30eb6b5596ee38c96c22227c7beb2fff24590255
3 // require_once('Tokenizer.php');
8 * Processes wiki markup
10 * There are two main entry points into the Parser class:
12 * produces HTML output
14 * produces altered wiki markup.
17 * objects: $wgLang, $wgDateFormatter, $wgLinkCache, $wgCurParser
19 * NOT $wgArticle, $wgUser or $wgTitle. Keep them away!
22 * $wgUseTex*, $wgUseDynamicDates*, $wgInterwikiMagic*,
23 * $wgNamespacesWithSubpages, $wgAllowExternalImages*,
26 * * only within ParserOptions
32 * Variable substitution O(N^2) attack
34 * Without countermeasures, it would be possible to attack the parser by saving
35 * a page filled with a large number of inclusions of large pages. The size of
36 * the generated page would be proportional to the square of the input size.
37 * Hence, we limit the number of inclusions of any given page, thus bringing any
38 * attack back to O(N).
40 define( 'MAX_INCLUDE_REPEAT', 100 );
41 define( 'MAX_INCLUDE_SIZE', 1000000 ); // 1 Million
43 # Allowed values for $mOutputType
44 define( 'OT_HTML', 1 );
45 define( 'OT_WIKI', 2 );
46 define( 'OT_MSG' , 3 );
48 # string parameter for extractTags which will cause it
49 # to strip HTML comments in addition to regular
50 # <XML>-style tags. This should not be anything we
51 # may want to use in wikisyntax
52 define( 'STRIP_COMMENTS', 'HTMLCommentStrip' );
54 # prefix for escaping, used in two functions at least
55 define( 'UNIQ_PREFIX', 'NaodW29');
57 # Constants needed for external link processing
58 define( 'URL_PROTOCOLS', 'http|https|ftp|irc|gopher|news|mailto' );
59 define( 'HTTP_PROTOCOLS', 'http|https' );
60 # Everything except bracket, space, or control characters
61 define( 'EXT_LINK_URL_CLASS', '[^]\\x00-\\x20\\x7F]' );
62 define( 'INVERSE_EXT_LINK_URL_CLASS', '[\]\\x00-\\x20\\x7F]' );
64 define( 'EXT_LINK_TEXT_CLASS', '[^\]\\x00-\\x1F\\x7F]' );
65 define( 'EXT_IMAGE_FNAME_CLASS', '[A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]' );
66 define( 'EXT_IMAGE_EXTENSIONS', 'gif|png|jpg|jpeg' );
67 define( 'EXT_LINK_BRACKETED', '/\[(('.URL_PROTOCOLS
.'):'.EXT_LINK_URL_CLASS
.'+) *('.EXT_LINK_TEXT_CLASS
.'*?)\]/S' );
68 define( 'EXT_IMAGE_REGEX',
69 '/^('.HTTP_PROTOCOLS
.':)'. # Protocol
70 '('.EXT_LINK_URL_CLASS
.'+)\\/'. # Hostname and path
71 '('.EXT_IMAGE_FNAME_CLASS
.'+)\\.((?i)'.EXT_IMAGE_EXTENSIONS
.')$/S' # Filename
83 # Cleared with clearState():
84 var $mOutput, $mAutonumber, $mDTopen, $mStripState = array();
85 var $mVariables, $mIncludeCount, $mArgStack, $mLastSection, $mInPre;
88 var $mOptions, $mTitle, $mOutputType,
89 $mTemplates, // cache of already loaded templates, avoids
90 // multiple SQL queries for the same string
91 $mTemplatePath; // stores an unsorted hash of all the templates already loaded
92 // in this path. Used for loop detection.
95 $this->mTemplates
= array();
96 $this->mTemplatePath
= array();
97 $this->mTagHooks
= array();
101 function clearState() {
102 $this->mOutput
= new ParserOutput
;
103 $this->mAutonumber
= 0;
104 $this->mLastSection
= "";
105 $this->mDTopen
= false;
106 $this->mVariables
= false;
107 $this->mIncludeCount
= array();
108 $this->mStripState
= array();
109 $this->mArgStack
= array();
110 $this->mInPre
= false;
113 # First pass--just handle <nowiki> sections, pass the rest off
114 # to internalParse() which does all the real work.
116 # Returns a ParserOutput
118 function parse( $text, &$title, $options, $linestart = true, $clearState = true ) {
120 $fname = 'Parser::parse';
121 wfProfileIn( $fname );
127 $this->mOptions
= $options;
128 $this->mTitle
=& $title;
129 $this->mOutputType
= OT_HTML
;
132 $text = $this->strip( $text, $this->mStripState
);
133 $text = $this->internalParse( $text, $linestart );
134 $text = $this->unstrip( $text, $this->mStripState
);
135 # Clean up special characters, only run once, next-to-last before doBlockLevels
138 # french spaces, last one Guillemet-left
139 # only if there is something before the space
140 '/(.) (?=\\?|:|;|!|\\302\\273)/i' => '\\1 \\2',
141 # french spaces, Guillemet-right
142 "/(\\302\\253) /i"=>"\\1 ",
143 '/<hr *>/i' => '<hr />',
144 '/<br *>/i' => '<br />',
145 '/<center *>/i' => '<div class="center">',
146 '/<\\/center *>/i' => '</div>',
147 # Clean up spare ampersands; note that we probably ought to be
148 # more careful about named entities.
149 '/&(?!:amp;|#[Xx][0-9A-fa-f]+;|#[0-9]+;|[a-zA-Z0-9]+;)/' => '&'
151 $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
154 # french spaces, last one Guillemet-left
155 '/ (\\?|:|;|!|\\302\\273)/i' => ' \\1',
156 # french spaces, Guillemet-right
157 '/(\\302\\253) /i' => '\\1 ',
158 '/<center *>/i' => '<div class="center">',
159 '/<\\/center *>/i' => '</div>'
161 $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
164 $text = $this->doBlockLevels( $text, $linestart );
165 $text = $this->unstripNoWiki( $text, $this->mStripState
);
167 $text = $this->tidy($text);
169 $this->mOutput
->setText( $text );
170 wfProfileOut( $fname );
171 return $this->mOutput
;
174 /* static */ function getRandomString() {
175 return dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff));
178 # Replaces all occurrences of <$tag>content</$tag> in the text
179 # with a random marker and returns the new text. the output parameter
180 # $content will be an associative array filled with data on the form
181 # $unique_marker => content.
183 # If $content is already set, the additional entries will be appended
185 # If $tag is set to STRIP_COMMENTS, the function will extract
186 # <!-- HTML comments -->
188 /* static */ function extractTags($tag, $text, &$content, $uniq_prefix = ''){
189 $rnd = $uniq_prefix . '-' . $tag . Parser
::getRandomString();
196 while ( '' != $text ) {
197 if($tag==STRIP_COMMENTS
) {
198 $p = preg_split( '/<!--/i', $text, 2 );
200 $p = preg_split( "/<\\s*$tag\\s*>/i", $text, 2 );
203 if ( ( count( $p ) < 2 ) ||
( '' == $p[1] ) ) {
206 if($tag==STRIP_COMMENTS
) {
207 $q = preg_split( '/-->/i', $p[1], 2 );
209 $q = preg_split( "/<\\/\\s*$tag\\s*>/i", $p[1], 2 );
211 $marker = $rnd . sprintf('%08X', $n++
);
212 $content[$marker] = $q[0];
213 $stripped .= $marker;
220 # Strips and renders <nowiki>, <pre>, <math>, <hiero>
221 # If $render is set, performs necessary rendering operations on plugins
222 # Returns the text, and fills an array with data needed in unstrip()
223 # If the $state is already a valid strip state, it adds to the state
225 # When $stripcomments is set, HTML comments <!-- like this -->
226 # will be stripped in addition to other tags. This is important
227 # for section editing, where these comments cause confusion when
228 # counting the sections in the wikisource
229 function strip( $text, &$state, $stripcomments = false ) {
230 $render = ($this->mOutputType
== OT_HTML
);
231 $html_content = array();
232 $nowiki_content = array();
233 $math_content = array();
234 $pre_content = array();
235 $comment_content = array();
236 $ext_content = array();
238 # Replace any instances of the placeholders
239 $uniq_prefix = UNIQ_PREFIX
;
240 #$text = str_replace( $uniq_prefix, wfHtmlEscapeFirst( $uniq_prefix ), $text );
243 global $wgRawHtml, $wgWhitelistEdit;
244 if( $wgRawHtml && $wgWhitelistEdit ) {
245 $text = Parser
::extractTags('html', $text, $html_content, $uniq_prefix);
246 foreach( $html_content as $marker => $content ) {
248 # Raw and unchecked for validity.
