<?php
-require_once('Tokenizer.php');
+// require_once('Tokenizer.php');
if( $GLOBALS['wgUseWikiHiero'] ){
require_once('extensions/wikihiero/wikihiero.php');
# Clean up special characters, only run once, next-to-last before doBlockLevels
if(!$wgUseTidy) {
$fixtags = array(
+ # french spaces, last one Guillemet-left
+ "/ (\\?|:|!|\\302\\273)/i"=>" \\1",
+ # french spaces, Guillemet-right
+ "/(\\302\\253) /i"=>"\\1 ",
"/<hr *>/i" => '<hr/>',
"/<br *>/i" => '<br/>',
"/<center *>/i"=>'<div class="center">',
$text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
} else {
$fixtags = array(
+ # french spaces, last one Guillemet-left
+ "/ (\\?|:|!|\\302\\273)/i"=>" \\1",
+ # french spaces, Guillemet-right
+ "/(\\302\\253) /i"=>"\\1 ",
"/<center *>/i"=>'<div class="center">',
"/<\\/center *>/i" => '</div>'
);
$render = ($this->mOutputType == OT_HTML);
$nowiki_content = array();
$hiero_content = array();
+ $timeline_content = array();
$math_content = array();
$pre_content = array();
$comment_content = array();
$hiero_content[$marker] = "<hiero>$content</hiero>";
}
}
+
+ $text = Parser::extractTags("timeline", $text, $timeline_content, $uniq_prefix);
+ foreach( $timeline_content as $marker => $content ){
+ if( $render && $GLOBALS['wgUseTimeline']){
+ $timeline_content[$marker] = renderTimeline( $content );
+ } else {
+ $timeline_content[$marker] = "<timeline>$content</timeline>";
+ }
+ }
$text = Parser::extractTags("math", $text, $math_content, $uniq_prefix);
foreach( $math_content as $marker => $content ){
if ( $state ) {
$state['nowiki'] = $state['nowiki'] + $nowiki_content;
$state['hiero'] = $state['hiero'] + $hiero_content;
+ $state['timeline'] = $state['timeline'] + $timeline_content;
$state['math'] = $state['math'] + $math_content;
$state['pre'] = $state['pre'] + $pre_content;
$state['comment'] = $state['comment'] + $comment_content;
$state = array(
'nowiki' => $nowiki_content,
'hiero' => $hiero_content,
+ 'timeline' => $timeline_content,
'math' => $math_content,
'pre' => $pre_content,
'comment' => $comment_content
$data = array () ;
$id = $this->mTitle->getArticleID() ;
- # For existing categories
- if( $id ) {
- $sql = "SELECT DISTINCT cur_title,cur_namespace FROM cur,links WHERE l_to={$id} AND l_from=cur_id";
- $res = wfQuery ( $sql, DB_READ ) ;
- while ( $x = wfFetchObject ( $res ) ) $data[] = $x ;
- } else {
- # For non-existing categories
- $t = wfStrencode( $this->mTitle->getPrefixedDBKey() );
- $sql = "SELECT DISTINCT cur_title,cur_namespace FROM cur,brokenlinks WHERE bl_to='$t' AND bl_from=cur_id" ;
- $res = wfQuery ( $sql, DB_READ ) ;
- while ( $x = wfFetchObject ( $res ) ) $data[] = $x ;
- }
+ # FIXME: add limits
+ $t = wfStrencode( $this->mTitle->getDBKey() );
+ $sql = "SELECT DISTINCT cur_title,cur_namespace FROM cur,categorylinks WHERE cl_to='$t' AND cl_from=cur_id ORDER BY cl_sortkey" ;
+ $res = wfQuery ( $sql, DB_READ ) ;
+ while ( $x = wfFetchObject ( $res ) ) $data[] = $x ;
# For all pages that link to this category
foreach ( $data AS $x )
wfFreeResult ( $res ) ;
# Showing subcategories
- if ( count ( $children ) > 0 )
- {
- asort ( $children ) ;
+ if ( count ( $children ) > 0 ) {
$r .= "<h2>".wfMsg("subcategories")."</h2>\n" ;
$r .= implode ( ", " , $children ) ;
}
# Showing pages in this category
- if ( count ( $articles ) > 0 )
- {
+ if ( count ( $articles ) > 0 ) {
$ti = $this->mTitle->getText() ;
- asort ( $articles ) ;
$h = wfMsg( "category_header", $ti );
$r .= "<h2>{$h}</h2>\n" ;
$r .= implode ( ", " , $articles ) ;
return $t ;
}
+ # Parses the text and adds the result to the strip state
+ # Returns the strip tag
+ function stripParse( $text, $linestart, $args )
+ {
+ $text = $this->strip( $text, $this->mStripState );
+ $text = $this->internalParse( $text, $linestart, $args, false );
+ if( $linestart ) {
+ $text = "\n" . $text;
