#
define( "MAX_INCLUDE_REPEAT", 5 );
+define( "MAX_INCLUDE_SIZE", 1000000 ); // 1 Million
# Allowed values for $mOutputType
define( "OT_HTML", 1 );
# 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" );
+define( 'STRIP_COMMENTS', 'HTMLCommentStrip' );
# prefix for escaping, used in two functions at least
-define( "UNIQ_PREFIX", "NaodW29");
+define( 'UNIQ_PREFIX', 'NaodW29');
+
+
+# Constants needed for external link processing
+
+define( 'URL_PROTOCOLS', 'http|https|ftp|irc|gopher|news|mailto' );
+define( 'HTTP_PROTOCOLS', 'http|https' );
+# Everything except bracket, space, or control characters
+define( 'EXT_LINK_URL_CLASS', '[^]\\x00-\\x20\\x7F]' );
+define( 'INVERSE_EXT_LINK_URL_CLASS', '[\]\\x00-\\x20\\x7F]' );
+# Including space
+define( 'EXT_LINK_TEXT_CLASS', '[^\]\\x00-\\x1F\\x7F]' );
+define( 'EXT_IMAGE_FNAME_CLASS', '[A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]' );
+define( 'EXT_IMAGE_EXTENSIONS', 'gif|png|jpg|jpeg' );
+define( 'EXT_LINK_BRACKETED', '/\[(('.URL_PROTOCOLS.'):'.EXT_LINK_URL_CLASS.'+) *('.EXT_LINK_TEXT_CLASS.'*?)\]/S' );
+define( 'EXT_IMAGE_REGEX',
+ '/^('.HTTP_PROTOCOLS.':)'. # Protocol
+ '('.EXT_LINK_URL_CLASS.'+)\\/'. # Hostname and path
+ '('.EXT_IMAGE_FNAME_CLASS.'+)\\.((?i)'.EXT_IMAGE_EXTENSIONS.')$/S' # Filename
+);
class Parser
{
# Persistent:
var $mTagHooks;
-
+
# Cleared with clearState():
var $mOutput, $mAutonumber, $mDTopen, $mStripState = array();
var $mVariables, $mIncludeCount, $mArgStack, $mLastSection, $mInPre;
# Temporary:
- var $mOptions, $mTitle, $mOutputType;
+ var $mOptions, $mTitle, $mOutputType,
+ $mTemplates, // cache of already loaded templates, avoids
+ // multiple SQL queries for the same string
+ $mTemplatePath; // stores an unsorted hash of all the templates already loaded
+ // in this path. Used for loop detection.
function Parser() {
+ $this->mTemplates = array();
+ $this->mTemplatePath = array();
$this->mTagHooks = array();
$this->clearState();
}
$fixtags = array(
# french spaces, last one Guillemet-left
# only if there is something before the space
- '/(.) (\\?|:|;|!|\\302\\273)/i' => '\\1 \\2',
+ '/(.) (\\?|:|;|!|\\302\\273)/i' => '\\1 \\2',
# french spaces, Guillemet-right
- "/(\\302\\253) /i"=>"\\1 ",
+ "/(\\302\\253) /i"=>"\\1 ",
'/<hr *>/i' => '<hr />',
'/<br *>/i' => '<br />',
'/<center *>/i' => '<div class="center">',
} else {
$fixtags = array(
# french spaces, last one Guillemet-left
- '/ (\\?|:|!|\\302\\273)/i' => ' \\1',
+ '/ (\\?|:|!|\\302\\273)/i' => ' \\1',
# french spaces, Guillemet-right
- '/(\\302\\253) /i' => '\\1 ',
+ '/(\\302\\253) /i' => '\\1 ',
+ '/([^> ]+(0(1|3|9);)[^< ]*)/i' => '<span class="diacrit">\\1</span>',
'/<center *>/i' => '<div class="center">',
'/<\\/center *>/i' => '</div>'
);
# counting the sections in the wikisource
function strip( $text, &$state, $stripcomments = false ) {
$render = ($this->mOutputType == OT_HTML);
+ $html_content = array();
$nowiki_content = array();
$math_content = array();
$pre_content = array();
$comment_content = array();
$ext_content = array();
-
+
# Replace any instances of the placeholders
$uniq_prefix = UNIQ_PREFIX;
#$text = str_replace( $uniq_prefix, wfHtmlEscapeFirst( $uniq_prefix ), $text );
+ # html
+ global $wgRawHtml;
+ if( $wgRawHtml ) {
+ $text = Parser::extractTags('html', $text, $html_content, $uniq_prefix);
+ foreach( $html_content as $marker => $content ) {
+ if ($render ) {
+ # Raw and unchecked for validity.
+ $html_content[$marker] = $content;
+ } else {
+ $html_content[$marker] = "<html>$content</html>";
+ }
+ }
+ }
# nowiki
$text = Parser::extractTags('nowiki', $text, $nowiki_content, $uniq_prefix);
- foreach( $nowiki_content as $marker => $content ){
+ foreach( $nowiki_content as $marker => $content ) {
if( $render ){
$nowiki_content[$marker] = wfEscapeHTMLTagsOnly( $content );
} else {
# Merge state with the pre-existing state, if there is one
if ( $state ) {
+ $state['html'] = $state['html'] + $html_content;
$state['nowiki'] = $state['nowiki'] + $nowiki_content;
$state['math'] = $state['math'] + $math_content;
$state['pre'] = $state['pre'] + $pre_content;
$state['comment'] = $state['comment'] + $comment_content;
-
+
foreach( $ext_content as $tag => $array ) {
if ( array_key_exists( $tag, $state ) ) {
$state[$tag] = $state[$tag] + $array;
}
} else {
$state = array(
+ 'html' => $html_content,
'nowiki' => $nowiki_content,
'math' => $math_content,
'pre' => $pre_content,
# Must expand in reverse order, otherwise nested tags will be corrupted
$contentDict = end( $state );
for ( $contentDict = end( $state ); $contentDict !== false; $contentDict = prev( $state ) ) {
- if( key($state) != 'nowiki') {
+ if( key($state) != 'nowiki' && key($state) != 'html') {
for ( $content = end( $contentDict ); $content !== false; $content = prev( $contentDict ) ) {
$text = str_replace( key( $contentDict ), $content, $text );
}
return $text;
}
- # always call this after unstrip() to preserve the order
+ # always call this after unstrip() to preserve the order
function unstripNoWiki( $text, &$state ) {