249 $html_content[$marker] = $content;
251 $html_content[$marker] = '<html>'.$content.'</html>';
257 $text = Parser
::extractTags('nowiki', $text, $nowiki_content, $uniq_prefix);
258 foreach( $nowiki_content as $marker => $content ) {
260 $nowiki_content[$marker] = wfEscapeHTMLTagsOnly( $content );
262 $nowiki_content[$marker] = '<nowiki>'.$content.'</nowiki>';
267 $text = Parser
::extractTags('math', $text, $math_content, $uniq_prefix);
268 foreach( $math_content as $marker => $content ){
270 if( $this->mOptions
->getUseTeX() ) {
271 $math_content[$marker] = renderMath( $content );
273 $math_content[$marker] = '<math>'.$content.'<math>';
276 $math_content[$marker] = '<math>'.$content.'</math>';
281 $text = Parser
::extractTags('pre', $text, $pre_content, $uniq_prefix);
282 foreach( $pre_content as $marker => $content ){
284 $pre_content[$marker] = '<pre>' . wfEscapeHTMLTagsOnly( $content ) . '</pre>';
286 $pre_content[$marker] = '<pre>'.$content.'</pre>';
292 $text = Parser
::extractTags(STRIP_COMMENTS
, $text, $comment_content, $uniq_prefix);
293 foreach( $comment_content as $marker => $content ){
294 $comment_content[$marker] = '<!--'.$content.'-->';
299 foreach ( $this->mTagHooks
as $tag => $callback ) {
300 $ext_contents[$tag] = array();
301 $text = Parser
::extractTags( $tag, $text, $ext_content[$tag], $uniq_prefix );
302 foreach( $ext_content[$tag] as $marker => $content ) {
304 $ext_content[$tag][$marker] = $callback( $content );
306 $ext_content[$tag][$marker] = "<$tag>$content</$tag>";
311 # Merge state with the pre-existing state, if there is one
313 $state['html'] = $state['html'] +
$html_content;
314 $state['nowiki'] = $state['nowiki'] +
$nowiki_content;
315 $state['math'] = $state['math'] +
$math_content;
316 $state['pre'] = $state['pre'] +
$pre_content;
317 $state['comment'] = $state['comment'] +
$comment_content;
319 foreach( $ext_content as $tag => $array ) {
320 if ( array_key_exists( $tag, $state ) ) {
321 $state[$tag] = $state[$tag] +
$array;
326 'html' => $html_content,
327 'nowiki' => $nowiki_content,
328 'math' => $math_content,
329 'pre' => $pre_content,
330 'comment' => $comment_content,
336 # always call unstripNoWiki() after this one
337 function unstrip( $text, &$state ) {
338 # Must expand in reverse order, otherwise nested tags will be corrupted
339 $contentDict = end( $state );
340 for ( $contentDict = end( $state ); $contentDict !== false; $contentDict = prev( $state ) ) {
341 if( key($state) != 'nowiki' && key($state) != 'html') {
342 for ( $content = end( $contentDict ); $content !== false; $content = prev( $contentDict ) ) {
343 $text = str_replace( key( $contentDict ), $content, $text );
350 # always call this after unstrip() to preserve the order
351 function unstripNoWiki( $text, &$state ) {
352 # Must expand in reverse order, otherwise nested tags will be corrupted
353 for ( $content = end($state['nowiki']); $content !== false; $content = prev( $state['nowiki'] ) ) {
354 $text = str_replace( key( $state['nowiki'] ), $content, $text );
359 for ( $content = end($state['html']); $content !== false; $content = prev( $state['html'] ) ) {
360 $text = str_replace( key( $state['html'] ), $content, $text );
367 # Add an item to the strip state
368 # Returns the unique tag which must be inserted into the stripped text
369 # The tag will be replaced with the original text in unstrip()
370 function insertStripItem( $text, &$state ) {
371 $rnd = UNIQ_PREFIX
. '-item' . Parser
::getRandomString();
380 $state['item'][$rnd] = $text;
384 # Return allowed HTML attributes
385 function getHTMLattrs () {
386 $htmlattrs = array( # Allowed attributes--no scripting, etc.
387 'title', 'align', 'lang', 'dir', 'width', 'height',
388 'bgcolor', 'clear', /* BR */ 'noshade', /* HR */
389 'cite', /* BLOCKQUOTE, Q */ 'size', 'face', 'color',
390 /* FONT */ 'type', 'start', 'value', 'compact',
391 /* For various lists, mostly deprecated but safe */
392 'summary', 'width', 'border', 'frame', 'rules',
393 'cellspacing', 'cellpadding', 'valign', 'char',
394 'charoff', 'colgroup', 'col', 'span', 'abbr', 'axis',
395 'headers', 'scope', 'rowspan', 'colspan', /* Tables */
396 'id', 'class', 'name', 'style' /* For CSS */
401 # Remove non approved attributes and javascript in css
402 function fixTagAttributes ( $t ) {
403 if ( trim ( $t ) == '' ) return '' ; # Saves runtime ;-)
404 $htmlattrs = $this->getHTMLattrs() ;
406 # Strip non-approved attributes from the tag
408 '/(\\w+)(\\s*=\\s*([^\\s\">]+|\"[^\">]*\"))?/e',
409 "(in_array(strtolower(\"\$1\"),\$htmlattrs)?(\"\$1\".((\"x\$3\" != \"x\")?\"=\$3\":'')):'')",
412 $t = str_replace ( '<></>' , '' , $t ) ; # This should fix bug 980557
414 # Strip javascript "expression" from stylesheets. Brute force approach:
415 # If anythin offensive is found, all attributes of the HTML tag are dropped
418 '/style\\s*=.*(expression|tps*:\/\/|url\\s*\().*/is',
419 wfMungeToUtf8( $t ) ) )
427 # interface with html tidy, used if $wgUseTidy = true
428 function tidy ( $text ) {
429 global $wgTidyConf, $wgTidyBin, $wgTidyOpts;
430 global $wgInputEncoding, $wgOutputEncoding;
431 $fname = 'Parser::tidy';
432 wfProfileIn( $fname );
435 switch(strtoupper($wgOutputEncoding)) {
437 $wgTidyOpts .= ($wgInputEncoding == $wgOutputEncoding)?
' -latin1':' -raw';
440 $wgTidyOpts .= ($wgInputEncoding == $wgOutputEncoding)?
' -utf8':' -raw';
443 $wgTidyOpts .= ' -raw';
446 $wrappedtext = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'.
447 ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>'.
448 '<head><title>test</title></head><body>'.$text.'</body></html>';
449 $descriptorspec = array(
450 0 => array('pipe', 'r'),
451 1 => array('pipe', 'w'),
452 2 => array('file', '/dev/null', 'a')
454 $process = proc_open("$wgTidyBin -config $wgTidyConf $wgTidyOpts", $descriptorspec, $pipes);
455 if (is_resource($process)) {
456 fwrite($pipes[0], $wrappedtext);
458 while (!feof($pipes[1])) {
459 $cleansource .= fgets($pipes[1], 1024);
462 $return_value = proc_close($process);
465 wfProfileOut( $fname );
467 if( $cleansource == '' && $text != '') {
468 wfDebug( "Tidy error detected!\n" );
469 return $text . "\n<!-- Tidy found serious XHTML errors -->\n";
475 # parse the wiki syntax used to render tables
476 function doTableStuff ( $t ) {
477 $fname = 'Parser::doTableStuff';
478 wfProfileIn( $fname );
480 $t = explode ( "\n" , $t ) ;
481 $td = array () ; # Is currently a td tag open?
482 $ltd = array () ; # Was it TD or TH?
483 $tr = array () ; # Is currently a tr tag open?
484 $ltr = array () ; # tr attributes
485 $indent_level = 0; # indent level of the table
486 foreach ( $t AS $k => $x )
489 $fc = substr ( $x , 0 , 1 ) ;
490 if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
491 $indent_level = strlen( $matches[1] );
493 str_repeat( '<dl><dd>', $indent_level ) .
494 '<table ' . $this->fixTagAttributes ( $matches[2] ) . '>' ;
495 array_push ( $td , false ) ;
496 array_push ( $ltd , '' ) ;
497 array_push ( $tr , false ) ;
498 array_push ( $ltr , '' ) ;
500 else if ( count ( $td ) == 0 ) { } # Don't do any of the following
501 else if ( '|}' == substr ( $x , 0 , 2 ) ) {
503 $l = array_pop ( $ltd ) ;
504 if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
505 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
507 $t[$k] = $z . str_repeat( '</dd></dl>', $indent_level );
509 else if ( '|-' == substr ( $x , 0 , 2 ) ) { # Allows for |---------------
510 $x = substr ( $x , 1 ) ;
511 while ( $x != '' && substr ( $x , 0 , 1 ) == '-' ) $x = substr ( $x , 1 ) ;
513 $l = array_pop ( $ltd ) ;
514 if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
515 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
518 array_push ( $tr , false ) ;
519 array_push ( $td , false ) ;
520 array_push ( $ltd , '' ) ;
521 array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
523 else if ( '|' == $fc ||
'!' == $fc ||
'|+' == substr ( $x , 0 , 2 ) ) { # Caption
524 if ( '|+' == substr ( $x , 0 , 2 ) ) {
526 $x = substr ( $x , 1 ) ;
528 $after = substr ( $x , 1 ) ;
529 if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;
530 $after = explode ( '||' , $after ) ;
532 foreach ( $after AS $theline )
537 $tra = array_pop ( $ltr ) ;
538 if ( !array_pop ( $tr ) ) $z = '<tr '.$tra.">\n" ;
539 array_push ( $tr , true ) ;
540 array_push ( $ltr , '' ) ;
543 $l = array_pop ( $ltd ) ;
544 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
545 if ( $fc == '|' ) $l = 'td' ;
546 else if ( $fc == '!' ) $l = 'th' ;
547 else if ( $fc == '+' ) $l = 'caption' ;
549 array_push ( $ltd , $l ) ;
550 $y = explode ( '|' , $theline , 2 ) ;
551 if ( count ( $y ) == 1 ) $y = "{$z}<{$l}>{$y[0]}" ;
552 else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
554 array_push ( $td , true ) ;
559 # Closing open td, tr && table
560 while ( count ( $td ) > 0 )
562 if ( array_pop ( $td ) ) $t[] = '</td>' ;
563 if ( array_pop ( $tr ) ) $t[] = '</tr>' ;
567 $t = implode ( "\n" , $t ) ;
568 # $t = $this->removeHTMLtags( $t );
569 wfProfileOut( $fname );
573 function internalParse( $text, $linestart, $args = array(), $isMain=true ) {
576 $fname = 'Parser::internalParse';
577 wfProfileIn( $fname );
579 $text = $this->removeHTMLtags( $text );
580 $text = $this->replaceVariables( $text, $args );
582 $text = $wgLang->convert($text);
584 $text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
586 $text = $this->doHeadings( $text );
587 if($this->mOptions
->getUseDynamicDates()) {
588 global $wgDateFormatter;
589 $text = $wgDateFormatter->reformat( $this->mOptions
->getDateFormat(), $text );
591 $text = $this->doAllQuotes( $text );
592 $text = $this->replaceExternalLinks( $text );
593 $text = $this->doMagicLinks( $text );
594 $text = $this->replaceInternalLinks ( $text );
595 $text = $this->replaceInternalLinks ( $text );
597 $text = $this->unstrip( $text, $this->mStripState
);
598 $text = $this->unstripNoWiki( $text, $this->mStripState
);
600 $text = $this->doTableStuff( $text );
601 $text = $this->formatHeadings( $text, $isMain );
602 $sk =& $this->mOptions
->getSkin();
603 $text = $sk->transformContent( $text );
605 wfProfileOut( $fname );
609 /* private */ function &doMagicLinks( &$text ) {
610 global $wgUseGeoMode;
611 $text = $this->magicISBN( $text );
612 if ( isset( $wgUseGeoMode ) && $wgUseGeoMode ) {
613 $text = $this->magicGEO( $text );
615 $text = $this->magicRFC( $text );
619 # Parse ^^ tokens and return html
620 /* private */ function doExponent ( $text ) {
621 $fname = 'Parser::doExponent';
622 wfProfileIn( $fname);
623 $text = preg_replace('/\^\^(.*)\^\^/','<small><sup>\\1</sup></small>', $text);
624 wfProfileOut( $fname);
628 # Parse headers and return html
629 /* private */ function doHeadings( $text ) {
630 $fname = 'Parser::doHeadings';
631 wfProfileIn( $fname );
632 for ( $i = 6; $i >= 1; --$i ) {
633 $h = substr( '======', 0, $i );
634 $text = preg_replace( "/^{$h}(.+){$h}(\\s|$)/m",
635 "<h{$i}>\\1</h{$i}>\\2", $text );
637 wfProfileOut( $fname );
641 /* private */ function doAllQuotes( $text ) {
642 $fname = 'Parser::doAllQuotes';
643 wfProfileIn( $fname );
645 $lines = explode( "\n", $text );
646 foreach ( $lines as $line ) {
647 $outtext .= $this->doQuotes ( $line ) . "\n";
649 $outtext = substr($outtext, 0,-1);
650 wfProfileOut( $fname );
654 /* private */ function doQuotes( $text ) {
655 $arr = preg_split ("/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
656 if (count ($arr) == 1)
660 # First, do some preliminary work. This may shift some apostrophes from
661 # being mark-up to being text. It also counts the number of occurrences
662 # of bold and italics mark-ups.
