var $mTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables;
# Cleared with clearState():
- var $mOutput, $mAutonumber, $mDTopen, $mStripState = array();
+ var $mOutput, $mAutonumber, $mDTopen, $mStripState;
var $mIncludeCount, $mArgStack, $mLastSection, $mInPre;
var $mInterwikiLinkHolders, $mLinkHolders, $mUniqPrefix;
var $mIncludeSizes;
$mTitle, // Title context, used for self-link rendering and similar things
$mOutputType, // Output type, one of the OT_xxx constants
$ot, // Shortcut alias, see setOutputType()
- $mRevisionId; // ID to display in {{REVISIONID}} tags
+ $mRevisionId, // ID to display in {{REVISIONID}} tags
+ $mRevisionTimestamp, // The timestamp of the specified revision ID
+ $mRevIdForTs; // The revision ID which was used to fetch the timestamp
/**#@-*/
}
$this->initialiseVariables();
-
$this->mFirstCall = false;
wfProfileOut( __METHOD__ );
}
$this->mLastSection = '';
$this->mDTopen = false;
$this->mIncludeCount = array();
- $this->mStripState = array();
+ $this->mStripState = new StripState;
$this->mArgStack = array();
$this->mInPre = false;
$this->mInterwikiLinkHolders = array(
'texts' => array(),
'titles' => array()
);
- $this->mRevisionId = null;
-
+ $this->mRevisionTimestamp = $this->mRevisionId = null;
+
/**
* Prefix for temporary replacement strings for the multipass parser.
* \x07 should never appear in input as it's disallowed in XML.
global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang;
$fname = 'Parser::parse-' . wfGetCaller();
+ wfProfileIn( __METHOD__ );
wfProfileIn( $fname );
if ( $clearState ) {
$this->mOptions = $options;
$this->mTitle =& $title;
$oldRevisionId = $this->mRevisionId;
+ $oldRevisionTimestamp = $this->mRevisionTimestamp;
if( $revid !== null ) {
$this->mRevisionId = $revid;
+ $this->mRevisionTimestamp = null;
}
$this->setOutputType( OT_HTML );
-
- //$text = $this->strip( $text, $this->mStripState );
- // VOODOO MAGIC FIX! Sometimes the above segfaults in PHP5.
- $x =& $this->mStripState;
-
- wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$x ) );
- $text = $this->strip( $text, $x );
- wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$x ) );
-
+ wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) );
+ $text = $this->strip( $text, $this->mStripState );
+ wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) );
$text = $this->internalParse( $text );
-
- $text = $this->unstrip( $text, $this->mStripState );
+ $text = $this->mStripState->unstripGeneral( $text );
# Clean up special characters, only run once, next-to-last before doBlockLevels
$fixtags = array(
# Side-effects: this calls $this->mOutput->setTitleText()
$text = $wgContLang->parserConvert( $text, $this );
- $text = $this->unstripNoWiki( $text, $this->mStripState );
+ $text = $this->mStripState->unstripNoWiki( $text );
wfRunHooks( 'ParserBeforeTidy', array( &$this, &$text ) );
}
$this->mOutput->setText( $text );
$this->mRevisionId = $oldRevisionId;
+ $this->mRevisionTimestamp = $oldRevisionTimestamp;
wfProfileOut( $fname );
+ wfProfileOut( __METHOD__ );
return $this->mOutput;
}
*/
function recursiveTagParse( $text ) {
wfProfileIn( __METHOD__ );
- $x =& $this->mStripState;
- wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$x ) );
- $text = $this->strip( $text, $x );
- wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$x ) );
+ wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) );
+ $text = $this->strip( $text, $this->mStripState );
+ wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) );
$text = $this->internalParse( $text );
wfProfileOut( __METHOD__ );
return $text;
$this->setOutputType( OT_PREPROCESS );
$this->mOptions = $options;
$this->mTitle = $title;
- $x =& $this->mStripState;
- wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$x ) );
- $text = $this->strip( $text, $x );
- wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$x ) );
+ wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) );
+ $text = $this->strip( $text, $this->mStripState );
+ wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) );
if ( $this->mOptions->getRemoveComments() ) {
$text = Sanitizer::removeHTMLcomments( $text );
}
$text = $this->replaceVariables( $text );
- $text = $this->unstrip( $text, $x );
- $text = $this->unstripNowiki( $text, $x );
+ $text = $this->mStripState->unstripBoth( $text );
wfProfileOut( __METHOD__ );
return $text;
}
* Strips and renders nowiki, pre, math, hiero
* If $render is set, performs necessary rendering operations on plugins
* Returns the text, and fills an array with data needed in unstrip()
- * If the $state is already a valid strip state, it adds to the state
+ *
+ * @param StripState $state
*
* @param bool $stripcomments when set, HTML comments <!-- like this -->
* will be stripped in addition to other tags. This is important
*
* @private
*/
- function strip( $text, &$state, $stripcomments = false , $dontstrip = array () ) {
+ function strip( $text, $state, $stripcomments = false , $dontstrip = array () ) {
+ global $wgContLang;
wfProfileIn( __METHOD__ );
$render = ($this->mOutputType == OT_HTML);
$uniq_prefix = $this->mUniqPrefix;
- $commentState = array();
+ $commentState = new ReplacementArray;
$elements = array_merge(
array( 'nowiki', 'gallery' ),
}
// Shouldn't happen otherwise. :)
case 'nowiki':
- $output = wfEscapeHTMLTagsOnly( $content );
+ $output = Xml::escapeTagsOnly( $content );
break;
case 'math':
- $output = MathRenderer::renderMath( $content );
+ $output = $wgContLang->armourMath( MathRenderer::renderMath( $content ) );
break;
case 'gallery':
$output = $this->renderImageGallery( $content, $params );
// Unstrip the output, because unstrip() is no longer recursive so
// it won't do it itself
- $output = $this->unstrip( $output, $state );
+ $output = $state->unstripBoth( $output );
if( !$stripcomments && $element == '!--' ) {
- $commentState[$marker] = $output;
+ $commentState->setPair( $marker, $output );
} elseif ( $element == 'html' || $element == 'nowiki' ) {
- $state['nowiki'][$marker] = $output;
+ $state->nowiki->setPair( $marker, $output );
} else {
- $state['general'][$marker] = $output;
+ $state->general->setPair( $marker, $output );
