<?php
-
/**
* File for Parser and related classes
*
* @package MediaWiki
+ * @subpackage Parser
*/
+/** */
require_once( 'Sanitizer.php' );
/**
* changes in an incompatible way, so the parser cache
* can automatically discard old data.
*/
-define( 'MW_PARSER_VERSION', '1.4.0' );
+define( 'MW_PARSER_VERSION', '1.5.0' );
/**
* Variable substitution O(N^2) attack
$text = $this->strip( $text, $x );
$text = $this->internalParse( $text, $linestart );
+
+ $dashReplace = array(
+ '/ - /' => " – ", # N dash
+ '/(?<=[0-9])-(?=[0-9])/' => "–", # N dash between numbers
+ '/ -- /' => " — " # M dash
+ );
+ $text = preg_replace( array_keys($dashReplace), array_values($dashReplace), $text );
+
+
$text = $this->unstrip( $text, $this->mStripState );
# Clean up special characters, only run once, next-to-last before doBlockLevels
+ global $wgUseTidy;
if(!$wgUseTidy) {
$fixtags = array(
# french spaces, last one Guillemet-left
# only if there is something before the space
- '/(.) (?=\\?|:|;|!|\\302\\273)/i' => '\\1 \\2',
+ '/(.) (?=\\?|:|;|!|\\302\\273)/' => '\\1 \\2',
# french spaces, Guillemet-right
- '/(\\302\\253) /i' => '\\1 ',
+ '/(\\302\\253) /' => '\\1 ',
'/<hr *>/i' => '<hr />',
'/<br *>/i' => '<br />',
'/<center *>/i' => '<div class="center">',
} else {
$fixtags = array(
# french spaces, last one Guillemet-left
- '/ (\\?|:|;|!|\\302\\273)/i' => ' \\1',
+ '/ (\\?|:|;|!|\\302\\273)/' => ' \\1',
# french spaces, Guillemet-right
- '/(\\302\\253) /i' => '\\1 ',
+ '/(\\302\\253) /' => '\\1 ',
'/<center *>/i' => '<div class="center">',
'/<\\/center *>/i' => '</div>'
);
$this->replaceLinkHolders( $text );
$text = $wgContLang->convert($text);
-
+ $this->mOutput->setTitleText($wgContLang->getParsedTitle());
$text = $this->unstripNoWiki( $text, $this->mStripState );
- global $wgUseTidy;
if ($wgUseTidy) {
$text = Parser::tidy($text);
}
while ( '' != $text ) {
if($tag==STRIP_COMMENTS) {
- $p = preg_split( '/<!--/i', $text, 2 );
+ $p = preg_split( '/<!--/', $text, 2 );
} else {
$p = preg_split( "/<\\s*$tag\\s*>/i", $text, 2 );
}
*/
function tidy ( $text ) {
global $wgTidyConf, $wgTidyBin, $wgTidyOpts;
- global $wgInputEncoding, $wgOutputEncoding;
$fname = 'Parser::tidy';
wfProfileIn( $fname );
$cleansource = '';
- $opts = '';
- switch(strtoupper($wgOutputEncoding)) {
- case 'ISO-8859-1':
- $opts .= ($wgInputEncoding == $wgOutputEncoding)? ' -latin1':' -raw';
- break;
- case 'UTF-8':
- $opts .= ($wgInputEncoding == $wgOutputEncoding)? ' -utf8':' -raw';
- break;
- default:
- $opts .= ' -raw';
- }
+ $opts = ' -utf8';
$wrappedtext = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'.
' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>'.
$fc = substr ( $x , 0 , 1 ) ;
if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
$indent_level = strlen( $matches[1] );
- $t[$k] = "\n" .
- str_repeat( '<dl><dd>', $indent_level ) .
+ $t[$k] = str_repeat( '<dl><dd>', $indent_level ) .
'<table' . Sanitizer::fixTagAttributes ( $matches[2], 'table' ) . '>' ;
array_push ( $td , false ) ;
array_push ( $ltd , '' ) ;
}
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);
$l = array_pop ( $ltd ) ;
if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
$text = $m[2];
+ # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
+ # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
+ # the real problem is with the $e1 regex
+ # See bug 1300.