670 # If there are ever four apostrophes, assume the first is supposed to
671 # be text, and the remaining three constitute mark-up for bold text.
672 if (strlen ($arr[$i]) == 4)
677 # If there are more than 5 apostrophes in a row, assume they're all
678 # text except for the last 5.
679 else if (strlen ($arr[$i]) > 5)
681 $arr[$i-1] .= str_repeat ("'", strlen ($arr[$i]) - 5);
684 # Count the number of occurrences of bold and italics mark-ups.
685 # We are not counting sequences of five apostrophes.
686 if (strlen ($arr[$i]) == 2) $numitalics++
; else
687 if (strlen ($arr[$i]) == 3) $numbold++
; else
688 if (strlen ($arr[$i]) == 5) { $numitalics++
; $numbold++
; }
693 # If there is an odd number of both bold and italics, it is likely
694 # that one of the bold ones was meant to be an apostrophe followed
695 # by italics. Which one we cannot know for certain, but it is more
696 # likely to be one that has a single-letter word before it.
697 if (($numbold %
2 == 1) && ($numitalics %
2 == 1))
700 $firstsingleletterword = -1;
701 $firstmultiletterword = -1;
705 if (($i %
2 == 1) and (strlen ($r) == 3))
707 $x1 = substr ($arr[$i-1], -1);
708 $x2 = substr ($arr[$i-1], -2, 1);
710 if ($firstspace == -1) $firstspace = $i;
711 } else if ($x2 == ' ') {
712 if ($firstsingleletterword == -1) $firstsingleletterword = $i;
714 if ($firstmultiletterword == -1) $firstmultiletterword = $i;
720 # If there is a single-letter word, use it!
721 if ($firstsingleletterword > -1)
723 $arr [ $firstsingleletterword ] = "''";
724 $arr [ $firstsingleletterword-1 ] .= "'";
726 # If not, but there's a multi-letter word, use that one.
727 else if ($firstmultiletterword > -1)
729 $arr [ $firstmultiletterword ] = "''";
730 $arr [ $firstmultiletterword-1 ] .= "'";
732 # ... otherwise use the first one that has neither.
733 # (notice that it is possible for all three to be -1 if, for example,
734 # there is only one pentuple-apostrophe in the line)
735 else if ($firstspace > -1)
737 $arr [ $firstspace ] = "''";
738 $arr [ $firstspace-1 ] .= "'";
742 # Now let's actually convert our apostrophic mush to HTML!
751 if ($state == 'both')
758 if (strlen ($r) == 2)
761 { $output .= '</i>'; $state = ''; }
762 else if ($state == 'bi')
763 { $output .= '</i>'; $state = 'b'; }
764 else if ($state == 'ib')
765 { $output .= '</b></i><b>'; $state = 'b'; }
766 else if ($state == 'both')
767 { $output .= '<b><i>'.$buffer.'</i>'; $state = 'b'; }
768 else # $state can be 'b' or ''
769 { $output .= '<i>'; $state .= 'i'; }
771 else if (strlen ($r) == 3)
774 { $output .= '</b>'; $state = ''; }
775 else if ($state == 'bi')
776 { $output .= '</i></b><i>'; $state = 'i'; }
777 else if ($state == 'ib')
778 { $output .= '</b>'; $state = 'i'; }
779 else if ($state == 'both')
780 { $output .= '<i><b>'.$buffer.'</b>'; $state = 'i'; }
781 else # $state can be 'i' or ''
782 { $output .= '<b>'; $state .= 'b'; }
784 else if (strlen ($r) == 5)
787 { $output .= '</b><i>'; $state = 'i'; }
788 else if ($state == 'i')
789 { $output .= '</i><b>'; $state = 'b'; }
790 else if ($state == 'bi')
791 { $output .= '</i></b>'; $state = ''; }
792 else if ($state == 'ib')
793 { $output .= '</b></i>'; $state = ''; }
794 else if ($state == 'both')
795 { $output .= '<i><b>'.$buffer.'</b></i>'; $state = ''; }
796 else # ($state == '')
797 { $buffer = ''; $state = 'both'; }
802 # Now close all remaining tags. Notice that the order is important.
803 if ($state == 'b' ||
$state == 'ib')
805 if ($state == 'i' ||
$state == 'bi' ||
$state == 'ib')
809 if ($state == 'both')
810 $output .= '<b><i>'.$buffer.'</i></b>';
815 # Note: we have to do external links before the internal ones,
816 # and otherwise take great care in the order of things here, so
817 # that we don't end up interpreting some URLs twice.
819 /* private */ function replaceExternalLinks( $text ) {
820 $fname = 'Parser::replaceExternalLinks';
821 wfProfileIn( $fname );
823 $sk =& $this->mOptions
->getSkin();
824 $linktrail = wfMsg('linktrail');
825 $bits = preg_split( EXT_LINK_BRACKETED
, $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
827 $s = $this->replaceFreeExternalLinks( array_shift( $bits ) );
830 while ( $i<count( $bits ) ) {
832 $protocol = $bits[$i++
];
834 $trail = $bits[$i++
];
836 # If the link text is an image URL, replace it with an <img> tag
837 # This happened by accident in the original parser, but some people used it extensively
838 $img = $this->maybeMakeImageLink( $text );
839 if ( $img !== false ) {
845 # No link text, e.g. [http://domain.tld/some.link]
847 # Autonumber if allowed
848 if ( strpos( HTTP_PROTOCOLS
, $protocol ) !== false ) {
849 $text = '[' . ++
$this->mAutonumber
. ']';
851 # Otherwise just use the URL
852 $text = htmlspecialchars( $url );
855 # Have link text, e.g. [http://domain.tld/some.link text]s
857 if ( preg_match( $linktrail, $trail, $m2 ) ) {
863 $encUrl = htmlspecialchars( $url );
864 # Bit in parentheses showing the URL for the printable version
865 if( $url == $text ||
preg_match( "!$protocol://" . preg_quote( $text, '/' ) . "/?$!", $url ) ) {
868 # Expand the URL for printable version
869 if ( ! $sk->suppressUrlExpansion() ) {
870 $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $encUrl ) . "</i>)</span>";
876 # Process the trail (i.e. everything after this link up until start of the next link),
877 # replacing any non-bracketed links
878 $trail = $this->replaceFreeExternalLinks( $trail );
880 $la = $sk->getExternalLinkAttributes( $url, $text );
882 # Use the encoded URL
883 # This means that users can paste URLs directly into the text
884 # Funny characters like ö aren't valid in URLs anyway
885 # This was changed in August 2004
886 $s .= "<a href=\"{$url}\" {$la}>{$text}</a>{$dtrail}{$paren}{$trail}";
889 wfProfileOut( $fname );
893 # Replace anything that looks like a URL with a link
894 function replaceFreeExternalLinks( $text ) {
895 $bits = preg_split( '/((?:'.URL_PROTOCOLS
.'):)/', $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
896 $s = array_shift( $bits );
899 $sk =& $this->mOptions
->getSkin();
901 while ( $i < count( $bits ) ){
902 $protocol = $bits[$i++
];
903 $remainder = $bits[$i++
];
905 if ( preg_match( '/^('.EXT_LINK_URL_CLASS
.'+)(.*)$/s', $remainder, $m ) ) {
906 # Found some characters after the protocol that look promising
907 $url = $protocol . $m[1];
910 # Move trailing punctuation to $trail
912 # If there is no left bracket, then consider right brackets fair game too
913 if ( strpos( $url, '(' ) === false ) {
917 $numSepChars = strspn( strrev( $url ), $sep );
918 if ( $numSepChars ) {
919 $trail = substr( $url, -$numSepChars ) . $trail;
920 $url = substr( $url, 0, -$numSepChars );
923 # Replace & from obsolete syntax with &
924 $url = str_replace( '&', '&', $url );
926 # Is this an external image?
927 $text = $this->maybeMakeImageLink( $url );
928 if ( $text === false ) {
929 # Not an image, make a link
930 $text = $sk->makeExternalLink( $url, $url );
932 $s .= $text . $trail;
934 $s .= $protocol . $remainder;
940 # make an image if it's allowed
941 function maybeMakeImageLink( $url ) {
942 $sk =& $this->mOptions
->getSkin();
944 if ( $this->mOptions
->getAllowExternalImages() ) {
945 if ( preg_match( EXT_IMAGE_REGEX
, $url ) ) {
947 $text = $sk->makeImage( htmlspecialchars( $url ) );
953 # The wikilinks [[ ]] are procedeed here.
