<?php
-// 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
+/**
+ * File for Parser and related classes
+ *
+ * @package MediaWiki
+ * @version $Id$
+ */
+
+/**
+ * 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]' );
+# 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
+);
+
+/**
+ * PHP Parser
+ *
+ * Processes wiki markup
+ *
+ * <pre>
+ * There are three main entry points into the Parser class:
+ * parse()
+ * produces HTML output
+ * preSaveTransform().
+ * produces altered wiki markup.
+ * transformMsg()
+ * performs brace substitution on MediaWiki messages
+ *
+ * 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
+ * </pre>
+ *
+ * @package MediaWiki
+ */
class Parser
{
+ /**#@+
+ * @access private
+ */
# Persistent:
var $mTagHooks;
# 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.
+ /**#@-*/
+
+ /**
+ * Constructor
+ *
+ * @access public
+ */
function Parser() {
$this->mTemplates = array();
$this->mTemplatePath = array();
$this->clearState();
}
+ /**
+ * Clear Parser state
+ *
+ * @access private
+ */
function clearState() {
$this->mOutput = new ParserOutput;
$this->mAutonumber = 0;
$this->mInPre = false;
}
- # First pass--just handle <nowiki> sections, pass the rest off
- # to internalParse() which does all the real work.
- #
- # Returns a ParserOutput
- #
+ /**
+ * First pass--just handle <nowiki> sections, pass the rest off
+ * to internalParse() which does all the real work.
+ *
+ * @access private
+ * @return ParserOutput a ParserOutput
+ */
function parse( $text, &$title, $options, $linestart = true, $clearState = true ) {
- global $wgUseTidy;
- $fname = "Parser::parse";
+ global $wgUseTidy, $wgContLang;
+ $fname = 'Parser::parse';
wfProfileIn( $fname );
if ( $clearState ) {
$stripState = NULL;
$text = $this->strip( $text, $this->mStripState );
+
$text = $this->internalParse( $text, $linestart );
$text = $this->unstrip( $text, $this->mStripState );
# Clean up special characters, only run once, next-to-last before doBlockLevels
$fixtags = array(
# 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>'
);
}
# only once and last
$text = $this->doBlockLevels( $text, $linestart );
+
+ $text = $wgContLang->convert($text);
+
$text = $this->unstripNoWiki( $text, $this->mStripState );
- if($wgUseTidy) {
- $text = $this->tidy($text);
- }
$this->mOutput->setText( $text );
wfProfileOut( $fname );
return $this->mOutput;
}
- /* static */ function getRandomString() {
+ /**
+ * Get a random string
+ *
+ * @access private
+ * @static
+ */
+ function getRandomString() {
return dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff));
}
- # Replaces all occurrences of <$tag>content</$tag> in the text
- # with a random marker and returns the new text. the output parameter
- # $content will be an associative array filled with data on the form
- # $unique_marker => content.
-
- # If $content is already set, the additional entries will be appended
-
- # If $tag is set to STRIP_COMMENTS, the function will extract
- # <!-- HTML comments -->
-
- /* static */ function extractTags($tag, $text, &$content, $uniq_prefix = ""){
+ /**
+ * Replaces all occurrences of <$tag>content</$tag> in the text
+ * with a random marker and returns the new text. the output parameter
+ * $content will be an associative array filled with data on the form
+ * $unique_marker => content.
+ *
+ * If $content is already set, the additional entries will be appended
+ * If $tag is set to STRIP_COMMENTS, the function will extract
+ * <!-- HTML comments -->
+ *
+ * @access private
+ * @static
+ */
+ function extractTags($tag, $text, &$content, $uniq_prefix = ''){
$rnd = $uniq_prefix . '-' . $tag . Parser::getRandomString();
if ( !$content ) {
$content = array( );
return $stripped;
}
- # Strips and renders <nowiki>, <pre>, <math>, <hiero>
- # If $render is set, performs necessary rendering operations on plugins
- # Returns the text, and fills an array with data needed in unstrip()
- # If the $state is already a valid strip state, it adds to the state
-
- # When $stripcomments is set, HTML comments <!-- like this -->
- # will be stripped in addition to other tags. This is important
- # for section editing, where these comments cause confusion when
- # counting the sections in the wikisource
+ /**
+ * Strips and renders nowiki, pre, math, hiero
+ * If $render is set, performs necessary rendering operations on plugins
+ * Returns the text, and fills an array with data needed in unstrip()
+ * If the $state is already a valid strip state, it adds to the state
+ *
+ * @param bool $stripcomments when set, HTML comments <!-- like this -->
+ * will be stripped in addition to other tags. This is important
+ * for section editing, where these comments cause confusion when
+ * counting the sections in the wikisource
+ *
+ * @access private
+ */
function strip( $text, &$state, $stripcomments = false ) {
$render = ($this->mOutputType == OT_HTML);
$html_content = array();
#$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.'-->';
}
}
return $text;
}
- # always call unstripNoWiki() after this one
+ /**
+ * restores pre, math, and heiro removed by strip()
+ *
+ * always call unstripNoWiki() after this one
+ * @access private
+ */
function unstrip( $text, &$state ) {
# Must expand in reverse order, otherwise nested tags will be corrupted
$contentDict = end( $state );
return $text;
}
- # always call this after unstrip() to preserve the order
+
+ /**
+ * always call this after unstrip() to preserve the order
+ *
+ * @access private
+ */
function unstripNoWiki( $text, &$state ) {
# Must expand in reverse order, otherwise nested tags will be corrupted
for ( $content = end($state['nowiki']); $content !== false; $content = prev( $state['nowiki'] ) ) {
return $text;
}
- # 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()
-
+ /**
+ * 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()
+ *
+ * @access private
+ */
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
-
- if ( $this->mTitle->getNamespace() != NS_CATEGORY ) 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 == 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 , $wgUser ;
- if ( !$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 =& $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 == 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"
- . 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
+ /**
+ * Return allowed HTML attributes
+ *
+ * @access private
+ */
function getHTMLattrs () {
$htmlattrs = array( # Allowed attributes--no scripting, etc.
'title', 'align', 'lang', 'dir', 'width', 'height',
return $htmlattrs ;
}
- # Remove non approved attributes and javascript in css
+ /**
+ * Remove non approved attributes and javascript in css
+ *
+ * @access private
+ */
function fixTagAttributes ( $t ) {
if ( trim ( $t ) == '' ) return '' ; # Saves runtime ;-)
$htmlattrs = $this->getHTMLattrs() ;
"(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
return trim ( $t ) ;
}
- # interface with html tidy, used if $wgUseTidy = true
+ /**
+ * interface with html tidy, used if $wgUseTidy = true
+ *
+ * @access public
+ * @static
+ */
function tidy ( $text ) {
global $wgTidyConf, $wgTidyBin, $wgTidyOpts;
global $wgInputEncoding, $wgOutputEncoding;
wfProfileIn( $fname );
$cleansource = '';
+ $opts = '';
switch(strtoupper($wgOutputEncoding)) {
case 'ISO-8859-1':
- $wgTidyOpts .= ($wgInputEncoding == $wgOutputEncoding)? ' -latin1':' -raw';
+ $opts .= ($wgInputEncoding == $wgOutputEncoding)? ' -latin1':' -raw';
break;
case 'UTF-8':
- $wgTidyOpts .= ($wgInputEncoding == $wgOutputEncoding)? ' -utf8':' -raw';
+ $opts .= ($wgInputEncoding == $wgOutputEncoding)? ' -utf8':' -raw';
break;
default:
- $wgTidyOpts .= ' -raw';
+ $opts .= ' -raw';
}
$wrappedtext = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'.