}
}
# a comment.)
if ( !$stripcomments ) {
// Put them all back and forget them
- $text = strtr( $text, $commentState );
+ $text = $commentState->replace( $text );
}
wfProfileOut( __METHOD__ );
*
* always call unstripNoWiki() after this one
* @private
+ * @deprecated use $this->mStripState->unstrip()
*/
function unstrip( $text, $state ) {
- if ( !isset( $state['general'] ) ) {
- return $text;
- }
-
- wfProfileIn( __METHOD__ );
- # TODO: good candidate for FSS
- $text = strtr( $text, $state['general'] );
- wfProfileOut( __METHOD__ );
- return $text;
+ return $state->unstripGeneral( $text );
}
/**
* Always call this after unstrip() to preserve the order
*
* @private
+ * @deprecated use $this->mStripState->unstrip()
*/
function unstripNoWiki( $text, $state ) {
- if ( !isset( $state['nowiki'] ) ) {
- return $text;
- }
-
- wfProfileIn( __METHOD__ );
- # TODO: good candidate for FSS
- $text = strtr( $text, $state['nowiki'] );
- wfProfileOut( __METHOD__ );
+ return $state->unstripNoWiki( $text );
+ }
- return $text;
+ /**
+ * @deprecated use $this->mStripState->unstripBoth()
+ */
+ function unstripForHTML( $text ) {
+ return $this->mStripState->unstripBoth( $text );
}
/**
*/
function insertStripItem( $text, &$state ) {
$rnd = $this->mUniqPrefix . '-item' . Parser::getRandomString();
- if ( !$state ) {
- $state = array();
- }
- $state['general'][$rnd] = $text;
+ $state->general->setPair( $rnd, $text );
return $rnd;
}
*
* @private
*/
- function doTableStuff ( $t ) {
+ function doTableStuff ( $text ) {
$fname = 'Parser::doTableStuff';
wfProfileIn( $fname );
- $t = explode ( "\n" , $t ) ;
- $td = array () ; # Is currently a td tag open?
- $ltd = array () ; # Was it TD or TH?
- $tr = array () ; # Is currently a tr tag open?
- $ltr = array () ; # tr attributes
- $has_opened_tr = array(); # Did this table open a <tr> element?
- $indent_level = 0; # indent level of the table
- foreach ( $t AS $k => $x )
+ $lines = explode ( "\n" , $text );
+ $td_history = array (); // Is currently a td tag open?
+ $last_tag_history = array (); // Save history of last lag activated (td, th or caption)
+ $tr_history = array (); // Is currently a tr tag open?
+ $tr_attributes = array (); // history of tr attributes
+ $has_opened_tr = array(); // Did this table open a <tr> element?
+ $indent_level = 0; // indent level of the table
+ foreach ( $lines as $key => $line )
{
- $x = trim ( $x ) ;
- $fc = substr ( $x , 0 , 1 ) ;
+ $line = trim ( $line );
+
+ if( $line == '' ) { // empty line, go to next line
+ continue;
+ }
+ $first_character = $line{0};
$matches = array();
- if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
+
+ if ( preg_match( '/^(:*)\{\|(.*)$/' , $line , $matches ) ) {
+ // First check if we are starting a new table
$indent_level = strlen( $matches[1] );
- $attributes = $this->unstripForHTML( $matches[2] );
+ $attributes = $this->mStripState->unstripBoth( $matches[2] );
+ $attributes = Sanitizer::fixTagAttributes ( $attributes , 'table' );
+
+ $lines[$key] = str_repeat( '<dl><dd>' , $indent_level ) . "<table{$attributes}>";
+ array_push ( $td_history , false );
+ array_push ( $last_tag_history , '' );
+ array_push ( $tr_history , false );
+ array_push ( $tr_attributes , '' );
+ array_push ( $has_opened_tr , false );
+ } else if ( count ( $td_history ) == 0 ) {
+ // Don't do any of the following
+ continue;
+ } else if ( substr ( $line , 0 , 2 ) == '|}' ) {
+ // We are ending a table
+ $line = '</table>' . substr ( $line , 2 );
+ $last_tag = array_pop ( $last_tag_history );
- $t[$k] = str_repeat( '<dl><dd>', $indent_level ) .