# Must expand in reverse order, otherwise nested tags will be corrupted
for ( $content = end($state['nowiki']); $content !== false; $content = prev( $state['nowiki'] ) ) {
$text = str_replace( key( $state['nowiki'] ), $content, $text );
}
+ global $wgRawHtml;
+ if ($wgRawHtml) {
+ for ( $content = end($state['html']); $content !== false; $content = prev( $state['html'] ) ) {
+ $text = str_replace( key( $state['html'] ), $content, $text );
+ }
+ }
+
return $text;
}
$rnd = UNIQ_PREFIX . '-item' . Parser::getRandomString();
if ( !$state ) {
$state = array(
+ 'html' => array(),
'nowiki' => array(),
'math' => array(),
'pre' => array()
# This method generates the list of subcategories and pages for a category
function oldCategoryMagic () {
- global $wgLang , $wgUser ;
+ global $wgLang ;
+ $fname = 'Parser::oldCategoryMagic';
+
if ( !$this->mOptions->getUseCategoryMagic() ) return ; # Doesn't use categories at all
- $cns = Namespace::getCategory() ;
- if ( $this->mTitle->getNamespace() != $cns ) return "" ; # This ain't a category page
+ if ( $this->mTitle->getNamespace() != NS_CATEGORY ) return "" ; # This ain't a category page
$r = "<br style=\"clear:both;\"/>\n";
- $sk =& $wgUser->getSkin() ;
+ $sk =& $this->mOptions->getSkin() ;
$articles = array() ;
$children = array() ;
$id = $this->mTitle->getArticleID() ;
# 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 ;
+ $dbr =& wfGetDB( DB_SLAVE );
+ $cur = $dbr->tableName( 'cur' );
+ $categorylinks = $dbr->tableName( 'categorylinks' );
+
+ $t = $dbr->strencode( $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 = $dbr->query( $sql, $fname ) ;
+ while ( $x = $dbr->fetchObject ( $res ) ) $data[] = $x ;
# For all pages that link to this category
foreach ( $data AS $x )
if ( $t != "" ) $t .= ":" ;
$t .= $x->cur_title ;
- if ( $x->cur_namespace == $cns ) {
+ if ( $x->cur_namespace == NS_CATEGORY ) {
array_push ( $children , $sk->makeLink ( $t ) ) ; # Subcategory
} else {
array_push ( $articles , $sk->makeLink ( $t ) ) ; # Page in this category
}
}
- wfFreeResult ( $res ) ;
+ $dbr->freeResult ( $res ) ;
# Showing subcategories
if ( count ( $children ) > 0 ) {
$r .= implode ( ', ' , $articles ) ;
}
-
return $r ;
}
function newCategoryMagic () {
- global $wgLang , $wgUser ;
+ global $wgLang;
if ( !$this->mOptions->getUseCategoryMagic() ) return ; # Doesn't use categories at all
- $cns = Namespace::getCategory() ;
- if ( $this->mTitle->getNamespace() != $cns ) return '' ; # This ain't a category page
+ if ( $this->mTitle->getNamespace() != NS_CATEGORY ) return '' ; # This ain't a category page
$r = "<br style=\"clear:both;\"/>\n";
- $sk =& $wgUser->getSkin() ;
+ $sk =& $this->mOptions->getSkin() ;
$articles = array() ;
$articles_start_char = array();
$id = $this->mTitle->getArticleID() ;
# FIXME: add limits
- $t = wfStrencode( $this->mTitle->getDBKey() );
- $sql = "SELECT DISTINCT cur_title,cur_namespace,cl_sortkey FROM
-cur,categorylinks WHERE cl_to='$t' AND cl_from=cur_id ORDER BY
-cl_sortkey" ;
- $res = wfQuery ( $sql, DB_READ ) ;
- while ( $x = wfFetchObject ( $res ) )
+ $dbr =& wfGetDB( DB_SLAVE );
+ $cur = $dbr->tableName( 'cur' );
+ $categorylinks = $dbr->tableName( 'categorylinks' );
+
+ $t = $dbr->strencode( $this->mTitle->getDBKey() );
+ $sql = "SELECT DISTINCT cur_title,cur_namespace,cl_sortkey FROM " .
+ "$cur,$categorylinks WHERE cl_to='$t' AND cl_from=cur_id ORDER BY cl_sortkey" ;
+ $res = $dbr->query ( $sql ) ;
+ while ( $x = $dbr->fetchObject ( $res ) )
{
$t = $ns = $wgLang->getNsText ( $x->cur_namespace ) ;
if ( $t != '' ) $t .= ':' ;
$t .= $x->cur_title ;
-
- if ( $x->cur_namespace == $cns ) {
+
+ if ( $x->cur_namespace == NS_CATEGORY ) {
$ctitle = str_replace( '_',' ',$x->cur_title );
array_push ( $children, $sk->makeKnownLink ( $t, $ctitle ) ) ; # Subcategory
array_push ( $children_start_char, $wgLang->firstChar( $x->cl_sortkey ) ) ;
}
} else {
- array_push ( $articles , $sk->makeLink ( $t ) ) ; # Page in this category
+ array_push ( $articles , $sk->makeKnownLink ( $t ) ) ; # Page in this category
array_push ( $articles_start_char, $wgLang->firstChar( $x->cl_sortkey ) ) ;
}
}
- wfFreeResult ( $res ) ;
+ $dbr->freeResult ( $res ) ;
$ti = $this->mTitle->getText() ;
$r .= '<h2>' . wfMsg( 'subcategories' ) . "</h2>\n"
. wfMsg( 'subcategorycount', count( $children ) );
if ( count ( $children ) > 6 ) {
-
+
// divide list into three equal chunks
$chunk = (int) (count ( $children ) / 3);
// get and display header
$r .= '<table width="100%"><tr valign="top">';
-
+
$startChunk = 0;
$endChunk = $chunk;
-
+
// loop through the chunks
- for($startChunk = 0, $endChunk = $chunk, $chunkIndex = 0;
+ for($startChunk = 0, $endChunk = $chunk, $chunkIndex = 0;
$chunkIndex < 3;
$chunkIndex++, $startChunk = $endChunk, $endChunk += $chunk + 1)
{
-
+
$r .= '<td><ul>';
// output all subcategories to category
for ($index = $startChunk ;
{
$r .= "</ul><h3>{$children_start_char[$index]}</h3>\n<ul>";
}
-
+
$r .= "<li>{$children[$index]}</li>";
}
$r .= '</ul></td>';
-
-
+
+
}
$r .= '</tr></table>';
} else {
// for short lists of subcategories to category.
