97b18c6b5e39238857c003c3159cd1dd4d90f8c8
3 * File for Parser and related classes
10 require_once( 'Sanitizer.php' );
11 require_once( 'HttpFunctions.php' );
14 * Update this version number when the ParserOutput format
15 * changes in an incompatible way, so the parser cache
16 * can automatically discard old data.
18 define( 'MW_PARSER_VERSION', '1.6.0' );
21 * Variable substitution O(N^2) attack
23 * Without countermeasures, it would be possible to attack the parser by saving
24 * a page filled with a large number of inclusions of large pages. The size of
25 * the generated page would be proportional to the square of the input size.
26 * Hence, we limit the number of inclusions of any given page, thus bringing any
27 * attack back to O(N).
30 define( 'MAX_INCLUDE_REPEAT', 100 );
31 define( 'MAX_INCLUDE_SIZE', 1000000 ); // 1 Million
33 define( 'RLH_FOR_UPDATE', 1 );
35 # Allowed values for $mOutputType
36 define( 'OT_HTML', 1 );
37 define( 'OT_WIKI', 2 );
38 define( 'OT_MSG' , 3 );
40 # string parameter for extractTags which will cause it
41 # to strip HTML comments in addition to regular
42 # <XML>-style tags. This should not be anything we
43 # may want to use in wikisyntax
44 define( 'STRIP_COMMENTS', 'HTMLCommentStrip' );
46 # Constants needed for external link processing
47 define( 'HTTP_PROTOCOLS', 'http:\/\/|https:\/\/' );
48 # Everything except bracket, space, or control characters
49 define( 'EXT_LINK_URL_CLASS', '[^][<>"\\x00-\\x20\\x7F]' );
51 define( 'EXT_LINK_TEXT_CLASS', '[^\]\\x00-\\x1F\\x7F]' );
52 define( 'EXT_IMAGE_FNAME_CLASS', '[A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]' );
53 define( 'EXT_IMAGE_EXTENSIONS', 'gif|png|jpg|jpeg' );
54 define( 'EXT_LINK_BRACKETED', '/\[(\b(' . wfUrlProtocols() . ')'.EXT_LINK_URL_CLASS
.'+) *('.EXT_LINK_TEXT_CLASS
.'*?)\]/S' );
55 define( 'EXT_IMAGE_REGEX',
56 '/^('.HTTP_PROTOCOLS
.')'. # Protocol
57 '('.EXT_LINK_URL_CLASS
.'+)\\/'. # Hostname and path
58 '('.EXT_IMAGE_FNAME_CLASS
.'+)\\.((?i)'.EXT_IMAGE_EXTENSIONS
.')$/S' # Filename
64 * Processes wiki markup
67 * There are three main entry points into the Parser class:
69 * produces HTML output
71 * produces altered wiki markup.
73 * performs brace substitution on MediaWiki messages
78 * NOT $wgArticle, $wgUser or $wgTitle. Keep them away!
81 * $wgUseTex*, $wgUseDynamicDates*, $wgInterwikiMagic*,
82 * $wgNamespacesWithSubpages, $wgAllowExternalImages*,
83 * $wgLocaltimezone, $wgAllowSpecialInclusion*
85 * * only within ParserOptions
98 # Cleared with clearState():
99 var $mOutput, $mAutonumber, $mDTopen, $mStripState = array();
100 var $mVariables, $mIncludeCount, $mArgStack, $mLastSection, $mInPre;
101 var $mInterwikiLinkHolders, $mLinkHolders, $mUniqPrefix;
104 var $mOptions, // ParserOptions object
105 $mTitle, // Title context, used for self-link rendering and similar things
106 $mOutputType, // Output type, one of the OT_xxx constants
107 $mTemplates, // cache of already loaded templates, avoids
108 // multiple SQL queries for the same string
109 $mTemplatePath, // stores an unsorted hash of all the templates already loaded
110 // in this path. Used for loop detection.
111 $mIWTransData = array(),
112 $mRevisionId; // ID to display in {{REVISIONID}} tags
122 $this->mTemplates
= array();
123 $this->mTemplatePath
= array();
124 $this->mTagHooks
= array();
133 function clearState() {
134 $this->mOutput
= new ParserOutput
;
135 $this->mAutonumber
= 0;
136 $this->mLastSection
= '';
137 $this->mDTopen
= false;
138 $this->mVariables
= false;
139 $this->mIncludeCount
= array();
140 $this->mStripState
= array();
141 $this->mArgStack
= array();
142 $this->mInPre
= false;
143 $this->mInterwikiLinkHolders
= array(
147 $this->mLinkHolders
= array(
148 'namespaces' => array(),
150 'queries' => array(),
154 $this->mRevisionId
= null;
155 $this->mUniqPrefix
= 'UNIQ' . Parser
::getRandomString();
157 wfRunHooks( 'ParserClearState', array( &$this ) );
161 * Convert wikitext to HTML
162 * Do not call this function recursively.
165 * @param string $text Text we want to parse
166 * @param Title &$title A title object
167 * @param array $options
168 * @param boolean $linestart
169 * @param boolean $clearState
170 * @param int $revid number to pass in {{REVISIONID}}
171 * @return ParserOutput a ParserOutput
173 function parse( $text, &$title, $options, $linestart = true, $clearState = true, $revid = null ) {
175 * First pass--just handle <nowiki> sections, pass the rest off
176 * to internalParse() which does all the real work.
179 global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang;
180 $fname = 'Parser::parse';
181 wfProfileIn( $fname );
187 $this->mOptions
= $options;
188 $this->mTitle
=& $title;
189 $this->mRevisionId
= $revid;
190 $this->mOutputType
= OT_HTML
;
192 $this->mStripState
= NULL;
194 //$text = $this->strip( $text, $this->mStripState );
195 // VOODOO MAGIC FIX! Sometimes the above segfaults in PHP5.
196 $x =& $this->mStripState
;
198 wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$x ) );
199 $text = $this->strip( $text, $x );
200 wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$x ) );
202 # Hook to suspend the parser in this state
203 if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$x ) ) ) {
204 wfProfileOut( $fname );
208 $text = $this->internalParse( $text );
210 $text = $this->unstrip( $text, $this->mStripState
);
212 # Clean up special characters, only run once, next-to-last before doBlockLevels
214 # french spaces, last one Guillemet-left
215 # only if there is something before the space
216 '/(.) (?=\\?|:|;|!|\\302\\273)/' => '\\1 \\2',
217 # french spaces, Guillemet-right
218 '/(\\302\\253) /' => '\\1 ',
219 '/<center *>(.*)<\\/center *>/i' => '<div class="center">\\1</div>',
221 $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
224 $text = $this->doBlockLevels( $text, $linestart );
226 $this->replaceLinkHolders( $text );
228 # the position of the convert() call should not be changed. it
229 # assumes that the links are all replaces and the only thing left
230 # is the <nowiki> mark.
231 $text = $wgContLang->convert($text);
232 $this->mOutput
->setTitleText($wgContLang->getParsedTitle());
234 $text = $this->unstripNoWiki( $text, $this->mStripState
);
236 wfRunHooks( 'ParserBeforeTidy', array( &$this, &$text ) );
238 $text = Sanitizer
::normalizeCharReferences( $text );
240 if (($wgUseTidy and $this->mOptions
->mTidy
) or $wgAlwaysUseTidy) {
241 $text = Parser
::tidy($text);
244 wfRunHooks( 'ParserAfterTidy', array( &$this, &$text ) );
246 $this->mOutput
->setText( $text );
247 wfProfileOut( $fname );
249 return $this->mOutput
;
253 * Get a random string
258 function getRandomString() {
259 return dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff));
263 * Replaces all occurrences of <$tag>content</$tag> in the text
264 * with a random marker and returns the new text. the output parameter
265 * $content will be an associative array filled with data on the form
266 * $unique_marker => content.
268 * If $content is already set, the additional entries will be appended
269 * If $tag is set to STRIP_COMMENTS, the function will extract
270 * <!-- HTML comments -->
275 function extractTagsAndParams($tag, $text, &$content, &$tags, &$params, $uniq_prefix = ''){
276 $rnd = $uniq_prefix . '-' . $tag . Parser
::getRandomString();
291 if( $tag == STRIP_COMMENTS
) {
292 $start = '/<!--()()/';
295 $start = "/<$tag(\\s+[^\\/>]*|\\s*)(\\/?)>/i";
296 $end = "/<\\/$tag\\s*>/i";
299 while ( '' != $text ) {
300 $p = preg_split( $start, $text, 2, PREG_SPLIT_DELIM_CAPTURE
);
302 if( count( $p ) < 4 ) {
309 $marker = $rnd . sprintf('%08X', $n++
);
310 $stripped .= $marker;
312 $tags[$marker] = "<$tag$attributes$empty>";
313 $params[$marker] = Sanitizer
::decodeTagAttributes( $attributes );
315 if ( $empty === '/' ) {
316 // Empty element tag, <tag />
317 $content[$marker] = null;
320 $q = preg_split( $end, $inside, 2 );
321 $content[$marker] = $q[0];
322 if( count( $q ) < 2 ) {
323 # No end tag -- let it run out to the end of the text.
334 * Wrapper function for extractTagsAndParams
335 * for cases where $tags and $params isn't needed
336 * i.e. where tags will never have params, like <nowiki>
341 function extractTags( $tag, $text, &$content, $uniq_prefix = '' ) {
342 $dummy_tags = array();
343 $dummy_params = array();
345 return Parser
::extractTagsAndParams( $tag, $text, $content,
346 $dummy_tags, $dummy_params, $uniq_prefix );
350 * Strips and renders nowiki, pre, math, hiero
351 * If $render is set, performs necessary rendering operations on plugins
352 * Returns the text, and fills an array with data needed in unstrip()
353 * If the $state is already a valid strip state, it adds to the state
355 * @param bool $stripcomments when set, HTML comments <!-- like this -->
356 * will be stripped in addition to other tags. This is important
357 * for section editing, where these comments cause confusion when
358 * counting the sections in the wikisource
362 function strip( $text, &$state, $stripcomments = false ) {
363 $render = ($this->mOutputType
== OT_HTML
);
364 $html_content = array();
365 $nowiki_content = array();
366 $math_content = array();
367 $pre_content = array();
368 $comment_content = array();
369 $ext_content = array();
371 $ext_params = array();
372 $gallery_content = array();
374 # Replace any instances of the placeholders
375 $uniq_prefix = $this->mUniqPrefix
;
376 #$text = str_replace( $uniq_prefix, wfHtmlEscapeFirst( $uniq_prefix ), $text );
381 $text = Parser
::extractTags('html', $text, $html_content, $uniq_prefix);
382 foreach( $html_content as $marker => $content ) {
384 # Raw and unchecked for validity.
385 $html_content[$marker] = $content;
387 $html_content[$marker] = '<html>'.$content.'</html>';
393 $text = Parser
::extractTags('nowiki', $text, $nowiki_content, $uniq_prefix);
394 foreach( $nowiki_content as $marker => $content ) {
396 $nowiki_content[$marker] = wfEscapeHTMLTagsOnly( $content );
398 $nowiki_content[$marker] = '<nowiki>'.$content.'</nowiki>';
403 if( $this->mOptions
->getUseTeX() ) {
404 $text = Parser
::extractTags('math', $text, $math_content, $uniq_prefix);
405 foreach( $math_content as $marker => $content ){
407 $math_content[$marker] = renderMath( $content );
409 $math_content[$marker] = '<math>'.$content.'</math>';
415 $text = Parser
::extractTags('pre', $text, $pre_content, $uniq_prefix);
416 foreach( $pre_content as $marker => $content ){
418 $pre_content[$marker] = '<pre>' . wfEscapeHTMLTagsOnly( $content ) . '</pre>';
420 $pre_content[$marker] = '<pre>'.$content.'</pre>';
425 $text = Parser
::extractTags('gallery', $text, $gallery_content, $uniq_prefix);
426 foreach( $gallery_content as $marker => $content ) {
427 require_once( 'ImageGallery.php' );
429 $gallery_content[$marker] = $this->renderImageGallery( $content );
431 $gallery_content[$marker] = '<gallery>'.$content.'</gallery>';
437 $text = Parser
::extractTags(STRIP_COMMENTS
, $text, $comment_content, $uniq_prefix);
438 foreach( $comment_content as $marker => $content ){
439 $comment_content[$marker] = '<!--'.$content.'-->';
444 foreach ( $this->mTagHooks
as $tag => $callback ) {
445 $ext_content[$tag] = array();
446 $text = Parser
::extractTagsAndParams( $tag, $text, $ext_content[$tag],
447 $ext_tags[$tag], $ext_params[$tag], $uniq_prefix );
448 foreach( $ext_content[$tag] as $marker => $content ) {
449 $full_tag = $ext_tags[$tag][$marker];
450 $params = $ext_params[$tag][$marker];
452 $ext_content[$tag][$marker] = call_user_func_array( $callback, array( $content, $params, &$this ) );
454 if ( is_null( $content ) ) {
456 $ext_content[$tag][$marker] = $full_tag;
458 $ext_content[$tag][$marker] = "$full_tag$content</$tag>";
464 # Merge state with the pre-existing state, if there is one
466 $state['html'] = $state['html'] +
$html_content;
467 $state['nowiki'] = $state['nowiki'] +
$nowiki_content;
468 $state['math'] = $state['math'] +
$math_content;
469 $state['pre'] = $state['pre'] +
$pre_content;
470 $state['comment'] = $state['comment'] +
$comment_content;
471 $state['gallery'] = $state['gallery'] +
$gallery_content;
473 foreach( $ext_content as $tag => $array ) {
474 if ( array_key_exists( $tag, $state ) ) {
475 $state[$tag] = $state[$tag] +
$array;
480 'html' => $html_content,
481 'nowiki' => $nowiki_content,
482 'math' => $math_content,
483 'pre' => $pre_content,
484 'comment' => $comment_content,
485 'gallery' => $gallery_content,
492 * restores pre, math, and hiero removed by strip()
494 * always call unstripNoWiki() after this one
497 function unstrip( $text, &$state ) {
498 if ( !is_array( $state ) ) {
502 # Must expand in reverse order, otherwise nested tags will be corrupted
503 foreach( array_reverse( $state, true ) as $tag => $contentDict ) {
504 if( $tag != 'nowiki' && $tag != 'html' ) {
505 foreach( array_reverse( $contentDict, true ) as $uniq => $content ) {
506 $text = str_replace( $uniq, $content, $text );
515 * always call this after unstrip() to preserve the order
519 function unstripNoWiki( $text, &$state ) {
520 if ( !is_array( $state ) ) {
524 # Must expand in reverse order, otherwise nested tags will be corrupted
525 for ( $content = end($state['nowiki']); $content !== false; $content = prev( $state['nowiki'] ) ) {
526 $text = str_replace( key( $state['nowiki'] ), $content, $text );
531 for ( $content = end($state['html']); $content !== false; $content = prev( $state['html'] ) ) {
532 $text = str_replace( key( $state['html'] ), $content, $text );
540 * Add an item to the strip state
541 * Returns the unique tag which must be inserted into the stripped text
542 * The tag will be replaced with the original text in unstrip()
546 function insertStripItem( $text, &$state ) {
547 $rnd = $this->mUniqPrefix
. '-item' . Parser
::getRandomString();
554 'comment' => array(),
555 'gallery' => array(),
558 $state['item'][$rnd] = $text;
563 * Interface with html tidy, used if $wgUseTidy = true.
564 * If tidy isn't able to correct the markup, the original will be
565 * returned in all its glory with a warning comment appended.
567 * Either the external tidy program or the in-process tidy extension
568 * will be used depending on availability. Override the default
569 * $wgTidyInternal setting to disable the internal if it's not working.
571 * @param string $text Hideous HTML input
572 * @return string Corrected HTML output
576 function tidy( $text ) {
577 global $wgTidyInternal;
578 $wrappedtext = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'.