+ }
+ return $this->insertStripItem( $text, $this->mStripState );
+ }
+
function internalParse( $text, $linestart, $args = array(), $isMain=true )
{
$fname = "Parser::internalParse";
$text = $this->removeHTMLtags( $text );
$text = $this->replaceVariables( $text, $args );
- # $text = preg_replace( "/(^|\n)-----*/", "\\1<hr>", $text );
+ $text = preg_replace( "/(^|\n)-----*/", "\\1<hr/>", $text );
$text = $this->doHeadings( $text );
if($this->mOptions->getUseDynamicDates()) {
global $wgDateFormatter;
$text = $wgDateFormatter->reformat( $this->mOptions->getDateFormat(), $text );
}
+ $text = $this->doAllQuotes( $text );
$text = $this->replaceExternalLinks( $text );
- $text = $this->doTokenizedParser ( $text );
+ $text = $this->replaceInternalLinks ( $text );
+ $text = $this->replaceInternalLinks ( $text );
+ //$text = $this->doTokenizedParser ( $text );
$text = $this->doTableStuff ( $text ) ;
+ $text = $this->magicISBN( $text );
+ $text = $this->magicRFC( $text );
$text = $this->formatHeadings( $text, $isMain );
$sk =& $this->mOptions->getSkin();
$text = $sk->transformContent( $text );
if ( !isset ( $this->categoryMagicDone ) ) {
- $text .= $this->categoryMagic () ;
- $this->categoryMagicDone = true ;
- }
+ $text .= $this->categoryMagic () ;
+ $this->categoryMagicDone = true ;
+ }
wfProfileOut( $fname );
return $text;
return $text;
}
+ /* private */ function doAllQuotes( $text )
+ {
+ $outtext = "";
+ $lines = explode( "\r\n", $text );
+ foreach ( $lines as $line ) {
+ $outtext .= $this->doQuotes ( "", $line, "" ) . "\r\n";
+ }
+ return $outtext;
+ }
+
+ /* private */ function doQuotes( $pre, $text, $mode )
+ {
+ if ( preg_match( "/^(.*)''(.*)$/sU", $text, $m ) ) {
+ $m1_strong = ($m[1] == "") ? "" : "<strong>{$m[1]}</strong>";
+ $m1_em = ($m[1] == "") ? "" : "<em>{$m[1]}</em>";
+ if ( substr ($m[2], 0, 1) == "'" ) {
+ $m[2] = substr ($m[2], 1);
+ if ($mode == "em") {
+ return $this->doQuotes ( $m[1], $m[2], ($m[1] == "") ? "both" : "emstrong" );
+ } else if ($mode == "strong") {
+ return $m1_strong . $this->doQuotes ( "", $m[2], "" );
+ } else if (($mode == "emstrong") || ($mode == "both")) {
+ return $this->doQuotes ( "", $pre.$m1_strong.$m[2], "em" );
+ } else if ($mode == "strongem") {
+ return "<strong>{$pre}{$m1_em}</strong>" . $this->doQuotes ( "", $m[2], "em" );
+ } else {
+ return $m[1] . $this->doQuotes ( "", $m[2], "strong" );
+ }
+ } else {
+ if ($mode == "strong") {
+ return $this->doQuotes ( $m[1], $m[2], ($m[1] == "") ? "both" : "strongem" );
+ } else if ($mode == "em") {
+ return $m1_em . $this->doQuotes ( "", $m[2], "" );
+ } else if ($mode == "emstrong") {
+ return "<em>{$pre}{$m1_strong}</em>" . $this->doQuotes ( "", $m[2], "strong" );
+ } else if (($mode == "strongem") || ($mode == "both")) {
+ return $this->doQuotes ( "", $pre.$m1_em.$m[2], "strong" );
+ } else {
+ return $m[1] . $this->doQuotes ( "", $m[2], "em" );
+ }
+ }
+ } else {
+ $text_strong = ($text == "") ? "" : "<strong>{$text}</strong>";
+ $text_em = ($text == "") ? "" : "<em>{$text}</em>";
+ if ($mode == "") {
+ return $pre . $text;
+ } else if ($mode == "em") {
+ return $pre . $text_em;
+ } else if ($mode == "strong") {
+ return $pre . $text_strong;
+ } else if ($mode == "strongem") {
+ return (($pre == "") && ($text == "")) ? "" : "<strong>{$pre}{$text_em}</strong>";
+ } else {
+ return (($pre == "") && ($text == "")) ? "" : "<em>{$pre}{$text_strong}</em>";
+ }
+ }
+ }
+
# 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.
return $s;
}
- /* private */ function handle4Quotes( &$state, $token )
- {
- /* This one makes some assumptions.
- * '''Caesar''''s army => <strong>Caesar</strong>'s army
- * ''''Caesar'''' was a roman emperor => '<strong>Caesar</strong>' was a roman emperor
- * These assumptions might be wrong, but any other assumption might be wrong, too.
