From 396b28f3d881f5debd888ba9bb9b47c2d478a76f Mon Sep 17 00:00:00 2001 From: Pierre Schmitz Date: Mon, 15 Dec 2008 18:02:47 +0100 Subject: update to Mediawiki 1.13.3; some cleanups --- includes/AutoLoader.php | 1 + includes/CacheManager.php | 159 - includes/CoreParserFunctions.php | 268 -- includes/Database.php | 2443 ------------- includes/DatabaseMysql.php | 6 - includes/DatabaseOracle.php | 697 ---- includes/DatabasePostgres.php | 1313 ------- includes/DateFormatter.php | 285 -- includes/DefaultSettings.php | 6 +- includes/Exception.php | 11 +- includes/HTMLForm.php | 107 - includes/IEContentAnalyzer.php | 823 +++++ includes/Image.php | 2142 ------------ includes/LoadBalancer.php | 653 ---- includes/MimeMagic.php | 27 + includes/Parser.php | 4913 -------------------------- includes/ParserCache.php | 119 - includes/ParserOptions.php | 145 - includes/ParserOutput.php | 189 - includes/ParserXML.php | 643 ---- includes/Parser_DiffTest.php | 85 - includes/Parser_OldPP.php | 4942 --------------------------- includes/Preprocessor.php | 154 - includes/Preprocessor_DOM.php | 1356 -------- includes/Preprocessor_Hash.php | 1471 -------- includes/Profiling.php | 353 -- includes/SearchTsearch2.php | 122 - includes/SiteStatsUpdate.php | 92 - includes/SpecialAllmessages.php | 219 -- includes/SpecialAllpages.php | 395 --- includes/SpecialAncientpages.php | 63 - includes/SpecialBlockip.php | 476 --- includes/SpecialBlockme.php | 38 - includes/SpecialBooksources.php | 113 - includes/SpecialBrokenRedirects.php | 95 - includes/SpecialCategories.php | 63 - includes/SpecialConfirmemail.php | 104 - includes/SpecialContributions.php | 434 --- includes/SpecialDeadendpages.php | 65 - includes/SpecialDisambiguations.php | 106 - includes/SpecialDoubleRedirects.php | 104 - includes/SpecialEmailuser.php | 222 -- includes/SpecialExport.php | 284 -- includes/SpecialFewestrevisions.php | 65 - includes/SpecialFilepath.php | 69 - includes/SpecialImagelist.php | 166 - includes/SpecialImport.php | 921 ----- includes/SpecialIpblocklist.php | 430 --- includes/SpecialListredirects.php | 60 - includes/SpecialListusers.php | 217 -- includes/SpecialLockdb.php | 134 - includes/SpecialLog.php | 527 --- includes/SpecialLonelypages.php | 60 - includes/SpecialLongpages.php | 33 - includes/SpecialMIMEsearch.php | 141 - includes/SpecialMergeHistory.php | 423 --- includes/SpecialMostcategories.php | 59 - includes/SpecialMostimages.php | 55 - includes/SpecialMostlinked.php | 93 - includes/SpecialMostlinkedcategories.php | 75 - includes/SpecialMostlinkedtemplates.php | 129 - includes/SpecialMostrevisions.php | 66 - includes/SpecialMovepage.php | 327 -- includes/SpecialNewimages.php | 207 -- includes/SpecialNewpages.php | 317 -- includes/SpecialPopularpages.php | 69 - includes/SpecialPreferences.php | 1047 ------ includes/SpecialPrefixindex.php | 149 - includes/SpecialProtectedpages.php | 287 -- includes/SpecialProtectedtitles.php | 219 -- includes/SpecialRandompage.php | 108 - includes/SpecialRandomredirect.php | 20 - includes/SpecialRecentchanges.php | 730 ---- includes/SpecialRecentchangeslinked.php | 190 - includes/SpecialResetpass.php | 165 - includes/SpecialRevisiondelete.php | 275 -- includes/SpecialSearch.php | 438 --- includes/SpecialShortpages.php | 92 - includes/SpecialSpecialpages.php | 61 - includes/SpecialStatistics.php | 93 - includes/SpecialUncategorizedcategories.php | 37 - includes/SpecialUncategorizedimages.php | 47 - includes/SpecialUncategorizedpages.php | 57 - includes/SpecialUncategorizedtemplates.php | 31 - includes/SpecialUndelete.php | 1075 ------ includes/SpecialUnlockdb.php | 110 - includes/SpecialUnusedcategories.php | 46 - includes/SpecialUnusedimages.php | 61 - includes/SpecialUnusedtemplates.php | 51 - includes/SpecialUnwatchedpages.php | 65 - includes/SpecialUpload.php | 1646 --------- includes/SpecialUploadMogile.php | 136 - includes/SpecialUserlogin.php | 863 ----- includes/SpecialUserlogout.php | 19 - includes/SpecialUserrights.php | 575 ---- includes/SpecialVersion.php | 355 -- includes/SpecialWantedcategories.php | 79 - includes/SpecialWantedpages.php | 133 - includes/SpecialWatchlist.php | 375 -- includes/SpecialWhatlinkshere.php | 322 -- includes/SpecialWithoutinterwiki.php | 93 - includes/StreamFile.php | 52 +- includes/Title.php | 8 +- includes/Utf8Case.php | 1505 -------- includes/XmlTypeCheck.php | 74 +- includes/api/ApiChangeRights.php | 155 - includes/api/ApiFormatBase.php | 11 +- includes/api/ApiMain.php | 4 +- includes/filerepo/FSRepo.php | 21 +- includes/filerepo/ICRepo.php | 309 -- includes/specials/SpecialImport.php | 30 +- includes/specials/SpecialUndelete.php | 35 +- includes/specials/SpecialUpload.php | 65 +- 113 files changed, 1080 insertions(+), 42188 deletions(-) delete mode 100644 includes/CacheManager.php delete mode 100644 includes/CoreParserFunctions.php delete mode 100644 includes/Database.php delete mode 100644 includes/DatabaseMysql.php delete mode 100644 includes/DatabaseOracle.php delete mode 100644 includes/DatabasePostgres.php delete mode 100644 includes/DateFormatter.php delete mode 100644 includes/HTMLForm.php create mode 100644 includes/IEContentAnalyzer.php delete mode 100644 includes/Image.php delete mode 100644 includes/LoadBalancer.php delete mode 100644 includes/Parser.php delete mode 100644 includes/ParserCache.php delete mode 100644 includes/ParserOptions.php delete mode 100644 includes/ParserOutput.php delete mode 100644 includes/ParserXML.php delete mode 100644 includes/Parser_DiffTest.php delete mode 100644 includes/Parser_OldPP.php delete mode 100644 includes/Preprocessor.php delete mode 100644 includes/Preprocessor_DOM.php delete mode 100644 includes/Preprocessor_Hash.php delete mode 100644 includes/Profiling.php delete mode 100644 includes/SearchTsearch2.php delete mode 100644 includes/SiteStatsUpdate.php delete mode 100644 includes/SpecialAllmessages.php delete mode 100644 includes/SpecialAllpages.php delete mode 100644 includes/SpecialAncientpages.php delete mode 100644 includes/SpecialBlockip.php delete mode 100644 includes/SpecialBlockme.php delete mode 100644 includes/SpecialBooksources.php delete mode 100644 includes/SpecialBrokenRedirects.php delete mode 100644 includes/SpecialCategories.php delete mode 100644 includes/SpecialConfirmemail.php delete mode 100644 includes/SpecialContributions.php delete mode 100644 includes/SpecialDeadendpages.php delete mode 100644 includes/SpecialDisambiguations.php delete mode 100644 includes/SpecialDoubleRedirects.php delete mode 100644 includes/SpecialEmailuser.php delete mode 100644 includes/SpecialExport.php delete mode 100644 includes/SpecialFewestrevisions.php delete mode 100644 includes/SpecialFilepath.php delete mode 100644 includes/SpecialImagelist.php delete mode 100644 includes/SpecialImport.php delete mode 100644 includes/SpecialIpblocklist.php delete mode 100644 includes/SpecialListredirects.php delete mode 100644 includes/SpecialListusers.php delete mode 100644 includes/SpecialLockdb.php delete mode 100644 includes/SpecialLog.php delete mode 100644 includes/SpecialLonelypages.php delete mode 100644 includes/SpecialLongpages.php delete mode 100644 includes/SpecialMIMEsearch.php delete mode 100644 includes/SpecialMergeHistory.php delete mode 100644 includes/SpecialMostcategories.php delete mode 100644 includes/SpecialMostimages.php delete mode 100644 includes/SpecialMostlinked.php delete mode 100644 includes/SpecialMostlinkedcategories.php delete mode 100644 includes/SpecialMostlinkedtemplates.php delete mode 100644 includes/SpecialMostrevisions.php delete mode 100644 includes/SpecialMovepage.php delete mode 100644 includes/SpecialNewimages.php delete mode 100644 includes/SpecialNewpages.php delete mode 100644 includes/SpecialPopularpages.php delete mode 100644 includes/SpecialPreferences.php delete mode 100644 includes/SpecialPrefixindex.php delete mode 100644 includes/SpecialProtectedpages.php delete mode 100755 includes/SpecialProtectedtitles.php delete mode 100644 includes/SpecialRandompage.php delete mode 100644 includes/SpecialRandomredirect.php delete mode 100644 includes/SpecialRecentchanges.php delete mode 100644 includes/SpecialRecentchangeslinked.php delete mode 100644 includes/SpecialResetpass.php delete mode 100644 includes/SpecialRevisiondelete.php delete mode 100644 includes/SpecialSearch.php delete mode 100644 includes/SpecialShortpages.php delete mode 100644 includes/SpecialSpecialpages.php delete mode 100644 includes/SpecialStatistics.php delete mode 100644 includes/SpecialUncategorizedcategories.php delete mode 100644 includes/SpecialUncategorizedimages.php delete mode 100644 includes/SpecialUncategorizedpages.php delete mode 100644 includes/SpecialUncategorizedtemplates.php delete mode 100644 includes/SpecialUndelete.php delete mode 100644 includes/SpecialUnlockdb.php delete mode 100644 includes/SpecialUnusedcategories.php delete mode 100644 includes/SpecialUnusedimages.php delete mode 100644 includes/SpecialUnusedtemplates.php delete mode 100644 includes/SpecialUnwatchedpages.php delete mode 100644 includes/SpecialUpload.php delete mode 100644 includes/SpecialUploadMogile.php delete mode 100644 includes/SpecialUserlogin.php delete mode 100644 includes/SpecialUserlogout.php delete mode 100644 includes/SpecialUserrights.php delete mode 100644 includes/SpecialVersion.php delete mode 100644 includes/SpecialWantedcategories.php delete mode 100644 includes/SpecialWantedpages.php delete mode 100644 includes/SpecialWatchlist.php delete mode 100644 includes/SpecialWhatlinkshere.php delete mode 100644 includes/SpecialWithoutinterwiki.php delete mode 100644 includes/Utf8Case.php delete mode 100644 includes/api/ApiChangeRights.php delete mode 100644 includes/filerepo/ICRepo.php (limited to 'includes') diff --git a/includes/AutoLoader.php b/includes/AutoLoader.php index 4f36784a..de75b41d 100644 --- a/includes/AutoLoader.php +++ b/includes/AutoLoader.php @@ -91,6 +91,7 @@ class AutoLoader { 'HTMLFileCache' => 'includes/HTMLFileCache.php', 'Http' => 'includes/HttpFunctions.php', '_HWLDF_WordAccumulator' => 'includes/DifferenceEngine.php', + 'IEContentAnalyzer' => 'includes/IEContentAnalyzer.php', 'ImageGallery' => 'includes/ImageGallery.php', 'ImageHistoryList' => 'includes/ImagePage.php', 'ImagePage' => 'includes/ImagePage.php', diff --git a/includes/CacheManager.php b/includes/CacheManager.php deleted file mode 100644 index b9e307f4..00000000 --- a/includes/CacheManager.php +++ /dev/null @@ -1,159 +0,0 @@ -mTitle =& $title; - $this->mFileCache = ''; - } - - function fileCacheName() { - global $wgFileCacheDirectory; - if( !$this->mFileCache ) { - $key = $this->mTitle->getPrefixedDbkey(); - $hash = md5( $key ); - $key = str_replace( '.', '%2E', urlencode( $key ) ); - - $hash1 = substr( $hash, 0, 1 ); - $hash2 = substr( $hash, 0, 2 ); - $this->mFileCache = "{$wgFileCacheDirectory}/{$hash1}/{$hash2}/{$key}.html"; - - if($this->useGzip()) - $this->mFileCache .= '.gz'; - - wfDebug( " fileCacheName() - {$this->mFileCache}\n" ); - } - return $this->mFileCache; - } - - function isFileCached() { - return file_exists( $this->fileCacheName() ); - } - - function fileCacheTime() { - return wfTimestamp( TS_MW, filemtime( $this->fileCacheName() ) ); - } - - function isFileCacheGood( $timestamp ) { - global $wgCacheEpoch; - - if( !$this->isFileCached() ) return false; - - $cachetime = $this->fileCacheTime(); - $good = (( $timestamp <= $cachetime ) && - ( $wgCacheEpoch <= $cachetime )); - - wfDebug(" isFileCacheGood() - cachetime $cachetime, touched {$timestamp} epoch {$wgCacheEpoch}, good $good\n"); - return $good; - } - - function useGzip() { - global $wgUseGzip; - return $wgUseGzip; - } - - /* In handy string packages */ - function fetchRawText() { - return file_get_contents( $this->fileCacheName() ); - } - - function fetchPageText() { - if( $this->useGzip() ) { - /* Why is there no gzfile_get_contents() or gzdecode()? */ - return implode( '', gzfile( $this->fileCacheName() ) ); - } else { - return $this->fetchRawText(); - } - } - - /* Working directory to/from output */ - function loadFromFileCache() { - global $wgOut, $wgMimeType, $wgOutputEncoding, $wgContLanguageCode; - wfDebug(" loadFromFileCache()\n"); - - $filename=$this->fileCacheName(); - $wgOut->sendCacheControl(); - - header( "Content-type: $wgMimeType; charset={$wgOutputEncoding}" ); - header( "Content-language: $wgContLanguageCode" ); - - if( $this->useGzip() ) { - if( wfClientAcceptsGzip() ) { - header( 'Content-Encoding: gzip' ); - } else { - /* Send uncompressed */ - readgzfile( $filename ); - return; - } - } - readfile( $filename ); - } - - function checkCacheDirs() { - $filename = $this->fileCacheName(); - $mydir2=substr($filename,0,strrpos($filename,'/')); # subdirectory level 2 - $mydir1=substr($mydir2,0,strrpos($mydir2,'/')); # subdirectory level 1 - - if(!file_exists($mydir1)) { mkdir($mydir1,0775); } # create if necessary - if(!file_exists($mydir2)) { mkdir($mydir2,0775); } - } - - function saveToFileCache( $origtext ) { - $text = $origtext; - if(strcmp($text,'') == 0) return ''; - - wfDebug(" saveToFileCache()\n", false); - - $this->checkCacheDirs(); - - $f = fopen( $this->fileCacheName(), 'w' ); - if($f) { - $now = wfTimestampNow(); - if( $this->useGzip() ) { - $rawtext = str_replace( '', - '\n", - $text ); - $text = gzencode( $rawtext ); - } else { - $text = str_replace( '', - '\n", - $text ); - } - fwrite( $f, $text ); - fclose( $f ); - if( $this->useGzip() ) { - if( wfClientAcceptsGzip() ) { - header( 'Content-Encoding: gzip' ); - return $text; - } else { - return $rawtext; - } - } else { - return $text; - } - } - return $text; - } - -} - -?