<?php
-
/**
* File for Parser and related classes
*
* @package MediaWiki
- * @version $Id$
*/
+/** */
+require_once( 'Sanitizer.php' );
+
+/**
+ * Update this version number when the ParserOutput format
+ * changes in an incompatible way, so the parser cache
+ * can automatically discard old data.
+ */
+define( 'MW_PARSER_VERSION', '1.4.0' );
+
/**
* Variable substitution O(N^2) attack
*
* Hence, we limit the number of inclusions of any given page, thus bringing any
* attack back to O(N).
*/
+
define( 'MAX_INCLUDE_REPEAT', 100 );
define( 'MAX_INCLUDE_SIZE', 1000000 ); // 1 Million
+define( 'RLH_FOR_UPDATE', 1 );
+
# Allowed values for $mOutputType
define( 'OT_HTML', 1 );
define( 'OT_WIKI', 2 );
define( 'URL_PROTOCOLS', 'http|https|ftp|irc|gopher|news|mailto' );
define( 'HTTP_PROTOCOLS', 'http|https' );
# Everything except bracket, space, or control characters
-define( 'EXT_LINK_URL_CLASS', '[^]<>\\x00-\\x20\\x7F]' );
+define( 'EXT_LINK_URL_CLASS', '[^]<>"\\x00-\\x20\\x7F]' );
# Including space
define( 'EXT_LINK_TEXT_CLASS', '[^\]\\x00-\\x1F\\x7F]' );
define( 'EXT_IMAGE_FNAME_CLASS', '[A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]' );
define( 'EXT_IMAGE_EXTENSIONS', 'gif|png|jpg|jpeg' );
-define( 'EXT_LINK_BRACKETED', '/\[(('.URL_PROTOCOLS.'):'.EXT_LINK_URL_CLASS.'+) *('.EXT_LINK_TEXT_CLASS.'*?)\]/S' );
+define( 'EXT_LINK_BRACKETED', '/\[(\b('.URL_PROTOCOLS.'):'.EXT_LINK_URL_CLASS.'+) *('.EXT_LINK_TEXT_CLASS.'*?)\]/S' );
define( 'EXT_IMAGE_REGEX',
'/^('.HTTP_PROTOCOLS.':)'. # Protocol
'('.EXT_LINK_URL_CLASS.'+)\\/'. # Hostname and path
* performs brace substitution on MediaWiki messages
*
* Globals used:
- * objects: $wgLang, $wgDateFormatter, $wgLinkCache, $wgCurParser
+ * objects: $wgLang, $wgDateFormatter, $wgLinkCache
*
* NOT $wgArticle, $wgUser or $wgTitle. Keep them away!
*
function clearState() {
$this->mOutput = new ParserOutput;
$this->mAutonumber = 0;
- $this->mLastSection = "";
+ $this->mLastSection = '';
$this->mDTopen = false;
$this->mVariables = false;
$this->mIncludeCount = array();
* to internalParse() which does all the real work.
*
* @access private
+ * @param string $text Text we want to parse
+ * @param Title &$title A title object
+ * @param array $options
+ * @param boolean $linestart
+ * @param boolean $clearState
* @return ParserOutput a ParserOutput
*/
function parse( $text, &$title, $options, $linestart = true, $clearState = true ) {
- global $wgUseTidy;
+ global $wgUseTidy, $wgContLang;
$fname = 'Parser::parse';
wfProfileIn( $fname );
$this->mOutputType = OT_HTML;
$stripState = NULL;
- $text = $this->strip( $text, $this->mStripState );
+ global $fnord; $fnord = 1;
+ //$text = $this->strip( $text, $this->mStripState );
+ // VOODOO MAGIC FIX! Sometimes the above segfaults in PHP5.
+ $x =& $this->mStripState;
+ $text = $this->strip( $text, $x );
+
$text = $this->internalParse( $text, $linestart );
$text = $this->unstrip( $text, $this->mStripState );
# Clean up special characters, only run once, next-to-last before doBlockLevels
# only if there is something before the space
'/(.) (?=\\?|:|;|!|\\302\\273)/i' => '\\1 \\2',
# french spaces, Guillemet-right
- "/(\\302\\253) /i"=>"\\1 ",
+ '/(\\302\\253) /i' => '\\1 ',
'/<hr *>/i' => '<hr />',
'/<br *>/i' => '<br />',
'/<center *>/i' => '<div class="center">',
'/<\\/center *>/i' => '</div>',
- # Clean up spare ampersands; note that we probably ought to be
- # more careful about named entities.
- '/&(?!:amp;|#[Xx][0-9A-fa-f]+;|#[0-9]+;|[a-zA-Z0-9]+;)/' => '&'
);
$text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
+ $text = Sanitizer::normalizeCharReferences( $text );
} else {
$fixtags = array(
# french spaces, last one Guillemet-left
}
# only once and last
$text = $this->doBlockLevels( $text, $linestart );
+
+ $this->replaceLinkHolders( $text );
+ $text = $wgContLang->convert($text);
+
$text = $this->unstripNoWiki( $text, $this->mStripState );
+ global $wgUseTidy;
+ if ($wgUseTidy) {
+ $text = Parser::tidy($text);
+ }
+
$this->mOutput->setText( $text );
wfProfileOut( $fname );
return $this->mOutput;
$pre_content = array();
$comment_content = array();
$ext_content = array();
+ $gallery_content = array();
# Replace any instances of the placeholders
$uniq_prefix = UNIQ_PREFIX;
}
}
+ # gallery
+ $text = Parser::extractTags('gallery', $text, $gallery_content, $uniq_prefix);
+ foreach( $gallery_content as $marker => $content ) {
+ require_once( 'ImageGallery.php' );
+ if ( $render ) {
+ $gallery_content[$marker] = Parser::renderImageGallery( $content );
+ } else {
+ $gallery_content[$marker] = '<gallery>'.$content.'</gallery>';
+ }
+ }
+
# Comments
if($stripcomments) {
$text = Parser::extractTags(STRIP_COMMENTS, $text, $comment_content, $uniq_prefix);
$state['math'] = $state['math'] + $math_content;
$state['pre'] = $state['pre'] + $pre_content;
$state['comment'] = $state['comment'] + $comment_content;
+ $state['gallery'] = $state['gallery'] + $gallery_content;
foreach( $ext_content as $tag => $array ) {
if ( array_key_exists( $tag, $state ) ) {
'math' => $math_content,
'pre' => $pre_content,
'comment' => $comment_content,
+ 'gallery' => $gallery_content,
) + $ext_content;
}
return $text;
}
/**
- * restores pre, math, and heiro removed by strip()
+ * restores pre, math, and hiero removed by strip()
*
* always call unstripNoWiki() after this one
* @access private
return $rnd;
}
- /**
- * Return allowed HTML attributes
- *
- * @access private
- */
- function getHTMLattrs () {
- $htmlattrs = array( # Allowed attributes--no scripting, etc.
- 'title', 'align', 'lang', 'dir', 'width', 'height',
- 'bgcolor', 'clear', /* BR */ 'noshade', /* HR */
- 'cite', /* BLOCKQUOTE, Q */ 'size', 'face', 'color',
- /* FONT */ 'type', 'start', 'value', 'compact',
- /* For various lists, mostly deprecated but safe */
- 'summary', 'width', 'border', 'frame', 'rules',
- 'cellspacing', 'cellpadding', 'valign', 'char',
- 'charoff', 'colgroup', 'col', 'span', 'abbr', 'axis',
- 'headers', 'scope', 'rowspan', 'colspan', /* Tables */
- 'id', 'class', 'name', 'style' /* For CSS */
- );
- return $htmlattrs ;
- }
-
- /**
- * Remove non approved attributes and javascript in css
- *
- * @access private
- */
- function fixTagAttributes ( $t ) {
- if ( trim ( $t ) == '' ) return '' ; # Saves runtime ;-)
- $htmlattrs = $this->getHTMLattrs() ;
-
- # Strip non-approved attributes from the tag
- $t = preg_replace(
- '/(\\w+)(\\s*=\\s*([^\\s\">]+|\"[^\">]*\"))?/e',
- "(in_array(strtolower(\"\$1\"),\$htmlattrs)?(\"\$1\".((\"x\$3\" != \"x\")?\"=\$3\":'')):'')",
- $t);
-
- $t = str_replace ( '<></>' , '' , $t ) ; # This should fix bug 980557
-
- # Strip javascript "expression" from stylesheets. Brute force approach:
- # If anythin offensive is found, all attributes of the HTML tag are dropped
-
- if( preg_match(
- '/style\\s*=.*(expression|tps*:\/\/|url\\s*\().*/is',
- wfMungeToUtf8( $t ) ) )
- {
- $t='';
- }
-
- return trim ( $t ) ;
- }
-
/**
* interface with html tidy, used if $wgUseTidy = true
*
$indent_level = strlen( $matches[1] );
$t[$k] = "\n" .
str_repeat( '<dl><dd>', $indent_level ) .