954 /* private */ function replaceInternalLinks( $s ) {
955 global $wgLang, $wgLinkCache;
956 global $wgNamespacesWithSubpages;
957 static $fname = 'Parser::replaceInternalLinks' ;
958 wfProfileIn( $fname );
960 wfProfileIn( $fname.'-setup' );
962 # the % is needed to support urlencoded titles as well
963 if ( !$tc ) { $tc = Title
::legalChars() . '#%'; }
964 $sk =& $this->mOptions
->getSkin();
966 $redirect = MagicWord
::get ( MAG_REDIRECT
) ;
968 $a = explode( '[[', ' ' . $s );
969 $s = array_shift( $a );
970 $s = substr( $s, 1 );
972 # Match a link having the form [[namespace:link|alternate]]trail
974 if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|([^]]+))?]](.*)\$/sD"; }
975 # Match the end of a line for a word that's not followed by whitespace,
976 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
977 static $e2 = '/^(.*?)([a-zA-Z\x80-\xff]+)$/sD';
979 $useLinkPrefixExtension = $wgLang->linkPrefixExtension();
980 # Special and Media are pseudo-namespaces; no pages actually exist in them
982 $nottalk = !Namespace::isTalk( $this->mTitle
->getNamespace() );
984 if ( $useLinkPrefixExtension ) {
985 if ( preg_match( $e2, $s, $m ) ) {
986 $first_prefix = $m[2];
989 $first_prefix = false;
995 wfProfileOut( $fname.'-setup' );
997 # start procedeeding each line
998 foreach ( $a as $line ) {
999 wfProfileIn( $fname.'-prefixhandling' );
1000 if ( $useLinkPrefixExtension ) {
1001 if ( preg_match( $e2, $s, $m ) ) {
1009 $prefix = $first_prefix;
1010 $first_prefix = false;
1013 wfProfileOut( $fname.'-prefixhandling' );
1015 if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
1017 # fix up urlencoded title texts
1018 if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
1020 } else { # Invalid form; output directly
1021 $s .= $prefix . '[[' . $line ;
1027 # :Foobar -- override special treatment of prefix (images, language links)
1028 # /Foobar -- convert to CurrentPage/Foobar
1029 # /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text
1031 # Look at the first character
1032 $c = substr($m[1],0,1);
1033 $noforce = ($c != ':');
1037 # / at end means we don't want the slash to be shown
1038 if(substr($m[1],-1,1)=='/') {
1039 $m[1]=substr($m[1],1,strlen($m[1])-2);
1042 $noslash=substr($m[1],1);
1045 # Some namespaces don't allow subpages
1046 if(!empty($wgNamespacesWithSubpages[$this->mTitle
->getNamespace()])) {
1047 # subpages allowed here
1048 $link = $this->mTitle
->getPrefixedText(). '/' . trim($noslash);
1051 } # this might be changed for ugliness reasons
1053 # no subpage allowed, use standard link
1057 } elseif( $noforce ) { # no subpage
1060 # We don't want to keep the first character
1061 $link = substr( $m[1], 1 );
1064 $wasblank = ( '' == $text );
1065 if( $wasblank ) $text = $link;
1067 $nt = Title
::newFromText( $link );
1069 $s .= $prefix . '[[' . $line;
1073 $ns = $nt->getNamespace();
1074 $iw = $nt->getInterWiki();
1076 # Link not escaped by : , create the various objects
1080 if( $iw && $this->mOptions
->getInterwikiMagic() && $nottalk && $wgLang->getLanguageName( $iw ) ) {
1081 array_push( $this->mOutput
->mLanguageLinks
, $nt->getFullText() );
1082 $tmp = $prefix . $trail ;
1083 $s .= (trim($tmp) == '')?
'': $tmp;
1087 if ( $ns == NS_IMAGE
) {
1088 $s .= $prefix . $sk->makeImageLinkObj( $nt, $text ) . $trail;
1089 $wgLinkCache->addImageLinkObj( $nt );
1093 if ( $ns == NS_CATEGORY
) {
1094 $t = $nt->getText() ;
1095 $nnt = Title
::newFromText ( Namespace::getCanonicalName(NS_CATEGORY
).':'.$t ) ;
1097 $wgLinkCache->suspend(); # Don't save in links/brokenlinks
1098 $pPLC=$sk->postParseLinkColour();
1099 $sk->postParseLinkColour( false );
1100 $t = $sk->makeLinkObj( $nnt, $t, '', '' , $prefix );
1101 $sk->postParseLinkColour( $pPLC );
1102 $wgLinkCache->resume();
1105 if ( $this->mTitle
->getNamespace() == NS_CATEGORY
) {
1106 $sortkey = $this->mTitle
->getText();
1108 $sortkey = $this->mTitle
->getPrefixedText();
1113 $wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
1114 $this->mOutput
->mCategoryLinks
[] = $t ;
1115 $s .= $prefix . $trail ;
1120 if( ( $nt->getPrefixedText() === $this->mTitle
->getPrefixedText() ) &&
1121 ( strpos( $link, '#' ) === FALSE ) ) {
1122 # Self-links are handled specially; generally de-link and change to bold.
1123 $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail );
1127 if( $ns == NS_MEDIA
) {
1128 $s .= $prefix . $sk->makeMediaLinkObj( $nt, $text ) . $trail;
1129 $wgLinkCache->addImageLinkObj( $nt );
1131 } elseif( $ns == NS_SPECIAL
) {
1132 $s .= $prefix . $sk->makeKnownLinkObj( $nt, $text, '', $trail );
1135 $s .= $sk->makeLinkObj( $nt, $text, '', $trail, $prefix );
1137 wfProfileOut( $fname );
1141 # Some functions here used by doBlockLevels()
1143 /* private */ function closeParagraph() {
1145 if ( '' != $this->mLastSection
) {
1146 $result = '</' . $this->mLastSection
. ">\n";
1148 $this->mInPre
= false;
1149 $this->mLastSection
= '';
1152 # getCommon() returns the length of the longest common substring
1153 # of both arguments, starting at the beginning of both.
1155 /* private */ function getCommon( $st1, $st2 ) {
1156 $fl = strlen( $st1 );
1157 $shorter = strlen( $st2 );
1158 if ( $fl < $shorter ) { $shorter = $fl; }
1160 for ( $i = 0; $i < $shorter; ++
$i ) {
1161 if ( $st1{$i} != $st2{$i} ) { break; }
1165 # These next three functions open, continue, and close the list
1166 # element appropriate to the prefix character passed into them.
1168 /* private */ function openList( $char ) {
1169 $result = $this->closeParagraph();
1171 if ( '*' == $char ) { $result .= '<ul><li>'; }
1172 else if ( '#' == $char ) { $result .= '<ol><li>'; }
1173 else if ( ':' == $char ) { $result .= '<dl><dd>'; }
1174 else if ( ';' == $char ) {
1175 $result .= '<dl><dt>';
1176 $this->mDTopen
= true;
1178 else { $result = '<!-- ERR 1 -->'; }
1183 /* private */ function nextItem( $char ) {
1184 if ( '*' == $char ||
'#' == $char ) { return '</li><li>'; }
1185 else if ( ':' == $char ||
';' == $char ) {
1187 if ( $this->mDTopen
) { $close = '</dt>'; }
1188 if ( ';' == $char ) {
1189 $this->mDTopen
= true;
1190 return $close . '<dt>';
1192 $this->mDTopen
= false;
1193 return $close . '<dd>';
1196 return '<!-- ERR 2 -->';
1199 /* private */ function closeList( $char ) {
1200 if ( '*' == $char ) { $text = '</li></ul>'; }
1201 else if ( '#' == $char ) { $text = '</li></ol>'; }
1202 else if ( ':' == $char ) {
1203 if ( $this->mDTopen
) {
1204 $this->mDTopen
= false;
1205 $text = '</dt></dl>';
1207 $text = '</dd></dl>';
1210 else { return '<!-- ERR 3 -->'; }
1214 /* private */ function doBlockLevels( $text, $linestart ) {
1215 $fname = 'Parser::doBlockLevels';
1216 wfProfileIn( $fname );
1218 # Parsing through the text line by line. The main thing
1219 # happening here is handling of block-level elements p, pre,
1220 # and making lists from lines starting with * # : etc.
1222 $textLines = explode( "\n", $text );
1224 $lastPrefix = $output = $lastLine = '';
1225 $this->mDTopen
= $inBlockElem = false;
1227 $paragraphStack = false;
1229 if ( !$linestart ) {
1230 $output .= array_shift( $textLines );
1232 foreach ( $textLines as $oLine ) {
1233 $lastPrefixLength = strlen( $lastPrefix );
1234 $preCloseMatch = preg_match('/<\\/pre/i', $oLine );
1235 $preOpenMatch = preg_match('/<pre/i', $oLine );
1236 if ( !$this->mInPre
) {
1237 # Multiple prefixes may abut each other for nested lists.
1238 $prefixLength = strspn( $oLine, '*#:;' );
1239 $pref = substr( $oLine, 0, $prefixLength );
1242 $pref2 = str_replace( ';', ':', $pref );
1243 $t = substr( $oLine, $prefixLength );
1244 $this->mInPre
= !empty($preOpenMatch);
1246 # Don't interpret any other prefixes in preformatted text
1248 $pref = $pref2 = '';
1253 if( $prefixLength && 0 == strcmp( $lastPrefix, $pref2 ) ) {
1254 # Same as the last item, so no need to deal with nesting or opening stuff
1255 $output .= $this->nextItem( substr( $pref, -1 ) );
1256 $paragraphStack = false;
1258 if ( substr( $pref, -1 ) == ';') {
1259 # The one nasty exception: definition lists work like this:
1260 # ; title : definition text
1261 # So we check for : in the remainder text to split up the
1262 # title and definition, without b0rking links.
1263 # FIXME: This is not foolproof. Something better in Tokenizer might help.
1264 if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
1266 $output .= $term . $this->nextItem( ':' );
1270 } elseif( $prefixLength ||
$lastPrefixLength ) {
1271 # Either open or close a level...
1272 $commonPrefixLength = $this->getCommon( $pref, $lastPrefix );
1273 $paragraphStack = false;
1275 while( $commonPrefixLength < $lastPrefixLength ) {
1276 $output .= $this->closeList( $lastPrefix{$lastPrefixLength-1} );
1277 --$lastPrefixLength;
1279 if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) {
1280 $output .= $this->nextItem( $pref{$commonPrefixLength-1} );
1282 while ( $prefixLength > $commonPrefixLength ) {
1283 $char = substr( $pref, $commonPrefixLength, 1 );
1284 $output .= $this->openList( $char );
1286 if ( ';' == $char ) {
1287 # FIXME: This is dupe of code above
1288 if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
1290 $output .= $term . $this->nextItem( ':' );
1294 ++
$commonPrefixLength;
1296 $lastPrefix = $pref2;
1298 if( 0 == $prefixLength ) {
1299 # No prefix (not in list)--go to paragraph mode
1300 $uniq_prefix = UNIQ_PREFIX
;
1301 // XXX: use a stack for nestable elements like span, table and div
1302 $openmatch = preg_match('/(<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<li|<\\/tr|<\\/td|<\\/th)/i', $t );
1303 $closematch = preg_match(
1304 '/(<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|'.