> diff --git a/includes/CoreParserFunctions.php b/includes/CoreParserFunctions.php deleted file mode 100644 index 61dbafe5..00000000 --- a/includes/CoreParserFunctions.php +++ /dev/null @@ -1,268 +0,0 @@ - false ); - } - } - - static function ns( $parser, $part1 = '' ) { - global $wgContLang; - $found = false; - if ( intval( $part1 ) || $part1 == "0" ) { - $text = $wgContLang->getNsText( intval( $part1 ) ); - $found = true; - } else { - $param = str_replace( ' ', '_', strtolower( $part1 ) ); - $index = Namespace::getCanonicalIndex( strtolower( $param ) ); - if ( !is_null( $index ) ) { - $text = $wgContLang->getNsText( $index ); - $found = true; - } - } - if ( $found ) { - return $text; - } else { - return array( 'found' => false ); - } - } - - static function urlencode( $parser, $s = '' ) { - return urlencode( $s ); - } - - static function lcfirst( $parser, $s = '' ) { - global $wgContLang; - return $wgContLang->lcfirst( $s ); - } - - static function ucfirst( $parser, $s = '' ) { - global $wgContLang; - return $wgContLang->ucfirst( $s ); - } - - static function lc( $parser, $s = '' ) { - global $wgContLang; - if ( is_callable( array( $parser, 'markerSkipCallback' ) ) ) { - return $parser->markerSkipCallback( $s, array( $wgContLang, 'lc' ) ); - } else { - return $wgContLang->lc( $s ); - } - } - - static function uc( $parser, $s = '' ) { - global $wgContLang; - if ( is_callable( array( $parser, 'markerSkipCallback' ) ) ) { - return $parser->markerSkipCallback( $s, array( $wgContLang, 'uc' ) ); - } else { - return $wgContLang->uc( $s ); - } - } - - static function localurl( $parser, $s = '', $arg = null ) { return self::urlFunction( 'getLocalURL', $s, $arg ); } - static function localurle( $parser, $s = '', $arg = null ) { return self::urlFunction( 'escapeLocalURL', $s, $arg ); } - static function fullurl( $parser, $s = '', $arg = null ) { return self::urlFunction( 'getFullURL', $s, $arg ); } - static function fullurle( $parser, $s = '', $arg = null ) { return self::urlFunction( 'escapeFullURL', $s, $arg ); } - - static function urlFunction( $func, $s = '', $arg = null ) { - $title = Title::newFromText( $s ); - # Due to order of execution of a lot of bits, the values might be encoded - # before arriving here; if that's true, then the title can't be created - # and the variable will fail. If we can't get a decent title from the first - # attempt, url-decode and try for a second. - if( is_null( $title ) ) - $title = Title::newFromUrl( urldecode( $s ) ); - if ( !is_null( $title ) ) { - if ( !is_null( $arg ) ) { - $text = $title->$func( $arg ); - } else { - $text = $title->$func(); - } - return $text; - } else { - return array( 'found' => false ); - } - } - - static function formatNum( $parser, $num = '' ) { - return $parser->getFunctionLang()->formatNum( $num ); - } - - static function grammar( $parser, $case = '', $word = '' ) { - return $parser->getFunctionLang()->convertGrammar( $word, $case ); - } - - static function plural( $parser, $text = '') { - $forms = array_slice( func_get_args(), 2); - $text = $parser->getFunctionLang()->parseFormattedNumber( $text ); - return $parser->getFunctionLang()->convertPlural( $text, $forms ); - } - - /** - * Override the title of the page when viewed, - * provided we've been given a title which - * will normalise to the canonical title - * - * @param Parser $parser Parent parser - * @param string $text Desired title text - * @return string - */ - static function displaytitle( $parser, $text = '' ) { - $text = trim( Sanitizer::decodeCharReferences( $text ) ); - $title = Title::newFromText( $text ); - if( $title instanceof Title && $title->getFragment() == '' && $title->equals( $parser->mTitle ) ) - $parser->mOutput->setDisplayTitle( $text ); - return ''; - } - - static function isRaw( $param ) { - static $mwRaw; - if ( !$mwRaw ) { - $mwRaw =& MagicWord::get( 'rawsuffix' ); - } - if ( is_null( $param ) ) { - return false; - } else { - return $mwRaw->match( $param ); - } - } - - static function statisticsFunction( $func, $raw = null ) { - if ( self::isRaw( $raw ) ) { - return call_user_func( array( 'SiteStats', $func ) ); - } else { - global $wgContLang; - return $wgContLang->formatNum( call_user_func( array( 'SiteStats', $func ) ) ); - } - } - - static function numberofpages( $parser, $raw = null ) { return self::statisticsFunction( 'pages', $raw ); } - static function numberofusers( $parser, $raw = null ) { return self::statisticsFunction( 'users', $raw ); } - static function numberofarticles( $parser, $raw = null ) { return self::statisticsFunction( 'articles', $raw ); } - static function numberoffiles( $parser, $raw = null ) { return self::statisticsFunction( 'images', $raw ); } - static function numberofadmins( $parser, $raw = null ) { return self::statisticsFunction( 'admins', $raw ); } - static function numberofedits( $parser, $raw = null ) { return self::statisticsFunction( 'edits', $raw ); } - - static function pagesinnamespace( $parser, $namespace = 0, $raw = null ) { - $count = SiteStats::pagesInNs( intval( $namespace ) ); - if ( self::isRaw( $raw ) ) { - global $wgContLang; - return $wgContLang->formatNum( $count ); - } else { - return $count; - } - } - - static function language( $parser, $arg = '' ) { - global $wgContLang; - $lang = $wgContLang->getLanguageName( strtolower( $arg ) ); - return $lang != '' ? $lang : $arg; - } - - static function pad( $string = '', $length = 0, $char = 0, $direction = STR_PAD_RIGHT ) { - $length = min( max( $length, 0 ), 500 ); - $char = substr( $char, 0, 1 ); - return ( $string !== '' && (int)$length > 0 && strlen( trim( (string)$char ) ) > 0 ) - ? str_pad( $string, $length, (string)$char, $direction ) - : $string; - } - - static function padleft( $parser, $string = '', $length = 0, $char = 0 ) { - return self::pad( $string, $length, $char, STR_PAD_LEFT ); - } - - static function padright( $parser, $string = '', $length = 0, $char = 0 ) { - return self::pad( $string, $length, $char ); - } - - static function anchorencode( $parser, $text ) { - $a = urlencode( $text ); - $a = strtr( $a, array( '%' => '.', '+' => '_' ) ); - # leave colons alone, however - $a = str_replace( '.3A', ':', $a ); - return $a; - } - - static function special( $parser, $text ) { - $title = SpecialPage::getTitleForAlias( $text ); - if ( $title ) { - return $title->getPrefixedText(); - } else { - return wfMsgForContent( 'nosuchspecialpage' ); - } - } - - public static function defaultsort( $parser, $text ) { - $text = trim( $text ); - if( strlen( $text ) > 0 ) - $parser->setDefaultSort( $text ); - return ''; - } - - public static function filepath( $parser, $name='', $option='' ) { - $file = wfFindFile( $name ); - if( $file ) { - $url = $file->getFullUrl(); - if( $option == 'nowiki' ) { - return "$url"; - } - return $url; - } else { - return ''; - } - } - - /** - * Parser function to extension tag adaptor - */ - public static function tagObj( $parser, $frame, $args ) { - $xpath = false; - if ( !count( $args ) ) { - return ''; - } - $tagName = strtolower( trim( $frame->expand( array_shift( $args ) ) ) ); - - if ( count( $args ) ) { - $inner = $frame->expand( array_shift( $args ) ); - } else { - $inner = null; - } - - $stripList = $parser->getStripList(); - if ( !in_array( $tagName, $stripList ) ) { - return '' . - wfMsg( 'unknown_extension_tag', $tagName ) . - ''; - } - - $attributes = array(); - foreach ( $args as $arg ) { - $bits = $arg->splitArg(); - if ( strval( $bits['index'] ) === '' ) { - $name = $frame->expand( $bits['name'], PPFrame::STRIP_COMMENTS ); - $value = trim( $frame->expand( $bits['value'] ) ); - if ( preg_match( '/^(?:["\'](.+)["\']|""|\'\')$/s', $value, $m ) ) { - $value = isset( $m[1] ) ? $m[1] : ''; - } - $attributes[$name] = $value; - } - } - - $params = array( - 'name' => $tagName, - 'inner' => $inner, - 'attributes' => $attributes, - 'close' => "", - ); - return $parser->extensionSubstitution( $params, $frame ); - } -} - diff --git a/includes/Database.php b/includes/Database.php deleted file mode 100644 index f8738288..00000000 --- a/includes/Database.php +++ /dev/null @@ -1,2443 +0,0 @@ -mData = $data; - } - - function isLOB() { - return false; - } - - function data() { - return $this->mData; - } -}; - -/** - * Utility class - * @addtogroup Database - * - * This allows us to distinguish a blob from a normal string and an array of strings - */ -class Blob { - private $mData; - function __construct($data) { - $this->mData = $data; - } - function fetch() { - return $this->mData; - } -}; - -/** - * Utility class. - * @addtogroup Database - */ -class MySQLField { - private $name, $tablename, $default, $max_length, $nullable, - $is_pk, $is_unique, $is_key, $type; - function __construct ($info) { - $this->name = $info->name; - $this->tablename = $info->table; - $this->default = $info->def; - $this->max_length = $info->max_length; - $this->nullable = !$info->not_null; - $this->is_pk = $info->primary_key; - $this->is_unique = $info->unique_key; - $this->is_multiple = $info->multiple_key; - $this->is_key = ($this->is_pk || $this->is_unique || $this->is_multiple); - $this->type = $info->type; - } - - function name() { - return $this->name; - } - - function tableName() { - return $this->tableName; - } - - function defaultValue() { - return $this->default; - } - - function maxLength() { - return $this->max_length; - } - - function nullable() { - return $this->nullable; - } - - function isKey() { - return $this->is_key; - } - - function isMultipleKey() { - return $this->is_multiple; - } - - function type() { - return $this->type; - } -} - -/****************************************************************************** - * Error classes - *****************************************************************************/ - -/** - * Database error base class - * @addtogroup Database - */ -class DBError extends MWException { - public $db; - - /** - * Construct a database error - * @param Database $db The database object which threw the error - * @param string $error A simple error message to be used for debugging - */ - function __construct( Database &$db, $error ) { - $this->db =& $db; - parent::__construct( $error ); - } -} - -/** - * @addtogroup Database - */ -class DBConnectionError extends DBError { - public $error; - - function __construct( Database &$db, $error = 'unknown error' ) { - $msg = 'DB connection error'; - if ( trim( $error ) != '' ) { - $msg .= ": $error"; - } - $this->error = $error; - parent::__construct( $db, $msg ); - } - - function useOutputPage() { - // Not likely to work - return false; - } - - function useMessageCache() { - // Not likely to work - return false; - } - - function getText() { - return $this->getMessage() . "\n"; - } - - function getLogMessage() { - # Don't send to the exception log - return false; - } - - function getPageTitle() { - global $wgSitename; - return "$wgSitename has a problem"; - } - - function getHTML() { - global $wgTitle, $wgUseFileCache, $title, $wgInputEncoding; - global $wgSitename, $wgServer, $wgMessageCache; - - # I give up, Brion is right. Getting the message cache to work when there is no DB is tricky. - # Hard coding strings instead. - - $noconnect = "

Sorry! This site is experiencing technical difficulties.

Try waiting a few minutes and reloading.

(Can't contact the database server: $1)

"; - $mainpage = 'Main Page'; - $searchdisabled = <<$wgSitename search is disabled for performance reasons. You can search via Google in the meantime. -Note that their indexes of $wgSitename content may be out of date.

', -EOT; - - $googlesearch = " - -
- -
- -\"Google\" - - - - -
WWW $wgServer
- - -
-
-
-"; - $cachederror = "The following is a cached copy of the requested page, and may not be up to date. "; - - # No database access - if ( is_object( $wgMessageCache ) ) { - $wgMessageCache->disable(); - } - - if ( trim( $this->error ) == '' ) { - $this->error = $this->db->getProperty('mServer'); - } - - $text = str_replace( '$1', $this->error, $noconnect ); - $text .= wfGetSiteNotice(); - - if($wgUseFileCache) { - if($wgTitle) { - $t =& $wgTitle; - } else { - if($title) { - $t = Title::newFromURL( $title ); - } elseif (@/**/$_REQUEST['search']) { - $search = $_REQUEST['search']; - return $searchdisabled . - str_replace( array( '$1', '$2' ), array( htmlspecialchars( $search ), - $wgInputEncoding ), $googlesearch ); - } else { - $t = Title::newFromText( $mainpage ); - } - } - - $cache = new HTMLFileCache( $t ); - if( $cache->isFileCached() ) { - // @todo, FIXME: $msg is not defined on the next line. - $msg = '

'.$msg."