-
+
$r .= "<h3>{$children_start_char[0]}</h3>\n";
$r .= '<ul><li>'.$children[0].'</li>';
for ($index = 1; $index < count($children); $index++ )
{
$r .= "</ul><h3>{$children_start_char[$index]}</h3>\n<ul>";
}
-
+
$r .= "<li>{$children[$index]}</li>";
}
$r .= '</ul>';
# Showing articles in this category
if ( count ( $articles ) > 6) {
$ti = $this->mTitle->getText() ;
-
+
// divide list into three equal chunks
$chunk = (int) (count ( $articles ) / 3);
-
+
// get and display header
$r .= '<table width="100%"><tr valign="top">';
-
+
// loop through the chunks
- for($startChunk = 0, $endChunk = $chunk, $chunkIndex = 0;
+ for($startChunk = 0, $endChunk = $chunk, $chunkIndex = 0;
$chunkIndex < 3;
$chunkIndex++, $startChunk = $endChunk, $endChunk += $chunk + 1)
{
-
+
$r .= '<td><ul>';
-
+
// output all articles in category
for ($index = $startChunk ;
$index < $endChunk && $index < count($articles);
{
$r .= "</ul><h3>{$articles_start_char[$index]}</h3>\n<ul>";
}
-
+
$r .= "<li>{$articles[$index]}</li>";
}
$r .= '</ul></td>';
-
-
+
+
}
$r .= '</tr></table>';
} elseif ( count ( $articles ) > 0) {
// for short lists of articles in categories.
$ti = $this->mTitle->getText() ;
-
+
$r .= '<h3>'.$articles_start_char[0]."</h3>\n";
$r .= '<ul><li>'.$articles[0].'</li>';
for ($index = 1; $index < count($articles); $index++ )
{
$r .= "</ul><h3>{$articles_start_char[$index]}</h3>\n<ul>";
}
-
+
$r .= "<li>{$articles[$index]}</li>";
}
$r .= '</ul>';
'/(\\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
global $wgInputEncoding, $wgOutputEncoding;
$fname = 'Parser::tidy';
wfProfileIn( $fname );
-
+
$cleansource = '';
switch(strtoupper($wgOutputEncoding)) {
case 'ISO-8859-1':
}
wfProfileOut( $fname );
-
+
if( $cleansource == '' && $text != '') {
wfDebug( "Tidy error detected!\n" );
return $text . "\n<!-- Tidy found serious XHTML errors -->\n";
# parse the wiki syntax used to render tables
function doTableStuff ( $t ) {
+ $fname = 'Parser::doTableStuff';
+ wfProfileIn( $fname );
+
$t = explode ( "\n" , $t ) ;
$td = array () ; # Is currently a td tag open?
- $ltd = array () ; # Was it TD or TH?
- $tr = array () ; # Is currently a tr tag open?
- $ltr = array () ; # tr attributes
- foreach ( $t AS $k => $x )
+ $ltd = array () ; # Was it TD or TH?
+ $tr = array () ; # Is currently a tr tag open?
+ $ltr = array () ; # tr attributes
+ $indent_level = 0; # indent level of the table
+ foreach ( $t AS $k => $x )
+ {
+ $x = trim ( $x ) ;
+ $fc = substr ( $x , 0 , 1 ) ;
+ if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) )
{
- $x = trim ( $x ) ;
- $fc = substr ( $x , 0 , 1 ) ;
- if ( '{|' == substr ( $x , 0 , 2 ) )
- {
- $t[$k] = "\n<table " . $this->fixTagAttributes ( substr ( $x , 3 ) ) . '>' ;
- array_push ( $td , false ) ;
- array_push ( $ltd , '' ) ;
- array_push ( $tr , false ) ;
- array_push ( $ltr , '' ) ;
- }
- else if ( count ( $td ) == 0 ) { } # Don't do any of the following
- else if ( '|}' == substr ( $x , 0 , 2 ) )
- {
- $z = "</table>\n" ;
- $l = array_pop ( $ltd ) ;
- if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
- if ( array_pop ( $td ) ) $z = "</{$l}>" . $z ;
- array_pop ( $ltr ) ;
- $t[$k] = $z ;
- }
- /* else if ( "|_" == substr ( $x , 0 , 2 ) ) # Caption
- {
- $z = trim ( substr ( $x , 2 ) ) ;
- $t[$k] = "<caption>{$z}</caption>\n" ;
- }*/
- else if ( '|-' == substr ( $x , 0 , 2 ) ) # Allows for |---------------
+ $indent_level = strlen( $matches[1] );
+ $t[$k] = "\n" .
+ str_repeat( "<dl><dd>", $indent_level ) .
+ "<table " . $this->fixTagAttributes ( $matches[2] ) . '>' ;
+ array_push ( $td , false ) ;
+ array_push ( $ltd , '' ) ;
+ array_push ( $tr , false ) ;
+ array_push ( $ltr , '' ) ;
+ }
+ else if ( count ( $td ) == 0 ) { } # Don't do any of the following
+ else if ( '|}' == substr ( $x , 0 , 2 ) )
+ {
+ $z = "</table>\n" ;
+ $l = array_pop ( $ltd ) ;
+ if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
+ if ( array_pop ( $td ) ) $z = "</{$l}>" . $z ;
+ array_pop ( $ltr ) ;
+ $t[$k] = $z . str_repeat( "</dd></dl>", $indent_level );
+ }
+ else if ( '|-' == substr ( $x , 0 , 2 ) ) # Allows for |---------------
+ {
+ $x = substr ( $x , 1 ) ;
+ while ( $x != '' && substr ( $x , 0 , 1 ) == '-' ) $x = substr ( $x , 1 ) ;
+ $z = '' ;
+ $l = array_pop ( $ltd ) ;
+ if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
+ if ( array_pop ( $td ) ) $z = "</{$l}>" . $z ;
+ array_pop ( $ltr ) ;
+ $t[$k] = $z ;
+ array_push ( $tr , false ) ;
+ array_push ( $td , false ) ;
+ array_push ( $ltd , '' ) ;
+ array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
+ }
+ else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) # Caption
+ {
+ if ( '|+' == substr ( $x , 0 , 2 ) )
{
+ $fc = '+' ;
$x = substr ( $x , 1 ) ;
- while ( $x != '' && substr ( $x , 0 , 1 ) == '-' ) $x = substr ( $x , 1 ) ;
- $z = '' ;
- $l = array_pop ( $ltd ) ;
- if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
- if ( array_pop ( $td ) ) $z = "</{$l}>" . $z ;
- array_pop ( $ltr ) ;
- $t[$k] = $z ;
- array_push ( $tr , false ) ;
- array_push ( $td , false ) ;
- array_push ( $ltd , '' ) ;
- array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
}
- else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) # Caption
+ $after = substr ( $x , 1 ) ;
+ if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;
+ $after = explode ( '||' , $after ) ;
+ $t[$k] = '' ;
+ foreach ( $after AS $theline )
{
- if ( '|+' == substr ( $x , 0 , 2 ) )
+ $z = '' ;
+ if ( $fc != '+' )
{
- $fc = '+' ;
- $x = substr ( $x , 1 ) ;