1305 '<td|<th|<div|<\\/div|<hr|<\\/pre|<\\/p|'.$uniq_prefix.'-pre|<\\/li|<\\/ul)/i', $t );
1306 if ( $openmatch or $closematch ) {
1307 $paragraphStack = false;
1308 $output .= $this->closeParagraph();
1309 if($preOpenMatch and !$preCloseMatch) {
1310 $this->mInPre
= true;
1312 if ( $closematch ) {
1313 $inBlockElem = false;
1315 $inBlockElem = true;
1317 } else if ( !$inBlockElem && !$this->mInPre
) {
1318 if ( ' ' == $t{0} and ( $this->mLastSection
== 'pre' or trim($t) != '' ) ) {
1320 if ($this->mLastSection
!= 'pre') {
1321 $paragraphStack = false;
1322 $output .= $this->closeParagraph().'<pre>';
1323 $this->mLastSection
= 'pre';
1325 $t = substr( $t, 1 );
1328 if ( '' == trim($t) ) {
1329 if ( $paragraphStack ) {
1330 $output .= $paragraphStack.'<br />';
1331 $paragraphStack = false;
1332 $this->mLastSection
= 'p';
1334 if ($this->mLastSection
!= 'p' ) {
1335 $output .= $this->closeParagraph();
1336 $this->mLastSection
= '';
1337 $paragraphStack = '<p>';
1339 $paragraphStack = '</p><p>';
1343 if ( $paragraphStack ) {
1344 $output .= $paragraphStack;
1345 $paragraphStack = false;
1346 $this->mLastSection
= 'p';
1347 } else if ($this->mLastSection
!= 'p') {
1348 $output .= $this->closeParagraph().'<p>';
1349 $this->mLastSection
= 'p';
1355 if ($paragraphStack === false) {
1359 while ( $prefixLength ) {
1360 $output .= $this->closeList( $pref2{$prefixLength-1} );
1363 if ( '' != $this->mLastSection
) {
1364 $output .= '</' . $this->mLastSection
. '>';
1365 $this->mLastSection
= '';
1368 wfProfileOut( $fname );
1372 # Return value of a magic variable (like PAGENAME)
1373 function getVariableValue( $index ) {
1374 global $wgLang, $wgSitename, $wgServer;
1377 case MAG_CURRENTMONTH
:
1378 return $wgLang->formatNum( date( 'm' ) );
1379 case MAG_CURRENTMONTHNAME
:
1380 return $wgLang->getMonthName( date('n') );
1381 case MAG_CURRENTMONTHNAMEGEN
:
1382 return $wgLang->getMonthNameGen( date('n') );
1383 case MAG_CURRENTDAY
:
1384 return $wgLang->formatNum( date('j') );
1386 return $this->mTitle
->getText();
1388 return $this->mTitle
->getPartialURL();
1390 # return Namespace::getCanonicalName($this->mTitle->getNamespace());
1391 return $wgLang->getNsText($this->mTitle
->getNamespace()); # Patch by Dori
1392 case MAG_CURRENTDAYNAME
:
1393 return $wgLang->getWeekdayName( date('w')+
1 );
1394 case MAG_CURRENTYEAR
:
1395 return $wgLang->formatNum( date( 'Y' ) );
1396 case MAG_CURRENTTIME
:
1397 return $wgLang->time( wfTimestampNow(), false );
1398 case MAG_NUMBEROFARTICLES
:
1399 return $wgLang->formatNum( wfNumberOfArticles() );
1409 # initialise the magic variables (like CURRENTMONTHNAME)
1410 function initialiseVariables() {
1411 global $wgVariableIDs;
1412 $this->mVariables
= array();
1413 foreach ( $wgVariableIDs as $id ) {
1414 $mw =& MagicWord
::get( $id );
1415 $mw->addToArray( $this->mVariables
, $this->getVariableValue( $id ) );
1419 /* private */ function replaceVariables( $text, $args = array() ) {
1420 global $wgLang, $wgScript, $wgArticlePath;
1422 # Prevent too big inclusions
1423 if(strlen($text)> MAX_INCLUDE_SIZE
)
1426 $fname = 'Parser::replaceVariables';
1427 wfProfileIn( $fname );
1430 $titleChars = Title
::legalChars();
1431 $nonBraceChars = str_replace( array( '{', '}' ), array( '', '' ), $titleChars );
1433 # This function is called recursively. To keep track of arguments we need a stack:
1434 array_push( $this->mArgStack
, $args );
1436 # PHP global rebinding syntax is a bit weird, need to use the GLOBALS array
1437 $GLOBALS['wgCurParser'] =& $this;
1439 if ( $this->mOutputType
== OT_HTML ||
$this->mOutputType
== OT_MSG
) {
1440 # Variable substitution
1441 $text = preg_replace_callback( "/{{([$nonBraceChars]*?)}}/", 'wfVariableSubstitution', $text );
1444 if ( $this->mOutputType
== OT_HTML
) {
1445 # Argument substitution
1446 $text = preg_replace_callback( "/(\\n?){{{([$titleChars]*?)}}}/", 'wfArgSubstitution', $text );
1448 # Template substitution
1449 $regex = '/(\\n?){{(['.$nonBraceChars.']*)(\\|.*?|)}}/s';
1450 $text = preg_replace_callback( $regex, 'wfBraceSubstitution', $text );
1452 array_pop( $this->mArgStack
);
1454 wfProfileOut( $fname );
1458 function variableSubstitution( $matches ) {
1459 if ( !$this->mVariables
) {
1460 $this->initialiseVariables();
1462 if ( array_key_exists( $matches[1], $this->mVariables
) ) {
1463 $text = $this->mVariables
[$matches[1]];
1464 $this->mOutput
->mContainsOldMagic
= true;
1466 $text = $matches[0];
1471 # Split template arguments
1472 function getTemplateArgs( $argsString ) {
1473 if ( $argsString === '' ) {
1477 $args = explode( '|', substr( $argsString, 1 ) );
1479 # If any of the arguments contains a '[[' but no ']]', it needs to be
1480 # merged with the next arg because the '|' character between belongs
1481 # to the link syntax and not the template parameter syntax.
1482 $argc = count($args);
1484 for ( $i = 0; $i < $argc-1; $i++
) {
1485 if ( substr_count ( $args[$i], '[[' ) != substr_count ( $args[$i], ']]' ) ) {
1486 $args[$i] .= '|'.$args[$i+
1];
1487 array_splice($args, $i+
1, 1);
1496 function braceSubstitution( $matches ) {
1497 global $wgLinkCache, $wgLang;
1498 $fname = 'Parser::braceSubstitution';
1505 # $newline is an optional newline character before the braces
1506 # $part1 is the bit before the first |, and must contain only title characters
1507 # $args is a list of arguments, starting from index 0, not including $part1
1509 $newline = $matches[1];
1510 $part1 = $matches[2];
1511 # If the third subpattern matched anything, it will start with |
1513 $args = $this->getTemplateArgs($matches[3]);
1514 $argc = count( $args );
1517 if ( strpos( $matches[0], '{{{' ) !== false ) {
1518 $text = $matches[0];
1525 $mwSubst =& MagicWord
::get( MAG_SUBST
);
1526 if ( $mwSubst->matchStartAndRemove( $part1 ) ) {
1527 if ( $this->mOutputType
!= OT_WIKI
) {
1528 # Invalid SUBST not replaced at PST time
1529 # Return without further processing
1530 $text = $matches[0];
1534 } elseif ( $this->mOutputType
== OT_WIKI
) {
1535 # SUBST not found in PST pass, do nothing
1536 $text = $matches[0];
1541 # MSG, MSGNW and INT
1544 $mwMsgnw =& MagicWord
::get( MAG_MSGNW
);
1545 if ( $mwMsgnw->matchStartAndRemove( $part1 ) ) {
1548 # Remove obsolete MSG:
1549 $mwMsg =& MagicWord
::get( MAG_MSG
);
1550 $mwMsg->matchStartAndRemove( $part1 );
1553 # Check if it is an internal message
1554 $mwInt =& MagicWord
::get( MAG_INT
);
1555 if ( $mwInt->matchStartAndRemove( $part1 ) ) {
1556 if ( $this->incrementIncludeCount( 'int:'.$part1 ) ) {
1557 $text = wfMsgReal( $part1, $args, true );
1565 # Check for NS: (namespace expansion)
1566 $mwNs = MagicWord
::get( MAG_NS
);
1567 if ( $mwNs->matchStartAndRemove( $part1 ) ) {
1568 if ( intval( $part1 ) ) {
1569 $text = $wgLang->getNsText( intval( $part1 ) );
1572 $index = Namespace::getCanonicalIndex( strtolower( $part1 ) );
1573 if ( !is_null( $index ) ) {
1574 $text = $wgLang->getNsText( $index );
1581 # LOCALURL and LOCALURLE
1583 $mwLocal = MagicWord
::get( MAG_LOCALURL
);
1584 $mwLocalE = MagicWord
::get( MAG_LOCALURLE
);
1586 if ( $mwLocal->matchStartAndRemove( $part1 ) ) {
1587 $func = 'getLocalURL';
1588 } elseif ( $mwLocalE->matchStartAndRemove( $part1 ) ) {
1589 $func = 'escapeLocalURL';
1594 if ( $func !== '' ) {
1595 $title = Title
::newFromText( $part1 );
1596 if ( !is_null( $title ) ) {
1598 $text = $title->$func( $args[0] );
1600 $text = $title->$func();
1607 # Internal variables
1608 if ( !$this->mVariables
) {
1609 $this->initialiseVariables();
1611 if ( !$found && array_key_exists( $part1, $this->mVariables
) ) {
1612 $text = $this->mVariables
[$part1];
1614 $this->mOutput
->mContainsOldMagic
= true;
1618 if ( !$found && $argc == 1 ) {
1619 $mwGrammar =& MagicWord
::get( MAG_GRAMMAR
);
1620 if ( $mwGrammar->matchStartAndRemove( $part1 ) ) {
1621 $text = $wgLang->convertGrammar( $args[0], $part1 );
1626 # Template table test
1628 # Did we encounter this template already? If yes, it is in the cache
1629 # and we need to check for loops.
1630 if ( isset( $this->mTemplates
[$part1] ) ) {
1631 # Infinite loop test
1632 if ( isset( $this->mTemplatePath
[$part1] ) ) {
1636 # set $text to cached message.
