else {
- $text .= '<' . str_replace( '>', '>', $x);
- }
- }
- }
- wfProfileOut( $fname );
- return $text;
- }
-
- /**
- * Remove '<!--', '-->', and everything between.
- * To avoid leaving blank lines, when a comment is both preceded
- * and followed by a newline (ignoring spaces), trim leading and
- * trailing spaces and one of the newlines.
- *
- * @access private
- */
- function removeHTMLcomments( $text ) {
- $fname='Parser::removeHTMLcomments';
- wfProfileIn( $fname );
- while (($start = strpos($text, '<!--')) !== false) {
- $end = strpos($text, '-->', $start + 4);
- if ($end === false) {
- # Unterminated comment; bail out
- break;
- }
-
- $end += 3;
-
- # Trim space and newline if the comment is both
- # preceded and followed by a newline
- $spaceStart = max($start - 1, 0);
- $spaceLen = $end - $spaceStart;
- while (substr($text, $spaceStart, 1) === ' ' && $spaceStart > 0) {
- $spaceStart--;
- $spaceLen++;
- }
- while (substr($text, $spaceStart + $spaceLen, 1) === ' ')
- $spaceLen++;
- if (substr($text, $spaceStart, 1) === "\n" and substr($text, $spaceStart + $spaceLen, 1) === "\n") {
- # Remove the comment, leading and trailing
- # spaces, and leave only one newline.
- $text = substr_replace($text, "\n", $spaceStart, $spaceLen + 1);
- }
- else {
- # Remove just the comment.
- $text = substr_replace($text, '', $start, $end - $start);
- }
- }
- wfProfileOut( $fname );
- return $text;
- }
-
/**
* This function accomplishes several tasks:
* 1) Auto-number headings if that option is enabled
*
* It loops through all headlines, collects the necessary data, then splits up the
* string and re-inserts the newly formatted headlines.
+ *
+ * @param string $text
+ * @param boolean $isMain
* @access private
*/
- /* private */ function formatHeadings( $text, $isMain=true ) {
- global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders;
+ function formatHeadings( $text, $isMain=true ) {
+ global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders, $wgInterwikiLinkHolders;
$doNumberHeadings = $this->mOptions->getNumberHeadings();
- $doShowToc = $this->mOptions->getShowToc();
+ $doShowToc = true;
$forceTocHere = false;
if( !$this->mTitle->userCanEdit() ) {
$showEditLink = 0;
- $rightClickHack = 0;
} else {
$showEditLink = $this->mOptions->getEditSection();
- $rightClickHack = $this->mOptions->getEditSectionOnRightClick();
}
# Inhibit editsection links if requested in the page
# do not add TOC
$mw =& MagicWord::get( MAG_NOTOC );
if( $mw->matchAndRemove( $text ) ) {
- $doShowToc = 0;
- }
-
- # never add the TOC to the Main Page. This is an entry page that should not
- # be more than 1-2 screens large anyway
- if( $this->mTitle->getPrefixedText() == wfMsg('mainpage') ) {
- $doShowToc = 0;
+ $doShowToc = false;
}
# Get all headlines for numbering them and adding funky stuff like [edit]
# if there are fewer than 4 headlines in the article, do not show TOC
if( $numMatches < 4 ) {
- $doShowToc = 0;
+ $doShowToc = false;
}
# if the string __TOC__ (not case-sensitive) occurs in the HTML,
# override above conditions and always show TOC at that place
+