- '<table ' . $this->fixTagAttributes ( $matches[2] ) . '>' ;
+ '<table' . Sanitizer::fixTagAttributes ( $matches[2], 'table' ) . '>' ;
array_push ( $td , false ) ;
array_push ( $ltd , '' ) ;
array_push ( $tr , false ) ;
}
else if ( count ( $td ) == 0 ) { } # Don't do any of the following
else if ( '|}' == substr ( $x , 0 , 2 ) ) {
- $z = "</table>\n" ;
+ $z = "</table>" . substr ( $x , 2) . "\n";
$l = array_pop ( $ltd ) ;
if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
array_push ( $tr , false ) ;
array_push ( $td , false ) ;
array_push ( $ltd , '' ) ;
- array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
+ array_push ( $ltr , Sanitizer::fixTagAttributes ( $x, 'tr' ) ) ;
}
else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) { # Caption
# $x is a table row
if ( $fc != '+' )
{
$tra = array_pop ( $ltr ) ;
- if ( !array_pop ( $tr ) ) $z = '<tr '.$tra.">\n" ;
+ if ( !array_pop ( $tr ) ) $z = '<tr'.$tra.">\n" ;
array_push ( $tr , true ) ;
array_push ( $ltr , '' ) ;
}
}
if ( count ( $y ) == 1 )
$y = "{$z}<{$l}>{$y[0]}" ;
- else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
+ else $y = $y = "{$z}<{$l}".Sanitizer::fixTagAttributes($y[0], $l).">{$y[1]}" ;
$t[$k] .= $y ;
array_push ( $td , true ) ;
}
}
$t = implode ( "\n" , $t ) ;
- # $t = $this->removeHTMLtags( $t );
+ # $t = Sanitizer::removeHTMLtags( $t );
wfProfileOut( $fname );
return $t ;
}
* @access private
*/
function internalParse( $text, $linestart, $args = array(), $isMain=true ) {
- global $wgContLang;
+ global $wgContLang;
$fname = 'Parser::internalParse';
wfProfileIn( $fname );
- $text = $this->removeHTMLtags( $text );
+ $text = Sanitizer::removeHTMLtags( $text );
$text = $this->replaceVariables( $text, $args );
- $text = $wgContLang->convert($text);
-
$text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
$text = $this->doHeadings( $text );
$text = $wgDateFormatter->reformat( $this->mOptions->getDateFormat(), $text );
}
$text = $this->doAllQuotes( $text );
- $text = $this->replaceInternalLinks ( $text );
- # Another call to replace links and images inside captions of images
- $text = $this->replaceInternalLinks ( $text );
- $text = $this->replaceExternalLinks( $text );
+ $text = $this->replaceInternalLinks( $text );
+ $text = $this->replaceExternalLinks( $text );
+
+ # replaceInternalLinks may sometimes leave behind
+ # absolute URLs, which have to be masked to hide them from replaceExternalLinks
+ $text = str_replace("http-noparse://","http://",$text);
+
$text = $this->doMagicLinks( $text );
$text = $this->doTableStuff( $text );
$text = $this->formatHeadings( $text, $isMain );
- $sk =& $this->mOptions->getSkin();
- $text = $sk->transformContent( $text );
wfProfileOut( $fname );
return $text;
* @access private
*/
function &doMagicLinks( &$text ) {
- global $wgUseGeoMode;
$text = $this->magicISBN( $text );
- if ( isset( $wgUseGeoMode ) && $wgUseGeoMode ) {
- $text = $this->magicGEO( $text );
- }
- $text = $this->magicRFC( $text );
+ $text = $this->magicRFC( $text, 'RFC ', 'rfcurl' );
+ $text = $this->magicRFC( $text, 'PMID ', 'pubmedurl' );
return $text;
}
*
* @access private
*/
- function doExponent ( $text ) {
+ function doExponent( $text ) {
$fname = 'Parser::doExponent';
- wfProfileIn( $fname);
+ wfProfileIn( $fname );
$text = preg_replace('/\^\^(.*)\^\^/','<small><sup>\\1</sup></small>', $text);
- wfProfileOut( $fname);
+ wfProfileOut( $fname );
return $text;
}
* @access private
*/
function doQuotes( $text ) {
- $arr = preg_split ("/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE);
- if (count ($arr) == 1)
+ $arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE );
+ if ( count( $arr ) == 1 )
return $text;
else
{
$i = 0;
$numbold = 0;
$numitalics = 0;
- foreach ($arr as $r)
+ foreach ( $arr as $r )
{
- if (($i % 2) == 1)
+ if ( ( $i % 2 ) == 1 )
{
# If there are ever four apostrophes, assume the first is supposed to
# be text, and the remaining three constitute mark-up for bold text.
- if (strlen ($arr[$i]) == 4)
+ if ( strlen( $arr[$i] ) == 4 )
{
$arr[$i-1] .= "'";
$arr[$i] = "'''";
}
# If there are more than 5 apostrophes in a row, assume they're all
# text except for the last 5.
- else if (strlen ($arr[$i]) > 5)
+ else if ( strlen( $arr[$i] ) > 5 )
{
- $arr[$i-1] .= str_repeat ("'", strlen ($arr[$i]) - 5);
+ $arr[$i-1] .= str_repeat( "'", strlen( $arr[$i] ) - 5 );
$arr[$i] = "'''''";
}
# Count the number of occurrences of bold and italics mark-ups.
# We are not counting sequences of five apostrophes.
- if (strlen ($arr[$i]) == 2) $numitalics++; else
- if (strlen ($arr[$i]) == 3) $numbold++; else
- if (strlen ($arr[$i]) == 5) { $numitalics++; $numbold++; }
+ if ( strlen( $arr[$i] ) == 2 ) $numitalics++; else
+ if ( strlen( $arr[$i] ) == 3 ) $numbold++; else
+ if ( strlen( $arr[$i] ) == 5 ) { $numitalics++; $numbold++; }
}
$i++;
}
# that one of the bold ones was meant to be an apostrophe followed
# by italics. Which one we cannot know for certain, but it is more
# likely to be one that has a single-letter word before it.
- if (($numbold % 2 == 1) && ($numitalics % 2 == 1))
+ if ( ( $numbold % 2 == 1 ) && ( $numitalics % 2 == 1 ) )
{
$i = 0;
$firstsingleletterword = -1;
$firstmultiletterword = -1;
$firstspace = -1;
- foreach ($arr as $r)
+ foreach ( $arr as $r )
{
- if (($i % 2 == 1) and (strlen ($r) == 3))
+ if ( ( $i % 2 == 1 ) and ( strlen( $r ) == 3 ) )
{
$x1 = substr ($arr[$i-1], -1);
$x2 = substr ($arr[$i-1], -2, 1);
/**
* Replace external links
*
- * Note: we have to do external links before the internal ones,
- * and otherwise take great care in the order of things here, so
- * that we don't end up interpreting some URLs twice.
+ * Note: this is all very hackish and the order of execution matters a lot.
+ * Make sure to run maintenance/parserTests.php if you change this code.
*
* @access private
*/
wfProfileIn( $fname );
$sk =& $this->mOptions->getSkin();
- $linktrail = wfMsgForContent('linktrail');
+ global $wgContLang;
+ $linktrail = $wgContLang->linkTrail();
+
$bits = preg_split( EXT_LINK_BRACKETED, $text, -1, PREG_SPLIT_DELIM_CAPTURE );
$s = $this->replaceFreeExternalLinks( array_shift( $bits ) );
$text = $bits[$i++];
$trail = $bits[$i++];
+ # The characters '<' and '>' (which were escaped by
+ # removeHTMLtags()) should not be included in
+ # URLs, per RFC 2396.
+ if (preg_match('/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE)) {
+ $text = substr($url, $m2[0][1]) . ' ' . $text;
+ $url = substr($url, 0, $m2[0][1]);
+ }
+
# If the link text is an image URL, replace it with an <img> tag
# This happened by accident in the original parser, but some people used it extensively
$img = $this->maybeMakeImageLink( $text );
$dtrail = '';
+ # Set linktype for CSS - if URL==text, link is essentially free
+ $linktype = ($text == $url) ? 'free' : 'text';
+
# No link text, e.g. [http://domain.tld/some.link]
if ( $text == '' ) {
# Autonumber if allowed
if ( strpos( HTTP_PROTOCOLS, $protocol ) !== false ) {
$text = '[' . ++$this->mAutonumber . ']';
+ $linktype = 'autonumber';
} else {
# Otherwise just use the URL
$text = htmlspecialchars( $url );
+ $linktype = 'free';
}
} else {
# Have link text, e.g. [http://domain.tld/some.link text]s
}
}
- $encUrl = htmlspecialchars( $url );
- # Bit in parentheses showing the URL for the printable version
- if( $url == $text || preg_match( "!$protocol://" . preg_quote( $text, '/' ) . "/?$!", $url ) ) {
- $paren = '';
- } else {
- # Expand the URL for printable version
- if ( ! $sk->suppressUrlExpansion() ) {
- $paren = "<span class='urlexpansion'> (<i>" . htmlspecialchars ( $encUrl ) . "</i>)</span>";
- } else {
- $paren = '';
- }
- }
+ # Replace & from obsolete syntax with &.
+ # All HTML entities will be escaped by makeExternalLink()
+ # or maybeMakeImageLink()
+ $url = str_replace( '&', '&', $url );
# Process the trail (i.e. everything after this link up until start of the next link),
# replacing any non-bracketed links
$trail = $this->replaceFreeExternalLinks( $trail );
- $la = $sk->getExternalLinkAttributes( $url, $text );
-
# Use the encoded URL
# This means that users can paste URLs directly into the text
# Funny characters like ö aren't valid in URLs anyway
# This was changed in August 2004
- $s .= "<a href=\"{$url}\"{$la}>{$text}</a>{$dtrail}{$paren}{$trail}";
+ $s .= $sk->makeExternalLink( $url, $text, false, $linktype ) . $dtrail . $trail;
}
wfProfileOut( $fname );
* @access private
*/
function replaceFreeExternalLinks( $text ) {
- $bits = preg_split( '/((?:'.URL_PROTOCOLS.'):)/', $text, -1, PREG_SPLIT_DELIM_CAPTURE );
+ $fname = 'Parser::replaceFreeExternalLinks';
+ wfProfileIn( $fname );
+
+ $bits = preg_split( '/(\b(?:'.URL_PROTOCOLS.'):)/S', $text, -1, PREG_SPLIT_DELIM_CAPTURE );
$s = array_shift( $bits );
$i = 0;
$url = $protocol . $m[1];
$trail = $m[2];
+ # The characters '<' and '>' (which were escaped by
+ # removeHTMLtags()) should not be included in
+ # URLs, per RFC 2396.