+ $tra = array_pop ( $ltr ) ;
+ if ( !array_pop ( $tr ) ) $z = "<tr {$tra}>\n" ;
+ array_push ( $tr , true ) ;
+ array_push ( $ltr , '' ) ;
}
- $after = substr ( $x , 1 ) ;
- if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;
- $after = explode ( '||' , $after ) ;
- $t[$k] = '' ;
- foreach ( $after AS $theline )
- {
- $z = '' ;
- if ( $fc != '+' )
- {
- $tra = array_pop ( $ltr ) ;
- if ( !array_pop ( $tr ) ) $z = "<tr {$tra}>\n" ;
- array_push ( $tr , true ) ;
- array_push ( $ltr , '' ) ;
- }
- $l = array_pop ( $ltd ) ;
- if ( array_pop ( $td ) ) $z = "</{$l}>" . $z ;
- if ( $fc == '|' ) $l = 'td' ;
- else if ( $fc == '!' ) $l = 'th' ;
- else if ( $fc == '+' ) $l = 'caption' ;
- else $l = '' ;
- array_push ( $ltd , $l ) ;
- $y = explode ( '|' , $theline , 2 ) ;
- if ( count ( $y ) == 1 ) $y = "{$z}<{$l}>{$y[0]}" ;
- else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
- $t[$k] .= $y ;
- array_push ( $td , true ) ;
- }
+ $l = array_pop ( $ltd ) ;
+ if ( array_pop ( $td ) ) $z = "</{$l}>" . $z ;
+ if ( $fc == '|' ) $l = 'td' ;
+ else if ( $fc == '!' ) $l = 'th' ;
+ else if ( $fc == '+' ) $l = 'caption' ;
+ else $l = '' ;
+ array_push ( $ltd , $l ) ;
+ $y = explode ( '|' , $theline , 2 ) ;
+ if ( count ( $y ) == 1 ) $y = "{$z}<{$l}>{$y[0]}" ;
+ else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
+ $t[$k] .= $y ;
+ array_push ( $td , true ) ;
}
}
+ }
# Closing open td, tr && table
while ( count ( $td ) > 0 )
$t = implode ( "\n" , $t ) ;
# $t = $this->removeHTMLtags( $t );
+ wfProfileOut( $fname );
return $t ;
}
# Parses the text and adds the result to the strip state
# Returns the strip tag
- function stripParse( $text, $newline, $args )
+ function stripParse( $text, $newline, $args )
{
$text = $this->strip( $text, $this->mStripState );
$text = $this->internalParse( $text, (bool)$newline, $args, false );
return $newline.$this->insertStripItem( $text, $this->mStripState );
}
-
+
function internalParse( $text, $linestart, $args = array(), $isMain=true ) {
$fname = 'Parser::internalParse';
wfProfileIn( $fname );
$text = $wgDateFormatter->reformat( $this->mOptions->getDateFormat(), $text );
}
$text = $this->doAllQuotes( $text );
- // $text = $this->doExponent( $text );
$text = $this->replaceExternalLinks( $text );
+ $text = $this->doMagicLinks( $text );
$text = $this->replaceInternalLinks ( $text );
$text = $this->replaceInternalLinks ( $text );
- //$text = $this->doTokenizedParser ( $text );
- $text = $this->doTableStuff ( $text ) ;
- $text = $this->magicISBN( $text );
- $text = $this->magicRFC( $text );
+
+ $text = $this->unstrip( $text, $this->mStripState );
+ $text = $this->unstripNoWiki( $text, $this->mStripState );
+
+ $text = $this->doTableStuff( $text );
$text = $this->formatHeadings( $text, $isMain );
$sk =& $this->mOptions->getSkin();
$text = $sk->transformContent( $text );
- if ( !isset ( $this->categoryMagicDone ) ) {
+ if ( $isMain && !isset ( $this->categoryMagicDone ) ) {
$text .= $this->categoryMagic () ;
$this->categoryMagicDone = true ;
}
-
+
wfProfileOut( $fname );
return $text;
}
-
+
+ /* private */ function &doMagicLinks( &$text ) {
+ $text = $this->magicISBN( $text );
+ $text = $this->magicGEO( $text );
+ $text = $this->magicRFC( $text );
+ return $text;
+ }
+
# Parse ^^ tokens and return html
/* private */ function doExponent ( $text )
{
$outtext = '';
$lines = explode( "\n", $text );
foreach ( $lines as $line ) {
- $outtext .= $this->doQuotes ( '', $line, '' ) . "\n";
+ $outtext .= $this->doQuotes ( $line ) . "\n";
}
$outtext = substr($outtext, 0,-1);
wfProfileOut( $fname );
return $outtext;
}
- /* private */ function doQuotes( $pre, $text, $mode ) {
- if ( preg_match( "/^(.*)''(.*)$/sU", $text, $m ) ) {
- $m1_strong = ($m[1] == "") ? "" : "<strong>{$m[1]}</strong>";
- $m1_em = ($m[1] == "") ? "" : "<em>{$m[1]}</em>";
- if ( substr ($m[2], 0, 1) == '\'' ) {
- $m[2] = substr ($m[2], 1);
- if ($mode == 'em') {
- return $this->doQuotes ( $m[1], $m[2], ($m[1] == '') ? 'both' : 'emstrong' );
- } else if ($mode == 'strong') {
- return $m1_strong . $this->doQuotes ( '', $m[2], '' );
- } else if (($mode == 'emstrong') || ($mode == 'both')) {
- return $this->doQuotes ( '', $pre.$m1_strong.$m[2], 'em' );
- } else if ($mode == 'strongem') {
- return "<strong>{$pre}{$m1_em}</strong>" . $this->doQuotes ( '', $m[2], 'em' );
- } else {
- return $m[1] . $this->doQuotes ( '', $m[2], 'strong' );
- }
- } else {
- if ($mode == 'strong') {
- return $this->doQuotes ( $m[1], $m[2], ($m[1] == '') ? 'both' : 'strongem' );
- } else if ($mode == 'em') {
- return $m1_em . $this->doQuotes ( '', $m[2], '' );
- } else if ($mode == 'emstrong') {
- return "<em>{$pre}{$m1_strong}</em>" . $this->doQuotes ( '', $m[2], 'strong' );
- } else if (($mode == 'strongem') || ($mode == 'both')) {
- return $this->doQuotes ( '', $pre.$m1_em.$m[2], 'strong' );
- } else {
- return $m[1] . $this->doQuotes ( '', $m[2], 'em' );
+ /* private */ function doQuotes( $text ) {
+ $arr = preg_split ("/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE);
+ if (count ($arr) == 1)
+ return $text;
+ else
+ {
+ # First, do some preliminary work. This may shift some apostrophes from
+ # being mark-up to being text. It also counts the number of occurrences
+ # of bold and italics mark-ups.
+ $i = 0;
+ $numbold = 0;
+ $numitalics = 0;
+ foreach ($arr as $r)
+ {
+ if (($i % 2) == 1)
+ {
+ # If there are ever four apostrophes, assume the first is supposed to
+ # be text, and the remaining three constitute mark-up for bold text.
