4 * File for Parser and related classes
11 * Variable substitution O(N^2) attack
13 * Without countermeasures, it would be possible to attack the parser by saving
14 * a page filled with a large number of inclusions of large pages. The size of
15 * the generated page would be proportional to the square of the input size.
16 * Hence, we limit the number of inclusions of any given page, thus bringing any
17 * attack back to O(N).
20 define( 'MAX_INCLUDE_REPEAT', 100 );
21 define( 'MAX_INCLUDE_SIZE', 1000000 ); // 1 Million
23 define( 'RLH_FOR_UPDATE', 1 );
25 # Allowed values for $mOutputType
26 define( 'OT_HTML', 1 );
27 define( 'OT_WIKI', 2 );
28 define( 'OT_MSG' , 3 );
30 # string parameter for extractTags which will cause it
31 # to strip HTML comments in addition to regular
32 # <XML>-style tags. This should not be anything we
33 # may want to use in wikisyntax
34 define( 'STRIP_COMMENTS', 'HTMLCommentStrip' );
36 # prefix for escaping, used in two functions at least
37 define( 'UNIQ_PREFIX', 'NaodW29');
39 # Constants needed for external link processing
40 define( 'URL_PROTOCOLS', 'http|https|ftp|irc|gopher|news|mailto' );
41 define( 'HTTP_PROTOCOLS', 'http|https' );
42 # Everything except bracket, space, or control characters
43 define( 'EXT_LINK_URL_CLASS', '[^]<>"\\x00-\\x20\\x7F]' );
45 define( 'EXT_LINK_TEXT_CLASS', '[^\]\\x00-\\x1F\\x7F]' );
46 define( 'EXT_IMAGE_FNAME_CLASS', '[A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]' );
47 define( 'EXT_IMAGE_EXTENSIONS', 'gif|png|jpg|jpeg' );
48 define( 'EXT_LINK_BRACKETED', '/\[(('.URL_PROTOCOLS
.'):'.EXT_LINK_URL_CLASS
.'+) *('.EXT_LINK_TEXT_CLASS
.'*?)\]/S' );
49 define( 'EXT_IMAGE_REGEX',
50 '/^('.HTTP_PROTOCOLS
.':)'. # Protocol
51 '('.EXT_LINK_URL_CLASS
.'+)\\/'. # Hostname and path
52 '('.EXT_IMAGE_FNAME_CLASS
.'+)\\.((?i)'.EXT_IMAGE_EXTENSIONS
.')$/S' # Filename
58 * Processes wiki markup
61 * There are three main entry points into the Parser class:
63 * produces HTML output
65 * produces altered wiki markup.
67 * performs brace substitution on MediaWiki messages
70 * objects: $wgLang, $wgDateFormatter, $wgLinkCache
72 * NOT $wgArticle, $wgUser or $wgTitle. Keep them away!
75 * $wgUseTex*, $wgUseDynamicDates*, $wgInterwikiMagic*,
76 * $wgNamespacesWithSubpages, $wgAllowExternalImages*,
79 * * only within ParserOptions
92 # Cleared with clearState():
93 var $mOutput, $mAutonumber, $mDTopen, $mStripState = array();
94 var $mVariables, $mIncludeCount, $mArgStack, $mLastSection, $mInPre;
97 var $mOptions, $mTitle, $mOutputType,
98 $mTemplates, // cache of already loaded templates, avoids
99 // multiple SQL queries for the same string
100 $mTemplatePath; // stores an unsorted hash of all the templates already loaded
101 // in this path. Used for loop detection.
111 $this->mTemplates
= array();
112 $this->mTemplatePath
= array();
113 $this->mTagHooks
= array();
122 function clearState() {
123 $this->mOutput
= new ParserOutput
;
124 $this->mAutonumber
= 0;
125 $this->mLastSection
= "";
126 $this->mDTopen
= false;
127 $this->mVariables
= false;
128 $this->mIncludeCount
= array();
129 $this->mStripState
= array();
130 $this->mArgStack
= array();
131 $this->mInPre
= false;
135 * First pass--just handle <nowiki> sections, pass the rest off
136 * to internalParse() which does all the real work.
139 * @return ParserOutput a ParserOutput
141 function parse( $text, &$title, $options, $linestart = true, $clearState = true ) {
142 global $wgUseTidy, $wgContLang;
143 $fname = 'Parser::parse';
144 wfProfileIn( $fname );
150 $this->mOptions
= $options;
151 $this->mTitle
=& $title;
152 $this->mOutputType
= OT_HTML
;
155 $text = $this->strip( $text, $this->mStripState
);
157 $text = $this->internalParse( $text, $linestart );
158 $text = $this->unstrip( $text, $this->mStripState
);
159 # Clean up special characters, only run once, next-to-last before doBlockLevels
162 # french spaces, last one Guillemet-left
163 # only if there is something before the space
164 '/(.) (?=\\?|:|;|!|\\302\\273)/i' => '\\1 \\2',
165 # french spaces, Guillemet-right
166 "/(\\302\\253) /i"=>"\\1 ",
167 '/<hr *>/i' => '<hr />',
168 '/<br *>/i' => '<br />',
169 '/<center *>/i' => '<div class="center">',
170 '/<\\/center *>/i' => '</div>',
171 # Clean up spare ampersands; note that we probably ought to be
172 # more careful about named entities.
173 '/&(?!:amp;|#[Xx][0-9A-fa-f]+;|#[0-9]+;|[a-zA-Z0-9]+;)/' => '&'
175 $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
178 # french spaces, last one Guillemet-left
179 '/ (\\?|:|;|!|\\302\\273)/i' => ' \\1',
180 # french spaces, Guillemet-right
181 '/(\\302\\253) /i' => '\\1 ',
182 '/<center *>/i' => '<div class="center">',
183 '/<\\/center *>/i' => '</div>'
185 $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
188 $text = $this->doBlockLevels( $text, $linestart );
190 $this->replaceLinkHolders( $text );
191 $text = $wgContLang->convert($text);
193 $text = $this->unstripNoWiki( $text, $this->mStripState
);
196 $text = Parser
::tidy($text);
199 $this->mOutput
->setText( $text );
200 wfProfileOut( $fname );
201 return $this->mOutput
;
205 * Get a random string
210 function getRandomString() {
211 return dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff));
215 * Replaces all occurrences of <$tag>content</$tag> in the text
216 * with a random marker and returns the new text. the output parameter
217 * $content will be an associative array filled with data on the form
218 * $unique_marker => content.
220 * If $content is already set, the additional entries will be appended
221 * If $tag is set to STRIP_COMMENTS, the function will extract
222 * <!-- HTML comments -->
227 function extractTags($tag, $text, &$content, $uniq_prefix = ''){
228 $rnd = $uniq_prefix . '-' . $tag . Parser
::getRandomString();
235 while ( '' != $text ) {
236 if($tag==STRIP_COMMENTS
) {
237 $p = preg_split( '/<!--/i', $text, 2 );
239 $p = preg_split( "/<\\s*$tag\\s*>/i", $text, 2 );
242 if ( ( count( $p ) < 2 ) ||
( '' == $p[1] ) ) {
245 if($tag==STRIP_COMMENTS
) {
246 $q = preg_split( '/-->/i', $p[1], 2 );
248 $q = preg_split( "/<\\/\\s*$tag\\s*>/i", $p[1], 2 );
250 $marker = $rnd . sprintf('%08X', $n++
);
251 $content[$marker] = $q[0];
252 $stripped .= $marker;
260 * Strips and renders nowiki, pre, math, hiero
261 * If $render is set, performs necessary rendering operations on plugins
262 * Returns the text, and fills an array with data needed in unstrip()
263 * If the $state is already a valid strip state, it adds to the state
265 * @param bool $stripcomments when set, HTML comments <!-- like this -->
266 * will be stripped in addition to other tags. This is important
267 * for section editing, where these comments cause confusion when
268 * counting the sections in the wikisource
272 function strip( $text, &$state, $stripcomments = false ) {
273 $render = ($this->mOutputType
== OT_HTML
);
274 $html_content = array();
275 $nowiki_content = array();
276 $math_content = array();
277 $pre_content = array();
278 $comment_content = array();
279 $ext_content = array();
280 $gallery_content = array();
282 # Replace any instances of the placeholders
283 $uniq_prefix = UNIQ_PREFIX
;
284 #$text = str_replace( $uniq_prefix, wfHtmlEscapeFirst( $uniq_prefix ), $text );
287 global $wgRawHtml, $wgWhitelistEdit;
288 if( $wgRawHtml && $wgWhitelistEdit ) {
289 $text = Parser
::extractTags('html', $text, $html_content, $uniq_prefix);
290 foreach( $html_content as $marker => $content ) {
292 # Raw and unchecked for validity.
293 $html_content[$marker] = $content;
295 $html_content[$marker] = '<html>'.$content.'</html>';
301 $text = Parser
::extractTags('nowiki', $text, $nowiki_content, $uniq_prefix);
302 foreach( $nowiki_content as $marker => $content ) {
304 $nowiki_content[$marker] = wfEscapeHTMLTagsOnly( $content );
306 $nowiki_content[$marker] = '<nowiki>'.$content.'</nowiki>';
311 $text = Parser
::extractTags('math', $text, $math_content, $uniq_prefix);
312 foreach( $math_content as $marker => $content ){
314 if( $this->mOptions
->getUseTeX() ) {
315 $math_content[$marker] = renderMath( $content );
317 $math_content[$marker] = '<math>'.$content.'<math>';
320 $math_content[$marker] = '<math>'.$content.'</math>';
325 $text = Parser
::extractTags('pre', $text, $pre_content, $uniq_prefix);
326 foreach( $pre_content as $marker => $content ){
328 $pre_content[$marker] = '<pre>' . wfEscapeHTMLTagsOnly( $content ) . '</pre>';
330 $pre_content[$marker] = '<pre>'.$content.'</pre>';
335 $text = Parser
::extractTags('gallery', $text, $gallery_content, $uniq_prefix);
336 foreach( $gallery_content as $marker => $content ) {
337 require_once( 'ImageGallery.php' );
339 $gallery_content[$marker] = Parser
::renderImageGallery( $content );
341 $gallery_content[$marker] = '<gallery>'.$content.'</gallery>';
347 $text = Parser
::extractTags(STRIP_COMMENTS
, $text, $comment_content, $uniq_prefix);
348 foreach( $comment_content as $marker => $content ){
349 $comment_content[$marker] = '<!--'.$content.'-->';
354 foreach ( $this->mTagHooks
as $tag => $callback ) {
355 $ext_contents[$tag] = array();
356 $text = Parser
::extractTags( $tag, $text, $ext_content[$tag], $uniq_prefix );
357 foreach( $ext_content[$tag] as $marker => $content ) {
359 $ext_content[$tag][$marker] = $callback( $content );
361 $ext_content[$tag][$marker] = "<$tag>$content</$tag>";
366 # Merge state with the pre-existing state, if there is one
368 $state['html'] = $state['html'] +
$html_content;
369 $state['nowiki'] = $state['nowiki'] +
$nowiki_content;
370 $state['math'] = $state['math'] +
$math_content;
371 $state['pre'] = $state['pre'] +
$pre_content;
372 $state['comment'] = $state['comment'] +
$comment_content;
373 $state['gallery'] = $state['gallery'] +
$gallery_content;
375 foreach( $ext_content as $tag => $array ) {
376 if ( array_key_exists( $tag, $state ) ) {
377 $state[$tag] = $state[$tag] +
$array;
382 'html' => $html_content,
383 'nowiki' => $nowiki_content,
384 'math' => $math_content,
385 'pre' => $pre_content,
386 'comment' => $comment_content,
387 'gallery' => $gallery_content,
394 * restores pre, math, and hiero removed by strip()
396 * always call unstripNoWiki() after this one
399 function unstrip( $text, &$state ) {
400 # Must expand in reverse order, otherwise nested tags will be corrupted
401 $contentDict = end( $state );
402 for ( $contentDict = end( $state ); $contentDict !== false; $contentDict = prev( $state ) ) {
403 if( key($state) != 'nowiki' && key($state) != 'html') {
404 for ( $content = end( $contentDict ); $content !== false; $content = prev( $contentDict ) ) {
405 $text = str_replace( key( $contentDict ), $content, $text );
414 * always call this after unstrip() to preserve the order
418 function unstripNoWiki( $text, &$state ) {
419 # Must expand in reverse order, otherwise nested tags will be corrupted
420 for ( $content = end($state['nowiki']); $content !== false; $content = prev( $state['nowiki'] ) ) {
421 $text = str_replace( key( $state['nowiki'] ), $content, $text );
426 for ( $content = end($state['html']); $content !== false; $content = prev( $state['html'] ) ) {
427 $text = str_replace( key( $state['html'] ), $content, $text );
435 * Add an item to the strip state
436 * Returns the unique tag which must be inserted into the stripped text
437 * The tag will be replaced with the original text in unstrip()
441 function insertStripItem( $text, &$state ) {
442 $rnd = UNIQ_PREFIX
. '-item' . Parser
::getRandomString();
451 $state['item'][$rnd] = $text;
456 * Return allowed HTML attributes
460 function getHTMLattrs () {
461 $htmlattrs = array( # Allowed attributes--no scripting, etc.