579 ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>'.
580 '<head><title>test</title></head><body>'.$text.'</body></html>';
581 if( $wgTidyInternal ) {
582 $correctedtext = Parser
::internalTidy( $wrappedtext );
584 $correctedtext = Parser
::externalTidy( $wrappedtext );
586 if( is_null( $correctedtext ) ) {
587 wfDebug( "Tidy error detected!\n" );
588 return $text . "\n<!-- Tidy found serious XHTML errors -->\n";
590 return $correctedtext;
594 * Spawn an external HTML tidy process and get corrected markup back from it.
599 function externalTidy( $text ) {
600 global $wgTidyConf, $wgTidyBin, $wgTidyOpts;
601 $fname = 'Parser::externalTidy';
602 wfProfileIn( $fname );
607 $descriptorspec = array(
608 0 => array('pipe', 'r'),
609 1 => array('pipe', 'w'),
610 2 => array('file', '/dev/null', 'a')
613 $process = proc_open("$wgTidyBin -config $wgTidyConf $wgTidyOpts$opts", $descriptorspec, $pipes);
614 if (is_resource($process)) {
615 fwrite($pipes[0], $text);
617 while (!feof($pipes[1])) {
618 $cleansource .= fgets($pipes[1], 1024);
621 proc_close($process);
624 wfProfileOut( $fname );
626 if( $cleansource == '' && $text != '') {
627 // Some kind of error happened, so we couldn't get the corrected text.
628 // Just give up; we'll use the source text and append a warning.
636 * Use the HTML tidy PECL extension to use the tidy library in-process,
637 * saving the overhead of spawning a new process. Currently written to
638 * the PHP 4.3.x version of the extension, may not work on PHP 5.
640 * 'pear install tidy' should be able to compile the extension module.
645 function internalTidy( $text ) {
647 $fname = 'Parser::internalTidy';
648 wfProfileIn( $fname );
650 tidy_load_config( $wgTidyConf );
651 tidy_set_encoding( 'utf8' );
652 tidy_parse_string( $text );
654 if( tidy_get_status() == 2 ) {
655 // 2 is magic number for fatal error
656 // http://www.php.net/manual/en/function.tidy-get-status.php
659 $cleansource = tidy_get_output();
661 wfProfileOut( $fname );
666 * parse the wiki syntax used to render tables
670 function doTableStuff ( $t ) {
671 $fname = 'Parser::doTableStuff';
672 wfProfileIn( $fname );
674 $t = explode ( "\n" , $t ) ;
675 $td = array () ; # Is currently a td tag open?
676 $ltd = array () ; # Was it TD or TH?
677 $tr = array () ; # Is currently a tr tag open?
678 $ltr = array () ; # tr attributes
679 $indent_level = 0; # indent level of the table
680 foreach ( $t AS $k => $x )
683 $fc = substr ( $x , 0 , 1 ) ;
684 if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
685 $indent_level = strlen( $matches[1] );
687 $attributes = $this->unstripForHTML( $matches[2] );
689 $t[$k] = str_repeat( '<dl><dd>', $indent_level ) .
690 '<table' . Sanitizer
::fixTagAttributes ( $attributes, 'table' ) . '>' ;
691 array_push ( $td , false ) ;
692 array_push ( $ltd , '' ) ;
693 array_push ( $tr , false ) ;
694 array_push ( $ltr , '' ) ;
696 else if ( count ( $td ) == 0 ) { } # Don't do any of the following
697 else if ( '|}' == substr ( $x , 0 , 2 ) ) {
698 $z = "</table>" . substr ( $x , 2);
699 $l = array_pop ( $ltd ) ;
700 if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
701 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
703 $t[$k] = $z . str_repeat( '</dd></dl>', $indent_level );
705 else if ( '|-' == substr ( $x , 0 , 2 ) ) { # Allows for |---------------
706 $x = substr ( $x , 1 ) ;
707 while ( $x != '' && substr ( $x , 0 , 1 ) == '-' ) $x = substr ( $x , 1 ) ;
709 $l = array_pop ( $ltd ) ;
710 if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
711 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
714 array_push ( $tr , false ) ;
715 array_push ( $td , false ) ;
716 array_push ( $ltd , '' ) ;
717 $attributes = $this->unstripForHTML( $x );
718 array_push ( $ltr , Sanitizer
::fixTagAttributes ( $attributes, 'tr' ) ) ;
720 else if ( '|' == $fc ||
'!' == $fc ||
'|+' == substr ( $x , 0 , 2 ) ) { # Caption
722 if ( '|+' == substr ( $x , 0 , 2 ) ) {
724 $x = substr ( $x , 1 ) ;
726 $after = substr ( $x , 1 ) ;
727 if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;
728 $after = explode ( '||' , $after ) ;
731 # Loop through each table cell
732 foreach ( $after AS $theline )
737 $tra = array_pop ( $ltr ) ;
738 if ( !array_pop ( $tr ) ) $z = '<tr'.$tra.">\n" ;
739 array_push ( $tr , true ) ;
740 array_push ( $ltr , '' ) ;
743 $l = array_pop ( $ltd ) ;
744 if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
745 if ( $fc == '|' ) $l = 'td' ;
746 else if ( $fc == '!' ) $l = 'th' ;
747 else if ( $fc == '+' ) $l = 'caption' ;
749 array_push ( $ltd , $l ) ;
752 $y = explode ( '|' , $theline , 2 ) ;
753 # Note that a '|' inside an invalid link should not
754 # be mistaken as delimiting cell parameters
755 if ( strpos( $y[0], '[[' ) !== false ) {
756 $y = array ($theline);
758 if ( count ( $y ) == 1 )
759 $y = "{$z}<{$l}>{$y[0]}" ;
761 $attributes = $this->unstripForHTML( $y[0] );
762 $y = "{$z}<{$l}".Sanitizer
::fixTagAttributes($attributes, $l).">{$y[1]}" ;
765 array_push ( $td , true ) ;
770 # Closing open td, tr && table
771 while ( count ( $td ) > 0 )
773 if ( array_pop ( $td ) ) $t[] = '</td>' ;
774 if ( array_pop ( $tr ) ) $t[] = '</tr>' ;
778 $t = implode ( "\n" , $t ) ;
779 wfProfileOut( $fname );
784 * Helper function for parse() that transforms wiki markup into
785 * HTML. Only called for $mOutputType == OT_HTML.
789 function internalParse( $text ) {
793 $fname = 'Parser::internalParse';
794 wfProfileIn( $fname );
796 # Remove <noinclude> tags and <includeonly> sections
797 $text = strtr( $text, array( '<onlyinclude>' => '' , '</onlyinclude>' => '' ) );
798 $text = strtr( $text, array( '<noinclude>' => '', '</noinclude>' => '') );
799 $text = preg_replace( '/<includeonly>.*?<\/includeonly>/s', '', $text );
801 $text = Sanitizer
::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ) );
802 $text = $this->replaceVariables( $text, $args );
804 $text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
806 $text = $this->doHeadings( $text );
807 if($this->mOptions
->getUseDynamicDates()) {
808 $df =& DateFormatter
::getInstance();
809 $text = $df->reformat( $this->mOptions
->getDateFormat(), $text );
811 $text = $this->doAllQuotes( $text );
812 $text = $this->replaceInternalLinks( $text );
813 $text = $this->replaceExternalLinks( $text );
815 # replaceInternalLinks may sometimes leave behind
816 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
817 $text = str_replace($this->mUniqPrefix
."NOPARSE", "", $text);
819 $text = $this->doMagicLinks( $text );
820 $text = $this->doTableStuff( $text );
821 $text = $this->formatHeadings( $text, $isMain );
823 $regex = '/<!--IW_TRANSCLUDE (\d+)-->/';
824 $text = preg_replace_callback($regex, array(&$this, 'scarySubstitution'), $text);
826 wfProfileOut( $fname );
830 function scarySubstitution($matches) {
831 # return "[[".$matches[0]."]]";
832 return $this->mIWTransData
[(int)$matches[0]];
836 * Replace special strings like "ISBN xxx" and "RFC xxx" with
837 * magic external links.
841 function &doMagicLinks( &$text ) {
842 $text = $this->magicISBN( $text );
843 $text = $this->magicRFC( $text, 'RFC ', 'rfcurl' );
844 $text = $this->magicRFC( $text, 'PMID ', 'pubmedurl' );
849 * Parse headers and return html
853 function doHeadings( $text ) {
854 $fname = 'Parser::doHeadings';
855 wfProfileIn( $fname );
856 for ( $i = 6; $i >= 1; --$i ) {
857 $h = str_repeat( '=', $i );
858 $text = preg_replace( "/^{$h}(.+){$h}(\\s|$)/m",
859 "<h{$i}>\\1</h{$i}>\\2", $text );
861 wfProfileOut( $fname );
866 * Replace single quotes with HTML markup
868 * @return string the altered text
870 function doAllQuotes( $text ) {
871 $fname = 'Parser::doAllQuotes';
872 wfProfileIn( $fname );
874 $lines = explode( "\n", $text );
875 foreach ( $lines as $line ) {
876 $outtext .= $this->doQuotes ( $line ) . "\n";
878 $outtext = substr($outtext, 0,-1);
879 wfProfileOut( $fname );
884 * Helper function for doAllQuotes()
887 function doQuotes( $text ) {
888 $arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
889 if ( count( $arr ) == 1 )
893 # First, do some preliminary work. This may shift some apostrophes from
894 # being mark-up to being text. It also counts the number of occurrences
895 # of bold and italics mark-ups.
899 foreach ( $arr as $r )
901 if ( ( $i %
2 ) == 1 )
903 # If there are ever four apostrophes, assume the first is supposed to
904 # be text, and the remaining three constitute mark-up for bold text.
905 if ( strlen( $arr[$i] ) == 4 )
910 # If there are more than 5 apostrophes in a row, assume they're all
911 # text except for the last 5.
912 else if ( strlen( $arr[$i] ) > 5 )
914 $arr[$i-1] .= str_repeat( "'", strlen( $arr[$i] ) - 5 );
917 # Count the number of occurrences of bold and italics mark-ups.
918 # We are not counting sequences of five apostrophes.
919 if ( strlen( $arr[$i] ) == 2 ) $numitalics++
; else
920 if ( strlen( $arr[$i] ) == 3 ) $numbold++
; else
921 if ( strlen( $arr[$i] ) == 5 ) { $numitalics++
; $numbold++
; }
926 # If there is an odd number of both bold and italics, it is likely
927 # that one of the bold ones was meant to be an apostrophe followed
928 # by italics. Which one we cannot know for certain, but it is more
929 # likely to be one that has a single-letter word before it.
930 if ( ( $numbold %
2 == 1 ) && ( $numitalics %
2 == 1 ) )
933 $firstsingleletterword = -1;
934 $firstmultiletterword = -1;
936 foreach ( $arr as $r )
938 if ( ( $i %
2 == 1 ) and ( strlen( $r ) == 3 ) )
940 $x1 = substr ($arr[$i-1], -1);
941 $x2 = substr ($arr[$i-1], -2, 1);
943 if ($firstspace == -1) $firstspace = $i;
944 } else if ($x2 == ' ') {
945 if ($firstsingleletterword == -1) $firstsingleletterword = $i;
947 if ($firstmultiletterword == -1) $firstmultiletterword = $i;
953 # If there is a single-letter word, use it!
954 if ($firstsingleletterword > -1)
956 $arr [ $firstsingleletterword ] = "''";
957 $arr [ $firstsingleletterword-1 ] .= "'";
959 # If not, but there's a multi-letter word, use that one.
960 else if ($firstmultiletterword > -1)
962 $arr [ $firstmultiletterword ] = "''";
963 $arr [ $firstmultiletterword-1 ] .= "'";
965 # ... otherwise use the first one that has neither.
966 # (notice that it is possible for all three to be -1 if, for example,
967 # there is only one pentuple-apostrophe in the line)
968 else if ($firstspace > -1)
970 $arr [ $firstspace ] = "''";
971 $arr [ $firstspace-1 ] .= "'";
975 # Now let's actually convert our apostrophic mush to HTML!
984 if ($state == 'both')
991 if (strlen ($r) == 2)
994 { $output .= '</i>'; $state = ''; }
995 else if ($state == 'bi')
996 { $output .= '</i>'; $state = 'b'; }
997 else if ($state == 'ib')
998 { $output .= '</b></i><b>'; $state = 'b'; }
999 else if ($state == 'both')
1000 { $output .= '<b><i>'.$buffer.'</i>'; $state = 'b'; }
1001 else # $state can be 'b' or ''
1002 { $output .= '<i>'; $state .= 'i'; }
1004 else if (strlen ($r) == 3)
1007 { $output .= '</b>'; $state = ''; }
1008 else if ($state == 'bi')
1009 { $output .= '</i></b><i>'; $state = 'i'; }
1010 else if ($state == 'ib')
1011 { $output .= '</b>'; $state = 'i'; }
1012 else if ($state == 'both')
1013 { $output .= '<i><b>'.$buffer.'</b>'; $state = 'i'; }
1014 else # $state can be 'i' or ''
1015 { $output .= '<b>'; $state .= 'b'; }
1017 else if (strlen ($r) == 5)
1020 { $output .= '</b><i>'; $state = 'i'; }
1021 else if ($state == 'i')
1022 { $output .= '</i><b>'; $state = 'b'; }
1023 else if ($state == 'bi')
1024 { $output .= '</i></b>'; $state = ''; }
1025 else if ($state == 'ib')
1026 { $output .= '</b></i>'; $state = ''; }
1027 else if ($state == 'both')
1028 { $output .= '<i><b>'.$buffer.'</b></i>'; $state = ''; }
1029 else # ($state == '')
1030 { $buffer = ''; $state = 'both'; }
1035 # Now close all remaining tags. Notice that the order is important.
1036 if ($state == 'b' ||
$state == 'ib')
1038 if ($state == 'i' ||
$state == 'bi' ||
$state == 'ib')
1042 if ($state == 'both')
1043 $output .= '<b><i>'.$buffer.'</i></b>';
1049 * Replace external links
1051 * Note: this is all very hackish and the order of execution matters a lot.
1052 * Make sure to run maintenance/parserTests.php if you change this code.
1056 function replaceExternalLinks( $text ) {
1058 $fname = 'Parser::replaceExternalLinks';
1059 wfProfileIn( $fname );
1061 $sk =& $this->mOptions
->getSkin();
1063 $bits = preg_split( EXT_LINK_BRACKETED
, $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
1065 $s = $this->replaceFreeExternalLinks( array_shift( $bits ) );
1068 while ( $i<count( $bits ) ) {
1070 $protocol = $bits[$i++
];
1071 $text = $bits[$i++
];
1072 $trail = $bits[$i++
];
1074 # The characters '<' and '>' (which were escaped by
1075 # removeHTMLtags()) should not be included in
1076 # URLs, per RFC 2396.
1077 if (preg_match('/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE
)) {
1078 $text = substr($url, $m2[0][1]) . ' ' . $text;
1079 $url = substr($url, 0, $m2[0][1]);
1082 # If the link text is an image URL, replace it with an <img> tag
1083 # This happened by accident in the original parser, but some people used it extensively
1084 $img = $this->maybeMakeExternalImage( $text );
1085 if ( $img !== false ) {
1091 # Set linktype for CSS - if URL==text, link is essentially free
1092 $linktype = ($text == $url) ?
'free' : 'text';
1094 # No link text, e.g. [http://domain.tld/some.link]
1095 if ( $text == '' ) {
1096 # Autonumber if allowed
1097 if ( strpos( HTTP_PROTOCOLS
, str_replace('/','\/', $protocol) ) !== false ) {
1098 $text = '[' . ++
$this->mAutonumber
. ']';
1099 $linktype = 'autonumber';
1101 # Otherwise just use the URL
1102 $text = htmlspecialchars( $url );
1106 # Have link text, e.g. [http://domain.tld/some.link text]s
1108 list( $dtrail, $trail ) = Linker
::splitTrail( $trail );
1111 $text = $wgContLang->markNoConversion($text);
1113 # Replace & from obsolete syntax with &.