$mw =& MagicWord::get( MAG_TOC );
- if ($mw->match( $text ) ) {
- $doShowToc = 1;
+ if($mw->match( $text ) ) {
+ $doShowToc = true;
$forceTocHere = true;
} else {
# if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
# override above conditions and always show TOC above first header
$mw =& MagicWord::get( MAG_FORCETOC );
if ($mw->matchAndRemove( $text ) ) {
- $doShowToc = 1;
+ $doShowToc = true;
}
}
-
+ # Never ever show TOC if no headers
+ if( $numMatches < 1 ) {
+ $doShowToc = false;
+ }
# We need this to perform operations on the HTML
$sk =& $this->mOptions->getSkin();
# Ugh .. the TOC should have neat indentation levels which can be
# passed to the skin functions. These are determined here
- $toclevel = 0;
$toc = '';
$full = '';
$head = array();
$sublevelCount = array();
+ $levelCount = array();
+ $toclevel = 0;
$level = 0;
$prevlevel = 0;
+ $toclevel = 0;
+ $prevtoclevel = 0;
+
foreach( $matches[3] as $headline ) {
$istemplate = 0;
- $templatetitle = "";
+ $templatetitle = '';
$templatesection = 0;
+ $numbering = '';
if (preg_match("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", $headline, $mat)) {
$istemplate = 1;
$headline = preg_replace("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", "", $headline);
}
- $numbering = '';
- if( $level ) {
+ if( $toclevel ) {
$prevlevel = $level;
+ $prevtoclevel = $toclevel;
}
$level = $matches[1][$headlineCount];
- if( ( $doNumberHeadings || $doShowToc ) && $prevlevel && $level > $prevlevel ) {
- # reset when we enter a new level
- $sublevelCount[$level] = 0;
- $toc .= $sk->tocIndent( $level - $prevlevel );
- $toclevel += $level - $prevlevel;
- }
- if( ( $doNumberHeadings || $doShowToc ) && $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]++;
+
if( $doNumberHeadings || $doShowToc ) {
+
+ if ( $level > $prevlevel ) {
+ # Increase TOC level
+ $toclevel++;
+ $sublevelCount[$toclevel] = 0;
+ $toc .= $sk->tocIndent();
+ }
+ elseif ( $level < $prevlevel && $toclevel > 1 ) {
+ # Decrease TOC level, find level to jump to
+
+ if ( $toclevel == 2 && $level <= $levelCount[1] ) {
+ # Can only go down to level 1
+ $toclevel = 1;
+ } else {
+ for ($i = $toclevel; $i > 0; $i--) {
+ if ( $levelCount[$i] == $level ) {
+ # Found last matching level
+ $toclevel = $i;
+ break;
+ }
+ elseif ( $levelCount[$i] < $level ) {
+ # Found first matching level below current level
+ $toclevel = $i + 1;
+ break;
+ }
+ }
+ }
+
+ $toc .= $sk->tocUnindent( $prevtoclevel - $toclevel );
+ }
+ else {
+ # No change in level, end TOC line
+ $toc .= $sk->tocLineEnd();
+ }
+
+ $levelCount[$toclevel] = $level;
+
+ # count number of headlines for each level
+ @$sublevelCount[$toclevel]++;
$dot = 0;
- for( $i = 1; $i <= $level; $i++ ) {
+ for( $i = 1; $i <= $toclevel; $i++ ) {
if( !empty( $sublevelCount[$i] ) ) {
if( $dot ) {
$numbering .= '.';
# 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
$canonized_headline = $this->unstrip( $headline, $this->mStripState );
- $canonized_headline = $this->unstripNoWiki( $headline, $this->mStripState );
+ $canonized_headline = $this->unstripNoWiki( $canonized_headline, $this->mStripState );