462 'title', 'align', 'lang', 'dir', 'width', 'height',
463 'bgcolor', 'clear', /* BR */ 'noshade', /* HR */
464 'cite', /* BLOCKQUOTE, Q */ 'size', 'face', 'color',
465 /* FONT */ 'type', 'start', 'value', 'compact',
466 /* For various lists, mostly deprecated but safe */
467 'summary', 'width', 'border', 'frame', 'rules',
468 'cellspacing', 'cellpadding', 'valign', 'char',
469 'charoff', 'colgroup', 'col', 'span', 'abbr', 'axis',
470 'headers', 'scope', 'rowspan', 'colspan', /* Tables */
471 'id', 'class', 'name', 'style' /* For CSS */
477 * Remove non approved attributes and javascript in css
481 function fixTagAttributes ( $t ) {
482 if ( trim ( $t ) == '' ) return '' ; # Saves runtime ;-)
483 $htmlattrs = $this->getHTMLattrs() ;
485 # Strip non-approved attributes from the tag
487 '/(\\w+)(\\s*=\\s*([^\\s\">]+|\"[^\">]*\"))?/e',
488 "(in_array(strtolower(\"\$1\"),\$htmlattrs)?(\"\$1\".((\"x\$3\" != \"x\")?\"=\$3\":'')):'')",
491 $t = str_replace ( '<></>' , '' , $t ) ; # This should fix bug 980557
493 # Strip javascript "expression" from stylesheets. Brute force approach:
494 # If anythin offensive is found, all attributes of the HTML tag are dropped
497 '/style\\s*=.*(expression|tps*:\/\/|url\\s*\().*/is',
498 wfMungeToUtf8( $t ) ) )
507 * interface with html tidy, used if $wgUseTidy = true
512 function tidy ( $text ) {
513 global $wgTidyConf, $wgTidyBin, $wgTidyOpts;
514 global $wgInputEncoding, $wgOutputEncoding;
515 $fname = 'Parser::tidy';
516 wfProfileIn( $fname );
520 switch(strtoupper($wgOutputEncoding)) {
522 $opts .= ($wgInputEncoding == $wgOutputEncoding)?
' -latin1':' -raw';
525 $opts .= ($wgInputEncoding == $wgOutputEncoding)?
' -utf8':' -raw';
531 $wrappedtext = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'.
532 ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>'.
533 '<head><title>test</title></head><body>'.$text.'</body></html>';
534 $descriptorspec = array(
535 0 => array('pipe', 'r'),
536 1 => array('pipe', 'w'),
537 2 => array('file', '/dev/null', 'a')
539 $process = proc_open("$wgTidyBin -config $wgTidyConf $wgTidyOpts$opts", $descriptorspec, $pipes);
540 if (is_resource($process)) {
541 fwrite($pipes[0], $wrappedtext);
543 while (!feof($pipes[1])) {
544 $cleansource .= fgets($pipes[1], 1024);
547 $return_value = proc_close($process);
550 wfProfileOut( $fname );
552 if( $cleansource == '' && $text != '') {
553 wfDebug( "Tidy error detected!\n" );
554 return $text . "\n<!-- Tidy found serious XHTML errors -->\n";
561 * parse the wiki syntax used to render tables
565 function doTableStuff ( $t ) {
566 $fname = 'Parser::doTableStuff';
567 wfProfileIn( $fname );
569 $t = explode ( "\n" , $t ) ;
570 $td = array () ; # Is currently a td tag open?
571 $ltd = array () ; # Was it TD or TH?
572 $tr = array () ; # Is currently a tr tag open?
573 $ltr = array () ; # tr attributes
574 $indent_level = 0; # indent level of the table
575 foreach ( $t AS $k => $x )
578 $fc = substr ( $x , 0 , 1 ) ;
579 if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
580 $indent_level = strlen( $matches[1] );
582 str_repeat( '<dl><dd>', $indent_level ) .
583 '<table ' . $this->fixTagAttributes ( $matches[2] ) . '>' ;
584 array_push ( $td , false ) ;
585 array_push ( $ltd , '' ) ;
586 array_push ( $tr , false ) ;
587 array_push ( $ltr , '' ) ;
589 else if ( count ( $td ) == 0 ) { } # Don't do any of the following
590 else if ( '|}' == substr ( $x , 0 , 2 ) ) {
592 $l = array_pop ( $ltd ) ;
593 if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
594 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
596 $t[$k] = $z . str_repeat( '</dd></dl>', $indent_level );
598 else if ( '|-' == substr ( $x , 0 , 2 ) ) { # Allows for |---------------
599 $x = substr ( $x , 1 ) ;
600 while ( $x != '' && substr ( $x , 0 , 1 ) == '-' ) $x = substr ( $x , 1 ) ;
602 $l = array_pop ( $ltd ) ;
603 if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
604 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
607 array_push ( $tr , false ) ;
608 array_push ( $td , false ) ;
609 array_push ( $ltd , '' ) ;
610 array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
612 else if ( '|' == $fc ||
'!' == $fc ||
'|+' == substr ( $x , 0 , 2 ) ) { # Caption
614 if ( '|+' == substr ( $x , 0 , 2 ) ) {
616 $x = substr ( $x , 1 ) ;
618 $after = substr ( $x , 1 ) ;
619 if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;
620 $after = explode ( '||' , $after ) ;
623 # Loop through each table cell
624 foreach ( $after AS $theline )
629 $tra = array_pop ( $ltr ) ;
630 if ( !array_pop ( $tr ) ) $z = '<tr '.$tra.">\n" ;
631 array_push ( $tr , true ) ;
632 array_push ( $ltr , '' ) ;
635 $l = array_pop ( $ltd ) ;
636 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
637 if ( $fc == '|' ) $l = 'td' ;
638 else if ( $fc == '!' ) $l = 'th' ;
639 else if ( $fc == '+' ) $l = 'caption' ;
641 array_push ( $ltd , $l ) ;
644 $y = explode ( '|' , $theline , 2 ) ;
645 # Note that a '|' inside an invalid link should not
646 # be mistaken as delimiting cell parameters
647 if ( strpos( $y[0], '[[' ) !== false ) {
648 $y = array ($theline);
650 if ( count ( $y ) == 1 )
651 $y = "{$z}<{$l}>{$y[0]}" ;
652 else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
654 array_push ( $td , true ) ;
659 # Closing open td, tr && table
660 while ( count ( $td ) > 0 )
662 if ( array_pop ( $td ) ) $t[] = '</td>' ;
663 if ( array_pop ( $tr ) ) $t[] = '</tr>' ;
667 $t = implode ( "\n" , $t ) ;
668 # $t = $this->removeHTMLtags( $t );
669 wfProfileOut( $fname );
674 * Helper function for parse() that transforms wiki markup into
675 * HTML. Only called for $mOutputType == OT_HTML.
679 function internalParse( $text, $linestart, $args = array(), $isMain=true ) {
682 $fname = 'Parser::internalParse';
683 wfProfileIn( $fname );
685 $text = $this->removeHTMLtags( $text );
686 $text = $this->replaceVariables( $text, $args );
688 $text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
690 $text = $this->doHeadings( $text );
691 if($this->mOptions
->getUseDynamicDates()) {
692 global $wgDateFormatter;
693 $text = $wgDateFormatter->reformat( $this->mOptions
->getDateFormat(), $text );
695 $text = $this->doAllQuotes( $text );
696 $text = $this->replaceInternalLinks ( $text );
697 $text = $this->replaceExternalLinks( $text );
699 # replaceInternalLinks may sometimes leave behind
700 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
701 $text = str_replace("http-noparse://","http://",$text);
703 $text = $this->doMagicLinks( $text );
704 $text = $this->doTableStuff( $text );
705 $text = $this->formatHeadings( $text, $isMain );
706 $sk =& $this->mOptions
->getSkin();
707 $text = $sk->transformContent( $text );
709 wfProfileOut( $fname );
714 * Replace special strings like "ISBN xxx" and "RFC xxx" with
715 * magic external links.
719 function &doMagicLinks( &$text ) {
720 global $wgUseGeoMode;
721 $text = $this->magicISBN( $text );
722 if ( isset( $wgUseGeoMode ) && $wgUseGeoMode ) {
723 $text = $this->magicGEO( $text );
725 $text = $this->magicRFC( $text );
730 * Parse ^^ tokens and return html
734 function doExponent ( $text ) {
735 $fname = 'Parser::doExponent';
736 wfProfileIn( $fname);
737 $text = preg_replace('/\^\^(.*)\^\^/','<small><sup>\\1</sup></small>', $text);
738 wfProfileOut( $fname);
743 * Parse headers and return html
747 function doHeadings( $text ) {
748 $fname = 'Parser::doHeadings';
749 wfProfileIn( $fname );
750 for ( $i = 6; $i >= 1; --$i ) {
751 $h = substr( '======', 0, $i );
752 $text = preg_replace( "/^{$h}(.+){$h}(\\s|$)/m",
753 "<h{$i}>\\1</h{$i}>\\2", $text );
755 wfProfileOut( $fname );
760 * Replace single quotes with HTML markup
762 * @return string the altered text
764 function doAllQuotes( $text ) {
765 $fname = 'Parser::doAllQuotes';
766 wfProfileIn( $fname );
768 $lines = explode( "\n", $text );
769 foreach ( $lines as $line ) {
770 $outtext .= $this->doQuotes ( $line ) . "\n";
772 $outtext = substr($outtext, 0,-1);
773 wfProfileOut( $fname );
778 * Helper function for doAllQuotes()
781 function doQuotes( $text ) {
782 $arr = preg_split ("/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
783 if (count ($arr) == 1)
787 # First, do some preliminary work. This may shift some apostrophes from
788 # being mark-up to being text. It also counts the number of occurrences
789 # of bold and italics mark-ups.
797 # If there are ever four apostrophes, assume the first is supposed to
798 # be text, and the remaining three constitute mark-up for bold text.
799 if (strlen ($arr[$i]) == 4)
804 # If there are more than 5 apostrophes in a row, assume they're all
805 # text except for the last 5.
806 else if (strlen ($arr[$i]) > 5)
808 $arr[$i-1] .= str_repeat ("'", strlen ($arr[$i]) - 5);
811 # Count the number of occurrences of bold and italics mark-ups.
812 # We are not counting sequences of five apostrophes.
813 if (strlen ($arr[$i]) == 2) $numitalics++
; else
814 if (strlen ($arr[$i]) == 3) $numbold++
; else
815 if (strlen ($arr[$i]) == 5) { $numitalics++
; $numbold++
; }
820 # If there is an odd number of both bold and italics, it is likely
821 # that one of the bold ones was meant to be an apostrophe followed
822 # by italics. Which one we cannot know for certain, but it is more
823 # likely to be one that has a single-letter word before it.
824 if (($numbold %
2 == 1) && ($numitalics %
2 == 1))
827 $firstsingleletterword = -1;
828 $firstmultiletterword = -1;
832 if (($i %
2 == 1) and (strlen ($r) == 3))
834 $x1 = substr ($arr[$i-1], -1);
835 $x2 = substr ($arr[$i-1], -2, 1);
837 if ($firstspace == -1) $firstspace = $i;
838 } else if ($x2 == ' ') {
839 if ($firstsingleletterword == -1) $firstsingleletterword = $i;
841 if ($firstmultiletterword == -1) $firstmultiletterword = $i;
847 # If there is a single-letter word, use it!
848 if ($firstsingleletterword > -1)
850 $arr [ $firstsingleletterword ] = "''";
851 $arr [ $firstsingleletterword-1 ] .= "'";
853 # If not, but there's a multi-letter word, use that one.
854 else if ($firstmultiletterword > -1)
856 $arr [ $firstmultiletterword ] = "''";
857 $arr [ $firstmultiletterword-1 ] .= "'";
859 # ... otherwise use the first one that has neither.
860 # (notice that it is possible for all three to be -1 if, for example,
861 # there is only one pentuple-apostrophe in the line)
862 else if ($firstspace > -1)
864 $arr [ $firstspace ] = "''";
865 $arr [ $firstspace-1 ] .= "'";
869 # Now let's actually convert our apostrophic mush to HTML!