+ if (strlen ($arr[$i]) == 4)
+ {
+ $arr[$i-1] .= "'";
+ $arr[$i] = "'''";
+ }
+ # If there are more than 5 apostrophes in a row, assume they're all
+ # text except for the last 5.
+ else if (strlen ($arr[$i]) > 5)
+ {
+ $arr[$i-1] .= str_repeat ("'", strlen ($arr[$i]) - 5);
+ $arr[$i] = "'''''";
+ }
+ # Count the number of occurrences of bold and italics mark-ups.
+ # We are not counting sequences of five apostrophes.
+ if (strlen ($arr[$i]) == 2) $numitalics++; else
+ if (strlen ($arr[$i]) == 3) $numbold++;
}
+ $i++;
}
- } else {
- $text_strong = ($text == '') ? '' : "<strong>{$text}</strong>";
- $text_em = ($text == '') ? '' : "<em>{$text}</em>";
- if ($mode == '') {
- return $pre . $text;
- } else if ($mode == 'em') {
- return $pre . $text_em;
- } else if ($mode == 'strong') {
- return $pre . $text_strong;
- } else if ($mode == 'strongem') {
- return (($pre == '') && ($text == '')) ? '' : "<strong>{$pre}{$text_em}</strong>";
- } else {
- return (($pre == '') && ($text == '')) ? '' : "<em>{$pre}{$text_strong}</em>";
+
+ # If there is an odd number of both bold and italics, it is likely
+ # that one of the bold ones was meant to be an apostrophe followed
+ # by italics. Which one we cannot know for certain, but it is more
+ # likely to be one that has a single-letter word before it.
+ if (($numbold % 2 == 1) && ($numitalics % 2 == 1))
+ {
+ $i = 0;
+ $firstsingleletterword = -1;
+ $firstmultiletterword = -1;
+ $firstspace = -1;
+ foreach ($arr as $r)
+ {
+ if (($i % 2 == 1) and (strlen ($r) == 3))
+ {
+ $x1 = substr ($arr[$i-1], -1);
+ $x2 = substr ($arr[$i-1], -2, 1);
+ if ($x1 == " ") {
+ if ($firstspace == -1) $firstspace = $i;
+ } else if ($x2 == " ") {
+ if ($firstsingleletterword == -1) $firstsingleletterword = $i;
+ } else {
+ if ($firstmultiletterword == -1) $firstmultiletterword = $i;
+ }
+ }
+ $i++;
+ }
+
+ # If there is a single-letter word, use it!
+ if ($firstsingleletterword > -1)
+ {
+ $arr [ $firstsingleletterword ] = "''";
+ $arr [ $firstsingleletterword-1 ] .= "'";
+ }
+ # If not, but there's a multi-letter word, use that one.
+ else if ($firstmultiletterword > -1)
+ {
+ $arr [ $firstmultiletterword ] = "''";
+ $arr [ $firstmultiletterword-1 ] .= "'";
+ }
+ # ... otherwise use the first one that has neither.
+ else
+ {
+ $arr [ $firstspace ] = "''";
+ $arr [ $firstspace-1 ] .= "'";
+ }
}
+
+ # Now let's actually convert our apostrophic mush to HTML!
+ $output = '';
+ $buffer = '';
+ $state = '';
+ $i = 0;
+ foreach ($arr as $r)
+ {
+ if (($i % 2) == 0)
+ {
+ if ($state == 'both')
+ $buffer .= $r;
+ else
+ $output .= $r;
+ }
+ else
+ {
+ if (strlen ($r) == 2)
+ {
+ if ($state == 'em')
+ { $output .= "</em>"; $state = ''; }
+ else if ($state == 'strongem')
+ { $output .= "</em>"; $state = 'strong'; }
+ else if ($state == 'emstrong')
+ { $output .= "</strong></em><strong>"; $state = 'strong'; }
+ else if ($state == 'both')
+ { $output .= "<strong><em>{$buffer}</em>"; $state = 'strong'; }
+ else # $state can be 'strong' or ''
+ { $output .= "<em>"; $state .= 'em'; }
+ }
+ else if (strlen ($r) == 3)
+ {
+ if ($state == 'strong')
+ { $output .= "</strong>"; $state = ''; }
+ else if ($state == 'strongem')
+ { $output .= "</em></strong><em>"; $state = 'em'; }
+ else if ($state == 'emstrong')
+ { $output .= "</strong>"; $state = 'em'; }
+ else if ($state == 'both')
+ { $output .= "<em><strong>{$buffer}</strong>"; $state = 'em'; }
+ else # $state can be 'em' or ''
+ { $output .= "<strong>"; $state .= 'strong'; }
+ }
+ else if (strlen ($r) == 5)
+ {
+ if ($state == 'strong')
+ { $output .= "</strong><em>"; $state = 'em'; }
+ else if ($state == 'em')
+ { $output .= "</em><strong>"; $state = 'strong'; }
+ else if ($state == 'strongem')
+ { $output .= "</em></strong>"; $state = ''; }
+ else if ($state == 'emstrong')
+ { $output .= "</strong></em>"; $state = ''; }
+ else if ($state == 'both')
+ { $output .= "<em><strong>{$buffer}</strong></em>"; $state = ''; }
+ else # ($state == '')
+ { $buffer = ''; $state = 'both'; }
+ }
+ }
+ $i++;
+ }
+ # Now close all remaining tags. Notice that the order is important.
+ if ($state == 'strong' || $state == 'emstrong')
+ $output .= "</strong>";
+ if ($state == 'em' || $state == 'strongem' || $state == 'emstrong')
+ $output .= "</em>";
+ if ($state == 'strongem')
+ $output .= "</strong>";
+ if ($state == 'both')
+ $output .= "<strong><em>{$buffer}</em></strong>";
+ return $output;
}
}
/* private */ function replaceExternalLinks( $text ) {
$fname = 'Parser::replaceExternalLinks';
wfProfileIn( $fname );
- $text = $this->subReplaceExternalLinks( $text, 'http', true );
- $text = $this->subReplaceExternalLinks( $text, 'https', true );
- $text = $this->subReplaceExternalLinks( $text, 'ftp', false );
- $text = $this->subReplaceExternalLinks( $text, 'irc', false );
- $text = $this->subReplaceExternalLinks( $text, 'gopher', false );
- $text = $this->subReplaceExternalLinks( $text, 'news', false );
- $text = $this->subReplaceExternalLinks( $text, 'mailto', false );
- wfProfileOut( $fname );
- return $text;
- }
- /* private */ function subReplaceExternalLinks( $s, $protocol, $autonumber ) {
- $unique = '4jzAfzB8hNvf4sqyO9Edd8pSmk9rE2in0Tgw3';
- $uc = "A-Za-z0-9_\\/~%\\-+&*#?!=()@\\x80-\\xFF";
-
- # this is the list of separators that should be ignored if they
- # are the last character of an URL but that should be included
- # if they occur within the URL, e.g. "go to www.foo.com, where .."
