// require_once('Tokenizer.php');
-# PHP Parser
-#
-# Processes wiki markup
-#
-# There are two main entry points into the Parser class: parse() and preSaveTransform().
-# The parse() function produces HTML output, preSaveTransform() produces altered wiki markup.
-#
-# Globals used:
-# objects: $wgLang, $wgDateFormatter, $wgLinkCache, $wgCurParser
-#
-# NOT $wgArticle, $wgUser or $wgTitle. Keep them away!
-#
-# settings: $wgUseTex*, $wgUseCategoryMagic*, $wgUseDynamicDates*, $wgInterwikiMagic*,
-# $wgNamespacesWithSubpages, $wgLanguageCode, $wgAllowExternalImages*,
-# $wgLocaltimezone
-#
-# * only within ParserOptions
-#
-#
-#----------------------------------------
-# Variable substitution O(N^2) attack
-#-----------------------------------------
-# Without countermeasures, it would be possible to attack the parser by saving a page
-# filled with a large number of inclusions of large pages. The size of the generated
-# page would be proportional to the square of the input size. Hence, we limit the number
-# of inclusions of any given page, thus bringing any attack back to O(N).
-#
-
-define( "MAX_INCLUDE_REPEAT", 20 );
-define( "MAX_INCLUDE_SIZE", 1000000 ); // 1 Million
+/**
+ * PHP Parser
+ *
+ * Processes wiki markup
+ *
+ * There are two main entry points into the Parser class:
+ * parse()
+ * produces HTML output
+ * preSaveTransform().
+ * produces altered wiki markup.
+ *
+ * Globals used:
+ * objects: $wgLang, $wgDateFormatter, $wgLinkCache, $wgCurParser
+ *
+ * NOT $wgArticle, $wgUser or $wgTitle. Keep them away!
+ *
+ * settings:
+ * $wgUseTex*, $wgUseDynamicDates*, $wgInterwikiMagic*,
+ * $wgNamespacesWithSubpages, $wgAllowExternalImages*,
+ * $wgLocaltimezone
+ *
+ * * only within ParserOptions
+ *
+ * @package MediaWiki
+ */
+
+/**
+ * Variable substitution O(N^2) attack
+ *
+* Without countermeasures, it would be possible to attack the parser by saving
+* a page filled with a large number of inclusions of large pages. The size of
+* the generated page would be proportional to the square of the input size.
+* Hence, we limit the number of inclusions of any given page, thus bringing any
+* attack back to O(N).
+*/
+define( 'MAX_INCLUDE_REPEAT', 100 );
+define( 'MAX_INCLUDE_SIZE', 1000000 ); // 1 Million
# Allowed values for $mOutputType
-define( "OT_HTML", 1 );
-define( "OT_WIKI", 2 );
-define( "OT_MSG", 3 );
+define( 'OT_HTML', 1 );
+define( 'OT_WIKI', 2 );
+define( 'OT_MSG' , 3 );
# string parameter for extractTags which will cause it
# to strip HTML comments in addition to regular
# prefix for escaping, used in two functions at least
define( 'UNIQ_PREFIX', 'NaodW29');
-
# Constants needed for external link processing
-
define( 'URL_PROTOCOLS', 'http|https|ftp|irc|gopher|news|mailto' );
define( 'HTTP_PROTOCOLS', 'http|https' );
# Everything except bracket, space, or control characters
'('.EXT_IMAGE_FNAME_CLASS.'+)\\.((?i)'.EXT_IMAGE_EXTENSIONS.')$/S' # Filename
);
+/**
+ * @todo document
+ * @package MediaWiki
+ */
class Parser
{
# Persistent:
# Temporary:
var $mOptions, $mTitle, $mOutputType,
$mTemplates, // cache of already loaded templates, avoids
- // multiple SQL queries for the same string
+ // 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.
+ // in this path. Used for loop detection.