878 if ($state == 'both')
885 if (strlen ($r) == 2)
888 { $output .= '</i>'; $state = ''; }
889 else if ($state == 'bi')
890 { $output .= '</i>'; $state = 'b'; }
891 else if ($state == 'ib')
892 { $output .= '</b></i><b>'; $state = 'b'; }
893 else if ($state == 'both')
894 { $output .= '<b><i>'.$buffer.'</i>'; $state = 'b'; }
895 else # $state can be 'b' or ''
896 { $output .= '<i>'; $state .= 'i'; }
898 else if (strlen ($r) == 3)
901 { $output .= '</b>'; $state = ''; }
902 else if ($state == 'bi')
903 { $output .= '</i></b><i>'; $state = 'i'; }
904 else if ($state == 'ib')
905 { $output .= '</b>'; $state = 'i'; }
906 else if ($state == 'both')
907 { $output .= '<i><b>'.$buffer.'</b>'; $state = 'i'; }
908 else # $state can be 'i' or ''
909 { $output .= '<b>'; $state .= 'b'; }
911 else if (strlen ($r) == 5)
914 { $output .= '</b><i>'; $state = 'i'; }
915 else if ($state == 'i')
916 { $output .= '</i><b>'; $state = 'b'; }
917 else if ($state == 'bi')
918 { $output .= '</i></b>'; $state = ''; }
919 else if ($state == 'ib')
920 { $output .= '</b></i>'; $state = ''; }
921 else if ($state == 'both')
922 { $output .= '<i><b>'.$buffer.'</b></i>'; $state = ''; }
923 else # ($state == '')
924 { $buffer = ''; $state = 'both'; }
929 # Now close all remaining tags. Notice that the order is important.
930 if ($state == 'b' ||
$state == 'ib')
932 if ($state == 'i' ||
$state == 'bi' ||
$state == 'ib')
936 if ($state == 'both')
937 $output .= '<b><i>'.$buffer.'</i></b>';
943 * Replace external links
945 * Note: this is all very hackish and the order of execution matters a lot.
946 * Make sure to run maintenance/parserTests.php if you change this code.
950 function replaceExternalLinks( $text ) {
951 $fname = 'Parser::replaceExternalLinks';
952 wfProfileIn( $fname );
954 $sk =& $this->mOptions
->getSkin();
955 $linktrail = wfMsgForContent('linktrail');
956 $bits = preg_split( EXT_LINK_BRACKETED
, $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
958 $s = $this->replaceFreeExternalLinks( array_shift( $bits ) );
961 while ( $i<count( $bits ) ) {
963 $protocol = $bits[$i++
];
965 $trail = $bits[$i++
];
967 # The characters '<' and '>' (which were escaped by
968 # removeHTMLtags()) should not be included in
969 # URLs, per RFC 2396.
970 if (preg_match('/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE
)) {
971 $text = substr($url, $m2[0][1]) . ' ' . $text;
972 $url = substr($url, 0, $m2[0][1]);
975 # If the link text is an image URL, replace it with an <img> tag
976 # This happened by accident in the original parser, but some people used it extensively
977 $img = $this->maybeMakeImageLink( $text );
978 if ( $img !== false ) {
984 # No link text, e.g. [http://domain.tld/some.link]
986 # Autonumber if allowed
987 if ( strpos( HTTP_PROTOCOLS
, $protocol ) !== false ) {
988 $text = '[' . ++
$this->mAutonumber
. ']';
990 # Otherwise just use the URL
991 $text = htmlspecialchars( $url );
994 # Have link text, e.g. [http://domain.tld/some.link text]s
996 if ( preg_match( $linktrail, $trail, $m2 ) ) {
1002 $encUrl = htmlspecialchars( $url );
1003 # Bit in parentheses showing the URL for the printable version
1004 if( $url == $text ||
preg_match( "!$protocol://" . preg_quote( $text, '/' ) . "/?$!", $url ) ) {
1007 # Expand the URL for printable version
1008 if ( ! $sk->suppressUrlExpansion() ) {
1009 $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $encUrl ) . "</i>)</span>";
1015 # Process the trail (i.e. everything after this link up until start of the next link),
1016 # replacing any non-bracketed links
1017 $trail = $this->replaceFreeExternalLinks( $trail );
1019 # Use the encoded URL
1020 # This means that users can paste URLs directly into the text
1021 # Funny characters like ö aren't valid in URLs anyway
1022 # This was changed in August 2004
1023 $s .= $sk->makeExternalLink( $url, $text, false ) . $dtrail. $paren . $trail;
1026 wfProfileOut( $fname );
1031 * Replace anything that looks like a URL with a link
1034 function replaceFreeExternalLinks( $text ) {
1035 $bits = preg_split( '/((?:'.URL_PROTOCOLS
.'):)/', $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
1036 $s = array_shift( $bits );
1039 $sk =& $this->mOptions
->getSkin();
1041 while ( $i < count( $bits ) ){
1042 $protocol = $bits[$i++
];
1043 $remainder = $bits[$i++
];
1045 if ( preg_match( '/^('.EXT_LINK_URL_CLASS
.'+)(.*)$/s', $remainder, $m ) ) {
1046 # Found some characters after the protocol that look promising
1047 $url = $protocol . $m[1];
1050 # The characters '<' and '>' (which were escaped by
1051 # removeHTMLtags()) should not be included in
1052 # URLs, per RFC 2396.
1053 if (preg_match('/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE
)) {
1054 $trail = substr($url, $m2[0][1]) . $trail;
1055 $url = substr($url, 0, $m2[0][1]);
1058 # Move trailing punctuation to $trail
1060 # If there is no left bracket, then consider right brackets fair game too
1061 if ( strpos( $url, '(' ) === false ) {
1065 $numSepChars = strspn( strrev( $url ), $sep );
1066 if ( $numSepChars ) {
1067 $trail = substr( $url, -$numSepChars ) . $trail;
1068 $url = substr( $url, 0, -$numSepChars );
1071 # Replace & from obsolete syntax with &.
1072 # All HTML entities will be escaped by makeExternalLink()
1073 # or maybeMakeImageLink()
1074 $url = str_replace( '&', '&', $url );
1076 # Is this an external image?
1077 $text = $this->maybeMakeImageLink( $url );
1078 if ( $text === false ) {
1079 # Not an image, make a link
1080 $text = $sk->makeExternalLink( $url, $url );
1082 $s .= $text . $trail;
1084 $s .= $protocol . $remainder;
1091 * make an image if it's allowed
1094 function maybeMakeImageLink( $url ) {
1095 $sk =& $this->mOptions
->getSkin();
1097 if ( $this->mOptions
->getAllowExternalImages() ) {
1098 if ( preg_match( EXT_IMAGE_REGEX
, $url ) ) {
1100 $text = $sk->makeImage( htmlspecialchars( $url ) );
1107 * Process [[ ]] wikilinks
1112 function replaceInternalLinks( $s ) {
1113 global $wgLang, $wgContLang, $wgLinkCache;
1114 global $wgDisableLangConversion;
1115 static $fname = 'Parser::replaceInternalLinks' ;
1117 wfProfileIn( $fname );
1119 wfProfileIn( $fname.'-setup' );
1121 # the % is needed to support urlencoded titles as well
1122 if ( !$tc ) { $tc = Title
::legalChars() . '#%'; }
1124 $sk =& $this->mOptions
->getSkin();
1125 global $wgUseOldExistenceCheck;
1126 # "Post-parse link colour check" works only on wiki text since it's now
1127 # in Parser. Enable it, then disable it when we're done.
1128 $saveParseColour = $sk->postParseLinkColour( !$wgUseOldExistenceCheck );
1130 $redirect = MagicWord
::get ( MAG_REDIRECT
) ;
1132 #split the entire text string on occurences of [[
1133 $a = explode( '[[', ' ' . $s );
1134 #get the first element (all text up to first [[), and remove the space we added
1135 $s = array_shift( $a );
1136 $s = substr( $s, 1 );
1138 # Match a link having the form [[namespace:link|alternate]]trail
1140 if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|([^]]+))?]](.*)\$/sD"; }
1141 # Match cases where there is no "]]", which might still be images
1142 static $e1_img = FALSE;
1143 if ( !$e1_img ) { $e1_img = "/^([{$tc}]+)\\|(.*)\$/sD"; }
1144 # Match the end of a line for a word that's not followed by whitespace,
1145 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
1146 static $e2 = '/^(.*?)([a-zA-Z\x80-\xff]+)$/sD';
1148 $useLinkPrefixExtension = $wgContLang->linkPrefixExtension();
1150 $nottalk = !Namespace::isTalk( $this->mTitle
->getNamespace() );
1152 if ( $useLinkPrefixExtension ) {
1153 if ( preg_match( $e2, $s, $m ) ) {
1154 $first_prefix = $m[2];
1157 $first_prefix = false;
1163 wfProfileOut( $fname.'-setup' );
1165 $checkVariantLink = sizeof($wgContLang->getVariants())>1;
1166 # Loop for each link
1167 for ($k = 0; isset( $a[$k] ); $k++
) {
1169 wfProfileIn( $fname.'-prefixhandling' );
1170 if ( $useLinkPrefixExtension ) {
1171 if ( preg_match( $e2, $s, $m ) ) {
1179 $prefix = $first_prefix;
1180 $first_prefix = false;
1183 wfProfileOut( $fname.'-prefixhandling' );
1185 $might_be_img = false;
1187 if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
1189 # fix up urlencoded title texts
1190 if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
1192 } elseif( preg_match($e1_img, $line, $m) ) { # Invalid, but might be an image with a link in its caption
1193 $might_be_img = true;
1195 if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
1197 } else { # Invalid form; output directly
1198 $s .= $prefix . '[[' . $line ;
1202 # Don't allow internal links to pages containing
1203 # PROTO: where PROTO is a valid URL protocol; these
1204 # should be external links.
1205 if (preg_match('/^((?:'.URL_PROTOCOLS
.'):)/', $m[1])) {
1206 $s .= $prefix . '[[' . $line ;
1210 # Make subpage if necessary
1211 $link = $this->maybeDoSubpageLink( $m[1], $text );
1213 $noforce = (substr($m[1], 0, 1) != ':');
1215 # Strip off leading ':'
1216 $link = substr($link, 1);
1219 $nt = Title
::newFromText( $this->unstripNoWiki($link, $this->mStripState
) );
1221 $s .= $prefix . '[[' . $line;
1225 #check other language variants of the link
1226 #if the article does not exist
1227 if( $nt->getArticleID() == 0
1228 && $checkVariantLink ) {
1229 $wgContLang->findVariantLink($link, $nt);
1232 $ns = $nt->getNamespace();
1233 $iw = $nt->getInterWiki();
1235 if ($might_be_img) { # if this is actually an invalid link
1236 if ($ns == NS_IMAGE
&& $noforce) { #but might be an image
1238 while (isset ($a[$k+
1]) ) {
1239 #look at the next 'line' to see if we can close it there
1240 $next_line = array_shift(array_splice( $a, $k +
1, 1) );
1241 if( preg_match("/^(.*?]].*?)]](.*)$/sD", $next_line, $m) ) {
1242 # the first ]] closes the inner link, the second the image
1244 $text .= '[[' . $m[1];
1247 } elseif( preg_match("/^.*?]].*$/sD", $next_line, $m) ) {
1248 #if there's exactly one ]] that's fine, we'll keep looking
1249 $text .= '[[' . $m[0];
1251 #if $next_line is invalid too, we need look no further
1252 $text .= '[[' . $next_line;
1257 # we couldn't find the end of this imageLink, so output it raw
1258 #but don't ignore what might be perfectly normal links in the text we've examined
1259 $text = $this->replaceInternalLinks($text);
1260 $s .= $prefix . '[[' . $link . '|' . $text;
1261 # note: no $trail, because without an end, there *is* no trail
1264 } else { #it's not an image, so output it raw
1265 $s .= $prefix . '[[' . $link . '|' . $text;
1266 # note: no $trail, because without an end, there *is* no trail
1271 $wasblank = ( '' == $text );
1272 if( $wasblank ) $text = $link;
1275 # Link not escaped by : , create the various objects
1279 if( $iw && $this->mOptions
->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) {
1280 array_push( $this->mOutput
->mLanguageLinks
, $nt->getFullText() );
1281 $tmp = $prefix . $trail ;
1282 $s .= (trim($tmp) == '')?