1 => array('pipe', 'w'),
2 => array('file', '/dev/null', 'a')
);
- $process = proc_open("$wgTidyBin -config $wgTidyConf $wgTidyOpts", $descriptorspec, $pipes);
+ $process = proc_open("$wgTidyBin -config $wgTidyConf $wgTidyOpts$opts", $descriptorspec, $pipes);
if (is_resource($process)) {
fwrite($pipes[0], $wrappedtext);
fclose($pipes[0]);
}
}
- # parse the wiki syntax used to render tables
+ /**
+ * parse the wiki syntax used to render tables
+ *
+ * @access private
+ */
function doTableStuff ( $t ) {
$fname = 'Parser::doTableStuff';
wfProfileIn( $fname );
{
$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
+ # $x is a table row
+ if ( '|+' == substr ( $x , 0 , 2 ) ) {
$fc = '+' ;
$x = substr ( $x , 1 ) ;
}
if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;
$after = explode ( '||' , $after ) ;
$t[$k] = '' ;
+
+ # Loop through each table cell
foreach ( $after AS $theline )
{
$z = '' ;
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' ;
else $l = '' ;
array_push ( $ltd , $l ) ;
+
+ # Cell parameters
$y = explode ( '|' , $theline , 2 ) ;
- if ( count ( $y ) == 1 ) $y = "{$z}<{$l}>{$y[0]}" ;
+ # Note that a '|' inside an invalid link should not
+ # be mistaken as delimiting cell parameters
+ if ( strpos( $y[0], '[[' ) !== false ) {
+ $y = array ($theline);
+ }
+ if ( count ( $y ) == 1 )
+ $y = "{$z}<{$l}>{$y[0]}" ;
else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
$t[$k] .= $y ;
array_push ( $td , true ) ;
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 );
- }
-
+ /**
+ * Helper function for parse() that transforms wiki markup into
+ * HTML. Only called for $mOutputType == OT_HTML.
+ *
+ * @access private
+ */
function internalParse( $text, $linestart, $args = array(), $isMain=true ) {
+ global $wgContLang;
+
$fname = 'Parser::internalParse';
wfProfileIn( $fname );
$text = $wgDateFormatter->reformat( $this->mOptions->getDateFormat(), $text );
}
$text = $this->doAllQuotes( $text );
+ $text = $this->replaceInternalLinks ( $text );
$text = $this->replaceExternalLinks( $text );
$text = $this->doMagicLinks( $text );
- $text = $this->replaceInternalLinks ( $text );
- $text = $this->replaceInternalLinks ( $text );
$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 ) {
+ /**
+ * Replace special strings like "ISBN xxx" and "RFC xxx" with
+ * magic external links.
+ *
+ * @access private
+ */
+ function &doMagicLinks( &$text ) {
+ global $wgUseGeoMode;
$text = $this->magicISBN( $text );
- $text = $this->magicGEO( $text );
+ if ( isset( $wgUseGeoMode ) && $wgUseGeoMode ) {
+ $text = $this->magicGEO( $text );
+ }
$text = $this->magicRFC( $text );
return $text;
}
- # Parse ^^ tokens and return html
- /* private */ function doExponent ( $text )
- {
+ /**
+ * Parse ^^ tokens and return html
+ *
+ * @access 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
- /* private */ function doHeadings( $text ) {
+ /**
+ * Parse headers and return html
+ *
+ * @access private
+ */
+ function doHeadings( $text ) {
$fname = 'Parser::doHeadings';
wfProfileIn( $fname );
for ( $i = 6; $i >= 1; --$i ) {
return $text;
}
- /* private */ function doAllQuotes( $text ) {
+ /**
+ * Replace single quotes with HTML markup
+ * @access private
+ * @return string the altered text
+ */
+ function doAllQuotes( $text ) {
$fname = 'Parser::doAllQuotes';
wfProfileIn( $fname );
$outtext = '';
return $outtext;
}
- /* private */ function doQuotes( $text ) {
+ /**
+ * Helper function for doAllQuotes()
+ * @access private
+ */
+ function doQuotes( $text ) {
$arr = preg_split ("/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE);
if (count ($arr) == 1)
return $text;
# Count the number of occurrences of bold and italics mark-ups.
# We are not counting sequences of five apostrophes.
if (strlen ($arr[$i]) == 2) $numitalics++; else
- if (strlen ($arr[$i]) == 3) $numbold++;
+ if (strlen ($arr[$i]) == 3) $numbold++; else
+ if (strlen ($arr[$i]) == 5) { $numitalics++; $numbold++; }
}
$i++;
}
{
$x1 = substr ($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;
$arr [ $firstmultiletterword-1 ] .= "'";
}
# ... otherwise use the first one that has neither.
- else
+ # (notice that it is possible for all three to be -1 if, for example,
+ # there is only one pentuple-apostrophe in the line)
+ else if ($firstspace > -1)
{
$arr [ $firstspace ] = "''";
$arr [ $firstspace-1 ] .= "'";
{
if (strlen ($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;
}
}
- # Note: we have to do external links before the internal ones,
- # and otherwise take great care in the order of things here, so
- # that we don't end up interpreting some URLs twice.
-
- /* private */ function replaceExternalLinks( $text ) {
+ /**
+ * Replace external links
+ *
+ * Note: we have to do external links before the internal ones,
+ * and otherwise take great care in the order of things here, so
+ * that we don't end up interpreting some URLs twice.
+ *
+ * @access 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;
- }
-
- # Replaces all external links with a given protocol
- /* private */ function subReplaceExternalLinks( $s, $protocol, $autonumber ) {
- $unique = '4jzAfzB8hNvf4sqyO9Edd8pSmk9rE2in0Tgw3';
-
- # URL character class
- $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 = ",;\.:";
+ $sk =& $this->mOptions->getSkin();
+ $linktrail = wfMsgForContent('linktrail');
+ $bits = preg_split( EXT_LINK_BRACKETED, $text, -1, PREG_SPLIT_DELIM_CAPTURE );
- # File name character class, used for identifying images
- $fnc = 'A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF';
- # Recognised image extensions
- $images = 'gif|png|jpg|jpeg';
+ $s = $this->replaceFreeExternalLinks( array_shift( $bits ) );
- # 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).
+ $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];
+ }
+ }
- # Regexp for matching image URLs
- $rxImageURL = "/(^|[^\\[])({$protocol}:)([{$uc}{$sep}]+)\\/([{$fnc}]+)\\." .