function Parser() {
$this->mTemplates = array();
#
function parse( $text, &$title, $options, $linestart = true, $clearState = true ) {
global $wgUseTidy;
- $fname = "Parser::parse";
+ $fname = 'Parser::parse';
wfProfileIn( $fname );
if ( $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 ",
'/<hr *>/i' => '<hr />',
'/ (\\?|:|;|!|\\302\\273)/i' => ' \\1',
# french spaces, Guillemet-right
'/(\\302\\253) /i' => '\\1 ',
- '/([^> ]+(0(1|3|9);)[^< ]*)/i' => '<span class="diacrit">\\1</span>',
'/<center *>/i' => '<div class="center">',
'/<\\/center *>/i' => '</div>'
);
# If $tag is set to STRIP_COMMENTS, the function will extract
# <!-- HTML comments -->
- /* static */ function extractTags($tag, $text, &$content, $uniq_prefix = ""){
+ /* static */ function extractTags($tag, $text, &$content, $uniq_prefix = ''){
$rnd = $uniq_prefix . '-' . $tag . Parser::getRandomString();
if ( !$content ) {
$content = array( );
#$text = str_replace( $uniq_prefix, wfHtmlEscapeFirst( $uniq_prefix ), $text );
# html
- global $wgRawHtml;
- if( $wgRawHtml ) {
+ global $wgRawHtml, $wgWhitelistEdit;
+ if( $wgRawHtml && $wgWhitelistEdit ) {
$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>";
+ $html_content[$marker] = '<html>'.$content.'</html>';
}
}
}
if( $render ){
$nowiki_content[$marker] = wfEscapeHTMLTagsOnly( $content );
} else {
- $nowiki_content[$marker] = "<nowiki>$content</nowiki>";
+ $nowiki_content[$marker] = '<nowiki>'.$content.'</nowiki>';
}
}
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>";
+ $math_content[$marker] = '<math>'.$content.'</math>';
}
}
if( $render ){
$pre_content[$marker] = '<pre>' . wfEscapeHTMLTagsOnly( $content ) . '</pre>';
} else {
- $pre_content[$marker] = "<pre>$content</pre>";
+ $pre_content[$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-->";
+ $comment_content[$marker] = '<!--'.$content.'-->';
}
}
# Add an item to the strip state
# Returns the unique tag which must be inserted into the stripped text
# The tag will be replaced with the original text in unstrip()
-
function insertStripItem( $text, &$state ) {
$rnd = UNIQ_PREFIX . '-item' . Parser::getRandomString();
if ( !$state ) {
return $rnd;
}
- # categoryMagic
- # generate a list of subcategories and pages for a category
- # depending on wfMsg("usenewcategorypage") it either calls the new
- # or the old code. The new code will not work properly for some
- # languages due to sorting issues, so they might want to turn it
- # off.
- function categoryMagic() {
- $msg = wfMsg('usenewcategorypage');
- if ( '0' == @$msg[0] )
- {
- return $this->oldCategoryMagic();
- } else {
- return $this->newCategoryMagic();
- }
- }
-
- # This method generates the list of subcategories and pages for a category
- function oldCategoryMagic () {
- global $wgLang ;
- $fname = 'Parser::oldCategoryMagic';
-
- if ( !$this->mOptions->getUseCategoryMagic() ) return ; # Doesn't use categories at all
-
- if ( $this->mTitle->getNamespace() != NS_CATEGORY ) return "" ; # This ain't a category page
-
- $r = "<br style=\"clear:both;\"/>\n";
-
-
- $sk =& $this->mOptions->getSkin() ;
-
- $articles = array() ;
- $children = array() ;
- $data = array () ;
- $id = $this->mTitle->getArticleID() ;
-
- # FIXME: add limits
- $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 )
- {
- $t = $wgLang->getNsText ( $x->cur_namespace ) ;
- if ( $t != '' ) $t .= ':' ;
- $t .= $x->cur_title ;
-
- if ( $x->cur_namespace == NS_CATEGORY ) {
- array_push ( $children , $sk->makeLink ( $t ) ) ; # Subcategory
- } else {
- array_push ( $articles , $sk->makeLink ( $t ) ) ; # Page in this category
- }
- }
- $dbr->freeResult ( $res ) ;
-
- # Showing subcategories
- if ( count ( $children ) > 0 ) {
- $r .= '<h2>'.wfMsg('subcategories')."</h2>\n" ;
- $r .= implode ( ', ' , $children ) ;
- }
-
- # Showing pages in this category
- if ( count ( $articles ) > 0 ) {
- $ti = $this->mTitle->getText() ;
- $h = wfMsg( 'category_header', $ti );
- $r .= "<h2>{$h}</h2>\n" ;
- $r .= implode ( ', ' , $articles ) ;
- }
-
- return $r ;
- }
-
- function newCategoryMagic () {
- global $wgLang;
- if ( !$this->mOptions->getUseCategoryMagic() ) return ; # Doesn't use categories at all
-
- if ( $this->mTitle->getNamespace() != NS_CATEGORY ) return '' ; # This ain't a category page
-
- $r = "<br style=\"clear:both;\"/>\n";
-
-
- $sk =& $this->mOptions->getSkin() ;
-
- $articles = array() ;
- $articles_start_char = array();
- $children = array() ;
- $children_start_char = array();
- $data = array () ;
- $id = $this->mTitle->getArticleID() ;
-
- # FIXME: add limits
- $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 == NS_CATEGORY ) {
- $ctitle = str_replace( '_',' ',$x->cur_title );
- array_push ( $children, $sk->makeKnownLink ( $t, $ctitle ) ) ; # Subcategory
-
- // If there's a link from Category:A to Category:B, the sortkey of the resulting
- // entry in the categorylinks table is Category:A, not A, which it SHOULD be.