1637 $text = $this->mTemplates
[$part1];
1641 # Load from database
1643 $title = Title
::newFromText( $part1, NS_TEMPLATE
);
1644 if ( !is_null( $title ) && !$title->isExternal() ) {
1645 # Check for excessive inclusion
1646 $dbk = $title->getPrefixedDBkey();
1647 if ( $this->incrementIncludeCount( $dbk ) ) {
1648 # This should never be reached.
1649 $article = new Article( $title );
1650 $articleContent = $article->getContentWithoutUsingSoManyDamnGlobals();
1651 if ( $articleContent !== false ) {
1653 $text = $articleContent;
1657 # If the title is valid but undisplayable, make a link to it
1658 if ( $this->mOutputType
== OT_HTML
&& !$found ) {
1659 $text = '[['.$title->getPrefixedText().']]';
1663 # Template cache array insertion
1664 $this->mTemplates
[$part1] = $text;
1668 # Recursive parsing, escaping and link table handling
1669 # Only for HTML output
1670 if ( $nowiki && $found && $this->mOutputType
== OT_HTML
) {
1671 $text = wfEscapeWikiText( $text );
1672 } elseif ( $this->mOutputType
== OT_HTML
&& $found && !$noparse) {
1673 # Clean up argument array
1674 $assocArgs = array();
1676 foreach( $args as $arg ) {
1677 $eqpos = strpos( $arg, '=' );
1678 if ( $eqpos === false ) {
1679 $assocArgs[$index++
] = $arg;
1681 $name = trim( substr( $arg, 0, $eqpos ) );
1682 $value = trim( substr( $arg, $eqpos+
1 ) );
1683 if ( $value === false ) {
1686 if ( $name !== false ) {
1687 $assocArgs[$name] = $value;
1692 # Do not enter included links in link table
1693 if ( !is_null( $title ) ) {
1694 $wgLinkCache->suspend();
1697 # Add a new element to the templace recursion path
1698 $this->mTemplatePath
[$part1] = 1;
1700 $text = $this->removeHTMLtags( $text );
1701 $text = $this->replaceVariables( $text, $assocArgs );
1703 # Resume the link cache and register the inclusion as a link
1704 if ( !is_null( $title ) ) {
1705 $wgLinkCache->resume();
1706 $wgLinkCache->addLinkObj( $title );
1710 # Empties the template path
1711 $this->mTemplatePath
= array();
1720 # Triple brace replacement -- used for template arguments
1721 function argSubstitution( $matches ) {
1722 $newline = $matches[1];
1723 $arg = trim( $matches[2] );
1724 $text = $matches[0];
1725 $inputArgs = end( $this->mArgStack
);
1727 if ( array_key_exists( $arg, $inputArgs ) ) {
1728 $text = $this->removeHTMLtags( $inputArgs[$arg] );
1729 $text = $this->replaceVariables( $text, array() );
1735 # Returns true if the function is allowed to include this entity
1736 function incrementIncludeCount( $dbk ) {
1737 if ( !array_key_exists( $dbk, $this->mIncludeCount
) ) {
1738 $this->mIncludeCount
[$dbk] = 0;
1740 if ( ++
$this->mIncludeCount
[$dbk] <= MAX_INCLUDE_REPEAT
) {
1748 # Cleans up HTML, removes dangerous tags and attributes
1749 /* private */ function removeHTMLtags( $text ) {
1750 global $wgUseTidy, $wgUserHtml;
1751 $fname = 'Parser::removeHTMLtags';
1752 wfProfileIn( $fname );
1755 $htmlpairs = array( # Tags that must be closed
1756 'b', 'del', 'i', 'ins', 'u', 'font', 'big', 'small', 'sub', 'sup', 'h1',
1757 'h2', 'h3', 'h4', 'h5', 'h6', 'cite', 'code', 'em', 's',
1758 'strike', 'strong', 'tt', 'var', 'div', 'center',
1759 'blockquote', 'ol', 'ul', 'dl', 'table', 'caption', 'pre',
1760 'ruby', 'rt' , 'rb' , 'rp', 'p'
1762 $htmlsingle = array(
1763 'br', 'hr', 'li', 'dt', 'dd'
1765 $htmlnest = array( # Tags that can be nested--??
1766 'table', 'tr', 'td', 'th', 'div', 'blockquote', 'ol', 'ul',
1767 'dl', 'font', 'big', 'small', 'sub', 'sup'
1769 $tabletags = array( # Can only appear inside table
1773 $htmlpairs = array();
1774 $htmlsingle = array();
1775 $htmlnest = array();
1776 $tabletags = array();
1779 $htmlsingle = array_merge( $tabletags, $htmlsingle );
1780 $htmlelements = array_merge( $htmlsingle, $htmlpairs );
1782 $htmlattrs = $this->getHTMLattrs () ;
1784 # Remove HTML comments
1785 $text = preg_replace( '/(\\n *<!--.*--> *|<!--.*?-->)/sU', '', $text );
1787 $bits = explode( '<', $text );
1788 $text = array_shift( $bits );
1790 $tagstack = array(); $tablestack = array();
1791 foreach ( $bits as $x ) {
1792 $prev = error_reporting( E_ALL
& ~
( E_NOTICE | E_WARNING
) );
1793 preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
1795 list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
1796 error_reporting( $prev );
1799 if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
1803 if ( ! in_array( $t, $htmlsingle ) &&
1804 ( $ot = @array_pop
( $tagstack ) ) != $t ) {
1805 @array_push
( $tagstack, $ot );
1808 if ( $t == 'table' ) {
1809 $tagstack = array_pop( $tablestack );
1814 # Keep track for later
1815 if ( in_array( $t, $tabletags ) &&
1816 ! in_array( 'table', $tagstack ) ) {
1818 } else if ( in_array( $t, $tagstack ) &&
1819 ! in_array ( $t , $htmlnest ) ) {
1821 } else if ( ! in_array( $t, $htmlsingle ) ) {
1822 if ( $t == 'table' ) {
1823 array_push( $tablestack, $tagstack );
1824 $tagstack = array();
1826 array_push( $tagstack, $t );
1828 # Strip non-approved attributes from the tag
1829 $newparams = $this->fixTagAttributes($params);
1833 $rest = str_replace( '>', '>', $rest );
1834 $text .= "<$slash$t $newparams$brace$rest";
1838 $text .= '<' . str_replace( '>', '>', $x);
1840 # Close off any remaining tags
1841 while ( is_array( $tagstack ) && ($t = array_pop( $tagstack )) ) {
1843 if ( $t == 'table' ) { $tagstack = array_pop( $tablestack ); }
1846 # this might be possible using tidy itself
1847 foreach ( $bits as $x ) {
1848 preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
1850 @list
( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
1851 if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
1852 $newparams = $this->fixTagAttributes($params);
1853 $rest = str_replace( '>', '>', $rest );
1854 $text .= "<$slash$t $newparams$brace$rest";
1856 $text .= '<' . str_replace( '>', '>', $x);
1860 wfProfileOut( $fname );
1865 # This function accomplishes several tasks:
1866 # 1) Auto-number headings if that option is enabled
1867 # 2) Add an [edit] link to sections for logged in users who have enabled the option
1868 # 3) Add a Table of contents on the top for users who have enabled the option
1869 # 4) Auto-anchor headings
1871 # It loops through all headlines, collects the necessary data, then splits up the
1872 # string and re-inserts the newly formatted headlines.
1873 /* private */ function formatHeadings( $text, $isMain=true ) {
1874 global $wgInputEncoding, $wgMaxTocLevel, $wgLang;
1876 $doNumberHeadings = $this->mOptions
->getNumberHeadings();
1877 $doShowToc = $this->mOptions
->getShowToc();
1878 $forceTocHere = false;
1879 if( !$this->mTitle
->userCanEdit() ) {
1881 $rightClickHack = 0;
1883 $showEditLink = $this->mOptions
->getEditSection();
1884 $rightClickHack = $this->mOptions
->getEditSectionOnRightClick();
1887 # Inhibit editsection links if requested in the page
1888 $esw =& MagicWord
::get( MAG_NOEDITSECTION
);
1889 if( $esw->matchAndRemove( $text ) ) {
1892 # if the string __NOTOC__ (not case-sensitive) occurs in the HTML,
1894 $mw =& MagicWord
::get( MAG_NOTOC
);
1895 if( $mw->matchAndRemove( $text ) ) {
1899 # never add the TOC to the Main Page. This is an entry page that should not
1900 # be more than 1-2 screens large anyway
1901 if( $this->mTitle
->getPrefixedText() == wfMsg('mainpage') ) {
1905 # Get all headlines for numbering them and adding funky stuff like [edit]
1906 # links - this is for later, but we need the number of headlines right now
1907 $numMatches = preg_match_all( '/<H([1-6])(.*?' . '>)(.*?)<\/H[1-6]>/i', $text, $matches );
1909 # if there are fewer than 4 headlines in the article, do not show TOC
1910 if( $numMatches < 4 ) {
1914 # if the string __TOC__ (not case-sensitive) occurs in the HTML,
1915 # override above conditions and always show TOC at that place
1916 $mw =& MagicWord
::get( MAG_TOC
);
1917 if ($mw->match( $text ) ) {
1919 $forceTocHere = true;
1921 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
1922 # override above conditions and always show TOC above first header
1923 $mw =& MagicWord
::get( MAG_FORCETOC
);
1924 if ($mw->matchAndRemove( $text ) ) {
1931 # We need this to perform operations on the HTML
1932 $sk =& $this->mOptions
->getSkin();
1937 # Ugh .. the TOC should have neat indentation levels which can be
1938 # passed to the skin functions. These are determined here
1943 $sublevelCount = array();
1946 foreach( $matches[3] as $headline ) {
1949 $prevlevel = $level;
1951 $level = $matches[1][$headlineCount];
1952 if( ( $doNumberHeadings ||
$doShowToc ) && $prevlevel && $level > $prevlevel ) {
1953 # reset when we enter a new level
1954 $sublevelCount[$level] = 0;
1955 $toc .= $sk->tocIndent( $level - $prevlevel );
1956 $toclevel +
= $level - $prevlevel;
1958 if( ( $doNumberHeadings ||
$doShowToc ) && $level < $prevlevel ) {
1959 # reset when we step back a level
1960 $sublevelCount[$level+
1]=0;
1961 $toc .= $sk->tocUnindent( $prevlevel - $level );
1962 $toclevel -= $prevlevel - $level;
1964 # count number of headlines for each level
1965 @$sublevelCount[$level]++
;
1966 if( $doNumberHeadings ||
$doShowToc ) {
1968 for( $i = 1; $i <= $level; $i++
) {
1969 if( !empty( $sublevelCount[$i] ) ) {
1973 $numbering .= $wgLang->formatNum( $sublevelCount[$i] );
1979 # The canonized header is a version of the header text safe to use for links
1980 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
1981 $canonized_headline = $this->unstrip( $headline, $this->mStripState
);
1982 $canonized_headline = $this->unstripNoWiki( $headline, $this->mStripState
);
1984 # Remove link placeholders by the link text.