- '<table' . Sanitizer::fixTagAttributes ( $attributes, 'table' ) . '>' ;
- array_push ( $td , false ) ;
- array_push ( $ltd , '' ) ;
- array_push ( $tr , false ) ;
- array_push ( $ltr , '' ) ;
- array_push ( $has_opened_tr, false );
- }
- else if ( count ( $td ) == 0 ) { } # Don't do any of the following
- else if ( '|}' == substr ( $x , 0 , 2 ) ) {
- $z = "</table>" . substr ( $x , 2);
- $l = array_pop ( $ltd ) ;
- if ( !array_pop ( $has_opened_tr ) ) $z = "<tr><td></td></tr>" . $z ;
- if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
- if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
- array_pop ( $ltr ) ;
- $t[$k] = $z . str_repeat( '</dd></dl>', $indent_level );
- }
- else if ( '|-' == substr ( $x , 0 , 2 ) ) { # Allows for |---------------
- $x = substr ( $x , 1 ) ;
- while ( $x != '' && substr ( $x , 0 , 1 ) == '-' ) $x = substr ( $x , 1 ) ;
- $z = '' ;
- $l = array_pop ( $ltd ) ;
+ if ( !array_pop ( $has_opened_tr ) ) {
+ $line = "<tr><td></td></tr>{$line}";
+ }
+
+ if ( array_pop ( $tr_history ) ) {
+ $line = "</tr>{$line}";
+ }
+
+ if ( array_pop ( $td_history ) ) {
+ $line = "</{$last_tag}>{$line}";
+ }
+ array_pop ( $tr_attributes );
+ $lines[$key] = $line . str_repeat( '</dd></dl>' , $indent_level );
+ } else if ( substr ( $line , 0 , 2 ) == '|-' ) {
+ // Now we have a table row
+ $line = preg_replace( '#^\|-+#', '', $line );
+
+ // Whats after the tag is now only attributes
+ $attributes = $this->mStripState->unstripBoth( $line );
+ $attributes = Sanitizer::fixTagAttributes ( $attributes , 'tr' );
+ array_pop ( $tr_attributes );
+ array_push ( $tr_attributes , $attributes );
+
+ $line = '';
+ $last_tag = array_pop ( $last_tag_history );
array_pop ( $has_opened_tr );
- array_push ( $has_opened_tr , true ) ;
- if ( array_pop ( $tr ) ) $z = '</tr>' . $z ;
- if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
- array_pop ( $ltr ) ;
- $t[$k] = $z ;
- array_push ( $tr , false ) ;
- array_push ( $td , false ) ;
- array_push ( $ltd , '' ) ;
- $attributes = $this->unstripForHTML( $x );
- array_push ( $ltr , Sanitizer::fixTagAttributes ( $attributes, 'tr' ) ) ;
+ array_push ( $has_opened_tr , true );
+
+ if ( array_pop ( $tr_history ) ) {
+ $line = '</tr>';
+ }
+
+ if ( array_pop ( $td_history ) ) {
+ $line = "</{$last_tag}>{$line}";
+ }
+
+ $lines[$key] = $line;
+ array_push ( $tr_history , false );
+ array_push ( $td_history , false );
+ array_push ( $last_tag_history , '' );
}
- else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) { # Caption
- # $x is a table row
- if ( '|+' == substr ( $x , 0 , 2 ) ) {
- $fc = '+' ;
- $x = substr ( $x , 1 ) ;
+ else if ( $first_character == '|' || $first_character == '!' || substr ( $line , 0 , 2 ) == '|+' ) {
+ // This might be cell elements, td, th or captions
+ if ( substr ( $line , 0 , 2 ) == '|+' ) {
+ $first_character = '+';
+ $line = substr ( $line , 1 );
+ }
+
+ $line = substr ( $line , 1 );
+
+ if ( $first_character == '!' ) {
+ $line = str_replace ( '!!' , '||' , $line );
}
- $after = substr ( $x , 1 ) ;
- if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;
// Split up multiple cells on the same line.
- // FIXME: This can result in improper nesting of tags processed
+ // FIXME : This can result in improper nesting of tags processed
// by earlier parser steps, but should avoid splitting up eg
// attribute values containing literal "||".
- $after = wfExplodeMarkup( '||', $after );
+ $cells = StringUtils::explodeMarkup( '||' , $line );
- $t[$k] = '' ;
+ $lines[$key] = '';
- # Loop through each table cell
- foreach ( $after AS $theline )
+ // Loop through each table cell
+ foreach ( $cells as $cell )
{
- $z = '' ;
- if ( $fc != '+' )
+ $previous = '';
+ if ( $first_character != '+' )
{
- $tra = array_pop ( $ltr ) ;
- if ( !array_pop ( $tr ) ) $z = '<tr'.$tra.">\n" ;
- array_push ( $tr , true ) ;
- array_push ( $ltr , '' ) ;
+ $tr_after = array_pop ( $tr_attributes );
+ if ( !array_pop ( $tr_history ) ) {
+ $previous = "<tr{$tr_after}>\n";
+ }
+ array_push ( $tr_history , true );
+ array_push ( $tr_attributes , '' );
array_pop ( $has_opened_tr );
- array_push ( $has_opened_tr , true ) ;
+ array_push ( $has_opened_tr , true );
}
- $l = array_pop ( $ltd ) ;