'': $tmp;
1286 if ( $ns == NS_IMAGE
) {
1287 # recursively parse links inside the image caption
1288 # actually, this will parse them in any other parameters, too,
1289 # but it might be hard to fix that, and it doesn't matter ATM
1290 $text = $this->replaceExternalLinks($text);
1291 $text = $this->replaceInternalLinks($text);
1293 # replace the image with a link-holder so that replaceExternalLinks() can't mess with it
1294 $s .= $prefix . $this->insertStripItem( $sk->makeImageLinkObj( $nt, $text ), $this->mStripState
) . $trail;
1295 $wgLinkCache->addImageLinkObj( $nt );
1299 if ( $ns == NS_CATEGORY
) {
1300 $t = $nt->getText() ;
1302 $wgLinkCache->suspend(); # Don't save in links/brokenlinks
1303 $pPLC=$sk->postParseLinkColour();
1304 $sk->postParseLinkColour( false );
1305 $t = $sk->makeLinkObj( $nt, $t, '', '' , $prefix );
1306 $sk->postParseLinkColour( $pPLC );
1307 $wgLinkCache->resume();
1310 if ( $this->mTitle
->getNamespace() == NS_CATEGORY
) {
1311 $sortkey = $this->mTitle
->getText();
1313 $sortkey = $this->mTitle
->getPrefixedText();
1318 $wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
1319 $this->mOutput
->mCategoryLinks
[] = $t ;
1320 $s .= $prefix . $trail ;
1325 if( ( $nt->getPrefixedText() === $this->mTitle
->getPrefixedText() ) &&
1326 ( $nt->getFragment() === '' ) ) {
1327 # Self-links are handled specially; generally de-link and change to bold.
1328 $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail );
1332 # Special and Media are pseudo-namespaces; no pages actually exist in them
1333 if( $ns == NS_MEDIA
) {
1334 $s .= $prefix . $sk->makeMediaLinkObj( $nt, $text, true ) . $trail;
1335 $wgLinkCache->addImageLinkObj( $nt );
1337 } elseif( $ns == NS_SPECIAL
) {
1338 $s .= $prefix . $sk->makeKnownLinkObj( $nt, $text, '', $trail );
1341 $s .= $sk->makeLinkObj( $nt, $text, '', $trail, $prefix );
1343 $sk->postParseLinkColour( $saveParseColour );
1344 wfProfileOut( $fname );
1349 * Handle link to subpage if necessary
1350 * @param string $target the source of the link
1351 * @param string &$text the link text, modified as necessary
1352 * @return string the full name of the link
1355 function maybeDoSubpageLink($target, &$text) {
1358 # :Foobar -- override special treatment of prefix (images, language links)
1359 # /Foobar -- convert to CurrentPage/Foobar
1360 # /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text
1361 global $wgNamespacesWithSubpages;
1363 $fname = 'Parser::maybeDoSubpageLink';
1364 wfProfileIn( $fname );
1365 # Look at the first character
1366 if( $target{0} == '/' ) {
1367 # / at end means we don't want the slash to be shown
1368 if(substr($target,-1,1)=='/') {
1369 $target=substr($target,1,-1);
1372 $noslash=substr($target,1);
1375 # Some namespaces don't allow subpages
1376 if(!empty($wgNamespacesWithSubpages[$this->mTitle
->getNamespace()])) {
1377 # subpages allowed here
1378 $ret = $this->mTitle
->getPrefixedText(). '/' . trim($noslash);
1379 if( '' === $text ) {
1381 } # this might be changed for ugliness reasons
1383 # no subpage allowed, use standard link
1391 wfProfileOut( $fname );
1396 * Used by doBlockLevels()
1399 /* private */ function closeParagraph() {
1401 if ( '' != $this->mLastSection
) {
1402 $result = '</' . $this->mLastSection
. ">\n";
1404 $this->mInPre
= false;
1405 $this->mLastSection
= '';
1408 # getCommon() returns the length of the longest common substring
1409 # of both arguments, starting at the beginning of both.
1411 /* private */ function getCommon( $st1, $st2 ) {
1412 $fl = strlen( $st1 );
1413 $shorter = strlen( $st2 );
1414 if ( $fl < $shorter ) { $shorter = $fl; }
1416 for ( $i = 0; $i < $shorter; ++
$i ) {
1417 if ( $st1{$i} != $st2{$i} ) { break; }
1421 # These next three functions open, continue, and close the list
1422 # element appropriate to the prefix character passed into them.
1424 /* private */ function openList( $char ) {
1425 $result = $this->closeParagraph();
1427 if ( '*' == $char ) { $result .= '<ul><li>'; }
1428 else if ( '#' == $char ) { $result .= '<ol><li>'; }
1429 else if ( ':' == $char ) { $result .= '<dl><dd>'; }
1430 else if ( ';' == $char ) {
1431 $result .= '<dl><dt>';
1432 $this->mDTopen
= true;
1434 else { $result = '<!-- ERR 1 -->'; }
1439 /* private */ function nextItem( $char ) {
1440 if ( '*' == $char ||
'#' == $char ) { return '</li><li>'; }
1441 else if ( ':' == $char ||
';' == $char ) {
1443 if ( $this->mDTopen
) { $close = '</dt>'; }
1444 if ( ';' == $char ) {
1445 $this->mDTopen
= true;
1446 return $close . '<dt>';
1448 $this->mDTopen
= false;
1449 return $close . '<dd>';
1452 return '<!-- ERR 2 -->';
1455 /* private */ function closeList( $char ) {
1456 if ( '*' == $char ) { $text = '</li></ul>'; }
1457 else if ( '#' == $char ) { $text = '</li></ol>'; }
1458 else if ( ':' == $char ) {
1459 if ( $this->mDTopen
) {
1460 $this->mDTopen
= false;
1461 $text = '</dt></dl>';
1463 $text = '</dd></dl>';
1466 else { return '<!-- ERR 3 -->'; }
1472 * Make lists from lines starting with ':', '*', '#', etc.
1475 * @return string the lists rendered as HTML
1477 function doBlockLevels( $text, $linestart ) {
1478 $fname = 'Parser::doBlockLevels';
1479 wfProfileIn( $fname );
1481 # Parsing through the text line by line. The main thing
1482 # happening here is handling of block-level elements p, pre,
1483 # and making lists from lines starting with * # : etc.
1485 $textLines = explode( "\n", $text );
1487 $lastPrefix = $output = $lastLine = '';
1488 $this->mDTopen
= $inBlockElem = false;
1490 $paragraphStack = false;
1492 if ( !$linestart ) {
1493 $output .= array_shift( $textLines );
1495 foreach ( $textLines as $oLine ) {
1496 $lastPrefixLength = strlen( $lastPrefix );
1497 $preCloseMatch = preg_match('/<\\/pre/i', $oLine );
1498 $preOpenMatch = preg_match('/<pre/i', $oLine );
1499 if ( !$this->mInPre
) {
1500 # Multiple prefixes may abut each other for nested lists.
1501 $prefixLength = strspn( $oLine, '*#:;' );
1502 $pref = substr( $oLine, 0, $prefixLength );
1505 $pref2 = str_replace( ';', ':', $pref );
1506 $t = substr( $oLine, $prefixLength );
1507 $this->mInPre
= !empty($preOpenMatch);
1509 # Don't interpret any other prefixes in preformatted text
1511 $pref = $pref2 = '';
1516 if( $prefixLength && 0 == strcmp( $lastPrefix, $pref2 ) ) {
1517 # Same as the last item, so no need to deal with nesting or opening stuff
1518 $output .= $this->nextItem( substr( $pref, -1 ) );
1519 $paragraphStack = false;
1521 if ( substr( $pref, -1 ) == ';') {
1522 # The one nasty exception: definition lists work like this:
1523 # ; title : definition text
1524 # So we check for : in the remainder text to split up the
1525 # title and definition, without b0rking links.
1526 if ($this->findColonNoLinks($t, $term, $t2) !== false) {
1528 $output .= $term . $this->nextItem( ':' );
1531 } elseif( $prefixLength ||
$lastPrefixLength ) {
1532 # Either open or close a level...
1533 $commonPrefixLength = $this->getCommon( $pref, $lastPrefix );
1534 $paragraphStack = false;
1536 while( $commonPrefixLength < $lastPrefixLength ) {
1537 $output .= $this->closeList( $lastPrefix{$lastPrefixLength-1} );
1538 --$lastPrefixLength;
1540 if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) {
1541 $output .= $this->nextItem( $pref{$commonPrefixLength-1} );
1543 while ( $prefixLength > $commonPrefixLength ) {
1544 $char = substr( $pref, $commonPrefixLength, 1 );
1545 $output .= $this->openList( $char );
1547 if ( ';' == $char ) {
1548 # FIXME: This is dupe of code above
1549 if ($this->findColonNoLinks($t, $term, $t2) !== false) {
1551 $output .= $term . $this->nextItem( ':' );
1554 ++
$commonPrefixLength;
1556 $lastPrefix = $pref2;
1558 if( 0 == $prefixLength ) {
1559 # No prefix (not in list)--go to paragraph mode
1560 $uniq_prefix = UNIQ_PREFIX
;
1561 // XXX: use a stack for nestable elements like span, table and div
1562 $openmatch = preg_match('/(<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<li|<\\/tr|<\\/td|<\\/th)/i', $t );
1563 $closematch = preg_match(
1564 '/(<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|'.
1565 '<td|<th|<div|<\\/div|<hr|<\\/pre|<\\/p|'.$uniq_prefix.'-pre|<\\/li|<\\/ul)/i', $t );
1566 if ( $openmatch or $closematch ) {
1567 $paragraphStack = false;
1568 $output .= $this->closeParagraph();
1569 if($preOpenMatch and !$preCloseMatch) {
1570 $this->mInPre
= true;
1572 if ( $closematch ) {
1573 $inBlockElem = false;
1575 $inBlockElem = true;
1577 } else if ( !$inBlockElem && !$this->mInPre
) {
1578 if ( ' ' == $t{0} and ( $this->mLastSection
== 'pre' or trim($t) != '' ) ) {
1580 if ($this->mLastSection
!= 'pre') {
1581 $paragraphStack = false;
1582 $output .= $this->closeParagraph().'<pre>';
1583 $this->mLastSection
= 'pre';
1585 $t = substr( $t, 1 );
1588 if ( '' == trim($t) ) {
1589 if ( $paragraphStack ) {
1590 $output .= $paragraphStack.'<br />';
1591 $paragraphStack = false;
1592 $this->mLastSection
= 'p';
1594 if ($this->mLastSection
!= 'p' ) {
1595 $output .= $this->closeParagraph();
1596 $this->mLastSection
= '';
1597 $paragraphStack = '<p>';
1599 $paragraphStack = '</p><p>';
1603 if ( $paragraphStack ) {
1604 $output .= $paragraphStack;
1605 $paragraphStack = false;
1606 $this->mLastSection
= 'p';
1607 } else if ($this->mLastSection
!= 'p') {
1608 $output .= $this->closeParagraph().'<p>';
1609 $this->mLastSection
= 'p';
1615 if ($paragraphStack === false) {
1619 while ( $prefixLength ) {
1620 $output .= $this->closeList( $pref2{$prefixLength-1} );
1623 if ( '' != $this->mLastSection
) {
1624 $output .= '</' . $this->mLastSection
. '>';
1625 $this->mLastSection
= '';
1628 wfProfileOut( $fname );
1633 * Split up a string on ':', ignoring any occurences inside
1634 * <a>..</a> or <span>...</span>
1635 * @param string $str the string to split
1636 * @param string &$before set to everything before the ':'
1637 * @param string &$after set to everything after the ':'
1638 * return string the position of the ':', or false if none found
1640 function findColonNoLinks($str, &$before, &$after) {
1641 # I wonder if we should make this count all tags, not just <a>
1642 # and <span>. That would prevent us from matching a ':' that
1643 # comes in the middle of italics other such formatting....