- // Workaround: If sortkey == "Category:".$title, than use $title for sorting,
- // else use sortkey...
- if ( ($ns.':'.$ctitle) == $x->cl_sortkey ) {
- array_push ( $children_start_char, $wgLang->firstChar( $x->cur_title ) );
- } else {
- array_push ( $children_start_char, $wgLang->firstChar( $x->cl_sortkey ) ) ;
- }
- } else {
- array_push ( $articles , $sk->makeKnownLink ( $t ) ) ; # Page in this category
- array_push ( $articles_start_char, $wgLang->firstChar( $x->cl_sortkey ) ) ;
- }
- }
- $dbr->freeResult ( $res ) ;
-
- $ti = $this->mTitle->getText() ;
-
- # Don't show subcategories section if there are none.
- if ( count ( $children ) > 0 )
- {
- # Showing subcategories
- $r .= '<h2>' . wfMsg( 'subcategories' ) . "</h2>\n";
-
- $numchild = count( $children );
- if($numchild == 1) {
- $r .= wfMsg( 'subcategorycount1', 1 );
- } else {
- $r .= wfMsg( 'subcategorycount' , $numchild );
- }
- unset($numchild);
-
- 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;
- $chunkIndex < 3;
- $chunkIndex++, $startChunk = $endChunk, $endChunk += $chunk + 1)
- {
-
- $r .= '<td><ul>';
- // output all subcategories to category
- for ($index = $startChunk ;
- $index < $endChunk && $index < count($children);
- $index++ )
- {
- // check for change of starting letter or begging of chunk
- if ( ($children_start_char[$index] != $children_start_char[$index - 1])
- || ($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); $index++ )
- {
- if ($children_start_char[$index] != $children_start_char[$index - 1])
- {
- $r .= "</ul><h3>{$children_start_char[$index]}</h3>\n<ul>";
- }
-
- $r .= "<li>{$children[$index]}</li>";
- }
- $r .= '</ul>';
- }
- } # END of if ( count($children) > 0 )
-
- $r .= '<h2>' . wfMsg( 'category_header', $ti ) . "</h2>\n";
-
- $numart = count( $articles );
- if($numart == 1) {
- $r .= wfMsg( 'categoryarticlecount1', 1 );
- } else {
- $r .= wfMsg( 'categoryarticlecount' , $numart );
- }
- unset($numart);
-
- # 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;
- $chunkIndex < 3;
- $chunkIndex++, $startChunk = $endChunk, $endChunk += $chunk + 1)
- {
-
- $r .= '<td><ul>';
-
- // output all articles in category
- for ($index = $startChunk ;
- $index < $endChunk && $index < count($articles);
- $index++ )
- {
- // check for change of starting letter or begging of chunk
- if ( ($articles_start_char[$index] != $articles_start_char[$index - 1])
- || ($index == $startChunk) )
- {
- $r .= "</ul><h3>{$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); $index++ )
- {
- if ($articles_start_char[$index] != $articles_start_char[$index - 1])
- {
- $r .= "</ul><h3>{$articles_start_char[$index]}</h3>\n<ul>";
- }
-
- $r .= "<li>{$articles[$index]}</li>";
- }
- $r .= '</ul>';
- }
-
-
- return $r ;
- }
-
# Return allowed HTML attributes
function getHTMLattrs () {
$htmlattrs = array( # Allowed attributes--no scripting, etc.
"(in_array(strtolower(\"\$1\"),\$htmlattrs)?(\"\$1\".((\"x\$3\" != \"x\")?\"=\$3\":'')):'')",
$t);
- $t = str_replace ( "<></>" , "" , $t ) ; # This should fix bug 980557
+ $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
{
$x = trim ( $x ) ;
$fc = substr ( $x , 0 , 1 ) ;
- if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) )
- {
+ if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
$indent_level = strlen( $matches[1] );
$t[$k] = "\n" .
- str_repeat( "<dl><dd>", $indent_level ) .