- if ( array_pop ( $td ) ) $z = '</'.$l.'>' . $z ;
- if ( $fc == '|' ) {
- $l = 'td' ;
- } else if ( $fc == '!' ) {
- $l = 'th' ;
- } else if ( $fc == '+' ) {
- $l = 'caption' ;
- } else {
- $l = '' ;
+ $last_tag = array_pop ( $last_tag_history );
+
+ if ( array_pop ( $td_history ) ) {
+ $previous = "</{$last_tag}>{$previous}";
}
- array_push ( $ltd , $l ) ;
-
- # Cell parameters
- $y = explode ( '|' , $theline , 2 ) ;
- # Note that a '|' inside an invalid link should not
- # be mistaken as delimiting cell parameters
- if ( strpos( $y[0], '[[' ) !== false ) {
- $y = array ($theline);
+
+ if ( $first_character == '|' ) {
+ $last_tag = 'td';
+ } else if ( $first_character == '!' ) {
+ $last_tag = 'th';
+ } else if ( $first_character == '+' ) {
+ $last_tag = 'caption';
+ } else {
+ $last_tag = '';
}
- if ( count ( $y ) == 1 )
- $y = "{$z}<{$l}>{$y[0]}" ;
+
+ array_push ( $last_tag_history , $last_tag );
+
+ // A cell could contain both parameters and data
+ $cell_data = explode ( '|' , $cell , 2 );
+
+ // Bug 553: Note that a '|' inside an invalid link should not
+ // be mistaken as delimiting cell parameters
+ if ( strpos( $cell_data[0], '[[' ) !== false ) {
+ $cell = "{$previous}<{$last_tag}>{$cell}";
+ } else if ( count ( $cell_data ) == 1 )
+ $cell = "{$previous}<{$last_tag}>{$cell_data[0]}";
else {
- $attributes = $this->unstripForHTML( $y[0] );
- $y = "{$z}<{$l}".Sanitizer::fixTagAttributes($attributes, $l).">{$y[1]}" ;
+ $attributes = $this->mStripState->unstripBoth( $cell_data[0] );
+ $attributes = Sanitizer::fixTagAttributes( $attributes , $last_tag );
+ $cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[1]}";
}
- $t[$k] .= $y ;
- array_push ( $td , true ) ;
+
+ $lines[$key] .= $cell;
+ array_push ( $td_history , true );
}
}
}
- # Closing open td, tr && table
- while ( count ( $td ) > 0 )
+ // Closing open td, tr && table
+ while ( count ( $td_history ) > 0 )
{
- $l = array_pop ( $ltd ) ;
- if ( array_pop ( $td ) ) $t[] = '</td>' ;
- if ( array_pop ( $tr ) ) $t[] = '</tr>' ;
- if ( !array_pop ( $has_opened_tr ) ) $t[] = "<tr><td></td></tr>" ;
- $t[] = '</table>' ;
+ if ( array_pop ( $td_history ) ) {
+ $lines[] = '</td>' ;
+ }
+ if ( array_pop ( $tr_history ) ) {
+ $lines[] = '</tr>' ;
+ }
+ if ( !array_pop ( $has_opened_tr ) ) {
+ $lines[] = "<tr><td></td></tr>" ;
+ }
+
+ $lines[] = '</table>' ;
+ }
+
+ $output = implode ( "\n" , $lines ) ;
+
+ // special case: don't return empty table
+ if( $output == "<table>\n<tr><td></td></tr>\n</table>" ) {
+ $output = '';
}
- $t = implode ( "\n" , $t ) ;
- # special case: don't return empty table
- if($t == "<table>\n<tr><td></td></tr>\n</table>")
- $t = '';
wfProfileOut( $fname );
- return $t ;
+
+ return $output;
}
/**
wfProfileIn( $fname );
# Hook to suspend the parser in this state
- $x =& $this->mStripState; // FIXME: Please check that this initialization is correct.
- if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$x ) ) ) {
+ if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
wfProfileOut( $fname );
return $text ;
}
# Remove <noinclude> tags and <includeonly> sections
$text = strtr( $text, array( '<onlyinclude>' => '' , '</onlyinclude>' => '' ) );
$text = strtr( $text, array( '<noinclude>' => '', '</noinclude>' => '') );
- $text = preg_replace( '/<includeonly>.*?<\/includeonly>/s', '', $text );
+ $text = StringUtils::delimiterReplace( '<includeonly>', '</includeonly>', '', $text );
$text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ) );
<a.*?</a> | # Skip link text
<.*?> | # Skip stuff inside HTML elements
(?:RFC|PMID)\s+([0-9]+) | # RFC or PMID, capture number as m[1]
- ISBN\s+([0-9Xx-]+) # ISBN, capture number as m[2]
+ ISBN\s+(\b[0-9Xx\ \-]+) # ISBN, capture number as m[2]
)!x', array( &$this, 'magicLinkCallback' ), $text );
wfProfileOut( __METHOD__ );
return $text;
wfProfileOut( "$fname-misc" );
wfProfileIn( "$fname-title" );
- $nt = Title::newFromText( $this->unstripNoWiki($link, $this->mStripState) );
+ $nt = Title::newFromText( $this->mStripState->unstripNoWiki($link) );
if( !$nt ) {
$s .= $prefix . '[[' . $line;
wfProfileOut( "$fname-title" );
* @return string less-or-more HTML with NOPARSE bits
*/
function armorLinks( $text ) {