\n" . - $cachederror . "

\n"; - - $tag = '
'; - $text = str_replace( - $tag, - $tag . $msg, - $cache->fetchPageText() ); - } - } - - return $text; - } -} - -/** - * @addtogroup Database - */ -class DBQueryError extends DBError { - public $error, $errno, $sql, $fname; - - function __construct( Database &$db, $error, $errno, $sql, $fname ) { - $message = "A database error has occurred\n" . - "Query: $sql\n" . - "Function: $fname\n" . - "Error: $errno $error\n"; - - parent::__construct( $db, $message ); - $this->error = $error; - $this->errno = $errno; - $this->sql = $sql; - $this->fname = $fname; - } - - function getText() { - if ( $this->useMessageCache() ) { - return wfMsg( 'dberrortextcl', htmlspecialchars( $this->getSQL() ), - htmlspecialchars( $this->fname ), $this->errno, htmlspecialchars( $this->error ) ) . "\n"; - } else { - return $this->getMessage(); - } - } - - function getSQL() { - global $wgShowSQLErrors; - if( !$wgShowSQLErrors ) { - return $this->msg( 'sqlhidden', 'SQL hidden' ); - } else { - return $this->sql; - } - } - - function getLogMessage() { - # Don't send to the exception log - return false; - } - - function getPageTitle() { - return $this->msg( 'databaseerror', 'Database error' ); - } - - function getHTML() { - if ( $this->useMessageCache() ) { - return wfMsgNoDB( 'dberrortext', htmlspecialchars( $this->getSQL() ), - htmlspecialchars( $this->fname ), $this->errno, htmlspecialchars( $this->error ) ); - } else { - return nl2br( htmlspecialchars( $this->getMessage() ) ); - } - } -} - -/** - * @addtogroup Database - */ -class DBUnexpectedError extends DBError {} - -/******************************************************************************/ - -/** - * Database abstraction object - * @addtogroup Database - */ -class Database { - -#------------------------------------------------------------------------------ -# Variables -#------------------------------------------------------------------------------ - - protected $mLastQuery = ''; - - protected $mServer, $mUser, $mPassword, $mConn = null, $mDBname; - protected $mOut, $mOpened = false; - - protected $mFailFunction; - protected $mTablePrefix; - protected $mFlags; - protected $mTrxLevel = 0; - protected $mErrorCount = 0; - protected $mLBInfo = array(); - -#------------------------------------------------------------------------------ -# Accessors -#------------------------------------------------------------------------------ - # These optionally set a variable and return the previous state - - /** - * Fail function, takes a Database as a parameter - * Set to false for default, 1 for ignore errors - */ - function failFunction( $function = NULL ) { - return wfSetVar( $this->mFailFunction, $function ); - } - - /** - * Output page, used for reporting errors - * FALSE means discard output - */ - function setOutputPage( $out ) { - $this->mOut = $out; - } - - /** - * Boolean, controls output of large amounts of debug information - */ - function debug( $debug = NULL ) { - return wfSetBit( $this->mFlags, DBO_DEBUG, $debug ); - } - - /** - * Turns buffering of SQL result sets on (true) or off (false). - * Default is "on" and it should not be changed without good reasons. - */ - function bufferResults( $buffer = NULL ) { - if ( is_null( $buffer ) ) { - return !(bool)( $this->mFlags & DBO_NOBUFFER ); - } else { - return !wfSetBit( $this->mFlags, DBO_NOBUFFER, !$buffer ); - } - } - - /** - * Turns on (false) or off (true) the automatic generation and sending - * of a "we're sorry, but there has been a database error" page on - * database errors. Default is on (false). When turned off, the - * code should use lastErrno() and lastError() to handle the - * situation as appropriate. - */ - function ignoreErrors( $ignoreErrors = NULL ) { - return wfSetBit( $this->mFlags, DBO_IGNORE, $ignoreErrors ); - } - - /** - * The current depth of nested transactions - * @param $level Integer: , default NULL. - */ - function trxLevel( $level = NULL ) { - return wfSetVar( $this->mTrxLevel, $level ); - } - - /** - * Number of errors logged, only useful when errors are ignored - */ - function errorCount( $count = NULL ) { - return wfSetVar( $this->mErrorCount, $count ); - } - - /** - * Properties passed down from the server info array of the load balancer - */ - function getLBInfo( $name = NULL ) { - if ( is_null( $name ) ) { - return $this->mLBInfo; - } else { - if ( array_key_exists( $name, $this->mLBInfo ) ) { - return $this->mLBInfo[$name]; - } else { - return NULL; - } - } - } - - function setLBInfo( $name, $value = NULL ) { - if ( is_null( $value ) ) { - $this->mLBInfo = $name; - } else { - $this->mLBInfo[$name] = $value; - } - } - - /** - * Returns true if this database supports (and uses) cascading deletes - */ - function cascadingDeletes() { - return false; - } - - /** - * Returns true if this database supports (and uses) triggers (e.g. on the page table) - */ - function cleanupTriggers() { - return false; - } - - /** - * Returns true if this database is strict about what can be put into an IP field. - * Specifically, it uses a NULL value instead of an empty string. - */ - function strictIPs() { - return false; - } - - /** - * Returns true if this database uses timestamps rather than integers - */ - function realTimestamps() { - return false; - } - - /** - * Returns true if this database does an implicit sort when doing GROUP BY - */ - function implicitGroupby() { - return true; - } - - /** - * Returns true if this database does an implicit order by when the column has an index - * For example: SELECT page_title FROM page LIMIT 1 - */ - function implicitOrderby() { - return true; - } - - /** - * Returns true if this database can do a native search on IP columns - * e.g. this works as expected: .. WHERE rc_ip = '127.42.12.102/32'; - */ - function searchableIPs() { - return false; - } - - /** - * Returns true if this database can use functional indexes - */ - function functionalIndexes() { - return false; - } - - /**#@+ - * Get function - */ - function lastQuery() { return $this->mLastQuery; } - function isOpen() { return $this->mOpened; } - /**#@-*/ - - function setFlag( $flag ) { - $this->mFlags |= $flag; - } - - function clearFlag( $flag ) { - $this->mFlags &= ~$flag; - } - - function getFlag( $flag ) { - return !!($this->mFlags & $flag); - } - - /** - * General read-only accessor - */ - function getProperty( $name ) { - return $this->$name; - } - -#------------------------------------------------------------------------------ -# Other functions -#------------------------------------------------------------------------------ - - /**@{{ - * Constructor. - * @param string $server database server host - * @param string $user database user name - * @param string $password database user password - * @param string $dbname database name - * @param failFunction - * @param $flags - * @param $tablePrefix String: database table prefixes. By default use the prefix gave in LocalSettings.php - */ - function __construct( $server = false, $user = false, $password = false, $dbName = false, - $failFunction = false, $flags = 0, $tablePrefix = 'get from global' ) { - - global $wgOut, $wgDBprefix, $wgCommandLineMode; - # Can't get a reference if it hasn't been set yet - if ( !isset( $wgOut ) ) { - $wgOut = NULL; - } - $this->mOut =& $wgOut; - - $this->mFailFunction = $failFunction; - $this->mFlags = $flags; - - if ( $this->mFlags & DBO_DEFAULT ) { - if ( $wgCommandLineMode ) { - $this->mFlags &= ~DBO_TRX; - } else { - $this->mFlags |= DBO_TRX; - } - } - - /* - // Faster read-only access - if ( wfReadOnly() ) { - $this->mFlags |= DBO_PERSISTENT; - $this->mFlags &= ~DBO_TRX; - }*/ - - /** Get the default table prefix*/ - if ( $tablePrefix == 'get from global' ) { - $this->mTablePrefix = $wgDBprefix; - } else { - $this->mTablePrefix = $tablePrefix; - } - - if ( $server ) { - $this->open( $server, $user, $password, $dbName ); - } - } - - /** - * @static - * @param failFunction - * @param $flags - */ - static function newFromParams( $server, $user, $password, $dbName, $failFunction = false, $flags = 0 ) - { - return new Database( $server, $user, $password, $dbName, $failFunction, $flags ); - } - - /** - * Usually aborts on failure - * If the failFunction is set to a non-zero integer, returns success - */ - function open( $server, $user, $password, $dbName ) { - global $wguname; - wfProfileIn( __METHOD__ ); - - # Test for missing mysql.so - # First try to load it - if (!@extension_loaded('mysql')) { - @dl('mysql.so'); - } - - # Fail now - # Otherwise we get a suppressed fatal error, which is very hard to track down - if ( !function_exists( 'mysql_connect' ) ) { - throw new DBConnectionError( $this, "MySQL functions missing, have you compiled PHP with the --with-mysql option?\n" ); - } - - $this->close(); - $this->mServer = $server; - $this->mUser = $user; - $this->mPassword = $password; - $this->mDBname = $dbName; - - $success = false; - - wfProfileIn("dbconnect-$server"); - - # LIVE PATCH by Tim, ask Domas for why: retry loop - $this->mConn = false; - $max = 3; - for ( $i = 0; $i < $max && !$this->mConn; $i++ ) { - if ( $i > 1 ) { - usleep( 1000 ); - } - if ( $this->mFlags & DBO_PERSISTENT ) { - @/**/$this->mConn = mysql_pconnect( $server, $user, $password ); - } else { - # Create a new connection... - @/**/$this->mConn = mysql_connect( $server, $user, $password, true ); - } - if ($this->mConn === false) { - #$iplus = $i + 1; - #wfLogDBError("Connect loop error $iplus of $max ($server): " . mysql_errno() . " - " . mysql_error()."\n"); - } - } - - wfProfileOut("dbconnect-$server"); - - if ( $dbName != '' ) { - if ( $this->mConn !== false ) { - $success = @/**/mysql_select_db( $dbName, $this->mConn ); - if ( !$success ) { - $error = "Error selecting database $dbName on server {$this->mServer} " . - "from client host {$wguname['nodename']}\n"; - wfLogDBError(" Error selecting database $dbName on server {$this->mServer} \n"); - wfDebug( $error ); - } - } else { - wfDebug( "DB connection error\n" ); - wfDebug( "Server: $server, User: $user, Password: " . - substr( $password, 0, 3 ) . "..., error: " . mysql_error() . "\n" ); - $success = false; - } - } else { - # Delay USE query - $success = (bool)$this->mConn; - } - - if ( $success ) { - $version = $this->getServerVersion(); - if ( version_compare( $version, '4.1' ) >= 0 ) { - // Tell the server we're communicating with it in UTF-8. - // This may engage various charset conversions. - global $wgDBmysql5; - if( $wgDBmysql5 ) { - $this->query( 'SET NAMES utf8', __METHOD__ ); - } - // Turn off strict mode - $this->query( "SET sql_mode = ''", __METHOD__ ); - } - - // Turn off strict mode if it is on - } else { - $this->reportConnectionError(); - } - - $this->mOpened = $success; - wfProfileOut( __METHOD__ ); - return $success; - } - /**@}}*/ - - /** - * Closes a database connection. - * if it is open : commits any open transactions - * - * @return bool operation success. true if already closed. - */ - function close() - { - $this->mOpened = false; - if ( $this->mConn ) { - if ( $this->trxLevel() ) { - $this->immediateCommit(); - } - return mysql_close( $this->mConn ); - } else { - return true; - } - } - - /** - * @param string $error fallback error message, used if none is given by MySQL - */ - function reportConnectionError( $error = 'Unknown error' ) { - $myError = $this->lastError(); - if ( $myError ) { - $error = $myError; - } - - if ( $this->mFailFunction ) { - # Legacy error handling method - if ( !is_int( $this->mFailFunction ) ) { - $ff = $this->mFailFunction; - $ff( $this, $error ); - } - } else { - # New method - wfLogDBError( "Connection error: $error\n" ); - throw new DBConnectionError( $this, $error ); - } - } - - /** - * Usually aborts on failure. If errors are explicitly ignored, returns success. - * - * @param $sql String: SQL query - * @param $fname String: Name of the calling function, for profiling/SHOW PROCESSLIST - * comment (you can use __METHOD__ or add some extra info) - * @param $tempIgnore Bool: Whether to avoid throwing an exception on errors... - * maybe best to catch the exception instead? - * @return true for a successful write query, ResultWrapper object for a successful read query, - * or false on failure if $tempIgnore set - * @throws DBQueryError Thrown when the database returns an error of any kind - */ - public function query( $sql, $fname = '', $tempIgnore = false ) { - global $wgProfiling; - - if ( $wgProfiling ) { - # generalizeSQL will probably cut down the query to reasonable - # logging size most of the time. The substr is really just a sanity check. - - # Who's been wasting my precious column space? -- TS - #$profName = 'query: ' . $fname . ' ' . substr( Database::generalizeSQL( $sql ), 0, 255 ); - - if ( is_null( $this->getLBInfo( 'master' ) ) ) { - $queryProf = 'query: ' . substr( Database::generalizeSQL( $sql ), 0, 255 ); - $totalProf = 'Database::query'; - } else { - $queryProf = 'query-m: ' . substr( Database::generalizeSQL( $sql ), 0, 255 ); - $totalProf = 'Database::query-master'; - } - wfProfileIn( $totalProf ); - wfProfileIn( $queryProf ); - } - - $this->mLastQuery = $sql; - - # Add a comment for easy SHOW PROCESSLIST interpretation - #if ( $fname ) { - global $wgUser; - if ( is_object( $wgUser ) && !($wgUser instanceof StubObject) ) { - $userName = $wgUser->getName(); - if ( mb_strlen( $userName ) > 15 ) { - $userName = mb_substr( $userName, 0, 15 ) . '...'; - } - $userName = str_replace( '/', '', $userName ); - } else { - $userName = ''; - } - $commentedSql = preg_replace('/\s/', " /* $fname $userName */ ", $sql, 1); - #} else { - # $commentedSql = $sql; - #} - - # If DBO_TRX is set, start a transaction - if ( ( $this->mFlags & DBO_TRX ) && !$this->trxLevel() && - $sql != 'BEGIN' && $sql != 'COMMIT' && $sql != 'ROLLBACK') { - // avoid establishing transactions for SHOW and SET statements too - - // that would delay transaction initializations to once connection - // is really used by application - $sqlstart = substr($sql,0,10); // very much worth it, benchmark certified(tm) - if (strpos($sqlstart,"SHOW ")!==0 and strpos($sqlstart,"SET ")!==0) - $this->begin(); - } - - if ( $this->debug() ) { - $sqlx = substr( $commentedSql, 0, 500 ); - $sqlx = strtr( $sqlx, "\t\n", ' ' ); - wfDebug( "SQL: $sqlx\n" ); - } - - # Do the query and handle errors - $ret = $this->doQuery( $commentedSql ); - - # Try reconnecting if the connection was lost - if ( false === $ret && ( $this->lastErrno() == 2013 || $this->lastErrno() == 2006 ) ) { - # Transaction is gone, like it or not - $this->mTrxLevel = 0; - wfDebug( "Connection lost, reconnecting...\n" ); - if ( $this->ping() ) { - wfDebug( "Reconnected\n" ); - $sqlx = substr( $commentedSql, 0, 500 ); - $sqlx = strtr( $sqlx, "\t\n", ' ' ); - global $wgRequestTime; - $elapsed = round( microtime(true) - $wgRequestTime, 3 ); - wfLogDBError( "Connection lost and reconnected after {$elapsed}s, query: $sqlx\n" ); - $ret = $this->doQuery( $commentedSql ); - } else { - wfDebug( "Failed\n" ); - } - } - - if ( false === $ret ) { - $this->reportQueryError( $this->lastError(), $this->lastErrno(), $sql, $fname, $tempIgnore ); - } - - if ( $wgProfiling ) { - wfProfileOut( $queryProf ); - wfProfileOut( $totalProf ); - } - return $this->resultObject( $ret ); - } - - /** - * The DBMS-dependent part of query() - * @param $sql String: SQL query. - * @return Result object to feed to fetchObject, fetchRow, ...; or false on failure - * @access private - */ - /*private*/ function doQuery( $sql ) { - if( $this->bufferResults() ) { - $ret = mysql_query( $sql, $this->mConn ); - } else { - $ret = mysql_unbuffered_query( $sql, $this->mConn ); - } - return $ret; - } - - /** - * @param $error - * @param $errno - * @param $sql - * @param string $fname - * @param bool $tempIgnore - */ - function reportQueryError( $error, $errno, $sql, $fname, $tempIgnore = false ) { - global $wgCommandLineMode; - # Ignore errors during error handling to avoid infinite recursion - $ignore = $this->ignoreErrors( true ); - ++$this->mErrorCount; - - if( $ignore || $tempIgnore ) { - wfDebug("SQL ERROR (ignored): $error\n"); - $this->ignoreErrors( $ignore ); - } else { - $sql1line = str_replace( "\n", "\\n", $sql ); - wfLogDBError("$fname\t{$this->mServer}\t$errno\t$error\t$sql1line\n"); - wfDebug("SQL ERROR: " . $error . "\n"); - throw new DBQueryError( $this, $error, $errno, $sql, $fname ); - } - } - - - /** - * Intended to be compatible with the PEAR::DB wrapper functions. - * http://pear.php.net/manual/en/package.database.db.intro-execute.php - * - * ? = scalar value, quoted as necessary - * ! = raw SQL bit (a function for instance) - * & = filename; reads the file and inserts as a blob - * (we don't use this though...) - */ - function prepare( $sql, $func = 'Database::prepare' ) { - /* MySQL doesn't support prepared statements (yet), so just - pack up the query for reference. We'll manually replace - the bits later. */ - return array( 'query' => $sql, 'func' => $func ); - } - - function freePrepared( $prepared ) { - /* No-op for MySQL */ - } - - /** - * Execute a prepared query with the various arguments - * @param string $prepared the prepared sql - * @param mixed $args Either an array here, or put scalars as varargs - */ - function execute( $prepared, $args = null ) { - if( !is_array( $args ) ) { - # Pull the var args - $args = func_get_args(); - array_shift( $args ); - } - $sql = $this->fillPrepared( $prepared['query'], $args ); - return $this->query( $sql, $prepared['func'] ); - } - - /** - * Prepare & execute an SQL statement, quoting and inserting arguments - * in the appropriate places. - * @param string $query - * @param string $args ... - */ - function safeQuery( $query, $args = null ) { - $prepared = $this->prepare( $query, 'Database::safeQuery' ); - if( !is_array( $args ) ) { - # Pull the var args - $args = func_get_args(); - array_shift( $args ); - } - $retval = $this->execute( $prepared, $args ); - $this->freePrepared( $prepared ); - return $retval; - } - - /** - * For faking prepared SQL statements on DBs that don't support - * it directly. - * @param string $preparedSql - a 'preparable' SQL statement - * @param array $args - array of arguments to fill it with - * @return string executable SQL - */ - function fillPrepared( $preparedQuery, $args ) { - reset( $args ); - $this->preparedArgs =& $args; - return preg_replace_callback( '/(\\\\[?!