1645 $fname = 'Parser::findColonNoLinks';
1646 wfProfileIn( $fname );
1649 $colon = strpos($str, ':', $pos);
1651 if ($colon !== false) {
1652 $before = substr($str, 0, $colon);
1653 $after = substr($str, $colon +
1);
1655 # Skip any ':' within <a> or <span> pairs
1656 $a = substr_count($before, '<a');
1657 $s = substr_count($before, '<span');
1658 $ca = substr_count($before, '</a>');
1659 $cs = substr_count($before, '</span>');
1661 if ($a <= $ca and $s <= $cs) {
1662 # Tags are balanced before ':'; ok
1667 } while ($colon !== false);
1668 wfProfileOut( $fname );
1673 * Return value of a magic variable (like PAGENAME)
1677 function getVariableValue( $index ) {
1678 global $wgContLang, $wgSitename, $wgServer;
1681 case MAG_CURRENTMONTH
:
1682 return $wgContLang->formatNum( date( 'm' ) );
1683 case MAG_CURRENTMONTHNAME
:
1684 return $wgContLang->getMonthName( date('n') );
1685 case MAG_CURRENTMONTHNAMEGEN
:
1686 return $wgContLang->getMonthNameGen( date('n') );
1687 case MAG_CURRENTDAY
:
1688 return $wgContLang->formatNum( date('j') );
1690 return $this->mTitle
->getText();
1692 return $this->mTitle
->getPartialURL();
1694 # return Namespace::getCanonicalName($this->mTitle->getNamespace());
1695 return $wgContLang->getNsText($this->mTitle
->getNamespace()); # Patch by Dori
1696 case MAG_CURRENTDAYNAME
:
1697 return $wgContLang->getWeekdayName( date('w')+
1 );
1698 case MAG_CURRENTYEAR
:
1699 return $wgContLang->formatNum( date( 'Y' ) );
1700 case MAG_CURRENTTIME
:
1701 return $wgContLang->time( wfTimestampNow(), false );
1702 case MAG_NUMBEROFARTICLES
:
1703 return $wgContLang->formatNum( wfNumberOfArticles() );
1714 * initialise the magic variables (like CURRENTMONTHNAME)
1718 function initialiseVariables() {
1719 $fname = 'Parser::initialiseVariables';
1720 wfProfileIn( $fname );
1721 global $wgVariableIDs;
1722 $this->mVariables
= array();
1723 foreach ( $wgVariableIDs as $id ) {
1724 $mw =& MagicWord
::get( $id );
1725 $mw->addToArray( $this->mVariables
, $this->getVariableValue( $id ) );
1727 wfProfileOut( $fname );
1731 * Replace magic variables, templates, and template arguments
1732 * with the appropriate text. Templates are substituted recursively,
1733 * taking care to avoid infinite loops.
1735 * Note that the substitution depends on value of $mOutputType:
1736 * OT_WIKI: only {{subst:}} templates
1737 * OT_MSG: only magic variables
1738 * OT_HTML: all templates and magic variables
1740 * @param string $tex The text to transform
1741 * @param array $args Key-value pairs representing template parameters to substitute
1744 function replaceVariables( $text, $args = array() ) {
1745 global $wgLang, $wgScript, $wgArticlePath;
1747 # Prevent too big inclusions
1748 if(strlen($text)> MAX_INCLUDE_SIZE
)
1751 $fname = 'Parser::replaceVariables';
1752 wfProfileIn( $fname );
1754 $titleChars = Title
::legalChars();
1756 # This function is called recursively. To keep track of arguments we need a stack:
1757 array_push( $this->mArgStack
, $args );
1759 # Variable substitution
1760 $text = preg_replace_callback( "/{{([$titleChars]*?)}}/", array( &$this, 'variableSubstitution' ), $text );
1762 if ( $this->mOutputType
== OT_HTML ||
$this->mOutputType
== OT_WIKI
) {
1763 # Argument substitution
1764 $text = preg_replace_callback( "/{{{([$titleChars]*?)}}}/", array( &$this, 'argSubstitution' ), $text );
1766 # Template substitution
1767 $regex = '/(\\n|{)?{{(['.$titleChars.']*)(\\|.*?|)}}/s';
1768 $text = preg_replace_callback( $regex, array( &$this, 'braceSubstitution' ), $text );
1770 array_pop( $this->mArgStack
);
1772 wfProfileOut( $fname );
1777 * Replace magic variables
1780 function variableSubstitution( $matches ) {
1781 if ( !$this->mVariables
) {
1782 $this->initialiseVariables();
1785 if ( $this->mOutputType
== OT_WIKI
) {
1786 # Do only magic variables prefixed by SUBST
1787 $mwSubst =& MagicWord
::get( MAG_SUBST
);
1788 if (!$mwSubst->matchStartAndRemove( $matches[1] ))
1790 # Note that if we don't substitute the variable below,
1791 # we don't remove the {{subst:}} magic word, in case
1792 # it is a template rather than a magic variable.
1794 if ( !$skip && array_key_exists( $matches[1], $this->mVariables
) ) {
1795 $text = $this->mVariables
[$matches[1]];
1796 $this->mOutput
->mContainsOldMagic
= true;
1798 $text = $matches[0];
1803 # Split template arguments
1804 function getTemplateArgs( $argsString ) {
1805 if ( $argsString === '' ) {
1809 $args = explode( '|', substr( $argsString, 1 ) );
1811 # If any of the arguments contains a '[[' but no ']]', it needs to be
1812 # merged with the next arg because the '|' character between belongs
1813 # to the link syntax and not the template parameter syntax.
1814 $argc = count($args);
1816 for ( $i = 0; $i < $argc-1; $i++
) {
1817 if ( substr_count ( $args[$i], '[[' ) != substr_count ( $args[$i], ']]' ) ) {
1818 $args[$i] .= '|'.$args[$i+
1];
1819 array_splice($args, $i+
1, 1);
1829 * Return the text of a template, after recursively
1830 * replacing any variables or templates within the template.
1832 * @param array $matches The parts of the template
1833 * $matches[1]: the title, i.e. the part before the |
1834 * $matches[2]: the parameters (including a leading |), if any
1835 * @return string the text of the template
1838 function braceSubstitution( $matches ) {
1839 global $wgLinkCache, $wgContLang;
1840 $fname = 'Parser::braceSubstitution';
1847 # Need to know if the template comes at the start of a line,
1848 # to treat the beginning of the template like the beginning
1849 # of a line for tables and block-level elements.
1850 $linestart = $matches[1];
1852 # $part1 is the bit before the first |, and must contain only title characters
1853 # $args is a list of arguments, starting from index 0, not including $part1
1855 $part1 = $matches[2];
1856 # If the third subpattern matched anything, it will start with |
1858 $args = $this->getTemplateArgs($matches[3]);
1859 $argc = count( $args );
1861 # Don't parse {{{}}} because that's only for template arguments
1862 if ( $linestart === '{' ) {
1863 $text = $matches[0];
1870 $mwSubst =& MagicWord
::get( MAG_SUBST
);
1871 if ( $mwSubst->matchStartAndRemove( $part1 ) xor ($this->mOutputType
== OT_WIKI
) ) {
1872 # One of two possibilities is true:
1873 # 1) Found SUBST but not in the PST phase
1874 # 2) Didn't find SUBST and in the PST phase
1875 # In either case, return without further processing
1876 $text = $matches[0];
1882 # MSG, MSGNW and INT
1885 $mwMsgnw =& MagicWord
::get( MAG_MSGNW
);
1886 if ( $mwMsgnw->matchStartAndRemove( $part1 ) ) {
1889 # Remove obsolete MSG:
1890 $mwMsg =& MagicWord
::get( MAG_MSG
);
1891 $mwMsg->matchStartAndRemove( $part1 );
1894 # Check if it is an internal message
1895 $mwInt =& MagicWord
::get( MAG_INT
);
1896 if ( $mwInt->matchStartAndRemove( $part1 ) ) {
1897 if ( $this->incrementIncludeCount( 'int:'.$part1 ) ) {
1898 $text = $linestart . wfMsgReal( $part1, $args, true );
1906 # Check for NS: (namespace expansion)
1907 $mwNs = MagicWord
::get( MAG_NS
);
1908 if ( $mwNs->matchStartAndRemove( $part1 ) ) {
1909 if ( intval( $part1 ) ) {
1910 $text = $linestart . $wgContLang->getNsText( intval( $part1 ) );
1913 $index = Namespace::getCanonicalIndex( strtolower( $part1 ) );
1914 if ( !is_null( $index ) ) {
1915 $text = $linestart . $wgContLang->getNsText( $index );
1922 # LOCALURL and LOCALURLE
1924 $mwLocal = MagicWord
::get( MAG_LOCALURL
);
1925 $mwLocalE = MagicWord
::get( MAG_LOCALURLE
);
1927 if ( $mwLocal->matchStartAndRemove( $part1 ) ) {
1928 $func = 'getLocalURL';
1929 } elseif ( $mwLocalE->matchStartAndRemove( $part1 ) ) {
1930 $func = 'escapeLocalURL';
1935 if ( $func !== '' ) {
1936 $title = Title
::newFromText( $part1 );
1937 if ( !is_null( $title ) ) {
1939 $text = $linestart . $title->$func( $args[0] );
1941 $text = $linestart . $title->$func();
1949 if ( !$found && $argc == 1 ) {
1950 $mwGrammar =& MagicWord
::get( MAG_GRAMMAR
);
1951 if ( $mwGrammar->matchStartAndRemove( $part1 ) ) {
1952 $text = $linestart . $wgContLang->convertGrammar( $args[0], $part1 );
1957 # Template table test
1959 # Did we encounter this template already? If yes, it is in the cache
1960 # and we need to check for loops.
1961 if ( !$found && isset( $this->mTemplates
[$part1] ) ) {
1962 # set $text to cached message.
1963 $text = $linestart . $this->mTemplates
[$part1];
1966 # Infinite loop test
1967 if ( isset( $this->mTemplatePath
[$part1] ) ) {
1970 $text .= '<!-- WARNING: template loop detected -->';
1974 # Load from database
1975 $itcamefromthedatabase = false;
1978 $part1 = $this->maybeDoSubpageLink( $part1, $subpage='' );
1979 if ($subpage !== '') {
1980 $ns = $this->mTitle
->getNamespace();
1982 $title = Title
::newFromText( $part1, $ns );
1983 if ( !is_null( $title ) && !$title->isExternal() ) {
1984 # Check for excessive inclusion
1985 $dbk = $title->getPrefixedDBkey();
1986 if ( $this->incrementIncludeCount( $dbk ) ) {
1987 # This should never be reached.
1988 $article = new Article( $title );
1989 $articleContent = $article->getContentWithoutUsingSoManyDamnGlobals();
1990 if ( $articleContent !== false ) {
1992 $text = $linestart . $articleContent;
1993 $itcamefromthedatabase = true;
1997 # If the title is valid but undisplayable, make a link to it
1998 if ( $this->mOutputType
== OT_HTML
&& !$found ) {
1999 $text = $linestart . '[['.$title->getPrefixedText().']]';
2003 # Template cache array insertion
2004 $this->mTemplates
[$part1] = $text;
2008 # Recursive parsing, escaping and link table handling
2009 # Only for HTML output
2010 if ( $nowiki && $found && $this->mOutputType
== OT_HTML
) {
2011 $text = wfEscapeWikiText( $text );
2012 } elseif ( ($this->mOutputType
== OT_HTML ||
$this->mOutputType
== OT_WIKI
) && $found && !$noparse) {
2013 # Clean up argument array
2014 $assocArgs = array();
2016 foreach( $args as $arg ) {
2017 $eqpos = strpos( $arg, '=' );
2018 if ( $eqpos === false ) {
2019 $assocArgs[$index++
] = $arg;
2021 $name = trim( substr( $arg, 0, $eqpos ) );
2022 $value = trim( substr( $arg, $eqpos+
1 ) );
2023 if ( $value === false ) {
2026 if ( $name !== false ) {
2027 $assocArgs[$name] = $value;
2032 # Add a new element to the templace recursion path
2033 $this->mTemplatePath
[$part1] = 1;
2035 $text = $this->strip( $text, $this->mStripState
);
2036 $text = $this->removeHTMLtags( $text );
2037 $text = $this->replaceVariables( $text, $assocArgs );
2039 # Resume the link cache and register the inclusion as a link
2040 if ( $this->mOutputType
== OT_HTML
&& !is_null( $title ) ) {
2041 $wgLinkCache->addLinkObj( $title );
2044 # If the template begins with a table or block-level
2045 # element, it should be treated as beginning a new line.
2046 if ($linestart !== '\n' && preg_match('/^({\\||:|;|#|\*)/', $text)) {
2047 $text = "\n" . $text;
2051 # Empties the template path
2052 $this->mTemplatePath
= array();
2056 # replace ==section headers==
2057 # XXX this needs to go away once we have a better parser.