- * So here we go */
- if ( $state["strong"] !== false ) {
- return $this->handle3Quotes( $state, $token ) . "'";
- } else {
- return "'" . $this->handle3Quotes( $state, $token );
- }
- }
-
-
- /* private */ function handle3Quotes( &$state, $token )
- {
- if ( $state["strong"] !== false ) {
- if ( $state["em"] !== false && $state["em"] > $state["strong"] )
- {
- # ''' lala ''lala '''
- $s = "</em></strong><em>";
- } else {
- $s = "</strong>";
- }
- $state["strong"] = FALSE;
- } else {
- $s = "<strong>";
- $state["strong"] = $token["pos"];
- }
- return $s;
- }
-
- /* private */ function handle2Quotes( &$state, $token )
- {
- if ( $state["em"] !== false ) {
- if ( $state["strong"] !== false && $state["strong"] > $state["em"] )
- {
- # ''lala'''lala'' ....'''
- $s = "</strong></em><strong>";
- } else {
- $s = "</em>";
- }
- $state["em"] = FALSE;
- } else {
- $s = "<em>";
- $state["em"] = $token["pos"];
-
- }
- return $s;
- }
-
- /* private */ function handle5Quotes( &$state, $token )
- {
- $s = "";
- if ( $state["em"] !== false && $state["strong"] !== false ) {
- if ( $state["em"] < $state["strong"] ) {
- $s .= "</strong></em>";
- } else {
- $s .= "</em></strong>";
- }
- $state["strong"] = $state["em"] = FALSE;
- } elseif ( $state["em"] !== false ) {
- $s .= "</em><strong>";
- $state["em"] = FALSE;
- $state["strong"] = $token["pos"];
- } elseif ( $state["strong"] !== false ) {
- $s .= "</strong><em>";
- $state["strong"] = FALSE;
- $state["em"] = $token["pos"];
- } else { # not $em and not $strong
- $s .= "<strong><em>";
- $state["strong"] = $state["em"] = $token["pos"];
- }
- return $s;
- }
- /* private */ function doTokenizedParser( $str )
- {
- global $wgLang; # for language specific parser hook
- global $wgUploadDirectory, $wgUseTimeline;
-
- $tokenizer=Tokenizer::newFromString( $str );
- $tokenStack = array();
-
- $s="";
- $state["em"] = FALSE;
- $state["strong"] = FALSE;
- $tagIsOpen = FALSE;
- $threeopen = false;
-
- # The tokenizer splits the text into tokens and returns them one by one.
- # Every call to the tokenizer returns a new token.
- while ( $token = $tokenizer->nextToken() )
- {
- switch ( $token["type"] )
- {
- case "text":
- # simple text with no further markup
- $txt = $token["text"];
- break;
- case "blank":
- # Text that contains blanks that have to be converted to
- # non-breakable spaces for French.
- # U+202F NARROW NO-BREAK SPACE might be a better choice, but
- # browser support for Unicode spacing is poor.
- $txt = str_replace( " ", " ", $token["text"] );
- break;
- case "[[[":
- # remember the tag opened with 3 [
- $threeopen = true;
- case "[[":
- # link opening tag.
- # FIXME : Treat orphaned open tags (stack not empty when text is over)
- $tagIsOpen = TRUE;
- array_push( $tokenStack, $token );
- $txt="";
- break;
-
- case "]]]":
- case "]]":
- # link close tag.
- # get text from stack, glue it together, and call the code to handle a
- # link
-
- if ( count( $tokenStack ) == 0 )
- {
- # stack empty. Found a ]] without an opening [[
- $txt = "]]";
- } else {
- $linkText = "";
- $lastToken = array_pop( $tokenStack );
- while ( !(($lastToken["type"] == "[[[") or ($lastToken["type"] == "[[")) )
- {
- if( !empty( $lastToken["text"] ) ) {
- $linkText = $lastToken["text"] . $linkText;
- }
- $lastToken = array_pop( $tokenStack );
- }
-
- $txt = $linkText ."]]";
-
- if( isset( $lastToken["text"] ) ) {
- $prefix = $lastToken["text"];
- } else {
- $prefix = "";
- }
- $nextToken = $tokenizer->previewToken();
- if ( $nextToken["type"] == "text" )
- {
- # Preview just looks at it. Now we have to fetch it.
- $nextToken = $tokenizer->nextToken();
- $txt .= $nextToken["text"];
- }
- $txt = $this->handleInternalLink( $this->unstrip($txt,$this->mStripState), $prefix );
-
- # did the tag start with 3 [ ?
- if($threeopen) {
- # show the first as text
- $txt = "[".$txt;
- $threeopen=false;
- }
-
- }
- $tagIsOpen = (count( $tokenStack ) != 0);
- break;
- case "----":
- $txt = "\n<hr />\n";
- break;
- case "'''":
- # This and the four next ones handle quotes
- $txt = $this->handle3Quotes( $state, $token );
- break;
- case "''":
- $txt = $this->handle2Quotes( $state, $token );
- break;
- case "'''''":
- $txt = $this->handle5Quotes( $state, $token );
- break;
- case "''''":
- $txt = $this->handle4Quotes( $state, $token );
- break;
- case "":
- # empty token
- $txt="";
- break;
- case "RFC ":
- if ( $tagIsOpen ) {
- $txt = "RFC ";
- } else {
- $txt = $this->doMagicRFC( $tokenizer );
- }
- break;
- case "ISBN ":
- if ( $tagIsOpen ) {
- $txt = "ISBN ";
- } else {
- $txt = $this->doMagicISBN( $tokenizer );
- }
- break;
- case "<timeline>":
- if ( $wgUseTimeline &&
- "" != ( $timelinesrc = $tokenizer->readAllUntil("</timeline>") ) )
- {
- $txt = renderTimeline( $timelinesrc );
- } else {
- $txt=$token["text"];
- }
- break;
- default:
- # Call language specific Hook.