1114 # All HTML entities will be escaped by makeExternalLink()
1115 # or maybeMakeExternalImage()
1116 $url = str_replace( '&', '&', $url );
1118 # Process the trail (i.e. everything after this link up until start of the next link),
1119 # replacing any non-bracketed links
1120 $trail = $this->replaceFreeExternalLinks( $trail );
1123 # Use the encoded URL
1124 # This means that users can paste URLs directly into the text
1125 # Funny characters like ö aren't valid in URLs anyway
1126 # This was changed in August 2004
1127 $s .= $sk->makeExternalLink( $url, $text, false, $linktype ) . $dtrail . $trail;
1130 wfProfileOut( $fname );
1135 * Replace anything that looks like a URL with a link
1138 function replaceFreeExternalLinks( $text ) {
1140 $fname = 'Parser::replaceFreeExternalLinks';
1141 wfProfileIn( $fname );
1143 $bits = preg_split( '/(\b(?:' . wfUrlProtocols() . '))/S', $text, -1, PREG_SPLIT_DELIM_CAPTURE
);
1144 $s = array_shift( $bits );
1147 $sk =& $this->mOptions
->getSkin();
1149 while ( $i < count( $bits ) ){
1150 $protocol = $bits[$i++
];
1151 $remainder = $bits[$i++
];
1153 if ( preg_match( '/^('.EXT_LINK_URL_CLASS
.'+)(.*)$/s', $remainder, $m ) ) {
1154 # Found some characters after the protocol that look promising
1155 $url = $protocol . $m[1];
1158 # The characters '<' and '>' (which were escaped by
1159 # removeHTMLtags()) should not be included in
1160 # URLs, per RFC 2396.
1161 if (preg_match('/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE
)) {
1162 $trail = substr($url, $m2[0][1]) . $trail;
1163 $url = substr($url, 0, $m2[0][1]);
1166 # Move trailing punctuation to $trail
1168 # If there is no left bracket, then consider right brackets fair game too
1169 if ( strpos( $url, '(' ) === false ) {
1173 $numSepChars = strspn( strrev( $url ), $sep );
1174 if ( $numSepChars ) {
1175 $trail = substr( $url, -$numSepChars ) . $trail;
1176 $url = substr( $url, 0, -$numSepChars );
1179 # Replace & from obsolete syntax with &.
1180 # All HTML entities will be escaped by makeExternalLink()
1181 # or maybeMakeExternalImage()
1182 $url = str_replace( '&', '&', $url );
1184 # Is this an external image?
1185 $text = $this->maybeMakeExternalImage( $url );
1186 if ( $text === false ) {
1187 # Not an image, make a link
1188 $text = $sk->makeExternalLink( $url, $wgContLang->markNoConversion($url), true, 'free' );
1190 $s .= $text . $trail;
1192 $s .= $protocol . $remainder;
1195 wfProfileOut( $fname );
1200 * make an image if it's allowed, either through the global
1201 * option or through the exception
1204 function maybeMakeExternalImage( $url ) {
1205 $sk =& $this->mOptions
->getSkin();
1206 $imagesfrom = $this->mOptions
->getAllowExternalImagesFrom();
1207 $imagesexception = !empty($imagesfrom);
1209 if ( $this->mOptions
->getAllowExternalImages()
1210 ||
( $imagesexception && strpos( $url, $imagesfrom ) === 0 ) ) {
1211 if ( preg_match( EXT_IMAGE_REGEX
, $url ) ) {
1213 $text = $sk->makeExternalImage( htmlspecialchars( $url ) );
1220 * Process [[ ]] wikilinks
1224 function replaceInternalLinks( $s ) {
1226 static $fname = 'Parser::replaceInternalLinks' ;
1228 wfProfileIn( $fname );
1230 wfProfileIn( $fname.'-setup' );
1232 # the % is needed to support urlencoded titles as well
1233 if ( !$tc ) { $tc = Title
::legalChars() . '#%'; }
1235 $sk =& $this->mOptions
->getSkin();
1237 #split the entire text string on occurences of [[
1238 $a = explode( '[[', ' ' . $s );
1239 #get the first element (all text up to first [[), and remove the space we added
1240 $s = array_shift( $a );
1241 $s = substr( $s, 1 );
1243 # Match a link having the form [[namespace:link|alternate]]trail
1245 if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|(.+?))?]](.*)\$/sD"; }
1246 # Match cases where there is no "]]", which might still be images
1247 static $e1_img = FALSE;
1248 if ( !$e1_img ) { $e1_img = "/^([{$tc}]+)\\|(.*)\$/sD"; }
1249 # Match the end of a line for a word that's not followed by whitespace,
1250 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
1251 $e2 = wfMsgForContent( 'linkprefix' );
1253 $useLinkPrefixExtension = $wgContLang->linkPrefixExtension();
1255 if( is_null( $this->mTitle
) ) {
1256 wfDebugDieBacktrace( 'nooo' );
1258 $nottalk = !$this->mTitle
->isTalkPage();
1260 if ( $useLinkPrefixExtension ) {
1261 if ( preg_match( $e2, $s, $m ) ) {
1262 $first_prefix = $m[2];
1264 $first_prefix = false;
1270 $selflink = $this->mTitle
->getPrefixedText();
1271 wfProfileOut( $fname.'-setup' );
1273 $checkVariantLink = sizeof($wgContLang->getVariants())>1;
1274 $useSubpages = $this->areSubpagesAllowed();
1276 # Loop for each link
1277 for ($k = 0; isset( $a[$k] ); $k++
) {
1279 if ( $useLinkPrefixExtension ) {
1280 wfProfileIn( $fname.'-prefixhandling' );
1281 if ( preg_match( $e2, $s, $m ) ) {
1289 $prefix = $first_prefix;
1290 $first_prefix = false;
1292 wfProfileOut( $fname.'-prefixhandling' );
1295 $might_be_img = false;
1297 if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
1299 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
1300 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
1301 # the real problem is with the $e1 regex
1304 # Still some problems for cases where the ] is meant to be outside punctuation,
1305 # and no image is in sight. See bug 2095.
1307 if( $text !== '' && preg_match( "/^\](.*)/s", $m[3], $n ) ) {
1308 $text .= ']'; # so that replaceExternalLinks($text) works later
1311 # fix up urlencoded title texts
1312 if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
1314 } elseif( preg_match($e1_img, $line, $m) ) { # Invalid, but might be an image with a link in its caption
1315 $might_be_img = true;
1317 if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
1319 } else { # Invalid form; output directly
1320 $s .= $prefix . '[[' . $line ;
1324 # Don't allow internal links to pages containing
1325 # PROTO: where PROTO is a valid URL protocol; these
1326 # should be external links.
1327 if (preg_match('/^(\b(?:' . wfUrlProtocols() . '))/', $m[1])) {
1328 $s .= $prefix . '[[' . $line ;
1332 # Make subpage if necessary
1333 if( $useSubpages ) {
1334 $link = $this->maybeDoSubpageLink( $m[1], $text );
1339 $noforce = (substr($m[1], 0, 1) != ':');
1341 # Strip off leading ':'
1342 $link = substr($link, 1);
1345 $nt = Title
::newFromText( $this->unstripNoWiki($link, $this->mStripState
) );
1347 $s .= $prefix . '[[' . $line;
1351 #check other language variants of the link
1352 #if the article does not exist
1353 if( $checkVariantLink
1354 && $nt->getArticleID() == 0 ) {
1355 $wgContLang->findVariantLink($link, $nt);
1358 $ns = $nt->getNamespace();
1359 $iw = $nt->getInterWiki();
1361 if ($might_be_img) { # if this is actually an invalid link
1362 if ($ns == NS_IMAGE
&& $noforce) { #but might be an image
1364 while (isset ($a[$k+
1]) ) {
1365 #look at the next 'line' to see if we can close it there
1366 $spliced = array_splice( $a, $k +
1, 1 );
1367 $next_line = array_shift( $spliced );
1368 if( preg_match("/^(.*?]].*?)]](.*)$/sD", $next_line, $m) ) {
1369 # the first ]] closes the inner link, the second the image
1371 $text .= '[[' . $m[1];
1374 } elseif( preg_match("/^.*?]].*$/sD", $next_line, $m) ) {
1375 #if there's exactly one ]] that's fine, we'll keep looking
1376 $text .= '[[' . $m[0];
1378 #if $next_line is invalid too, we need look no further
1379 $text .= '[[' . $next_line;
1384 # we couldn't find the end of this imageLink, so output it raw
1385 #but don't ignore what might be perfectly normal links in the text we've examined
1386 $text = $this->replaceInternalLinks($text);
1387 $s .= $prefix . '[[' . $link . '|' . $text;
1388 # note: no $trail, because without an end, there *is* no trail
1391 } else { #it's not an image, so output it raw
1392 $s .= $prefix . '[[' . $link . '|' . $text;
1393 # note: no $trail, because without an end, there *is* no trail
1398 $wasblank = ( '' == $text );
1399 if( $wasblank ) $text = $link;
1402 # Link not escaped by : , create the various objects
1406 if( $iw && $this->mOptions
->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) {
1407 $this->mOutput
->addLanguageLink( $nt->getFullText() );
1408 $s = rtrim($s . "\n");
1409 $s .= trim($prefix . $trail, "\n") == '' ?
'': $prefix . $trail;
1413 if ( $ns == NS_IMAGE
) {
1414 wfProfileIn( "$fname-image" );
1415 if ( !wfIsBadImage( $nt->getDBkey() ) ) {
1416 # recursively parse links inside the image caption
1417 # actually, this will parse them in any other parameters, too,
1418 # but it might be hard to fix that, and it doesn't matter ATM
1419 $text = $this->replaceExternalLinks($text);
1420 $text = $this->replaceInternalLinks($text);
1422 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
1423 $s .= $prefix . $this->armorLinks( $this->makeImage( $nt, $text ) ) . $trail;
1424 $this->mOutput
->addImage( $nt->getDBkey() );
1426 wfProfileOut( "$fname-image" );
1429 wfProfileOut( "$fname-image" );
1433 if ( $ns == NS_CATEGORY
) {
1434 wfProfileIn( "$fname-category" );
1435 $s = rtrim($s . "\n"); # bug 87
1438 if ( $this->mTitle
->getNamespace() == NS_CATEGORY
) {
1439 $sortkey = $this->mTitle
->getText();
1441 $sortkey = $this->mTitle
->getPrefixedText();
1446 $sortkey = Sanitizer
::decodeCharReferences( $sortkey );
1447 $sortkey = $wgContLang->convertCategoryKey( $sortkey );
1448 $this->mOutput
->addCategory( $nt->getDBkey(), $sortkey );
1451 * Strip the whitespace Category links produce, see bug 87
1452 * @todo We might want to use trim($tmp, "\n") here.
1454 $s .= trim($prefix . $trail, "\n") == '' ?
'': $prefix . $trail;
1456 wfProfileOut( "$fname-category" );
1461 if( ( $nt->getPrefixedText() === $selflink ) &&
1462 ( $nt->getFragment() === '' ) ) {
1463 # Self-links are handled specially; generally de-link and change to bold.
1464 $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail );
1468 # Special and Media are pseudo-namespaces; no pages actually exist in them
1469 if( $ns == NS_MEDIA
) {
1470 $link = $sk->makeMediaLinkObj( $nt, $text );
1471 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
1472 $s .= $prefix . $this->armorLinks( $link ) . $trail;
1473 $this->mOutput
->addImage( $nt->getDBkey() );
1475 } elseif( $ns == NS_SPECIAL
) {
1476 $s .= $this->makeKnownLinkHolder( $nt, $text, '', $trail, $prefix );
1478 } elseif( $ns == NS_IMAGE
) {
1479 $img = Image
::newFromTitle( $nt );
1480 if( $img->exists() ) {
1481 // Force a blue link if the file exists; may be a remote
1482 // upload on the shared repository, and we want to see its
1483 // auto-generated page.
1484 $s .= $this->makeKnownLinkHolder( $nt, $text, '', $trail, $prefix );
1488 $s .= $this->makeLinkHolder( $nt, $text, '', $trail, $prefix );
1490 wfProfileOut( $fname );
1495 * Make a link placeholder. The text returned can be later resolved to a real link with
1496 * replaceLinkHolders(). This is done for two reasons: firstly to avoid further
1497 * parsing of interwiki links, and secondly to allow all extistence checks and
1498 * article length checks (for stub links) to be bundled into a single query.
1501 function makeLinkHolder( &$nt, $text = '', $query = '', $trail = '', $prefix = '' ) {
1502 if ( ! is_object($nt) ) {
1504 $retVal = "<!-- ERROR -->{$prefix}{$text}{$trail}";
1506 # Separate the link trail from the rest of the link
1507 list( $inside, $trail ) = Linker
::splitTrail( $trail );
1509 if ( $nt->isExternal() ) {
1510 $nr = array_push( $this->mInterwikiLinkHolders
['texts'], $prefix.$text.$inside );
1511 $this->mInterwikiLinkHolders
['titles'][] = $nt;
1512 $retVal = '<!--IWLINK '. ($nr-1) ."-->{$trail}";
1514 $nr = array_push( $this->mLinkHolders
['namespaces'], $nt->getNamespace() );
1515 $this->mLinkHolders
['dbkeys'][] = $nt->getDBkey();
1516 $this->mLinkHolders
['queries'][] = $query;
1517 $this->mLinkHolders
['texts'][] = $prefix.$text.$inside;
1518 $this->mLinkHolders
['titles'][] = $nt;
1520 $retVal = '<!--LINK '. ($nr-1) ."-->{$trail}";
1527 * Render a forced-blue link inline; protect against double expansion of
1528 * URLs if we're in a mode that prepends full URL prefixes to internal links.
1529 * Since this little disaster has to split off the trail text to avoid
1530 * breaking URLs in the following text without breaking trails on the
1531 * wiki links, it's been made into a horrible function.
1534 * @param string $text
1535 * @param string $query
1536 * @param string $trail
1537 * @param string $prefix
1538 * @return string HTML-wikitext mix oh yuck
1540 function makeKnownLinkHolder( $nt, $text = '', $query = '', $trail = '', $prefix = '' ) {
1541 list( $inside, $trail ) = Linker
::splitTrail( $trail );
1542 $sk =& $this->mOptions
->getSkin();
1543 $link = $sk->makeKnownLinkObj( $nt, $text, $query, $inside, $prefix );
1544 return $this->armorLinks( $link ) . $trail;
1548 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
1549 * going to go through further parsing steps before inline URL expansion.
1551 * In particular this is important when using action=render, which causes
1552 * full URLs to be included.
1554 * Oh man I hate our multi-layer parser!
1556 * @param string more-or-less HTML
1557 * @return string less-or-more HTML with NOPARSE bits
1559 function armorLinks( $text ) {
1560 return preg_replace( "/\b(" . wfUrlProtocols() . ')/',
1561 "{$this->mUniqPrefix}NOPARSE$1", $text );
1565 * Return true if subpage links should be expanded on this page.