- return preg_replace( "/\b(" . wfUrlProtocols() . ')/',
+ return preg_replace( '/\b(' . wfUrlProtocols() . ')/',
"{$this->mUniqPrefix}NOPARSE$1", $text );
}
case 'revisionid':
return $this->mRevisionId;
case 'revisionday':
- return intval( substr( wfRevisionTimestamp( $this->mRevisionId ), 6, 2 ) );
+ return intval( substr( $this->getRevisionTimestamp(), 6, 2 ) );
case 'revisionday2':
- return substr( wfRevisionTimestamp( $this->mRevisionId ), 6, 2 );
+ return substr( $this->getRevisionTimestamp(), 6, 2 );
case 'revisionmonth':
- return intval( substr( wfRevisionTimestamp( $this->mRevisionId ), 4, 2 ) );
+ return intval( substr( $this->getRevisionTimestamp(), 4, 2 ) );
case 'revisionyear':
- return substr( wfRevisionTimestamp( $this->mRevisionId ), 0, 4 );
+ return substr( $this->getRevisionTimestamp(), 0, 4 );
case 'revisiontimestamp':
- return wfRevisionTimestamp( $this->mRevisionId );
+ return $this->getRevisionTimestamp();
case 'namespace':
return str_replace('_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) );
case 'namespacee':
case 'localdow':
return $varCache[$index] = $wgContLang->formatNum( $localDayOfWeek );
case 'numberofarticles':
- return $varCache[$index] = $wgContLang->formatNum( wfNumberOfArticles() );
+ return $varCache[$index] = $wgContLang->formatNum( SiteStats::articles() );
case 'numberoffiles':
- return $varCache[$index] = $wgContLang->formatNum( wfNumberOfFiles() );
+ return $varCache[$index] = $wgContLang->formatNum( SiteStats::images() );
case 'numberofusers':
- return $varCache[$index] = $wgContLang->formatNum( wfNumberOfUsers() );
+ return $varCache[$index] = $wgContLang->formatNum( SiteStats::users() );
case 'numberofpages':
- return $varCache[$index] = $wgContLang->formatNum( wfNumberOfPages() );
+ return $varCache[$index] = $wgContLang->formatNum( SiteStats::pages() );
case 'numberofadmins':
- return $varCache[$index] = $wgContLang->formatNum( wfNumberOfAdmins() );
+ return $varCache[$index] = $wgContLang->formatNum( SiteStats::admins() );
case 'currenttimestamp':
return $varCache[$index] = wfTimestampNow();
case 'localtimestamp':
$noparse = false; # Unsafe HTML tags should not be stripped, etc.
$noargs = false; # Don't replace triple-brace arguments in $text
$replaceHeadings = false; # Make the edit section links go to the template not the article
+ $headingOffset = 0; # Skip headings when number, to account for those that weren't transcluded.
$isHTML = false; # $text is HTML, armour it against wikitext transformation
$forceRawInterwiki = false; # Force interwiki transclusion to be done in raw mode not rendered
if ( !is_null( $title ) ) {
$titleText = $title->getPrefixedText();
- $checkVariantLink = sizeof($wgContLang->getVariants())>1;
# Check for language variants if the template is not found
- if($checkVariantLink && $title->getArticleID() == 0){
+ if($wgContLang->hasVariants() && $title->getArticleID() == 0){
$wgContLang->findVariantLink($part1, $title);
}
if ( !$noparse ) {
# If there are any <onlyinclude> tags, only include them
if ( in_string( '<onlyinclude>', $text ) && in_string( '</onlyinclude>', $text ) ) {
- $m = array();
- preg_match_all( '/<onlyinclude>(.*?)\n?<\/onlyinclude>/s', $text, $m );
- $text = '';
- foreach ($m[1] as $piece)
- $text .= $piece;
+ $replacer = new OnlyIncludeReplacer;
+ StringUtils::delimiterReplaceCallback( '<onlyinclude>', '</onlyinclude>',
+ array( &$replacer, 'replace' ), $text );
+ $text = $replacer->output;
}
# Remove <noinclude> sections and <includeonly> tags
- $text = preg_replace( '/<noinclude>.*?<\/noinclude>/s', '', $text );
+ $text = StringUtils::delimiterReplace( '<noinclude>', '</noinclude>', '', $text );
$text = strtr( $text, array( '<includeonly>' => '' , '</includeonly>' => '' ) );
if( $this->ot['html'] || $this->ot['pre'] ) {
$m = preg_split('/(^={1,6}.*?={1,6}\s*?$)/m', $text, -1,
PREG_SPLIT_DELIM_CAPTURE);
$text = '';
- $nsec = 0;
+ $nsec = $headingOffset;
for( $i = 0; $i < count($m); $i += 2 ) {
$text .= $m[$i];
if (!isset($m[$i + 1]) || $m[$i + 1] == "") continue;
# 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( $canonized_headline, $this->mStripState );
+ $canonized_headline = $this->mStripState->unstripBoth( $headline );