&]|[?!&])/', - array( &$this, 'fillPreparedArg' ), $preparedQuery ); - } - - /** - * preg_callback func for fillPrepared() - * The arguments should be in $this->preparedArgs and must not be touched - * while we're doing this. - * - * @param array $matches - * @return string - * @private - */ - function fillPreparedArg( $matches ) { - switch( $matches[1] ) { - case '\\?': return '?'; - case '\\!': return '!'; - case '\\&': return '&'; - } - list( /* $n */ , $arg ) = each( $this->preparedArgs ); - switch( $matches[1] ) { - case '?': return $this->addQuotes( $arg ); - case '!': return $arg; - case '&': - # return $this->addQuotes( file_get_contents( $arg ) ); - throw new DBUnexpectedError( $this, '& mode is not implemented. If it\'s really needed, uncomment the line above.' ); - default: - throw new DBUnexpectedError( $this, 'Received invalid match. This should never happen!' ); - } - } - - /**#@+ - * @param mixed $res A SQL result - */ - /** - * Free a result object - */ - function freeResult( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - if ( !@/**/mysql_free_result( $res ) ) { - throw new DBUnexpectedError( $this, "Unable to free MySQL result" ); - } - } - - /** - * Fetch the next row from the given result object, in object form. - * Fields can be retrieved with $row->fieldname, with fields acting like - * member variables. - * - * @param $res SQL result object as returned from Database::query(), etc. - * @return MySQL row object - * @throws DBUnexpectedError Thrown if the database returns an error - */ - function fetchObject( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - @/**/$row = mysql_fetch_object( $res ); - if( $this->lastErrno() ) { - throw new DBUnexpectedError( $this, 'Error in fetchObject(): ' . htmlspecialchars( $this->lastError() ) ); - } - return $row; - } - - /** - * Fetch the next row from the given result object, in associative array - * form. Fields are retrieved with $row['fieldname']. - * - * @param $res SQL result object as returned from Database::query(), etc. - * @return MySQL row object - * @throws DBUnexpectedError Thrown if the database returns an error - */ - function fetchRow( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - @/**/$row = mysql_fetch_array( $res ); - if ( $this->lastErrno() ) { - throw new DBUnexpectedError( $this, 'Error in fetchRow(): ' . htmlspecialchars( $this->lastError() ) ); - } - return $row; - } - - /** - * Get the number of rows in a result object - */ - function numRows( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - @/**/$n = mysql_num_rows( $res ); - if( $this->lastErrno() ) { - throw new DBUnexpectedError( $this, 'Error in numRows(): ' . htmlspecialchars( $this->lastError() ) ); - } - return $n; - } - - /** - * Get the number of fields in a result object - * See documentation for mysql_num_fields() - */ - function numFields( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - return mysql_num_fields( $res ); - } - - /** - * Get a field name in a result object - * See documentation for mysql_field_name(): - * http://www.php.net/mysql_field_name - */ - function fieldName( $res, $n ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - return mysql_field_name( $res, $n ); - } - - /** - * Get the inserted value of an auto-increment row - * - * The value inserted should be fetched from nextSequenceValue() - * - * Example: - * $id = $dbw->nextSequenceValue('page_page_id_seq'); - * $dbw->insert('page',array('page_id' => $id)); - * $id = $dbw->insertId(); - */ - function insertId() { return mysql_insert_id( $this->mConn ); } - - /** - * Change the position of the cursor in a result object - * See mysql_data_seek() - */ - function dataSeek( $res, $row ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - return mysql_data_seek( $res, $row ); - } - - /** - * Get the last error number - * See mysql_errno() - */ - function lastErrno() { - if ( $this->mConn ) { - return mysql_errno( $this->mConn ); - } else { - return mysql_errno(); - } - } - - /** - * Get a description of the last error - * See mysql_error() for more details - */ - function lastError() { - if ( $this->mConn ) { - # Even if it's non-zero, it can still be invalid - wfSuppressWarnings(); - $error = mysql_error( $this->mConn ); - if ( !$error ) { - $error = mysql_error(); - } - wfRestoreWarnings(); - } else { - $error = mysql_error(); - } - if( $error ) { - $error .= ' (' . $this->mServer . ')'; - } - return $error; - } - /** - * Get the number of rows affected by the last write query - * See mysql_affected_rows() for more details - */ - function affectedRows() { return mysql_affected_rows( $this->mConn ); } - /**#@-*/ // end of template : @param $result - - /** - * Simple UPDATE wrapper - * Usually aborts on failure - * If errors are explicitly ignored, returns success - * - * This function exists for historical reasons, Database::update() has a more standard - * calling convention and feature set - */ - function set( $table, $var, $value, $cond, $fname = 'Database::set' ) - { - $table = $this->tableName( $table ); - $sql = "UPDATE $table SET $var = '" . - $this->strencode( $value ) . "' WHERE ($cond)"; - return (bool)$this->query( $sql, $fname ); - } - - /** - * Simple SELECT wrapper, returns a single field, input must be encoded - * Usually aborts on failure - * If errors are explicitly ignored, returns FALSE on failure - */ - function selectField( $table, $var, $cond='', $fname = 'Database::selectField', $options = array() ) { - if ( !is_array( $options ) ) { - $options = array( $options ); - } - $options['LIMIT'] = 1; - - $res = $this->select( $table, $var, $cond, $fname, $options ); - if ( $res === false || !$this->numRows( $res ) ) { - return false; - } - $row = $this->fetchRow( $res ); - if ( $row !== false ) { - $this->freeResult( $res ); - return $row[0]; - } else { - return false; - } - } - - /** - * Returns an optional USE INDEX clause to go after the table, and a - * string to go at the end of the query - * - * @private - * - * @param array $options an associative array of options to be turned into - * an SQL query, valid keys are listed in the function. - * @return array - */ - function makeSelectOptions( $options ) { - $preLimitTail = $postLimitTail = ''; - $startOpts = ''; - - $noKeyOptions = array(); - foreach ( $options as $key => $option ) { - if ( is_numeric( $key ) ) { - $noKeyOptions[$option] = true; - } - } - - if ( isset( $options['GROUP BY'] ) ) $preLimitTail .= " GROUP BY {$options['GROUP BY']}"; - if ( isset( $options['HAVING'] ) ) $preLimitTail .= " HAVING {$options['HAVING']}"; - if ( isset( $options['ORDER BY'] ) ) $preLimitTail .= " ORDER BY {$options['ORDER BY']}"; - - //if (isset($options['LIMIT'])) { - // $tailOpts .= $this->limitResult('', $options['LIMIT'], - // isset($options['OFFSET']) ? $options['OFFSET'] - // : false); - //} - - if ( isset( $noKeyOptions['FOR UPDATE'] ) ) $postLimitTail .= ' FOR UPDATE'; - if ( isset( $noKeyOptions['LOCK IN SHARE MODE'] ) ) $postLimitTail .= ' LOCK IN SHARE MODE'; - if ( isset( $noKeyOptions['DISTINCT'] ) || isset( $noKeyOptions['DISTINCTROW'] ) ) $startOpts .= 'DISTINCT'; - - # Various MySQL extensions - if ( isset( $noKeyOptions['STRAIGHT_JOIN'] ) ) $startOpts .= ' /*! STRAIGHT_JOIN */'; - if ( isset( $noKeyOptions['HIGH_PRIORITY'] ) ) $startOpts .= ' HIGH_PRIORITY'; - if ( isset( $noKeyOptions['SQL_BIG_RESULT'] ) ) $startOpts .= ' SQL_BIG_RESULT'; - if ( isset( $noKeyOptions['SQL_BUFFER_RESULT'] ) ) $startOpts .= ' SQL_BUFFER_RESULT'; - if ( isset( $noKeyOptions['SQL_SMALL_RESULT'] ) ) $startOpts .= ' SQL_SMALL_RESULT'; - if ( isset( $noKeyOptions['SQL_CALC_FOUND_ROWS'] ) ) $startOpts .= ' SQL_CALC_FOUND_ROWS'; - if ( isset( $noKeyOptions['SQL_CACHE'] ) ) $startOpts .= ' SQL_CACHE'; - if ( isset( $noKeyOptions['SQL_NO_CACHE'] ) ) $startOpts .= ' SQL_NO_CACHE'; - - if ( isset( $options['USE INDEX'] ) && ! is_array( $options['USE INDEX'] ) ) { - $useIndex = $this->useIndexClause( $options['USE INDEX'] ); - } else { - $useIndex = ''; - } - - return array( $startOpts, $useIndex, $preLimitTail, $postLimitTail ); - } - - /** - * SELECT wrapper - * - * @param mixed $table Array or string, table name(s) (prefix auto-added) - * @param mixed $vars Array or string, field name(s) to be retrieved - * @param mixed $conds Array or string, condition(s) for WHERE - * @param string $fname Calling function name (use __METHOD__) for logs/profiling - * @param array $options Associative array of options (e.g. array('GROUP BY' => 'page_title')), - * see Database::makeSelectOptions code for list of supported stuff - * @return mixed Database result resource (feed to Database::fetchObject or whatever), or false on failure - */ - function select( $table, $vars, $conds='', $fname = 'Database::select', $options = array() ) - { - if( is_array( $vars ) ) { - $vars = implode( ',', $vars ); - } - if( !is_array( $options ) ) { - $options = array( $options ); - } - if( is_array( $table ) ) { - if ( isset( $options['USE INDEX'] ) && is_array( $options['USE INDEX'] ) ) - $from = ' FROM ' . $this->tableNamesWithUseIndex( $table, $options['USE INDEX'] ); - else - $from = ' FROM ' . implode( ',', array_map( array( &$this, 'tableName' ), $table ) ); - } elseif ($table!='') { - if ($table{0}==' ') { - $from = ' FROM ' . $table; - } else { - $from = ' FROM ' . $this->tableName( $table ); - } - } else { - $from = ''; - } - - list( $startOpts, $useIndex, $preLimitTail, $postLimitTail ) = $this->makeSelectOptions( $options ); - - if( !empty( $conds ) ) { - if ( is_array( $conds ) ) { - $conds = $this->makeList( $conds, LIST_AND ); - } - $sql = "SELECT $startOpts $vars $from $useIndex WHERE $conds $preLimitTail"; - } else { - $sql = "SELECT $startOpts $vars $from $useIndex $preLimitTail"; - } - - if (isset($options['LIMIT'])) - $sql = $this->limitResult($sql, $options['LIMIT'], - isset($options['OFFSET']) ? $options['OFFSET'] : false); - $sql = "$sql $postLimitTail"; - - if (isset($options['EXPLAIN'])) { - $sql = 'EXPLAIN ' . $sql; - } - return $this->query( $sql, $fname ); - } - - /** - * Single row SELECT wrapper - * Aborts or returns FALSE on error - * - * $vars: the selected variables - * $conds: a condition map, terms are ANDed together. - * Items with numeric keys are taken to be literal conditions - * Takes an array of selected variables, and a condition map, which is ANDed - * e.g: selectRow( "page", array( "page_id" ), array( "page_namespace" => - * NS_MAIN, "page_title" => "Astronomy" ) ) would return an object where - * $obj- >page_id is the ID of the Astronomy article - * - * @todo migrate documentation to phpdocumentor format - */ - function selectRow( $table, $vars, $conds, $fname = 'Database::selectRow', $options = array() ) { - $options['LIMIT'] = 1; - $res = $this->select( $table, $vars, $conds, $fname, $options ); - if ( $res === false ) - return false; - if ( !$this->numRows($res) ) { - $this->freeResult($res); - return false; - } - $obj = $this->fetchObject( $res ); - $this->freeResult( $res ); - return $obj; - - } - - /** - * Estimate rows in dataset - * Returns estimated count, based on EXPLAIN output - * Takes same arguments as Database::select() - */ - - function estimateRowCount( $table, $vars='*', $conds='', $fname = 'Database::estimateRowCount', $options = array() ) { - $options['EXPLAIN']=true; - $res = $this->select ($table, $vars, $conds, $fname, $options ); - if ( $res === false ) - return false; - if (!$this->numRows($res)) { - $this->freeResult($res); - return 0; - } - - $rows=1; - - while( $plan = $this->fetchObject( $res ) ) { - $rows *= ($plan->rows > 0)?$plan->rows:1; // avoid resetting to zero - } - - $this->freeResult($res); - return $rows; - } - - - /** - * Removes most variables from an SQL query and replaces them with X or N for numbers. - * It's only slightly flawed. Don't use for anything important. - * - * @param string $sql A SQL Query - * @static - */ - static function generalizeSQL( $sql ) { - # This does the same as the regexp below would do, but in such a way - # as to avoid crashing php on some large strings. - # $sql = preg_replace ( "/'([^\\\\']|\\\\.)*'|\"([^\\\\\"]|\\\\.)*\"/", "'X'", $sql); - - $sql = str_replace ( "\\\\", '', $sql); - $sql = str_replace ( "\\'", '', $sql); - $sql = str_replace ( "\\\"", '', $sql); - $sql = preg_replace ("/'.*'/s", "'X'", $sql); - $sql = preg_replace ('/".*"/s', "'X'", $sql); - - # All newlines, tabs, etc replaced by single space - $sql = preg_replace ( '/\s+/', ' ', $sql); - - # All numbers => N - $sql = preg_replace ('/-?[0-9]+/s', 'N', $sql); - - return $sql; - } - - /** - * Determines whether a field exists in a table - * Usually aborts on failure - * If errors are explicitly ignored, returns NULL on failure - */ - function fieldExists( $table, $field, $fname = 'Database::fieldExists' ) { - $table = $this->tableName( $table ); - $res = $this->query( 'DESCRIBE '.$table, $fname ); - if ( !$res ) { - return NULL; - } - - $found = false; - - while ( $row = $this->fetchObject( $res ) ) { - if ( $row->Field == $field ) { - $found = true; - break; - } - } - return $found; - } - - /** - * Determines whether an index exists - * Usually aborts on failure - * If errors are explicitly ignored, returns NULL on failure - */ - function indexExists( $table, $index, $fname = 'Database::indexExists' ) { - $info = $this->indexInfo( $table, $index, $fname ); - if ( is_null( $info ) ) { - return NULL; - } else { - return $info !== false; - } - } - - - /** - * Get information about an index into an object - * Returns false if the index does not exist - */ - function indexInfo( $table, $index, $fname = 'Database::indexInfo' ) { - # SHOW INDEX works in MySQL 3.23.58, but SHOW INDEXES does not. - # SHOW INDEX should work for 3.x and up: - # http://dev.mysql.com/doc/mysql/en/SHOW_INDEX.html - $table = $this->tableName( $table ); - $sql = 'SHOW INDEX FROM '.$table; - $res = $this->query( $sql, $fname ); - if ( !$res ) { - return NULL; - } - - $result = array(); - while ( $row = $this->fetchObject( $res ) ) { - if ( $row->Key_name == $index ) { - $result[] = $row; - } - } - $this->freeResult($res); - - return empty($result) ? false : $result; - } - - /** - * Query whether a given table exists - */ - function tableExists( $table ) { - $table = $this->tableName( $table ); - $old = $this->ignoreErrors( true ); - $res = $this->query( "SELECT 1 FROM $table LIMIT 1" ); - $this->ignoreErrors( $old ); - if( $res ) { - $this->freeResult( $res ); - return true; - } else { - return false; - } - } - - /** - * mysql_fetch_field() wrapper - * Returns false if the field doesn't exist - * - * @param $table - * @param $field - */ - function fieldInfo( $table, $field ) { - $table = $this->tableName( $table ); - $res = $this->query( "SELECT * FROM $table LIMIT 1" ); - $n = mysql_num_fields( $res->result ); - for( $i = 0; $i < $n; $i++ ) { - $meta = mysql_fetch_field( $res->result, $i ); - if( $field == $meta->name ) { - return new MySQLField($meta); - } - } - return false; - } - - /** - * mysql_field_type() wrapper - */ - function fieldType( $res, $index ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - return mysql_field_type( $res, $index ); - } - - /** - * Determines if a given index is unique - */ - function indexUnique( $table, $index ) { - $indexInfo = $this->indexInfo( $table, $index ); - if ( !