- $txt = $wgLang->processToken( $token, $tokenStack );
- if ( NULL == $txt ) {
- # An unkown token. Highlight.
- $txt = "<font color=\"#FF0000\"><b>".$token["type"]."</b></font>";
- $txt .= "<font color=\"#FFFF00\"><b>".$token["text"]."</b></font>";
- }
- break;
- }
- # If we're parsing the interior of a link, don't append the interior to $s,
- # but push it to the stack so it can be processed when a ]] token is found.
- if ( $tagIsOpen && $txt != "" ) {
- $token["type"] = "text";
- $token["text"] = $txt;
- array_push( $tokenStack, $token );
- } else {
- $s .= $txt;
- }
- } #end while
- if ( count( $tokenStack ) != 0 )
- {
- # still objects on stack. opened [[ tag without closing ]] tag.
- $txt = "";
- while ( $lastToken = array_pop( $tokenStack ) )
- {
- if ( $lastToken["type"] == "text" )
- {
- $txt = $lastToken["text"] . $txt;
- } else {
- $txt = $lastToken["type"] . $txt;
- }
- }
- $s .= $txt;
- }
- return $s;
- }
-
- /* private */ function handleInternalLink( $line, $prefix )
+ /* private */ function replaceInternalLinks( $s )
{
global $wgLang, $wgLinkCache;
global $wgNamespacesWithSubpages, $wgLanguageCode;
- static $fname = "Parser::handleInternalLink" ;
+ static $fname = "Parser::replaceInternalLink" ;
wfProfileIn( $fname );
wfProfileIn( "$fname-setup" );
static $tc = FALSE;
- if ( !$tc ) { $tc = Title::legalChars() . "#"; }
+ # the % is needed to support urlencoded titles as well
+ if ( !$tc ) { $tc = Title::legalChars() . "#%"; }
$sk =& $this->mOptions->getSkin();
+ $a = explode( "[[", " " . $s );
+ $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"; }
$nottalk = !Namespace::isTalk( $this->mTitle->getNamespace() );
+ if ( $wgLang->linkPrefixExtension() && preg_match( $e2, $s, $m ) ) {
+ $new_prefix = $m[2];
+ $s = $m[1];
+ } else {
+ $new_prefix="";
+ }
+
wfProfileOut( "$fname-setup" );
- $s = "";
- if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
- $text = $m[2];
- $trail = $m[3];
- } else { # Invalid form; output directly
- $s .= $prefix . "[[" . $line ;
- return $s;
- }
+ foreach ( $a as $line ) {
+ $prefix = $new_prefix;
- /* Valid link forms:
- Foobar -- normal
- :Foobar -- override special treatment of prefix (images, language links)
- /Foobar -- convert to CurrentPage/Foobar
- /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text
- */
- $c = substr($m[1],0,1);
- $noforce = ($c != ":");
- if( $c == "/" ) { # subpage
- if(substr($m[1],-1,1)=="/") { # / at end means we don't want the slash to be shown
- $m[1]=substr($m[1],1,strlen($m[1])-2);
- $noslash=$m[1];
- } else {
- $noslash=substr($m[1],1);
+ 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];
+ } else { # Invalid form; output directly
+ $s .= $prefix . "[[" . $line ;
+ wfProfileOut( $fname );
+ continue;
}
- if($wgNamespacesWithSubpages[$this->mTitle->getNamespace()]) { # subpages allowed here
- $link = $this->mTitle->getPrefixedText(). "/" . trim($noslash);
- if( "" == $text ) {
- $text= $m[1];
- } # this might be changed for ugliness reasons
+
+ /* Valid link forms:
+ Foobar -- normal
+ :Foobar -- override special treatment of prefix (images, language links)
+ /Foobar -- convert to CurrentPage/Foobar
+ /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text
+ */
+ $c = substr($m[1],0,1);
+ $noforce = ($c != ":");
+ if( $c == "/" ) { # subpage
+ if(substr($m[1],-1,1)=="/") { # / at end means we don't want the slash to be shown
+ $m[1]=substr($m[1],1,strlen($m[1])-2);
+ $noslash=$m[1];
+ } else {
+ $noslash=substr($m[1],1);
+ }
+ if(!empty($wgNamespacesWithSubpages[$this->mTitle->getNamespace()])) { # subpages allowed here
+ $link = $this->mTitle->getPrefixedText(). "/" . trim($noslash);
+ if( "" == $text ) {
+ $text= $m[1];
+ } # this might be changed for ugliness reasons
+ } else {
+ $link = $noslash; # no subpage allowed, use standard link
+ }
+ } elseif( $noforce ) { # no subpage
+ $link = $m[1];