- "<table " . $this->fixTagAttributes ( $matches[2] ) . '>' ;
+ 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 ) )
- {
+ 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 ;
+ if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
array_pop ( $ltr ) ;
- $t[$k] = $z . str_repeat( "</dd></dl>", $indent_level );
+ $t[$k] = $z . str_repeat( '</dd></dl>', $indent_level );
}
- else if ( '|-' == substr ( $x , 0 , 2 ) ) # Allows for |---------------
- {
+ 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 ;
+ if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
array_pop ( $ltr ) ;
$t[$k] = $z ;
array_push ( $tr , false ) ;
array_push ( $ltd , '' ) ;
array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
}
- else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) # Caption
- {
- if ( '|+' == substr ( $x , 0 , 2 ) )
- {
+ else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) { # Caption
+ if ( '|+' == substr ( $x , 0 , 2 ) ) {
$fc = '+' ;
$x = substr ( $x , 1 ) ;
}
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 , '' ) ;
}
$l = array_pop ( $ltd ) ;
- if ( array_pop ( $td ) ) $z = "</{$l}>" . $z ;
+ if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
if ( $fc == '|' ) $l = 'td' ;
else if ( $fc == '!' ) $l = 'th' ;
else if ( $fc == '+' ) $l = 'caption' ;
return $t ;
}
- # Parses the text and adds the result to the strip state
- # Returns the strip tag
- 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 ) {
+ global $wgLang;
+
$fname = 'Parser::internalParse';
wfProfileIn( $fname );
$text = $this->removeHTMLtags( $text );
$text = $this->replaceVariables( $text, $args );
+ $text = $wgLang->convert($text);
+
$text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
$text = $this->doHeadings( $text );
$sk =& $this->mOptions->getSkin();
$text = $sk->transformContent( $text );
- if ( $isMain && !isset ( $this->categoryMagicDone ) ) {
- $text .= $this->categoryMagic () ;
- $this->categoryMagicDone = true ;
- }
-
wfProfileOut( $fname );
return $text;
}
{
$x1 = substr ($arr[$i-1], -1);
$x2 = substr ($arr[$i-1], -2, 1);
- if ($x1 == " ") {
+ if ($x1 == ' ') {
if ($firstspace == -1) $firstspace = $i;
- } else if ($x2 == " ") {
+ } else if ($x2 == ' ') {
if ($firstsingleletterword == -1) $firstsingleletterword = $i;
} else {
if ($firstmultiletterword == -1) $firstmultiletterword = $i;
{
if (strlen ($r) == 2)
{
- if ($state == 'em')
- { $output .= "</em>"; $state = ''; }
- else if ($state == 'strongem')
- { $output .= "</em>"; $state = 'strong'; }
- else if ($state == 'emstrong')
- { $output .= "</strong></em><strong>"; $state = 'strong'; }
+ if ($state == 'i')
+ { $output .= '</i>'; $state = ''; }
+ else if ($state == 'bi')
+ { $output .= '</i>'; $state = 'b'; }
+ else if ($state == 'ib')
+ { $output .= '</b></i><b>'; $state = 'b'; }
else if ($state == 'both')
- { $output .= "<strong><em>{$buffer}</em>"; $state = 'strong'; }
- else # $state can be 'strong' or ''
- { $output .= "<em>"; $state .= 'em'; }
+ { $output .= '<b><i>'.$buffer.'</i>'; $state = 'b'; }
+ else # $state can be 'b' or ''
+ { $output .= '<i>'; $state .= 'i'; }
}
else if (strlen ($r) == 3)
{
- if ($state == 'strong')
- { $output .= "</strong>"; $state = ''; }
- else if ($state == 'strongem')
- { $output .= "</em></strong><em>"; $state = 'em'; }
- else if ($state == 'emstrong')
- { $output .= "</strong>"; $state = 'em'; }
+ if ($state == 'b')
+ { $output .= '</b>'; $state = ''; }
+ else if ($state == 'bi')
+ { $output .= '</i></b><i>'; $state = 'i'; }
+ else if ($state == 'ib')
+ { $output .= '</b>'; $state = 'i'; }
else if ($state == 'both')
- { $output .= "<em><strong>{$buffer}</strong>"; $state = 'em'; }
- else # $state can be 'em' or ''
- { $output .= "<strong>"; $state .= 'strong'; }
+ { $output .= '<i><b>'.$buffer.'</b>'; $state = 'i'; }
+ else # $state can be 'i' or ''
+ { $output .= '<b>'; $state .= 'b'; }
}
else if (strlen ($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>"; $state = ''; }
+ if ($state == 'b')
+ { $output .= '</b><i>'; $state = 'i'; }
+ else if ($state == 'i')
+ { $output .= '</i><b>'; $state = 'b'; }
+ else if ($state == 'bi')
+ { $output .= '</i></b>'; $state = ''; }
+ else if ($state == 'ib')
+ { $output .= '</b></i>'; $state = ''; }
else if ($state == 'both')
- { $output .= "<em><strong>{$buffer}</strong></em>"; $state = ''; }
+ { $output .= '<i><b>'.$buffer.'</b></i>'; $state = ''; }
else # ($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 == 'b' || $state == 'ib')
+ $output .= '</b>';
+ if ($state == 'i' || $state == 'bi' || $state == 'ib')
+ $output .= '</i>';
+ if ($state == 'bi')
+ $output .= '</b>';
if ($state == 'both')
- $output .= "<strong><em>{$buffer}</em></strong>";
+ $output .= '<b><i>'.$buffer.'</i></b>';
return $output;
}
}
if ( $text == '' ) {
# Autonumber if allowed
if ( strpos( HTTP_PROTOCOLS, $protocol ) !== false ) {
- $text = "[" . ++$this->mAutonumber . "]";
+ $text = '[' . ++$this->mAutonumber . ']';
} else {
# Otherwise just use the URL
$text = htmlspecialchars( $url );
$encUrl = htmlspecialchars( $url );
# Bit in parentheses showing the URL for the printable version
- if( $url == $text || preg_match( "!$protocol://" . preg_quote( $text, "/" ) . "/?$!", $url ) ) {
+ if( $url == $text || preg_match( "!$protocol://" . preg_quote( $text, '/' ) . "/?$!", $url ) ) {
$paren = '';
} else {
# Expand the URL for printable version
}
return $text;
}
-
+
+ # The wikilinks [[ ]] are procedeed here.