- # in this case, the last comma should not become part of the URL,
- # but in "www.foo.com/123,2342,32.htm" it should.
- $sep = ",;\.:";
- $fnc = 'A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF';
- $images = 'gif|png|jpg|jpeg';
-
- # PLEASE NOTE: The curly braces { } are not part of the regex,
- # they are interpreted as part of the string (used to tell PHP
- # that the content of the string should be inserted there).
- $e1 = "/(^|[^\\[])({$protocol}:)([{$uc}{$sep}]+)\\/([{$fnc}]+)\\." .
- "((?i){$images})([^{$uc}]|$)/";
-
- $e2 = "/(^|[^\\[])({$protocol}:)(([".$uc."]|[".$sep."][".$uc."])+)([^". $uc . $sep. "]|[".$sep."]|$)/";
$sk =& $this->mOptions->getSkin();
+ $linktrail = wfMsg('linktrail');
+ $bits = preg_split( EXT_LINK_BRACKETED, $text, -1, PREG_SPLIT_DELIM_CAPTURE );
- if ( $autonumber and $this->mOptions->getAllowExternalImages() ) { # Use img tags only for HTTP urls
- $s = preg_replace( $e1, '\\1' . $sk->makeImage( "{$unique}:\\3" .
- '/\\4.\\5', '\\4.\\5' ) . '\\6', $s );
- }
- $s = preg_replace( $e2, '\\1' . "<a href=\"{$unique}:\\3\"" .
- $sk->getExternalLinkAttributes( "{$unique}:\\3", wfEscapeHTML(
- "{$unique}:\\3" ) ) . ">" . wfEscapeHTML( "{$unique}:\\3" ) .
- '</a>\\5', $s );
- $s = str_replace( $unique, $protocol, $s );
-
- $a = explode( "[{$protocol}:", " " . $s );
- $s = array_shift( $a );
- $s = substr( $s, 1 );
-
- $e1 = "/^([{$uc}"."{$sep}]+)](.*)\$/sD";
- $e2 = "/^([{$uc}"."{$sep}]+)\\s+([^\\]]+)](.*)\$/sD";
+ $s = $this->replaceFreeExternalLinks( array_shift( $bits ) );
- foreach ( $a as $line ) {
- if ( preg_match( $e1, $line, $m ) ) {
- $link = "{$protocol}:{$m[1]}";
- $trail = $m[2];
- if ( $autonumber ) { $text = "[" . ++$this->mAutonumber . "]"; }
- else { $text = wfEscapeHTML( $link ); }
- } else if ( preg_match( $e2, $line, $m ) ) {
- $link = "{$protocol}:{$m[1]}";
- $text = $m[2];
- $trail = $m[3];
+ $i = 0;
+ while ( $i<count( $bits ) ) {
+ $url = $bits[$i++];
+ $protocol = $bits[$i++];
+ $text = $bits[$i++];
+ $trail = $bits[$i++];
+
+ # If the link text is an image URL, replace it with an <img> tag
+ # This happened by accident in the original parser, but some people used it extensively
+ $img = $this->maybeMakeImageLink( $text );
+ if ( $img !== false ) {
+ $text = $img;
+ }
+
+ $dtrail = '';
+
+ # No link text, e.g. [http://domain.tld/some.link]
+ if ( $text == '' ) {
+ # Autonumber if allowed
+ if ( strpos( HTTP_PROTOCOLS, $protocol ) !== false ) {
+ $text = "[" . ++$this->mAutonumber . "]";
+ } else {
+ # Otherwise just use the URL
+ $text = wfEscapeHTML( $url );
+ }
} else {
- $s .= "[{$protocol}:" . $line;
- continue;
+ # Have link text, e.g. [http://domain.tld/some.link text]s
+ # Check for trail
+ if ( preg_match( $linktrail, $trail, $m2 ) ) {
+ $dtrail = $m2[1];
+ $trail = $m2[2];
+ }
}
- if( $link == $text || preg_match( "!$protocol://" . preg_quote( $text, "/" ) . "/?$!", $link ) ) {
+
+ $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
- $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $link ) . "</i>)</span>";
+ $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $encUrl ) . "</i>)</span>";
}
- $la = $sk->getExternalLinkAttributes( $link, $text );
- $s .= "<a href='{$link}'{$la}>{$text}</a>{$paren}{$trail}";
+ # Process the trail (i.e. everything after this link up until start of the next link),
+ # replacing any non-bracketed links
+ $trail = $this->replaceFreeExternalLinks( $trail );
+
+ $la = $sk->getExternalLinkAttributes( $url, $text );
+
+ # Use the encoded URL
+ # 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
+ $s .= "<a href=\"{$url}\" {$la}>{$text}</a>{$dtrail}{$paren}{$trail}";
}
+
+ wfProfileOut( $fname );
return $s;
}
+ # Replace anything that looks like a URL with a link
+ function replaceFreeExternalLinks( $text ) {
+ $bits = preg_split( '/((?:'.URL_PROTOCOLS.'):)/', $text, -1, PREG_SPLIT_DELIM_CAPTURE );
+ $s = array_shift( $bits );
+ $i = 0;
+
+ $sk =& $this->mOptions->getSkin();
+
+ while ( $i < count( $bits ) ){
+ $protocol = $bits[$i++];
+ $remainder = $bits[$i++];
+
+ if ( preg_match( '/^('.EXT_LINK_URL_CLASS.'+)(.*)$/s', $remainder, $m ) ) {
+ # Found some characters after the protocol that look promising
+ $url = $protocol . $m[1];
+ $trail = $m[2];
+
+ # Move trailing punctuation to $trail
+ $sep = ',;\.:!?';
+ # If there is no left bracket, then consider right brackets fair game too
+ if ( strpos( $url, '(' ) === false ) {
+ $sep .= ')';
+ }
+
+ $numSepChars = strspn( strrev( $url ), $sep );
+ if ( $numSepChars ) {
+ $trail = substr( $url, -$numSepChars ) . $trail;
+ $url = substr( $url, 0, -$numSepChars );
+ }
+
+ # Replace & from obsolete syntax with &
+ $url = str_replace( '&', '&', $url );
+
+ # Is this an external image?