- "((?i){$images})([^{$uc}]|$)/";
+ $encUrl = htmlspecialchars( $url );
+ # Bit in parentheses showing the URL for the printable version
+ if( $url == $text || preg_match( "!$protocol://" . preg_quote( $text, '/' ) . "/?$!", $url ) ) {
+ $paren = '';
+ } else {
+ # Expand the URL for printable version
+ if ( ! $sk->suppressUrlExpansion() ) {
+ $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $encUrl ) . "</i>)</span>";
+ } else {
+ $paren = '';
+ }
+ }
- # Regexp for matching non-delimited URLs
- $rxFreeURL = "/(^|[^\\[])({$protocol}:)(([".$uc."]|[".$sep."][".$uc."])+)([^". $uc . $sep. "]|[".$sep."]|$)/";
- $sk =& $this->mOptions->getSkin();
+ # Process the trail (i.e. everything after this link up until start of the next link),
+ # replacing any non-bracketed links
+ $trail = $this->replaceFreeExternalLinks( $trail );
- # Replace image URLs with <img> tags, but only for HTTP and HTTPS ($autonumber=true)
- if ( $autonumber and $this->mOptions->getAllowExternalImages() ) {
- $s = preg_replace( $rxImageURL, '\\1' . $sk->makeImage( "{$unique}:\\3" .
- '/\\4.\\5', '\\4.\\5' ) . '\\6', $s );
+ # 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 .= $sk->makeExternalLink( $url, $text, false ) . $dtrail. $paren . $trail;
}
- # Replace free URLs
- $s = preg_replace( $rxFreeURL, '\\1' . "<a href=\"{$unique}:\\3\"" .
- $sk->getExternalLinkAttributes( "{$unique}:\\3", wfEscapeHTML(
- "{$unique}:\\3" ) ) . ">" . wfEscapeHTML( "{$unique}:\\3" ) .
- '</a>\\5', $s );
- $s = str_replace( $unique, $protocol, $s );
+ wfProfileOut( $fname );
+ return $s;
+ }
- # Search for external links with square brackets by splitting with explode()
- $a = explode( "[{$protocol}:", " " . $s );
- $s = array_shift( $a );
- $s = substr( $s, 1 );
+ /**
+ * Replace anything that looks like a URL with a link
+ * @access private
+ */
+ function replaceFreeExternalLinks( $text ) {
+ $bits = preg_split( '/((?:'.URL_PROTOCOLS.'):)/', $text, -1, PREG_SPLIT_DELIM_CAPTURE );
+ $s = array_shift( $bits );
+ $i = 0;
- # Regexp for URL in square brackets
- $rxWithoutText = "/^([{$uc}{$sep}]+)\\](.*)\$/sD";
- # Regexp for URL with link text in square brackets
- $rxWithText = "/^([{$uc}{$sep}]+)\\s+([^\\]]+)\\](.*)\$/sD";
+ $sk =& $this->mOptions->getSkin();
- # Now interpret each instance of [protocol:
- foreach ( $a as $line ) {
+ while ( $i < count( $bits ) ){
+ $protocol = $bits[$i++];
+ $remainder = $bits[$i++];
- # CASE 1: Link in square brackets, e.g.
- # some text [http://domain.tld/some.link] more text
- if ( preg_match( $rxWithoutText, $line, $m ) ) {
- $link = "{$protocol}:{$m[1]}";
+ if ( preg_match( '/^('.EXT_LINK_URL_CLASS.'+)(.*)$/s', $remainder, $m ) ) {
+ # Found some characters after the protocol that look promising
+ $url = $protocol . $m[1];
$trail = $m[2];
- if ( $autonumber ) { $text = "[" . ++$this->mAutonumber . "]"; }
- else { $text = wfEscapeHTML( $link ); }
- }
- # 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( $rxWithText, $line, $m ) ) {
- $link = "{$protocol}:{$m[1]}";
- $text = $m[2];
- $dtrail = '';
- $trail = $m[3];
- if ( preg_match( wfMsg ('linktrail'), $trail, $m2 ) ) {
- $dtrail = $m2[1];
- $trail = $m2[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 .= ')';
}
- }
- # CASE 3: Nothing matches, just output the source text
- else {
- $s .= "[{$protocol}:" . $line;
- continue;
- }
+ $numSepChars = strspn( strrev( $url ), $sep );
+ if ( $numSepChars ) {
+ $trail = substr( $url, -$numSepChars ) . $trail;
+ $url = substr( $url, 0, -$numSepChars );
+ }
- if( $link == $text || preg_match( "!$protocol://" . preg_quote( $text, "/" ) . "/?$!", $link ) ) {
- $paren = '';
+ # 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 {
- # Expand the URL for printable version
- $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $link ) . "</i>)</span>";
+ $s .= $protocol . $remainder;
}
- $la = $sk->getExternalLinkAttributes( $link, $text );
- $s .= "<a href='{$link}'{$la}>{$text}</a>{$dtrail}{$paren}{$trail}";
-
}
return $s;
}
-
- /* private */ function replaceInternalLinks( $s ) {
- global $wgLang, $wgLinkCache;
- global $wgNamespacesWithSubpages, $wgLanguageCode;
+ /**
+ * make an image if it's allowed
+ * @access private
+ */
+ 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;
+ }
+
+ /**
+ * Process [[ ]] wikilinks
+ *
+ * @access private
+ */
+
+ function replaceInternalLinks( $s ) {
+ global $wgLang, $wgContLang, $wgLinkCache;
static $fname = 'Parser::replaceInternalLinks' ;
+ # use a counter to prevent too much unknown links from
+ # being checked for different language variants.
+ static $convertCount;
wfProfileIn( $fname );
wfProfileIn( $fname.'-setup' );
$redirect = MagicWord::get ( MAG_REDIRECT ) ;
+ #split the entire text string on occurences of [[
$a = explode( '[[', ' ' . $s );
+ #get the first element (all text up to first [[), and remove the space we added
$s = array_shift( $a );
$s = substr( $s, 1 );
# Match a link having the form [[namespace:link|alternate]]trail
static $e1 = FALSE;
if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|([^]]+))?]](.*)\$/sD"; }
+ # Match cases where there is no "]]", which might still be images
+ static $e1_img = FALSE;
+ if ( !$e1_img ) { $e1_img = "/^([{$tc}]+)\\|(.*)\$/sD"; }
# Match the end of a line for a word that's not followed by whitespace,
# e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
static $e2 = '/^(.*?)([a-zA-Z\x80-\xff]+)$/sD';
- $useLinkPrefixExtension = $wgLang->linkPrefixExtension();
- # Special and Media are pseudo-namespaces; no pages actually exist in them
+ $useLinkPrefixExtension = $wgContLang->linkPrefixExtension();
$nottalk = !Namespace::isTalk( $this->mTitle->getNamespace() );
wfProfileOut( $fname.'-setup' );
- foreach ( $a as $line ) {
+ # Loop for each link
+ for ($k = 0; isset( $a[$k] ); $k++) {
+ $line = $a[$k];
wfProfileIn( $fname.'-prefixhandling' );
if ( $useLinkPrefixExtension ) {
if ( preg_match( $e2, $s, $m ) ) {
}
wfProfileOut( $fname.'-prefixhandling' );
+ $might_be_img = false;
+
if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
$text = $m[2];
# fix up urlencoded title texts
if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
$trail = $m[3];
+ } elseif( preg_match($e1_img, $line, $m) ) { # Invalid, but might be an image with a link in its caption
+ $might_be_img = true;
+ $text = $m[2];
+ if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
+ $trail = "";
} else { # Invalid form; output directly
$s .= $prefix . '[[' . $line ;
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
- */
- $c = substr($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
- $m[1]=substr($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
- $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
- }
- } elseif( $noforce ) { # no subpage
- $link = $m[1];
- } else {
- $link = substr( $m[1], 1 );
+ # Don't allow internal links to pages containing
+ # PROTO: where PROTO is a valid URL protocol; these
+ # should be external links.