/* private */ function replaceInternalLinks( $s ) {
global $wgLang, $wgLinkCache;
- global $wgNamespacesWithSubpages, $wgLanguageCode;
+ global $wgNamespacesWithSubpages;
static $fname = 'Parser::replaceInternalLinks' ;
wfProfileIn( $fname );
wfProfileOut( $fname.'-setup' );
+ # start procedeeding each line
foreach ( $a as $line ) {
wfProfileIn( $fname.'-prefixhandling' );
if ( $useLinkPrefixExtension ) {
continue;
}
- /* Valid link forms:
- Foobar -- normal
- :Foobar -- override special treatment of prefix (images, language links)
- /Foobar -- convert to CurrentPage/Foobar
- /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text
- */
+ # Valid link forms:
+ # Foobar -- normal
+ # :Foobar -- override special treatment of prefix (images, language links)
+ # /Foobar -- convert to CurrentPage/Foobar
+ # /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text
+
+ # Look at the first character
$c = substr($m[1],0,1);
$noforce = ($c != ':');
- if( $c == '/' ) { # subpage
- if(substr($m[1],-1,1)=='/') { # / at end means we don't want the slash to be shown
+
+ # subpage
+ if( $c == '/' ) {
+ # / at end means we don't want the slash to be shown
+ if(substr($m[1],-1,1)=='/') {
$m[1]=substr($m[1],1,strlen($m[1])-2);
$noslash=$m[1];
} else {
$noslash=substr($m[1],1);
}
- if(!empty($wgNamespacesWithSubpages[$this->mTitle->getNamespace()])) { # subpages allowed here
+
+ # Some namespaces don't allow subpages
+ if(!empty($wgNamespacesWithSubpages[$this->mTitle->getNamespace()])) {
+ # subpages allowed here
$link = $this->mTitle->getPrefixedText(). '/' . trim($noslash);
if( '' == $text ) {
$text= $m[1];
} # this might be changed for ugliness reasons
} else {
- $link = $noslash; # no subpage allowed, use standard link
+ # no subpage allowed, use standard link
+ $link = $noslash;
}
+
} elseif( $noforce ) { # no subpage
$link = $m[1];
} else {
+ # We don't want to keep the first character
$link = substr( $m[1], 1 );
}
+
$wasblank = ( '' == $text );
- if( $wasblank )
- $text = $link;
+ if( $wasblank ) $text = $link;
$nt = Title::newFromText( $link );
if( !$nt ) {
$s .= $prefix . '[[' . $line;
continue;
}
+
$ns = $nt->getNamespace();
$iw = $nt->getInterWiki();
+
+ # Link not escaped by : , create the various objects
if( $noforce ) {
+
+ # Interwikis
if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgLang->getLanguageName( $iw ) ) {
array_push( $this->mOutput->mLanguageLinks, $nt->getFullText() );
$tmp = $prefix . $trail ;
$s .= (trim($tmp) == '')? '': $tmp;
continue;
}
+
if ( $ns == NS_IMAGE ) {
$s .= $prefix . $sk->makeImageLinkObj( $nt, $text ) . $trail;
$wgLinkCache->addImageLinkObj( $nt );
continue;
}
+
if ( $ns == NS_CATEGORY ) {
$t = $nt->getText() ;
- $nnt = Title::newFromText ( Namespace::getCanonicalName(NS_CATEGORY).":".$t ) ;
+ $nnt = Title::newFromText ( Namespace::getCanonicalName(NS_CATEGORY).':'.$t ) ;
$wgLinkCache->suspend(); # Don't save in links/brokenlinks
$pPLC=$sk->postParseLinkColour();
$sk->postParseLinkColour( $pPLC );
$wgLinkCache->resume();
- $sortkey = $wasblank ? $this->mTitle->getPrefixedText() : $text;
+ if ( $wasblank ) {
+ if ( $this->mTitle->getNamespace() == NS_CATEGORY ) {
+ $sortkey = $this->mTitle->getText();
+ } else {
+ $sortkey = $this->mTitle->getPrefixedText();
+ }
+ } else {
+ $sortkey = $text;
+ }
$wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
$this->mOutput->mCategoryLinks[] = $t ;
$s .= $prefix . $trail ;
continue;