# Remove link placeholders by the link text.
# <!--LINK number-->
$canonized_headline = preg_replace( '/<!--LINK ([0-9]*)-->/e',
"\$wgLinkHolders['texts'][\$1]",
$canonized_headline );
+ $canonized_headline = preg_replace( '/<!--IWLINK ([0-9]*)-->/e',
+ "\$wgInterwikiLinkHolders[\$1]",
+ $canonized_headline );
# strip out HTML
$canonized_headline = preg_replace( '/<.*?' . '>/','',$canonized_headline );
@$refers[$canonized_headline]++;
$refcount[$headlineCount]=$refers[$canonized_headline];
- # Prepend the number to the heading text
-
- if( $doNumberHeadings || $doShowToc ) {
- $tocline = $numbering . ' ' . $tocline;
-
- # Don't number the heading if it is the only one (looks silly)
- if( $doNumberHeadings && count( $matches[3] ) > 1) {
- # the two are different if the line contains a link
- $headline=$numbering . ' ' . $headline;
- }
+ # Don't number the heading if it is the only one (looks silly)
+ if( $doNumberHeadings && count( $matches[3] ) > 1) {
+ # the two are different if the line contains a link
+ $headline=$numbering . ' ' . $headline;
}
# Create the anchor for linking from the TOC to the section
$anchor .= '_' . $refcount[$headlineCount];
}
if( $doShowToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) {
- $toc .= $sk->tocLine($anchor,$tocline,$toclevel);
+ $toc .= $sk->tocLine($anchor, $tocline, $numbering, $toclevel);
}
if( $showEditLink && ( !$istemplate || $templatetitle !== "" ) ) {
if ( empty( $head[$headlineCount] ) ) {
$head[$headlineCount] .= $sk->editSectionLink($this->mTitle, $sectionCount+1);
}
- # Add the edit section span
- if( $rightClickHack ) {
- if( $istemplate )
- $headline = $sk->editSectionScriptForOther($templatetitle, $templatesection, $headline);
- else
- $headline = $sk->editSectionScript($this->mTitle, $sectionCount+1,$headline);
- }
-
# give headline the correct <h#> tag
@$head[$headlineCount] .= "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount] .$headline.'</h'.$level.'>';
if( $doShowToc ) {
$toclines = $headlineCount;
- $toc .= $sk->tocUnindent( $toclevel );
- $toc = $sk->tocTable( $toc );
+ $toc .= $sk->tocUnindent( $toclevel - 1 );
+ $toc = $sk->tocList( $toc );
}
# split up and insert constructed headlines
return $text;
}
$text = substr( array_shift( $a ), 1);
- $valid = '0123456789-ABCDEFGHIJKLMNOPQRSTUVWXYZ';
+ $valid = '0123456789-Xx';
foreach ( $a as $x ) {
$isbn = $blank = '' ;
}
$num = str_replace( '-', '', $isbn );
$num = str_replace( ' ', '', $num );
+ $num = str_replace( 'x', 'X', $num );
if ( '' == $num ) {
$text .= "ISBN $blank$x";
return $text;
}
- /**
- * Return an HTML link for the "GEO ..." text
- * @access private
- */
- function magicGEO( $text ) {
- global $wgLang, $wgUseGeoMode;
- $fname = 'Parser::magicGEO';
- wfProfileIn( $fname );
-
- # These next five lines are only for the ~35000 U.S. Census Rambot pages...
- $directions = array ( 'N' => 'North' , 'S' => 'South' , 'E' => 'East' , 'W' => 'West' ) ;
- $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['N']}, (\d+)°(\d+)'(\d+)\" {$directions['W']}/" , "(GEO +\$1.\$2.\$3:-\$4.\$5.\$6)" , $text ) ;
- $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['N']}, (\d+)°(\d+)'(\d+)\" {$directions['E']}/" , "(GEO +\$1.\$2.\$3:+\$4.\$5.\$6)" , $text ) ;
- $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['S']}, (\d+)°(\d+)'(\d+)\" {$directions['W']}/" , "(GEO +\$1.\$2.\$3:-\$4.\$5.\$6)" , $text ) ;
- $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['S']}, (\d+)°(\d+)'(\d+)\" {$directions['E']}/" , "(GEO +\$1.\$2.\$3:+\$4.\$5.\$6)" , $text ) ;
-
- $a = split( 'GEO ', ' '.$text );
- if ( count ( $a ) < 2 ) {
- wfProfileOut( $fname );
- return $text;
- }
- $text = substr( array_shift( $a ), 1);
- $valid = '0123456789.+-:';
-
- foreach ( $a as $x ) {
- $geo = $blank = '' ;
- while ( ' ' == $x{0} ) {
- $blank .= ' ';
- $x = substr( $x, 1 );
- }
- while ( strstr( $valid, $x{0} ) != false ) {
- $geo .= $x{0};
- $x = substr( $x, 1 );
- }
- $num = str_replace( '+', '', $geo );
- $num = str_replace( ' ', '', $num );
-
- if ( '' == $num || count ( explode ( ':' , $num , 3 ) ) < 2 ) {
- $text .= "GEO $blank$x";
- } else {
- $titleObj = Title::makeTitle( NS_SPECIAL, 'Geo' );
- $text .= '<a href="' .