+ if (preg_match('/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE)) {
+ $trail = substr($url, $m2[0][1]) . $trail;
+ $url = substr($url, 0, $m2[0][1]);
+ }
+
# Move trailing punctuation to $trail
$sep = ',;\.:!?';
# If there is no left bracket, then consider right brackets fair game too
$url = substr( $url, 0, -$numSepChars );
}
- # Replace & from obsolete syntax with &
+ # Replace & from obsolete syntax with &.
+ # All HTML entities will be escaped by makeExternalLink()
+ # or maybeMakeImageLink()
$url = str_replace( '&', '&', $url );
# Is this an external image?
$text = $this->maybeMakeImageLink( $url );
if ( $text === false ) {
# Not an image, make a link
- $text = $sk->makeExternalLink( $url, $url );
+ $text = $sk->makeExternalLink( $url, $url, true, 'free' );
}
$s .= $text . $trail;
} else {
$s .= $protocol . $remainder;
}
}
+ wfProfileOut();
return $s;
}
*/
function replaceInternalLinks( $s ) {
global $wgLang, $wgContLang, $wgLinkCache;
+ global $wgDisableLangConversion;
static $fname = 'Parser::replaceInternalLinks' ;
+
wfProfileIn( $fname );
wfProfileIn( $fname.'-setup' );
static $tc = FALSE;
# the % is needed to support urlencoded titles as well
if ( !$tc ) { $tc = Title::legalChars() . '#%'; }
+
$sk =& $this->mOptions->getSkin();
+ global $wgUseOldExistenceCheck;
+ # "Post-parse link colour check" works only on wiki text since it's now
+ # in Parser. Enable it, then disable it when we're done.
+ $saveParseColour = $sk->postParseLinkColour( !$wgUseOldExistenceCheck );
$redirect = MagicWord::get ( MAG_REDIRECT ) ;
+ #split the entire text string on occurences of [[
$a = explode( '[[', ' ' . $s );
+ #get the first element (all text up to first [[), and remove the space we added
$s = array_shift( $a );
$s = substr( $s, 1 );
# Match a link having the form [[namespace:link|alternate]]trail
static $e1 = FALSE;
- if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|([^]]+))?]](.*)\$/sD"; }
+ if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|(.+?))?]](.*)\$/sD"; }
+ # Match cases where there is no "]]", which might still be images
+ static $e1_img = FALSE;
+ if ( !$e1_img ) { $e1_img = "/^([{$tc}]+)\\|(.*)\$/sD"; }
# Match the end of a line for a word that's not followed by whitespace,
# e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
static $e2 = '/^(.*?)([a-zA-Z\x80-\xff]+)$/sD';
$useLinkPrefixExtension = $wgContLang->linkPrefixExtension();
- # Special and Media are pseudo-namespaces; no pages actually exist in them
- $nottalk = !Namespace::isTalk( $this->mTitle->getNamespace() );
+ if( is_null( $this->mTitle ) ) {
+ wfDebugDieBacktrace( 'nooo' );
+ }
+ $nottalk = !$this->mTitle->isTalkPage();
if ( $useLinkPrefixExtension ) {
if ( preg_match( $e2, $s, $m ) ) {
$prefix = '';
}
+ $selflink = $this->mTitle->getPrefixedText();
wfProfileOut( $fname.'-setup' );
- # start procedeeding each line
- foreach ( $a as $line ) {
- wfProfileIn( $fname.'-prefixhandling' );
+ $checkVariantLink = sizeof($wgContLang->getVariants())>1;
+ $useSubpages = $this->areSubpagesAllowed();
+
+ # Loop for each link
+ for ($k = 0; isset( $a[$k] ); $k++) {
+ $line = $a[$k];
if ( $useLinkPrefixExtension ) {
+ wfProfileIn( $fname.'-prefixhandling' );
if ( preg_match( $e2, $s, $m ) ) {
$prefix = $m[2];
$s = $m[1];
$prefix = $first_prefix;
$first_prefix = false;
}
+ wfProfileOut( $fname.'-prefixhandling' );
}
- wfProfileOut( $fname.'-prefixhandling' );
+ $might_be_img = false;
+
if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
$text = $m[2];
# fix up urlencoded title texts
if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
$trail = $m[3];
+ } elseif( preg_match($e1_img, $line, $m) ) { # Invalid, but might be an image with a link in its caption
+ $might_be_img = true;
+ $text = $m[2];
+ if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
+ $trail = "";
} else { # Invalid form; output directly
$s .= $prefix . '[[' . $line ;
continue;
# Don't allow internal links to pages containing
# PROTO: where PROTO is a valid URL protocol; these
# should be external links.
- if (preg_match('/((?:'.URL_PROTOCOLS.'):)/', $m[1])) {
+ if (preg_match('/^(\b(?:'.URL_PROTOCOLS.'):)/', $m[1])) {
$s .= $prefix . '[[' . $line ;
continue;
}
# Make subpage if necessary
- $link = $this->maybeDoSubpageLink( $m[1], $text );
+ if( $useSubpages ) {
+ $link = $this->maybeDoSubpageLink( $m[1], $text );
+ } else {
+ $link = $m[1];
+ }
$noforce = (substr($m[1], 0, 1) != ':');
if (!$noforce) {
$link = substr($link, 1);
}
- $wasblank = ( '' == $text );
- if( $wasblank ) $text = $link;
-
- $nt = Title::newFromText( $link );
+ $nt =& Title::newFromText( $this->unstripNoWiki($link, $this->mStripState) );
if( !$nt ) {
$s .= $prefix . '[[' . $line;
continue;
}
- //check other language variants of the link
- //if the article does not exist
- if($nt->getArticleID() == 0) {
- global $wgContLang;
- $variants = $wgContLang->getVariants();
- $varnt = false;
- if(sizeof($variants) > 1) {
- foreach ( $variants as $v ) {
- if($v == $wgContLang->getPreferredVariant())
- continue;
- $varlink = $wgContLang->autoConvert($link, $v);
- $varnt = Title::newFromText($varlink);
- if($varnt && $varnt->getArticleID()>0) {
+ #check other language variants of the link
+ #if the article does not exist
+ if( $checkVariantLink
+ && $nt->getArticleID() == 0 ) {
+ $wgContLang->findVariantLink($link, $nt);
+ }
+
+ $ns = $nt->getNamespace();
+ $iw = $nt->getInterWiki();
+
+ if ($might_be_img) { # if this is actually an invalid link
+ if ($ns == NS_IMAGE && $noforce) { #but might be an image
+ $found = false;
+ while (isset ($a[$k+1]) ) {
+ #look at the next 'line' to see if we can close it there
+ $next_line = array_shift(array_splice( $a, $k + 1, 1) );
+ if( preg_match("/^(.*?]].*?)]](.*)$/sD", $next_line, $m) ) {
+ # the first ]] closes the inner link, the second the image
+ $found = true;
+ $text .= '[[' . $m[1];
+ $trail = $m[2];
+ break;
+ } elseif( preg_match("/^.*?]].*$/sD", $next_line, $m) ) {
+ #if there's exactly one ]] that's fine, we'll keep looking
+ $text .= '[[' . $m[0];
+ } else {
+ #if $next_line is invalid too, we need look no further
+ $text .= '[[' . $next_line;
break;
}
}
- }
- if($varnt && $varnt->getArticleID()>0) {
- $nt = $varnt;
- $link = $varlink;
+ if ( !$found ) {
+ # we couldn't find the end of this imageLink, so output it raw
+ #but don't ignore what might be perfectly normal links in the text we've examined
+ $text = $this->replaceInternalLinks($text);
+ $s .= $prefix . '[[' . $link . '|' . $text;
+ # note: no $trail, because without an end, there *is* no trail
+ continue;
+ }
+ } else { #it's not an image, so output it raw
+ $s .= $prefix . '[[' . $link . '|' . $text;
+ # note: no $trail, because without an end, there *is* no trail
+ continue;
}
}
- $ns = $nt->getNamespace();
- $iw = $nt->getInterWiki();
+ $wasblank = ( '' == $text );
+ if( $wasblank ) $text = $link;
+
# Link not escaped by : , create the various objects
if( $noforce ) {
}
if ( $ns == NS_IMAGE ) {
- $s .= $prefix . $sk->makeImageLinkObj( $nt, $text ) . $trail;
+ wfProfileIn( "$fname-image" );
+
+ # recursively parse links inside the image caption
+ # actually, this will parse them in any other parameters, too,
+ # but it might be hard to fix that, and it doesn't matter ATM
+ $text = $this->replaceExternalLinks($text);
+ $text = $this->replaceInternalLinks($text);
+
+ # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
+ $s .= $prefix . str_replace('http://', 'http-noparse://', $sk->makeImageLinkObj( $nt, $text ) ) . $trail;
$wgLinkCache->addImageLinkObj( $nt );
+
+ wfProfileOut( "$fname-image" );
continue;
}
if ( $ns == NS_CATEGORY ) {
- $t = $nt->getText() ;
- $nnt = Title::newFromText ( Namespace::getCanonicalName(NS_CATEGORY).':'.$t ) ;
+ wfProfileIn( "$fname-category" );
+ $t = $nt->getText();
$wgLinkCache->suspend(); # Don't save in links/brokenlinks
$pPLC=$sk->postParseLinkColour();
$sk->postParseLinkColour( false );
- $t = $sk->makeLinkObj( $nnt, $t, '', '' , $prefix );
+ $t = $sk->makeLinkObj( $nt, $t, '', '' , $prefix );
$sk->postParseLinkColour( $pPLC );
$wgLinkCache->resume();
$sortkey = $text;
}
$wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
- $this->mOutput->mCategoryLinks[] = $t ;
+ $this->mOutput->addCategoryLink( $t );
$s .= $prefix . $trail ;
+
+ wfProfileOut( "$fname-category" );
continue;