# Remove link placeholders by the link text.
# <!--LINK number-->
$refers[$headlineCount] = $canonized_headline;
# count how many in assoc. array so we can track dupes in anchors
- @$refers[$canonized_headline]++;
+ isset( $refers[$canonized_headline] ) ? $refers[$canonized_headline]++ : $refers[$canonized_headline] = 1;
$refcount[$headlineCount]=$refers[$canonized_headline];
# Don't number the heading if it is the only one (looks silly)
$toc .= $sk->tocLine($anchor, $tocline, $numbering, $toclevel);
}
# give headline the correct <h#> tag
- @$head[$headlineCount] .= "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount];
+ $head[$headlineCount] = "<a name=\"$anchor\"></a><h".$level.$matches[2][$headlineCount];
if( $showEditLink && ( !$istemplate || $templatetitle !== "" ) ) {
if ( empty( $head[$headlineCount] ) ) {
$this->clearState();
}
- $stripState = false;
+ $stripState = new StripState;
$pairs = array(
"\r\n" => "\n",
);
$text = str_replace( array_keys( $pairs ), array_values( $pairs ), $text );
$text = $this->strip( $text, $stripState, true, array( 'gallery' ) );
$text = $this->pstPass2( $text, $stripState, $user );
- $text = $this->unstrip( $text, $stripState );
- $text = $this->unstripNoWiki( $text, $stripState );
+ $text = $stripState->unstripBoth( $text );
return $text;
}
*/
function setHook( $tag, $callback ) {
$tag = strtolower( $tag );
- $oldVal = @$this->mTagHooks[$tag];
+ $oldVal = isset( $this->mTagHooks[$tag] ) ? $this->mTagHooks[$tag] : null;
$this->mTagHooks[$tag] = $callback;
return $oldVal;
* @return The old callback function for this name, if any
*/
function setFunctionHook( $id, $callback, $flags = 0 ) {
- $oldVal = @$this->mFunctionHooks[$id];
+ $oldVal = isset( $this->mFunctionHooks[$id] ) ? $this->mFunctionHooks[$id] : null;
$this->mFunctionHooks[$id] = $callback;
# Add to function cache
*/
function replaceLinkHolders( &$text, $options = 0 ) {
global $wgUser;
- global $wgOutputReplace;
global $wgContLang;
$fname = 'Parser::replaceLinkHolders';
}
wfProfileOut( $fname.'-check' );
- # Do a second query for different language variants of links (if needed)
+ # Do a second query for different language variants of links and categories
if($wgContLang->hasVariants()){
$linkBatch = new LinkBatch();
- $variantMap = array(); // maps $pdbkey_Variant => $pdbkey_original
+ $variantMap = array(); // maps $pdbkey_Variant => $keys (of link holders)
+ $categoryMap = array(); // maps $category_variant => $category (dbkeys)
+ $varCategories = array(); // category replacements oldDBkey => newDBkey
+
+ $categories = $this->mOutput->getCategoryLinks();
// Add variants of links to link batch
foreach ( $this->mLinkHolders['namespaces'] as $key => $ns ) {
continue;
$pdbk = $title->getPrefixedDBkey();
+ $titleText = $title->getText();
// generate all variants of the link title text
- $allTextVariants = $wgContLang->convertLinkToAllVariants($title->getText());
+ $allTextVariants = $wgContLang->convertLinkToAllVariants($titleText);
// if link was not found (in first query), add all variants to query
if ( !isset($colours[$pdbk]) ){
foreach($allTextVariants as $textVariant){
- $variantTitle = Title::makeTitle( $ns, $textVariant );
- if(is_null($variantTitle)) continue;
- $linkBatch->addObj( $variantTitle );
- $variantMap[$variantTitle->getPrefixedDBkey()][] = $key;
+ if($textVariant != $titleText){
+ $variantTitle = Title::makeTitle( $ns, $textVariant );
+ if(is_null($variantTitle)) continue;
+ $linkBatch->addObj( $variantTitle );
+ $variantMap[$variantTitle->getPrefixedDBkey()][] = $key;
+ }
}
}
}
+ // process categories, check if a category exists in some variant
+ foreach( $categories as $category){
+ $variants = $wgContLang->convertLinkToAllVariants($category);
+ foreach($variants as $variant){
+ $variantTitle = Title::newFromDBkey( Title::makeName(NS_CATEGORY,$variant) );
+ if(is_null($variantTitle)) continue;
+ $linkBatch->addObj( $variantTitle );
+ $categoryMap[$variant] = $category;
+ }
+ }
+
if(!$linkBatch->isEmpty()){
// construct query
$variantTitle = Title::makeTitle( $s->page_namespace, $s->page_title );
$varPdbk = $variantTitle->getPrefixedDBkey();
+ $vardbk = $variantTitle->getDBkey();
$linkCache->addGoodLinkObj( $s->page_id, $variantTitle );
$this->mOutput->addLink( $variantTitle, $s->page_id );
- $holderKeys = $variantMap[$varPdbk];
+ $holderKeys = array();
+ if(isset($variantMap[$varPdbk]))
+ $holderKeys = $variantMap[$varPdbk];
// loop over link holders
foreach($holderKeys as $key){
}
}
}
+
+ // check if the object is a variant of a category
+ if(isset($categoryMap[$vardbk])){
+ $oldkey = $categoryMap[$vardbk];
+ if($oldkey != $vardbk)
+ $varCategories[$oldkey]=$vardbk;
+ }
+ }
+
+ // rebuild the categories in original order (if there are replacements)
+ if(count($varCategories)>0){
+ $newCats = array();
+ $originalCats = $this->mOutput->getCategories();
+ foreach($originalCats as $cat => $sortkey){
+ // make the replacement
+ if( array_key_exists($cat,$varCategories) )
+ $newCats[$varCategories[$cat]] = $sortkey;
+ else $newCats[$cat] = $sortkey;
+ }
+ $this->mOutput->setCategoryLinks($newCats);
}
}
}