1985 # <!--LINK namespace page_title link text with suffix-->
1987 # link text with suffix
1988 $canonized_headline = preg_replace( '/<!--LINK [0-9]* [^ ]* *(.*?)-->/','$1', $canonized_headline );
1990 $canonized_headline = preg_replace( '/<.*?' . '>/','',$canonized_headline );
1991 $tocline = trim( $canonized_headline );
1992 $canonized_headline = urlencode( do_html_entity_decode( str_replace(' ', '_', $tocline), ENT_COMPAT
, $wgInputEncoding ) );
1993 $replacearray = array(
1997 $canonized_headline = str_replace(array_keys($replacearray),array_values($replacearray),$canonized_headline);
1998 $refer[$headlineCount] = $canonized_headline;
2000 # count how many in assoc. array so we can track dupes in anchors
2001 @$refers[$canonized_headline]++
;
2002 $refcount[$headlineCount]=$refers[$canonized_headline];
2004 # Prepend the number to the heading text
2006 if( $doNumberHeadings ||
$doShowToc ) {
2007 $tocline = $numbering . ' ' . $tocline;
2009 # Don't number the heading if it is the only one (looks silly)
2010 if( $doNumberHeadings && count( $matches[3] ) > 1) {
2011 # the two are different if the line contains a link
2012 $headline=$numbering . ' ' . $headline;
2016 # Create the anchor for linking from the TOC to the section
2017 $anchor = $canonized_headline;
2018 if($refcount[$headlineCount] > 1 ) {
2019 $anchor .= '_' . $refcount[$headlineCount];
2021 if( $doShowToc && ( !isset($wgMaxTocLevel) ||
$toclevel<$wgMaxTocLevel ) ) {
2022 $toc .= $sk->tocLine($anchor,$tocline,$toclevel);
2024 if( $showEditLink ) {
2025 if ( empty( $head[$headlineCount] ) ) {
2026 $head[$headlineCount] = '';
2028 $head[$headlineCount] .= $sk->editSectionLink($headlineCount+
1);
2031 # Add the edit section span
2032 if( $rightClickHack ) {
2033 $headline = $sk->editSectionScript($headlineCount+
1,$headline);
2036 # give headline the correct <h#> tag
2037 @$head[$headlineCount] .= "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount] .$headline.'</h'.$level.'>';
2043 $toclines = $headlineCount;
2044 $toc .= $sk->tocUnindent( $toclevel );
2045 $toc = $sk->tocTable( $toc );
2048 # split up and insert constructed headlines
2050 $blocks = preg_split( '/<H[1-6].*?' . '>.*?<\/H[1-6]>/i', $text );
2053 foreach( $blocks as $block ) {
2054 if( $showEditLink && $headlineCount > 0 && $i == 0 && $block != "\n" ) {
2055 # This is the [edit] link that appears for the top block of text when
2056 # section editing is enabled
2058 # Disabled because it broke block formatting
2059 # For example, a bullet point in the top line
2060 # $full .= $sk->editSectionLink(0);
2063 if( $doShowToc && !$i && $isMain && !$forceTocHere) {
2064 # Top anchor now in skin
2068 if( !empty( $head[$i] ) ) {
2074 $mw =& MagicWord
::get( MAG_TOC
);
2075 return $mw->replace( $toc, $full );
2081 # Return an HTML link for the "ISBN 123456" text
2082 /* private */ function magicISBN( $text ) {
2084 $fname = 'Parser::magicISBN';
2085 wfProfileIn( $fname );
2087 $a = split( 'ISBN ', ' '.$text );
2088 if ( count ( $a ) < 2 ) {
2089 wfProfileOut( $fname );
2092 $text = substr( array_shift( $a ), 1);
2093 $valid = '0123456789-ABCDEFGHIJKLMNOPQRSTUVWXYZ';
2095 foreach ( $a as $x ) {
2096 $isbn = $blank = '' ;
2097 while ( ' ' == $x{0} ) {
2099 $x = substr( $x, 1 );
2101 if ( $x == '' ) { # blank isbn
2102 $text .= "ISBN $blank";
2105 while ( strstr( $valid, $x{0} ) != false ) {
2107 $x = substr( $x, 1 );
2109 $num = str_replace( '-', '', $isbn );
2110 $num = str_replace( ' ', '', $num );
2113 $text .= "ISBN $blank$x";
2115 $titleObj = Title
::makeTitle( NS_SPECIAL
, 'Booksources' );
2116 $text .= '<a href="' .
2117 $titleObj->escapeLocalUrl( 'isbn='.$num ) .
2118 "\" class=\"internal\">ISBN $isbn</a>";
2122 wfProfileOut( $fname );
2126 # Return an HTML link for the "GEO ..." text
2127 /* private */ function magicGEO( $text ) {
2128 global $wgLang, $wgUseGeoMode;
2129 $fname = 'Parser::magicGEO';
2130 wfProfileIn( $fname );
2132 # These next five lines are only for the ~35000 U.S. Census Rambot pages...
2133 $directions = array ( 'N' => 'North' , 'S' => 'South' , 'E' => 'East' , 'W' => 'West' ) ;
2134 $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['N']}, (\d+)°(\d+)'(\d+)\" {$directions['W']}/" , "(GEO +\$1.\$2.\$3:-\$4.\$5.\$6)" , $text ) ;
2135 $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['N']}, (\d+)°(\d+)'(\d+)\" {$directions['E']}/" , "(GEO +\$1.\$2.\$3:+\$4.\$5.\$6)" , $text ) ;
2136 $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['S']}, (\d+)°(\d+)'(\d+)\" {$directions['W']}/" , "(GEO +\$1.\$2.\$3:-\$4.\$5.\$6)" , $text ) ;
2137 $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['S']}, (\d+)°(\d+)'(\d+)\" {$directions['E']}/" , "(GEO +\$1.\$2.\$3:+\$4.\$5.\$6)" , $text ) ;
2139 $a = split( 'GEO ', ' '.$text );
2140 if ( count ( $a ) < 2 ) {
2141 wfProfileOut( $fname );
2144 $text = substr( array_shift( $a ), 1);
2145 $valid = '0123456789.+-:';
2147 foreach ( $a as $x ) {
2148 $geo = $blank = '' ;
2149 while ( ' ' == $x{0} ) {
2151 $x = substr( $x, 1 );
2153 while ( strstr( $valid, $x{0} ) != false ) {
2155 $x = substr( $x, 1 );
2157 $num = str_replace( '+', '', $geo );
2158 $num = str_replace( ' ', '', $num );
2160 if ( '' == $num ||
count ( explode ( ':' , $num , 3 ) ) < 2 ) {
2161 $text .= "GEO $blank$x";
2163 $titleObj = Title
::makeTitle( NS_SPECIAL
, 'Geo' );
2164 $text .= '<a href="' .
2165 $titleObj->escapeLocalUrl( 'coordinates='.$num ) .
2166 "\" class=\"internal\">GEO $geo</a>";
2170 wfProfileOut( $fname );
2174 # Return an HTML link for the "RFC 1234" text
2175 /* private */ function magicRFC( $text ) {
2178 $a = split( 'RFC ', ' '.$text );
2179 if ( count ( $a ) < 2 ) return $text;
2180 $text = substr( array_shift( $a ), 1);
2181 $valid = '0123456789';
2183 foreach ( $a as $x ) {
2184 $rfc = $blank = '' ;
2185 while ( ' ' == $x{0} ) {
2187 $x = substr( $x, 1 );
2189 while ( strstr( $valid, $x{0} ) != false ) {
2191 $x = substr( $x, 1 );
2195 $text .= "RFC $blank$x";
2197 $url = wfmsg( 'rfcurl' );
2198 $url = str_replace( '$1', $rfc, $url);
2199 $sk =& $this->mOptions
->getSkin();
2200 $la = $sk->getExternalLinkAttributes( $url, 'RFC '.$rfc );
2201 $text .= "<a href='{$url}'{$la}>RFC {$rfc}</a>{$x}";
2207 function preSaveTransform( $text, &$title, &$user, $options, $clearState = true ) {
2208 $this->mOptions
= $options;
2209 $this->mTitle
=& $title;
2210 $this->mOutputType
= OT_WIKI
;
2212 if ( $clearState ) {
2213 $this->clearState();
2216 $stripState = false;
2220 $text = str_replace(array_keys($pairs), array_values($pairs), $text);
2224 "/<br.+(clear|break)=[\"']?(all|both)[\"']?\\/?>/i" => '<br style="clear:both;"/>',
2225 "/<br *?>/i" => "<br />",
2227 $text = preg_replace(array_keys($pairs), array_values($pairs), $text);
2229 $text = $this->strip( $text, $stripState, false );
2230 $text = $this->pstPass2( $text, $user );
2231 $text = $this->unstrip( $text, $stripState );
2232 $text = $this->unstripNoWiki( $text, $stripState );
2236 /* private */ function pstPass2( $text, &$user ) {
2237 global $wgLang, $wgLocaltimezone, $wgCurParser;
2239 # Variable replacement
2240 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
2241 $text = $this->replaceVariables( $text );
2245 $n = $user->getName();
2246 $k = $user->getOption( 'nickname' );
2247 if ( '' == $k ) { $k = $n; }
2248 if(isset($wgLocaltimezone)) {
2249 $oldtz = getenv('TZ'); putenv('TZ='.$wgLocaltimezone);
2251 /* Note: this is an ugly timezone hack for the European wikis */
2252 $d = $wgLang->timeanddate( date( 'YmdHis' ), false ) .