}
}
- if( ( $nt->getPrefixedText() == $this->mTitle->getPrefixedText() ) &&
- ( strpos( $link, '#' ) == FALSE ) ) {
+
+ if( ( $nt->getPrefixedText() === $this->mTitle->getPrefixedText() ) &&
+ ( strpos( $link, '#' ) === FALSE ) ) {
# Self-links are handled specially; generally de-link and change to bold.
$s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail );
continue;
return '<!-- ERR 2 -->';
}
- /* private */function closeList( $char ) {
+ /* private */ function closeList( $char ) {
if ( '*' == $char ) { $text = '</li></ul>'; }
else if ( '#' == $char ) { $text = '</li></ol>'; }
else if ( ':' == $char ) {
# FIXME: This is dupe of code above
if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
$term = $match[1];
- $output .= $term . $this->nextItem( ":" );
+ $output .= $term . $this->nextItem( ':' );
$t = $match[2];
}
}
$inBlockElem = true;
}
} else if ( !$inBlockElem && !$this->mInPre ) {
- if ( " " == $t{0} and ( $this->mLastSection == 'pre' or trim($t) != '' ) ) {
+ if ( ' ' == $t{0} and ( $this->mLastSection == 'pre' or trim($t) != '' ) ) {
// pre
if ($this->mLastSection != 'pre') {
$paragraphStack = false;
$output .= $this->closeParagraph().'<pre>';
$this->mLastSection = 'pre';
}
+ $t = substr( $t, 1 );
} else {
// paragraph
if ( '' == trim($t) ) {
# PHP global rebinding syntax is a bit weird, need to use the GLOBALS array
$GLOBALS['wgCurParser'] =& $this;
-
- if ( $this->mOutputType == OT_HTML ) {
+ if ( $this->mOutputType == OT_HTML || $this->mOutputType == OT_MSG ) {
# Variable substitution
$text = preg_replace_callback( "/{{([$nonBraceChars]*?)}}/", 'wfVariableSubstitution', $text );
-
+ }
+
+ if ( $this->mOutputType == OT_HTML ) {
# Argument substitution
$text = preg_replace_callback( "/(\\n?){{{([$titleChars]*?)}}}/", 'wfArgSubstitution', $text );
}
$argc = count($args);
$i = 0;
for ( $i = 0; $i < $argc-1; $i++ ) {
- if ( substr_count ( $args[$i], "[[" ) != substr_count ( $args[$i], "]]" ) ) {
- $args[$i] .= "|".$args[$i+1];
+ if ( substr_count ( $args[$i], '[[' ) != substr_count ( $args[$i], ']]' ) ) {
+ $args[$i] .= '|'.$args[$i+1];
array_splice($args, $i+1, 1);
$i--;
$argc--;
$this->mOutput->mContainsOldMagic = true;
}
+ # GRAMMAR
+ if ( !$found && $argc == 1 ) {
+ $mwGrammar =& MagicWord::get( MAG_GRAMMAR );
+ if ( $mwGrammar->matchStartAndRemove( $part1 ) ) {
+ $text = $wgLang->convertGrammar( $args[0], $part1 );
+ $found = true;
+ }
+ }
+
# Template table test
# Did we encounter this template already? If yes, it is in the cache
if ( $articleContent !== false ) {
$found = true;
$text = $articleContent;
-
}
}
# If the title is valid but undisplayable, make a link to it
if ( $this->mOutputType == OT_HTML && !$found ) {
- $text = '[[' . $title->getPrefixedText() . ']]';
+ $text = '[['.$title->getPrefixedText().']]';
$found = true;
}
# Add a new element to the templace recursion path
$this->mTemplatePath[$part1] = 1;
- $text = $this->stripParse( $text, $newline, $assocArgs );
+ $text = $this->removeHTMLtags( $text );
+ $text = $this->replaceVariables( $text, $assocArgs );
# Resume the link cache and register the inclusion as a link
if ( !is_null( $title ) ) {
$wgLinkCache->addLinkObj( $title );
}
}
- # Empties the template path
- $this->mTemplatePath = array();
+
+ # Empties the template path
+ $this->mTemplatePath = array();
if ( !$found ) {
return $matches[0];
$inputArgs = end( $this->mArgStack );
if ( array_key_exists( $arg, $inputArgs ) ) {
- $text = $this->stripParse( $inputArgs[$arg], $newline, array() );
+ $text = $this->removeHTMLtags( $inputArgs[$arg] );
+ $text = $this->replaceVariables( $text, array() );
}
return $text;
$htmlattrs = $this->getHTMLattrs () ;
# Remove HTML comments
- $text = preg_replace( '/(\\n *<!--.*--> *(?=\\n)|<!--.*-->)/sU', '$2', $text );
+ $text = preg_replace( '/(\\n *<!--.*--> *|<!--.*?-->)/sU', '', $text );
$bits = explode( '<', $text );
$text = array_shift( $bits );
}
-/*
- *
- * This function accomplishes several tasks:
- * 1) Auto-number headings if that option is enabled
- * 2) Add an [edit] link to sections for logged in users who have enabled the option
- * 3) Add a Table of contents on the top for users who have enabled the option
- * 4) Auto-anchor headings
- *
- * It loops through all headlines, collects the necessary data, then splits up the
- * string and re-inserts the newly formatted headlines.