1568 function areSubpagesAllowed() {
1569 # Some namespaces don't allow subpages
1570 global $wgNamespacesWithSubpages;
1571 return !empty($wgNamespacesWithSubpages[$this->mTitle
->getNamespace()]);
1575 * Handle link to subpage if necessary
1576 * @param string $target the source of the link
1577 * @param string &$text the link text, modified as necessary
1578 * @return string the full name of the link
1581 function maybeDoSubpageLink($target, &$text) {
1584 # :Foobar -- override special treatment of prefix (images, language links)
1585 # /Foobar -- convert to CurrentPage/Foobar
1586 # /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text
1587 # ../ -- convert to CurrentPage, from CurrentPage/CurrentSubPage
1588 # ../Foobar -- convert to CurrentPage/Foobar, from CurrentPage/CurrentSubPage
1590 $fname = 'Parser::maybeDoSubpageLink';
1591 wfProfileIn( $fname );
1592 $ret = $target; # default return value is no change
1594 # Some namespaces don't allow subpages,
1595 # so only perform processing if subpages are allowed
1596 if( $this->areSubpagesAllowed() ) {
1597 # Look at the first character
1598 if( $target != '' && $target{0} == '/' ) {
1599 # / at end means we don't want the slash to be shown
1600 if( substr( $target, -1, 1 ) == '/' ) {
1601 $target = substr( $target, 1, -1 );
1604 $noslash = substr( $target, 1 );
1607 $ret = $this->mTitle
->getPrefixedText(). '/' . trim($noslash);
1608 if( '' === $text ) {
1610 } # this might be changed for ugliness reasons
1612 # check for .. subpage backlinks
1614 $nodotdot = $target;
1615 while( strncmp( $nodotdot, "../", 3 ) == 0 ) {
1617 $nodotdot = substr( $nodotdot, 3 );
1619 if($dotdotcount > 0) {
1620 $exploded = explode( '/', $this->mTitle
->GetPrefixedText() );
1621 if( count( $exploded ) > $dotdotcount ) { # not allowed to go below top level page
1622 $ret = implode( '/', array_slice( $exploded, 0, -$dotdotcount ) );
1623 # / at the end means don't show full path
1624 if( substr( $nodotdot, -1, 1 ) == '/' ) {
1625 $nodotdot = substr( $nodotdot, 0, -1 );
1626 if( '' === $text ) {
1630 $nodotdot = trim( $nodotdot );
1631 if( $nodotdot != '' ) {
1632 $ret .= '/' . $nodotdot;
1639 wfProfileOut( $fname );
1644 * Used by doBlockLevels()
1647 /* private */ function closeParagraph() {
1649 if ( '' != $this->mLastSection
) {
1650 $result = '</' . $this->mLastSection
. ">\n";
1652 $this->mInPre
= false;
1653 $this->mLastSection
= '';
1656 # getCommon() returns the length of the longest common substring
1657 # of both arguments, starting at the beginning of both.
1659 /* private */ function getCommon( $st1, $st2 ) {
1660 $fl = strlen( $st1 );
1661 $shorter = strlen( $st2 );
1662 if ( $fl < $shorter ) { $shorter = $fl; }
1664 for ( $i = 0; $i < $shorter; ++
$i ) {
1665 if ( $st1{$i} != $st2{$i} ) { break; }
1669 # These next three functions open, continue, and close the list
1670 # element appropriate to the prefix character passed into them.
1672 /* private */ function openList( $char ) {
1673 $result = $this->closeParagraph();
1675 if ( '*' == $char ) { $result .= '<ul><li>'; }
1676 else if ( '#' == $char ) { $result .= '<ol><li>'; }
1677 else if ( ':' == $char ) { $result .= '<dl><dd>'; }
1678 else if ( ';' == $char ) {
1679 $result .= '<dl><dt>';
1680 $this->mDTopen
= true;
1682 else { $result = '<!-- ERR 1 -->'; }
1687 /* private */ function nextItem( $char ) {
1688 if ( '*' == $char ||
'#' == $char ) { return '</li><li>'; }
1689 else if ( ':' == $char ||
';' == $char ) {
1691 if ( $this->mDTopen
) { $close = '</dt>'; }
1692 if ( ';' == $char ) {
1693 $this->mDTopen
= true;
1694 return $close . '<dt>';
1696 $this->mDTopen
= false;
1697 return $close . '<dd>';
1700 return '<!-- ERR 2 -->';
1703 /* private */ function closeList( $char ) {
1704 if ( '*' == $char ) { $text = '</li></ul>'; }
1705 else if ( '#' == $char ) { $text = '</li></ol>'; }
1706 else if ( ':' == $char ) {
1707 if ( $this->mDTopen
) {
1708 $this->mDTopen
= false;
1709 $text = '</dt></dl>';
1711 $text = '</dd></dl>';
1714 else { return '<!-- ERR 3 -->'; }
1720 * Make lists from lines starting with ':', '*', '#', etc.
1723 * @return string the lists rendered as HTML
1725 function doBlockLevels( $text, $linestart ) {
1726 $fname = 'Parser::doBlockLevels';
1727 wfProfileIn( $fname );
1729 # Parsing through the text line by line. The main thing
1730 # happening here is handling of block-level elements p, pre,
1731 # and making lists from lines starting with * # : etc.
1733 $textLines = explode( "\n", $text );
1735 $lastPrefix = $output = '';
1736 $this->mDTopen
= $inBlockElem = false;
1738 $paragraphStack = false;
1740 if ( !$linestart ) {
1741 $output .= array_shift( $textLines );
1743 foreach ( $textLines as $oLine ) {
1744 $lastPrefixLength = strlen( $lastPrefix );
1745 $preCloseMatch = preg_match('/<\\/pre/i', $oLine );
1746 $preOpenMatch = preg_match('/<pre/i', $oLine );
1747 if ( !$this->mInPre
) {
1748 # Multiple prefixes may abut each other for nested lists.
1749 $prefixLength = strspn( $oLine, '*#:;' );
1750 $pref = substr( $oLine, 0, $prefixLength );
1753 $pref2 = str_replace( ';', ':', $pref );
1754 $t = substr( $oLine, $prefixLength );
1755 $this->mInPre
= !empty($preOpenMatch);
1757 # Don't interpret any other prefixes in preformatted text
1759 $pref = $pref2 = '';
1764 if( $prefixLength && 0 == strcmp( $lastPrefix, $pref2 ) ) {
1765 # Same as the last item, so no need to deal with nesting or opening stuff
1766 $output .= $this->nextItem( substr( $pref, -1 ) );
1767 $paragraphStack = false;
1769 if ( substr( $pref, -1 ) == ';') {
1770 # The one nasty exception: definition lists work like this:
1771 # ; title : definition text
1772 # So we check for : in the remainder text to split up the
1773 # title and definition, without b0rking links.
1775 if ($this->findColonNoLinks($t, $term, $t2) !== false) {
1777 $output .= $term . $this->nextItem( ':' );
1780 } elseif( $prefixLength ||
$lastPrefixLength ) {
1781 # Either open or close a level...
1782 $commonPrefixLength = $this->getCommon( $pref, $lastPrefix );
1783 $paragraphStack = false;
1785 while( $commonPrefixLength < $lastPrefixLength ) {
1786 $output .= $this->closeList( $lastPrefix{$lastPrefixLength-1} );
1787 --$lastPrefixLength;
1789 if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) {
1790 $output .= $this->nextItem( $pref{$commonPrefixLength-1} );
1792 while ( $prefixLength > $commonPrefixLength ) {
1793 $char = substr( $pref, $commonPrefixLength, 1 );
1794 $output .= $this->openList( $char );
1796 if ( ';' == $char ) {
1797 # FIXME: This is dupe of code above
1798 if ($this->findColonNoLinks($t, $term, $t2) !== false) {
1800 $output .= $term . $this->nextItem( ':' );
1803 ++
$commonPrefixLength;
1805 $lastPrefix = $pref2;
1807 if( 0 == $prefixLength ) {
1808 wfProfileIn( "$fname-paragraph" );
1809 # No prefix (not in list)--go to paragraph mode
1810 // XXX: use a stack for nestable elements like span, table and div
1811 $openmatch = preg_match('/(<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
1812 $closematch = preg_match(
1813 '/(<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|'.
1814 '<td|<th|<div|<\\/div|<hr|<\\/pre|<\\/p|'.$this->mUniqPrefix
.'-pre|<\\/li|<\\/ul)/iS', $t );
1815 if ( $openmatch or $closematch ) {
1816 $paragraphStack = false;
1817 $output .= $this->closeParagraph();
1818 if ( $preOpenMatch and !$preCloseMatch ) {
1819 $this->mInPre
= true;
1821 if ( $closematch ) {
1822 $inBlockElem = false;
1824 $inBlockElem = true;
1826 } else if ( !$inBlockElem && !$this->mInPre
) {
1827 if ( ' ' == $t{0} and ( $this->mLastSection
== 'pre' or trim($t) != '' ) ) {
1829 if ($this->mLastSection
!= 'pre') {
1830 $paragraphStack = false;
1831 $output .= $this->closeParagraph().'<pre>';
1832 $this->mLastSection
= 'pre';
1834 $t = substr( $t, 1 );
1837 if ( '' == trim($t) ) {
1838 if ( $paragraphStack ) {
1839 $output .= $paragraphStack.'<br />';
1840 $paragraphStack = false;
1841 $this->mLastSection
= 'p';
1843 if ($this->mLastSection
!= 'p' ) {
1844 $output .= $this->closeParagraph();
1845 $this->mLastSection
= '';
1846 $paragraphStack = '<p>';
1848 $paragraphStack = '</p><p>';
1852 if ( $paragraphStack ) {
1853 $output .= $paragraphStack;
1854 $paragraphStack = false;
1855 $this->mLastSection
= 'p';
1856 } else if ($this->mLastSection
!= 'p') {
1857 $output .= $this->closeParagraph().'<p>';
1858 $this->mLastSection
= 'p';
1863 wfProfileOut( "$fname-paragraph" );
1865 // somewhere above we forget to get out of pre block (bug 785)
1866 if($preCloseMatch && $this->mInPre
) {
1867 $this->mInPre
= false;
1869 if ($paragraphStack === false) {
1873 while ( $prefixLength ) {
1874 $output .= $this->closeList( $pref2{$prefixLength-1} );
1877 if ( '' != $this->mLastSection
) {
1878 $output .= '</' . $this->mLastSection
. '>';
1879 $this->mLastSection
= '';
1882 wfProfileOut( $fname );
1887 * Split up a string on ':', ignoring any occurences inside
1888 * <a>..</a> or <span>...</span>
1889 * @param string $str the string to split
1890 * @param string &$before set to everything before the ':'
1891 * @param string &$after set to everything after the ':'
1892 * return string the position of the ':', or false if none found
1894 function findColonNoLinks($str, &$before, &$after) {
1895 # I wonder if we should make this count all tags, not just <a>
1896 # and <span>. That would prevent us from matching a ':' that
1897 # comes in the middle of italics other such formatting....
1899 $fname = 'Parser::findColonNoLinks';
1900 wfProfileIn( $fname );
1903 $colon = strpos($str, ':', $pos);
1905 if ($colon !== false) {
1906 $before = substr($str, 0, $colon);
1907 $after = substr($str, $colon +
1);
1909 # Skip any ':' within <a> or <span> pairs
1910 $a = substr_count($before, '<a');
1911 $s = substr_count($before, '<span');
1912 $ca = substr_count($before, '</a>');
1913 $cs = substr_count($before, '</span>');
1915 if ($a <= $ca and $s <= $cs) {
1916 # Tags are balanced before ':'; ok
1921 } while ($colon !== false);
1922 wfProfileOut( $fname );
1927 * Return value of a magic variable (like PAGENAME)
1931 function getVariableValue( $index ) {
1932 global $wgContLang, $wgSitename, $wgServer, $wgServerName, $wgScriptPath;
1935 * Some of these require message or data lookups and can be
1936 * expensive to check many times.
1938 static $varCache = array();
1939 if ( wfRunHooks( 'ParserGetVariableValueVarCache', array( &$this, &$varCache ) ) )
1940 if ( isset( $varCache[$index] ) )
1941 return $varCache[$index];
1944 wfRunHooks( 'ParserGetVariableValueTs', array( &$this, &$ts ) );
1947 case MAG_CURRENTMONTH
:
1948 return $varCache[$index] = $wgContLang->formatNum( date( 'm', $ts ) );
1949 case MAG_CURRENTMONTHNAME
:
1950 return $varCache[$index] = $wgContLang->getMonthName( date( 'n', $ts ) );
1951 case MAG_CURRENTMONTHNAMEGEN
:
1952 return $varCache[$index] = $wgContLang->getMonthNameGen( date( 'n', $ts ) );
1953 case MAG_CURRENTMONTHABBREV
:
1954 return $varCache[$index] = $wgContLang->getMonthAbbreviation( date( 'n', $ts ) );
1955 case MAG_CURRENTDAY
:
1956 return $varCache[$index] = $wgContLang->formatNum( date( 'j', $ts ) );
1957 case MAG_CURRENTDAY2
:
1958 return $varCache[$index] = $wgContLang->formatNum( date( 'd', $ts ) );
1960 return $this->mTitle
->getText();
1962 return $this->mTitle
->getPartialURL();
1963 case MAG_FULLPAGENAME
:
1964 return $this->mTitle
->getPrefixedText();
1965 case MAG_FULLPAGENAMEE
:
1966 return $this->mTitle
->getPrefixedURL();
1967 case MAG_REVISIONID
:
1968 return $this->mRevisionId
;
1970 return $wgContLang->getNsText( $this->mTitle
->getNamespace() );
1971 case MAG_NAMESPACEE
:
1972 return wfUrlencode( $wgContLang->getNsText( $this->mTitle
->getNamespace() ) );
1973 case MAG_CURRENTDAYNAME
:
1974 return $varCache[$index] = $wgContLang->getWeekdayName( date( 'w', $ts ) +
1 );
1975 case MAG_CURRENTYEAR
:
1976 return $varCache[$index] = $wgContLang->formatNum( date( 'Y', $ts ), true );
1977 case MAG_CURRENTTIME
:
1978 return $varCache[$index] = $wgContLang->time( wfTimestamp( TS_MW
, $ts ), false, false );
1979 case MAG_CURRENTWEEK
:
1980 // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
1981 // int to remove the padding
1982 return $varCache[$index] = $wgContLang->formatNum( (int)date( 'W', $ts ) );
1983 case MAG_CURRENTDOW
:
1984 return $varCache[$index] = $wgContLang->formatNum( date( 'w', $ts ) );
1985 case MAG_NUMBEROFARTICLES
:
1986 return $varCache[$index] = $wgContLang->formatNum( wfNumberOfArticles() );
1987 case MAG_NUMBEROFFILES
:
1988 return $varCache[$index] = $wgContLang->formatNum( wfNumberOfFiles() );
1993 case MAG_SERVERNAME
:
1994 return $wgServerName;
1995 case MAG_SCRIPTPATH
:
1996 return $wgScriptPath;
1999 if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$varCache, &$index, &$ret ) ) )
2007 * initialise the magic variables (like CURRENTMONTHNAME)
2011 function initialiseVariables() {
2012 $fname = 'Parser::initialiseVariables';
2013 wfProfileIn( $fname );
2014 global $wgVariableIDs;
2015 $this->mVariables
= array();
2016 foreach ( $wgVariableIDs as $id ) {
2017 $mw =& MagicWord
::get( $id );
2018 $mw->addToArray( $this->mVariables
, $id );
2020 wfProfileOut( $fname );
2024 * parse any parentheses in format ((title|part|part))
2025 * and call callbacks to get a replacement text for any found piece
2027 * @param string $text The text to parse
2028 * @param array $callbacks rules in form:
2029 * '{' => array( # opening parentheses
2030 * 'end' => '}', # closing parentheses
2031 * 'cb' => array(2 => callback, # replacement callback to call if {{..}} is found
2032 * 4 => callback # replacement callback to call if {{{{..}}}} is found
2037 function replace_callback ($text, $callbacks) {
2038 $openingBraceStack = array(); # this array will hold a stack of parentheses which are not closed yet
2039 $lastOpeningBrace = -1; # last not closed parentheses
2041 for ($i = 0; $i < strlen($text); $i++
) {
2042 # check for any opening brace
2045 foreach ($callbacks as $key => $value) {
2046 $pos = strpos ($text, $key, $i);
2047 if (false !== $pos && (-1 == $nextPos ||
$pos < $nextPos)) {
2053 if ($lastOpeningBrace >= 0) {
2054 $pos = strpos ($text, $openingBraceStack[$lastOpeningBrace]['braceEnd'], $i);
2056 if (false !== $pos && (-1 == $nextPos ||
$pos < $nextPos)){
2061 $pos = strpos ($text, '|', $i);
2063 if (false !== $pos && (-1 == $nextPos ||
$pos < $nextPos)){
2074 # found openning brace, lets add it to parentheses stack
2075 if (null != $rule) {
2076 $piece = array('brace' => $text[$i],
2077 'braceEnd' => $rule['end'],
2082 # count openning brace characters
2083 while ($i+
1 < strlen($text) && $text[$i+
1] == $piece['brace']) {
2088 $piece['startAt'] = $i+
1;
2089 $piece['partStart'] = $i+
1;
2091 # we need to add to stack only if openning brace count is enough for any given rule
2092 foreach ($rule['cb'] as $cnt => $fn) {
2093 if ($piece['count'] >= $cnt) {
2094 $lastOpeningBrace ++
;
2095 $openingBraceStack[$lastOpeningBrace] = $piece;
2102 else if ($lastOpeningBrace >= 0) {
2103 # first check if it is a closing brace
2104 if ($openingBraceStack[$lastOpeningBrace]['braceEnd'] == $text[$i]) {
2105 # lets check if it is enough characters for closing brace
2107 while ($i+
$count < strlen($text) && $text[$i+
$count] == $text[$i])
2110 # if there are more closing parentheses than opening ones, we parse less
2111 if ($openingBraceStack[$lastOpeningBrace]['count'] < $count)
2112 $count = $openingBraceStack[$lastOpeningBrace]['count'];
2114 # check for maximum matching characters (if there are 5 closing characters, we will probably need only 3 - depending on the rules)
2116 $matchingCallback = null;
2117 foreach ($callbacks[$openingBraceStack[$lastOpeningBrace]['brace']]['cb'] as $cnt => $fn) {
2118 if ($count >= $cnt && $matchingCount < $cnt) {
2119 $matchingCount = $cnt;
2120 $matchingCallback = $fn;
2124 if ($matchingCount == 0) {
2129 # lets set a title or last part (if '|' was found)
2130 if (null === $openingBraceStack[$lastOpeningBrace]['parts'])
2131 $openingBraceStack[$lastOpeningBrace]['title'] = substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], $i - $openingBraceStack[$lastOpeningBrace]['partStart']);
2133 $openingBraceStack[$lastOpeningBrace]['parts'][] = substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], $i - $openingBraceStack[$lastOpeningBrace]['partStart']);
2135 $pieceStart = $openingBraceStack[$lastOpeningBrace]['startAt'] - $matchingCount;
2136 $pieceEnd = $i +
$matchingCount;
2138 if( is_callable( $matchingCallback ) ) {
2140 'text' => substr($text, $pieceStart, $pieceEnd - $pieceStart),
2141 'title' => trim($openingBraceStack[$lastOpeningBrace]['title']),
2142 'parts' => $openingBraceStack[$lastOpeningBrace]['parts'],
2143 'lineStart' => (($pieceStart > 0) && ($text[$pieceStart-1] == '\n')),
2145 # finally we can call a user callback and replace piece of text
2146 $replaceWith = call_user_func( $matchingCallback, $cbArgs );
2147 $text = substr($text, 0, $pieceStart) . $replaceWith . substr($text, $pieceEnd);
2148 $i = $pieceStart +
strlen($replaceWith) - 1;
2151 # null value for callback means that parentheses should be parsed, but not replaced
2152 $i +
= $matchingCount - 1;
2155 # reset last openning parentheses, but keep it in case there are unused characters
2156 $piece = array('brace' => $openingBraceStack[$lastOpeningBrace]['brace'],
2157 'braceEnd' => $openingBraceStack[$lastOpeningBrace]['braceEnd'],
2158 'count' => $openingBraceStack[$lastOpeningBrace]['count'],
2161 'startAt' => $openingBraceStack[$lastOpeningBrace]['startAt']);
2162 $openingBraceStack[$lastOpeningBrace--] = null;
2164 if ($matchingCount < $piece['count']) {
2165 $piece['count'] -= $matchingCount;
2166 $piece['startAt'] -= $matchingCount;
2167 $piece['partStart'] = $piece['startAt'];
2168 # do we still qualify for any callback with remaining count?