+ if (preg_match('/((?:'.URL_PROTOCOLS.'):)/', $m[1])) {
+ $s .= $prefix . '[[' . $line ;
+ continue;
}
- $wasblank = ( '' == $text );
- if( $wasblank )
- $text = $link;
+ # Make subpage if necessary
+ $link = $this->maybeDoSubpageLink( $m[1], $text );
+
+ $noforce = (substr($m[1], 0, 1) != ':');
+ if (!$noforce) {
+ # Strip off leading ':'
+ $link = substr($link, 1);
+ }
+
$nt = Title::newFromText( $link );
if( !$nt ) {
$s .= $prefix . '[[' . $line;
continue;
}
+
+ //check other language variants of the link
+ //if the article does not exist
+ global $wgContLang;
+ $variants = $wgContLang->getVariants();
+
+ if(sizeof($variants) > 1 && $convertCount < 200) {
+ $varnt = false;
+ if($nt->getArticleID() == 0) {
+ foreach ( $variants as $v ) {
+ if($v == $wgContLang->getPreferredVariant())
+ continue;
+ $convertCount ++;
+ $varlink = $wgContLang->autoConvert($link, $v);
+ $varnt = Title::newFromText($varlink);
+ if($varnt && $varnt->getArticleID()>0) {
+ break;
+ }
+ }
+ }
+ if($varnt && $varnt->getArticleID()>0) {
+ $nt = $varnt;
+ $link = $varlink;
+ }
+ }
+
$ns = $nt->getNamespace();
$iw = $nt->getInterWiki();
+
+ if ($might_be_img) { # if this is actually an invalid link
+ if ($ns == NS_IMAGE && $noforce) { #but might be an image
+ $found = false;
+ while (isset ($a[$k+1]) ) {
+ #look at the next 'line' to see if we can close it there
+ $next_line = array_shift(array_splice( $a, $k + 1, 1) );
+ if( preg_match("/^(.*?]].*?)]](.*)$/sD", $next_line, $m) ) {
+ # the first ]] closes the inner link, the second the image
+ $found = true;
+ $text .= '[[' . $m[1];
+ $trail = $m[2];
+ break;
+ } elseif( preg_match("/^.*?]].*$/sD", $next_line, $m) ) {
+ #if there's exactly one ]] that's fine, we'll keep looking
+ $text .= '[[' . $m[0];
+ } else {
+ #if $next_line is invalid too, we need look no further
+ $text .= '[[' . $next_line;
+ break;
+ }
+ }
+ if ( !$found ) {
+ # we couldn't find the end of this imageLink, so output it raw
+ #but don't ignore what might be perfectly normal links in the text we've examined
+ $text = $this->replaceInternalLinks($text);
+ $s .= $prefix . '[[' . $link . '|' . $text;
+ # note: no $trail, because without an end, there *is* no trail
+ continue;
+ }
+ } else { #it's not an image, so output it raw
+ $s .= $prefix . '[[' . $link . '|' . $text;
+ # note: no $trail, because without an end, there *is* no trail
+ continue;
+ }
+ }
+
+ $wasblank = ( '' == $text );
+ if( $wasblank ) $text = $link;
+
+
+ # Link not escaped by : , create the various objects
if( $noforce ) {
- if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgLang->getLanguageName( $iw ) ) {
+
+ # Interwikis
+ if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgContLang->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;
+ # recursively parse links inside the image caption
+ # actually, this will parse them in any other parameters, too,
+ # but it might be hard to fix that, and it doesn't matter ATM
+ $text = $this->replaceExternalLinks($text);
+ $text = $this->replaceInternalLinks($text);
+
+ # replace the image with a link-holder so that replaceExternalLinks() can't mess with it
+ $s .= $prefix . $this->insertStripItem( $sk->makeImageLinkObj( $nt, $text ), $this->mStripState ) . $trail;
$wgLinkCache->addImageLinkObj( $nt );
continue;
}
+
if ( $ns == NS_CATEGORY ) {
$t = $nt->getText() ;
- $nnt = Title::newFromText ( Namespace::getCanonicalName(NS_CATEGORY).":".$t ) ;
$wgLinkCache->suspend(); # Don't save in links/brokenlinks
$pPLC=$sk->postParseLinkColour();
$sk->postParseLinkColour( false );
- $t = $sk->makeLinkObj( $nnt, $t, '', '' , $prefix );
+ $t = $sk->makeLinkObj( $nt, $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 ) ) {
+
+ $text = $wgContLang->convert($text);
+
+ 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;
}
+ # Special and Media are pseudo-namespaces; no pages actually exist in them
if( $ns == NS_MEDIA ) {
$s .= $prefix . $sk->makeMediaLinkObj( $nt, $text ) . $trail;
$wgLinkCache->addImageLinkObj( $nt );
return $s;
}
- # Some functions here used by doBlockLevels()
- #
+ /**
+ * Handle link to subpage if necessary
+ * @param $target string the source of the link
+ * @param &$text the link text, modified as necessary
+ * @return string the full name of the link
+ * @access private
+ */
+ function maybeDoSubpageLink($target, &$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
+ global $wgNamespacesWithSubpages;
+
+ $fname = 'Parser::maybeDoSubpageLink';
+ wfProfileIn( $fname );
+ # Look at the first character
+ if( $target{0} == '/' ) {
+ # / at end means we don't want the slash to be shown
+ if(substr($target,-1,1)=='/') {
+ $target=substr($target,1,-1);
+ $noslash=$target;
+ } else {
+ $noslash=substr($target,1);
+ }
+
+ # Some namespaces don't allow subpages
+ if(!empty($wgNamespacesWithSubpages[$this->mTitle->getNamespace()])) {
+ # subpages allowed here
+ $ret = $this->mTitle->getPrefixedText(). '/' . trim($noslash);
+ if( '' === $text ) {
+ $text = $target;
+ } # this might be changed for ugliness reasons
+ } else {
+ # no subpage allowed, use standard link
+ $ret = $target;
+ }
+ } else {
+ # no subpage
+ $ret = $target;
+ }
+
+ wfProfileOut( $fname );
+ return $ret;
+ }
+
+ /**#@+
+ * Used by doBlockLevels()
+ * @access private
+ */
/* private */ function closeParagraph() {
$result = '';
if ( '' != $this->mLastSection ) {
# 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 ) {
else { return '<!-- ERR 3 -->'; }
return $text."\n";
}
-
- /* private */ function doBlockLevels( $text, $linestart ) {
+ /**#@-*/
+
+ /**
+ * Make lists from lines starting with ':', '*', '#', etc.
