* File for Parser and related classes
*
* @package MediaWiki
- * @version $Id$
*/
+/**
+ * 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
*
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!
*
$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 );
$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;
$text = $wgDateFormatter->reformat( $this->mOptions->getDateFormat(), $text );
}
$text = $this->doAllQuotes( $text );
+ $text = $this->replaceInternalLinks( $text );
$text = $this->replaceExternalLinks( $text );
- $text = $this->replaceInternalLinks ( $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;
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 ) );
$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 = '';
- }
- }
-
# 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 );
# This means that users can paste URLs directly into the text
# Funny characters like ö aren't valid in URLs anyway
# This was changed in August 2004
- $s .= $sk->makeExternalLink( $url, $text, false ) . $dtrail. $paren . $trail;
+ $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;
$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;
}
global $wgLang, $wgContLang, $wgLinkCache;
global $wgDisableLangConversion;
static $fname = 'Parser::replaceInternalLinks' ;
- # use a counter to prevent too much unknown links from
- # being checked for different language variants.
+
wfProfileIn( $fname );
wfProfileIn( $fname.'-setup' );
# 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"; }
$prefix = '';
}
+ $selflink = $this->mTitle->getPrefixedText();
wfProfileOut( $fname.'-setup' );
$checkVariantLink = sizeof($wgContLang->getVariants())>1;
+ $useSubpages = $this->areSubpagesAllowed();
+
# Loop for each link
for ($k = 0; isset( $a[$k] ); $k++) {
$line = $a[$k];
- wfProfileIn( $fname.'-prefixhandling' );
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;
# 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);
}
- $nt = Title::newFromText( $this->unstripNoWiki($link, $this->mStripState) );
+ $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
- && $checkVariantLink ) {
+ if( $checkVariantLink
+ && $nt->getArticleID() == 0 ) {
$wgContLang->findVariantLink($link, $nt);
}
}
if ( $ns == NS_IMAGE ) {
+ wfProfileIn( "$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);
- # replace the image with a link-holder so that replaceExternalLinks() can't mess with it
- $s .= $prefix . $this->insertStripItem( $sk->makeImageLinkObj( $nt, $text ), $this->mStripState ) . $trail;
+ # 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 = $nt->getText() ;
+ wfProfileIn( "$fname-category" );
+ $t = $nt->getText();
$wgLinkCache->suspend(); # Don't save in links/brokenlinks
$pPLC=$sk->postParseLinkColour();
$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() ) &&
+ 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 );
# 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 ) {
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 string $target the source of the link
# :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";
*/
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;
# Empties the template path
$this->mTemplatePath = array();
if ( !$found ) {
+ wfProfileOut( $fname );
return $matches[0];
} else {
# replace ==section headers==
# Empties the template path
$this->mTemplatePath = array();
+
if ( !$found ) {
+ wfProfileOut( $fname );
return $matches[0];
} else {
+ wfProfileOut( $fname );
return $text;
}
}
'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'
+ 'ruby', 'rt' , 'rb' , 'rp', 'p', 'span'
);
$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'
+ 'dl', 'font', 'big', 'small', 'sub', 'sup', 'span'
);
$tabletags = array( # Can only appear inside table
'td', 'th', 'tr'
* @access private
*/
/* private */ function formatHeadings( $text, $isMain=true ) {
- global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders;
+ 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 )
$headline = $sk->editSectionScriptForOther($templatetitle, $templatesection, $headline);
else
- $headline = $sk->editSectionScript($this->title, $sectionCount+1,$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
* @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, $text );
# Context links: [[|name]] and [[name (context)|]]
#
*/
function replaceLinkHolders( &$text, $options = 0 ) {
global $wgUser, $wgLinkCache, $wgUseOldExistenceCheck, $wgLinkHolders;
+ global $wgInterwikiLinkHolders;
+ global $outputReplace;
if ( $wgUseOldExistenceCheck ) {
return array();
if ( !empty( $wgLinkHolders['namespaces'] ) ) {
wfProfileIn( $fname.'-check' );
$dbr =& wfGetDB( DB_SLAVE );
- $cur = $dbr->tableName( 'cur' );
+ $page = $dbr->tableName( 'page' );
$sk = $wgUser->getSkin();
$threshold = $wgUser->getOption('stubthreshold');
# Not in the link cache, add it to the query
if ( !isset( $current ) ) {
$current = $val;
- $query = "SELECT cur_id, cur_namespace, cur_title";
+ $tables = $page;
+ $join = '';
+ $query = "SELECT page_id, page_namespace, page_title";
if ( $threshold > 0 ) {
- $query .= ", LENGTH(cur_text) AS cur_len, cur_is_redirect";
- }
- $query .= " FROM $cur WHERE (cur_namespace=$val AND cur_title IN(";
+ $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;
- $query .= ")) OR (cur_namespace=$val AND cur_title IN(";
+ $query .= ")) OR (page_namespace=$val AND page_title IN(";
} else {
$query .= ', ';
}
# 1 = known
# 2 = stub
while ( $s = $dbr->fetchObject($res) ) {
- $title = Title::makeTitle( $s->cur_namespace, $s->cur_title );
+ $title = Title::makeTitle( $s->page_namespace, $s->page_title );
$pdbk = $title->getPrefixedDBkey();
- $wgLinkCache->addGoodLink( $s->cur_id, $pdbk );
+ $wgLinkCache->addGoodLink( $s->page_id, $pdbk );
if ( $threshold > 0 ) {
- $size = $s->cur_len;
- if ( $s->cur_is_redirect || $s->cur_namespace != 0 || $length < $threshold ) {
+ $size = $s->page_len;
+ if ( $s->page_is_redirect || $s->page_namespace != 0 || $length < $threshold ) {
$colours[$pdbk] = 1;
} else {
$colours[$pdbk] = 2;
# Construct search and replace arrays
wfProfileIn( $fname.'-construct' );
- global $outputReplace;
$outputReplace = array();
foreach ( $wgLinkHolders['namespaces'] as $key => $ns ) {
$pdbk = $pdbks[$key];
"outputReplaceMatches",
$text);
wfProfileOut( $fname.'-replace' );
+ }
+ if ( !empty( $wgInterwikiLinkHolders ) ) {
wfProfileIn( $fname.'-interwiki' );
- global $wgInterwikiLinkHolders;
$outputReplace = $wgInterwikiLinkHolders;
$text = preg_replace_callback(
'/<!--IWLINK (.*?)-->/',
"outputReplaceMatches",
- $text);
+ $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( $this->mVersion, MW_PARSER_VERSION, "lt" );
+ }
}
/**
# Get user options
function initialiseFromUser( &$userInput ) {
global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages;
-
$fname = 'ParserOptions::initialiseFromUser';
wfProfileIn( $fname );
if ( !$userInput ) {
* Callback function used by Parser::replaceLinkHolders()
* to substitute link placeholders.
*/
-function &outputReplaceMatches($matches) {
+function &outputReplaceMatches( $matches ) {
global $outputReplace;
return $outputReplace[$matches[1]];
}
-
-# 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 );
-}
-
/**
* Return the total number of articles
*/
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
);