2169 foreach ($callbacks[$piece['brace']]['cb'] as $cnt => $fn) {
2170 if ($piece['count'] >= $cnt) {
2171 $lastOpeningBrace ++
;
2172 $openingBraceStack[$lastOpeningBrace] = $piece;
2180 # lets set a title if it is a first separator, or next part otherwise
2181 if ($text[$i] == '|') {
2182 if (null === $openingBraceStack[$lastOpeningBrace]['parts']) {
2183 $openingBraceStack[$lastOpeningBrace]['title'] = substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], $i - $openingBraceStack[$lastOpeningBrace]['partStart']);
2184 $openingBraceStack[$lastOpeningBrace]['parts'] = array();
2187 $openingBraceStack[$lastOpeningBrace]['parts'][] = substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], $i - $openingBraceStack[$lastOpeningBrace]['partStart']);
2189 $openingBraceStack[$lastOpeningBrace]['partStart'] = $i +
1;
2198 * Replace magic variables, templates, and template arguments
2199 * with the appropriate text. Templates are substituted recursively,
2200 * taking care to avoid infinite loops.
2202 * Note that the substitution depends on value of $mOutputType:
2203 * OT_WIKI: only {{subst:}} templates
2204 * OT_MSG: only magic variables
2205 * OT_HTML: all templates and magic variables
2207 * @param string $tex The text to transform
2208 * @param array $args Key-value pairs representing template parameters to substitute
2211 function replaceVariables( $text, $args = array() ) {
2212 # Prevent too big inclusions
2213 if( strlen( $text ) > MAX_INCLUDE_SIZE
) {
2217 $fname = 'Parser::replaceVariables';
2218 wfProfileIn( $fname );
2220 # This function is called recursively. To keep track of arguments we need a stack:
2221 array_push( $this->mArgStack
, $args );
2223 $braceCallbacks = array();
2224 $braceCallbacks[2] = array( &$this, 'braceSubstitution' );
2225 if ( $this->mOutputType
== OT_HTML ||
$this->mOutputType
== OT_WIKI
) {
2226 $braceCallbacks[3] = array( &$this, 'argSubstitution' );
2228 $callbacks = array();
2229 $callbacks['{'] = array('end' => '}', 'cb' => $braceCallbacks);
2230 $callbacks['['] = array('end' => ']', 'cb' => array(2=>null));
2231 $text = $this->replace_callback ($text, $callbacks);
2233 array_pop( $this->mArgStack
);
2235 wfProfileOut( $fname );
2240 * Replace magic variables
2243 function variableSubstitution( $matches ) {
2244 $fname = 'Parser::variableSubstitution';
2245 $varname = $matches[1];
2246 wfProfileIn( $fname );
2247 if ( !$this->mVariables
) {
2248 $this->initialiseVariables();
2251 if ( $this->mOutputType
== OT_WIKI
) {
2252 # Do only magic variables prefixed by SUBST
2253 $mwSubst =& MagicWord
::get( MAG_SUBST
);
2254 if (!$mwSubst->matchStartAndRemove( $varname ))
2256 # Note that if we don't substitute the variable below,
2257 # we don't remove the {{subst:}} magic word, in case
2258 # it is a template rather than a magic variable.
2260 if ( !$skip && array_key_exists( $varname, $this->mVariables
) ) {
2261 $id = $this->mVariables
[$varname];
2262 $text = $this->getVariableValue( $id );
2263 $this->mOutput
->mContainsOldMagic
= true;
2265 $text = $matches[0];
2267 wfProfileOut( $fname );
2271 # Split template arguments
2272 function getTemplateArgs( $argsString ) {
2273 if ( $argsString === '' ) {
2277 $args = explode( '|', substr( $argsString, 1 ) );
2279 # If any of the arguments contains a '[[' but no ']]', it needs to be
2280 # merged with the next arg because the '|' character between belongs
2281 # to the link syntax and not the template parameter syntax.
2282 $argc = count($args);
2284 for ( $i = 0; $i < $argc-1; $i++
) {
2285 if ( substr_count ( $args[$i], '[[' ) != substr_count ( $args[$i], ']]' ) ) {
2286 $args[$i] .= '|'.$args[$i+
1];
2287 array_splice($args, $i+
1, 1);
2297 * Return the text of a template, after recursively
2298 * replacing any variables or templates within the template.
2300 * @param array $piece The parts of the template
2301 * $piece['text']: matched text
2302 * $piece['title']: the title, i.e. the part before the |
2303 * $piece['parts']: the parameter array
2304 * @return string the text of the template
2307 function braceSubstitution( $piece ) {
2309 $fname = 'Parser::braceSubstitution';
2310 wfProfileIn( $fname );
2320 # $part1 is the bit before the first |, and must contain only title characters
2321 # $args is a list of arguments, starting from index 0, not including $part1
2323 $part1 = $piece['title'];
2324 # If the third subpattern matched anything, it will start with |
2326 if (null == $piece['parts']) {
2327 $replaceWith = $this->variableSubstitution (array ($piece['text'], $piece['title']));
2328 if ($replaceWith != $piece['text']) {
2329 $text = $replaceWith;
2335 $args = (null == $piece['parts']) ?
array() : $piece['parts'];
2336 $argc = count( $args );
2340 $mwSubst =& MagicWord
::get( MAG_SUBST
);
2341 if ( $mwSubst->matchStartAndRemove( $part1 ) xor ($this->mOutputType
== OT_WIKI
) ) {
2342 # One of two possibilities is true:
2343 # 1) Found SUBST but not in the PST phase
2344 # 2) Didn't find SUBST and in the PST phase
2345 # In either case, return without further processing
2346 $text = $piece['text'];
2352 # MSG, MSGNW and INT
2355 $mwMsgnw =& MagicWord
::get( MAG_MSGNW
);
2356 if ( $mwMsgnw->matchStartAndRemove( $part1 ) ) {
2359 # Remove obsolete MSG:
2360 $mwMsg =& MagicWord
::get( MAG_MSG
);
2361 $mwMsg->matchStartAndRemove( $part1 );
2364 # Check if it is an internal message
2365 $mwInt =& MagicWord
::get( MAG_INT
);
2366 if ( $mwInt->matchStartAndRemove( $part1 ) ) {
2367 if ( $this->incrementIncludeCount( 'int:'.$part1 ) ) {
2368 $text = $linestart . wfMsgReal( $part1, $args, true );
2376 # Check for NS: (namespace expansion)
2377 $mwNs = MagicWord
::get( MAG_NS
);
2378 if ( $mwNs->matchStartAndRemove( $part1 ) ) {
2379 if ( intval( $part1 ) ) {
2380 $text = $linestart . $wgContLang->getNsText( intval( $part1 ) );
2383 $index = Namespace::getCanonicalIndex( strtolower( $part1 ) );
2384 if ( !is_null( $index ) ) {
2385 $text = $linestart . $wgContLang->getNsText( $index );
2392 # LCFIRST, UCFIRST, LC and UC
2394 $lcfirst =& MagicWord
::get( MAG_LCFIRST
);
2395 $ucfirst =& MagicWord
::get( MAG_UCFIRST
);
2396 $lc =& MagicWord
::get( MAG_LC
);
2397 $uc =& MagicWord
::get( MAG_UC
);
2398 if ( $lcfirst->matchStartAndRemove( $part1 ) ) {
2399 $text = $linestart . $wgContLang->lcfirst( $part1 );
2401 } else if ( $ucfirst->matchStartAndRemove( $part1 ) ) {
2402 $text = $linestart . $wgContLang->ucfirst( $part1 );
2404 } else if ( $lc->matchStartAndRemove( $part1 ) ) {
2405 $text = $linestart . $wgContLang->lc( $part1 );
2407 } else if ( $uc->matchStartAndRemove( $part1 ) ) {
2408 $text = $linestart . $wgContLang->uc( $part1 );
2413 # LOCALURL and FULLURL
2415 $mwLocal =& MagicWord
::get( MAG_LOCALURL
);
2416 $mwLocalE =& MagicWord
::get( MAG_LOCALURLE
);
2417 $mwFull =& MagicWord
::get( MAG_FULLURL
);
2418 $mwFullE =& MagicWord
::get( MAG_FULLURLE
);
2421 if ( $mwLocal->matchStartAndRemove( $part1 ) ) {
2422 $func = 'getLocalURL';
2423 } elseif ( $mwLocalE->matchStartAndRemove( $part1 ) ) {
2424 $func = 'escapeLocalURL';
2425 } elseif ( $mwFull->matchStartAndRemove( $part1 ) ) {
2426 $func = 'getFullURL';
2427 } elseif ( $mwFullE->matchStartAndRemove( $part1 ) ) {
2428 $func = 'escapeFullURL';
2433 if ( $func !== false ) {
2434 $title = Title
::newFromText( $part1 );
2435 if ( !is_null( $title ) ) {
2437 $text = $linestart . $title->$func( $args[0] );
2439 $text = $linestart . $title->$func();
2447 if ( !$found && $argc == 1 ) {
2448 $mwGrammar =& MagicWord
::get( MAG_GRAMMAR
);
2449 if ( $mwGrammar->matchStartAndRemove( $part1 ) ) {
2450 $text = $linestart . $wgContLang->convertGrammar( $args[0], $part1 );
2456 if ( !$found && $argc >= 2 ) {
2457 $mwPluralForm =& MagicWord
::get( MAG_PLURAL
);
2458 if ( $mwPluralForm->matchStartAndRemove( $part1 ) ) {
2459 if ($argc==2) {$args[2]=$args[1];}
2460 $text = $linestart . $wgContLang->convertPlural( $part1, $args[0], $args[1], $args[2]);
2465 # Template table test
2467 # Did we encounter this template already? If yes, it is in the cache
2468 # and we need to check for loops.
2469 if ( !$found && isset( $this->mTemplates
[$part1] ) ) {
2472 # Infinite loop test
2473 if ( isset( $this->mTemplatePath
[$part1] ) ) {
2476 $text = $linestart .
2477 '{{' . $part1 . '}}' .
2478 '<!-- WARNING: template loop detected -->';
2479 wfDebug( "$fname: template loop broken at '$part1'\n" );
2481 # set $text to cached message.