}
}
-
- if( ( $nt->getPrefixedText() === $this->mTitle->getPrefixedText() ) &&
- ( strpos( $link, '#' ) === FALSE ) ) {
+
+ if( ( $nt->getPrefixedText() === $selflink ) &&
+ ( $nt->getFragment() === '' ) ) {
# Self-links are handled specially; generally de-link and change to bold.
$s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail );
continue;
}
+ # Special and Media are pseudo-namespaces; no pages actually exist in them
if( $ns == NS_MEDIA ) {
- $s .= $prefix . $sk->makeMediaLinkObj( $nt, $text ) . $trail;
+ $s .= $prefix . $sk->makeMediaLinkObj( $nt, $text, true ) . $trail;
$wgLinkCache->addImageLinkObj( $nt );
continue;
} elseif( $ns == NS_SPECIAL ) {
}
$s .= $sk->makeLinkObj( $nt, $text, '', $trail, $prefix );
}
+ $sk->postParseLinkColour( $saveParseColour );
wfProfileOut( $fname );
return $s;
}
+ /**
+ * Return true if subpage links should be expanded on this page.
+ * @return bool
+ */
+ function areSubpagesAllowed() {
+ # Some namespaces don't allow subpages
+ global $wgNamespacesWithSubpages;
+ return !empty($wgNamespacesWithSubpages[$this->mTitle->getNamespace()]);
+ }
+
/**
* Handle link to subpage if necessary
- * @param $target string the source of the link
- * @param &$text the link text, modified as necessary
+ * @param string $target the source of the link
+ * @param string &$text the link text, modified as necessary
* @return string the full name of the link
* @access private
*/
# :Foobar -- override special treatment of prefix (images, language links)
# /Foobar -- convert to CurrentPage/Foobar
# /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text
- global $wgNamespacesWithSubpages;
+ # ../ -- convert to CurrentPage, from CurrentPage/CurrentSubPage
+ # ../Foobar -- convert to CurrentPage/Foobar, from CurrentPage/CurrentSubPage
$fname = 'Parser::maybeDoSubpageLink';
wfProfileIn( $fname );
- # Look at the first character
- if( $target{0} == '/' ) {
- # / at end means we don't want the slash to be shown
- if(substr($target,-1,1)=='/') {
- $target=substr($target,1,-1);
- $noslash=$target;
- } else {
- $noslash=substr($target,1);
- }
+ $ret = $target; # default return value is no change
+
+ # Some namespaces don't allow subpages,
+ # so only perform processing if subpages are allowed
+ if( $this->areSubpagesAllowed() ) {
+ # Look at the first character
+ if( $target != '' && $target{0} == '/' ) {
+ # / at end means we don't want the slash to be shown
+ if( substr( $target, -1, 1 ) == '/' ) {
+ $target = substr( $target, 1, -1 );
+ $noslash = $target;
+ } else {
+ $noslash = substr( $target, 1 );
+ }
- # Some namespaces don't allow subpages
- if(!empty($wgNamespacesWithSubpages[$this->mTitle->getNamespace()])) {
- # subpages allowed here
$ret = $this->mTitle->getPrefixedText(). '/' . trim($noslash);
if( '' === $text ) {
$text = $target;
} # this might be changed for ugliness reasons
} else {
- # no subpage allowed, use standard link
- $ret = $target;
+ # check for .. subpage backlinks
+ $dotdotcount = 0;
+ $nodotdot = $target;
+ while( strncmp( $nodotdot, "../", 3 ) == 0 ) {
+ ++$dotdotcount;
+ $nodotdot = substr( $nodotdot, 3 );
+ }
+ if($dotdotcount > 0) {
+ $exploded = explode( '/', $this->mTitle->GetPrefixedText() );
+ if( count( $exploded ) > $dotdotcount ) { # not allowed to go below top level page
+ $ret = implode( '/', array_slice( $exploded, 0, -$dotdotcount ) );
+ # / at the end means don't show full path
+ if( substr( $nodotdot, -1, 1 ) == '/' ) {
+ $nodotdot = substr( $nodotdot, 0, -1 );
+ if( '' === $text ) {
+ $text = $nodotdot;
+ }
+ }
+ $nodotdot = trim( $nodotdot );
+ if( $nodotdot != '' ) {
+ $ret .= '/' . $nodotdot;
+ }
+ }
+ }
}
- } else {
- # no subpage
- $ret = $target;
}
wfProfileOut( $fname );
$lastPrefix = $pref2;
}
if( 0 == $prefixLength ) {
+ wfProfileIn( "$fname-paragraph" );
# No prefix (not in list)--go to paragraph mode
$uniq_prefix = UNIQ_PREFIX;
// XXX: use a stack for nestable elements like span, table and div
- $openmatch = preg_match('/(<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<li|<\\/tr|<\\/td|<\\/th)/i', $t );
+ $openmatch = preg_match('/(<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
$closematch = preg_match(
'/(<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|'.
- '<td|<th|<div|<\\/div|<hr|<\\/pre|<\\/p|'.$uniq_prefix.'-pre|<\\/li|<\\/ul)/i', $t );
+ '<td|<th|<div|<\\/div|<hr|<\\/pre|<\\/p|'.$uniq_prefix.'-pre|<\\/li|<\\/ul)/iS', $t );
if ( $openmatch or $closematch ) {
$paragraphStack = false;
$output .= $this->closeParagraph();
}
}
}
+ wfProfileOut( "$fname-paragraph" );
}
if ($paragraphStack === false) {
$output .= $t."\n";
/**
* Split up a string on ':', ignoring any occurences inside
* <a>..</a> or <span>...</span>
- * @param $str string the string to split
- * @param &$before string set to everything before the ':'
- * @param &$after string set to everything after the ':'
+ * @param string $str the string to split
+ * @param string &$before set to everything before the ':'
+ * @param string &$after set to everything after the ':'
* return string the position of the ':', or false if none found
*/
function findColonNoLinks($str, &$before, &$after) {
*/
function getVariableValue( $index ) {
global $wgContLang, $wgSitename, $wgServer;
-
+
+ /**
+ * Some of these require message or data lookups and can be
+ * expensive to check many times.
+ */
+ static $varCache = array();
+ if( isset( $varCache[$index] ) ) return $varCache[$index];
+
switch ( $index ) {
case MAG_CURRENTMONTH:
- return $wgContLang->formatNum( date( 'm' ) );
+ return $varCache[$index] = $wgContLang->formatNum( date( 'm' ) );
case MAG_CURRENTMONTHNAME:
- return $wgContLang->getMonthName( date('n') );
+ return $varCache[$index] = $wgContLang->getMonthName( date('n') );
case MAG_CURRENTMONTHNAMEGEN:
- return $wgContLang->getMonthNameGen( date('n') );
+ return $varCache[$index] = $wgContLang->getMonthNameGen( date('n') );
case MAG_CURRENTDAY:
- return $wgContLang->formatNum( date('j') );
+ return $varCache[$index] = $wgContLang->formatNum( date('j') );
case MAG_PAGENAME:
return $this->mTitle->getText();
case MAG_PAGENAMEE:
# return Namespace::getCanonicalName($this->mTitle->getNamespace());
return $wgContLang->getNsText($this->mTitle->getNamespace()); # Patch by Dori
case MAG_CURRENTDAYNAME:
- return $wgContLang->getWeekdayName( date('w')+1 );
+ return $varCache[$index] = $wgContLang->getWeekdayName( date('w')+1 );
case MAG_CURRENTYEAR:
- return $wgContLang->formatNum( date( 'Y' ) );
+ return $varCache[$index] = $wgContLang->formatNum( date( 'Y' ) );
case MAG_CURRENTTIME:
- return $wgContLang->time( wfTimestampNow(), false );
+ return $varCache[$index] = $wgContLang->time( wfTimestampNow(), false );
+ case MAG_CURRENTWEEK:
+ return $varCache[$index] = $wgContLang->formatNum( date('W') );
+ case MAG_CURRENTDOW:
+ return $varCache[$index] = $wgContLang->formatNum( date('w') );
case MAG_NUMBEROFARTICLES:
- return $wgContLang->formatNum( wfNumberOfArticles() );
+ return $varCache[$index] = $wgContLang->formatNum( wfNumberOfArticles() );
case MAG_SITENAME:
return $wgSitename;
case MAG_SERVER:
$this->mVariables = array();
foreach ( $wgVariableIDs as $id ) {
$mw =& MagicWord::get( $id );
- $mw->addToArray( $this->mVariables, $this->getVariableValue( $id ) );
+ $mw->addToArray( $this->mVariables, $id );
}
wfProfileOut( $fname );
}
global $wgLang, $wgScript, $wgArticlePath;
# Prevent too big inclusions
- if(strlen($text)> MAX_INCLUDE_SIZE)
- return $text;
+ if( strlen( $text ) > MAX_INCLUDE_SIZE ) {
+ return $text;
+ }
$fname = 'Parser::replaceVariables';
wfProfileIn( $fname );
# This function is called recursively. To keep track of arguments we need a stack:
array_push( $this->mArgStack, $args );
- # PHP global rebinding syntax is a bit weird, need to use the GLOBALS array
- $GLOBALS['wgCurParser'] =& $this;
-
# Variable substitution
- $text = preg_replace_callback( "/{{([$titleChars]*?)}}/", 'wfVariableSubstitution', $text );
+ $text = preg_replace_callback( "/{{([$titleChars]*?)}}/", array( &$this, 'variableSubstitution' ), $text );
if ( $this->mOutputType == OT_HTML || $this->mOutputType == OT_WIKI ) {
# Argument substitution
- $text = preg_replace_callback( "/{{{([$titleChars]*?)}}}/", 'wfArgSubstitution', $text );
+ $text = preg_replace_callback( "/{{{([$titleChars]*?)}}}/", array( &$this, 'argSubstitution' ), $text );
}
# Template substitution
$regex = '/(\\n|{)?{{(['.$titleChars.']*)(\\|.*?|)}}/s';
- $text = preg_replace_callback( $regex, 'wfBraceSubstitution', $text );
+ $text = preg_replace_callback( $regex, array( &$this, 'braceSubstitution' ), $text );
array_pop( $this->mArgStack );
* @access private
*/
function variableSubstitution( $matches ) {
+ $fname = 'parser::variableSubstitution';
+ $varname = $matches[1];
+ wfProfileIn( $fname );
if ( !$this->mVariables ) {
$this->initialiseVariables();
}
if ( $this->mOutputType == OT_WIKI ) {
# Do only magic variables prefixed by SUBST
$mwSubst =& MagicWord::get( MAG_SUBST );
- if (!$mwSubst->matchStartAndRemove( $matches[1] ))
+ if (!$mwSubst->matchStartAndRemove( $varname ))
$skip = true;