2058 if ( $this->mOutputType
!= OT_WIKI
&& $itcamefromthedatabase ) {
2059 if( !is_null( $title ) )
2060 $encodedname = base64_encode($title->getPrefixedDBkey());
2062 $encodedname = base64_encode("");
2063 $m = preg_split('/(^={1,6}.*?={1,6}\s*?$)/m', $text, -1,
2064 PREG_SPLIT_DELIM_CAPTURE
);
2067 for( $i = 0; $i < count($m); $i +
= 2 ) {
2069 if (!isset($m[$i +
1]) ||
$m[$i +
1] == "") continue;
2071 if( strstr($hl, "<!--MWTEMPLATESECTION") ) {
2075 preg_match('/^(={1,6})(.*?)(={1,6})\s*?$/m', $hl, $m2);
2076 $text .= $m2[1] . $m2[2] . "<!--MWTEMPLATESECTION="
2077 . $encodedname . "&" . base64_encode("$nsec") . "-->" . $m2[3];
2084 # Empties the template path
2085 $this->mTemplatePath
= array();
2094 * Triple brace replacement -- used for template arguments
2097 function argSubstitution( $matches ) {
2098 $arg = trim( $matches[1] );
2099 $text = $matches[0];
2100 $inputArgs = end( $this->mArgStack
);
2102 if ( array_key_exists( $arg, $inputArgs ) ) {
2103 $text = $inputArgs[$arg];
2110 * Returns true if the function is allowed to include this entity
2113 function incrementIncludeCount( $dbk ) {
2114 if ( !array_key_exists( $dbk, $this->mIncludeCount
) ) {
2115 $this->mIncludeCount
[$dbk] = 0;
2117 if ( ++
$this->mIncludeCount
[$dbk] <= MAX_INCLUDE_REPEAT
) {
2126 * Cleans up HTML, removes dangerous tags and attributes, and
2127 * removes HTML comments
2130 function removeHTMLtags( $text ) {
2131 global $wgUseTidy, $wgUserHtml;
2132 $fname = 'Parser::removeHTMLtags';
2133 wfProfileIn( $fname );
2136 $htmlpairs = array( # Tags that must be closed
2137 'b', 'del', 'i', 'ins', 'u', 'font', 'big', 'small', 'sub', 'sup', 'h1',
2138 'h2', 'h3', 'h4', 'h5', 'h6', 'cite', 'code', 'em', 's',
2139 'strike', 'strong', 'tt', 'var', 'div', 'center',
2140 'blockquote', 'ol', 'ul', 'dl', 'table', 'caption', 'pre',
2141 'ruby', 'rt' , 'rb' , 'rp', 'p'
2143 $htmlsingle = array(
2144 'br', 'hr', 'li', 'dt', 'dd'
2146 $htmlnest = array( # Tags that can be nested--??
2147 'table', 'tr', 'td', 'th', 'div', 'blockquote', 'ol', 'ul',
2148 'dl', 'font', 'big', 'small', 'sub', 'sup'
2150 $tabletags = array( # Can only appear inside table
2154 $htmlpairs = array();
2155 $htmlsingle = array();
2156 $htmlnest = array();
2157 $tabletags = array();
2160 $htmlsingle = array_merge( $tabletags, $htmlsingle );
2161 $htmlelements = array_merge( $htmlsingle, $htmlpairs );
2163 $htmlattrs = $this->getHTMLattrs () ;
2165 # Remove HTML comments
2166 $text = $this->removeHTMLcomments( $text );
2168 $bits = explode( '<', $text );
2169 $text = array_shift( $bits );
2171 $tagstack = array(); $tablestack = array();
2172 foreach ( $bits as $x ) {
2173 $prev = error_reporting( E_ALL
& ~
( E_NOTICE | E_WARNING
) );
2174 preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
2176 list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
2177 error_reporting( $prev );
2180 if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
2184 if ( ! in_array( $t, $htmlsingle ) &&
2185 ( $ot = @array_pop
( $tagstack ) ) != $t ) {
2186 @array_push
( $tagstack, $ot );
2189 if ( $t == 'table' ) {
2190 $tagstack = array_pop( $tablestack );
2195 # Keep track for later
2196 if ( in_array( $t, $tabletags ) &&
2197 ! in_array( 'table', $tagstack ) ) {
2199 } else if ( in_array( $t, $tagstack ) &&
2200 ! in_array ( $t , $htmlnest ) ) {
2202 } else if ( ! in_array( $t, $htmlsingle ) ) {
2203 if ( $t == 'table' ) {
2204 array_push( $tablestack, $tagstack );
2205 $tagstack = array();
2207 array_push( $tagstack, $t );
2209 # Strip non-approved attributes from the tag
2210 $newparams = $this->fixTagAttributes($params);
2214 $rest = str_replace( '>', '>', $rest );
2215 $text .= "<$slash$t $newparams$brace$rest";
2219 $text .= '<' . str_replace( '>', '>', $x);
2221 # Close off any remaining tags
2222 while ( is_array( $tagstack ) && ($t = array_pop( $tagstack )) ) {
2224 if ( $t == 'table' ) { $tagstack = array_pop( $tablestack ); }
2227 # this might be possible using tidy itself
2228 foreach ( $bits as $x ) {
2229 preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
2231 @list
( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
2232 if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
2233 $newparams = $this->fixTagAttributes($params);
2234 $rest = str_replace( '>', '>', $rest );
2235 $text .= "<$slash$t $newparams$brace$rest";
2237 $text .= '<' . str_replace( '>', '>', $x);
2241 wfProfileOut( $fname );
2246 * Remove '<!--', '-->', and everything between.
2247 * To avoid leaving blank lines, when a comment is both preceded
2248 * and followed by a newline (ignoring spaces), trim leading and
2249 * trailing spaces and one of the newlines.
2253 function removeHTMLcomments( $text ) {
2254 $fname='Parser::removeHTMLcomments';
2255 wfProfileIn( $fname );
2256 while (($start = strpos($text, '<!--')) !== false) {
2257 $end = strpos($text, '-->', $start +
4);
2258 if ($end === false) {
2259 # Unterminated comment; bail out
2265 # Trim space and newline if the comment is both
2266 # preceded and followed by a newline
2267 $spaceStart = max($start - 1, 0);
2268 $spaceLen = $end - $spaceStart;
2269 while (substr($text, $spaceStart, 1) === ' ' && $spaceStart > 0) {
2273 while (substr($text, $spaceStart +
$spaceLen, 1) === ' ')
2275 if (substr($text, $spaceStart, 1) === "\n" and substr($text, $spaceStart +
$spaceLen, 1) === "\n") {
2276 # Remove the comment, leading and trailing
2277 # spaces, and leave only one newline.
2278 $text = substr_replace($text, "\n", $spaceStart, $spaceLen +
1);
2281 # Remove just the comment.
2282 $text = substr_replace($text, '', $start, $end - $start);
2285 wfProfileOut( $fname );
2290 * This function accomplishes several tasks:
2291 * 1) Auto-number headings if that option is enabled
2292 * 2) Add an [edit] link to sections for logged in users who have enabled the option
2293 * 3) Add a Table of contents on the top for users who have enabled the option
2294 * 4) Auto-anchor headings
2296 * It loops through all headlines, collects the necessary data, then splits up the
2297 * string and re-inserts the newly formatted headlines.
2300 /* private */ function formatHeadings( $text, $isMain=true ) {
2301 global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders;
2303 $doNumberHeadings = $this->mOptions
->getNumberHeadings();
2304 $doShowToc = $this->mOptions
->getShowToc();
2305 $forceTocHere = false;
2306 if( !$this->mTitle
->userCanEdit() ) {
2308 $rightClickHack = 0;
2310 $showEditLink = $this->mOptions
->getEditSection();
2311 $rightClickHack = $this->mOptions
->getEditSectionOnRightClick();
2314 # Inhibit editsection links if requested in the page
2315 $esw =& MagicWord
::get( MAG_NOEDITSECTION
);
2316 if( $esw->matchAndRemove( $text ) ) {
2319 # if the string __NOTOC__ (not case-sensitive) occurs in the HTML,
2321 $mw =& MagicWord
::get( MAG_NOTOC
);
2322 if( $mw->matchAndRemove( $text ) ) {
2326 # never add the TOC to the Main Page. This is an entry page that should not
2327 # be more than 1-2 screens large anyway
2328 if( $this->mTitle
->getPrefixedText() == wfMsg('mainpage') ) {
2332 # Get all headlines for numbering them and adding funky stuff like [edit]
2333 # links - this is for later, but we need the number of headlines right now
2334 $numMatches = preg_match_all( '/<H([1-6])(.*?' . '>)(.*?)<\/H[1-6]>/i', $text, $matches );
2336 # if there are fewer than 4 headlines in the article, do not show TOC
2337 if( $numMatches < 4 ) {
2341 # if the string __TOC__ (not case-sensitive) occurs in the HTML,
2342 # override above conditions and always show TOC at that place
2343 $mw =& MagicWord
::get( MAG_TOC
);
2344 if ($mw->match( $text ) ) {
2346 $forceTocHere = true;
2348 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
2349 # override above conditions and always show TOC above first header
2350 $mw =& MagicWord
::get( MAG_FORCETOC
);
2351 if ($mw->matchAndRemove( $text ) ) {
2358 # We need this to perform operations on the HTML
2359 $sk =& $this->mOptions
->getSkin();
2363 $sectionCount = 0; # headlineCount excluding template sections
2365 # Ugh .. the TOC should have neat indentation levels which can be
2366 # passed to the skin functions. These are determined here
2371 $sublevelCount = array();
2374 foreach( $matches[3] as $headline ) {
2376 $templatetitle = "";
2377 $templatesection = 0;
2379 if (preg_match("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", $headline, $mat)) {
2381 $templatetitle = base64_decode($mat[1]);
2382 $templatesection = 1 +
(int)base64_decode($mat[2]);
2383 $headline = preg_replace("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", "", $headline);
2388 $prevlevel = $level;
2390 $level = $matches[1][$headlineCount];
2391 if( ( $doNumberHeadings ||
$doShowToc ) && $prevlevel && $level > $prevlevel ) {
2392 # reset when we enter a new level
2393 $sublevelCount[$level] = 0;
2394 $toc .= $sk->tocIndent( $level - $prevlevel );
2395 $toclevel +
= $level - $prevlevel;
2397 if( ( $doNumberHeadings ||
$doShowToc ) && $level < $prevlevel ) {
2398 # reset when we step back a level
2399 $sublevelCount[$level+
1]=0;
2400 $toc .= $sk->tocUnindent( $prevlevel - $level );
2401 $toclevel -= $prevlevel - $level;
2403 # count number of headlines for each level
2404 @$sublevelCount[$level]++
;
2405 if( $doNumberHeadings ||
$doShowToc ) {
2407 for( $i = 1; $i <= $level; $i++
) {
2408 if( !empty( $sublevelCount[$i] ) ) {
2412 $numbering .= $wgContLang->formatNum( $sublevelCount[$i] );
2418 # The canonized header is a version of the header text safe to use for links
2419 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
2420 $canonized_headline = $this->unstrip( $headline, $this->mStripState
);
2421 $canonized_headline = $this->unstripNoWiki( $headline, $this->mStripState
);
2423 # Remove link placeholders by the link text.