2482 $text = $linestart . $this->mTemplates
[$part1];
2486 # Load from database
2487 $replaceHeadings = false;
2489 $lastPathLevel = $this->mTemplatePath
;
2492 $part1 = $this->maybeDoSubpageLink( $part1, $subpage='' );
2493 if ($subpage !== '') {
2494 $ns = $this->mTitle
->getNamespace();
2496 $title = Title
::newFromText( $part1, $ns );
2499 $interwiki = Title
::getInterwikiLink($title->getInterwiki());
2500 if ($interwiki != '' && $title->isTrans()) {
2501 return $this->scarytransclude($title, $interwiki);
2505 if ( !is_null( $title ) && !$title->isExternal() ) {
2506 # Check for excessive inclusion
2507 $dbk = $title->getPrefixedDBkey();
2508 if ( $this->incrementIncludeCount( $dbk ) ) {
2509 if ( $title->getNamespace() == NS_SPECIAL
&& $this->mOptions
->getAllowSpecialInclusion() ) {
2510 # Capture special page output
2511 $text = SpecialPage
::capturePath( $title );
2512 if ( is_string( $text ) ) {
2516 $this->disableCache();
2519 $articleContent = $this->fetchTemplate( $title );
2520 if ( $articleContent !== false ) {
2522 $text = $articleContent;
2523 $replaceHeadings = true;
2528 # If the title is valid but undisplayable, make a link to it
2529 if ( $this->mOutputType
== OT_HTML
&& !$found ) {
2530 $text = '[['.$title->getPrefixedText().']]';
2534 # Template cache array insertion
2536 $this->mTemplates
[$part1] = $text;
2537 $text = $linestart . $text;
2542 # Recursive parsing, escaping and link table handling
2543 # Only for HTML output
2544 if ( $nowiki && $found && $this->mOutputType
== OT_HTML
) {
2545 $text = wfEscapeWikiText( $text );
2546 } elseif ( ($this->mOutputType
== OT_HTML ||
$this->mOutputType
== OT_WIKI
) && $found && !$noparse) {
2547 # Clean up argument array
2548 $assocArgs = array();
2550 foreach( $args as $arg ) {
2551 $eqpos = strpos( $arg, '=' );
2552 if ( $eqpos === false ) {
2553 $assocArgs[$index++
] = $arg;
2555 $name = trim( substr( $arg, 0, $eqpos ) );
2556 $value = trim( substr( $arg, $eqpos+
1 ) );
2557 if ( $value === false ) {
2560 if ( $name !== false ) {
2561 $assocArgs[$name] = $value;
2566 # Add a new element to the templace recursion path
2567 $this->mTemplatePath
[$part1] = 1;
2569 # If there are any <onlyinclude> tags, only include them
2570 if ( in_string( '<onlyinclude>', $text ) && in_string( '</onlyinclude>', $text ) ) {
2571 preg_match_all( '/<onlyinclude>(.*?)\n?<\/onlyinclude>/s', $text, $m );
2573 foreach ($m[1] as $piece)
2576 # Remove <noinclude> sections and <includeonly> tags
2577 $text = preg_replace( '/<noinclude>.*?<\/noinclude>/s', '', $text );
2578 $text = strtr( $text, array( '<includeonly>' => '' , '</includeonly>' => '' ) );
2580 if( $this->mOutputType
== OT_HTML
) {
2581 # Strip <nowiki>, <pre>, etc.
2582 $text = $this->strip( $text, $this->mStripState
);
2583 $text = Sanitizer
::removeHTMLtags( $text, array( &$this, 'replaceVariables' ), $assocArgs );
2585 $text = $this->replaceVariables( $text, $assocArgs );
2587 # If the template begins with a table or block-level
2588 # element, it should be treated as beginning a new line.
2589 if (!$piece['lineStart'] && preg_match('/^({\\||:|;|#|\*)/', $text)) {
2590 $text = "\n" . $text;
2593 # Prune lower levels off the recursion check path
2594 $this->mTemplatePath
= $lastPathLevel;
2597 wfProfileOut( $fname );
2598 return $piece['text'];
2601 # Replace raw HTML by a placeholder
2602 # Add a blank line preceding, to prevent it from mucking up
2603 # immediately preceding headings
2604 $text = "\n\n" . $this->insertStripItem( $text, $this->mStripState
);
2606 # replace ==section headers==
2607 # XXX this needs to go away once we have a better parser.
2608 if ( $this->mOutputType
!= OT_WIKI
&& $replaceHeadings ) {
2609 if( !is_null( $title ) )
2610 $encodedname = base64_encode($title->getPrefixedDBkey());
2612 $encodedname = base64_encode("");
2613 $m = preg_split('/(^={1,6}.*?={1,6}\s*?$)/m', $text, -1,
2614 PREG_SPLIT_DELIM_CAPTURE
);
2617 for( $i = 0; $i < count($m); $i +
= 2 ) {
2619 if (!isset($m[$i +
1]) ||
$m[$i +
1] == "") continue;
2621 if( strstr($hl, "<!--MWTEMPLATESECTION") ) {
2625 preg_match('/^(={1,6})(.*?)(={1,6})\s*?$/m', $hl, $m2);
2626 $text .= $m2[1] . $m2[2] . "<!--MWTEMPLATESECTION="
2627 . $encodedname . "&" . base64_encode("$nsec") . "-->" . $m2[3];
2635 # Prune lower levels off the recursion check path
2636 $this->mTemplatePath
= $lastPathLevel;
2639 wfProfileOut( $fname );
2640 return $piece['text'];
2642 wfProfileOut( $fname );
2648 * Fetch the unparsed text of a template and register a reference to it.
2650 function fetchTemplate( $title ) {
2652 // Loop to fetch the article, with up to 1 redirect
2653 for ( $i = 0; $i < 2 && is_object( $title ); $i++
) {
2654 $rev = Revision
::newFromTitle( $title );
2655 $this->mOutput
->addTemplate( $title, $title->getArticleID() );
2659 $text = $rev->getText();
2660 if ( $text === false ) {
2664 $title = Title
::newFromRedirect( $text );
2670 * Translude an interwiki link.
2672 function scarytransclude($title, $interwiki) {
2673 global $wgEnableScaryTranscluding;
2675 if (!$wgEnableScaryTranscluding)
2676 return wfMsg('scarytranscludedisabled');
2678 $articlename = "Template:" . $title->getDBkey();
2679 $url = str_replace('$1', urlencode($articlename), $interwiki);
2680 if (strlen($url) > 255)
2681 return wfMsg('scarytranscludetoolong');
2682 $text = $this->fetchScaryTemplateMaybeFromCache($url);
2683 $this->mIWTransData
[] = $text;
2684 return "<!--IW_TRANSCLUDE ".(count($this->mIWTransData
) - 1)."-->";
2687 function fetchScaryTemplateMaybeFromCache($url) {
2688 $dbr =& wfGetDB(DB_SLAVE
);
2689 $obj = $dbr->selectRow('transcache', array('tc_time', 'tc_contents'),
2690 array('tc_url' => $url));
2692 $time = $obj->tc_time
;
2693 $text = $obj->tc_contents
;
2694 if ($time && $time < (time() +
(60*60))) {
2699 $text = wfGetHTTP($url . '?action=render');
2701 return wfMsg('scarytranscludefailed', $url);
2703 $dbw =& wfGetDB(DB_MASTER
);
2704 $dbw->replace('transcache', array(), array(
2706 'tc_time' => time(),
2707 'tc_contents' => $text));
2713 * Triple brace replacement -- used for template arguments
2716 function argSubstitution( $matches ) {
2717 $arg = trim( $matches['title'] );
2718 $text = $matches['text'];
2719 $inputArgs = end( $this->mArgStack
);
2721 if ( array_key_exists( $arg, $inputArgs ) ) {
2722 $text = $inputArgs[$arg];
2723 } else if ($this->mOutputType
== OT_HTML
&& null != $matches['parts'] && count($matches['parts']) > 0) {
2724 $text = $matches['parts'][0];
2731 * Returns true if the function is allowed to include this entity
2734 function incrementIncludeCount( $dbk ) {
2735 if ( !array_key_exists( $dbk, $this->mIncludeCount
) ) {
2736 $this->mIncludeCount
[$dbk] = 0;
2738 if ( ++
$this->mIncludeCount
[$dbk] <= MAX_INCLUDE_REPEAT
) {
2746 * This function accomplishes several tasks:
2747 * 1) Auto-number headings if that option is enabled
2748 * 2) Add an [edit] link to sections for logged in users who have enabled the option
2749 * 3) Add a Table of contents on the top for users who have enabled the option
2750 * 4) Auto-anchor headings
2752 * It loops through all headlines, collects the necessary data, then splits up the
2753 * string and re-inserts the newly formatted headlines.
2755 * @param string $text
2756 * @param boolean $isMain
2759 function formatHeadings( $text, $isMain=true ) {
2760 global $wgMaxTocLevel, $wgContLang;
2762 $doNumberHeadings = $this->mOptions
->getNumberHeadings();
2764 $forceTocHere = false;
2765 if( !$this->mTitle
->userCanEdit() ) {
2768 $showEditLink = $this->mOptions
->getEditSection();
2771 # Inhibit editsection links if requested in the page
2772 $esw =& MagicWord
::get( MAG_NOEDITSECTION
);
2773 if( $esw->matchAndRemove( $text ) ) {
2776 # if the string __NOTOC__ (not case-sensitive) occurs in the HTML,
2778 $mw =& MagicWord
::get( MAG_NOTOC
);
2779 if( $mw->matchAndRemove( $text ) ) {
2783 # Get all headlines for numbering them and adding funky stuff like [edit]
2784 # links - this is for later, but we need the number of headlines right now
2785 $numMatches = preg_match_all( '/<H([1-6])(.*?'.'>)(.*?)<\/H[1-6] *>/i', $text, $matches );
2787 # if there are fewer than 4 headlines in the article, do not show TOC
2788 if( $numMatches < 4 ) {
2792 # if the string __TOC__ (not case-sensitive) occurs in the HTML,
2793 # override above conditions and always show TOC at that place
2795 $mw =& MagicWord
::get( MAG_TOC
);
2796 if($mw->match( $text ) ) {
2798 $forceTocHere = true;
2800 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
2801 # override above conditions and always show TOC above first header
2802 $mw =& MagicWord
::get( MAG_FORCETOC
);
2803 if ($mw->matchAndRemove( $text ) ) {
2808 # Never ever show TOC if no headers
2809 if( $numMatches < 1 ) {
2813 # We need this to perform operations on the HTML
2814 $sk =& $this->mOptions
->getSkin();
2818 $sectionCount = 0; # headlineCount excluding template sections
2820 # Ugh .. the TOC should have neat indentation levels which can be
2821 # passed to the skin functions. These are determined here
2825 $sublevelCount = array();
2826 $levelCount = array();
2833 foreach( $matches[3] as $headline ) {
2835 $templatetitle = '';
2836 $templatesection = 0;
2839 if (preg_match("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", $headline, $mat)) {
2841 $templatetitle = base64_decode($mat[1]);
2842 $templatesection = 1 +
(int)base64_decode($mat[2]);
2843 $headline = preg_replace("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", "", $headline);
2847 $prevlevel = $level;
2848 $prevtoclevel = $toclevel;
2850 $level = $matches[1][$headlineCount];
2852 if( $doNumberHeadings ||
$doShowToc ) {
2854 if ( $level > $prevlevel ) {
2855 # Increase TOC level
2857 $sublevelCount[$toclevel] = 0;
2858 $toc .= $sk->tocIndent();
2860 elseif ( $level < $prevlevel && $toclevel > 1 ) {
2861 # Decrease TOC level, find level to jump to
2863 if ( $toclevel == 2 && $level <= $levelCount[1] ) {
2864 # Can only go down to level 1
2867 for ($i = $toclevel; $i > 0; $i--) {
2868 if ( $levelCount[$i] == $level ) {
2869 # Found last matching level
2873 elseif ( $levelCount[$i] < $level ) {
2874 # Found first matching level below current level
2881 $toc .= $sk->tocUnindent( $prevtoclevel - $toclevel );
2884 # No change in level, end TOC line
2885 $toc .= $sk->tocLineEnd();
2888 $levelCount[$toclevel] = $level;
2890 # count number of headlines for each level
2891 @$sublevelCount[$toclevel]++
;
2893 for( $i = 1; $i <= $toclevel; $i++
) {
2894 if( !empty( $sublevelCount[$i] ) ) {
2898 $numbering .= $wgContLang->formatNum( $sublevelCount[$i] );
2904 # The canonized header is a version of the header text safe to use for links
2905 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
2906 $canonized_headline = $this->unstrip( $headline, $this->mStripState
);
2907 $canonized_headline = $this->unstripNoWiki( $canonized_headline, $this->mStripState
);
2909 # Remove link placeholders by the link text.
2910 # <!--LINK number-->
2912 # link text with suffix
2913 $canonized_headline = preg_replace( '/<!--LINK ([0-9]*)-->/e',
2914 "\$this->mLinkHolders['texts'][\$1]",
2915 $canonized_headline );
2916 $canonized_headline = preg_replace( '/<!--IWLINK ([0-9]*)-->/e',
2917 "\$this->mInterwikiLinkHolders['texts'][\$1]",
2918 $canonized_headline );
2921 $canonized_headline = preg_replace( '/<.*?' . '>/','',$canonized_headline );
2922 $tocline = trim( $canonized_headline );
2923 $canonized_headline = Sanitizer
::escapeId( $tocline );
2924 $refers[$headlineCount] = $canonized_headline;
2926 # count how many in assoc. array so we can track dupes in anchors
2927 @$refers[$canonized_headline]++
;
2928 $refcount[$headlineCount]=$refers[$canonized_headline];
2930 # Don't number the heading if it is the only one (looks silly)
2931 if( $doNumberHeadings && count( $matches[3] ) > 1) {
2932 # the two are different if the line contains a link
2933 $headline=$numbering . ' ' . $headline;
2936 # Create the anchor for linking from the TOC to the section
2937 $anchor = $canonized_headline;
2938 if($refcount[$headlineCount] > 1 ) {
2939 $anchor .= '_' . $refcount[$headlineCount];
2941 if( $doShowToc && ( !isset($wgMaxTocLevel) ||
$toclevel<$wgMaxTocLevel ) ) {
2942 $toc .= $sk->tocLine($anchor, $tocline, $numbering, $toclevel);
2944 if( $showEditLink && ( !$istemplate ||
$templatetitle !== "" ) ) {
2945 if ( empty( $head[$headlineCount] ) ) {
2946 $head[$headlineCount] = '';
2949 $head[$headlineCount] .= $sk->editSectionLinkForOther($templatetitle, $templatesection);
2951 $head[$headlineCount] .= $sk->editSectionLink($this->mTitle
, $sectionCount+
1);
2954 # give headline the correct <h#> tag
2955 @$head[$headlineCount] .= "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount] .$headline.'</h'.$level.'>';
2963 $toc .= $sk->tocUnindent( $toclevel - 1 );
2964 $toc = $sk->tocList( $toc );
2967 # split up and insert constructed headlines
2969 $blocks = preg_split( '/<H[1-6].*?' . '>.*?<\/H[1-6]>/i', $text );
2972 foreach( $blocks as $block ) {
2973 if( $showEditLink && $headlineCount > 0 && $i == 0 && $block != "\n" ) {
2974 # This is the [edit] link that appears for the top block of text when
2975 # section editing is enabled
2977 # Disabled because it broke block formatting
2978 # For example, a bullet point in the top line
2979 # $full .= $sk->editSectionLink(0);
2982 if( $doShowToc && !$i && $isMain && !$forceTocHere) {
2983 # Top anchor now in skin
2987 if( !empty( $head[$i] ) ) {
2993 $mw =& MagicWord
::get( MAG_TOC
);
2994 return $mw->replace( $toc, $full );
3001 * Return an HTML link for the "ISBN 123456" text
3004 function magicISBN( $text ) {
3005 $fname = 'Parser::magicISBN';
3006 wfProfileIn( $fname );
3008 $a = split( 'ISBN ', ' '.$text );
3009 if ( count ( $a ) < 2 ) {
3010 wfProfileOut( $fname );
3013 $text = substr( array_shift( $a ), 1);
3014 $valid = '0123456789-Xx';
3016 foreach ( $a as $x ) {
3017 $isbn = $blank = '' ;
3018 while ( ' ' == $x{0} ) {
3020 $x = substr( $x, 1 );
3022 if ( $x == '' ) { # blank isbn
3023 $text .= "ISBN $blank";
3026 while ( strstr( $valid, $x{0} ) != false ) {
3028 $x = substr( $x, 1 );
3030 $num = str_replace( '-', '', $isbn );
3031 $num = str_replace( ' ', '', $num );
3032 $num = str_replace( 'x', 'X', $num );
3035 $text .= "ISBN $blank$x";
3037 $titleObj = Title
::makeTitle( NS_SPECIAL
, 'Booksources' );
3038 $text .= '<a href="' .
3039 $titleObj->escapeLocalUrl( 'isbn='.$num ) .