# Note that if we don't substitute the variable below,
# we don't remove the {{subst:}} magic word, in case
# it is a template rather than a magic variable.
}
- if ( !$skip && array_key_exists( $matches[1], $this->mVariables ) ) {
- $text = $this->mVariables[$matches[1]];
+ if ( !$skip && array_key_exists( $varname, $this->mVariables ) ) {
+ $id = $this->mVariables[$varname];
+ $text = $this->getVariableValue( $id );
$this->mOutput->mContainsOldMagic = true;
} else {
$text = $matches[0];
}
+ wfProfileOut( $fname );
return $text;
}
function braceSubstitution( $matches ) {
global $wgLinkCache, $wgContLang;
$fname = 'Parser::braceSubstitution';
+ wfProfileIn( $fname );
+
$found = false;
$nowiki = false;
$noparse = false;
# Did we encounter this template already? If yes, it is in the cache
# and we need to check for loops.
if ( !$found && isset( $this->mTemplates[$part1] ) ) {
- # set $text to cached message.
- $text = $linestart . $this->mTemplates[$part1];
$found = true;
# Infinite loop test
if ( isset( $this->mTemplatePath[$part1] ) ) {
$noparse = true;
$found = true;
- $text .= '<!-- WARNING: template loop detected -->';
+ $text = $linestart .
+ "\{\{$part1}}" .
+ '<!-- WARNING: template loop detected -->';
+ wfDebug( "$fname: template loop broken at '$part1'\n" );
+ } else {
+ # set $text to cached message.
+ $text = $linestart . $this->mTemplates[$part1];
}
}
# Load from database
$itcamefromthedatabase = false;
+ $lastPathLevel = $this->mTemplatePath;
if ( !$found ) {
$ns = NS_TEMPLATE;
$part1 = $this->maybeDoSubpageLink( $part1, $subpage='' );
}
# Template cache array insertion
- $this->mTemplates[$part1] = $text;
+ if( $found ) {
+ $this->mTemplates[$part1] = $text;
+ }
}
}
$this->mTemplatePath[$part1] = 1;
$text = $this->strip( $text, $this->mStripState );
- $text = $this->removeHTMLtags( $text );
+ $text = Sanitizer::removeHTMLtags( $text );
$text = $this->replaceVariables( $text, $assocArgs );
# Resume the link cache and register the inclusion as a link
$text = "\n" . $text;
}
}
-
- # Empties the template path
- $this->mTemplatePath = array();
+ # Prune lower levels off the recursion check path
+ $this->mTemplatePath = $lastPathLevel;
+
if ( !$found ) {
+ wfProfileOut( $fname );
return $matches[0];
} else {
# replace ==section headers==
}
}
}
-
- # Empties the template path
- $this->mTemplatePath = array();
+ # Prune lower levels off the recursion check path
+ $this->mTemplatePath = $lastPathLevel;
+
if ( !$found ) {
+ wfProfileOut( $fname );
return $matches[0];
} else {
+ wfProfileOut( $fname );
return $text;
}
}
}
}
-
- /**
- * Cleans up HTML, removes dangerous tags and attributes, and
- * removes HTML comments
- * @access private
- */
- function removeHTMLtags( $text ) {
- global $wgUseTidy, $wgUserHtml;
- $fname = 'Parser::removeHTMLtags';
- wfProfileIn( $fname );
-
- if( $wgUserHtml ) {
- $htmlpairs = array( # Tags that must be closed
- 'b', 'del', 'i', 'ins', 'u', 'font', 'big', 'small', 'sub', 'sup', 'h1',
- 'h2', 'h3', 'h4', 'h5', 'h6', 'cite', 'code', 'em', 's',
- 'strike', 'strong', 'tt', 'var', 'div', 'center',
- 'blockquote', 'ol', 'ul', 'dl', 'table', 'caption', 'pre',
- 'ruby', 'rt' , 'rb' , 'rp', 'p'
- );
- $htmlsingle = array(
- 'br', 'hr', 'li', 'dt', 'dd'
- );
- $htmlnest = array( # Tags that can be nested--??
- 'table', 'tr', 'td', 'th', 'div', 'blockquote', 'ol', 'ul',
- 'dl', 'font', 'big', 'small', 'sub', 'sup'
- );
- $tabletags = array( # Can only appear inside table
- 'td', 'th', 'tr'
- );
- } else {
- $htmlpairs = array();
- $htmlsingle = array();
- $htmlnest = array();
- $tabletags = array();
- }
-
- $htmlsingle = array_merge( $tabletags, $htmlsingle );
- $htmlelements = array_merge( $htmlsingle, $htmlpairs );
-
- $htmlattrs = $this->getHTMLattrs () ;
-
- # Remove HTML comments
- $text = $this->removeHTMLcomments( $text );
-
- $bits = explode( '<', $text );
- $text = array_shift( $bits );
- if(!$wgUseTidy) {
- $tagstack = array(); $tablestack = array();
- foreach ( $bits as $x ) {
- $prev = error_reporting( E_ALL & ~( E_NOTICE | E_WARNING ) );
- preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
- $x, $regs );
- list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
- error_reporting( $prev );
-
- $badtag = 0 ;
- if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
- # Check our stack
- if ( $slash ) {
- # Closing a tag...
- if ( ! in_array( $t, $htmlsingle ) &&
- ( $ot = @array_pop( $tagstack ) ) != $t ) {
- @array_push( $tagstack, $ot );
- $badtag = 1;
- } else {
- if ( $t == 'table' ) {
- $tagstack = array_pop( $tablestack );
- }
- $newparams = '';
- }
- } else {
- # Keep track for later
- if ( in_array( $t, $tabletags ) &&
- ! in_array( 'table', $tagstack ) ) {
- $badtag = 1;
- } else if ( in_array( $t, $tagstack ) &&
- ! in_array ( $t , $htmlnest ) ) {
- $badtag = 1 ;
- } else if ( ! in_array( $t, $htmlsingle ) ) {
- if ( $t == 'table' ) {
- array_push( $tablestack, $tagstack );
- $tagstack = array();
- }
- array_push( $tagstack, $t );
- }
- # Strip non-approved attributes from the tag
- $newparams = $this->fixTagAttributes($params);
-
- }
- if ( ! $badtag ) {
- $rest = str_replace( '>', '>', $rest );
- $text .= "<$slash$t $newparams$brace$rest";
- continue;
- }
- }
- $text .= '<' . str_replace( '>', '>', $x);
- }
- # Close off any remaining tags
- while ( is_array( $tagstack ) && ($t = array_pop( $tagstack )) ) {
- $text .= "</$t>\n";
- if ( $t == 'table' ) { $tagstack = array_pop( $tablestack ); }
- }
- } else {
- # this might be possible using tidy itself
- foreach ( $bits as $x ) {
- preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
- $x, $regs );
- @list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
- if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
- $newparams = $this->fixTagAttributes($params);
- $rest = str_replace( '>', '>', $rest );
- $text .= "<$slash$t $newparams$brace$rest";
- } else {
- $text .= '<' . str_replace( '>', '>', $x);
- }
- }
- }
- wfProfileOut( $fname );
- return $text;
- }
-
- /**
- * Remove '<!--', '-->', and everything between.
- * To avoid leaving blank lines, when a comment is both preceded
- * and followed by a newline (ignoring spaces), trim leading and
- * trailing spaces and one of the newlines.
- *
- * @access private
- */
- function removeHTMLcomments( $text ) {
- $fname='Parser::removeHTMLcomments';
- wfProfileIn( $fname );
- while (($start = strpos($text, '<!--')) !== false) {
- $end = strpos($text, '-->', $start + 4);
- if ($end === false) {
- # Unterminated comment; bail out
- break;
- }
-
- $end += 3;
-
- # Trim space and newline if the comment is both
- # preceded and followed by a newline
- $spaceStart = max($start - 1, 0);
- $spaceLen = $end - $spaceStart;
- while (substr($text, $spaceStart, 1) === ' ' && $spaceStart > 0) {
- $spaceStart--;
- $spaceLen++;
- }
- while (substr($text, $spaceStart + $spaceLen, 1) === ' ')
- $spaceLen++;
- if (substr($text, $spaceStart, 1) === "\n" and substr($text, $spaceStart + $spaceLen, 1) === "\n") {
- # Remove the comment, leading and trailing
- # spaces, and leave only one newline.