+ *
+ * @access private
+ * @return string the lists rendered as HTML
+ */
+ function doBlockLevels( $text, $linestart ) {
$fname = 'Parser::doBlockLevels';
wfProfileIn( $fname );
}
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( $pref, -1 ) == ';') {
# The one nasty exception: definition lists work like this:
# ; title : definition text
# So we check for : in the remainder text to split up the
# title and definition, without b0rking links.
- # FIXME: This is not foolproof. Something better in Tokenizer might help.
- if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
- $term = $match[1];
+ if ($this->findColonNoLinks($t, $term, $t2) !== false) {
+ $t = $t2;
$output .= $term . $this->nextItem( ':' );
- $t = $match[2];
}
}
} elseif( $prefixLength || $lastPrefixLength ) {
if ( ';' == $char ) {
# FIXME: This is dupe of code above
- if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
- $term = $match[1];
- $output .= $term . $this->nextItem( ":" );
- $t = $match[2];
+ if ($this->findColonNoLinks($t, $term, $t2) !== false) {
+ $t = $t2;
+ $output .= $term . $this->nextItem( ':' );
}
}
++$commonPrefixLength;
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 $output;
}
- # Return value of a magic variable (like PAGENAME)
+ /**
+ * Split up a string on ':', ignoring any occurences inside
+ * <a>..</a> or <span>...</span>
+ * @param $str string the string to split
+ * @param &$before string set to everything before the ':'
+ * @param &$after string set to everything after the ':'
+ * return string the position of the ':', or false if none found
+ */
+ function findColonNoLinks($str, &$before, &$after) {
+ # I wonder if we should make this count all tags, not just <a>
+ # and <span>. That would prevent us from matching a ':' that
+ # comes in the middle of italics other such formatting....
+ # -- Wil
+ $fname = 'Parser::findColonNoLinks';
+ wfProfileIn( $fname );
+ $pos = 0;
+ do {
+ $colon = strpos($str, ':', $pos);
+
+ if ($colon !== false) {
+ $before = substr($str, 0, $colon);
+ $after = substr($str, $colon + 1);
+
+ # Skip any ':' within <a> or <span> pairs
+ $a = substr_count($before, '<a');
+ $s = substr_count($before, '<span');
+ $ca = substr_count($before, '</a>');
+ $cs = substr_count($before, '</span>');
+
+ if ($a <= $ca and $s <= $cs) {
+ # Tags are balanced before ':'; ok
+ break;
+ }
+ $pos = $colon + 1;
+ }
+ } while ($colon !== false);
+ wfProfileOut( $fname );
+ return $colon;
+ }
+
+ /**
+ * Return value of a magic variable (like PAGENAME)
+ *
+ * @access private
+ */
function getVariableValue( $index ) {
- global $wgLang, $wgSitename, $wgServer;
+ global $wgContLang, $wgSitename, $wgServer;
switch ( $index ) {
case MAG_CURRENTMONTH:
- return $wgLang->formatNum( date( 'm' ) );
+ return $wgContLang->formatNum( date( 'm' ) );
case MAG_CURRENTMONTHNAME:
- return $wgLang->getMonthName( date('n') );
+ return $wgContLang->getMonthName( date('n') );
case MAG_CURRENTMONTHNAMEGEN:
- return $wgLang->getMonthNameGen( date('n') );
+ return $wgContLang->getMonthNameGen( date('n') );
case MAG_CURRENTDAY:
- return $wgLang->formatNum( date('j') );
+ return $wgContLang->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 $wgContLang->getNsText($this->mTitle->getNamespace()); # Patch by Dori
case MAG_CURRENTDAYNAME:
- return $wgLang->getWeekdayName( date('w')+1 );
+ return $wgContLang->getWeekdayName( date('w')+1 );
case MAG_CURRENTYEAR:
- return $wgLang->formatNum( date( 'Y' ) );
+ return $wgContLang->formatNum( date( 'Y' ) );
case MAG_CURRENTTIME:
- return $wgLang->time( wfTimestampNow(), false );
+ return $wgContLang->time( wfTimestampNow(), false );
case MAG_NUMBEROFARTICLES:
- return $wgLang->formatNum( wfNumberOfArticles() );
+ return $wgContLang->formatNum( wfNumberOfArticles() );
case MAG_SITENAME:
return $wgSitename;
case MAG_SERVER:
}
}
- # initialise the magic variables (like CURRENTMONTHNAME)
+ /**
+ * initialise the magic variables (like CURRENTMONTHNAME)
+ *
+ * @access private
+ */
function initialiseVariables() {
+ $fname = 'Parser::initialiseVariables';
+ wfProfileIn( $fname );
global $wgVariableIDs;
$this->mVariables = array();
foreach ( $wgVariableIDs as $id ) {
$mw =& MagicWord::get( $id );
$mw->addToArray( $this->mVariables, $this->getVariableValue( $id ) );
}
+ wfProfileOut( $fname );
}
- /* private */ function replaceVariables( $text, $args = array() ) {
+ /**
+ * Replace magic variables, templates, and template arguments
+ * with the appropriate text. Templates are substituted recursively,
+ * taking care to avoid infinite loops.
+ *
+ * Note that the substitution depends on value of $mOutputType:
+ * OT_WIKI: only {{subst:}} templates
+ * OT_MSG: only magic variables
+ * OT_HTML: all templates and magic variables
+ *
+ * @param string $tex The text to transform
+ * @param array $args Key-value pairs representing template parameters to substitute
+ * @access 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 );
- $bail = false;
$titleChars = Title::legalChars();
- $nonBraceChars = str_replace( array( '{', '}' ), array( '', '' ), $titleChars );
# This function is called recursively. To keep track of arguments we need a stack:
array_push( $this->mArgStack, $args );
# PHP global rebinding syntax is a bit weird, need to use the GLOBALS array
$GLOBALS['wgCurParser'] =& $this;
-
- if ( $this->mOutputType == OT_HTML ) {
- # Variable substitution
- $text = preg_replace_callback( "/{{([$nonBraceChars]*?)}}/", 'wfVariableSubstitution', $text );
-
+ # Variable substitution
+ $text = preg_replace_callback( "/{{([$titleChars]*?)}}/", 'wfVariableSubstitution', $text );
+
+ if ( $this->mOutputType == OT_HTML || $this->mOutputType == OT_WIKI ) {
# Argument substitution
- $text = preg_replace_callback( "/(\\n?){{{([$titleChars]*?)}}}/", 'wfArgSubstitution', $text );
+ $text = preg_replace_callback( "/{{{([$titleChars]*?)}}}/", 'wfArgSubstitution', $text );
}
# Template substitution
- $regex = '/(\\n?){{(['.$nonBraceChars.']*)(\\|.*?|)}}/s';
+ $regex = '/(\\n|{)?{{(['.$titleChars.']*)(\\|.*?|)}}/s';
$text = preg_replace_callback( $regex, 'wfBraceSubstitution', $text );
array_pop( $this->mArgStack );
return $text;
}
+ /**
+ * Replace magic variables
+ * @access private
+ */
function variableSubstitution( $matches ) {
if ( !$this->mVariables ) {
$this->initialiseVariables();
}
- if ( array_key_exists( $matches[1], $this->mVariables ) ) {
+ $skip = false;
+ if ( $this->mOutputType == OT_WIKI ) {
+ # Do only magic variables prefixed by SUBST
+ $mwSubst =& MagicWord::get( MAG_SUBST );
+ if (!$mwSubst->matchStartAndRemove( $matches[1] ))
+ $skip = true;
+ # Note that if we don't substitute the variable below,
+ # we don't remove the {{subst:}} magic word, in case
+ # it is a template rather than a magic variable.