$indexInfo ) { - return NULL; - } - return !$indexInfo[0]->Non_unique; - } - - /** - * INSERT wrapper, inserts an array into a table - * - * $a may be a single associative array, or an array of these with numeric keys, for - * multi-row insert. - * - * Usually aborts on failure - * If errors are explicitly ignored, returns success - */ - function insert( $table, $a, $fname = 'Database::insert', $options = array() ) { - # No rows to insert, easy just return now - if ( !count( $a ) ) { - return true; - } - - $table = $this->tableName( $table ); - if ( !is_array( $options ) ) { - $options = array( $options ); - } - if ( isset( $a[0] ) && is_array( $a[0] ) ) { - $multi = true; - $keys = array_keys( $a[0] ); - } else { - $multi = false; - $keys = array_keys( $a ); - } - - $sql = 'INSERT ' . implode( ' ', $options ) . - " INTO $table (" . implode( ',', $keys ) . ') VALUES '; - - if ( $multi ) { - $first = true; - foreach ( $a as $row ) { - if ( $first ) { - $first = false; - } else { - $sql .= ','; - } - $sql .= '(' . $this->makeList( $row ) . ')'; - } - } else { - $sql .= '(' . $this->makeList( $a ) . ')'; - } - return (bool)$this->query( $sql, $fname ); - } - - /** - * Make UPDATE options for the Database::update function - * - * @private - * @param array $options The options passed to Database::update - * @return string - */ - function makeUpdateOptions( $options ) { - if( !is_array( $options ) ) { - $options = array( $options ); - } - $opts = array(); - if ( in_array( 'LOW_PRIORITY', $options ) ) - $opts[] = $this->lowPriorityOption(); - if ( in_array( 'IGNORE', $options ) ) - $opts[] = 'IGNORE'; - return implode(' ', $opts); - } - - /** - * UPDATE wrapper, takes a condition array and a SET array - * - * @param string $table The table to UPDATE - * @param array $values An array of values to SET - * @param array $conds An array of conditions (WHERE). Use '*' to update all rows. - * @param string $fname The Class::Function calling this function - * (for the log) - * @param array $options An array of UPDATE options, can be one or - * more of IGNORE, LOW_PRIORITY - * @return bool - */ - function update( $table, $values, $conds, $fname = 'Database::update', $options = array() ) { - $table = $this->tableName( $table ); - $opts = $this->makeUpdateOptions( $options ); - $sql = "UPDATE $opts $table SET " . $this->makeList( $values, LIST_SET ); - if ( $conds != '*' ) { - $sql .= " WHERE " . $this->makeList( $conds, LIST_AND ); - } - return $this->query( $sql, $fname ); - } - - /** - * Makes an encoded list of strings from an array - * $mode: - * LIST_COMMA - comma separated, no field names - * LIST_AND - ANDed WHERE clause (without the WHERE) - * LIST_OR - ORed WHERE clause (without the WHERE) - * LIST_SET - comma separated with field names, like a SET clause - * LIST_NAMES - comma separated field names - */ - function makeList( $a, $mode = LIST_COMMA ) { - if ( !is_array( $a ) ) { - throw new DBUnexpectedError( $this, 'Database::makeList called with incorrect parameters' ); - } - - $first = true; - $list = ''; - foreach ( $a as $field => $value ) { - if ( !$first ) { - if ( $mode == LIST_AND ) { - $list .= ' AND '; - } elseif($mode == LIST_OR) { - $list .= ' OR '; - } else { - $list .= ','; - } - } else { - $first = false; - } - if ( ($mode == LIST_AND || $mode == LIST_OR) && is_numeric( $field ) ) { - $list .= "($value)"; - } elseif ( ($mode == LIST_SET) && is_numeric( $field ) ) { - $list .= "$value"; - } elseif ( ($mode == LIST_AND || $mode == LIST_OR) && is_array($value) ) { - if( count( $value ) == 0 ) { - // Empty input... or should this throw an error? - $list .= '0'; - } elseif( count( $value ) == 1 ) { - // Special-case single values, as IN isn't terribly efficient - $list .= $field." = ".$this->addQuotes( $value[0] ); - } else { - $list .= $field." IN (".$this->makeList($value).") "; - } - } elseif( is_null($value) ) { - if ( $mode == LIST_AND || $mode == LIST_OR ) { - $list .= "$field IS "; - } elseif ( $mode == LIST_SET ) { - $list .= "$field = "; - } - $list .= 'NULL'; - } else { - if ( $mode == LIST_AND || $mode == LIST_OR || $mode == LIST_SET ) { - $list .= "$field = "; - } - $list .= $mode == LIST_NAMES ? $value : $this->addQuotes( $value ); - } - } - return $list; - } - - /** - * Change the current database - */ - function selectDB( $db ) { - $this->mDBname = $db; - return mysql_select_db( $db, $this->mConn ); - } - - /** - * Format a table name ready for use in constructing an SQL query - * - * This does two important things: it quotes table names which as necessary, - * and it adds a table prefix if there is one. - * - * All functions of this object which require a table name call this function - * themselves. Pass the canonical name to such functions. This is only needed - * when calling query() directly. - * - * @param string $name database table name - */ - function tableName( $name ) { - global $wgSharedDB; - # Skip quoted literals - if ( $name{0} != '`' ) { - if ( $this->mTablePrefix !== '' && strpos( $name, '.' ) === false ) { - $name = "{$this->mTablePrefix}$name"; - } - if ( isset( $wgSharedDB ) && "{$this->mTablePrefix}user" == $name ) { - $name = "`$wgSharedDB`.`$name`"; - } else { - # Standard quoting - $name = "`$name`"; - } - } - return $name; - } - - /** - * Fetch a number of table names into an array - * This is handy when you need to construct SQL for joins - * - * Example: - * extract($dbr->tableNames('user','watchlist')); - * $sql = "SELECT wl_namespace,wl_title FROM $watchlist,$user - * WHERE wl_user=user_id AND wl_user=$nameWithQuotes"; - */ - public function tableNames() { - $inArray = func_get_args(); - $retVal = array(); - foreach ( $inArray as $name ) { - $retVal[$name] = $this->tableName( $name ); - } - return $retVal; - } - - /** - * Fetch a number of table names into an zero-indexed numerical array - * This is handy when you need to construct SQL for joins - * - * Example: - * list( $user, $watchlist ) = $dbr->tableNamesN('user','watchlist'); - * $sql = "SELECT wl_namespace,wl_title FROM $watchlist,$user - * WHERE wl_user=user_id AND wl_user=$nameWithQuotes"; - */ - public function tableNamesN() { - $inArray = func_get_args(); - $retVal = array(); - foreach ( $inArray as $name ) { - $retVal[] = $this->tableName( $name ); - } - return $retVal; - } - - /** - * @private - */ - function tableNamesWithUseIndex( $tables, $use_index ) { - $ret = array(); - - foreach ( $tables as $table ) - if ( @$use_index[$table] !== null ) - $ret[] = $this->tableName( $table ) . ' ' . $this->useIndexClause( implode( ',', (array)$use_index[$table] ) ); - else - $ret[] = $this->tableName( $table ); - - return implode( ',', $ret ); - } - - /** - * Wrapper for addslashes() - * @param string $s String to be slashed. - * @return string slashed string. - */ - function strencode( $s ) { - return mysql_real_escape_string( $s, $this->mConn ); - } - - /** - * If it's a string, adds quotes and backslashes - * Otherwise returns as-is - */ - function addQuotes( $s ) { - if ( is_null( $s ) ) { - return 'NULL'; - } else { - # This will also quote numeric values. This should be harmless, - # and protects against weird problems that occur when they really - # _are_ strings such as article titles and string->number->string - # conversion is not 1:1. - return "'" . $this->strencode( $s ) . "'"; - } - } - - /** - * Escape string for safe LIKE usage - */ - function escapeLike( $s ) { - $s=$this->strencode( $s ); - $s=str_replace(array('%','_'),array('\%','\_'),$s); - return $s; - } - - /** - * Returns an appropriately quoted sequence value for inserting a new row. - * MySQL has autoincrement fields, so this is just NULL. But the PostgreSQL - * subclass will return an integer, and save the value for insertId() - */ - function nextSequenceValue( $seqName ) { - return NULL; - } - - /** - * USE INDEX clause - * PostgreSQL doesn't have them and returns "" - */ - function useIndexClause( $index ) { - return "FORCE INDEX ($index)"; - } - - /** - * REPLACE query wrapper - * PostgreSQL simulates this with a DELETE followed by INSERT - * $row is the row to insert, an associative array - * $uniqueIndexes is an array of indexes. Each element may be either a - * field name or an array of field names - * - * It may be more efficient to leave off unique indexes which are unlikely to collide. - * However if you do this, you run the risk of encountering errors which wouldn't have - * occurred in MySQL - * - * @todo migrate comment to phodocumentor format - */ - function replace( $table, $uniqueIndexes, $rows, $fname = 'Database::replace' ) { - $table = $this->tableName( $table ); - - # Single row case - if ( !is_array( reset( $rows ) ) ) { - $rows = array( $rows ); - } - - $sql = "REPLACE INTO $table (" . implode( ',', array_keys( $rows[0] ) ) .') VALUES '; - $first = true; - foreach ( $rows as $row ) { - if ( $first ) { - $first = false; - } else { - $sql .= ','; - } - $sql .= '(' . $this->makeList( $row ) . ')'; - } - return $this->query( $sql, $fname ); - } - - /** - * DELETE where the condition is a join - * MySQL does this with a multi-table DELETE syntax, PostgreSQL does it with sub-selects - * - * For safety, an empty $conds will not delete everything. If you want to delete all rows where the - * join condition matches, set $conds='*' - * - * DO NOT put the join condition in $conds - * - * @param string $delTable The table to delete from. - * @param string $joinTable The other table. - * @param string $delVar The variable to join on, in the first table. - * @param string $joinVar The variable to join on, in the second table. - * @param array $conds Condition array of field names mapped to variables, ANDed together in the WHERE clause - */ - function deleteJoin( $delTable, $joinTable, $delVar, $joinVar, $conds, $fname = 'Database::deleteJoin' ) { - if ( !$conds ) { - throw new DBUnexpectedError( $this, 'Database::deleteJoin() called with empty $conds' ); - } - - $delTable = $this->tableName( $delTable ); - $joinTable = $this->tableName( $joinTable ); - $sql = "DELETE $delTable FROM $delTable, $joinTable WHERE $delVar=$joinVar "; - if ( $conds != '*' ) { - $sql .= ' AND ' . $this->makeList( $conds, LIST_AND ); - } - - return $this->query( $sql, $fname ); - } - - /** - * Returns the size of a text field, or -1 for "unlimited" - */ - function textFieldSize( $table, $field ) { - $table = $this->tableName( $table ); - $sql = "SHOW COLUMNS FROM $table LIKE \"$field\";"; - $res = $this->query( $sql, 'Database::textFieldSize' ); - $row = $this->fetchObject( $res ); - $this->freeResult( $res ); - - $m = array(); - if ( preg_match( '/\((.*)\)/', $row->Type, $m ) ) { - $size = $m[1]; - } else { - $size = -1; - } - return $size; - } - - /** - * @return string Returns the text of the low priority option if it is supported, or a blank string otherwise - */ - function lowPriorityOption() { - return 'LOW_PRIORITY'; - } - - /** - * DELETE query wrapper - * - * Use $conds == "*" to delete all rows - */ - function delete( $table, $conds, $fname = 'Database::delete' ) { - if ( !$conds ) { - throw new DBUnexpectedError( $this, 'Database::delete() called with no conditions' ); - } - $table = $this->tableName( $table ); - $sql = "DELETE FROM $table"; - if ( $conds != '*' ) { - $sql .= ' WHERE ' . $this->makeList( $conds, LIST_AND ); - } - return $this->query( $sql, $fname ); - } - - /** - * INSERT SELECT wrapper - * $varMap must be an associative array of the form array( 'dest1' => 'source1', ...) - * Source items may be literals rather than field names, but strings should be quoted with Database::addQuotes() - * $conds may be "*" to copy the whole table - * srcTable may be an array of tables. - */ - function insertSelect( $destTable, $srcTable, $varMap, $conds, $fname = 'Database::insertSelect', - $insertOptions = array(), $selectOptions = array() ) - { - $destTable = $this->tableName( $destTable ); - if ( is_array( $insertOptions ) ) { - $insertOptions = implode( ' ', $insertOptions ); - } - if( !is_array( $selectOptions ) ) { - $selectOptions = array( $selectOptions ); - } - list( $startOpts, $useIndex, $tailOpts ) = $this->makeSelectOptions( $selectOptions ); - if( is_array( $srcTable ) ) { - $srcTable = implode( ',', array_map( array( &$this, 'tableName' ), $srcTable ) ); - } else { - $srcTable = $this->tableName( $srcTable ); - } - $sql = "INSERT $insertOptions INTO $destTable (" . implode( ',', array_keys( $varMap ) ) . ')' . - " SELECT $startOpts " . implode( ',', $varMap ) . - " FROM $srcTable $useIndex "; - if ( $conds != '*' ) { - $sql .= ' WHERE ' . $this->makeList( $conds, LIST_AND ); - } - $sql .= " $tailOpts"; - return $this->query( $sql, $fname ); - } - - /** - * Construct a LIMIT query with optional offset - * This is used for query pages - * $sql string SQL query we will append the limit too - * $limit integer the SQL limit - * $offset integer the SQL offset (default false) - */ - function limitResult($sql, $limit, $offset=false) { - if( !is_numeric($limit) ) { - throw new DBUnexpectedError( $this, "Invalid non-numeric limit passed to limitResult()\n" ); - } - return " $sql LIMIT " - . ( (is_numeric($offset) && $offset != 0) ? "{$offset}," : "" ) - . "{$limit} "; - } - function limitResultForUpdate($sql, $num) { - return $this->limitResult($sql, $num, 0); - } - - /** - * Returns an SQL expression for a simple conditional. - * Uses IF on MySQL. - * - * @param string $cond SQL expression which will result in a boolean value - * @param string $trueVal SQL expression to return if true - * @param string $falseVal SQL expression to return if false - * @return string SQL fragment - */ - function conditional( $cond, $trueVal, $falseVal ) { - return " IF($cond, $trueVal, $falseVal) "; - } - - /** - * Determines if the last failure was due to a deadlock - */ - function wasDeadlock() { - return $this->lastErrno() == 1213; - } - - /** - * Perform a deadlock-prone transaction. - * - * This function invokes a callback function to perform a set of write - * queries. If a deadlock occurs during the processing, the transaction - * will be rolled back and the callback function will be called again. - * - * Usage: - * $dbw->deadlockLoop( callback, ... ); - * - * Extra arguments are passed through to the specified callback function. - * - * Returns whatever the callback function returned on its successful, - * iteration, or false on error, for example if the retry limit was - * reached. - */ - function deadlockLoop() { - $myFname = 'Database::deadlockLoop'; - - $this->begin(); - $args = func_get_args(); - $function = array_shift( $args ); - $oldIgnore = $this->ignoreErrors( true ); - $tries = DEADLOCK_TRIES; - if ( is_array( $function ) ) { - $fname = $function[0]; - } else { - $fname = $function; - } - do { - $retVal = call_user_func_array( $function, $args ); - $error = $this->lastError(); - $errno = $this->lastErrno(); - $sql = $this->lastQuery(); - - if ( $errno ) { - if ( $this->wasDeadlock() ) { - # Retry - usleep( mt_rand( DEADLOCK_DELAY_MIN, DEADLOCK_DELAY_MAX ) ); - } else { - $this->reportQueryError( $error, $errno, $sql, $fname ); - } - } - } while( $this->wasDeadlock() && --$tries > 0 ); - $this->ignoreErrors( $oldIgnore ); - if ( $tries <= 0 ) { - $this->query( 'ROLLBACK', $myFname ); - $this->reportQueryError( $error, $errno, $sql, $fname ); - return false; - } else { - $this->query( 'COMMIT', $myFname ); - return $retVal; - } - } - - /** - * Do a SELECT MASTER_POS_WAIT() - * - * @param string $file the binlog file - * @param string $pos the binlog position - * @param integer $timeout the maximum number of seconds to wait for synchronisation - */ - function masterPosWait( $file, $pos, $timeout ) { - $fname = 'Database::masterPosWait'; - wfProfileIn( $fname ); - - - # Commit any open transactions - $this->immediateCommit(); - - # Call doQuery() directly, to avoid opening a transaction if DBO_TRX is set - $encFile = $this->strencode( $file ); - $sql = "SELECT MASTER_POS_WAIT('$encFile', $pos, $timeout)"; - $res = $this->doQuery( $sql ); - if ( $res && $row = $this->fetchRow( $res ) ) { - $this->freeResult( $res ); - wfProfileOut( $fname ); - return $row[0]; - } else { - wfProfileOut( $fname ); - return false; - } - } - - /** - * Get the position of the master from SHOW SLAVE STATUS - */ - function getSlavePos() { - $res = $this->query( 'SHOW SLAVE STATUS', 'Database::getSlavePos' ); - $row = $this->fetchObject( $res ); - if ( $row ) { - return array( $row->Master_Log_File, $row->Read_Master_Log_Pos ); - } else { - return array( false, false ); - } - } - - /** - * Get the position of the master from SHOW MASTER STATUS - */ - function getMasterPos() { - $res = $this->query( 'SHOW MASTER STATUS', 'Database::getMasterPos' ); - $row = $this->fetchObject( $res ); - if ( $row ) { - return array( $row->File, $row->Position ); - } else { - return array( false, false ); - } - } - - /** - * Begin a transaction, committing any previously open transaction - */ - function begin( $fname = 'Database::begin' ) { - $this->query( 'BEGIN', $fname ); - $this->mTrxLevel = 1; - } - - /** - * End a transaction - */ - function commit( $fname = 'Database::commit' ) { - $this->query( 'COMMIT', $fname ); - $this->mTrxLevel = 0; - } - - /** - * Rollback a transaction. - * No-op on non-transactional databases. - */ - function rollback( $fname = 'Database::rollback' ) { - $this->query( 'ROLLBACK', $fname, true ); - $this->mTrxLevel = 0; - } - - /** - * Begin a transaction, committing any previously open transaction - * @deprecated use begin() - */ - function immediateBegin( $fname = 'Database::immediateBegin' ) { - $this->begin(); - } - - /** - * Commit transaction, if one is open - * @deprecated use commit() - */ - function immediateCommit( $fname = 'Database::immediateCommit' ) { - $this->commit(); - } - - /** - * Return MW-style timestamp used for MySQL schema - */ - function timestamp( $ts=0 ) { - return wfTimestamp(TS_MW,$ts); - } - - /** - * Local database timestamp format or null - */ - function timestampOrNull( $ts = null ) { - if( is_null( $ts ) ) { - return null; - } else { - return $this->timestamp( $ts ); - } - } - - /** - * @todo document - */ - function resultObject( $result ) { - if( empty( $result ) ) { - return false; - } elseif ( $result instanceof ResultWrapper ) { - return $result; - } elseif ( $result === true ) { - // Successful write query - return $result; - } else { - return new ResultWrapper( $this, $result ); - } - } - - /** - * Return aggregated value alias - */ - function aggregateValue ($valuedata,$valuename='value') { - return $valuename; - } - - /** - * @return string wikitext of a link to the server software's web site - */ - function getSoftwareLink() { - return "[http://www.mysql.com/ MySQL]"; - } - - /** - * @return string Version information from the database - */ - function getServerVersion() { - return mysql_get_server_info( $this->mConn ); - } - - /** - * Ping the server and try to reconnect if it there is no connection - */ - function ping() { - if( function_exists( 'mysql_ping' ) ) { - return mysql_ping( $this->mConn ); - } else { - wfDebug( "Tried to call mysql_ping but this is ancient PHP version. Faking it!\n" ); - return true; - } - } - - /** - * Get slave lag. - * At the moment, this will only work if the DB user has the PROCESS privilege - */ - function getLag() { - $res = $this->query( 'SHOW PROCESSLIST' ); - # Find slave SQL thread - while ( $row = $this->fetchObject( $res ) ) { - /* This should work for most situations - when default db - * for thread is not specified, it had no events executed, - * and therefore it doesn't know yet how lagged it is. - * - * Relay log I/O thread does not select databases. - */ - if ( $row->User == 'system user' && - $row->State != 'Waiting for master to send event' && - $row->State != 'Connecting to master' && - $row->State != 'Queueing master event to the relay log' && - $row->State != 'Waiting for master update' && - $row->State != 'Requesting binlog dump' - ) { - # This is it, return the time (except -ve) - if ( $row->Time > 0x7fffffff ) { - return false; - } else { - return $row->Time; - } - } - } - return false; - } - - /** - * Get status information from SHOW STATUS in an associative array - */ - function getStatus($which="%") { - $res = $this->query( "SHOW STATUS LIKE '{$which}'" ); - $status = array(); - while ( $row = $this->fetchObject( $res ) ) { - $status[$row->Variable_name] = $row->Value; - } - return $status; - } - - /** - * Return the maximum number of items allowed in a list, or 0 for unlimited. - */ - function maxListLen() { - return 0; - } - - function encodeBlob($b) { - return $b; - } - - function decodeBlob($b) { - return $b; - } - - /** - * Override database's default connection timeout. - * May be useful for very long batch queries such as - * full-wiki dumps, where a single query reads out - * over hours or days. - * @param int $timeout in seconds - */ - public function setTimeout( $timeout ) { - $this->query( "SET net_read_timeout=$timeout" ); - $this->query( "SET net_write_timeout=$timeout" ); - } - - /** - * Read and execute SQL commands from a file. - * Returns true on success, error string on failure - * @param string $filename File name to open - * @param callback $lineCallback Optional function called before reading each line - * @param callback $resultCallback Optional function called for each MySQL result - */ - function sourceFile( $filename, $lineCallback = false, $resultCallback = false ) { - $fp = fopen( $filename, 'r' ); - if ( false === $fp ) { - return "Could not open \"{$filename}\".\n"; - } - $error = $this->sourceStream( $fp, $lineCallback, $resultCallback ); - fclose( $fp ); - return $error; - } - - /** - * Read and execute commands from an open file handle - * Returns true on success, error string on failure - * @param string $fp File handle - * @param callback $lineCallback Optional function called before reading each line - * @param callback $resultCallback Optional function called for each MySQL result - */ - function sourceStream( $fp, $lineCallback = false, $resultCallback = false ) { - $cmd = ""; - $done = false; - $dollarquote = false; - - while ( ! feof( $fp ) ) { - if ( $lineCallback ) { - call_user_func( $lineCallback ); - } - $line = trim( fgets( $fp, 1024 ) ); - $sl = strlen( $line ) - 1; - - if ( $sl < 0 ) { continue; } - if ( '-' == $line{0} && '-' == $line{1} ) { continue; } - - ## Allow dollar quoting for function declarations - if (substr($line,0,4) == '$mw$') { - if ($dollarquote) { - $dollarquote = false; - $done = true; - } - else { - $dollarquote = true; - } - } - else if (!$dollarquote) { - if ( ';' == $line{$sl} && ($sl < 2 || ';' != $line{$sl - 1})) { - $done = true; - $line = substr( $line, 0, $sl ); - } - } - - if ( '' != $cmd ) { $cmd .= ' '; } - $cmd .= "$line\n"; - - if ( $done ) { - $cmd = str_replace(';;', ";", $cmd); - $cmd = $this->replaceVars( $cmd ); - $res = $this->query( $cmd, __METHOD__, true ); - if ( $resultCallback ) { - call_user_func( $resultCallback, $res ); - } - - if ( false === $res ) { - $err = $this->lastError(); - return "Query \"{$cmd}\" failed with error code \"$err\".\n"; - } - - $cmd = ''; - $done = false; - } - } - return true; - } - - - /** - * Replace variables in sourced SQL - */ - protected function replaceVars( $ins ) { - $varnames = array( - 'wgDBserver', 'wgDBname', 'wgDBintlname', 'wgDBuser', - 'wgDBpassword', 'wgDBsqluser', 'wgDBsqlpassword', - 'wgDBadminuser', 'wgDBadminpassword', 'wgDBTableOptions', - ); - - // Ordinary variables - foreach ( $varnames as $var ) { - if( isset( $GLOBALS[$var] ) ) { - $val = addslashes( $GLOBALS[$var] ); // FIXME: safety check? - $ins = str_replace( '{$' . $var . '}', $val, $ins ); - $ins = str_replace( '/*$' . $var . '*/`', '`' . $val, $ins ); - $ins = str_replace( '/*$' . $var . '*/', $val, $ins ); - } - } - - // Table prefixes - $ins = preg_replace_callback( '/\/\*(?:\$wgDBprefix|_)\*\/([a-z_]*)/', - array( &$this, 'tableNameCallback' ), $ins ); - return $ins; - } - - /** - * Table name callback - * @private - */ - protected function tableNameCallback( $matches ) { - return $this->tableName( $matches[1] ); - } - - /* - * Build a concatenation list to feed into a SQL query - */ - function buildConcat( $stringList ) { - return 'CONCAT(' . implode( ',', $stringList ) . ')'; - } - -} - -/** - * Database abstraction object for mySQL - * Inherit all methods and properties of Database::Database() - * - * @addtogroup Database - * @see Database - */ -class DatabaseMysql extends Database { - # Inherit all -} - - -/** - * Result wrapper for grabbing data queried by someone else - * @addtogroup Database - */ -class ResultWrapper implements Iterator { - var $db, $result, $pos = 0, $currentRow = null; - - /** - * Create a new result object from a result resource and a Database object - */ - function ResultWrapper( $database, $result ) { - $this->db = $database; - if ( $result instanceof ResultWrapper ) { - $this->result = $result->result; - } else { - $this->result = $result; - } - } - - /** - * Get the number of rows in a result object - */ - function numRows() { - return $this->db->numRows( $this->result ); - } - - /** - * Fetch the next row from the given result object, in object form. - * Fields can be retrieved with $row->fieldname, with fields acting like - * member variables. - * - * @param $res SQL result object as returned from Database::query(), etc. - * @return MySQL row object - * @throws DBUnexpectedError Thrown if the database returns an error - */ - function fetchObject() { - return $this->db->fetchObject( $this->result ); - } - - /** - * Fetch the next row from the given result object, in associative array - * form. Fields are retrieved with $row['fieldname']. - * - * @param $res SQL result object as returned from Database::query(), etc. - * @return MySQL row object - * @throws DBUnexpectedError Thrown if the database returns an error - */ - function fetchRow() { - return $this->db->fetchRow( $this->result ); - } - - /** - * Free a result object - */ - function free() { - $this->db->freeResult( $this->result ); - unset( $this->result ); - unset( $this->db ); - } - - /** - * Change the position of the cursor in a result object - * See mysql_data_seek() - */ - function seek( $row ) { - $this->db->dataSeek( $this->result, $row ); - } - - /********************* - * Iterator functions - * Note that using these in combination with the non-iterator functions - * above may cause rows to be skipped or repeated. - */ - - function rewind() { - if ($this->numRows()) { - $this->db->dataSeek($this->result, 0); - } - $this->pos = 0; - $this->currentRow = null; - } - - function current() { - if ( is_null( $this->currentRow ) ) { - $this->next(); - } - return $this->currentRow; - } - - function key() { - return $this->pos; - } - - function next() { - $this->pos++; - $this->currentRow = $this->fetchObject(); - return $this->currentRow; - } - - function valid() { - return $this->current() !== false; - } -} - - diff --git a/includes/DatabaseMysql.php b/includes/DatabaseMysql.php deleted file mode 100644 index 79e917b3..00000000 --- a/includes/DatabaseMysql.php +++ /dev/null @@ -1,6 +0,0 @@ - diff --git a/includes/DatabaseOracle.php b/includes/DatabaseOracle.php deleted file mode 100644 index 38485481..00000000 --- a/includes/DatabaseOracle.php +++ /dev/null @@ -1,697 +0,0 @@ -mData = $data; - } - - function getData() { - return $this->mData; - } -} - -/** - * The oci8 extension is fairly weak and doesn't support oci_num_rows, among - * other things. We use a wrapper class to handle that and other - * Oracle-specific bits, like converting column names back to lowercase. - * @addtogroup Database - */ -class ORAResult { - private $rows; - private $cursor; - private $stmt; - private $nrows; - private $db; - - function __construct(&$db, $stmt) { - $this->db =& $db; - if (($this->nrows = oci_fetch_all($stmt, $this->rows, 0, -1, OCI_FETCHSTATEMENT_BY_ROW | OCI_NUM)) === false) { - $e = oci_error($stmt); - $db->reportQueryError($e['message'], $e['code'], '', __FUNCTION__); - return; - } - - $this->cursor = 0; - $this->stmt = $stmt; - } - - function free() { - oci_free_statement($this->stmt); - } - - function seek($row) { - $this->cursor = min($row, $this->nrows); - } - - function numRows() { - return $this->nrows; - } - - function numFields() { - return oci_num_fields($this->stmt); - } - - function fetchObject() { - if ($this->cursor >= $this->nrows) - return false; - - $row = $this->rows[$this->cursor++]; - $ret = new stdClass(); - foreach ($row as $k => $v) { - $lc = strtolower(oci_field_name($this->stmt, $k + 1)); - $ret->$lc = $v; - } - - return $ret; - } - - function fetchAssoc() { - if ($this->cursor >= $this->nrows) - return false; - - $row = $this->rows[$this->cursor++]; - $ret = array(); - foreach ($row as $k => $v) { - $lc = strtolower(oci_field_name($this->stmt, $k + 1)); - $ret[$lc] = $v; - $ret[$k] = $v; - } - return $ret; - } -} - -/** - * @addtogroup Database - */ -class DatabaseOracle extends Database { - var $mInsertId = NULL; - var $mLastResult = NULL; - var $numeric_version = NULL; - var $lastResult = null; - var $cursor = 0; - var $mAffectedRows; - - function DatabaseOracle($server = false, $user = false, $password = false, $dbName = false, - $failFunction = false, $flags = 0 ) - { - - global $wgOut; - # Can't get a reference if it hasn't been set yet - if ( !isset( $wgOut ) ) { - $wgOut = NULL; - } - $this->mOut =& $wgOut; - $this->mFailFunction = $failFunction; - $this->mFlags = $flags; - $this->open( $server, $user, $password, $dbName); - - } - - function cascadingDeletes() { - return true; - } - function cleanupTriggers() { - return true; - } - function strictIPs() { - return true; - } - function realTimestamps() { - return true; - } - function implicitGroupby() { - return false; - } - function implicitOrderby() { - return false; - } - function searchableIPs() { - return true; - } - - static function newFromParams( $server = false, $user = false, $password = false, $dbName = false, - $failFunction = false, $flags = 0) - { - return new DatabaseOracle( $server, $user, $password, $dbName, $failFunction, $flags ); - } - - /** - * Usually aborts on failure - * If the failFunction is set to a non-zero integer, returns success - */ - function open( $server, $user, $password, $dbName ) { - if ( !function_exists( 'oci_connect' ) ) { - throw new DBConnectionError( $this, "Oracle functions missing, have you compiled PHP with the --with-oci8 option?\n (Note: if you recently installed PHP, you may need to restart your webserver and database)\n" ); - } - - # Needed for proper UTF-8 functionality - putenv("NLS_LANG=AMERICAN_AMERICA.AL32UTF8"); - - $this->close(); - $this->mServer = $server; - $this->mUser = $user; - $this->mPassword = $password; - $this->mDBname = $dbName; - - if (!strlen($user)) { ## e.g. the class is being loaded - return; - } - - error_reporting( E_ALL ); - $this->mConn = oci_connect($user, $password, $dbName); - - if ($this->mConn == false) { - wfDebug("DB connection error\n"); - wfDebug("Server: $server, Database: $dbName, User: $user, Password: " . substr( $password, 0, 3 ) . "...\n"); - wfDebug($this->lastError()."\n"); - return false; - } - - $this->mOpened = true; - return $this->mConn; - } - - /** - * Closes a database connection, if it is open - * Returns success, true if already closed - */ - function close() { - $this->mOpened = false; - if ( $this->mConn ) { - return oci_close( $this->mConn ); - } else { - return true; - } - } - - function execFlags() { - return $this->mTrxLevel ? OCI_DEFAULT : OCI_COMMIT_ON_SUCCESS; - } - - function doQuery($sql) { - wfDebug("SQL: [$sql]\n"); - if (!mb_check_encoding($sql)) { - throw new MWException("SQL encoding is invalid"); - } - - if (($this->mLastResult = $stmt = oci_parse($this->mConn, $sql)) === false) { - $e = oci_error($this->mConn); - $this->reportQueryError($e['message'], $e['code'], $sql, __FUNCTION__); - } - - if (oci_execute($stmt, $this->execFlags()) == false) { - $e = oci_error($stmt); - $this->reportQueryError($e['message'], $e['code'], $sql, __FUNCTION__); - } - if (oci_statement_type($stmt) == "SELECT") - return new ORAResult($this, $stmt); - else { - $this->mAffectedRows = oci_num_rows($stmt); - return true; - } - } - - function queryIgnore($sql, $fname = '') { - return $this->query($sql, $fname, true); - } - - function freeResult($res) { - $res->free(); - } - - function fetchObject($res) { - return $res->fetchObject(); - } - - function fetchRow($res) { - return $res->fetchAssoc(); - } - - function numRows($res) { - return $res->numRows(); - } - - function numFields($res) { - return $res->numFields(); - } - - function fieldName($stmt, $n) { - return pg_field_name($stmt, $n); - } - - /** - * This must be called after nextSequenceVal - */ - function insertId() { - return $this->mInsertId; - } - - function dataSeek($res, $row) { - $res->seek($row); - } - - function lastError() { - if ($this->mConn === false) - $e = oci_error(); - else - $e = oci_error($this->mConn); - return $e['message']; - } - - function lastErrno() { - if ($this->mConn === false) - $e = oci_error(); - else - $e = oci_error($this->mConn); - return $e['code']; - } - - function affectedRows() { - return $this->mAffectedRows; - } - - /** - * Returns information about an index - * If errors are explicitly ignored, returns NULL on failure - */ - function indexInfo( $table, $index, $fname = 'Database::indexExists' ) { - return false; - } - - function indexUnique ($table, $index, $fname = 'Database::indexUnique' ) { - return false; - } - - function insert( $table, $a, $fname = 'Database::insert', $options = array() ) { - if (!is_array($options)) - $options = array($options); - - #if (in_array('IGNORE', $options)) - # $oldIgnore = $this->ignoreErrors(true); - - # IGNORE is performed using single-row inserts, ignoring errors in each - # FIXME: need some way to distiguish between key collision and other types of error - //$oldIgnore = $this->ignoreErrors(true); - if (!is_array(reset($a))) { - $a = array($a); - } - foreach ($a as $row) { - $this->insertOneRow($table, $row, $fname); - } - //$this->ignoreErrors($oldIgnore); - $retVal = true; - - //if (in_array('IGNORE', $options)) - // $this->ignoreErrors($oldIgnore); - - return $retVal; - } - - function insertOneRow($table, $row, $fname) { - // "INSERT INTO tables (a, b, c)" - $sql = "INSERT INTO " . $this->tableName($table) . " (" . join(',', array_keys($row)) . ')'; - $sql .