} else {
- $link = $noslash; # no subpage allowed, use standard link
+ $link = substr( $m[1], 1 );
}
- } elseif( $noforce ) { # no subpage
- $link = $m[1];
- } else {
- $link = substr( $m[1], 1 );
- }
- if( "" == $text )
+ $wasblank = ( "" == $text );
+ if( $wasblank )
$text = $link;
- $nt = Title::newFromText( $link );
- if( !$nt ) {
- $s .= $prefix . "[[" . $line;
- return $s;
- }
- $ns = $nt->getNamespace();
- $iw = $nt->getInterWiki();
- if( $noforce ) {
- if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgLang->getLanguageName( $iw ) ) {
- array_push( $this->mOutput->mLanguageLinks, $nt->getPrefixedText() );
- $s .= $prefix . $trail ;
- return (trim($s) == '')? '': $s;
- }
- if( $ns == $image ) {
- $s .= $prefix . $sk->makeImageLinkObj( $nt, $text ) . $trail;
- $wgLinkCache->addImageLinkObj( $nt );
- return $s;
+ $nt = Title::newFromText( $link );
+ if( !$nt ) {
+ $s .= $prefix . "[[" . $line;
+ wfProfileOut( $fname );
+ continue;
}
- if ( $ns == $category ) {
- $t = $nt->getText() ;
- $nnt = Title::newFromText ( Namespace::getCanonicalName($category).":".$t ) ;
- $t = $sk->makeLinkObj( $nnt, $t, "", "" , $prefix );
- $this->mOutput->mCategoryLinks[] = $t ;
- $s .= $prefix . $trail ;
- return $s ;
+ $ns = $nt->getNamespace();
+ $iw = $nt->getInterWiki();
+ if( $noforce ) {
+ if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgLang->getLanguageName( $iw ) ) {
+ array_push( $this->mOutput->mLanguageLinks, $nt->getPrefixedText() );
+ $s .= $prefix . $trail ;
+ wfProfileOut( $fname );
+ $s .= (trim($s) == '')? '': $s;
+ continue;
+ }
+ if ( $ns == $image ) {
+ $s .= $prefix . $sk->makeImageLinkObj( $nt, $text ) . $trail;
+ $wgLinkCache->addImageLinkObj( $nt );
+ wfProfileOut( $fname );
+ continue;
+ }
+ if ( $ns == $category ) {
+ $t = $nt->getText() ;
+ $nnt = Title::newFromText ( Namespace::getCanonicalName($category).":".$t ) ;
+
+ $wgLinkCache->suspend(); # Don't save in links/brokenlinks
+ $t = $sk->makeLinkObj( $nnt, $t, "", "" , $prefix );
+ $wgLinkCache->resume();
+
+ $sortkey = $wasblank ? $this->mTitle->getPrefixedText() : $text;
+ $wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
+ $this->mOutput->mCategoryLinks[] = $t ;
+ $s .= $prefix . $trail ;
+ wfProfileOut( $fname );
+ continue;
+ }
+ }
+ if( ( $nt->getPrefixedText() == $this->mTitle->getPrefixedText() ) &&
+ ( strpos( $link, "#" ) == FALSE ) ) {
+ # Self-links are handled specially; generally de-link and change to bold.
+ $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, "", $trail );
+ wfProfileOut( $fname );
+ continue;
}
- }
- if( ( $nt->getPrefixedText() == $this->mTitle->getPrefixedText() ) &&
- ( strpos( $link, "#" ) == FALSE ) ) {
- # Self-links are handled specially; generally de-link and change to bold.
- $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, "", $trail );
- return $s;
- }
- if( $ns == $media ) {
- $s .= $prefix . $sk->makeMediaLinkObj( $nt, $text ) . $trail;
- $wgLinkCache->addImageLinkObj( $nt );
- return $s;
- } elseif( $ns == $special ) {
- $s .= $prefix . $sk->makeKnownLinkObj( $nt, $text, "", $trail );
- return $s;
+ if( $ns == $media ) {
+ $s .= $prefix . $sk->makeMediaLinkObj( $nt, $text ) . $trail;
+ $wgLinkCache->addImageLinkObj( $nt );
+ wfProfileOut( $fname );
+ continue;
+ } elseif( $ns == $special ) {
+ $s .= $prefix . $sk->makeKnownLinkObj( $nt, $text, "", $trail );
+ wfProfileOut( $fname );
+ continue;
+ }
+ $s .= $sk->makeLinkObj( $nt, $text, "", $trail , $prefix );
}
- $s .= $sk->makeLinkObj( $nt, $text, "", $trail , $prefix );
-
wfProfileOut( $fname );
return $s;
}
# 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|<div|<pre|<tr|<td|<p|<ul|<li)/i", $t );
+ $openmatch = preg_match("/(<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<td|<p|<ul|<li)/i", $t );
$closematch = preg_match(
"/(<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|".
