* @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
*
* 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( 'INVERSE_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]' );
* 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!
*
* @return ParserOutput a ParserOutput
*/
function parse( $text, &$title, $options, $linestart = true, $clearState = true ) {
- global $wgUseTidy;
+ global $wgUseTidy, $wgContLang;
$fname = 'Parser::parse';
wfProfileIn( $fname );
$stripState = NULL;
$text = $this->strip( $text, $this->mStripState );
+
$text = $this->internalParse( $text, $linestart );
$text = $this->unstrip( $text, $this->mStripState );
# Clean up special characters, only run once, next-to-last before doBlockLevels
}
# 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
array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
}
else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) { # Caption
+ # $x is a table row
if ( '|+' == substr ( $x , 0 , 2 ) ) {
$fc = '+' ;
$x = substr ( $x , 1 ) ;
if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;
$after = explode ( '||' , $after ) ;
$t[$k] = '' ;
+
+ # Loop through each table cell
foreach ( $after AS $theline )
{
$z = '' ;
else if ( $fc == '+' ) $l = 'caption' ;
else $l = '' ;
array_push ( $ltd , $l ) ;
+
+ # Cell parameters
$y = explode ( '|' , $theline , 2 ) ;
- if ( count ( $y ) == 1 ) $y = "{$z}<{$l}>{$y[0]}" ;
+ # Note that a '|' inside an invalid link should not
+ # be mistaken as delimiting cell parameters
+ if ( strpos( $y[0], '[[' ) !== false ) {
+ $y = array ($theline);
+ }
+ if ( count ( $y ) == 1 )
+ $y = "{$z}<{$l}>{$y[0]}" ;
else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
$t[$k] .= $y ;
array_push ( $td , true ) ;
* @access private
*/
function internalParse( $text, $linestart, $args = array(), $isMain=true ) {
- global $wgContLang;
+ global $wgContLang;
$fname = 'Parser::internalParse';
wfProfileIn( $fname );
$text = $this->removeHTMLtags( $text );
$text = $this->replaceVariables( $text, $args );
- $text = $wgContLang->convert($text);
-
$text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
$text = $this->doHeadings( $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->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 );
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
*/
$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 );
# 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 ) . $dtrail. $paren . $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( '/((?:'.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?
$s .= $protocol . $remainder;
}
}
+ wfProfileOut();
return $s;
}
*
* @access private
*/
+
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"; }
+ # 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() );
$prefix = '';
}
+ $selflink = $this->mTitle->getPrefixedText();
wfProfileOut( $fname.'-setup' );
- # start procedeeding each line
- foreach ( $a as $line ) {
- wfProfileIn( $fname.'-prefixhandling' );
+ $checkVariantLink = sizeof($wgContLang->getVariants())>1;
+ # 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('/^((?:'.URL_PROTOCOLS.'):)/', $m[1])) {
$s .= $prefix . '[[' . $line ;
continue;
}
$link = substr($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( $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 .= '[[' . $next_line;
+ break;
+ }
+ }
+ if ( !$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;
+ }
+ }
+
+ $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);
+
+ # 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;
$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;
}
/**
* 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/ -- convert to CurrentPage/Foobar, strip the initial / from text
global $wgNamespacesWithSubpages;
+ $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
$ret = $target;
}
+ wfProfileOut( $fname );
return $ret;
}
# ; title : definition text
# So we check for : in the remainder text to split up the
# title and definition, without b0rking links.
- # FIXME: This is not foolproof. Something better in Tokenizer might help.
- if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
- $term = $match[1];
+ if ($this->findColonNoLinks($t, $term, $t2) !== false) {
+ $t = $t2;
$output .= $term . $this->nextItem( ':' );
- $t = $match[2];
}
}
} elseif( $prefixLength || $lastPrefixLength ) {
if ( ';' == $char ) {
# FIXME: This is dupe of code above
- if( preg_match( '/^(.*?(?:\s| )):(.*)$/', $t, $match ) ) {
- $term = $match[1];
+ if ($this->findColonNoLinks($t, $term, $t2) !== false) {
+ $t = $t2;
$output .= $term . $this->nextItem( ':' );
- $t = $match[2];
}
}
++$commonPrefixLength;
$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";
return $output;
}
+ /**
+ * Split up a string on ':', ignoring any occurences inside
+ * <a>..</a> or <span>...</span>
+ * @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) {
+ # I wonder if we should make this count all tags, not just <a>
+ # and <span>. That would prevent us from matching a ':' that
+ # comes in the middle of italics other such formatting....
+ # -- Wil
+ $fname = 'Parser::findColonNoLinks';
+ wfProfileIn( $fname );
+ $pos = 0;
+ do {
+ $colon = strpos($str, ':', $pos);
+
+ if ($colon !== false) {
+ $before = substr($str, 0, $colon);
+ $after = substr($str, $colon + 1);
+
+ # Skip any ':' within <a> or <span> pairs
+ $a = substr_count($before, '<a');
+ $s = substr_count($before, '<span');
+ $ca = substr_count($before, '</a>');
+ $cs = substr_count($before, '</span>');
+
+ if ($a <= $ca and $s <= $cs) {
+ # Tags are balanced before ':'; ok
+ break;
+ }
+ $pos = $colon + 1;
+ }
+ } while ($colon !== false);
+ wfProfileOut( $fname );
+ return $colon;
+ }
+
/**
* Return value of a magic variable (like PAGENAME)
*
*/
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_NUMBEROFARTICLES:
- return $wgContLang->formatNum( wfNumberOfArticles() );
+ return $varCache[$index] = $wgContLang->formatNum( wfNumberOfArticles() );
case MAG_SITENAME:
return $wgSitename;
case MAG_SERVER:
* @access private
*/
function initialiseVariables() {
- $fname = 'Parser::initialiseVaraibles';
+ $fname = 'Parser::initialiseVariables';
wfProfileIn( $fname );
global $wgVariableIDs;
$this->mVariables = array();
$mw =& MagicWord::get( $id );
$mw->addToArray( $this->mVariables, $this->getVariableValue( $id ) );
}
- $fname = 'Parser::initialiseVaraibles';
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 );
# Empties the template path
$this->mTemplatePath = array();
+
if ( !$found ) {
return $matches[0];
} else {
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
* @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 );
+ }
- $text = preg_replace( '/~~~~~/', $d, $text );
+ $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 );
$this->mTagHooks[$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 );
+ $cur = $dbr->tableName( 'cur' );
+ $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;
+ $query = "SELECT cur_id, cur_namespace, cur_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(";
+ } elseif ( $current != $val ) {
+ $current = $val;
+ $query .= ")) OR (cur_namespace=$val AND cur_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->cur_namespace, $s->cur_title );
+ $pdbk = $title->getPrefixedDBkey();
+ $wgLinkCache->addGoodLink( $s->cur_id, $pdbk );
+
+ if ( $threshold > 0 ) {
+ $size = $s->cur_len;
+ if ( $s->cur_is_redirect || $s->cur_namespace != 0 || $length < $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 ( isset( $matches[3] ) ) {
+ $label = $matches[3];
+ } else {
+ $label = '';
+ }
+ $ig->add( Image::newFromTitle( $nt ), $label );
+ $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 ) {
}
-# 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
);
dépôts / lhc / web / wiklou.git / blobdiff