// 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", 5 );
-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
# <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
+);
+
+/**
+ * @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 />',
} 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 ',
- '/([^> ]+(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>';
}
}
}
-
+
# nowiki
$text = Parser::extractTags('nowiki', $text, $nowiki_content, $uniq_prefix);
foreach( $nowiki_content as $marker => $content ) {
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.'-->';
}
}
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'] ) ) {
# 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 , $wgUser ;
- $fname = 'Parser::oldCategoryMagic';
-
- if ( !$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
-
- $r = "<br style=\"clear:both;\"/>\n";
-
-
- $sk =& $wgUser->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 == $cns ) {
- 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 , $wgUser ;
- if ( !$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
-
- $r = "<br style=\"clear:both;\"/>\n";
-
-
- $sk =& $wgUser->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 == $cns ) {
- $ctitle = str_replace( '_',' ',$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"
- . 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;
- $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" .
- wfMsg( 'categoryarticlecount', count( $articles ) );
-
- # 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.
'/(\\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
$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 ( '{|' == substr ( $x , 0 , 2 ) )
- {
- $t[$k] = "\n<table " . $this->fixTagAttributes ( substr ( $x , 2 ) ) . '>' ;
+ if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
+ $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 ) )
- {
+ 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 ;
+ $t[$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 );
$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->magicGEO( $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 ( $isMain && !isset ( $this->categoryMagicDone ) ) {
- $text .= $this->categoryMagic () ;
- $this->categoryMagicDone = true ;
- }
-
wfProfileOut( $fname );
return $text;
}
+ /* private */ function &doMagicLinks( &$text ) {
+ global $wgUseGeoMode;
+ $text = $this->magicISBN( $text );
+ if ( isset( $wgUseGeoMode ) && $wgUseGeoMode ) {
+ $text = $this->magicGEO( $text );
+ }
+ $text = $this->magicRFC( $text );
+ return $text;
+ }
+
# Parse ^^ tokens and return html
- /* private */ function doExponent ( $text )
- {
+ /* private */ function doExponent ( $text ) {
$fname = 'Parser::doExponent';
wfProfileIn( $fname);
$text = preg_replace('/\^\^(.*)\^\^/','<small><sup>\\1</sup></small>', $text);
return $text;
}
- # Parse headers and return html
+ # Parse headers and return html
/* private */ function doHeadings( $text ) {
$fname = 'Parser::doHeadings';
wfProfileIn( $fname );
$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) $numbold++; else
+ if (strlen ($arr[$i]) == 5) { $numitalics++; $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 [ $firstmultiletterword-1 ] .= "'";
+ }
+ # ... otherwise use the first one that has neither.
+ # (notice that it is possible for all three to be -1 if, for example,
+ # there is only one pentuple-apostrophe in the line)
+ else if ($firstspace > -1)
+ {
+ $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 == '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')
+ { $output .= '<b><i>'.$buffer.'</i>'; $state = 'b'; }
+ else # $state can be 'b' or ''
+ { $output .= '<i>'; $state .= 'i'; }
+ }
+ else if (strlen ($r) == 3)
+ {
+ 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')
+ { $output .= '<i><b>'.$buffer.'</b>'; $state = 'i'; }
+ else # $state can be 'i' or ''
+ { $output .= '<b>'; $state .= 'b'; }
+ }
+ else if (strlen ($r) == 5)
+ {
+ 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')
+ { $output .= '<i><b>'.$buffer.'</b></i>'; $state = ''; }
+ else # ($state == '')
+ { $buffer = ''; $state = 'both'; }
+ }
+ }
+ $i++;
+ }
+ # Now close all remaining tags. Notice that the order is important.
+ if ($state == 'b' || $state == 'ib')
+ $output .= '</b>';
+ if ($state == 'i' || $state == 'bi' || $state == 'ib')
+ $output .= '</i>';
+ if ($state == 'bi')
+ $output .= '</b>';
+ if ($state == 'both')
+ $output .= '<b><i>'.$buffer.'</i></b>';
+ 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 );
-
- # Regexp for URL in square brackets
- $e1 = "/^([{$uc}{$sep}]+)\\](.*)\$/sD";
- # Regexp for URL with link text in square brackets
- $e2 = "/^([{$uc}{$sep}]+)\\s+([^\\]]+)\\](.*)\$/sD";
-
- foreach ( $a as $line ) {
-
- # CASE 1: Link in square brackets, e.g.