- "<\\/div|<hr|<\\/td|<\\/pre|<\\/p|".$uniq_prefix."-pre|<\\/li|<\\/ul)/i", $t );
+ "<div|<\\/div|<hr|<\\/td|<\\/pre|<\\/p|".$uniq_prefix."-pre|<\\/li|<\\/ul)/i", $t );
if ( $openmatch or $closematch ) {
$paragraphStack = false;
$output .= $this->closeParagraph();
} else {
$inBlockElem = true;
}
- } else if ( !$inBlockElem ) {
- if ( " " == $t{0} ) {
+ } else if ( !$inBlockElem && !$this->mInPre ) {
+ if ( " " == $t{0} and trim($t) != '' ) {
// pre
if ($this->mLastSection != 'pre') {
$paragraphStack = false;
$this->initialiseVariables();
}
$titleChars = Title::legalChars();
- $regex = "/(\\n?){{([$titleChars]*?)(\\|.*?|)}}/s";
# 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;
+
+ # Argument substitution
+ if ( $this->mOutputType == OT_HTML ) {
+ $text = preg_replace_callback( "/(\\n?){{{([$titleChars]*?)}}}/", "wfArgSubstitution", $text );
+ }
+ # Double brace substitution
+ $regex = "/(\\n?){{([$titleChars]*)(\\|.*?|)}}/s";
$text = preg_replace_callback( $regex, "wfBraceSubstitution", $text );
array_pop( $this->mArgStack );
+ wfProfileOut( $fname );
return $text;
}
$fname = "Parser::braceSubstitution";
$found = false;
$nowiki = false;
+ $noparse = false;
+
$title = NULL;
# $newline is an optional newline character before the braces
$args = array();
}
$argc = count( $args );
+
+ # {{{}}}
+ if ( strpos( $matches[0], "{{{" ) !== false ) {
+ $text = $matches[0];
+ $found = true;
+ $noparse = true;
+ }
# SUBST
- $mwSubst =& MagicWord::get( MAG_SUBST );
- if ( $mwSubst->matchStartAndRemove( $part1 ) ) {
- if ( $this->mOutputType != OT_WIKI ) {
- # Invalid SUBST not replaced at PST time
- # Return without further processing
+ if ( !$found ) {
+ $mwSubst =& MagicWord::get( MAG_SUBST );
+ if ( $mwSubst->matchStartAndRemove( $part1 ) ) {
+ if ( $this->mOutputType != OT_WIKI ) {
+ # Invalid SUBST not replaced at PST time
+ # Return without further processing
+ $text = $matches[0];
+ $found = true;
+ $noparse= true;
+ }
+ } elseif ( $this->mOutputType == OT_WIKI ) {
+ # SUBST not found in PST pass, do nothing
$text = $matches[0];
$found = true;
}
- } elseif ( $this->mOutputType == OT_WIKI ) {
- # SUBST not found in PST pass, do nothing
- $text = $matches[0];
- $found = true;
}
# MSG, MSGNW and INT
$found = true;
$this->mOutput->mContainsOldMagic = true;
}
-
+/*
# Arguments input from the caller
$inputArgs = end( $this->mArgStack );
if ( !$found && array_key_exists( $part1, $inputArgs ) ) {
$text = $inputArgs[$part1];
$found = true;
}
-
+*/
# Load from database
if ( !$found ) {
$title = Title::newFromText( $part1, NS_TEMPLATE );
# Only for HTML output
if ( $nowiki && $found && $this->mOutputType == OT_HTML ) {
$text = wfEscapeWikiText( $text );
- } elseif ( $this->mOutputType == OT_HTML && $found ) {
+ } elseif ( $this->mOutputType == OT_HTML && $found && !$noparse) {
# Clean up argument array
$assocArgs = array();
$index = 1;
}
# Run full parser on the included text
- $text = $this->strip( $text, $this->mStripState );
- $text = $this->internalParse( $text, (bool)$newline, $assocArgs, false );
-
- # Add the result to the strip state for re-inclusion after
- # the rest of the processing
- $text = $this->insertStripItem( $text, $this->mStripState );
+ $text = $this->stripParse( $text, (bool)$newline, $assocArgs );
# Resume the link cache and register the inclusion as a link
if ( !is_null( $title ) ) {
}
}
+ # Triple brace replacement -- used for template arguments
+ function argSubstitution( $matches )
+ {
+ $newline = $matches[1];
+ $arg = trim( $matches[2] );
+ $text = $matches[0];
+ $inputArgs = end( $this->mArgStack );
+
+ if ( array_key_exists( $arg, $inputArgs ) ) {
+ $text = $this->stripParse( $inputArgs[$arg], (bool)$newline, array() );
+ }
+
+ return $text;
+ }
+
# Returns true if the function is allowed to include this entity
function incrementIncludeCount( $dbk )
{
$canonized_headline = preg_replace( "/<.*?" . ">/","",$canonized_headline );
$tocline = trim( $canonized_headline );
$canonized_headline = preg_replace("/[ \\?&\\/<>\\(\\)\\[\\]=,+']+/", '_', urlencode( do_html_entity_decode( $tocline, ENT_COMPAT, $wgInputEncoding ) ) );
+ # strip out urlencoded (inserted for french spaces, e.g. first space in 'something : something')
+ $canonized_headline = str_replace('%C2%A0','_', $canonized_headline);
$refer[$headlineCount] = $canonized_headline;
# count how many in assoc. array so we can track dupes in anchors
return $full;
}
- /* private */ function doMagicISBN( &$tokenizer )
+ /* private */ function magicISBN( $text )
{
global $wgLang;
- # Check whether next token is a text token
- # If yes, fetch it and convert the text into a
- # Special::BookSources link
- $token = $tokenizer->previewToken();
- while ( $token["type"] == "" )
- {
- $tokenizer->nextToken();
- $token = $tokenizer->previewToken();
- }
- if ( $token["type"] == "text" )
- {
- $token = $tokenizer->nextToken();
- $x = $token["text"];
- $valid = "0123456789-ABCDEFGHIJKLMNOPQRSTUVWXYZ";
+ $a = split( "ISBN ", " $text" );
+ if ( count ( $a ) < 2 ) return $text;
+ $text = substr( array_shift( $a ), 1);
+ $valid = "0123456789-ABCDEFGHIJKLMNOPQRSTUVWXYZ";
+ foreach ( $a as $x ) {
$isbn = $blank = "" ;
while ( " " == $x{0} ) {
$blank .= " ";
$num = str_replace( " ", "", $num );
if ( "" == $num ) {
- $text = "ISBN $blank$x";
+ $text .= "ISBN $blank$x";
} else {
$titleObj = Title::makeTitle( NS_SPECIAL, "Booksources" );
- $text = "<a href=\"" .