2424 # <!--LINK number-->
2426 # link text with suffix
2427 $canonized_headline = preg_replace( '/<!--LINK ([0-9]*)-->/e',
2428 "\$wgLinkHolders['texts'][\$1]",
2429 $canonized_headline );
2432 $canonized_headline = preg_replace( '/<.*?' . '>/','',$canonized_headline );
2433 $tocline = trim( $canonized_headline );
2434 $canonized_headline = urlencode( do_html_entity_decode( str_replace(' ', '_', $tocline), ENT_COMPAT
, $wgInputEncoding ) );
2435 $replacearray = array(
2439 $canonized_headline = str_replace(array_keys($replacearray),array_values($replacearray),$canonized_headline);
2440 $refer[$headlineCount] = $canonized_headline;
2442 # count how many in assoc. array so we can track dupes in anchors
2443 @$refers[$canonized_headline]++
;
2444 $refcount[$headlineCount]=$refers[$canonized_headline];
2446 # Prepend the number to the heading text
2448 if( $doNumberHeadings ||
$doShowToc ) {
2449 $tocline = $numbering . ' ' . $tocline;
2451 # Don't number the heading if it is the only one (looks silly)
2452 if( $doNumberHeadings && count( $matches[3] ) > 1) {
2453 # the two are different if the line contains a link
2454 $headline=$numbering . ' ' . $headline;
2458 # Create the anchor for linking from the TOC to the section
2459 $anchor = $canonized_headline;
2460 if($refcount[$headlineCount] > 1 ) {
2461 $anchor .= '_' . $refcount[$headlineCount];
2463 if( $doShowToc && ( !isset($wgMaxTocLevel) ||
$toclevel<$wgMaxTocLevel ) ) {
2464 $toc .= $sk->tocLine($anchor,$tocline,$toclevel);
2466 if( $showEditLink && ( !$istemplate ||
$templatetitle !== "" ) ) {
2467 if ( empty( $head[$headlineCount] ) ) {
2468 $head[$headlineCount] = '';
2471 $head[$headlineCount] .= $sk->editSectionLinkForOther($templatetitle, $templatesection);
2473 $head[$headlineCount] .= $sk->editSectionLink($this->mTitle
, $sectionCount+
1);
2476 # Add the edit section span
2477 if( $rightClickHack ) {
2479 $headline = $sk->editSectionScriptForOther($templatetitle, $templatesection, $headline);
2481 $headline = $sk->editSectionScript($this->mTitle
, $sectionCount+
1,$headline);
2484 # give headline the correct <h#> tag
2485 @$head[$headlineCount] .= "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount] .$headline.'</h'.$level.'>';
2493 $toclines = $headlineCount;
2494 $toc .= $sk->tocUnindent( $toclevel );
2495 $toc = $sk->tocTable( $toc );
2498 # split up and insert constructed headlines
2500 $blocks = preg_split( '/<H[1-6].*?' . '>.*?<\/H[1-6]>/i', $text );
2503 foreach( $blocks as $block ) {
2504 if( $showEditLink && $headlineCount > 0 && $i == 0 && $block != "\n" ) {
2505 # This is the [edit] link that appears for the top block of text when
2506 # section editing is enabled
2508 # Disabled because it broke block formatting
2509 # For example, a bullet point in the top line
2510 # $full .= $sk->editSectionLink(0);
2513 if( $doShowToc && !$i && $isMain && !$forceTocHere) {
2514 # Top anchor now in skin
2518 if( !empty( $head[$i] ) ) {
2524 $mw =& MagicWord
::get( MAG_TOC
);
2525 return $mw->replace( $toc, $full );
2532 * Return an HTML link for the "ISBN 123456" text
2535 function magicISBN( $text ) {
2537 $fname = 'Parser::magicISBN';
2538 wfProfileIn( $fname );
2540 $a = split( 'ISBN ', ' '.$text );
2541 if ( count ( $a ) < 2 ) {
2542 wfProfileOut( $fname );
2545 $text = substr( array_shift( $a ), 1);
2546 $valid = '0123456789-ABCDEFGHIJKLMNOPQRSTUVWXYZ';
2548 foreach ( $a as $x ) {
2549 $isbn = $blank = '' ;
2550 while ( ' ' == $x{0} ) {
2552 $x = substr( $x, 1 );
2554 if ( $x == '' ) { # blank isbn
2555 $text .= "ISBN $blank";
2558 while ( strstr( $valid, $x{0} ) != false ) {
2560 $x = substr( $x, 1 );
2562 $num = str_replace( '-', '', $isbn );
2563 $num = str_replace( ' ', '', $num );
2566 $text .= "ISBN $blank$x";
2568 $titleObj = Title
::makeTitle( NS_SPECIAL
, 'Booksources' );
2569 $text .= '<a href="' .
2570 $titleObj->escapeLocalUrl( 'isbn='.$num ) .
2571 "\" class=\"internal\">ISBN $isbn</a>";
2575 wfProfileOut( $fname );
2580 * Return an HTML link for the "GEO ..." text
2583 function magicGEO( $text ) {
2584 global $wgLang, $wgUseGeoMode;
2585 $fname = 'Parser::magicGEO';
2586 wfProfileIn( $fname );
2588 # These next five lines are only for the ~35000 U.S. Census Rambot pages...
2589 $directions = array ( 'N' => 'North' , 'S' => 'South' , 'E' => 'East' , 'W' => 'West' ) ;
2590 $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['N']}, (\d+)°(\d+)'(\d+)\" {$directions['W']}/" , "(GEO +\$1.\$2.\$3:-\$4.\$5.\$6)" , $text ) ;
2591 $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['N']}, (\d+)°(\d+)'(\d+)\" {$directions['E']}/" , "(GEO +\$1.\$2.\$3:+\$4.\$5.\$6)" , $text ) ;
2592 $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['S']}, (\d+)°(\d+)'(\d+)\" {$directions['W']}/" , "(GEO +\$1.\$2.\$3:-\$4.\$5.\$6)" , $text ) ;
2593 $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['S']}, (\d+)°(\d+)'(\d+)\" {$directions['E']}/" , "(GEO +\$1.\$2.\$3:+\$4.\$5.\$6)" , $text ) ;
2595 $a = split( 'GEO ', ' '.$text );
2596 if ( count ( $a ) < 2 ) {
2597 wfProfileOut( $fname );
2600 $text = substr( array_shift( $a ), 1);
2601 $valid = '0123456789.+-:';
2603 foreach ( $a as $x ) {
2604 $geo = $blank = '' ;
2605 while ( ' ' == $x{0} ) {
2607 $x = substr( $x, 1 );
2609 while ( strstr( $valid, $x{0} ) != false ) {
2611 $x = substr( $x, 1 );
2613 $num = str_replace( '+', '', $geo );
2614 $num = str_replace( ' ', '', $num );
2616 if ( '' == $num ||
count ( explode ( ':' , $num , 3 ) ) < 2 ) {
2617 $text .= "GEO $blank$x";
2619 $titleObj = Title
::makeTitle( NS_SPECIAL
, 'Geo' );
2620 $text .= '<a href="' .
2621 $titleObj->escapeLocalUrl( 'coordinates='.$num ) .
2622 "\" class=\"internal\">GEO $geo</a>";
2626 wfProfileOut( $fname );
2631 * Return an HTML link for the "RFC 1234" text
2633 * @param string $text text to be processed
2635 function magicRFC( $text ) {
2638 $valid = '0123456789';
2641 $a = split( 'RFC ', ' '.$text );
2642 if ( count ( $a ) < 2 ) return $text;
2643 $text = substr( array_shift( $a ), 1);
2645 /* Check if RFC keyword is preceed by [[.
2646 * This test is made here cause of the array_shift above
2647 * that prevent the test to be done in the foreach.
2649 if(substr($text, -2) == '[[') { $internal = true; }
2651 foreach ( $a as $x ) {
2652 /* token might be empty if we have RFC RFC 1234 */
2658 $rfc = $blank = '' ;
2660 /** remove and save whitespaces in $blank */
2661 while ( $x{0} == ' ' ) {
2663 $x = substr( $x, 1 );
2666 /** remove and save the rfc number in $rfc */
2667 while ( strstr( $valid, $x{0} ) != false ) {
2669 $x = substr( $x, 1 );
2673 /* call back stripped spaces*/
2674 $text .= "RFC $blank$x";
2675 } elseif( $internal) {
2677 $text .= "RFC $rfc$x";
2679 /* build the external link*/
2680 $url = wfmsg( 'rfcurl' );
2681 $url = str_replace( '$1', $rfc, $url);
2682 $sk =& $this->mOptions
->getSkin();
2683 $la = $sk->getExternalLinkAttributes( $url, 'RFC '.$rfc );
2684 $text .= "<a href='{$url}'{$la}>RFC {$rfc}</a>{$x}";
2687 /* Check if the next RFC keyword is preceed by [[ */
2688 $internal = (substr($x,-2) == '[[');
2694 * Transform wiki markup when saving a page by doing \r\n -> \n
2695 * conversion, substitting signatures, {{subst:}} templates, etc.
2697 * @param string $text the text to transform
2698 * @param Title &$title the Title object for the current article
2699 * @param User &$user the User object describing the current user
2700 * @param ParserOptions $options parsing options
2701 * @param bool $clearState whether to clear the parser state first
2702 * @return string the altered wiki markup
2705 function preSaveTransform( $text, &$title, &$user, $options, $clearState = true ) {
2706 $this->mOptions
= $options;
2707 $this->mTitle
=& $title;
2708 $this->mOutputType
= OT_WIKI
;
2710 if ( $clearState ) {
2711 $this->clearState();
2714 $stripState = false;
2718 $text = str_replace(array_keys($pairs), array_values($pairs), $text);
2719 $text = $this->strip( $text, $stripState, false );
2720 $text = $this->pstPass2( $text, $user );
2721 $text = $this->unstrip( $text, $stripState );
2722 $text = $this->unstripNoWiki( $text, $stripState );
2727 * Pre-save transform helper function
2730 function pstPass2( $text, &$user ) {
2731 global $wgLang, $wgContLang, $wgLocaltimezone;
2733 # Variable replacement
2734 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
2735 $text = $this->replaceVariables( $text );
2739 $n = $user->getName();
2740 $k = $user->getOption( 'nickname' );
2741 if ( '' == $k ) { $k = $n; }
2742 if(isset($wgLocaltimezone)) {
2743 $oldtz = getenv('TZ'); putenv('TZ='.$wgLocaltimezone);
2745 /* Note: this is an ugly timezone hack for the European wikis */
2746 $d = $wgContLang->timeanddate( date( 'YmdHis' ), false ) .
2747 ' (' . date( 'T' ) . ')';
2748 if(isset($wgLocaltimezone)) putenv('TZ='.$oldtzs);
2750 $text = preg_replace( '/~~~~~/', $d, $text );
2751 $text = preg_replace( '/~~~~/', '[[' . $wgContLang->getNsText( NS_USER
) . ":$n|$k]] $d", $text );
2752 $text = preg_replace( '/~~~/', '[[' . $wgContLang->getNsText( NS_USER
) . ":$n|$k]]", $text );
2754 # Context links: [[|name]] and [[name (context)|]]
2756 $tc = "[&;%\\-,.\\(\\)' _0-9A-Za-z\\/:\\x80-\\xff]";
2757 $np = "[&;%\\-,.' _0-9A-Za-z\\/:\\x80-\\xff]"; # No parens
2758 $namespacechar = '[ _0-9A-Za-z\x80-\xff]'; # Namespaces can use non-ascii!
2759 $conpat = "/^({$np}+) \\(({$tc}+)\\)$/";
2761 $p1 = "/\[\[({$np}+) \\(({$np}+)\\)\\|]]/"; # [[page (context)|]]
2762 $p2 = "/\[\[\\|({$tc}+)]]/"; # [[|page]]
2763 $p3 = "/\[\[(:*$namespacechar+):({$np}+)\\|]]/"; # [[namespace:page|]] and [[:namespace:page|]]
2764 $p4 = "/\[\[(:*$namespacechar+):({$np}+) \\(({$np}+)\\)\\|]]/"; # [[ns:page (cont)|]] and [[:ns:page (cont)|]]
2766 $t = $this->mTitle
->getText();
2767 if ( preg_match( $conpat, $t, $m ) ) {
2770 $text = preg_replace( $p4, '[[\\1:\\2 (\\3)|\\2]]', $text );
2771 $text = preg_replace( $p1, '[[\\1 (\\2)|\\1]]', $text );
2772 $text = preg_replace( $p3, '[[\\1:\\2|\\2]]', $text );
2774 if ( '' == $context ) {
2775 $text = preg_replace( $p2, '[[\\1]]', $text );
2777 $text = preg_replace( $p2, "[[\\1 ({$context})|\\1]]", $text );
2780 # Trim trailing whitespace
2781 # MAG_END (__END__) tag allows for trailing
2782 # whitespace to be deliberately included
2783 $text = rtrim( $text );
2784 $mw =& MagicWord
::get( MAG_END
);
2785 $mw->matchAndRemove( $text );
2791 * Set up some variables which are usually set up in parse()
2792 * so that an external function can call some class members with confidence
2795 function startExternalParse( &$title, $options, $outputType, $clearState = true ) {
2796 $this->mTitle
=& $title;
2797 $this->mOptions
= $options;
2798 $this->mOutputType
= $outputType;
2799 if ( $clearState ) {
2800 $this->clearState();
2805 * Transform a MediaWiki message by replacing magic variables.
2807 * @param string $text the text to transform
2808 * @param ParserOptions $options options
2809 * @return string the text with variables substituted
2812 function transformMsg( $text, $options ) {
2814 static $executing = false;
2816 # Guard against infinite recursion
2822 $this->mTitle
= $wgTitle;
2823 $this->mOptions
= $options;
2824 $this->mOutputType
= OT_MSG
;
2825 $this->clearState();
2826 $text = $this->replaceVariables( $text );
2833 * Create an HTML-style tag, e.g. <yourtag>special text</yourtag>
2834 * Callback will be called with the text within
2835 * Transform and return the text within
2838 function setHook( $tag, $callback ) {
2839 $oldVal = @$this->mTagHooks
[$tag];
2840 $this->mTagHooks
[$tag] = $callback;
2845 * Replace <!--LINK--> link placeholders with actual links, in the buffer
2846 * Placeholders created in Skin::makeLinkObj()
2847 * Returns an array of links found, indexed by PDBK:
2851 * $options is a bit field, RLH_FOR_UPDATE to select for update
2853 function replaceLinkHolders( &$text, $options = 0 ) {
2854 global $wgUser, $wgLinkCache, $wgUseOldExistenceCheck, $wgLinkHolders;
2856 if ( $wgUseOldExistenceCheck ) {
2860 $fname = 'Parser::replaceLinkHolders';
2861 wfProfileIn( $fname );
2866 #if ( !empty( $tmpLinks[0] ) ) { #TODO
2867 if ( !empty( $wgLinkHolders['namespaces'] ) ) {
2868 wfProfileIn( $fname.'-check' );
2869 $dbr =& wfGetDB( DB_SLAVE
);
2870 $cur = $dbr->tableName( 'cur' );
2871 $sk = $wgUser->getSkin();
2872 $threshold = $wgUser->getOption('stubthreshold');
2875 asort( $wgLinkHolders['namespaces'] );
2879 foreach ( $wgLinkHolders['namespaces'] as $key => $val ) {
2881 $title = $wgLinkHolders['titles'][$key];
2883 # Skip invalid entries.