= " VALUES ("; - - // for each value, append ":key" - $first = true; - $returning = ''; - foreach ($row as $col => $val) { - if (is_object($val)) { - $what = "EMPTY_BLOB()"; - assert($returning === ''); - $returning = " RETURNING $col INTO :bval"; - $blobcol = $col; - } else - $what = ":$col"; - - if ($first) - $sql .= "$what"; - else - $sql.= ", $what"; - $first = false; - } - $sql .= ") $returning"; - - $stmt = oci_parse($this->mConn, $sql); - foreach ($row as $col => $val) { - if (!is_object($val)) { - if (oci_bind_by_name($stmt, ":$col", $row[$col]) === false) - $this->reportQueryError($this->lastErrno(), $this->lastError(), $sql, __METHOD__); - } - } - - if (($bval = oci_new_descriptor($this->mConn, OCI_D_LOB)) === false) { - $e = oci_error($stmt); - throw new DBUnexpectedError($this, "Cannot create LOB descriptor: " . $e['message']); - } - - if (strlen($returning)) - oci_bind_by_name($stmt, ":bval", $bval, -1, SQLT_BLOB); - - if (oci_execute($stmt, OCI_DEFAULT) === false) { - $e = oci_error($stmt); - $this->reportQueryError($e['message'], $e['code'], $sql, __METHOD__); - } - if (strlen($returning)) { - $bval->save($row[$blobcol]->getData()); - $bval->free(); - } - if (!$this->mTrxLevel) - oci_commit($this->mConn); - - oci_free_statement($stmt); - } - - function tableName( $name ) { - # Replace reserved words with better ones - switch( $name ) { - case 'user': - return 'mwuser'; - case 'text': - return 'pagecontent'; - default: - return $name; - } - } - - /** - * Return the next in a sequence, save the value for retrieval via insertId() - */ - function nextSequenceValue($seqName) { - $res = $this->query("SELECT $seqName.nextval FROM dual"); - $row = $this->fetchRow($res); - $this->mInsertId = $row[0]; - $this->freeResult($res); - return $this->mInsertId; - } - - /** - * Oracle does not have a "USE INDEX" clause, so return an empty string - */ - function useIndexClause($index) { - return ''; - } - - # REPLACE query wrapper - # Oracle simulates this with a DELETE followed by INSERT - # $row is the row to insert, an associative array - # $uniqueIndexes is an array of indexes. Each element may be either a - # field name or an array of field names - # - # It may be more efficient to leave off unique indexes which are unlikely to collide. - # However if you do this, you run the risk of encountering errors which wouldn't have - # occurred in MySQL - function replace( $table, $uniqueIndexes, $rows, $fname = 'Database::replace' ) { - $table = $this->tableName($table); - - if (count($rows)==0) { - return; - } - - # Single row case - if (!is_array(reset($rows))) { - $rows = array($rows); - } - - foreach( $rows as $row ) { - # Delete rows which collide - if ( $uniqueIndexes ) { - $sql = "DELETE FROM $table WHERE "; - $first = true; - foreach ( $uniqueIndexes as $index ) { - if ( $first ) { - $first = false; - $sql .= "("; - } else { - $sql .= ') OR ('; - } - if ( is_array( $index ) ) { - $first2 = true; - foreach ( $index as $col ) { - if ( $first2 ) { - $first2 = false; - } else { - $sql .= ' AND '; - } - $sql .= $col.'=' . $this->addQuotes( $row[$col] ); - } - } else { - $sql .= $index.'=' . $this->addQuotes( $row[$index] ); - } - } - $sql .= ')'; - $this->query( $sql, $fname ); - } - - # Now insert the row - $sql = "INSERT INTO $table (" . $this->makeList( array_keys( $row ), LIST_NAMES ) .') VALUES (' . - $this->makeList( $row, LIST_COMMA ) . ')'; - $this->query($sql, $fname); - } - } - - # DELETE where the condition is a join - function deleteJoin( $delTable, $joinTable, $delVar, $joinVar, $conds, $fname = "Database::deleteJoin" ) { - if ( !$conds ) { - throw new DBUnexpectedError($this, 'Database::deleteJoin() called with empty $conds' ); - } - - $delTable = $this->tableName( $delTable ); - $joinTable = $this->tableName( $joinTable ); - $sql = "DELETE FROM $delTable WHERE $delVar IN (SELECT $joinVar FROM $joinTable "; - if ( $conds != '*' ) { - $sql .= 'WHERE ' . $this->makeList( $conds, LIST_AND ); - } - $sql .= ')'; - - $this->query( $sql, $fname ); - } - - # Returns the size of a text field, or -1 for "unlimited" - function textFieldSize( $table, $field ) { - $table = $this->tableName( $table ); - $sql = "SELECT t.typname as ftype,a.atttypmod as size - FROM pg_class c, pg_attribute a, pg_type t - WHERE relname='$table' AND a.attrelid=c.oid AND - a.atttypid=t.oid and a.attname='$field'"; - $res =$this->query($sql); - $row=$this->fetchObject($res); - if ($row->ftype=="varchar") { - $size=$row->size-4; - } else { - $size=$row->size; - } - $this->freeResult( $res ); - return $size; - } - - function lowPriorityOption() { - return ''; - } - - function limitResult($sql, $limit, $offset) { - if ($offset === false) - $offset = 0; - return "SELECT * FROM ($sql) WHERE rownum >= (1 + $offset) AND rownum < 1 + $limit + $offset"; - } - - /** - * Returns an SQL expression for a simple conditional. - * Uses CASE on Oracle - * - * @param string $cond SQL expression which will result in a boolean value - * @param string $trueVal SQL expression to return if true - * @param string $falseVal SQL expression to return if false - * @return string SQL fragment - */ - function conditional( $cond, $trueVal, $falseVal ) { - return " (CASE WHEN $cond THEN $trueVal ELSE $falseVal END) "; - } - - function wasDeadlock() { - return $this->lastErrno() == 'OCI-00060'; - } - - function timestamp($ts = 0) { - return wfTimestamp(TS_ORACLE, $ts); - } - - /** - * Return aggregated value function call - */ - function aggregateValue ($valuedata,$valuename='value') { - return $valuedata; - } - - function reportQueryError($error, $errno, $sql, $fname, $tempIgnore = false) { - # Ignore errors during error handling to avoid infinite - # recursion - $ignore = $this->ignoreErrors(true); - ++$this->mErrorCount; - - if ($ignore || $tempIgnore) { -echo "error ignored! query = [$sql]\n"; - wfDebug("SQL ERROR (ignored): $error\n"); - $this->ignoreErrors( $ignore ); - } - else { -echo "error!\n"; - $message = "A database error has occurred\n" . - "Query: $sql\n" . - "Function: $fname\n" . - "Error: $errno $error\n"; - throw new DBUnexpectedError($this, $message); - } - } - - /** - * @return string wikitext of a link to the server software's web site - */ - function getSoftwareLink() { - return "[http://www.oracle.com/ Oracle]"; - } - - /** - * @return string Version information from the database - */ - function getServerVersion() { - return oci_server_version($this->mConn); - } - - /** - * Query whether a given table exists (in the given schema, or the default mw one if not given) - */ - function tableExists($table) { - $etable= $this->addQuotes($table); - $SQL = "SELECT 1 FROM user_tables WHERE table_name='$etable'"; - $res = $this->query($SQL); - $count = $res ? oci_num_rows($res) : 0; - if ($res) - $this->freeResult($res); - return $count; - } - - /** - * Query whether a given column exists in the mediawiki schema - */ - function fieldExists( $table, $field ) { - return true; // XXX - } - - function fieldInfo( $table, $field ) { - return false; // XXX - } - - function begin( $fname = '' ) { - $this->mTrxLevel = 1; - } - function immediateCommit( $fname = '' ) { - return true; - } - function commit( $fname = '' ) { - oci_commit($this->mConn); - $this->mTrxLevel = 0; - } - - /* Not even sure why this is used in the main codebase... */ - function limitResultForUpdate($sql, $num) { - return $sql; - } - - function strencode($s) { - return str_replace("'", "''", $s); - } - - function encodeBlob($b) { - return new ORABlob($b); - } - function decodeBlob($b) { - return $b; //return $b->load(); - } - - function addQuotes( $s ) { - global $wgLang; - $s = $wgLang->checkTitleEncoding($s); - return "'" . $this->strencode($s) . "'"; - } - - function quote_ident( $s ) { - return $s; - } - - /* For now, does nothing */ - function selectDB( $db ) { - return true; - } - - /** - * Returns an optional USE INDEX clause to go after the table, and a - * string to go at the end of the query - * - * @private - * - * @param array $options an associative array of options to be turned into - * an SQL query, valid keys are listed in the function. - * @return array - */ - function makeSelectOptions( $options ) { - $preLimitTail = $postLimitTail = ''; - $startOpts = ''; - - $noKeyOptions = array(); - foreach ( $options as $key => $option ) { - if ( is_numeric( $key ) ) { - $noKeyOptions[$option] = true; - } - } - - if ( isset( $options['GROUP BY'] ) ) $preLimitTail .= " GROUP BY {$options['GROUP BY']}"; - if ( isset( $options['ORDER BY'] ) ) $preLimitTail .= " ORDER BY {$options['ORDER BY']}"; - - if (isset($options['LIMIT'])) { - // $tailOpts .= $this->limitResult('', $options['LIMIT'], - // isset($options['OFFSET']) ? $options['OFFSET'] - // : false); - } - - #if ( isset( $noKeyOptions['FOR UPDATE'] ) ) $tailOpts .= ' FOR UPDATE'; - #if ( isset( $noKeyOptions['LOCK IN SHARE MODE'] ) ) $tailOpts .= ' LOCK IN SHARE MODE'; - if ( isset( $noKeyOptions['DISTINCT'] ) || isset( $noKeyOptions['DISTINCTROW'] ) ) $startOpts .= 'DISTINCT'; - - if ( isset( $options['USE INDEX'] ) && ! is_array( $options['USE INDEX'] ) ) { - $useIndex = $this->useIndexClause( $options['USE INDEX'] ); - } else { - $useIndex = ''; - } - - return array( $startOpts, $useIndex, $preLimitTail, $postLimitTail ); - } - - public function setTimeout( $timeout ) { - // @todo fixme no-op - } - - function ping() { - wfDebug( "Function ping() not written for DatabaseOracle.php yet"); - return true; - } - - /** - * How lagged is this slave? - * - * @return int - */ - public function getLag() { - # Not implemented for Oracle - return 0; - } - -} // end DatabaseOracle class - - diff --git a/includes/DatabasePostgres.php b/includes/DatabasePostgres.php deleted file mode 100644 index 01213715..00000000 --- a/includes/DatabasePostgres.php +++ /dev/null @@ -1,1313 +0,0 @@ -query(sprintf($q, - $db->addQuotes($wgDBmwschema), - $db->addQuotes($table), - $db->addQuotes($field))); - $row = $db->fetchObject($res); - if (!$row) - return null; - $n = new PostgresField; - $n->type = $row->typname; - $n->nullable = ($row->attnotnull == 'f'); - $n->name = $field; - $n->tablename = $table; - $n->max_length = $row->attlen; - return $n; - } - - function name() { - return $this->name; - } - - function tableName() { - return $this->tablename; - } - - function type() { - return $this->type; - } - - function nullable() { - return $this->nullable; - } - - function maxLength() { - return $this->max_length; - } -} - -/** - * @addtogroup Database - */ -class DatabasePostgres extends Database { - var $mInsertId = NULL; - var $mLastResult = NULL; - var $numeric_version = NULL; - - function DatabasePostgres($server = false, $user = false, $password = false, $dbName = false, - $failFunction = false, $flags = 0 ) - { - - global $wgOut; - # Can't get a reference if it hasn't been set yet - if ( !isset( $wgOut ) ) { - $wgOut = NULL; - } - $this->mOut =& $wgOut; - $this->mFailFunction = $failFunction; - $this->mFlags = $flags; - $this->open( $server, $user, $password, $dbName); - - } - - function cascadingDeletes() { - return true; - } - function cleanupTriggers() { - return true; - } - function strictIPs() { - return true; - } - function realTimestamps() { - return true; - } - function implicitGroupby() { - return false; - } - function implicitOrderby() { - return false; - } - function searchableIPs() { - return true; - } - function functionalIndexes() { - return true; - } - - function hasConstraint( $name ) { - global $wgDBmwschema; - $SQL = "SELECT 1 FROM pg_catalog.pg_constraint c, pg_catalog.pg_namespace n WHERE c.connamespace = n.oid AND conname = '" . pg_escape_string( $name ) . "' AND n.nspname = '" . pg_escape_string($wgDBmwschema) ."'"; - return $this->numRows($res = $this->doQuery($SQL)); - } - - static function newFromParams( $server, $user, $password, $dbName, $failFunction = false, $flags = 0) - { - return new DatabasePostgres( $server, $user, $password, $dbName, $failFunction, $flags ); - } - - /** - * Usually aborts on failure - * If the failFunction is set to a non-zero integer, returns success - */ - function open( $server, $user, $password, $dbName ) { - # Test for Postgres support, to avoid suppressed fatal error - if ( !function_exists( 'pg_connect' ) ) { - throw new DBConnectionError( $this, "Postgres functions missing, have you compiled PHP with the --with-pgsql option?\n (Note: if you recently installed PHP, you may need to restart your webserver and database)\n" ); - } - - global $wgDBport; - - if (!strlen($user)) { ## e.g. the class is being loaded - return; - } - - $this->close(); - $this->mServer = $server; - $this->mPort = $port = $wgDBport; - $this->mUser = $user; - $this->mPassword = $password; - $this->mDBname = $dbName; - - $hstring=""; - if ($server!=false && $server!="") { - $hstring="host=$server "; - } - if ($port!=false && $port!="") { - $hstring .= "port=$port "; - } - - error_reporting( E_ALL ); - @$this->mConn = pg_connect("$hstring dbname=$dbName user=$user password=$password"); - - if ( $this->mConn == false ) { - wfDebug( "DB connection error\n" ); - wfDebug( "Server: $server, Database: $dbName, User: $user, Password: " . substr( $password, 0, 3 ) . "...\n" ); - wfDebug( $this->lastError()."\n" ); - return false; - } - - $this->mOpened = true; - - global $wgCommandLineMode; - ## If called from the command-line (e.g. importDump), only show errors - if ($wgCommandLineMode) { - $this->doQuery("SET client_min_messages = 'ERROR'"); - } - - global $wgDBmwschema, $wgDBts2schema; - if (isset( $wgDBmwschema ) && isset( $wgDBts2schema ) - && $wgDBmwschema !== 'mediawiki' - && preg_match( '/^\w+$/', $wgDBmwschema ) - && preg_match( '/^\w+$/', $wgDBts2schema ) - ) { - $safeschema = $this->quote_ident($wgDBmwschema); - $safeschema2 = $this->quote_ident($wgDBts2schema); - $this->doQuery("SET search_path = $safeschema, $wgDBts2schema, public"); - } - - return $this->mConn; - } - - - function initial_setup($password, $dbName) { - // If this is the initial connection, setup the schema stuff and possibly create the user - global $wgDBname, $wgDBuser, $wgDBpassword, $wgDBsuperuser, $wgDBmwschema, $wgDBts2schema; - - print "
  • Checking the version of Postgres..."; - $version = $this->getServerVersion(); - $PGMINVER = '8.1'; - if ($this->numeric_version < $PGMINVER) { - print "FAILED. Required version is $PGMINVER. You have $this->numeric_version ($version)
  • \n"; - dieout(""); - } - print "version $this->numeric_version is OK.\n"; - - $safeuser = $this->quote_ident($wgDBuser); - // Are we connecting as a superuser for the first time? - if ($wgDBsuperuser) { - // Are we really a superuser? Check out our rights - $SQL = "SELECT - CASE WHEN usesuper IS TRUE THEN - CASE WHEN usecreatedb IS TRUE THEN 3 ELSE 1 END - ELSE CASE WHEN usecreatedb IS TRUE THEN 2 ELSE 0 END - END AS rights - FROM pg_catalog.pg_user WHERE usename = " . $this->addQuotes($wgDBsuperuser); - $rows = $this->numRows($res = $this->doQuery($SQL)); - if (!$rows) { - print "
  • ERROR: Could not read permissions for user \"$wgDBsuperuser\"
  • \n"; - dieout(''); - } - $perms = pg_fetch_result($res, 0, 0); - - $SQL = "SELECT 1 FROM pg_catalog.pg_user WHERE usename = " . $this->addQuotes($wgDBuser); - $rows = $this->numRows($this->doQuery($SQL)); - if ($rows) { - print "
  • User \"$wgDBuser\" already exists, skipping account creation.