+ $text .= "<a href=\"" .
$titleObj->escapeLocalUrl( "isbn={$num}" ) .
"\" class=\"internal\">ISBN $isbn</a>";
$text .= $x;
}
- } else {
- $text = "ISBN ";
}
return $text;
}
- /* private */ function doMagicRFC( &$tokenizer )
+ /* private */ function magicRFC( $text )
{
global $wgLang;
- # Check whether next token is a text token
- # If yes, fetch it and convert the text into a
- # link to an RFC source
- $token = $tokenizer->previewToken();
- while ( $token["type"] == "" )
- {
- $tokenizer->nextToken();
- $token = $tokenizer->previewToken();
- }
- if ( $token["type"] == "text" )
- {
- $token = $tokenizer->nextToken();
- $x = $token["text"];
- $valid = "0123456789";
+ $a = split( "RFC ", " $text" );
+ if ( count ( $a ) < 2 ) return $text;
+ $text = substr( array_shift( $a ), 1);
+ $valid = "0123456789";
+ foreach ( $a as $x ) {
$rfc = $blank = "" ;
while ( " " == $x{0} ) {
$blank .= " ";
$url = str_replace( "$1", $rfc, $url);
$sk =& $this->mOptions->getSkin();
$la = $sk->getExternalLinkAttributes( $url, "RFC {$rfc}" );
- $text = "<a href='{$url}'{$la}>RFC {$rfc}</a>{$x}";
+ $text .= "<a href='{$url}'{$la}>RFC {$rfc}</a>{$x}";
}
- } else {
- $text = "RFC ";
}
return $text;
}
class ParserOutput
{
var $mText, $mLanguageLinks, $mCategoryLinks, $mContainsOldMagic;
+ var $mTouched; # Used for caching
function ParserOutput( $text = "", $languageLinks = array(), $categoryLinks = array(),
$containsOldMagic = false )
$this->mLanguageLinks = $languageLinks;
$this->mCategoryLinks = $categoryLinks;
$this->mContainsOldMagic = $containsOldMagic;
+ $this->mTouched = "";
}
function getText() { return $this->mText; }
function getLanguageLinks() { return $this->mLanguageLinks; }
function getCategoryLinks() { return $this->mCategoryLinks; }
+ function getTouched() { return $this->mTouched; }
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 setTouched( $t ) { return wfSetVar( $this->mTouched, $t ); }
function merge( $other ) {
$this->mLanguageLinks = array_merge( $this->mLanguageLinks, $other->mLanguageLinks );
function wfBraceSubstitution( $matches )
{
global $wgCurParser;
+ $titleChars = Title::legalChars();
+
+ # not really nested stuff, just multiple includes separated by titlechars
+ if(preg_match("/^([^}{]*)}}([^}{]*{{)(.*)$/s", $matches[2], $m)) {
+ $text = wfInternalBraceSubstitution( $m[1] );
+ $string = $text.$m[2].$m[3];
+ while(preg_match("/^([^}{]*){{([$titleChars]*?)(}}[^}{]*{{.*)?$/s", $string, $m)) {
+ $text = wfInternalBraceSubstitution( $m[2] );
+ $trail = !empty($m[3])? preg_replace("/^}}/", '', $m[3]):'';
+ $string = $m[1].$text.$trail;
+ }
+ return $string;
+ }
+
+ # Double brace substitution, expand bar in {{foo{{bar}}}}
+ $i = 0;
+ while(preg_match("/{{([$titleChars]*?)}}/", $matches[2], $internalmatches) and $i < 30) {
+ $text = wfInternalBraceSubstitution( $internalmatches[1] );
+ $matches[0] = str_replace($internalmatches[0], $text , $matches[0]);
+ $matches[2] = str_replace($internalmatches[0], $text , $matches[2]);
+ $i++;
+ }
+
return $wgCurParser->braceSubstitution( $matches );
}
+function wfArgSubstitution( $matches )
+{
+ global $wgCurParser;
+ return $wgCurParser->argSubstitution( $matches );
+}
+
+# XXX: i don't think this is the most elegant way to do it..