- $text = substr_replace($text, "\n", $spaceStart, $spaceLen + 1);
- }
- else {
- # Remove just the comment.
- $text = substr_replace($text, '', $start, $end - $start);
- }
- }
- wfProfileOut( $fname );
- return $text;
- }
-
/**
* This function accomplishes several tasks:
* 1) Auto-number headings if that option is enabled
*
* It loops through all headlines, collects the necessary data, then splits up the
* string and re-inserts the newly formatted headlines.
+ *
+ * @param string $text
+ * @param boolean $isMain
* @access private
*/
- /* private */ function formatHeadings( $text, $isMain=true ) {
- global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders;
+ function formatHeadings( $text, $isMain=true ) {
+ global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders, $wgInterwikiLinkHolders;
$doNumberHeadings = $this->mOptions->getNumberHeadings();
$doShowToc = $this->mOptions->getShowToc();
# Get all headlines for numbering them and adding funky stuff like [edit]
# links - this is for later, but we need the number of headlines right now
- $numMatches = preg_match_all( '/<H([1-6])(.*?' . '>)(.*?)<\/H[1-6]>/i', $text, $matches );
+ $numMatches = preg_match_all( '/<H([1-6])(.*?'.'>)(.*?)<\/H[1-6] *>/i', $text, $matches );
# if there are fewer than 4 headlines in the article, do not show TOC
if( $numMatches < 4 ) {
# if the string __TOC__ (not case-sensitive) occurs in the HTML,
# override above conditions and always show TOC at that place
+
$mw =& MagicWord::get( MAG_TOC );
- if ($mw->match( $text ) ) {
+ if($mw->match( $text ) ) {
$doShowToc = 1;
$forceTocHere = true;
} else {
}
}
-
+ # Never ever show TOC if no headers
+ if( $numMatches < 1 ) {
+ $doShowToc = 0;
+ }
# We need this to perform operations on the HTML
$sk =& $this->mOptions->getSkin();
# Ugh .. the TOC should have neat indentation levels which can be
# passed to the skin functions. These are determined here
- $toclevel = 0;
$toc = '';
$full = '';
$head = array();
$sublevelCount = array();
+ $levelCount = array();
+ $toclevel = 0;
$level = 0;
$prevlevel = 0;
+ $toclevel = 0;
+ $prevtoclevel = 0;
+
foreach( $matches[3] as $headline ) {
$istemplate = 0;
- $templatetitle = "";
+ $templatetitle = '';
$templatesection = 0;
+ $numbering = '';
if (preg_match("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", $headline, $mat)) {
$istemplate = 1;
$headline = preg_replace("/<!--MWTEMPLATESECTION=([^&]+)&([^_]+)-->/", "", $headline);
}
- $numbering = '';
- if( $level ) {
+ if( $toclevel ) {
$prevlevel = $level;
+ $prevtoclevel = $toclevel;
}
$level = $matches[1][$headlineCount];
- if( ( $doNumberHeadings || $doShowToc ) && $prevlevel && $level > $prevlevel ) {
- # reset when we enter a new level
- $sublevelCount[$level] = 0;
- $toc .= $sk->tocIndent( $level - $prevlevel );
- $toclevel += $level - $prevlevel;
- }
- if( ( $doNumberHeadings || $doShowToc ) && $level < $prevlevel ) {
- # reset when we step back a level
- $sublevelCount[$level+1]=0;
- $toc .= $sk->tocUnindent( $prevlevel - $level );
- $toclevel -= $prevlevel - $level;
- }
- # count number of headlines for each level
- @$sublevelCount[$level]++;
+
if( $doNumberHeadings || $doShowToc ) {
+
+ if ( $level > $prevlevel ) {
+ # Increase TOC level
+ $toclevel++;
+ $sublevelCount[$toclevel] = 0;
+ $toc .= $sk->tocIndent();
+ }
+ elseif ( $level < $prevlevel && $toclevel > 1 ) {
+ # Decrease TOC level, find level to jump to
+
+ if ( $toclevel == 2 && $level <= $levelCount[1] ) {
+ # Can only go down to level 1
+ $toclevel = 1;
+ } else {
+ for ($i = $toclevel; $i > 0; $i--) {
+ if ( $levelCount[$i] == $level ) {
+ # Found last matching level
+ $toclevel = $i;
+ break;
+ }
+ elseif ( $levelCount[$i] < $level ) {
+ # Found first matching level below current level
+ $toclevel = $i + 1;
+ break;
+ }
+ }
+ }
+
+ $toc .= $sk->tocUnindent( $prevtoclevel - $toclevel );
+ }
+ else {
+ # No change in level, end TOC line
+ $toc .= $sk->tocLineEnd();
+ }
+
+ $levelCount[$toclevel] = $level;
+
+ # count number of headlines for each level
+ @$sublevelCount[$toclevel]++;
$dot = 0;
- for( $i = 1; $i <= $level; $i++ ) {
+ for( $i = 1; $i <= $toclevel; $i++ ) {
if( !empty( $sublevelCount[$i] ) ) {
if( $dot ) {
$numbering .= '.';
# The canonized header is a version of the header text safe to use for links
# Avoid insertion of weird stuff like <math> by expanding the relevant sections
$canonized_headline = $this->unstrip( $headline, $this->mStripState );
- $canonized_headline = $this->unstripNoWiki( $headline, $this->mStripState );
+ $canonized_headline = $this->unstripNoWiki( $canonized_headline, $this->mStripState );
# Remove link placeholders by the link text.
# <!--LINK number-->
$canonized_headline = preg_replace( '/<!--LINK ([0-9]*)-->/e',
"\$wgLinkHolders['texts'][\$1]",
$canonized_headline );
+ $canonized_headline = preg_replace( '/<!--IWLINK ([0-9]*)-->/e',
+ "\$wgInterwikiLinkHolders[\$1]",
+ $canonized_headline );
# strip out HTML
$canonized_headline = preg_replace( '/<.*?' . '>/','',$canonized_headline );
@$refers[$canonized_headline]++;
$refcount[$headlineCount]=$refers[$canonized_headline];
- # Prepend the number to the heading text
-
- if( $doNumberHeadings || $doShowToc ) {
- $tocline = $numbering . ' ' . $tocline;
-
- # Don't number the heading if it is the only one (looks silly)
- if( $doNumberHeadings && count( $matches[3] ) > 1) {
- # the two are different if the line contains a link
- $headline=$numbering . ' ' . $headline;
- }
+ # Don't number the heading if it is the only one (looks silly)
+ if( $doNumberHeadings && count( $matches[3] ) > 1) {
+ # the two are different if the line contains a link
+ $headline=$numbering . ' ' . $headline;
}
# Create the anchor for linking from the TOC to the section
$anchor .= '_' . $refcount[$headlineCount];
}
if( $doShowToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) {
- $toc .= $sk->tocLine($anchor,$tocline,$toclevel);
+ $toc .= $sk->tocLine($anchor, $tocline, $numbering, $toclevel);
}
if( $showEditLink && ( !$istemplate || $templatetitle !== "" ) ) {
if ( empty( $head[$headlineCount] ) ) {
if( $istemplate )
$head[$headlineCount] .= $sk->editSectionLinkForOther($templatetitle, $templatesection);
else
- $head[$headlineCount] .= $sk->editSectionLink($sectionCount+1);
+ $head[$headlineCount] .= $sk->editSectionLink($this->mTitle, $sectionCount+1);
}
# Add the edit section span
if( $istemplate )
$headline = $sk->editSectionScriptForOther($templatetitle, $templatesection, $headline);
else
- $headline = $sk->editSectionScript($sectionCount+1,$headline);
+ $headline = $sk->editSectionScript($this->mTitle, $sectionCount+1,$headline);
}
# give headline the correct <h#> tag
if( $doShowToc ) {
$toclines = $headlineCount;
- $toc .= $sk->tocUnindent( $toclevel );
- $toc = $sk->tocTable( $toc );
+ $toc .= $sk->tocUnindent( $toclevel - 1 );
+ $toc = $sk->tocList( $toc );
}
# split up and insert constructed headlines
return $text;
}
- /**
- * Return an HTML link for the "GEO ..." text
- * @access private
- */
- function magicGEO( $text ) {
- global $wgLang, $wgUseGeoMode;
- $fname = 'Parser::magicGEO';
- wfProfileIn( $fname );
-
- # These next five lines are only for the ~35000 U.S. Census Rambot pages...