+ }
+ if ( !$skip && array_key_exists( $matches[1], $this->mVariables ) ) {
$text = $this->mVariables[$matches[1]];
$this->mOutput->mContainsOldMagic = true;
} else {
$argc = count($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--;
return $args;
}
+ /**
+ * Return the text of a template, after recursively
+ * replacing any variables or templates within the template.
+ *
+ * @param array $matches The parts of the template
+ * $matches[1]: the title, i.e. the part before the |
+ * $matches[2]: the parameters (including a leading |), if any
+ * @return string the text of the template
+ * @access private
+ */
function braceSubstitution( $matches ) {
- global $wgLinkCache, $wgLang;
+ global $wgLinkCache, $wgContLang;
$fname = 'Parser::braceSubstitution';
$found = false;
$nowiki = false;
$title = NULL;
- # $newline is an optional newline character before the braces
+ # Need to know if the template comes at the start of a line,
+ # to treat the beginning of the template like the beginning
+ # of a line for tables and block-level elements.
+ $linestart = $matches[1];
+
# $part1 is the bit before the first |, and must contain only title characters
# $args is a list of arguments, starting from index 0, not including $part1
- $newline = $matches[1];
$part1 = $matches[2];
# If the third subpattern matched anything, it will start with |
$args = $this->getTemplateArgs($matches[3]);
$argc = count( $args );
- # {{{}}}
- if ( strpos( $matches[0], '{{{' ) !== false ) {
+ # Don't parse {{{}}} because that's only for template arguments
+ if ( $linestart === '{' ) {
$text = $matches[0];
$found = true;
$noparse = true;
# SUBST
if ( !$found ) {
$mwSubst =& MagicWord::get( MAG_SUBST );
- if ( $mwSubst->matchStartAndRemove( $part1 ) ) {
- if ( $this->mOutputType != OT_WIKI ) {
- # Invalid SUBST not replaced at PST time
- # Return without further processing
- $text = $matches[0];
- $found = true;
- $noparse= true;
- }
- } elseif ( $this->mOutputType == OT_WIKI ) {
- # SUBST not found in PST pass, do nothing
+ if ( $mwSubst->matchStartAndRemove( $part1 ) xor ($this->mOutputType == OT_WIKI) ) {
+ # One of two possibilities is true:
+ # 1) Found SUBST but not in the PST phase
+ # 2) Didn't find SUBST and in the PST phase
+ # In either case, return without further processing
$text = $matches[0];
$found = true;
+ $noparse = true;
}
}
$mwInt =& MagicWord::get( MAG_INT );
if ( $mwInt->matchStartAndRemove( $part1 ) ) {
if ( $this->incrementIncludeCount( 'int:'.$part1 ) ) {
- $text = wfMsgReal( $part1, $args, true );
+ $text = $linestart . wfMsgReal( $part1, $args, true );
$found = true;
}
}
$mwNs = MagicWord::get( MAG_NS );
if ( $mwNs->matchStartAndRemove( $part1 ) ) {
if ( intval( $part1 ) ) {
- $text = $wgLang->getNsText( intval( $part1 ) );
+ $text = $linestart . $wgContLang->getNsText( intval( $part1 ) );
$found = true;
} else {
$index = Namespace::getCanonicalIndex( strtolower( $part1 ) );
if ( !is_null( $index ) ) {
- $text = $wgLang->getNsText( $index );
+ $text = $linestart . $wgContLang->getNsText( $index );
$found = true;
}
}
$title = Title::newFromText( $part1 );
if ( !is_null( $title ) ) {
if ( $argc > 0 ) {
- $text = $title->$func( $args[0] );
+ $text = $linestart . $title->$func( $args[0] );
} else {
- $text = $title->$func();
+ $text = $linestart . $title->$func();
}
$found = true;
}
}
}
- # Internal variables
- if ( !$this->mVariables ) {
- $this->initialiseVariables();
- }
- if ( !$found && array_key_exists( $part1, $this->mVariables ) ) {
- $text = $this->mVariables[$part1];
- $found = true;
- $this->mOutput->mContainsOldMagic = true;
+ # GRAMMAR
+ if ( !$found && $argc == 1 ) {
+ $mwGrammar =& MagicWord::get( MAG_GRAMMAR );
+ if ( $mwGrammar->matchStartAndRemove( $part1 ) ) {
+ $text = $linestart . $wgContLang->convertGrammar( $args[0], $part1 );
+ $found = true;
+ }
}
- # Template table test
+ # Template table test
# Did we encounter this template already? If yes, it is in the cache
# and we need to check for loops.
- if ( isset( $this->mTemplates[$part1] ) ) {
+ if ( !$found && isset( $this->mTemplates[$part1] ) ) {
+ # set $text to cached message.
+ $text = $linestart . $this->mTemplates[$part1];
+ $found = true;
+
# Infinite loop test
if ( isset( $this->mTemplatePath[$part1] ) ) {
$noparse = true;
$found = true;
+ $text .= '<!-- WARNING: template loop detected -->';
}
- # set $text to cached message.
- $text = $this->mTemplates[$part1];
- $found = true;
}
# Load from database
+ $itcamefromthedatabase = false;
if ( !$found ) {
- $title = Title::newFromText( $part1, NS_TEMPLATE );
+ $ns = NS_TEMPLATE;
+ $part1 = $this->maybeDoSubpageLink( $part1, $subpage='' );
+ if ($subpage !== '') {
+ $ns = $this->mTitle->getNamespace();
+ }
+ $title = Title::newFromText( $part1, $ns );
if ( !is_null( $title ) && !$title->isExternal() ) {
# Check for excessive inclusion
$dbk = $title->getPrefixedDBkey();
$articleContent = $article->getContentWithoutUsingSoManyDamnGlobals();
if ( $articleContent !== false ) {
$found = true;
- $text = $articleContent;
-
+ $text = $linestart . $articleContent;
+ $itcamefromthedatabase = true;
}
}
# If the title is valid but undisplayable, make a link to it
if ( $this->mOutputType == OT_HTML && !$found ) {
- $text = '[[' . $title->getPrefixedText() . ']]';
+ $text = $linestart . '[['.$title->getPrefixedText().']]';
$found = true;
}
- # Template cache array insertion
+ # Template cache array insertion
$this->mTemplates[$part1] = $text;
}
}
# Only for HTML output
if ( $nowiki && $found && $this->mOutputType == OT_HTML ) {
$text = wfEscapeWikiText( $text );
- } elseif ( $this->mOutputType == OT_HTML && $found && !$noparse) {
+ } elseif ( ($this->mOutputType == OT_HTML || $this->mOutputType == OT_WIKI) && $found && !$noparse) {
# Clean up argument array
$assocArgs = array();
$index = 1;
}
}
- # Do not enter included links in link table
- if ( !is_null( $title ) ) {
- $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->internalParse( $text, $newline, $assocArgs );
- # I replaced the line below with the line above, as it former seems to cause several bugs
- #$text = $this->stripParse( $text, $newline, $assocArgs );
+ $text = $this->strip( $text, $this->mStripState );
+ $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->resume();
+ if ( $this->mOutputType == OT_HTML && !is_null( $title ) ) {
$wgLinkCache->addLinkObj( $title );
}
+
+ # If the template begins with a table or block-level
+ # element, it should be treated as beginning a new line.