+ if (preg_match( "/^\](.*)/", $m[3], $n ) ) {
+ $text .= ']'; # so that replaceExternalLinks($text) works later
+ $m[3] = $n[1];
+ }
# fix up urlencoded title texts
if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
$trail = $m[3];
# Interwikis
if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) {
array_push( $this->mOutput->mLanguageLinks, $nt->getFullText() );
- $tmp = $prefix . $trail ;
- $s .= (trim($tmp) == '')? '': $tmp;
+ $s = rtrim($s . "\n");
+ $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail;
continue;
}
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);
-
- # 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 );
-
+ if ( !wfIsBadImage( $nt->getDBkey() ) ) {
+ # 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;
+ }
wfProfileOut( "$fname-image" );
- continue;
+
}
if ( $ns == NS_CATEGORY ) {
wfProfileIn( "$fname-category" );
$t = $nt->getText();
+ $s = rtrim($s . "\n"); # bug 87
$wgLinkCache->suspend(); # Don't save in links/brokenlinks
$pPLC=$sk->postParseLinkColour();
}
$wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
$this->mOutput->addCategoryLink( $t );
- $s .= $prefix . $trail ;
+
+ /**
+ * Strip the whitespace Category links produce, see bug 87
+ * @todo We might want to use trim($tmp, "\n") here.
+ */
+ $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail;
wfProfileOut( "$fname-category" );
continue;
* @access private
*/
function getVariableValue( $index ) {
- global $wgContLang, $wgSitename, $wgServer;
+ global $wgContLang, $wgSitename, $wgServer, $wgArticle;
/**
* Some of these require message or data lookups and can be
return $varCache[$index] = $wgContLang->getMonthName( date('n') );
case MAG_CURRENTMONTHNAMEGEN:
return $varCache[$index] = $wgContLang->getMonthNameGen( date('n') );
+ case MAG_CURRENTMONTHABBREV:
+ return $varCache[$index] = $wgContLang->getMonthAbbreviation( date('n') );
case MAG_CURRENTDAY:
return $varCache[$index] = $wgContLang->formatNum( date('j') );
case MAG_PAGENAME:
return $this->mTitle->getText();
case MAG_PAGENAMEE:
return $this->mTitle->getPartialURL();
+ case MAG_REVISIONID:
+ return $wgArticle->getRevIdFetched();
case MAG_NAMESPACE:
# return Namespace::getCanonicalName($this->mTitle->getNamespace());
return $wgContLang->getNsText($this->mTitle->getNamespace()); # Patch by Dori
case MAG_CURRENTDAYNAME:
return $varCache[$index] = $wgContLang->getWeekdayName( date('w')+1 );
case MAG_CURRENTYEAR:
- return $varCache[$index] = $wgContLang->formatNum( date( 'Y' ) );
+ return $varCache[$index] = $wgContLang->formatNum( date( 'Y' ), true );
case MAG_CURRENTTIME:
return $varCache[$index] = $wgContLang->time( wfTimestampNow(), false );
case MAG_CURRENTWEEK:
# merged with the next arg because the '|' character between belongs
# to the link syntax and not the template parameter syntax.
$argc = count($args);
- $i = 0;
+
for ( $i = 0; $i < $argc-1; $i++ ) {
if ( substr_count ( $args[$i], '[[' ) != substr_count ( $args[$i], ']]' ) ) {
$args[$i] .= '|'.$args[$i+1];
global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders, $wgInterwikiLinkHolders;
$doNumberHeadings = $this->mOptions->getNumberHeadings();
- $doShowToc = $this->mOptions->getShowToc();
+ $doShowToc = true;
$forceTocHere = false;
if( !$this->mTitle->userCanEdit() ) {
$showEditLink = 0;
- $rightClickHack = 0;
} else {
$showEditLink = $this->mOptions->getEditSection();
- $rightClickHack = $this->mOptions->getEditSectionOnRightClick();
}
# Inhibit editsection links if requested in the page
# do not add TOC
$mw =& MagicWord::get( MAG_NOTOC );
if( $mw->matchAndRemove( $text ) ) {
- $doShowToc = 0;
- }
-
- # never add the TOC to the Main Page. This is an entry page that should not
- # be more than 1-2 screens large anyway
- if( $this->mTitle->getPrefixedText() == wfMsg('mainpage') ) {
- $doShowToc = 0;
+ $doShowToc = false;
}
# Get all headlines for numbering them and adding funky stuff like [edit]
# if there are fewer than 4 headlines in the article, do not show TOC
if( $numMatches < 4 ) {
- $doShowToc = 0;
+ $doShowToc = false;
}
# if the string __TOC__ (not case-sensitive) occurs in the HTML,
$mw =& MagicWord::get( MAG_TOC );
if($mw->match( $text ) ) {
- $doShowToc = 1;
+ $doShowToc = true;
$forceTocHere = true;
} else {
# if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
# override above conditions and always show TOC above first header
$mw =& MagicWord::get( MAG_FORCETOC );
if ($mw->matchAndRemove( $text ) ) {
- $doShowToc = 1;
+ $doShowToc = true;
}
}
# Never ever show TOC if no headers
if( $numMatches < 1 ) {
- $doShowToc = 0;
+ $doShowToc = false;
}
# We need this to perform operations on the HTML
$head[$headlineCount] .= $sk->editSectionLink($this->mTitle, $sectionCount+1);
}
- # Add the edit section span
- if( $rightClickHack ) {
- if( $istemplate )
- $headline = $sk->editSectionScriptForOther($templatetitle, $templatesection, $headline);
- else
- $headline = $sk->editSectionScript($this->mTitle, $sectionCount+1,$headline);
- }
-
# give headline the correct <h#> tag
@$head[$headlineCount] .= "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount] .$headline.'</h'.$level.'>';
return $text;
}
$text = substr( array_shift( $a ), 1);
- $valid = '0123456789-ABCDEFGHIJKLMNOPQRSTUVWXYZ';
+ $valid = '0123456789-Xx';
foreach ( $a as $x ) {
$isbn = $blank = '' ;
}
$num = str_replace( '-', '', $isbn );
$num = str_replace( ' ', '', $num );
+ $num = str_replace( 'x', 'X', $num );
if ( '' == $num ) {
$text .= "ISBN $blank$x";
/**
* Return an HTML link for the "RFC 1234" text
+ *
* @access private
- * @param string $text text to be processed
+ * @param string $text Text to be processed
+ * @param string $keyword Magic keyword to use (default RFC)
+ * @param string $urlmsg Interface message to use (default rfcurl)
+ * @return string
*/
function magicRFC( $text, $keyword='RFC ', $urlmsg='rfcurl' ) {
global $wgLang;
$text .= $keyword.$id.$x;
} else {
/* build the external link*/
- $url = wfmsg( $urlmsg );
- $url = str_replace( '$1', $id, $url);
+ $url = wfMsg( $urlmsg, $id);
$sk =& $this->mOptions->getSkin();
$la = $sk->getExternalLinkAttributes( $url, $keyword.$id );
$text .= "<a href='{$url}'{$la}>{$keyword}{$id}</a>{$x}";
$oldtz = getenv( 'TZ' );
putenv( 'TZ='.$wgLocaltimezone );
}
- /* Note: this is an ugly timezone hack for the European wikis */
- $d = $wgContLang->timeanddate( date( 'YmdHis' ), false ) .
+
+ /* Note: This is the timestamp saved as hardcoded wikitext to
+ * the database, we use $wgContLang here in order to give
+ * everyone the same signiture and use the default one rather
+ * than the one selected in each users preferences.
+ */
+ $d = $wgContLang->timeanddate( wfTimestampNow(), false, false) .
' (' . date( 'T' ) . ')';
if ( isset( $wgLocaltimezone ) ) {
putenv( 'TZ='.$oldtzs );
# 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 .= ', page_len, page_is_redirect';
}
- $query .= " FROM $tables WHERE $join (page_namespace=$val AND page_title IN(";
+ $query .= " FROM $page WHERE (page_namespace=$val AND page_title IN(";
} elseif ( $current != $val ) {
$current = $val;
$query .= ")) OR (page_namespace=$val AND page_title IN(";
* 'A tree'.