- # some text [http://domain.tld/some.link] more text
- if ( preg_match( $e1, $line, $m ) ) {
- $link = "{$protocol}:{$m[1]}";
- $trail = $m[2];
- if ( $autonumber ) { $text = "[" . ++$this->mAutonumber . "]"; }
- else { $text = wfEscapeHTML( $link ); }
- }
+ $s = $this->replaceFreeExternalLinks( array_shift( $bits ) );
- # CASE 2: Link with link text and text directly following it, e.g.
- # This is a collection of [http://domain.tld/some.link link]s
- else if ( preg_match( $e2, $line, $m ) ) {
- $link = "{$protocol}:{$m[1]}";
- $text = $m[2];
- $dtrail = '';
- $trail = $m[3];
- if ( preg_match( wfMsg ('linktrail'), $trail, $m2 ) ) {
+ $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 = htmlspecialchars( $url );
+ }
+ } else {
+ # 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];
}
}
- # CASE 3: Nothing matches, just output the source text
- else {
- $s .= "[{$protocol}:" . $line;
- continue;
- }
-
- 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>";
+ if ( ! $sk->suppressUrlExpansion() ) {
+ $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $encUrl ) . "</i>)</span>";
+ } else {
+ $paren = '';
+ }
}
- $la = $sk->getExternalLinkAttributes( $link, $text );
- $s .= "<a href='{$link}'{$la}>{$text}</a>{$dtrail}{$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;
+ }
+
+ # make an image if it's allowed
+ 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;
+ }
+
+ # The wikilinks [[ ]] are procedeed here.
/* private */ function replaceInternalLinks( $s ) {
global $wgLang, $wgLinkCache;
- global $wgNamespacesWithSubpages, $wgLanguageCode;
+ global $wgNamespacesWithSubpages;
static $fname = 'Parser::replaceInternalLinks' ;
wfProfileIn( $fname );
if ( !$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.'-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 == $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;
+ 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;
}
- 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;
}
# These next three functions open, continue, and close the list
# element appropriate to the prefix character passed into them.
#
- /* private */ function openList( $char )
- {
+ /* private */ function openList( $char ) {
$result = $this->closeParagraph();
if ( '*' == $char ) { $result .= '<ul><li>'; }
/* private */ function nextItem( $char ) {
if ( '*' == $char || '#' == $char ) { return '</li><li>'; }
else if ( ':' == $char || ';' == $char ) {
- $close = "</dd>";
+ $close = '</dd>';
if ( $this->mDTopen ) { $close = '</dt>'; }
if ( ';' == $char ) {
$this->mDTopen = true;
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 ) {
}
foreach ( $textLines as $oLine ) {
$lastPrefixLength = strlen( $lastPrefix );
- $preCloseMatch = preg_match("/<\\/pre/i", $oLine );
- $preOpenMatch = preg_match("/<pre/i", $oLine );
+ $preCloseMatch = preg_match('/<\\/pre/i', $oLine );
+ $preOpenMatch = preg_match('/<pre/i', $oLine );
if ( !$this->mInPre ) {
# Multiple prefixes may abut each other for nested lists.
$prefixLength = strspn( $oLine, '*#:;' );
$output .= $this->nextItem( substr( $pref, -1 ) );
$paragraphStack = false;
- if ( ";" == substr( $pref, -1 ) ) {
+ if ( substr( $pref, -1 ) == ';') {
# The one nasty exception: definition lists work like this:
# ; title : definition text
# So we check for : in the remainder text to split up the
# FIXME: This is dupe of code above
if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
$term = $match[1];
- $output .= $term . $this->nextItem( ":" );
+ $output .= $term . $this->nextItem( ':' );
$t = $match[2];
}
}
if($preOpenMatch and !$preCloseMatch) {
$this->mInPre = true;
}
- if ( $closematch ) {
+ if ( $closematch ) {
$inBlockElem = false;
} else {
$inBlockElem = true;
}
} else if ( !$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) ) {
return $wgLang->formatNum( date('j') );
case MAG_PAGENAME:
return $this->mTitle->getText();
+ case MAG_PAGENAMEE:
+ return $this->mTitle->getPartialURL();
case MAG_NAMESPACE:
# return Namespace::getCanonicalName($this->mTitle->getNamespace());
- return $wgLang->getNsText($this->mTitle->getNamespace()); // Patch by Dori
+ return $wgLang->getNsText($this->mTitle->getNamespace()); # Patch by Dori
case MAG_CURRENTDAYNAME:
return $wgLang->getWeekdayName( date('w')+1 );
case MAG_CURRENTYEAR:
/* private */ function replaceVariables( $text, $args = array() ) {
global $wgLang, $wgScript, $wgArticlePath;
- # Prevent too big inclusions
- if(strlen($text)> MAX_INCLUDE_SIZE)
- return $text;
+ # Prevent too big inclusions
+ if(strlen($text)> MAX_INCLUDE_SIZE)
+ return $text;
$fname = 'Parser::replaceVariables';
wfProfileIn( $fname );
# 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 );
}
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';
$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 );
# {{{}}}
$this->mOutput->mContainsOldMagic = true;