# Construct search and replace arrays
wfProfileIn( $fname.'-construct' );
- $wgOutputReplace = array();
+ $replacePairs = array();
foreach ( $this->mLinkHolders['namespaces'] as $key => $ns ) {
$pdbk = $pdbks[$key];
$searchkey = "<!--LINK $key-->";
$linkCache->addBadLinkObj( $title );
$colours[$pdbk] = 0;
$this->mOutput->addLink( $title, 0 );
- $wgOutputReplace[$searchkey] = $sk->makeBrokenLinkObj( $title,
+ $replacePairs[$searchkey] = $sk->makeBrokenLinkObj( $title,
$this->mLinkHolders['texts'][$key],
$this->mLinkHolders['queries'][$key] );
} elseif ( $colours[$pdbk] == 1 ) {
- $wgOutputReplace[$searchkey] = $sk->makeKnownLinkObj( $title,
+ $replacePairs[$searchkey] = $sk->makeKnownLinkObj( $title,
$this->mLinkHolders['texts'][$key],
$this->mLinkHolders['queries'][$key] );
} elseif ( $colours[$pdbk] == 2 ) {
- $wgOutputReplace[$searchkey] = $sk->makeStubLinkObj( $title,
+ $replacePairs[$searchkey] = $sk->makeStubLinkObj( $title,
$this->mLinkHolders['texts'][$key],
$this->mLinkHolders['queries'][$key] );
}
}
+ $replacer = new HashtableReplacer( $replacePairs, 1 );
wfProfileOut( $fname.'-construct' );
# Do the thing
wfProfileIn( $fname.'-replace' );
-
$text = preg_replace_callback(
'/(<!--LINK .*?-->)/',
- "wfOutputReplaceMatches",
+ $replacer->cb(),
$text);
wfProfileOut( $fname.'-replace' );
if ( !empty( $this->mInterwikiLinkHolders['texts'] ) ) {
wfProfileIn( $fname.'-interwiki' );
# Make interwiki link HTML
- $wgOutputReplace = array();
+ $replacePairs = array();
foreach( $this->mInterwikiLinkHolders['texts'] as $key => $link ) {
$title = $this->mInterwikiLinkHolders['titles'][$key];
- $wgOutputReplace[$key] = $sk->makeLinkObj( $title, $link );
+ $replacePairs[$key] = $sk->makeLinkObj( $title, $link );
}
+ $replacer = new HashtableReplacer( $replacePairs, 1 );
$text = preg_replace_callback(
'/<!--IWLINK (.*?)-->/',
- "wfOutputReplaceMatches",
+ $replacer->cb(),
$text );
wfProfileOut( $fname.'-interwiki' );
}
*/
function renderPreTag( $text, $attribs ) {
// Backwards-compatibility hack
- $content = preg_replace( '!<nowiki>(.*?)</nowiki>!is', '\\1', $text );
+ $content = StringUtils::delimiterReplace( '<nowiki>', '</nowiki>', '$1', $text, 'i' );
$attribs = Sanitizer::validateTagAttributes( $attribs, 'pre' );
return wfOpenElement( 'pre', $attribs ) .
- wfEscapeHTMLTagsOnly( $content ) .
+ Xml::escapeTagsOnly( $content ) .
'</pre>';
}
# make sure there are no placeholders in thumbnail attributes
# that are later expanded to html- so expand them now and
# remove the tags
- $alt = $this->unstrip($alt, $this->mStripState);
+ $alt = $this->mStripState->unstripBoth( $alt );
$alt = Sanitizer::stripAllTags( $alt );
# Linker does the rest
*/
function attributeStripCallback( &$text, $args ) {
$text = $this->replaceVariables( $text, $args );
- $text = $this->unstripForHTML( $text );
+ $text = $this->mStripState->unstripBoth( $text );
return $text;
}
- function unstripForHTML( $text ) {
- $text = $this->unstrip( $text, $this->mStripState );
- $text = $this->unstripNoWiki( $text, $this->mStripState );
- return $text;
- }
/**#@-*/
/**#@+
private function extractSections( $text, $section, $mode, $newtext='' ) {
# strip NOWIKI etc. to avoid confusion (true-parameter causes HTML
# comments to be stripped as well)
- $striparray = array();
+ $stripState = new StripState;
$oldOutputType = $this->mOutputType;
$oldOptions = $this->mOptions;
$this->mOptions = new ParserOptions();
$this->setOutputType( OT_WIKI );
- $striptext = $this->strip( $text, $striparray, true );
+ $striptext = $this->strip( $text, $stripState, true );
$this->setOutputType( $oldOutputType );
$this->mOptions = $oldOptions;
}
}
# reinsert stripped tags
- $rv = $this->unstrip( $rv, $striparray );
- $rv = $this->unstripNoWiki( $rv, $striparray );
- $rv = trim( $rv );
+ $rv = trim( $stripState->unstripBoth( $rv ) );
return $rv;
}
return $this->extractSections( $oldtext, $section, "replace", $text );
}
+ /**
+ * Get the timestamp associated with the current revision, adjusted for
+ * the default server-local timestamp
+ */
+ function getRevisionTimestamp() {
+ if ( is_null( $this->mRevisionTimestamp ) ) {
+ wfProfileIn( __METHOD__ );
+ global $wgContLang;
+ $dbr =& wfGetDB( DB_SLAVE );
+ $timestamp = $dbr->selectField( 'revision', 'rev_timestamp',
+ array( 'rev_id' => $this->mRevisionId ), __METHOD__ );
+
+ // Normalize timestamp to internal MW format for timezone processing.
+ // This has the added side-effect of replacing a null value with
+ // the current time, which gives us more sensible behavior for
+ // previews.
+ $timestamp = wfTimestamp( TS_MW, $timestamp );
+
+ // The cryptic '' timezone parameter tells to use the site-default
+ // timezone offset instead of the user settings.
+ //
+ // Since this value will be saved into the parser cache, served
+ // to other users, and potentially even used inside links and such,
+ // it needs to be consistent for all visitors.
+ $this->mRevisionTimestamp = $wgContLang->userAdjust( $timestamp, '' );
+
+ wfProfileOut( __METHOD__ );
+ }
+ return $this->mRevisionTimestamp;
+ }
}
/**
}
}
-/**
- * Callback function used by Parser::replaceLinkHolders()
- * to substitute link placeholders.