*/
function renderImageGallery( $text ) {
+ # Setup the parser
+ global $wgUser, $wgParser, $wgTitle;
+ $parserOptions = ParserOptions::newFromUser( $wgUser );
+
global $wgLinkCache;
$ig = new ImageGallery();
$ig->setShowBytes( false );
$label = '';
}
- # FIXME: Use the full wiki parser and add its links
- # to the page's links.
- $html = $this->mOptions->mSkin->formatComment( $label );
+ $html = $wgParser->parse( $label , $wgTitle, $parserOptions );
+ $html = $html->mText;
- $ig->add( Image::newFromTitle( $nt ), $html );
+ $ig->add( new Image( $nt ), $html );
$wgLinkCache->addImageLinkObj( $nt );
}
return $ig->toHTML();
var $mText, $mLanguageLinks, $mCategoryLinks, $mContainsOldMagic;
var $mCacheTime; # Used in ParserCache
var $mVersion; # Compatibility check
+ var $mTitleText; # title text of the chosen language variant
function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(),
- $containsOldMagic = false )
+ $containsOldMagic = false, $titletext = '' )
{
$this->mText = $text;
$this->mLanguageLinks = $languageLinks;
$this->mContainsOldMagic = $containsOldMagic;
$this->mCacheTime = '';
$this->mVersion = MW_PARSER_VERSION;
+ $this->mTitleText = $titletext;
}
function getText() { return $this->mText; }
function getLanguageLinks() { return $this->mLanguageLinks; }
function getCategoryLinks() { return array_keys( $this->mCategoryLinks ); }
function getCacheTime() { return $this->mCacheTime; }
+ function getTitleText() { return $this->mTitleText; }
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 setTitleText( $t ) { return wfSetVar ($this->mTitleText, $t); }
+
function addCategoryLink( $c ) { $this->mCategoryLinks[$c] = 1; }
function merge( $other ) {
var $mSkin; # Reference to the preferred skin
var $mDateFormat; # Date format index
var $mEditSection; # Create "edit section" links
- var $mEditSectionOnRightClick; # Generate JavaScript to edit section on right click
var $mNumberHeadings; # Automatically number headings
- var $mShowToc; # Show table of contents
function getUseTeX() { return $this->mUseTeX; }
function getUseDynamicDates() { return $this->mUseDynamicDates; }
function getSkin() { return $this->mSkin; }
function getDateFormat() { return $this->mDateFormat; }
function getEditSection() { return $this->mEditSection; }
- function getEditSectionOnRightClick() { return $this->mEditSectionOnRightClick; }
function getNumberHeadings() { return $this->mNumberHeadings; }
- function getShowToc() { return $this->mShowToc; }
function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX, $x ); }
function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates, $x ); }
function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages, $x ); }
function setDateFormat( $x ) { return wfSetVar( $this->mDateFormat, $x ); }
function setEditSection( $x ) { return wfSetVar( $this->mEditSection, $x ); }
- function setEditSectionOnRightClick( $x ) { return wfSetVar( $this->mEditSectionOnRightClick, $x ); }
function setNumberHeadings( $x ) { return wfSetVar( $this->mNumberHeadings, $x ); }
- function setShowToc( $x ) { return wfSetVar( $this->mShowToc, $x ); }
function setSkin( &$x ) { $this->mSkin =& $x; }
wfProfileOut( $fname.'-skin' );
$this->mDateFormat = $user->getOption( 'date' );
$this->mEditSection = $user->getOption( 'editsection' );
- $this->mEditSectionOnRightClick = $user->getOption( 'editsectiononrightclick' );
$this->mNumberHeadings = $user->getOption( 'numberheadings' );
- $this->mShowToc = $user->getOption( 'showtoc' );
wfProfileOut( $fname );
}
}
dépôts / lhc / web / wiklou.git / blobdiff