- *
- */
-
+ # This function accomplishes several tasks:
+ # 1) Auto-number headings if that option is enabled
+ # 2) Add an [edit] link to sections for logged in users who have enabled the option
+ # 3) Add a Table of contents on the top for users who have enabled the option
+ # 4) Auto-anchor headings
+ #
+ # It loops through all headlines, collects the necessary data, then splits up the
+ # string and re-inserts the newly formatted headlines.
/* private */ function formatHeadings( $text, $isMain=true ) {
- global $wgInputEncoding, $wgMaxTocLevel;
+ global $wgInputEncoding, $wgMaxTocLevel, $wgLang;
$doNumberHeadings = $this->mOptions->getNumberHeadings();
$doShowToc = $this->mOptions->getShowToc();
if( $dot ) {
$numbering .= '.';
}
- $numbering .= $sublevelCount[$i];
+ $numbering .= $wgLang->formatNum( $sublevelCount[$i] );
$dot = 1;
}
}
$canonized_headline = $this->unstrip( $headline, $this->mStripState );
$canonized_headline = $this->unstripNoWiki( $headline, $this->mStripState );
+ # Remove link placeholders by the link text.
+ # <!--LINK namespace page_title link text with suffix-->
+ # turns into
+ # link text with suffix
+ $canonized_headline = preg_replace( '/<!--LINK [0-9]* [^ ]* *(.*?)-->/','$1', $canonized_headline );
# strip out HTML
$canonized_headline = preg_replace( '/<.*?' . '>/','',$canonized_headline );
$tocline = trim( $canonized_headline );
}
# give headline the correct <h#> tag
- @$head[$headlineCount] .= "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount] .$headline."</h".$level.">";
+ @$head[$headlineCount] .= "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount] .$headline.'</h'.$level.'>';
$headlineCount++;
}
$fname = 'Parser::magicISBN';
wfProfileIn( $fname );
- $a = split( 'ISBN ', " $text" );
+ $a = split( 'ISBN ', ' '.$text );
if ( count ( $a ) < 2 ) {
wfProfileOut( $fname );
return $text;
$blank .= ' ';
$x = substr( $x, 1 );
}
+ if ( $x == '' ) { # blank isbn
+ $text .= "ISBN $blank";
+ continue;
+ }
while ( strstr( $valid, $x{0} ) != false ) {
$isbn .= $x{0};
$x = substr( $x, 1 );
} else {
$titleObj = Title::makeTitle( NS_SPECIAL, 'Booksources' );
$text .= '<a href="' .
- $titleObj->escapeLocalUrl( "isbn={$num}" ) .
+ $titleObj->escapeLocalUrl( 'isbn='.$num ) .
"\" class=\"internal\">ISBN $isbn</a>";
$text .= $x;
}
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" ) ;
+ $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" );
+ $a = split( 'GEO ', ' '.$text );
if ( count ( $a ) < 2 ) {
wfProfileOut( $fname );
return $text;
$num = str_replace( '+', '', $geo );
$num = str_replace( ' ', '', $num );
- if ( '' == $num || count ( explode ( ":" , $num , 3 ) ) < 2 ) {
+ 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}" ) .
+ $titleObj->escapeLocalUrl( 'coordinates='.$num ) .