3040 "\" class=\"internal\">ISBN $isbn</a>";
3044 wfProfileOut( $fname );
3049 * Return an HTML link for the "RFC 1234" text
3052 * @param string $text Text to be processed
3053 * @param string $keyword Magic keyword to use (default RFC)
3054 * @param string $urlmsg Interface message to use (default rfcurl)
3057 function magicRFC( $text, $keyword='RFC ', $urlmsg='rfcurl' ) {
3059 $valid = '0123456789';
3062 $a = split( $keyword, ' '.$text );
3063 if ( count ( $a ) < 2 ) {
3066 $text = substr( array_shift( $a ), 1);
3068 /* Check if keyword is preceed by [[.
3069 * This test is made here cause of the array_shift above
3070 * that prevent the test to be done in the foreach.
3072 if ( substr( $text, -2 ) == '[[' ) {
3076 foreach ( $a as $x ) {
3077 /* token might be empty if we have RFC RFC 1234 */
3085 /** remove and save whitespaces in $blank */
3086 while ( $x{0} == ' ' ) {
3088 $x = substr( $x, 1 );
3091 /** remove and save the rfc number in $id */
3092 while ( strstr( $valid, $x{0} ) != false ) {
3094 $x = substr( $x, 1 );
3098 /* call back stripped spaces*/
3099 $text .= $keyword.$blank.$x;
3100 } elseif( $internal ) {
3102 $text .= $keyword.$id.$x;
3104 /* build the external link*/
3105 $url = wfMsg( $urlmsg, $id);
3106 $sk =& $this->mOptions
->getSkin();
3107 $la = $sk->getExternalLinkAttributes( $url, $keyword.$id );
3108 $text .= "<a href='{$url}'{$la}>{$keyword}{$id}</a>{$x}";
3111 /* Check if the next RFC keyword is preceed by [[ */
3112 $internal = ( substr($x,-2) == '[[' );
3118 * Transform wiki markup when saving a page by doing \r\n -> \n
3119 * conversion, substitting signatures, {{subst:}} templates, etc.
3121 * @param string $text the text to transform
3122 * @param Title &$title the Title object for the current article
3123 * @param User &$user the User object describing the current user
3124 * @param ParserOptions $options parsing options
3125 * @param bool $clearState whether to clear the parser state first
3126 * @return string the altered wiki markup
3129 function preSaveTransform( $text, &$title, &$user, $options, $clearState = true ) {
3130 $this->mOptions
= $options;
3131 $this->mTitle
=& $title;
3132 $this->mOutputType
= OT_WIKI
;
3134 if ( $clearState ) {
3135 $this->clearState();
3138 $stripState = false;
3142 $text = str_replace( array_keys( $pairs ), array_values( $pairs ), $text );
3143 $text = $this->strip( $text, $stripState, true );
3144 $text = $this->pstPass2( $text, $user );
3145 $text = $this->unstrip( $text, $stripState );
3146 $text = $this->unstripNoWiki( $text, $stripState );
3151 * Pre-save transform helper function
3154 function pstPass2( $text, &$user ) {
3155 global $wgContLang, $wgLocaltimezone;
3157 /* Note: This is the timestamp saved as hardcoded wikitext to
3158 * the database, we use $wgContLang here in order to give
3159 * everyone the same signiture and use the default one rather
3160 * than the one selected in each users preferences.
3162 if ( isset( $wgLocaltimezone ) ) {
3163 $oldtz = getenv( 'TZ' );
3164 putenv( 'TZ='.$wgLocaltimezone );
3166 $d = $wgContLang->timeanddate( date( 'YmdHis' ), false, false) .
3167 ' (' . date( 'T' ) . ')';
3168 if ( isset( $wgLocaltimezone ) ) {
3169 putenv( 'TZ='.$oldtz );
3172 # Variable replacement
3173 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
3174 $text = $this->replaceVariables( $text );
3177 $sigText = $this->getUserSig( $user );
3178 $text = strtr( $text, array(
3180 '~~~~' => "$sigText $d",
3184 # Context links: [[|name]] and [[name (context)|]]
3186 global $wgLegalTitleChars;
3187 $tc = "[$wgLegalTitleChars]";
3188 $np = str_replace( array( '(', ')' ), array( '', '' ), $tc ); # No parens
3190 $namespacechar = '[ _0-9A-Za-z\x80-\xff]'; # Namespaces can use non-ascii!
3191 $conpat = "/^({$np}+) \\(({$tc}+)\\)$/";
3193 $p1 = "/\[\[({$np}+) \\(({$np}+)\\)\\|]]/"; # [[page (context)|]]
3194 $p2 = "/\[\[\\|({$tc}+)]]/"; # [[|page]]
3195 $p3 = "/\[\[(:*$namespacechar+):({$np}+)\\|]]/"; # [[namespace:page|]] and [[:namespace:page|]]
3196 $p4 = "/\[\[(:*$namespacechar+):({$np}+) \\(({$np}+)\\)\\|]]/"; # [[ns:page (cont)|]] and [[:ns:page (cont)|]]
3198 $t = $this->mTitle
->getText();
3199 if ( preg_match( $conpat, $t, $m ) ) {
3202 $text = preg_replace( $p4, '[[\\1:\\2 (\\3)|\\2]]', $text );
3203 $text = preg_replace( $p1, '[[\\1 (\\2)|\\1]]', $text );
3204 $text = preg_replace( $p3, '[[\\1:\\2|\\2]]', $text );
3206 if ( '' == $context ) {
3207 $text = preg_replace( $p2, '[[\\1]]', $text );
3209 $text = preg_replace( $p2, "[[\\1 ({$context})|\\1]]", $text );
3212 # Trim trailing whitespace
3213 # MAG_END (__END__) tag allows for trailing
3214 # whitespace to be deliberately included
3215 $text = rtrim( $text );
3216 $mw =& MagicWord
::get( MAG_END
);
3217 $mw->matchAndRemove( $text );
3223 * Fetch the user's signature text, if any, and normalize to
3224 * validated, ready-to-insert wikitext.
3230 function getUserSig( &$user ) {
3233 $username = $user->getName();
3234 $nickname = $user->getOption( 'nickname' );
3235 $nickname = $nickname === '' ?
$username : $nickname;
3237 if( $user->getBoolOption( 'fancysig' ) !== false ) {
3238 # Sig. might contain markup; validate this
3239 if( $this->validateSig( $nickname ) !== false ) {
3240 # Validated; clean up (if needed) and return it
3241 return( $this->cleanSig( $nickname ) );
3243 # Failed to validate; fall back to the default
3244 $nickname = $username;
3245 wfDebug( "Parser::getUserSig: $username has bad XML tags in signature.\n" );
3249 # If we're still here, make it a link to the user page
3250 $userpage = $user->getUserPage();
3251 return( '[[' . $userpage->getPrefixedText() . '|' . wfEscapeWikiText( $nickname ) . ']]' );
3255 * Check that the user's signature contains no bad XML
3257 * @param string $text
3258 * @return mixed An expanded string, or false if invalid.
3260 function validateSig( $text ) {
3261 return( wfIsWellFormedXmlFragment( $text ) ?
$text : false );
3265 * Clean up signature text
3267 * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures
3268 * 2) Substitute all transclusions
3270 * @param string $text
3271 * @return string Signature text
3273 function cleanSig( $text ) {
3274 $substWord = MagicWord
::get( MAG_SUBST
);
3275 $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
3276 $substText = '{{' . $substWord->getSynonym( 0 );
3278 $text = preg_replace( $substRegex, $substText, $text );
3279 $text = preg_replace( '/~{3,5}/', '', $text );
3280 $text = $this->replaceVariables( $text );
3286 * Set up some variables which are usually set up in parse()
3287 * so that an external function can call some class members with confidence
3290 function startExternalParse( &$title, $options, $outputType, $clearState = true ) {
3291 $this->mTitle
=& $title;
3292 $this->mOptions
= $options;
3293 $this->mOutputType
= $outputType;
3294 if ( $clearState ) {
3295 $this->clearState();
3300 * Transform a MediaWiki message by replacing magic variables.
3302 * @param string $text the text to transform
3303 * @param ParserOptions $options options
3304 * @return string the text with variables substituted
3307 function transformMsg( $text, $options ) {
3309 static $executing = false;
3311 $fname = "Parser::transformMsg";
3313 # Guard against infinite recursion
3319 wfProfileIn($fname);
3321 $this->mTitle
= $wgTitle;
3322 $this->mOptions
= $options;
3323 $this->mOutputType
= OT_MSG
;
3324 $this->clearState();
3325 $text = $this->replaceVariables( $text );
3328 wfProfileOut($fname);
3333 * Create an HTML-style tag, e.g. <yourtag>special text</yourtag>
3334 * Callback will be called with the text within
3335 * Transform and return the text within
3339 * @param mixed $tag The tag to use, e.g. 'hook' for <hook>
3340 * @param mixed $callback The callback function (and object) to use for the tag
3342 * @return The old value of the mTagHooks array associated with the hook
3344 function setHook( $tag, $callback ) {
3345 $oldVal = @$this->mTagHooks
[$tag];
3346 $this->mTagHooks
[$tag] = $callback;
3352 * Replace <!--LINK--> link placeholders with actual links, in the buffer
3353 * Placeholders created in Skin::makeLinkObj()
3354 * Returns an array of links found, indexed by PDBK:
3358 * $options is a bit field, RLH_FOR_UPDATE to select for update
3360 function replaceLinkHolders( &$text, $options = 0 ) {
3362 global $wgOutputReplace;
3364 $fname = 'Parser::replaceLinkHolders';
3365 wfProfileIn( $fname );
3369 $sk =& $this->mOptions
->getSkin();
3370 $linkCache =& LinkCache
::singleton();
3372 if ( !empty( $this->mLinkHolders
['namespaces'] ) ) {
3373 wfProfileIn( $fname.'-check' );
3374 $dbr =& wfGetDB( DB_SLAVE
);
3375 $page = $dbr->tableName( 'page' );
3376 $threshold = $wgUser->getOption('stubthreshold');
3379 asort( $this->mLinkHolders
['namespaces'] );
3383 foreach ( $this->mLinkHolders
['namespaces'] as $key => $ns ) {
3385 $title = $this->mLinkHolders
['titles'][$key];
3387 # Skip invalid entries.
3388 # Result will be ugly, but prevents crash.
3389 if ( is_null( $title ) ) {
3392 $pdbk = $pdbks[$key] = $title->getPrefixedDBkey();
3394 # Check if it's a static known link, e.g. interwiki
3395 if ( $title->isAlwaysKnown() ) {
3396 $colours[$pdbk] = 1;
3397 } elseif ( ( $id = $linkCache->getGoodLinkID( $pdbk ) ) != 0 ) {
3398 $colours[$pdbk] = 1;
3399 $this->mOutput
->addLink( $title, $id );
3400 } elseif ( $linkCache->isBadLink( $pdbk ) ) {
3401 $colours[$pdbk] = 0;
3403 # Not in the link cache, add it to the query
3404 if ( !isset( $current ) ) {
3406 $query = "SELECT page_id, page_namespace, page_title";
3407 if ( $threshold > 0 ) {
3408 $query .= ', page_len, page_is_redirect';
3410 $query .= " FROM $page WHERE (page_namespace=$ns AND page_title IN(";
3411 } elseif ( $current != $ns ) {
3413 $query .= ")) OR (page_namespace=$ns AND page_title IN(";
3418 $query .= $dbr->addQuotes( $this->mLinkHolders
['dbkeys'][$key] );
3423 if ( $options & RLH_FOR_UPDATE
) {
3424 $query .= ' FOR UPDATE';
3427 $res = $dbr->query( $query, $fname );
3429 # Fetch data and form into an associative array
3430 # non-existent = broken
3433 while ( $s = $dbr->fetchObject($res) ) {
3434 $title = Title
::makeTitle( $s->page_namespace
, $s->page_title
);
3435 $pdbk = $title->getPrefixedDBkey();
3436 $linkCache->addGoodLinkObj( $s->page_id
, $title );
3437 $this->mOutput
->addLink( $title, $s->page_id
);
3439 if ( $threshold > 0 ) {
3440 $size = $s->page_len
;
3441 if ( $s->page_is_redirect ||
$s->page_namespace
!= 0 ||
$size >= $threshold ) {
3442 $colours[$pdbk] = 1;
3444 $colours[$pdbk] = 2;
3447 $colours[$pdbk] = 1;
3451 wfProfileOut( $fname.'-check' );
3453 # Construct search and replace arrays
3454 wfProfileIn( $fname.'-construct' );
3455 $wgOutputReplace = array();
3456 foreach ( $this->mLinkHolders
['namespaces'] as $key => $ns ) {
3457 $pdbk = $pdbks[$key];
3458 $searchkey = "<!--LINK $key-->";
3459 $title = $this->mLinkHolders
['titles'][$key];
3460 if ( empty( $colours[$pdbk] ) ) {
3461 $linkCache->addBadLinkObj( $title );
3462 $colours[$pdbk] = 0;
3463 $this->mOutput
->addLink( $title, 0 );
3464 $wgOutputReplace[$searchkey] = $sk->makeBrokenLinkObj( $title,
3465 $this->mLinkHolders
['texts'][$key],
3466 $this->mLinkHolders
['queries'][$key] );
3467 } elseif ( $colours[$pdbk] == 1 ) {
3468 $wgOutputReplace[$searchkey] = $sk->makeKnownLinkObj( $title,
3469 $this->mLinkHolders
['texts'][$key],
3470 $this->mLinkHolders
['queries'][$key] );
3471 } elseif ( $colours[$pdbk] == 2 ) {
3472 $wgOutputReplace[$searchkey] = $sk->makeStubLinkObj( $title,
3473 $this->mLinkHolders
['texts'][$key],
3474 $this->mLinkHolders
['queries'][$key] );
3477 wfProfileOut( $fname.'-construct' );
3480 wfProfileIn( $fname.'-replace' );
3482 $text = preg_replace_callback(
3483 '/(<!--LINK .*?-->)/',
3484 "wfOutputReplaceMatches",
3487 wfProfileOut( $fname.'-replace' );
3490 # Now process interwiki link holders
3491 # This is quite a bit simpler than internal links
3492 if ( !empty( $this->mInterwikiLinkHolders
['texts'] ) ) {
3493 wfProfileIn( $fname.'-interwiki' );
3494 # Make interwiki link HTML
3495 $wgOutputReplace = array();
3496 foreach( $this->mInterwikiLinkHolders
['texts'] as $key => $link ) {
3497 $title = $this->mInterwikiLinkHolders
['titles'][$key];
3498 $wgOutputReplace[$key] = $sk->makeLinkObj( $title, $link );
3501 $text = preg_replace_callback(
3502 '/<!--IWLINK (.*?)-->/',
3503 "wfOutputReplaceMatches",
3505 wfProfileOut( $fname.'-interwiki' );
3508 wfProfileOut( $fname );
3513 * Replace <!--LINK--> link placeholders with plain text of links
3514 * (not HTML-formatted).
3515 * @param string $text
3518 function replaceLinkHoldersText( $text ) {
3520 global $wgOutputReplace;
3522 $fname = 'Parser::replaceLinkHoldersText';
3523 wfProfileIn( $fname );
3525 $text = preg_replace_callback(
3526 '/<!--(LINK|IWLINK) (.*?)-->/',
3527 array( &$this, 'replaceLinkHoldersTextCallback' ),
3530 wfProfileOut( $fname );
3535 * @param array $matches
3539 function replaceLinkHoldersTextCallback( $matches ) {
3540 $type = $matches[1];
3542 if( $type == 'LINK' ) {
3543 if( isset( $this->mLinkHolders
['texts'][$key] ) ) {
3544 return $this->mLinkHolders
['texts'][$key];
3546 } elseif( $type == 'IWLINK' ) {
3547 if( isset( $this->mInterwikiLinkHolders
['texts'][$key] ) ) {
3548 return $this->mInterwikiLinkHolders
['texts'][$key];
3555 * Renders an image gallery from a text with one line per image.