- $directions = array ( 'N' => 'North' , 'S' => 'South' , 'E' => 'East' , 'W' => 'West' ) ;
- $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['N']}, (\d+)°(\d+)'(\d+)\" {$directions['W']}/" , "(GEO +\$1.\$2.\$3:-\$4.\$5.\$6)" , $text ) ;
- $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['N']}, (\d+)°(\d+)'(\d+)\" {$directions['E']}/" , "(GEO +\$1.\$2.\$3:+\$4.\$5.\$6)" , $text ) ;
- $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['S']}, (\d+)°(\d+)'(\d+)\" {$directions['W']}/" , "(GEO +\$1.\$2.\$3:-\$4.\$5.\$6)" , $text ) ;
- $text = preg_replace ( "/(\d+)°(\d+)'(\d+)\" {$directions['S']}, (\d+)°(\d+)'(\d+)\" {$directions['E']}/" , "(GEO +\$1.\$2.\$3:+\$4.\$5.\$6)" , $text ) ;
-
- $a = split( 'GEO ', ' '.$text );
- if ( count ( $a ) < 2 ) {
- wfProfileOut( $fname );
- return $text;
- }
- $text = substr( array_shift( $a ), 1);
- $valid = '0123456789.+-:';
-
- foreach ( $a as $x ) {
- $geo = $blank = '' ;
- while ( ' ' == $x{0} ) {
- $blank .= ' ';
- $x = substr( $x, 1 );
- }
- while ( strstr( $valid, $x{0} ) != false ) {
- $geo .= $x{0};
- $x = substr( $x, 1 );
- }
- $num = str_replace( '+', '', $geo );
- $num = str_replace( ' ', '', $num );
-
- if ( '' == $num || count ( explode ( ':' , $num , 3 ) ) < 2 ) {
- $text .= "GEO $blank$x";
- } else {
- $titleObj = Title::makeTitle( NS_SPECIAL, 'Geo' );
- $text .= '<a href="' .
- $titleObj->escapeLocalUrl( 'coordinates='.$num ) .
- "\" class=\"internal\">GEO $geo</a>";
- $text .= $x;
- }
- }
- wfProfileOut( $fname );
- return $text;
- }
-
/**
* Return an HTML link for the "RFC 1234" text
* @access private
* @param string $text text to be processed
*/
- function magicRFC( $text ) {
+ function magicRFC( $text, $keyword='RFC ', $urlmsg='rfcurl' ) {
global $wgLang;
$valid = '0123456789';
$internal = false;
- $a = split( 'RFC ', ' '.$text );
- if ( count ( $a ) < 2 ) return $text;
+ $a = split( $keyword, ' '.$text );
+ if ( count ( $a ) < 2 ) {
+ return $text;
+ }
$text = substr( array_shift( $a ), 1);
- /* Check if RFC keyword is preceed by [[.
+ /* Check if keyword is preceed by [[.
* This test is made here cause of the array_shift above
* that prevent the test to be done in the foreach.
*/
- if(substr($text, -2) == '[[') { $internal = true; }
+ if ( substr( $text, -2 ) == '[[' ) {
+ $internal = true;
+ }
foreach ( $a as $x ) {
/* token might be empty if we have RFC RFC 1234 */
- if($x=='') {
- $text.='RFC ';
+ if ( $x=='' ) {
+ $text.=$keyword;
continue;
}
- $rfc = $blank = '' ;
+ $id = $blank = '' ;
/** remove and save whitespaces in $blank */
while ( $x{0} == ' ' ) {
$x = substr( $x, 1 );
}
- /** remove and save the rfc number in $rfc */
+ /** remove and save the rfc number in $id */
while ( strstr( $valid, $x{0} ) != false ) {
- $rfc .= $x{0};
+ $id .= $x{0};
$x = substr( $x, 1 );
}
- if ( $rfc == '') {
+ if ( $id == '' ) {
/* call back stripped spaces*/
- $text .= "RFC $blank$x";
- } elseif( $internal) {
+ $text .= $keyword.$blank.$x;
+ } elseif( $internal ) {
/* normal link */
- $text .= "RFC $rfc$x";
+ $text .= $keyword.$id.$x;
} else {
/* build the external link*/
- $url = wfmsg( 'rfcurl' );
- $url = str_replace( '$1', $rfc, $url);
+ $url = wfmsg( $urlmsg );
+ $url = str_replace( '$1', $id, $url);
$sk =& $this->mOptions->getSkin();
- $la = $sk->getExternalLinkAttributes( $url, 'RFC '.$rfc );
- $text .= "<a href='{$url}'{$la}>RFC {$rfc}</a>{$x}";
+ $la = $sk->getExternalLinkAttributes( $url, $keyword.$id );
+ $text .= "<a href='{$url}'{$la}>{$keyword}{$id}</a>{$x}";
}
/* Check if the next RFC keyword is preceed by [[ */
- $internal = (substr($x,-2) == '[[');
+ $internal = ( substr($x,-2) == '[[' );
}
return $text;
}
$pairs = array(
"\r\n" => "\n",
);
- $text = str_replace(array_keys($pairs), array_values($pairs), $text);
- // now with regexes
- /*
- $pairs = array(
- "/<br.+(clear|break)=[\"']?(all|both)[\"']?\\/?>/i" => '<br style="clear:both;"/>',
- "/<br *?>/i" => "<br />",
- );
- $text = preg_replace(array_keys($pairs), array_values($pairs), $text);
- */
+ $text = str_replace( array_keys( $pairs ), array_values( $pairs ), $text );
$text = $this->strip( $text, $stripState, false );
$text = $this->pstPass2( $text, $user );
$text = $this->unstrip( $text, $stripState );
* @access private
*/
function pstPass2( $text, &$user ) {
- global $wgLang, $wgContLang, $wgLocaltimezone, $wgCurParser;
+ global $wgLang, $wgContLang, $wgLocaltimezone;
# Variable replacement
# Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
$n = $user->getName();
$k = $user->getOption( 'nickname' );
if ( '' == $k ) { $k = $n; }
- if(isset($wgLocaltimezone)) {
- $oldtz = getenv('TZ'); putenv('TZ='.$wgLocaltimezone);
+ if ( isset( $wgLocaltimezone ) ) {
+ $oldtz = getenv( 'TZ' );
+ putenv( 'TZ='.$wgLocaltimezone );
}
/* Note: this is an ugly timezone hack for the European wikis */
$d = $wgContLang->timeanddate( date( 'YmdHis' ), false ) .
' (' . date( 'T' ) . ')';
- if(isset($wgLocaltimezone)) putenv('TZ='.$oldtzs);
+ if ( isset( $wgLocaltimezone ) ) {
+ putenv( 'TZ='.$oldtzs );
+ }
+ if( $user->getOption( 'fancysig' ) ) {
+ $sigText = $k;
+ } else {
+ $sigText = '[[' . $wgContLang->getNsText( NS_USER ) . ":$n|$k]]";
+ }
$text = preg_replace( '/~~~~~/', $d, $text );
- $text = preg_replace( '/~~~~/', '[[' . $wgContLang->getNsText( NS_USER ) . ":$n|$k]] $d", $text );
- $text = preg_replace( '/~~~/', '[[' . $wgContLang->getNsText( NS_USER ) . ":$n|$k]]", $text );
+ $text = preg_replace( '/~~~~/', "$sigText $d", $text );
+ $text = preg_replace( '/~~~/', $sigText, $text );
# Context links: [[|name]] and [[name (context)|]]
#
$this->mTagHooks[$tag] = $callback;
return $oldVal;
}
+
+ /**
+ * Replace <!--LINK--> link placeholders with actual links, in the buffer
+ * Placeholders created in Skin::makeLinkObj()
+ * Returns an array of links found, indexed by PDBK:
+ * 0 - broken
+ * 1 - normal link
+ * 2 - stub
+ * $options is a bit field, RLH_FOR_UPDATE to select for update
+ */
+ function replaceLinkHolders( &$text, $options = 0 ) {
+ global $wgUser, $wgLinkCache, $wgUseOldExistenceCheck, $wgLinkHolders;
+ global $wgInterwikiLinkHolders;
+ global $outputReplace;
+
+ if ( $wgUseOldExistenceCheck ) {
+ return array();
+ }
+
+ $fname = 'Parser::replaceLinkHolders';
+ wfProfileIn( $fname );
+
+ $pdbks = array();
+ $colours = array();
+
+ #if ( !empty( $tmpLinks[0] ) ) { #TODO
+ if ( !empty( $wgLinkHolders['namespaces'] ) ) {
+ wfProfileIn( $fname.'-check' );
+ $dbr =& wfGetDB( DB_SLAVE );
+ $page = $dbr->tableName( 'page' );
+ $sk = $wgUser->getSkin();
+ $threshold = $wgUser->getOption('stubthreshold');
+
+ # Sort by namespace
+ asort( $wgLinkHolders['namespaces'] );
+
+ # Generate query
+ $query = false;
+ foreach ( $wgLinkHolders['namespaces'] as $key => $val ) {
+ # Make title object
+ $title = $wgLinkHolders['titles'][$key];
+
+ # Skip invalid entries.
+ # Result will be ugly, but prevents crash.