- */
-function &wfOutputReplaceMatches( $matches ) {
- global $wgOutputReplace;
- return $wgOutputReplace[$matches[1]];
-}
+class OnlyIncludeReplacer {
+ var $output = '';
-/**
- * Return the total number of articles
- */
-function wfNumberOfArticles() {
- global $wgNumberOfArticles;
-
- wfLoadSiteStats();
- return $wgNumberOfArticles;
-}
-
-/**
- * Return the number of files
- */
-function wfNumberOfFiles() {
- $fname = 'wfNumberOfFiles';
-
- wfProfileIn( $fname );
- $dbr =& wfGetDB( DB_SLAVE );
- $numImages = $dbr->selectField('site_stats', 'ss_images', array(), $fname );
- wfProfileOut( $fname );
-
- return $numImages;
-}
-
-/**
- * Return the number of user accounts
- * @return integer
- */
-function wfNumberOfUsers() {
- wfProfileIn( 'wfNumberOfUsers' );
- $dbr =& wfGetDB( DB_SLAVE );
- $count = $dbr->selectField( 'site_stats', 'ss_users', array(), 'wfNumberOfUsers' );
- wfProfileOut( 'wfNumberOfUsers' );
- return (int)$count;
+ function replace( $matches ) {
+ if ( substr( $matches[1], -1 ) == "\n" ) {
+ $this->output .= substr( $matches[1], 0, -1 );
+ } else {
+ $this->output .= $matches[1];
+ }
+ }
}
-/**
- * Return the total number of pages
- * @return integer
- */
-function wfNumberOfPages() {
- wfProfileIn( 'wfNumberOfPages' );
- $dbr =& wfGetDB( DB_SLAVE );
- $count = $dbr->selectField( 'site_stats', 'ss_total_pages', array(), 'wfNumberOfPages' );
- wfProfileOut( 'wfNumberOfPages' );
- return (int)$count;
-}
+class StripState {
+ var $general, $nowiki;
-/**
- * Return the total number of admins
- *
- * @return integer
- */
-function wfNumberOfAdmins() {
- static $admins = -1;
- wfProfileIn( 'wfNumberOfAdmins' );
- if( $admins == -1 ) {
- $dbr =& wfGetDB( DB_SLAVE );
- $admins = $dbr->selectField( 'user_groups', 'COUNT(*)', array( 'ug_group' => 'sysop' ), 'wfNumberOfAdmins' );
+ function __construct() {
+ $this->general = new ReplacementArray;
+ $this->nowiki = new ReplacementArray;
}
- wfProfileOut( 'wfNumberOfAdmins' );
- return (int)$admins;
-}
-/**
- * Count the number of pages in a particular namespace
- *
- * @param $ns Namespace
- * @return integer
- */
-function wfPagesInNs( $ns ) {
- static $pageCount = array();
- wfProfileIn( 'wfPagesInNs' );
- if( !isset( $pageCount[$ns] ) ) {
- $dbr =& wfGetDB( DB_SLAVE );
- $pageCount[$ns] = $dbr->selectField( 'page', 'COUNT(*)', array( 'page_namespace' => $ns ), 'wfPagesInNs' );
+ function unstripGeneral( $text ) {
+ wfProfileIn( __METHOD__ );
+ $text = $this->general->replace( $text );
+ wfProfileOut( __METHOD__ );
+ return $text;
}
- wfProfileOut( 'wfPagesInNs' );
- return (int)$pageCount[$ns];
-}
-/**
- * Get various statistics from the database
- * @private
- */
-function wfLoadSiteStats() {
- global $wgNumberOfArticles, $wgTotalViews, $wgTotalEdits;
- $fname = 'wfLoadSiteStats';
-
- if ( -1 != $wgNumberOfArticles ) return;
- $dbr =& wfGetDB( DB_SLAVE );
- $s = $dbr->selectRow( 'site_stats',
- array( 'ss_total_views', 'ss_total_edits', 'ss_good_articles' ),
- array( 'ss_row_id' => 1 ), $fname
- );
-
- if ( $s === false ) {
- return;
- } else {
- $wgTotalViews = $s->ss_total_views;
- $wgTotalEdits = $s->ss_total_edits;
- $wgNumberOfArticles = $s->ss_good_articles;
+ function unstripNoWiki( $text ) {
+ wfProfileIn( __METHOD__ );
+ $text = $this->nowiki->replace( $text );
+ wfProfileOut( __METHOD__ );
+ return $text;
}
-}
-/**
- * Get revision timestamp from the database considering timecorrection
- *
- * @param $id Int: page revision id
- * @return integer
- */
-function wfRevisionTimestamp( $id ) {
- global $wgContLang;
- $fname = 'wfRevisionTimestamp';
-
- wfProfileIn( $fname );
- $dbr =& wfGetDB( DB_SLAVE );
- $timestamp = $dbr->selectField( 'revision', 'rev_timestamp',
- array( 'rev_id' => $id ), __METHOD__ );
- $timestamp = $wgContLang->userAdjust( $timestamp );
- wfProfileOut( $fname );
-
- return $timestamp;
-}
-
-/**
- * Escape html tags
- * Basically replacing " > and < with HTML entities ( ", >, <)
- *
- * @param $in String: text that might contain HTML tags.
- * @return string Escaped string
- */
-function wfEscapeHTMLTagsOnly( $in ) {
- return str_replace(
- array( '"', '>', '<' ),
- array( '"', '>', '<' ),
- $in );
+ function unstripBoth( $text ) {
+ wfProfileIn( __METHOD__ );
+ $text = $this->general->replace( $text );
+ $text = $this->nowiki->replace( $text );
+ wfProfileOut( __METHOD__ );
+ return $text;
+ }
}
?>