2253 ' (' . date( 'T' ) . ')';
2254 if(isset($wgLocaltimezone)) putenv('TZ='.$oldtzs);
2256 $text = preg_replace( '/~~~~~/', $d, $text );
2257 $text = preg_replace( '/~~~~/', '[[' . $wgLang->getNsText( NS_USER
) . ":$n|$k]] $d", $text );
2258 $text = preg_replace( '/~~~/', '[[' . $wgLang->getNsText( NS_USER
) . ":$n|$k]]", $text );
2260 # Context links: [[|name]] and [[name (context)|]]
2262 $tc = "[&;%\\-,.\\(\\)' _0-9A-Za-z\\/:\\x80-\\xff]";
2263 $np = "[&;%\\-,.' _0-9A-Za-z\\/:\\x80-\\xff]"; # No parens
2264 $namespacechar = '[ _0-9A-Za-z\x80-\xff]'; # Namespaces can use non-ascii!
2265 $conpat = "/^({$np}+) \\(({$tc}+)\\)$/";
2267 $p1 = "/\[\[({$np}+) \\(({$np}+)\\)\\|]]/"; # [[page (context)|]]
2268 $p2 = "/\[\[\\|({$tc}+)]]/"; # [[|page]]
2269 $p3 = "/\[\[(:*$namespacechar+):({$np}+)\\|]]/"; # [[namespace:page|]] and [[:namespace:page|]]
2270 $p4 = "/\[\[(:*$namespacechar+):({$np}+) \\(({$np}+)\\)\\|]]/"; # [[ns:page (cont)|]] and [[:ns:page (cont)|]]
2272 $t = $this->mTitle
->getText();
2273 if ( preg_match( $conpat, $t, $m ) ) {
2276 $text = preg_replace( $p4, '[[\\1:\\2 (\\3)|\\2]]', $text );
2277 $text = preg_replace( $p1, '[[\\1 (\\2)|\\1]]', $text );
2278 $text = preg_replace( $p3, '[[\\1:\\2|\\2]]', $text );
2280 if ( '' == $context ) {
2281 $text = preg_replace( $p2, '[[\\1]]', $text );
2283 $text = preg_replace( $p2, "[[\\1 ({$context})|\\1]]", $text );
2287 $mw =& MagicWord::get( MAG_SUBST );
2288 $wgCurParser = $this->fork();
2289 $text = $mw->substituteCallback( $text, "wfBraceSubstitution" );
2290 $this->merge( $wgCurParser );
2293 # Trim trailing whitespace
2294 # MAG_END (__END__) tag allows for trailing
2295 # whitespace to be deliberately included
2296 $text = rtrim( $text );
2297 $mw =& MagicWord
::get( MAG_END
);
2298 $mw->matchAndRemove( $text );
2303 # Set up some variables which are usually set up in parse()
2304 # so that an external function can call some class members with confidence
2305 function startExternalParse( &$title, $options, $outputType, $clearState = true ) {
2306 $this->mTitle
=& $title;
2307 $this->mOptions
= $options;
2308 $this->mOutputType
= $outputType;
2309 if ( $clearState ) {
2310 $this->clearState();
2314 function transformMsg( $text, $options ) {
2316 static $executing = false;
2318 # Guard against infinite recursion
2324 $this->mTitle
= $wgTitle;
2325 $this->mOptions
= $options;
2326 $this->mOutputType
= OT_MSG
;
2327 $this->clearState();
2328 $text = $this->replaceVariables( $text );
2334 # Create an HTML-style tag, e.g. <yourtag>special text</yourtag>
2335 # Callback will be called with the text within
2336 # Transform and return the text within
2337 function setHook( $tag, $callback ) {
2338 $oldVal = @$this->mTagHooks
[$tag];
2339 $this->mTagHooks
[$tag] = $callback;
2346 * @package MediaWiki
2350 var $mText, $mLanguageLinks, $mCategoryLinks, $mContainsOldMagic;
2351 var $mCacheTime; # Used in ParserCache
2353 function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(),
2354 $containsOldMagic = false )
2356 $this->mText
= $text;
2357 $this->mLanguageLinks
= $languageLinks;
2358 $this->mCategoryLinks
= $categoryLinks;
2359 $this->mContainsOldMagic
= $containsOldMagic;
2360 $this->mCacheTime
= '';
2363 function getText() { return $this->mText
; }
2364 function getLanguageLinks() { return $this->mLanguageLinks
; }
2365 function getCategoryLinks() { return $this->mCategoryLinks
; }
2366 function getCacheTime() { return $this->mCacheTime
; }
2367 function containsOldMagic() { return $this->mContainsOldMagic
; }
2368 function setText( $text ) { return wfSetVar( $this->mText
, $text ); }
2369 function setLanguageLinks( $ll ) { return wfSetVar( $this->mLanguageLinks
, $ll ); }
2370 function setCategoryLinks( $cl ) { return wfSetVar( $this->mCategoryLinks
, $cl ); }
2371 function setContainsOldMagic( $com ) { return wfSetVar( $this->mContainsOldMagic
, $com ); }
2372 function setCacheTime( $t ) { return wfSetVar( $this->mCacheTime
, $t ); }
2374 function merge( $other ) {
2375 $this->mLanguageLinks
= array_merge( $this->mLanguageLinks
, $other->mLanguageLinks
);
2376 $this->mCategoryLinks
= array_merge( $this->mCategoryLinks
, $this->mLanguageLinks
);
2377 $this->mContainsOldMagic
= $this->mContainsOldMagic ||
$other->mContainsOldMagic
;
2383 * Set options of the Parser
2385 * @package MediaWiki
2389 # All variables are private
2390 var $mUseTeX; # Use texvc to expand <math> tags
2391 var $mUseDynamicDates; # Use $wgDateFormatter to format dates
2392 var $mInterwikiMagic; # Interlanguage links are removed and returned in an array
2393 var $mAllowExternalImages; # Allow external images inline
2394 var $mSkin; # Reference to the preferred skin
2395 var $mDateFormat; # Date format index
2396 var $mEditSection; # Create "edit section" links
2397 var $mEditSectionOnRightClick; # Generate JavaScript to edit section on right click
2398 var $mNumberHeadings; # Automatically number headings
2399 var $mShowToc; # Show table of contents
2401 function getUseTeX() { return $this->mUseTeX
; }
2402 function getUseDynamicDates() { return $this->mUseDynamicDates
; }
2403 function getInterwikiMagic() { return $this->mInterwikiMagic
; }
2404 function getAllowExternalImages() { return $this->mAllowExternalImages
; }
2405 function getSkin() { return $this->mSkin
; }
2406 function getDateFormat() { return $this->mDateFormat
; }
2407 function getEditSection() { return $this->mEditSection
; }
2408 function getEditSectionOnRightClick() { return $this->mEditSectionOnRightClick
; }
2409 function getNumberHeadings() { return $this->mNumberHeadings
; }
2410 function getShowToc() { return $this->mShowToc
; }
2412 function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX
, $x ); }
2413 function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates
, $x ); }
2414 function setInterwikiMagic( $x ) { return wfSetVar( $this->mInterwikiMagic
, $x ); }
2415 function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages
, $x ); }
2416 function setDateFormat( $x ) { return wfSetVar( $this->mDateFormat
, $x ); }
2417 function setEditSection( $x ) { return wfSetVar( $this->mEditSection
, $x ); }
2418 function setEditSectionOnRightClick( $x ) { return wfSetVar( $this->mEditSectionOnRightClick
, $x ); }
2419 function setNumberHeadings( $x ) { return wfSetVar( $this->mNumberHeadings
, $x ); }
2420 function setShowToc( $x ) { return wfSetVar( $this->mShowToc
, $x ); }
2422 function setSkin( &$x ) { $this->mSkin
=& $x; }
2424 # Get parser options
2425 /* static */ function newFromUser( &$user ) {
2426 $popts = new ParserOptions
;
2427 $popts->initialiseFromUser( $user );
2432 function initialiseFromUser( &$userInput ) {
2433 global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages;
2435 $fname = 'ParserOptions::initialiseFromUser';
2436 wfProfileIn( $fname );
2437 if ( !$userInput ) {
2439 $user->setLoaded( true );
2441 $user =& $userInput;
2444 $this->mUseTeX
= $wgUseTeX;
2445 $this->mUseDynamicDates
= $wgUseDynamicDates;
2446 $this->mInterwikiMagic
= $wgInterwikiMagic;
2447 $this->mAllowExternalImages
= $wgAllowExternalImages;
2448 wfProfileIn( $fname.'-skin' );
2449 $this->mSkin
=& $user->getSkin();
2450 wfProfileOut( $fname.'-skin' );
2451 $this->mDateFormat
= $user->getOption( 'date' );
2452 $this->mEditSection
= $user->getOption( 'editsection' );
2453 $this->mEditSectionOnRightClick
= $user->getOption( 'editsectiononrightclick' );
2454 $this->mNumberHeadings
= $user->getOption( 'numberheadings' );
2455 $this->mShowToc
= $user->getOption( 'showtoc' );
2456 wfProfileOut( $fname );
2462 # Regex callbacks, used in Parser::replaceVariables
2463 function wfBraceSubstitution( $matches ) {
2464 global $wgCurParser;
2465 return $wgCurParser->braceSubstitution( $matches );
2468 function wfArgSubstitution( $matches ) {
2469 global $wgCurParser;
2470 return $wgCurParser->argSubstitution( $matches );
2473 function wfVariableSubstitution( $matches ) {
2474 global $wgCurParser;
2475 return $wgCurParser->variableSubstitution( $matches );
2479 * Return the total number of articles
2481 function wfNumberOfArticles() {
2482 global $wgNumberOfArticles;
2485 return $wgNumberOfArticles;
2489 * Get various statistics from the database
2492 function wfLoadSiteStats() {
2493 global $wgNumberOfArticles, $wgTotalViews, $wgTotalEdits;
2494 $fname = 'wfLoadSiteStats';
2496 if ( -1 != $wgNumberOfArticles ) return;
2497 $dbr =& wfGetDB( DB_SLAVE
);
2498 $s = $dbr->getArray( 'site_stats',
2499 array( 'ss_total_views', 'ss_total_edits', 'ss_good_articles' ),
2500 array( 'ss_row_id' => 1 ), $fname
2503 if ( $s === false ) {
2506 $wgTotalViews = $s->ss_total_views
;
2507 $wgTotalEdits = $s->ss_total_edits
;
2508 $wgNumberOfArticles = $s->ss_good_articles
;
2512 function wfEscapeHTMLTagsOnly( $in ) {
2514 array( '"', '>', '<' ),
2515 array( '"', '>', '<' ),