+ $text = $this->maybeMakeImageLink( $url );
+ if ( $text === false ) {
+ # Not an image, make a link
+ $text = $sk->makeExternalLink( $url, $url );
+ }
+ $s .= $text . $trail;
+ } else {
+ $s .= $protocol . $remainder;
+ }
+ }
+ return $s;
+ }
+
+ function maybeMakeImageLink( $url ) {
+ $sk =& $this->mOptions->getSkin();
+ $text = false;
+ if ( $this->mOptions->getAllowExternalImages() ) {
+ if ( preg_match( EXT_IMAGE_REGEX, $url ) ) {
+ # Image found
+ $text = $sk->makeImage( htmlspecialchars( $url ) );
+ }
+ }
+ return $text;
+ }
/* private */ function replaceInternalLinks( $s ) {
global $wgLang, $wgLinkCache;
if ( !$tc ) { $tc = Title::legalChars() . '#%'; }
$sk =& $this->mOptions->getSkin();
+ $redirect = MagicWord::get ( MAG_REDIRECT ) ;
+
$a = explode( '[[', ' ' . $s );
$s = array_shift( $a );
$s = substr( $s, 1 );
$useLinkPrefixExtension = $wgLang->linkPrefixExtension();
# Special and Media are pseudo-namespaces; no pages actually exist in them
- static $image = FALSE;
- static $special = FALSE;
- static $media = FALSE;
- static $category = FALSE;
- if ( !$image ) { $image = Namespace::getImage(); }
- if ( !$special ) { $special = Namespace::getSpecial(); }
- if ( !$media ) { $media = Namespace::getMedia(); }
- if ( !$category ) { $category = Namespace::getCategory(); }
$nottalk = !Namespace::isTalk( $this->mTitle->getNamespace() );
}
}
wfProfileOut( $fname.'-prefixhandling' );
-
+
if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
$text = $m[2];
# fix up urlencoded title texts
$iw = $nt->getInterWiki();
if( $noforce ) {
if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgLang->getLanguageName( $iw ) ) {
- array_push( $this->mOutput->mLanguageLinks, $nt->getPrefixedText() );
+ array_push( $this->mOutput->mLanguageLinks, $nt->getFullText() );
$tmp = $prefix . $trail ;
$s .= (trim($tmp) == '')? '': $tmp;
continue;
}
- if ( $ns == $image ) {
+ if ( $ns == NS_IMAGE ) {
$s .= $prefix . $sk->makeImageLinkObj( $nt, $text ) . $trail;
$wgLinkCache->addImageLinkObj( $nt );
continue;
}
- if ( $ns == $category ) {
+ if ( $ns == NS_CATEGORY ) {
$t = $nt->getText() ;
- $nnt = Title::newFromText ( Namespace::getCanonicalName($category).":".$t ) ;
+ $nnt = Title::newFromText ( Namespace::getCanonicalName(NS_CATEGORY).":".$t ) ;
$wgLinkCache->suspend(); # Don't save in links/brokenlinks
+ $pPLC=$sk->postParseLinkColour();
+ $sk->postParseLinkColour( false );
$t = $sk->makeLinkObj( $nnt, $t, '', '' , $prefix );
+ $sk->postParseLinkColour( $pPLC );
$wgLinkCache->resume();
$sortkey = $wasblank ? $this->mTitle->getPrefixedText() : $text;
continue;
}
- if( $ns == $media ) {
+ if( $ns == NS_MEDIA ) {
$s .= $prefix . $sk->makeMediaLinkObj( $nt, $text ) . $trail;
$wgLinkCache->addImageLinkObj( $nt );
continue;
- } elseif( $ns == $special ) {
+ } elseif( $ns == NS_SPECIAL ) {
$s .= $prefix . $sk->makeKnownLinkObj( $nt, $text, '', $trail );
continue;
}
/* 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.
$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 ) {
# Multiple prefixes may abut each other for nested lists.
$prefixLength = strspn( $oLine, '*#:;' );
$pref = substr( $oLine, 0, $prefixLength );
-
+
# eh?
$pref2 = str_replace( ';', ':', $pref );
$t = substr( $oLine, $prefixLength );
+ $this->mInPre = !empty($preOpenMatch);
} else {
# Don't interpret any other prefixes in preformatted text
$prefixLength = 0;
$paragraphStack = false;
$output .= $this->closeParagraph();
if($preOpenMatch and !$preCloseMatch) {
- $this->mInPre = true;
+ $this->mInPre = true;
}
if ( $closematch ) {
$inBlockElem = false;
$inBlockElem = true;
}
} else if ( !$inBlockElem && !$this->mInPre ) {
- if ( " " == $t{0} and trim($t) != '' ) {
+ if ( " " == $t{0} and ( $this->mLastSection == 'pre' or trim($t) != '' ) ) {
// pre
if ($this->mLastSection != 'pre') {
$paragraphStack = false;
switch ( $index ) {
case MAG_CURRENTMONTH:
- return date( 'm' );
+ return $wgLang->formatNum( date( 'm' ) );
case MAG_CURRENTMONTHNAME:
return $wgLang->getMonthName( date('n') );
case MAG_CURRENTMONTHNAMEGEN:
return $wgLang->getMonthNameGen( date('n') );
case MAG_CURRENTDAY:
- return date('j');
+ return $wgLang->formatNum( date('j') );
case MAG_PAGENAME:
return $this->mTitle->getText();
case MAG_NAMESPACE:
case MAG_CURRENTDAYNAME:
return $wgLang->getWeekdayName( date('w')+1 );
case MAG_CURRENTYEAR:
- return date( 'Y' );
+ return $wgLang->formatNum( date( 'Y' ) );
case MAG_CURRENTTIME:
return $wgLang->time( wfTimestampNow(), false );
case MAG_NUMBEROFARTICLES:
- return wfNumberOfArticles();
+ return $wgLang->formatNum( wfNumberOfArticles() );
case MAG_SITENAME:
return $wgSitename;
case MAG_SERVER:
/* private */ function replaceVariables( $text, $args = array() ) {
global $wgLang, $wgScript, $wgArticlePath;
+ # Prevent too big inclusions
+ if(strlen($text)> MAX_INCLUDE_SIZE)
+ return $text;
+
$fname = 'Parser::replaceVariables';
wfProfileIn( $fname );
$bail = false;
- if ( !$this->mVariables ) {
- $this->initialiseVariables();
- }
$titleChars = Title::legalChars();
$nonBraceChars = str_replace( array( '{', '}' ), array( '', '' ), $titleChars );
# PHP global rebinding syntax is a bit weird, need to use the GLOBALS array
$GLOBALS['wgCurParser'] =& $this;
-
+
if ( $this->mOutputType == OT_HTML ) {
# Variable substitution
$text = preg_replace_callback( "/{{([$nonBraceChars]*?)}}/", 'wfVariableSubstitution', $text );
-
+
# Argument substitution
$text = preg_replace_callback( "/(\\n?){{{([$titleChars]*?)}}}/", 'wfArgSubstitution', $text );
}
}
function variableSubstitution( $matches ) {
+ if ( !$this->mVariables ) {
+ $this->initialiseVariables();
+ }
if ( array_key_exists( $matches[1], $this->mVariables ) ) {
$text = $this->mVariables[$matches[1]];
$this->mOutput->mContainsOldMagic = true;
return $text;
}
+ # Split template arguments
+ function getTemplateArgs( $argsString ) {
+ if ( $argsString === '' ) {
+ return array();
+ }
+
+ $args = explode( '|', substr( $argsString, 1 ) );
+
+ # If any of the arguments contains a '[[' but no ']]', it needs to be
+ # 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];
+ array_splice($args, $i+1, 1);
+ $i--;
+ $argc--;
+ }
+ }
+
+ return $args;
+ }
+
function braceSubstitution( $matches ) {
global $wgLinkCache, $wgLang;
$fname = 'Parser::braceSubstitution';
$found = false;
$nowiki = false;
$noparse = false;
-
+
$title = NULL;
# $newline is an optional newline character before the braces
$newline = $matches[1];
$part1 = $matches[2];
# If the third subpattern matched anything, it will start with |
- if ( $matches[3] !== '' ) {
- $args = explode( '|', substr( $matches[3], 1 ) );
- } else {
- $args = array();
- }
+
+ $args = $this->getTemplateArgs($matches[3]);
$argc = count( $args );
-
+
# {{{}}}
if ( strpos( $matches[0], '{{{' ) !== false ) {
$text = $matches[0];
}
# Internal variables
+ if ( !$this->mVariables ) {
+ $this->initialiseVariables();
+ }
if ( !$found && array_key_exists( $part1, $this->mVariables ) ) {
$text = $this->mVariables[$part1];
$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];
+
+ # Template table test
+
+ # Did we encounter this template already? If yes, it is in the cache
+ # and we need to check for loops.