3556 * text labels may be given by using |-style alternative text. E.g.
3557 * Image:one.jpg|The number "1"
3558 * Image:tree.jpg|A tree
3559 * given as text will return the HTML of a gallery with two images,
3560 * labeled 'The number "1"' and
3563 function renderImageGallery( $text ) {
3565 $parserOptions = new ParserOptions
;
3566 $localParser = new Parser();
3568 $ig = new ImageGallery();
3569 $ig->setShowBytes( false );
3570 $ig->setShowFilename( false );
3571 $lines = explode( "\n", $text );
3573 foreach ( $lines as $line ) {
3574 # match lines like these:
3575 # Image:someimage.jpg|This is some image
3576 preg_match( "/^([^|]+)(\\|(.*))?$/", $line, $matches );
3578 if ( count( $matches ) == 0 ) {
3581 $nt =& Title
::newFromText( $matches[1] );
3582 if( is_null( $nt ) ) {
3583 # Bogus title. Ignore these so we don't bomb out later.
3586 if ( isset( $matches[3] ) ) {
3587 $label = $matches[3];
3592 $pout = $localParser->parse( $label , $this->mTitle
, $parserOptions );
3593 $html = $pout->getText();
3595 $ig->add( new Image( $nt ), $html );
3596 $this->mOutput
->addImage( $nt->getDBkey() );
3598 return $ig->toHTML();
3602 * Parse image options text and use it to make an image
3604 function makeImage( &$nt, $options ) {
3605 global $wgContLang, $wgUseImageResize, $wgUser;
3609 # Check if the options text is of the form "options|alt text"
3611 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
3612 # * left no resizing, just left align. label is used for alt= only
3613 # * right same, but right aligned
3614 # * none same, but not aligned
3615 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
3616 # * center center the image
3617 # * framed Keep original image size, no magnify-button.
3619 $part = explode( '|', $options);
3621 $mwThumb =& MagicWord
::get( MAG_IMG_THUMBNAIL
);
3622 $mwManualThumb =& MagicWord
::get( MAG_IMG_MANUALTHUMB
);
3623 $mwLeft =& MagicWord
::get( MAG_IMG_LEFT
);
3624 $mwRight =& MagicWord
::get( MAG_IMG_RIGHT
);
3625 $mwNone =& MagicWord
::get( MAG_IMG_NONE
);
3626 $mwWidth =& MagicWord
::get( MAG_IMG_WIDTH
);
3627 $mwCenter =& MagicWord
::get( MAG_IMG_CENTER
);
3628 $mwFramed =& MagicWord
::get( MAG_IMG_FRAMED
);
3631 $width = $height = $framed = $thumb = false;
3632 $manual_thumb = '' ;
3634 foreach( $part as $key => $val ) {
3635 if ( $wgUseImageResize && ! is_null( $mwThumb->matchVariableStartToEnd($val) ) ) {
3637 } elseif ( ! is_null( $match = $mwManualThumb->matchVariableStartToEnd($val) ) ) {
3638 # use manually specified thumbnail
3640 $manual_thumb = $match;
3641 } elseif ( ! is_null( $mwRight->matchVariableStartToEnd($val) ) ) {
3642 # remember to set an alignment, don't render immediately
3644 } elseif ( ! is_null( $mwLeft->matchVariableStartToEnd($val) ) ) {
3645 # remember to set an alignment, don't render immediately
3647 } elseif ( ! is_null( $mwCenter->matchVariableStartToEnd($val) ) ) {
3648 # remember to set an alignment, don't render immediately
3650 } elseif ( ! is_null( $mwNone->matchVariableStartToEnd($val) ) ) {
3651 # remember to set an alignment, don't render immediately
3653 } elseif ( $wgUseImageResize && ! is_null( $match = $mwWidth->matchVariableStartToEnd($val) ) ) {
3654 wfDebug( "MAG_IMG_WIDTH match: $match\n" );
3655 # $match is the image width in pixels
3656 if ( preg_match( '/^([0-9]*)x([0-9]*)$/', $match, $m ) ) {
3657 $width = intval( $m[1] );
3658 $height = intval( $m[2] );
3660 $width = intval($match);
3662 } elseif ( ! is_null( $mwFramed->matchVariableStartToEnd($val) ) ) {
3668 # Strip bad stuff out of the alt text
3669 $alt = $this->replaceLinkHoldersText( $caption );
3670 $alt = Sanitizer
::stripAllTags( $alt );
3672 # Linker does the rest
3673 $sk =& $this->mOptions
->getSkin();
3674 return $sk->makeImageLinkObj( $nt, $caption, $alt, $align, $width, $height, $framed, $thumb, $manual_thumb );
3678 * Set a flag in the output object indicating that the content is dynamic and
3679 * shouldn't be cached.
3681 function disableCache() {
3682 $this->mOutput
->mCacheTime
= -1;
3686 * Callback from the Sanitizer for expanding items found in HTML attribute
3687 * values, so they can be safely tested and escaped.
3688 * @param string $text
3689 * @param array $args
3693 function attributeStripCallback( &$text, $args ) {
3694 $text = $this->replaceVariables( $text, $args );
3695 $text = $this->unstripForHTML( $text );
3699 function unstripForHTML( $text ) {
3700 $text = $this->unstrip( $text, $this->mStripState
);
3701 $text = $this->unstripNoWiki( $text, $this->mStripState
);
3709 function Title( $x = NULL ) { return wfSetVar( $this->mTitle
, $x ); }
3710 function Options( $x = NULL ) { return wfSetVar( $this->mOptions
, $x ); }
3711 function OutputType( $x = NULL ) { return wfSetVar( $this->mOutputType
, $x ); }
3717 function getTags() { return array_keys( $this->mTagHooks
); }
3723 * @package MediaWiki
3727 var $mText, # The output text
3728 $mLanguageLinks, # List of the full text of language links, in the order they appear
3729 $mCategories, # Map of category names to sort keys
3730 $mContainsOldMagic, # Boolean variable indicating if the input contained variables like {{CURRENTDAY}}
3731 $mCacheTime, # Timestamp on this article, or -1 for uncacheable. Used in ParserCache.
3732 $mVersion, # Compatibility check
3733 $mTitleText, # title text of the chosen language variant
3734 $mLinks, # 2-D map of NS/DBK to ID for the links in the document. ID=zero for broken.
3735 $mTemplates, # 2-D map of NS/DBK to ID for the template references. ID=zero for broken.
3736 $mImages; # DB keys of the images used, in the array key only
3738 function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(),
3739 $containsOldMagic = false, $titletext = '' )
3741 $this->mText
= $text;
3742 $this->mLanguageLinks
= $languageLinks;
3743 $this->mCategories
= $categoryLinks;
3744 $this->mContainsOldMagic
= $containsOldMagic;
3745 $this->mCacheTime
= '';
3746 $this->mVersion
= MW_PARSER_VERSION
;
3747 $this->mTitleText
= $titletext;
3748 $this->mLinks
= array();
3749 $this->mTemplates
= array();
3750 $this->mImages
= array();
3753 function getText() { return $this->mText
; }
3754 function getLanguageLinks() { return $this->mLanguageLinks
; }
3755 function getCategoryLinks() { return array_keys( $this->mCategories
); }
3756 function &getCategories() { return $this->mCategories
; }
3757 function getCacheTime() { return $this->mCacheTime
; }
3758 function getTitleText() { return $this->mTitleText
; }
3759 function &getLinks() { return $this->mLinks
; }
3760 function &getTemplates() { return $this->mTemplates
; }
3761 function &getImages() { return $this->mImages
; }
3763 function containsOldMagic() { return $this->mContainsOldMagic
; }
3764 function setText( $text ) { return wfSetVar( $this->mText
, $text ); }
3765 function setLanguageLinks( $ll ) { return wfSetVar( $this->mLanguageLinks
, $ll ); }
3766 function setCategoryLinks( $cl ) { return wfSetVar( $this->mCategories
, $cl ); }
3767 function setContainsOldMagic( $com ) { return wfSetVar( $this->mContainsOldMagic
, $com ); }
3768 function setCacheTime( $t ) { return wfSetVar( $this->mCacheTime
, $t ); }
3769 function setTitleText( $t ) { return wfSetVar ($this->mTitleText
, $t); }
3771 function addCategory( $c, $sort ) { $this->mCategories
[$c] = $sort; }
3772 function addImage( $name ) { $this->mImages
[$name] = 1; }
3773 function addLanguageLink( $t ) { $this->mLanguageLinks
[] = $t; }
3775 function addLink( $title, $id ) {
3776 $ns = $title->getNamespace();
3777 $dbk = $title->getDBkey();
3778 if ( !isset( $this->mLinks
[$ns] ) ) {
3779 $this->mLinks
[$ns] = array();
3781 $this->mLinks
[$ns][$dbk] = $id;
3784 function addTemplate( $title, $id ) {
3785 $ns = $title->getNamespace();
3786 $dbk = $title->getDBkey();
3787 if ( !isset( $this->mTemplates
[$ns] ) ) {
3788 $this->mTemplates
[$ns] = array();
3790 $this->mTemplates
[$ns][$dbk] = $id;
3797 function merge( $other ) {
3798 $this->mLanguageLinks = array_merge( $this->mLanguageLinks, $other->mLanguageLinks );
3799 $this->mCategories = array_merge( $this->mCategories, $this->mLanguageLinks );
3800 $this->mContainsOldMagic = $this->mContainsOldMagic || $other->mContainsOldMagic;
3804 * Return true if this cached output object predates the global or
3805 * per-article cache invalidation timestamps, or if it comes from
3806 * an incompatible older version.
3808 * @param string $touched the affected article's last touched timestamp
3812 function expired( $touched ) {
3813 global $wgCacheEpoch;
3814 return $this->getCacheTime() == -1 ||
// parser says it's uncacheable
3815 $this->getCacheTime() < $touched ||
3816 $this->getCacheTime() <= $wgCacheEpoch ||
3817 !isset( $this->mVersion
) ||
3818 version_compare( $this->mVersion
, MW_PARSER_VERSION
, "lt" );
3823 * Set options of the Parser
3825 * @package MediaWiki
3829 # All variables are private
3830 var $mUseTeX; # Use texvc to expand <math> tags
3831 var $mUseDynamicDates; # Use DateFormatter to format dates
3832 var $mInterwikiMagic; # Interlanguage links are removed and returned in an array
3833 var $mAllowExternalImages; # Allow external images inline
3834 var $mAllowExternalImagesFrom; # If not, any exception?
3835 var $mSkin; # Reference to the preferred skin
3836 var $mDateFormat; # Date format index
3837 var $mEditSection; # Create "edit section" links
3838 var $mNumberHeadings; # Automatically number headings
3839 var $mAllowSpecialInclusion; # Allow inclusion of special pages
3840 var $mTidy; # Ask for tidy cleanup
3842 function getUseTeX() { return $this->mUseTeX
; }
3843 function getUseDynamicDates() { return $this->mUseDynamicDates
; }
3844 function getInterwikiMagic() { return $this->mInterwikiMagic
; }
3845 function getAllowExternalImages() { return $this->mAllowExternalImages
; }
3846 function getAllowExternalImagesFrom() { return $this->mAllowExternalImagesFrom
; }
3847 function &getSkin() { return $this->mSkin
; }
3848 function getDateFormat() { return $this->mDateFormat
; }
3849 function getEditSection() { return $this->mEditSection
; }
3850 function getNumberHeadings() { return $this->mNumberHeadings
; }
3851 function getAllowSpecialInclusion() { return $this->mAllowSpecialInclusion
; }
3852 function getTidy() { return $this->mTidy
; }
3854 function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX
, $x ); }
3855 function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates
, $x ); }
3856 function setInterwikiMagic( $x ) { return wfSetVar( $this->mInterwikiMagic
, $x ); }
3857 function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages
, $x ); }
3858 function setAllowExternalImagesFrom( $x ) { return wfSetVar( $this->mAllowExternalImagesFrom
, $x ); }
3859 function setDateFormat( $x ) { return wfSetVar( $this->mDateFormat
, $x ); }
3860 function setEditSection( $x ) { return wfSetVar( $this->mEditSection
, $x ); }
3861 function setNumberHeadings( $x ) { return wfSetVar( $this->mNumberHeadings
, $x ); }
3862 function setAllowSpecialInclusion( $x ) { return wfSetVar( $this->mAllowSpecialInclusion
, $x ); }
3863 function setTidy( $x ) { return wfSetVar( $this->mTidy
, $x); }
3864 function setSkin( &$x ) { $this->mSkin
=& $x; }
3866 function ParserOptions() {
3868 $this->initialiseFromUser( $wgUser );
3872 * Get parser options
3875 function newFromUser( &$user ) {
3876 $popts = new ParserOptions
;
3877 $popts->initialiseFromUser( $user );
3881 /** Get user options */
3882 function initialiseFromUser( &$userInput ) {
3883 global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages,
3884 $wgAllowExternalImagesFrom, $wgAllowSpecialInclusion;
3885 $fname = 'ParserOptions::initialiseFromUser';
3886 wfProfileIn( $fname );
3887 if ( !$userInput ) {
3889 $user->setLoaded( true );
3891 $user =& $userInput;
3894 $this->mUseTeX
= $wgUseTeX;
3895 $this->mUseDynamicDates
= $wgUseDynamicDates;
3896 $this->mInterwikiMagic
= $wgInterwikiMagic;
3897 $this->mAllowExternalImages
= $wgAllowExternalImages;
3898 $this->mAllowExternalImagesFrom
= $wgAllowExternalImagesFrom;
3899 wfProfileIn( $fname.'-skin' );
3900 $this->mSkin
=& $user->getSkin();
3901 wfProfileOut( $fname.'-skin' );
3902 $this->mDateFormat
= $user->getOption( 'date' );
3903 $this->mEditSection
= true;
3904 $this->mNumberHeadings
= $user->getOption( 'numberheadings' );
3905 $this->mAllowSpecialInclusion
= $wgAllowSpecialInclusion;
3906 $this->mTidy
= false;
3907 wfProfileOut( $fname );
3912 * Callback function used by Parser::replaceLinkHolders()
3913 * to substitute link placeholders.
3915 function &wfOutputReplaceMatches( $matches ) {
3916 global $wgOutputReplace;
3917 return $wgOutputReplace[$matches[1]];
3921 * Return the total number of articles
3923 function wfNumberOfArticles() {
3924 global $wgNumberOfArticles;
3927 return $wgNumberOfArticles;
3931 * Return the number of files
3933 function wfNumberOfFiles() {
3934 $fname = 'Parser::wfNumberOfFiles';
3936 wfProfileIn( $fname );
3937 $dbr =& wfGetDB( DB_SLAVE
);
3938 $res = $dbr->selectField('image', 'COUNT(*)', array(), $fname );
3939 wfProfileOut( $fname );
3945 * Get various statistics from the database
3948 function wfLoadSiteStats() {
3949 global $wgNumberOfArticles, $wgTotalViews, $wgTotalEdits;
3950 $fname = 'wfLoadSiteStats';
3952 if ( -1 != $wgNumberOfArticles ) return;
3953 $dbr =& wfGetDB( DB_SLAVE
);
3954 $s = $dbr->selectRow( 'site_stats',
3955 array( 'ss_total_views', 'ss_total_edits', 'ss_good_articles' ),
3956 array( 'ss_row_id' => 1 ), $fname
3959 if ( $s === false ) {
3962 $wgTotalViews = $s->ss_total_views
;
3963 $wgTotalEdits = $s->ss_total_edits
;
3964 $wgNumberOfArticles = $s->ss_good_articles
;
3970 * Basically replacing " > and < with HTML entities ( ", >, <)
3972 * @param string $in Text that might contain HTML tags
3973 * @return string Escaped string
3975 function wfEscapeHTMLTagsOnly( $in ) {
3977 array( '"', '>', '<' ),
3978 array( '"', '>', '<' ),