+ if ($linestart !== '\n' && preg_match('/^({\\||:|;|#|\*)/', $text)) {
+ $text = "\n" . $text;
+ }
}
- # Empties the template path
- $this->mTemplatePath = array();
+ # Empties the template path
+ $this->mTemplatePath = array();
+ if ( !$found ) {
+ return $matches[0];
+ } else {
+ # replace ==section headers==
+ # XXX this needs to go away once we have a better parser.
+ if ( $this->mOutputType != OT_WIKI && $itcamefromthedatabase ) {
+ if( !is_null( $title ) )
+ $encodedname = base64_encode($title->getPrefixedDBkey());
+ else
+ $encodedname = base64_encode("");
+ $m = preg_split('/(^={1,6}.*?={1,6}\s*?$)/m', $text, -1,
+ PREG_SPLIT_DELIM_CAPTURE);
+ $text = '';
+ $nsec = 0;
+ for( $i = 0; $i < count($m); $i += 2 ) {
+ $text .= $m[$i];
+ if (!isset($m[$i + 1]) || $m[$i + 1] == "") continue;
+ $hl = $m[$i + 1];
+ if( strstr($hl, "<!--MWTEMPLATESECTION") ) {
+ $text .= $hl;
+ continue;
+ }
+ preg_match('/^(={1,6})(.*?)(={1,6})\s*?$/m', $hl, $m2);
+ $text .= $m2[1] . $m2[2] . "<!--MWTEMPLATESECTION="
+ . $encodedname . "&" . base64_encode("$nsec") . "-->" . $m2[3];
+
+ $nsec++;
+ }
+ }
+ }
+
+ # Empties the template path
+ $this->mTemplatePath = array();
if ( !$found ) {
return $matches[0];
} else {
}
}
- # Triple brace replacement -- used for template arguments
+ /**
+ * Triple brace replacement -- used for template arguments
+ * @access private
+ */
function argSubstitution( $matches ) {
- $newline = $matches[1];
- $arg = trim( $matches[2] );
+ $arg = trim( $matches[1] );
$text = $matches[0];
$inputArgs = end( $this->mArgStack );
if ( array_key_exists( $arg, $inputArgs ) ) {
- $text = $this->stripParse( $inputArgs[$arg], $newline, array() );
+ $text = $inputArgs[$arg];
}
return $text;
}
- # Returns true if the function is allowed to include this entity
+ /**
+ * Returns true if the function is allowed to include this entity
+ * @access private
+ */
function incrementIncludeCount( $dbk ) {
if ( !array_key_exists( $dbk, $this->mIncludeCount ) ) {
$this->mIncludeCount[$dbk] = 0;
}
- # Cleans up HTML, removes dangerous tags and attributes
- /* private */ function removeHTMLtags( $text ) {
+ /**
+ * Cleans up HTML, removes dangerous tags and attributes, and
+ * removes HTML comments
+ * @access private
+ */
+ function removeHTMLtags( $text ) {
global $wgUseTidy, $wgUserHtml;
$fname = 'Parser::removeHTMLtags';
wfProfileIn( $fname );
$htmlattrs = $this->getHTMLattrs () ;
# Remove HTML comments
- $text = preg_replace( '/(\\n *<!--.*--> *(?=\\n)|<!--.*-->)/sU', '$2', $text );
+ $text = $this->removeHTMLcomments( $text );
$bits = explode( '<', $text );
$text = array_shift( $bits );
return $text;
}
+ /**
+ * Remove '<!--', '-->', and everything between.
+ * To avoid leaving blank lines, when a comment is both preceded
+ * and followed by a newline (ignoring spaces), trim leading and
+ * trailing spaces and one of the newlines.
+ *
+ * @access private
+ */
+ function removeHTMLcomments( $text ) {
+ $fname='Parser::removeHTMLcomments';
+ wfProfileIn( $fname );
+ while (($start = strpos($text, '<!--')) !== false) {
+ $end = strpos($text, '-->', $start + 4);
+ if ($end === false) {
+ # Unterminated comment; bail out
+ break;
+ }
-/*
- *
- * 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.
- *
- */
+ $end += 3;
+ # Trim space and newline if the comment is both
+ # preceded and followed by a newline
+ $spaceStart = max($start - 1, 0);
+ $spaceLen = $end - $spaceStart;
+ while (substr($text, $spaceStart, 1) === ' ' && $spaceStart > 0) {
+ $spaceStart--;
+ $spaceLen++;
+ }
+ while (substr($text, $spaceStart + $spaceLen, 1) === ' ')
+ $spaceLen++;
+ if (substr($text, $spaceStart, 1) === "\n" and substr($text, $spaceStart + $spaceLen, 1) === "\n") {
+ # Remove the comment, leading and trailing
+ # spaces, and leave only one newline.
+ $text = substr_replace($text, "\n", $spaceStart, $spaceLen + 1);
+ }
+ else {
+ # Remove just the comment.
+ $text = substr_replace($text, '', $start, $end - $start);
+ }
+ }
+ wfProfileOut( $fname );
+ return $text;
+ }
+
+ /**
+ * This function accomplishes several tasks:
+ * 1) Auto-number headings if that option is enabled
+ * 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.