2884 # Result will be ugly, but prevents crash.
2885 if ( is_null( $title ) ) {
2888 $pdbk = $pdbks[$key] = $title->getPrefixedDBkey();
2890 # Check if it's in the link cache already
2891 if ( $wgLinkCache->getGoodLinkID( $pdbk ) ) {
2892 $colours[$pdbk] = 1;
2893 } elseif ( $wgLinkCache->isBadLink( $pdbk ) ) {
2894 $colours[$pdbk] = 0;
2896 # Not in the link cache, add it to the query
2897 if ( !isset( $current ) ) {
2899 $query = "SELECT cur_id, cur_namespace, cur_title";
2900 if ( $threshold > 0 ) {
2901 $query .= ", LENGTH(cur_text) AS cur_len, cur_is_redirect";
2903 $query .= " FROM $cur WHERE (cur_namespace=$val AND cur_title IN(";
2904 } elseif ( $current != $val ) {
2906 $query .= ")) OR (cur_namespace=$val AND cur_title IN(";
2911 $query .= $dbr->addQuotes( $wgLinkHolders['dbkeys'][$key] );
2916 if ( $options & RLH_FOR_UPDATE
) {
2917 $query .= ' FOR UPDATE';
2920 $res = $dbr->query( $query, $fname );
2922 # Fetch data and form into an associative array
2923 # non-existent = broken
2926 while ( $s = $dbr->fetchObject($res) ) {
2927 $title = Title
::makeTitle( $s->cur_namespace
, $s->cur_title
);
2928 $pdbk = $title->getPrefixedDBkey();
2929 $wgLinkCache->addGoodLink( $s->cur_id
, $pdbk );
2931 if ( $threshold > 0 ) {
2932 $size = $s->cur_len
;
2933 if ( $s->cur_is_redirect ||
$s->cur_namespace
!= 0 ||
$length < $threshold ) {
2934 $colours[$pdbk] = 1;
2936 $colours[$pdbk] = 2;
2939 $colours[$pdbk] = 1;
2943 wfProfileOut( $fname.'-check' );
2945 # Construct search and replace arrays
2946 wfProfileIn( $fname.'-construct' );
2947 global $outputReplace;
2948 $outputReplace = array();
2949 foreach ( $wgLinkHolders['namespaces'] as $key => $ns ) {
2950 $pdbk = $pdbks[$key];
2951 $searchkey = '<!--LINK '.$key.'-->';
2952 $title = $wgLinkHolders['titles'][$key];
2953 if ( empty( $colours[$pdbk] ) ) {
2954 $wgLinkCache->addBadLink( $pdbk );
2955 $colours[$pdbk] = 0;
2956 $outputReplace[$searchkey] = $sk->makeBrokenLinkObj( $title,
2957 $wgLinkHolders['texts'][$key],
2958 $wgLinkHolders['queries'][$key] );
2959 } elseif ( $colours[$pdbk] == 1 ) {
2960 $outputReplace[$searchkey] = $sk->makeKnownLinkObj( $title,
2961 $wgLinkHolders['texts'][$key],
2962 $wgLinkHolders['queries'][$key] );
2963 } elseif ( $colours[$pdbk] == 2 ) {
2964 $outputReplace[$searchkey] = $sk->makeStubLinkObj( $title,
2965 $wgLinkHolders['texts'][$key],
2966 $wgLinkHolders['queries'][$key] );
2969 wfProfileOut( $fname.'-construct' );
2972 wfProfileIn( $fname.'-replace' );
2974 $text = preg_replace_callback(
2975 '/(<!--LINK .*?-->)/',
2976 "outputReplaceMatches",
2978 wfProfileOut( $fname.'-replace' );
2980 wfProfileIn( $fname.'-interwiki' );
2981 global $wgInterwikiLinkHolders;
2982 $outputReplace = $wgInterwikiLinkHolders;
2983 $text = preg_replace_callback(
2984 '/<!--IWLINK (.*?)-->/',
2985 "outputReplaceMatches",
2987 wfProfileOut( $fname.'-interwiki' );
2990 wfProfileOut( $fname );
2994 * Renders an image gallery from a text with one line per image.
2995 * text labels may be given by using |-style alternative text. E.g.
2996 * Image:one.jpg|The number "1"
2997 * Image:tree.jpg|A tree
2998 * given as text will return the HTML of a gallery with two images,
2999 * labeled 'The number "1"' and
3002 function renderImageGallery( $text ) {
3003 global $wgLinkCache;
3004 $ig = new ImageGallery();
3005 $ig->setShowBytes( false );
3006 $ig->setShowFilename( false );
3007 $lines = explode( "\n", $text );
3008 foreach ( $lines as $line ) {
3009 preg_match( "/^([^|]+)(\\|(.*))?$/", $line, $matches );
3011 if ( count( $matches ) == 0 ) {
3014 $nt = Title
::newFromURL( $matches[1] );
3015 if ( isset( $matches[3] ) ) {
3016 $label = $matches[3];
3020 $ig->add( Image
::newFromTitle( $nt ), $label );
3021 $wgLinkCache->addImageLinkObj( $nt );
3023 return $ig->toHTML();
3029 * @package MediaWiki
3033 var $mText, $mLanguageLinks, $mCategoryLinks, $mContainsOldMagic;
3034 var $mCacheTime; # Used in ParserCache
3036 function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(),
3037 $containsOldMagic = false )
3039 $this->mText
= $text;
3040 $this->mLanguageLinks
= $languageLinks;
3041 $this->mCategoryLinks
= $categoryLinks;
3042 $this->mContainsOldMagic
= $containsOldMagic;
3043 $this->mCacheTime
= '';
3046 function getText() { return $this->mText
; }
3047 function getLanguageLinks() { return $this->mLanguageLinks
; }
3048 function getCategoryLinks() { return $this->mCategoryLinks
; }
3049 function getCacheTime() { return $this->mCacheTime
; }
3050 function containsOldMagic() { return $this->mContainsOldMagic
; }
3051 function setText( $text ) { return wfSetVar( $this->mText
, $text ); }
3052 function setLanguageLinks( $ll ) { return wfSetVar( $this->mLanguageLinks
, $ll ); }
3053 function setCategoryLinks( $cl ) { return wfSetVar( $this->mCategoryLinks
, $cl ); }
3054 function setContainsOldMagic( $com ) { return wfSetVar( $this->mContainsOldMagic
, $com ); }
3055 function setCacheTime( $t ) { return wfSetVar( $this->mCacheTime
, $t ); }
3057 function merge( $other ) {
3058 $this->mLanguageLinks
= array_merge( $this->mLanguageLinks
, $other->mLanguageLinks
);
3059 $this->mCategoryLinks
= array_merge( $this->mCategoryLinks
, $this->mLanguageLinks
);
3060 $this->mContainsOldMagic
= $this->mContainsOldMagic ||
$other->mContainsOldMagic
;
3066 * Set options of the Parser
3068 * @package MediaWiki
3072 # All variables are private
3073 var $mUseTeX; # Use texvc to expand <math> tags
3074 var $mUseDynamicDates; # Use $wgDateFormatter to format dates
3075 var $mInterwikiMagic; # Interlanguage links are removed and returned in an array
3076 var $mAllowExternalImages; # Allow external images inline
3077 var $mSkin; # Reference to the preferred skin
3078 var $mDateFormat; # Date format index
3079 var $mEditSection; # Create "edit section" links
3080 var $mEditSectionOnRightClick; # Generate JavaScript to edit section on right click
3081 var $mNumberHeadings; # Automatically number headings
3082 var $mShowToc; # Show table of contents
3084 function getUseTeX() { return $this->mUseTeX
; }
3085 function getUseDynamicDates() { return $this->mUseDynamicDates
; }
3086 function getInterwikiMagic() { return $this->mInterwikiMagic
; }
3087 function getAllowExternalImages() { return $this->mAllowExternalImages
; }
3088 function getSkin() { return $this->mSkin
; }
3089 function getDateFormat() { return $this->mDateFormat
; }
3090 function getEditSection() { return $this->mEditSection
; }
3091 function getEditSectionOnRightClick() { return $this->mEditSectionOnRightClick
; }
3092 function getNumberHeadings() { return $this->mNumberHeadings
; }
3093 function getShowToc() { return $this->mShowToc
; }
3095 function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX
, $x ); }
3096 function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates
, $x ); }
3097 function setInterwikiMagic( $x ) { return wfSetVar( $this->mInterwikiMagic
, $x ); }
3098 function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages
, $x ); }
3099 function setDateFormat( $x ) { return wfSetVar( $this->mDateFormat
, $x ); }
3100 function setEditSection( $x ) { return wfSetVar( $this->mEditSection
, $x ); }
3101 function setEditSectionOnRightClick( $x ) { return wfSetVar( $this->mEditSectionOnRightClick
, $x ); }
3102 function setNumberHeadings( $x ) { return wfSetVar( $this->mNumberHeadings
, $x ); }
3103 function setShowToc( $x ) { return wfSetVar( $this->mShowToc
, $x ); }
3105 function setSkin( &$x ) { $this->mSkin
=& $x; }
3107 # Get parser options
3108 /* static */ function newFromUser( &$user ) {
3109 $popts = new ParserOptions
;
3110 $popts->initialiseFromUser( $user );
3115 function initialiseFromUser( &$userInput ) {
3116 global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages;
3118 $fname = 'ParserOptions::initialiseFromUser';
3119 wfProfileIn( $fname );
3120 if ( !$userInput ) {
3122 $user->setLoaded( true );
3124 $user =& $userInput;
3127 $this->mUseTeX
= $wgUseTeX;
3128 $this->mUseDynamicDates
= $wgUseDynamicDates;
3129 $this->mInterwikiMagic
= $wgInterwikiMagic;
3130 $this->mAllowExternalImages
= $wgAllowExternalImages;
3131 wfProfileIn( $fname.'-skin' );
3132 $this->mSkin
=& $user->getSkin();
3133 wfProfileOut( $fname.'-skin' );
3134 $this->mDateFormat
= $user->getOption( 'date' );
3135 $this->mEditSection
= $user->getOption( 'editsection' );
3136 $this->mEditSectionOnRightClick
= $user->getOption( 'editsectiononrightclick' );
3137 $this->mNumberHeadings
= $user->getOption( 'numberheadings' );
3138 $this->mShowToc
= $user->getOption( 'showtoc' );
3139 wfProfileOut( $fname );
3146 * Callback function used by Parser::replaceLinkHolders()
3147 * to substitute link placeholders.
3149 function &outputReplaceMatches($matches) {
3150 global $outputReplace;
3151 return $outputReplace[$matches[1]];
3155 * Return the total number of articles
3157 function wfNumberOfArticles() {
3158 global $wgNumberOfArticles;
3161 return $wgNumberOfArticles;
3165 * Get various statistics from the database
3168 function wfLoadSiteStats() {
3169 global $wgNumberOfArticles, $wgTotalViews, $wgTotalEdits;
3170 $fname = 'wfLoadSiteStats';
3172 if ( -1 != $wgNumberOfArticles ) return;
3173 $dbr =& wfGetDB( DB_SLAVE
);
3174 $s = $dbr->selectRow( 'site_stats',
3175 array( 'ss_total_views', 'ss_total_edits', 'ss_good_articles' ),
3176 array( 'ss_row_id' => 1 ), $fname
3179 if ( $s === false ) {
3182 $wgTotalViews = $s->ss_total_views
;
3183 $wgTotalEdits = $s->ss_total_edits
;
3184 $wgNumberOfArticles = $s->ss_good_articles
;
3188 function wfEscapeHTMLTagsOnly( $in ) {
3190 array( '"', '>', '<' ),
3191 array( '"', '>', '<' ),