"\" class=\"internal\">GEO $geo</a>";
$text .= $x;
}
$url = wfmsg( 'rfcurl' );
$url = str_replace( '$1', $rfc, $url);
$sk =& $this->mOptions->getSkin();
- $la = $sk->getExternalLinkAttributes( $url, "RFC {$rfc}" );
+ $la = $sk->getExternalLinkAttributes( $url, 'RFC '.$rfc );
$text .= "<a href='{$url}'{$la}>RFC {$rfc}</a>{$x}";
}
}
$p1 = "/\[\[({$np}+) \\(({$np}+)\\)\\|]]/"; # [[page (context)|]]
$p2 = "/\[\[\\|({$tc}+)]]/"; # [[|page]]
- $p3 = "/\[\[($namespacechar+):({$np}+)\\|]]/"; # [[namespace:page|]]
- $p4 = "/\[\[($namespacechar+):({$np}+) \\(({$np}+)\\)\\|]]/";
- # [[ns:page (cont)|]]
- $context = "";
+ $p3 = "/\[\[(:*$namespacechar+):({$np}+)\\|]]/"; # [[namespace:page|]] and [[:namespace:page|]]
+ $p4 = "/\[\[(:*$namespacechar+):({$np}+) \\(({$np}+)\\)\\|]]/"; # [[ns:page (cont)|]] and [[:ns:page (cont)|]]
+ $context = '';
$t = $this->mTitle->getText();
if ( preg_match( $conpat, $t, $m ) ) {
$context = $m[2];
}
}
+/**
+ * @todo document
+ * @package MediaWiki
+ */
class ParserOutput
{
var $mText, $mLanguageLinks, $mCategoryLinks, $mContainsOldMagic;
var $mCacheTime; # Used in ParserCache
- function ParserOutput( $text = "", $languageLinks = array(), $categoryLinks = array(),
+ function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(),
$containsOldMagic = false )
{
$this->mText = $text;
$this->mLanguageLinks = $languageLinks;
$this->mCategoryLinks = $categoryLinks;
$this->mContainsOldMagic = $containsOldMagic;
- $this->mCacheTime = "";
+ $this->mCacheTime = '';
}
function getText() { return $this->mText; }
}
+/**
+ * Set options of the Parser
+ * @todo document
+ * @package MediaWiki
+ */
class ParserOptions
{
# All variables are private
var $mUseTeX; # Use texvc to expand <math> tags
- var $mUseCategoryMagic; # Treat [[Category:xxxx]] tags specially
var $mUseDynamicDates; # Use $wgDateFormatter to format dates
var $mInterwikiMagic; # Interlanguage links are removed and returned in an array
var $mAllowExternalImages; # Allow external images inline
var $mShowToc; # Show table of contents
function getUseTeX() { return $this->mUseTeX; }
- function getUseCategoryMagic() { return $this->mUseCategoryMagic; }
function getUseDynamicDates() { return $this->mUseDynamicDates; }
function getInterwikiMagic() { return $this->mInterwikiMagic; }
function getAllowExternalImages() { return $this->mAllowExternalImages; }
function getShowToc() { return $this->mShowToc; }
function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX, $x ); }
- function setUseCategoryMagic( $x ) { return wfSetVar( $this->mUseCategoryMagic, $x ); }
function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates, $x ); }
function setInterwikiMagic( $x ) { return wfSetVar( $this->mInterwikiMagic, $x ); }
function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages, $x ); }
# Get user options
function initialiseFromUser( &$userInput ) {
- global $wgUseTeX, $wgUseCategoryMagic, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages;
-
+ global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages;
+
+ $fname = 'ParserOptions::initialiseFromUser';
+ wfProfileIn( $fname );
if ( !$userInput ) {
$user = new User;
$user->setLoaded( true );
}
$this->mUseTeX = $wgUseTeX;
- $this->mUseCategoryMagic = $wgUseCategoryMagic;
$this->mUseDynamicDates = $wgUseDynamicDates;
$this->mInterwikiMagic = $wgInterwikiMagic;
$this->mAllowExternalImages = $wgAllowExternalImages;
+ wfProfileIn( $fname.'-skin' );
$this->mSkin =& $user->getSkin();
+ 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 );
}
return $wgCurParser->variableSubstitution( $matches );
}
-# Return the total number of articles
+/**
+ * Return the total number of articles
+ */
function wfNumberOfArticles() {
global $wgNumberOfArticles;
return $wgNumberOfArticles;
}
-# Get various statistics from the database
-/* private */ function wfLoadSiteStats() {
+/**
+ * Get various statistics from the database
+ * @private
+ */
+function wfLoadSiteStats() {
global $wgNumberOfArticles, $wgTotalViews, $wgTotalEdits;
$fname = 'wfLoadSiteStats';