+ if ( is_null( $title ) ) {
+ continue;
+ }
+ $pdbk = $pdbks[$key] = $title->getPrefixedDBkey();
+
+ # Check if it's in the link cache already
+ if ( $wgLinkCache->getGoodLinkID( $pdbk ) ) {
+ $colours[$pdbk] = 1;
+ } elseif ( $wgLinkCache->isBadLink( $pdbk ) ) {
+ $colours[$pdbk] = 0;
+ } else {
+ # Not in the link cache, add it to the query
+ if ( !isset( $current ) ) {
+ $current = $val;
+ $tables = $page;
+ $join = '';
+ $query = "SELECT page_id, page_namespace, page_title";
+ if ( $threshold > 0 ) {
+ $textTable = $dbr->tableName( 'text' );
+ $query .= ', LENGTH(old_text) AS page_len, page_is_redirect';
+ $tables .= ", $textTable";
+ $join = 'page_latest=old_id AND';
+ }
+ $query .= " FROM $tables WHERE $join (page_namespace=$val AND page_title IN(";
+ } elseif ( $current != $val ) {
+ $current = $val;
+ $query .= ")) OR (page_namespace=$val AND page_title IN(";
+ } else {
+ $query .= ', ';
+ }
+
+ $query .= $dbr->addQuotes( $wgLinkHolders['dbkeys'][$key] );
+ }
+ }
+ if ( $query ) {
+ $query .= '))';
+ if ( $options & RLH_FOR_UPDATE ) {
+ $query .= ' FOR UPDATE';
+ }
+
+ $res = $dbr->query( $query, $fname );
+
+ # Fetch data and form into an associative array
+ # non-existent = broken
+ # 1 = known
+ # 2 = stub
+ while ( $s = $dbr->fetchObject($res) ) {
+ $title = Title::makeTitle( $s->page_namespace, $s->page_title );
+ $pdbk = $title->getPrefixedDBkey();
+ $wgLinkCache->addGoodLink( $s->page_id, $pdbk );
+
+ if ( $threshold > 0 ) {
+ $size = $s->page_len;
+ if ( $s->page_is_redirect || $s->page_namespace != 0 || $size >= $threshold ) {
+ $colours[$pdbk] = 1;
+ } else {
+ $colours[$pdbk] = 2;
+ }
+ } else {
+ $colours[$pdbk] = 1;
+ }
+ }
+ }
+ wfProfileOut( $fname.'-check' );
+
+ # Construct search and replace arrays
+ wfProfileIn( $fname.'-construct' );
+ $outputReplace = array();
+ foreach ( $wgLinkHolders['namespaces'] as $key => $ns ) {
+ $pdbk = $pdbks[$key];
+ $searchkey = '<!--LINK '.$key.'-->';
+ $title = $wgLinkHolders['titles'][$key];
+ if ( empty( $colours[$pdbk] ) ) {
+ $wgLinkCache->addBadLink( $pdbk );
+ $colours[$pdbk] = 0;
+ $outputReplace[$searchkey] = $sk->makeBrokenLinkObj( $title,
+ $wgLinkHolders['texts'][$key],
+ $wgLinkHolders['queries'][$key] );
+ } elseif ( $colours[$pdbk] == 1 ) {
+ $outputReplace[$searchkey] = $sk->makeKnownLinkObj( $title,
+ $wgLinkHolders['texts'][$key],
+ $wgLinkHolders['queries'][$key] );
+ } elseif ( $colours[$pdbk] == 2 ) {
+ $outputReplace[$searchkey] = $sk->makeStubLinkObj( $title,
+ $wgLinkHolders['texts'][$key],
+ $wgLinkHolders['queries'][$key] );
+ }
+ }
+ wfProfileOut( $fname.'-construct' );
+
+ # Do the thing
+ wfProfileIn( $fname.'-replace' );
+
+ $text = preg_replace_callback(
+ '/(<!--LINK .*?-->)/',
+ "outputReplaceMatches",
+ $text);
+ wfProfileOut( $fname.'-replace' );
+ }
+
+ if ( !empty( $wgInterwikiLinkHolders ) ) {
+ wfProfileIn( $fname.'-interwiki' );
+ $outputReplace = $wgInterwikiLinkHolders;
+ $text = preg_replace_callback(
+ '/<!--IWLINK (.*?)-->/',
+ "outputReplaceMatches",
+ $text );
+ wfProfileOut( $fname.'-interwiki' );
+ }
+
+ wfProfileOut( $fname );
+ return $colours;
+ }
+
+ /**
+ * Renders an image gallery from a text with one line per image.
+ * text labels may be given by using |-style alternative text. E.g.
+ * Image:one.jpg|The number "1"
+ * Image:tree.jpg|A tree
+ * given as text will return the HTML of a gallery with two images,
+ * labeled 'The number "1"' and
+ * 'A tree'.
+ */
+ function renderImageGallery( $text ) {
+ global $wgLinkCache;
+ $ig = new ImageGallery();
+ $ig->setShowBytes( false );
+ $ig->setShowFilename( false );
+ $lines = explode( "\n", $text );
+
+ foreach ( $lines as $line ) {
+ # match lines like these:
+ # Image:someimage.jpg|This is some image
+ preg_match( "/^([^|]+)(\\|(.*))?$/", $line, $matches );
+ # Skip empty lines
+ if ( count( $matches ) == 0 ) {
+ continue;
+ }
+ $nt = Title::newFromURL( $matches[1] );
+ if( is_null( $nt ) ) {
+ # Bogus title. Ignore these so we don't bomb out later.
+ continue;
+ }
+ if ( isset( $matches[3] ) ) {
+ $label = $matches[3];
+ } else {
+ $label = '';
+ }
+
+ # FIXME: Use the full wiki parser and add its links
+ # to the page's links.
+ $html = $this->mOptions->mSkin->formatComment( $label );
+
+ $ig->add( Image::newFromTitle( $nt ), $html );
+ $wgLinkCache->addImageLinkObj( $nt );
+ }
+ return $ig->toHTML();
+ }
}
/**
{
var $mText, $mLanguageLinks, $mCategoryLinks, $mContainsOldMagic;
var $mCacheTime; # Used in ParserCache
+ var $mVersion; # Compatibility check
function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(),
$containsOldMagic = false )
$this->mCategoryLinks = $categoryLinks;
$this->mContainsOldMagic = $containsOldMagic;
$this->mCacheTime = '';
+ $this->mVersion = MW_PARSER_VERSION;
}
- function getText() { return $this->mText; }
- function getLanguageLinks() { return $this->mLanguageLinks; }
- function getCategoryLinks() { return $this->mCategoryLinks; }
- function getCacheTime() { return $this->mCacheTime; }
- function containsOldMagic() { return $this->mContainsOldMagic; }
- function setText( $text ) { return wfSetVar( $this->mText, $text ); }
- function setLanguageLinks( $ll ) { return wfSetVar( $this->mLanguageLinks, $ll ); }
- function setCategoryLinks( $cl ) { return wfSetVar( $this->mCategoryLinks, $cl ); }
+ function getText() { return $this->mText; }
+ function getLanguageLinks() { return $this->mLanguageLinks; }
+ function getCategoryLinks() { return array_keys( $this->mCategoryLinks ); }
+ function getCacheTime() { return $this->mCacheTime; }
+ function containsOldMagic() { return $this->mContainsOldMagic; }
+ function setText( $text ) { return wfSetVar( $this->mText, $text ); }
+ function setLanguageLinks( $ll ) { return wfSetVar( $this->mLanguageLinks, $ll ); }
+ function setCategoryLinks( $cl ) { return wfSetVar( $this->mCategoryLinks, $cl ); }
function setContainsOldMagic( $com ) { return wfSetVar( $this->mContainsOldMagic, $com ); }
- function setCacheTime( $t ) { return wfSetVar( $this->mCacheTime, $t ); }
+ function setCacheTime( $t ) { return wfSetVar( $this->mCacheTime, $t ); }
+ function addCategoryLink( $c ) { $this->mCategoryLinks[$c] = 1; }
function merge( $other ) {
$this->mLanguageLinks = array_merge( $this->mLanguageLinks, $other->mLanguageLinks );
$this->mContainsOldMagic = $this->mContainsOldMagic || $other->mContainsOldMagic;
}
+ /**
+ * Return true if this cached output object predates the global or
+ * per-article cache invalidation timestamps, or if it comes from
+ * an incompatible older version.
+ *
+ * @param string $touched the affected article's last touched timestamp
+ * @return bool
+ * @access public
+ */
+ function expired( $touched ) {
+ global $wgCacheEpoch;
+ return $this->getCacheTime() <= $touched ||
+ $this->getCacheTime() <= $wgCacheEpoch ||
+ !isset( $this->mVersion ) ||
+ version_compare( $this->mVersion, MW_PARSER_VERSION, "lt" );
+ }
}
/**
function setSkin( &$x ) { $this->mSkin =& $x; }
- # Get parser options
- /* static */ function newFromUser( &$user ) {
+ /**
+ * Get parser options
+ * @static
+ */
+ function newFromUser( &$user ) {
$popts = new ParserOptions;
$popts->initialiseFromUser( $user );
return $popts;
}
- # Get user options
+ /** Get user options */
function initialiseFromUser( &$userInput ) {
global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages;
-
$fname = 'ParserOptions::initialiseFromUser';
wfProfileIn( $fname );
if ( !$userInput ) {
$this->mShowToc = $user->getOption( 'showtoc' );
wfProfileOut( $fname );
}
-
-
-}
-
-# Regex callbacks, used in Parser::replaceVariables
-function wfBraceSubstitution( $matches ) {
- global $wgCurParser;
- return $wgCurParser->braceSubstitution( $matches );
-}
-
-function wfArgSubstitution( $matches ) {
- global $wgCurParser;
- return $wgCurParser->argSubstitution( $matches );
}
-function wfVariableSubstitution( $matches ) {
- global $wgCurParser;
- return $wgCurParser->variableSubstitution( $matches );
+/**
+ * Callback function used by Parser::replaceLinkHolders()
+ * to substitute link placeholders.
+ */
+function &outputReplaceMatches( $matches ) {
+ global $outputReplace;
+ return $outputReplace[$matches[1]];
}
/**
if ( -1 != $wgNumberOfArticles ) return;
$dbr =& wfGetDB( DB_SLAVE );
- $s = $dbr->getArray( 'site_stats',
+ $s = $dbr->selectRow( 'site_stats',
array( 'ss_total_views', 'ss_total_edits', 'ss_good_articles' ),
array( 'ss_row_id' => 1 ), $fname
);
}
}
+/**
+ * Escape html tags
+ * Basicly replacing " > and < with HTML entities ( ", >, <)
+ *
+ * @param string $in Text that might contain HTML tags
+ * @return string Escaped string
+ */
function wfEscapeHTMLTagsOnly( $in ) {
return str_replace(
array( '"', '>', '<' ),