+ if ( isset( $this->mTemplates[$part1] ) ) {
+ # Infinite loop test
+ if ( isset( $this->mTemplatePath[$part1] ) ) {
+ $noparse = true;
+ $found = true;
+ }
+ # set $text to cached message.
+ $text = $this->mTemplates[$part1];
$found = true;
}
-*/
+
# Load from database
if ( !$found ) {
$title = Title::newFromText( $part1, NS_TEMPLATE );
# Check for excessive inclusion
$dbk = $title->getPrefixedDBkey();
if ( $this->incrementIncludeCount( $dbk ) ) {
+ # This should never be reached.
$article = new Article( $title );
$articleContent = $article->getContentWithoutUsingSoManyDamnGlobals();
if ( $articleContent !== false ) {
$text = '[[' . $title->getPrefixedText() . ']]';
$found = true;
}
+
+ # Template cache array insertion
+ $this->mTemplates[$part1] = $text;
}
}
$wgLinkCache->suspend();
}
- # Run full parser on the included text
+ # Add a new element to the templace recursion path
+ $this->mTemplatePath[$part1] = 1;
+
$text = $this->stripParse( $text, $newline, $assocArgs );
# Resume the link cache and register the inclusion as a link
$wgLinkCache->addLinkObj( $title );
}
}
+ # Empties the template path
+ $this->mTemplatePath = array();
if ( !$found ) {
return $matches[0];
if ( array_key_exists( $arg, $inputArgs ) ) {
$text = $this->stripParse( $inputArgs[$arg], $newline, array() );
}
-
+
return $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',
} else {
$text .= '<' . str_replace( '>', '>', $x);
}
- }
+ }
}
wfProfileOut( $fname );
return $text;
*/
/* private */ function formatHeadings( $text, $isMain=true ) {
- global $wgInputEncoding;
-
+ global $wgInputEncoding, $wgMaxTocLevel;
+
$doNumberHeadings = $this->mOptions->getNumberHeadings();
$doShowToc = $this->mOptions->getShowToc();
+ $forceTocHere = false;
if( !$this->mTitle->userCanEdit() ) {
$showEditLink = 0;
$rightClickHack = 0;
$doShowToc = 0;
}
- # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
- # override above conditions and always show TOC
- $mw =& MagicWord::get( MAG_FORCETOC );
- if ($mw->matchAndRemove( $text ) ) {
+ # 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;
+ $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;
+ }
}
+
# We need this to perform operations on the HTML
$sk =& $this->mOptions->getSkin();
if($refcount[$headlineCount] > 1 ) {
$anchor .= '_' . $refcount[$headlineCount];
}
- if( $doShowToc ) {
+ if( $doShowToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) {
$toc .= $sk->tocLine($anchor,$tocline,$toclevel);
}
if( $showEditLink ) {
# $full .= $sk->editSectionLink(0);
}
$full .= $block;
- if( $doShowToc && !$i && $isMain) {
+ if( $doShowToc && !$i && $isMain && !$forceTocHere) {
# Top anchor now in skin
$full = $full.$toc;
}
}
$i++;
}
-
- return $full;
+ if($forceTocHere) {
+ $mw =& MagicWord::get( MAG_TOC );
+ return $mw->replace( $toc, $full );
+ } else {
+ return $full;
+ }
}
# Return an HTML link for the "ISBN 123456" text
/* private */ function magicISBN( $text ) {
global $wgLang;
+ $fname = 'Parser::magicISBN';
+ wfProfileIn( $fname );
$a = split( 'ISBN ', " $text" );
- if ( count ( $a ) < 2 ) return $text;
+ if ( count ( $a ) < 2 ) {
+ wfProfileOut( $fname );
+ return $text;
+ }
$text = substr( array_shift( $a ), 1);
$valid = '0123456789-ABCDEFGHIJKLMNOPQRSTUVWXYZ';
$text .= $x;
}
}
+ wfProfileOut( $fname );
return $text;
}
-
+
+ # Return an HTML link for the "GEO ..." text
+ /* private */ function magicGEO( $text ) {
+ global $wgLang, $wgUseGeoMode;
+ if ( !isset ( $wgUseGeoMode ) || !$wgUseGeoMode ) return $text ;
+ $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
/* private */ function magicRFC( $text ) {
global $wgLang;
if(isset($wgLocaltimezone)) putenv('TZ='.$oldtzs);
$text = preg_replace( '/~~~~~/', $d, $text );
- $text = preg_replace( '/~~~~/', '[[' . $wgLang->getNsText(
- Namespace::getUser() ) . ":$n|$k]] $d", $text );
- $text = preg_replace( '/~~~/', '[[' . $wgLang->getNsText(
- Namespace::getUser() ) . ":$n|$k]]", $text );
+ $text = preg_replace( '/~~~~/', '[[' . $wgLang->getNsText( NS_USER ) . ":$n|$k]] $d", $text );
+ $text = preg_replace( '/~~~/', '[[' . $wgLang->getNsText( NS_USER ) . ":$n|$k]]", $text );
# Context links: [[|name]] and [[name (context)|]]
#