- $titleObj->escapeLocalUrl( 'coordinates='.$num ) .
- "\" class=\"internal\">GEO $geo</a>";
- $text .= $x;
- }
- }
- wfProfileOut( $fname );
- return $text;
- }
-
/**
* Return an HTML link for the "RFC 1234" text
+ *
* @access private
- * @param string $text text to be processed
+ * @param string $text Text to be processed
+ * @param string $keyword Magic keyword to use (default RFC)
+ * @param string $urlmsg Interface message to use (default rfcurl)
+ * @return string
*/
function magicRFC( $text, $keyword='RFC ', $urlmsg='rfcurl' ) {
global $wgLang;
$text .= $keyword.$id.$x;
} else {
/* build the external link*/
- $url = wfmsg( $urlmsg );
- $url = str_replace( '$1', $id, $url);
+ $url = wfMsg( $urlmsg, $id);
$sk =& $this->mOptions->getSkin();
$la = $sk->getExternalLinkAttributes( $url, $keyword.$id );
$text .= "<a href='{$url}'{$la}>{$keyword}{$id}</a>{$x}";
$oldtz = getenv( 'TZ' );
putenv( 'TZ='.$wgLocaltimezone );
}
- /* Note: this is an ugly timezone hack for the European wikis */
- $d = $wgContLang->timeanddate( date( 'YmdHis' ), false ) .
+
+ /* Note: This is the timestamp saved as hardcoded wikitext to
+ * the database, we use $wgContLang here in order to give
+ * everyone the same signiture and use the default one rather
+ * than the one selected in each users preferences.
+ */
+ $d = $wgContLang->timeanddate( wfTimestampNow(), false, false) .
' (' . date( 'T' ) . ')';
if ( isset( $wgLocaltimezone ) ) {
putenv( 'TZ='.$oldtzs );
# Not in the link cache, add it to the query
if ( !isset( $current ) ) {
$current = $val;
- $tables = $page;
- $join = '';
$query = "SELECT page_id, page_namespace, page_title";
if ( $threshold > 0 ) {
- $textTable = $dbr->tableName( 'text' );
- $query .= ', LENGTH(old_text) AS page_len, page_is_redirect';
- $tables .= ", $textTable";
- $join = 'page_latest=old_id AND';
+ $query .= ', page_len, page_is_redirect';
}
- $query .= " FROM $tables WHERE $join (page_namespace=$val AND page_title IN(";
+ $query .= " FROM $page WHERE (page_namespace=$val AND page_title IN(";
} elseif ( $current != $val ) {
$current = $val;
$query .= ")) OR (page_namespace=$val AND page_title IN(";
if ( $threshold > 0 ) {
$size = $s->page_len;
- if ( $s->page_is_redirect || $s->page_namespace != 0 || $length < $threshold ) {
+ if ( $s->page_is_redirect || $s->page_namespace != 0 || $size >= $threshold ) {
$colours[$pdbk] = 1;
} else {
$colours[$pdbk] = 2;
* 'A tree'.
*/
function renderImageGallery( $text ) {
+ # Setup the parser
+ global $wgUser, $wgParser, $wgTitle;
+ $parserOptions = ParserOptions::newFromUser( $wgUser );
+
global $wgLinkCache;
$ig = new ImageGallery();
$ig->setShowBytes( false );
$label = '';
}
- # FIXME: Use the full wiki parser and add its links
- # to the page's links.