+ * @access private
+ */
/* private */ function formatHeadings( $text, $isMain=true ) {
- global $wgInputEncoding, $wgMaxTocLevel;
+ global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders;
$doNumberHeadings = $this->mOptions->getNumberHeadings();
$doShowToc = $this->mOptions->getShowToc();
# headline counter
$headlineCount = 0;
+ $sectionCount = 0; # headlineCount excluding template sections
# Ugh .. the TOC should have neat indentation levels which can be
# passed to the skin functions. These are determined here
$level = 0;
$prevlevel = 0;
foreach( $matches[3] as $headline ) {
+ $istemplate = 0;
+ $templatetitle = "";
+ $templatesection = 0;
+
+ if (preg_match("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", $headline, $mat)) {
+ $istemplate = 1;
+ $templatetitle = base64_decode($mat[1]);
+ $templatesection = 1 + (int)base64_decode($mat[2]);
+ $headline = preg_replace("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", "", $headline);
+ }
+
$numbering = '';
if( $level ) {
$prevlevel = $level;
if( $dot ) {
$numbering .= '.';
}
- $numbering .= $sublevelCount[$i];
+ $numbering .= $wgContLang->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 number-->
+ # turns into
+ # link text with suffix
+ $canonized_headline = preg_replace( '/<!--LINK ([0-9]*)-->/e',
+ "\$wgLinkHolders['texts'][\$1]",
+ $canonized_headline );
+
# strip out HTML
$canonized_headline = preg_replace( '/<.*?' . '>/','',$canonized_headline );
$tocline = trim( $canonized_headline );
if( $doShowToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) {
$toc .= $sk->tocLine($anchor,$tocline,$toclevel);
}
- if( $showEditLink ) {
+ if( $showEditLink && ( !$istemplate || $templatetitle !== "" ) ) {
if ( empty( $head[$headlineCount] ) ) {
$head[$headlineCount] = '';
}
- $head[$headlineCount] .= $sk->editSectionLink($headlineCount+1);
+ if( $istemplate )
+ $head[$headlineCount] .= $sk->editSectionLinkForOther($templatetitle, $templatesection);
+ else
+ $head[$headlineCount] .= $sk->editSectionLink($sectionCount+1);
}
# Add the edit section span
if( $rightClickHack ) {
- $headline = $sk->editSectionScript($headlineCount+1,$headline);
+ if( $istemplate )
+ $headline = $sk->editSectionScriptForOther($templatetitle, $templatesection, $headline);
+ else
+ $headline = $sk->editSectionScript($sectionCount+1,$headline);
}
# give headline the correct <h#> tag
- @$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.'>';
$headlineCount++;
+ if( !$istemplate )
+ $sectionCount++;
}
if( $doShowToc ) {
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
}
}
- # Return an HTML link for the "ISBN 123456" text
- /* private */ function magicISBN( $text ) {
+ /**
+ * Return an HTML link for the "ISBN 123456" text
+ * @access private
+ */
+ function magicISBN( $text ) {
global $wgLang;
$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 $text;
}
- # Return an HTML link for the "GEO ..." text
- /* private */ function magicGEO( $text ) {
+ /**
+ * Return an HTML link for the "GEO ..." text
+ * @access 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;
}
return $text;
}
- # Return an HTML link for the "RFC 1234" text
- /* private */ function magicRFC( $text ) {
+ /**
+ * Return an HTML link for the "RFC 1234" text
+ * @access private
+ * @param string $text text to be processed
+ */
+ function magicRFC( $text ) {
global $wgLang;
+
+ $valid = '0123456789';
+ $internal = false;
$a = split( 'RFC ', ' '.$text );
if ( count ( $a ) < 2 ) return $text;
$text = substr( array_shift( $a ), 1);
- $valid = '0123456789';
+
+ /* Check if RFC keyword is preceed by [[.
+ * This test is made here cause of the array_shift above
+ * that prevent the test to be done in the foreach.
+ */
+ if(substr($text, -2) == '[[') { $internal = true; }
foreach ( $a as $x ) {
+ /* token might be empty if we have RFC RFC 1234 */
+ if($x=='') {
+ $text.='RFC ';
+ continue;
+ }
+
$rfc = $blank = '' ;
- while ( ' ' == $x{0} ) {
+
+ /** remove and save whitespaces in $blank */
+ while ( $x{0} == ' ' ) {
$blank .= ' ';
$x = substr( $x, 1 );
}
+
+ /** remove and save the rfc number in $rfc */
while ( strstr( $valid, $x{0} ) != false ) {
$rfc .= $x{0};
$x = substr( $x, 1 );
}
- if ( '' == $rfc ) {
+ if ( $rfc == '') {
+ /* call back stripped spaces*/
$text .= "RFC $blank$x";
+ } elseif( $internal) {
+ /* normal link */
+ $text .= "RFC $rfc$x";
} else {
+ /* build the external link*/
$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}";
}
+
+ /* Check if the next RFC keyword is preceed by [[ */
+ $internal = (substr($x,-2) == '[[');
}
return $text;
}
+ /**
+ * Transform wiki markup when saving a page by doing \r\n -> \n
+ * conversion, substitting signatures, {{subst:}} templates, etc.
+ *
+ * @param string $text the text to transform
+ * @param Title &$title the Title object for the current article
+ * @param User &$user the User object describing the current user
+ * @param ParserOptions $options parsing options
+ * @param bool $clearState whether to clear the parser state first
+ * @return string the altered wiki markup
+ * @access public
+ */
function preSaveTransform( $text, &$title, &$user, $options, $clearState = true ) {
$this->mOptions = $options;
$this->mTitle =& $title;
return $text;
}
- /* private */ function pstPass2( $text, &$user ) {
- global $wgLang, $wgLocaltimezone, $wgCurParser;
+ /**
+ * Pre-save transform helper function
+ * @access private
+ */
+ function pstPass2( $text, &$user ) {
+ global $wgLang, $wgContLang, $wgLocaltimezone, $wgCurParser;
# Variable replacement
# Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
$oldtz = getenv('TZ'); putenv('TZ='.$wgLocaltimezone);
}
/* Note: this is an ugly timezone hack for the European wikis */
- $d = $wgLang->timeanddate( date( 'YmdHis' ), false ) .
+ $d = $wgContLang->timeanddate( date( 'YmdHis' ), false ) .
' (' . date( 'T' ) . ')';
if(isset($wgLocaltimezone)) putenv('TZ='.$oldtzs);
$text = preg_replace( '/~~~~~/', $d, $text );
- $text = preg_replace( '/~~~~/', '[[' . $wgLang->getNsText( NS_USER ) . ":$n|$k]] $d", $text );
- $text = preg_replace( '/~~~/', '[[' . $wgLang->getNsText( NS_USER ) . ":$n|$k]]", $text );
+ $text = preg_replace( '/~~~~/', '[[' . $wgContLang->getNsText( NS_USER ) . ":$n|$k]] $d", $text );
+ $text = preg_replace( '/~~~/', '[[' . $wgContLang->getNsText( NS_USER ) . ":$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];
$text = preg_replace( $p2, "[[\\1 ({$context})|\\1]]", $text );
}
- /*
- $mw =& MagicWord::get( MAG_SUBST );
- $wgCurParser = $this->fork();
- $text = $mw->substituteCallback( $text, "wfBraceSubstitution" );
- $this->merge( $wgCurParser );
- */
-
# Trim trailing whitespace
# MAG_END (__END__) tag allows for trailing
# whitespace to be deliberately included
return $text;
}
- # Set up some variables which are usually set up in parse()
- # so that an external function can call some class members with confidence
+ /**
+ * Set up some variables which are usually set up in parse()
+ * so that an external function can call some class members with confidence
+ * @access public
+ */
function startExternalParse( &$title, $options, $outputType, $clearState = true ) {
$this->mTitle =& $title;
$this->mOptions = $options;
}
}
+ /**
+ * Transform a MediaWiki message by replacing magic variables.
+ *
+ * @param string $text the text to transform
+ * @param ParserOptions $options options
+ * @return string the text with variables substituted
+ * @access public
+ */
function transformMsg( $text, $options ) {
global $wgTitle;
static $executing = false;
return $text;
}
- # Create an HTML-style tag, e.g. <yourtag>special text</yourtag>
- # Callback will be called with the text within
- # Transform and return the text within
+ /**
+ * Create an HTML-style tag, e.g. <yourtag>special text</yourtag>
+ * Callback will be called with the text within
+ * Transform and return the text within
+ * @access public
+ */
function setHook( $tag, $callback ) {
$oldVal = @$this->mTagHooks[$tag];
$this->mTagHooks[$tag] = $callback;
}
}
+/**
+ * @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 );
+}
+
?>
dépôts / lhc / web / wiklou.git / blobdiff