}
- # Template table test
+ # GRAMMAR
+ if ( !$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
# and we need to check for loops.
if ( isset( $this->mTemplatePath[$part1] ) ) {
$noparse = true;
$found = true;
- }
+ }
# set $text to cached message.
$text = $this->mTemplates[$part1];
$found = true;
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;
}
- # Template cache array insertion
+ # Template cache array insertion
$this->mTemplates[$part1] = $text;
}
}
$wgLinkCache->suspend();
}
- # Add a new element to the templace recursion path
+ # Add a new element to the templace recursion path
$this->mTemplatePath[$part1] = 1;
- # Run full parser on the included text
- $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++;
}
foreach( $blocks as $block ) {
if( $showEditLink && $headlineCount > 0 && $i == 0 && $block != "\n" ) {
- # This is the [edit] link that appears for the top block of text when
+ # This is the [edit] link that appears for the top block of text when
# section editing is enabled
# Disabled because it broke block formatting
$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;
}
# 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" ) ;
+ $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}" );
- $text .= "<a href='{$url}'{$la}>RFC {$rfc}</a>{$x}";
+ $la = $sk->getExternalLinkAttributes( $url, 'RFC '.$rfc );
+ $text .= "<a href='{$url}'{$la}>RFC {$rfc}</a>{$x}";
}
}
return $text;
if(isset($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)|]]
#
$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 ); }
function setNumberHeadings( $x ) { return wfSetVar( $this->mNumberHeadings, $x ); }
function setShowToc( $x ) { return wfSetVar( $this->mShowToc, $x ); }
- function setSkin( &$x ) { $this->mSkin =& $x; }
+ function setSkin( &$x ) { $this->mSkin =& $x; }
+ # Get parser options
/* static */ function newFromUser( &$user ) {
$popts = new ParserOptions;
$popts->initialiseFromUser( $user );
return $popts;
}
+ # Get user options
function initialiseFromUser( &$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 );
}
}
# Regex callbacks, used in Parser::replaceVariables
-function wfBraceSubstitution( $matches )
-{
+function wfBraceSubstitution( $matches ) {
global $wgCurParser;
return $wgCurParser->braceSubstitution( $matches );
}
-function wfArgSubstitution( $matches )
-{
+function wfArgSubstitution( $matches ) {
global $wgCurParser;
return $wgCurParser->argSubstitution( $matches );
}
-function wfVariableSubstitution( $matches )
-{
+function wfVariableSubstitution( $matches ) {
global $wgCurParser;
return $wgCurParser->variableSubstitution( $matches );
}
+/**
+ * Return the total number of articles
+ */
+function wfNumberOfArticles() {
+ global $wgNumberOfArticles;
+
+ wfLoadSiteStats();
+ return $wgNumberOfArticles;
+}
+
+/**
+ * Get various statistics from the database
+ * @private
+ */
+function wfLoadSiteStats() {
+ global $wgNumberOfArticles, $wgTotalViews, $wgTotalEdits;
+ $fname = 'wfLoadSiteStats';
+
+ if ( -1 != $wgNumberOfArticles ) return;
+ $dbr =& wfGetDB( DB_SLAVE );
+ $s = $dbr->getArray( 'site_stats',
+ array( 'ss_total_views', 'ss_total_edits', 'ss_good_articles' ),
+ array( 'ss_row_id' => 1 ), $fname
+ );
+
+ if ( $s === false ) {
+ return;
+ } else {
+ $wgTotalViews = $s->ss_total_views;
+ $wgTotalEdits = $s->ss_total_edits;
+ $wgNumberOfArticles = $s->ss_good_articles;
+ }
+}
+
+function wfEscapeHTMLTagsOnly( $in ) {
+ return str_replace(
+ array( '"', '>', '<' ),
+ array( '"', '>', '<' ),
+ $in );
+}
+
+
?>