- $html = $this->mOptions->mSkin->formatComment( $label );
+ $html = $wgParser->parse( $label , $wgTitle, $parserOptions );
+ $html = $html->mText;
- $ig->add( Image::newFromTitle( $nt ), $html );
+ $ig->add( new Image( $nt ), $html );
$wgLinkCache->addImageLinkObj( $nt );
}
return $ig->toHTML();
var $mText, $mLanguageLinks, $mCategoryLinks, $mContainsOldMagic;
var $mCacheTime; # Used in ParserCache
var $mVersion; # Compatibility check
+ var $mTitleText; # title text of the chosen language variant
function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(),
- $containsOldMagic = false )
+ $containsOldMagic = false, $titletext = '' )
{
$this->mText = $text;
$this->mLanguageLinks = $languageLinks;
$this->mContainsOldMagic = $containsOldMagic;
$this->mCacheTime = '';
$this->mVersion = MW_PARSER_VERSION;
+ $this->mTitleText = $titletext;
}
function getText() { return $this->mText; }
function getLanguageLinks() { return $this->mLanguageLinks; }
function getCategoryLinks() { return array_keys( $this->mCategoryLinks ); }
function getCacheTime() { return $this->mCacheTime; }
+ function getTitleText() { return $this->mTitleText; }
function containsOldMagic() { return $this->mContainsOldMagic; }
function setText( $text ) { return wfSetVar( $this->mText, $text ); }
function setLanguageLinks( $ll ) { return wfSetVar( $this->mLanguageLinks, $ll ); }
function setCategoryLinks( $cl ) { return wfSetVar( $this->mCategoryLinks, $cl ); }
function setContainsOldMagic( $com ) { return wfSetVar( $this->mContainsOldMagic, $com ); }
function setCacheTime( $t ) { return wfSetVar( $this->mCacheTime, $t ); }
+ function setTitleText( $t ) { return wfSetVar ($this->mTitleText, $t); }
+
function addCategoryLink( $c ) { $this->mCategoryLinks[$c] = 1; }
function merge( $other ) {
var $mSkin; # Reference to the preferred skin
var $mDateFormat; # Date format index
var $mEditSection; # Create "edit section" links
- var $mEditSectionOnRightClick; # Generate JavaScript to edit section on right click
var $mNumberHeadings; # Automatically number headings
- var $mShowToc; # Show table of contents
function getUseTeX() { return $this->mUseTeX; }
function getUseDynamicDates() { return $this->mUseDynamicDates; }
function getSkin() { return $this->mSkin; }
function getDateFormat() { return $this->mDateFormat; }
function getEditSection() { return $this->mEditSection; }
- function getEditSectionOnRightClick() { return $this->mEditSectionOnRightClick; }
function getNumberHeadings() { return $this->mNumberHeadings; }
- function getShowToc() { return $this->mShowToc; }
function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX, $x ); }
function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates, $x ); }
function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages, $x ); }
function setDateFormat( $x ) { return wfSetVar( $this->mDateFormat, $x ); }
function setEditSection( $x ) { return wfSetVar( $this->mEditSection, $x ); }
- function setEditSectionOnRightClick( $x ) { return wfSetVar( $this->mEditSectionOnRightClick, $x ); }
function setNumberHeadings( $x ) { return wfSetVar( $this->mNumberHeadings, $x ); }
- function setShowToc( $x ) { return wfSetVar( $this->mShowToc, $x ); }
function setSkin( &$x ) { $this->mSkin =& $x; }
- # Get parser options
- /* static */ function newFromUser( &$user ) {
+ /**
+ * Get parser options
+ * @static
+ */
+ function newFromUser( &$user ) {
$popts = new ParserOptions;
$popts->initialiseFromUser( $user );
return $popts;
}
- # Get user options
+ /** Get user options */
function initialiseFromUser( &$userInput ) {
global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages;
$fname = 'ParserOptions::initialiseFromUser';
wfProfileOut( $fname.'-skin' );
$this->mDateFormat = $user->getOption( 'date' );
$this->mEditSection = $user->getOption( 'editsection' );
- $this->mEditSectionOnRightClick = $user->getOption( 'editsectiononrightclick' );
$this->mNumberHeadings = $user->getOption( 'numberheadings' );
- $this->mShowToc = $user->getOption( 'showtoc' );
wfProfileOut( $fname );
}
-
-
}
/**
}
}
+/**
+ * Escape html tags
+ * Basicly replacing " > and < with HTML entities ( ", >, <)
+ *
+ * @param string $in Text that might contain HTML tags
+ * @return string Escaped string
+ */
function wfEscapeHTMLTagsOnly( $in ) {
return str_replace(
array( '"', '>', '<' ),