+function wfInternalBraceSubstitution( $part1 ) {
+ global $wgLinkCache, $wgLang, $wgCurParser;
+ $fname = "wfInternalBraceSubstitution";
+ $found = false;
+ $nowiki = false;
+ $noparse = false;
+
+ $title = NULL;
+
+ # $newline is an optional newline character before the braces
+ # $part1 is the bit before the first |, and must contain only title characters
+ # $args is a list of arguments, starting from index 0, not including $part1
+
+ # SUBST
+ if ( !$found ) {
+ $mwSubst =& MagicWord::get( MAG_SUBST );
+ if ( $mwSubst->matchStartAndRemove( $part1 ) ) {
+ if ( $wgCurParser->mOutputType != OT_WIKI ) {
+ # Invalid SUBST not replaced at PST time
+ # Return without further processing
+ $text = $matches[0];
+ $found = true;
+ $noparse= true;
+ }
+ } elseif ( $wgCurParser->mOutputType == OT_WIKI ) {
+ # SUBST not found in PST pass, do nothing
+ $text = $matches[0];
+ $found = true;
+ }
+ }
+
+ # MSG, MSGNW and INT
+ if ( !$found ) {
+ # Check for MSGNW:
+ $mwMsgnw =& MagicWord::get( MAG_MSGNW );
+ if ( $mwMsgnw->matchStartAndRemove( $part1 ) ) {
+ $nowiki = true;
+ } else {
+ # Remove obsolete MSG:
+ $mwMsg =& MagicWord::get( MAG_MSG );
+ $mwMsg->matchStartAndRemove( $part1 );
+ }
+
+ # Check if it is an internal message
+ $mwInt =& MagicWord::get( MAG_INT );
+ if ( $mwInt->matchStartAndRemove( $part1 ) ) {
+ if ( $wgCurParser->incrementIncludeCount( "int:$part1" ) ) {
+ $text = wfMsgReal( $part1, array(), true );
+ $found = true;
+ }
+ }
+ }
+
+ # NS
+ if ( !$found ) {
+ # Check for NS: (namespace expansion)
+ $mwNs = MagicWord::get( MAG_NS );
+ if ( $mwNs->matchStartAndRemove( $part1 ) ) {
+ if ( intval( $part1 ) ) {
+ $text = $wgLang->getNsText( intval( $part1 ) );
+ $found = true;
+ } else {
+ $index = Namespace::getCanonicalIndex( strtolower( $part1 ) );
+ if ( !is_null( $index ) ) {
+ $text = $wgLang->getNsText( $index );
+ $found = true;
+ }
+ }
+ }
+ }
+
+ # LOCALURL and LOCALURLE
+ if ( !$found ) {
+ $mwLocal = MagicWord::get( MAG_LOCALURL );
+ $mwLocalE = MagicWord::get( MAG_LOCALURLE );
+
+ if ( $mwLocal->matchStartAndRemove( $part1 ) ) {
+ $func = 'getLocalURL';
+ } elseif ( $mwLocalE->matchStartAndRemove( $part1 ) ) {
+ $func = 'escapeLocalURL';
+ } else {
+ $func = '';
+ }
+
+ if ( $func !== '' ) {
+ $title = Title::newFromText( $part1 );
+ if ( !is_null( $title ) ) {
+ $text = $title->$func();
+ $found = true;
+ }
+ }
+ }
+
+ # Internal variables
+ if ( !$found && array_key_exists( $part1, $wgCurParser->mVariables ) ) {
+ $text = $wgCurParser->mVariables[$part1];
+ $found = true;
+ $wgCurParser->mOutput->mContainsOldMagic = true;
+ }
+
+ # Load from database
+ if ( !$found ) {
+ $title = Title::newFromText( $part1, NS_TEMPLATE );
+ if ( !is_null( $title ) && !$title->isExternal() ) {
+ # Check for excessive inclusion
+ $dbk = $title->getPrefixedDBkey();
+ if ( $wgCurParser->incrementIncludeCount( $dbk ) ) {
+ $article = new Article( $title );
+ $articleContent = $article->getContentWithoutUsingSoManyDamnGlobals();
+ if ( $articleContent !== false ) {
+ $found = true;
+ $text = $articleContent;
+
+ }
+ }
+
+ # If the title is valid but undisplayable, make a link to it
+ if ( $wgCurParser->mOutputType == OT_HTML && !$found ) {
+ $text = "[[" . $title->getPrefixedText() . "]]";
+ $found = true;
+ }
+ }
+ }
+
+ if ( !$found ) {
+ return $matches[0];
+ } else {
+ return $text;
+ }
+}
?>