  • "; - } - else { - if ($perms != 1 and $perms != 3) { - print "
  • ERROR: the user \"$wgDBsuperuser\" cannot create other users. "; - print 'Please use a different Postgres user.
  • '; - dieout(''); - } - print "
  • Creating user $wgDBuser..."; - $safepass = $this->addQuotes($wgDBpassword); - $SQL = "CREATE USER $safeuser NOCREATEDB PASSWORD $safepass"; - $this->doQuery($SQL); - print "OK
  • \n"; - } - // User now exists, check out the database - if ($dbName != $wgDBname) { - $SQL = "SELECT 1 FROM pg_catalog.pg_database WHERE datname = " . $this->addQuotes($wgDBname); - $rows = $this->numRows($this->doQuery($SQL)); - if ($rows) { - print "
  • Database \"$wgDBname\" already exists, skipping database creation.
  • "; - } - else { - if ($perms < 2) { - print "
  • ERROR: the user \"$wgDBsuperuser\" cannot create databases. "; - print 'Please use a different Postgres user.
  • '; - dieout(''); - } - print "
  • Creating database $wgDBname..."; - $safename = $this->quote_ident($wgDBname); - $SQL = "CREATE DATABASE $safename OWNER $safeuser "; - $this->doQuery($SQL); - print "OK
  • \n"; - // Hopefully tsearch2 and plpgsql are in template1... - } - - // Reconnect to check out tsearch2 rights for this user - print "
  • Connecting to \"$wgDBname\" as superuser \"$wgDBsuperuser\" to check rights..."; - - $hstring=""; - if ($this->mServer!=false && $this->mServer!="") { - $hstring="host=$this->mServer "; - } - if ($this->mPort!=false && $this->mPort!="") { - $hstring .= "port=$this->mPort "; - } - - @$this->mConn = pg_connect("$hstring dbname=$wgDBname user=$wgDBsuperuser password=$password"); - if ( $this->mConn == false ) { - print "FAILED TO CONNECT!
  • "; - dieout(""); - } - print "OK\n"; - } - - if ($this->numeric_version < 8.3) { - // Tsearch2 checks - print "
  • Checking that tsearch2 is installed in the database \"$wgDBname\"..."; - if (! $this->tableExists("pg_ts_cfg", $wgDBts2schema)) { - print "FAILED. tsearch2 must be installed in the database \"$wgDBname\"."; - print "Please see this article"; - print " for instructions or ask on #postgresql on irc.freenode.net
  • \n"; - dieout(""); - } - print "OK\n"; - print "
  • Ensuring that user \"$wgDBuser\" has select rights on the tsearch2 tables..."; - foreach (array('cfg','cfgmap','dict','parser') as $table) { - $SQL = "GRANT SELECT ON pg_ts_$table TO $safeuser"; - $this->doQuery($SQL); - } - print "OK
  • \n"; - } - - // Setup the schema for this user if needed - $result = $this->schemaExists($wgDBmwschema); - $safeschema = $this->quote_ident($wgDBmwschema); - if (!$result) { - print "
  • Creating schema $wgDBmwschema ..."; - $result = $this->doQuery("CREATE SCHEMA $safeschema AUTHORIZATION $safeuser"); - if (!$result) { - print "FAILED.
  • \n"; - dieout(""); - } - print "OK\n"; - } - else { - print "
  • Schema already exists, explicitly granting rights...\n"; - $safeschema2 = $this->addQuotes($wgDBmwschema); - $SQL = "SELECT 'GRANT ALL ON '||pg_catalog.quote_ident(relname)||' TO $safeuser;'\n". - "FROM pg_catalog.pg_class p, pg_catalog.pg_namespace n\n". - "WHERE relnamespace = n.oid AND n.nspname = $safeschema2\n". - "AND p.relkind IN ('r','S','v')\n"; - $SQL .= "UNION\n"; - $SQL .= "SELECT 'GRANT ALL ON FUNCTION '||pg_catalog.quote_ident(proname)||'('||\n". - "pg_catalog.oidvectortypes(p.proargtypes)||') TO $safeuser;'\n". - "FROM pg_catalog.pg_proc p, pg_catalog.pg_namespace n\n". - "WHERE p.pronamespace = n.oid AND n.nspname = $safeschema2"; - $res = $this->doQuery($SQL); - if (!$res) { - print "FAILED. Could not set rights for the user.
  • \n"; - dieout(""); - } - $this->doQuery("SET search_path = $safeschema"); - $rows = $this->numRows($res); - while ($rows) { - $rows--; - $this->doQuery(pg_fetch_result($res, $rows, 0)); - } - print "OK"; - } - - // Install plpgsql if needed - $this->setup_plpgsql(); - - $wgDBsuperuser = ''; - return true; // Reconnect as regular user - - } // end superuser - - if (!defined('POSTGRES_SEARCHPATH')) { - - if ($this->numeric_version < 8.3) { - // Do we have the basic tsearch2 table? - print "
  • Checking for tsearch2 in the schema \"$wgDBts2schema\"..."; - if (! $this->tableExists("pg_ts_dict", $wgDBts2schema)) { - print "FAILED. Make sure tsearch2 is installed. See this article"; - print " for instructions.
  • \n"; - dieout(""); - } - print "OK\n"; - - // Does this user have the rights to the tsearch2 tables? - $ctype = pg_fetch_result($this->doQuery("SHOW lc_ctype"),0,0); - print "
  • Checking tsearch2 permissions..."; - // Let's check all four, just to be safe - error_reporting( 0 ); - $ts2tables = array('cfg','cfgmap','dict','parser'); - $safetsschema = $this->quote_ident($wgDBts2schema); - foreach ( $ts2tables AS $tname ) { - $SQL = "SELECT count(*) FROM $safetsschema.pg_ts_$tname"; - $res = $this->doQuery($SQL); - if (!$res) { - print "FAILED to access pg_ts_$tname. Make sure that the user ". - "\"$wgDBuser\" has SELECT access to all four tsearch2 tables
  • \n"; - dieout(""); - } - } - $SQL = "SELECT ts_name FROM $safetsschema.pg_ts_cfg WHERE locale = '$ctype'"; - $SQL .= " ORDER BY CASE WHEN ts_name <> 'default' THEN 1 ELSE 0 END"; - $res = $this->doQuery($SQL); - error_reporting( E_ALL ); - if (!$res) { - print "FAILED. Could not determine the tsearch2 locale information\n"; - dieout(""); - } - print "OK"; - - // Will the current locale work? Can we force it to? - print "
  • Verifying tsearch2 locale with $ctype..."; - $rows = $this->numRows($res); - $resetlocale = 0; - if (!$rows) { - print "not found
  • \n"; - print "
  • Attempting to set default tsearch2 locale to \"$ctype\"..."; - $resetlocale = 1; - } - else { - $tsname = pg_fetch_result($res, 0, 0); - if ($tsname != 'default') { - print "not set to default ($tsname)"; - print "
  • Attempting to change tsearch2 default locale to \"$ctype\"..."; - $resetlocale = 1; - } - } - if ($resetlocale) { - $SQL = "UPDATE $safetsschema.pg_ts_cfg SET locale = '$ctype' WHERE ts_name = 'default'"; - $res = $this->doQuery($SQL); - if (!$res) { - print "FAILED. "; - print "Please make sure that the locale in pg_ts_cfg for \"default\" is set to \"$ctype\"
  • \n"; - dieout(""); - } - print "OK"; - } - - // Final test: try out a simple tsearch2 query - $SQL = "SELECT $safetsschema.to_tsvector('default','MediaWiki tsearch2 testing')"; - $res = $this->doQuery($SQL); - if (!$res) { - print "FAILED. Specifically, \"$SQL\" did not work."; - dieout(""); - } - print "OK"; - } - - // Install plpgsql if needed - $this->setup_plpgsql(); - - // Does the schema already exist? Who owns it? - $result = $this->schemaExists($wgDBmwschema); - if (!$result) { - print "
  • Creating schema $wgDBmwschema ..."; - error_reporting( 0 ); - $safeschema = $this->quote_ident($wgDBmwschema); - $result = $this->doQuery("CREATE SCHEMA $safeschema"); - error_reporting( E_ALL ); - if (!$result) { - print "FAILED. The user \"$wgDBuser\" must be able to access the schema. ". - "You can try making them the owner of the database, or try creating the schema with a ". - "different user, and then grant access to the \"$wgDBuser\" user.
  • \n"; - dieout(""); - } - print "OK\n"; - } - else if ($result != $wgDBuser) { - print "
  • Schema \"$wgDBmwschema\" exists but is not owned by \"$wgDBuser\". Not ideal.
  • \n"; - } - else { - print "
  • Schema \"$wgDBmwschema\" exists and is owned by \"$wgDBuser\". Excellent.
  • \n"; - } - - // Always return GMT time to accomodate the existing integer-based timestamp assumption - print "
  • Setting the timezone to GMT for user \"$wgDBuser\" ..."; - $SQL = "ALTER USER $safeuser SET timezone = 'GMT'"; - $result = pg_query($this->mConn, $SQL); - if (!$result) { - print "FAILED.
  • \n"; - dieout(""); - } - print "OK\n"; - // Set for the rest of this session - $SQL = "SET timezone = 'GMT'"; - $result = pg_query($this->mConn, $SQL); - if (!$result) { - print "
  • Failed to set timezone
  • \n"; - dieout(""); - } - - print "
  • Setting the datestyle to ISO, YMD for user \"$wgDBuser\" ..."; - $SQL = "ALTER USER $safeuser SET datestyle = 'ISO, YMD'"; - $result = pg_query($this->mConn, $SQL); - if (!$result) { - print "FAILED.
  • \n"; - dieout(""); - } - print "OK\n"; - // Set for the rest of this session - $SQL = "SET datestyle = 'ISO, YMD'"; - $result = pg_query($this->mConn, $SQL); - if (!$result) { - print "
  • Failed to set datestyle
  • \n"; - dieout(""); - } - - // Fix up the search paths if needed - print "
  • Setting the search path for user \"$wgDBuser\" ..."; - $path = $this->quote_ident($wgDBmwschema); - if ($wgDBts2schema !== $wgDBmwschema) - $path .= ", ". $this->quote_ident($wgDBts2schema); - if ($wgDBmwschema !== 'public' and $wgDBts2schema !== 'public') - $path .= ", public"; - $SQL = "ALTER USER $safeuser SET search_path = $path"; - $result = pg_query($this->mConn, $SQL); - if (!$result) { - print "FAILED.
  • \n"; - dieout(""); - } - print "OK\n"; - // Set for the rest of this session - $SQL = "SET search_path = $path"; - $result = pg_query($this->mConn, $SQL); - if (!$result) { - print "
  • Failed to set search_path
  • \n"; - dieout(""); - } - define( "POSTGRES_SEARCHPATH", $path ); - } - } - - - function setup_plpgsql() { - print "
  • Checking for Pl/Pgsql ..."; - $SQL = "SELECT 1 FROM pg_catalog.pg_language WHERE lanname = 'plpgsql'"; - $rows = $this->numRows($this->doQuery($SQL)); - if ($rows < 1) { - // plpgsql is not installed, but if we have a pg_pltemplate table, we should be able to create it - print "not installed. Attempting to install Pl/Pgsql ..."; - $SQL = "SELECT 1 FROM pg_catalog.pg_class c JOIN pg_catalog.pg_namespace n ON (n.oid = c.relnamespace) ". - "WHERE relname = 'pg_pltemplate' AND nspname='pg_catalog'"; - $rows = $this->numRows($this->doQuery($SQL)); - if ($rows >= 1) { - $olde = error_reporting(0); - error_reporting($olde - E_WARNING); - $result = $this->doQuery("CREATE LANGUAGE plpgsql"); - error_reporting($olde); - if (!$result) { - print "FAILED. You need to install the language plpgsql in the database $wgDBname
  • "; - dieout(""); - } - } - else { - print "FAILED. You need to install the language plpgsql in the database $wgDBname"; - dieout(""); - } - } - print "OK\n"; - } - - - /** - * Closes a database connection, if it is open - * Returns success, true if already closed - */ - function close() { - $this->mOpened = false; - if ( $this->mConn ) { - return pg_close( $this->mConn ); - } else { - return true; - } - } - - function doQuery( $sql ) { - if (function_exists('mb_convert_encoding')) { - return $this->mLastResult=pg_query( $this->mConn , mb_convert_encoding($sql,'UTF-8') ); - } - return $this->mLastResult=pg_query( $this->mConn , $sql); - } - - function queryIgnore( $sql, $fname = '' ) { - return $this->query( $sql, $fname, true ); - } - - function freeResult( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - if ( !@pg_free_result( $res ) ) { - throw new DBUnexpectedError($this, "Unable to free Postgres result\n" ); - } - } - - function fetchObject( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - @$row = pg_fetch_object( $res ); - # FIXME: HACK HACK HACK HACK debug - - # TODO: - # hashar : not sure if the following test really trigger if the object - # fetching failed. - if( pg_last_error($this->mConn) ) { - throw new DBUnexpectedError($this, 'SQL error: ' . htmlspecialchars( pg_last_error($this->mConn) ) ); - } - return $row; - } - - function fetchRow( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - @$row = pg_fetch_array( $res ); - if( pg_last_error($this->mConn) ) { - throw new DBUnexpectedError($this, 'SQL error: ' . htmlspecialchars( pg_last_error($this->mConn) ) ); - } - return $row; - } - - function numRows( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - @$n = pg_num_rows( $res ); - if( pg_last_error($this->mConn) ) { - throw new DBUnexpectedError($this, 'SQL error: ' . htmlspecialchars( pg_last_error($this->mConn) ) ); - } - return $n; - } - function numFields( $res ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - return pg_num_fields( $res ); - } - function fieldName( $res, $n ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - return pg_field_name( $res, $n ); - } - - /** - * This must be called after nextSequenceVal - */ - function insertId() { - return $this->mInsertId; - } - - function dataSeek( $res, $row ) { - if ( $res instanceof ResultWrapper ) { - $res = $res->result; - } - return pg_result_seek( $res, $row ); - } - - function lastError() { - if ( $this->mConn ) { - return pg_last_error(); - } - else { - return "No database connection"; - } - } - function lastErrno() { - return pg_last_error() ? 1 : 0; - } - - function affectedRows() { - if( !isset( $this->mLastResult ) or ! $this->mLastResult ) - return 0; - - return pg_affected_rows( $this->mLastResult ); - } - - /** - * Estimate rows in dataset - * Returns estimated count, based on EXPLAIN output - * This is not necessarily an accurate estimate, so use sparingly - * Returns -1 if count cannot be found - * Takes same arguments as Database::select() - */ - - function estimateRowCount( $table, $vars='*', $conds='', $fname = 'Database::estimateRowCount', $options = array() ) { - $options['EXPLAIN'] = true; - $res = $this->select( $table, $vars, $conds, $fname, $options ); - $rows = -1; - if ( $res ) { - $row = $this->fetchRow( $res ); - $count = array(); - if( preg_match( '/rows=(\d+)/', $row[0], $count ) ) { - $rows = $count[1]; - } - $this->freeResult($res); - } - return $rows; - } - - - /** - * Returns information about an index - * If errors are explicitly ignored, returns NULL on failure - */ - function indexInfo( $table, $index, $fname = 'Database::indexExists' ) { - $sql = "SELECT indexname FROM pg_indexes WHERE tablename='$table'"; - $res = $this->query( $sql, $fname ); - if ( !$res ) { - return NULL; - } - while ( $row = $this->fetchObject( $res ) ) { - if ( $row->indexname == $index ) { - return $row; - } - } - return false; - } - - function indexUnique ($table, $index, $fname = 'Database::indexUnique' ) { - $sql = "SELECT indexname FROM pg_indexes WHERE tablename='{$table}'". - " AND indexdef LIKE 'CREATE UNIQUE%({$index})'"; - $res = $this->query( $sql, $fname ); - if ( !$res ) - return NULL; - while ($row = $this->fetchObject( $res )) - return true; - return false; - - } - - /** - * INSERT wrapper, inserts an array into a table - * - * $args may be a single associative array, or an array of these with numeric keys, - * for multi-row insert (Postgres version 8.2 and above only). - * - * @param array $table String: Name of the table to insert to. - * @param array $args Array: Items to insert into the table. - * @param array $fname String: Name of the function, for profiling - * @param mixed $options String or Array. Valid options: IGNORE - * - * @return bool Success of insert operation. IGNORE always returns true. - */ - function insert( $table, $args, $fname = 'DatabasePostgres::insert', $options = array() ) { - global $wgDBversion; - - $table = $this->tableName( $table ); - if (! isset( $wgDBversion ) ) { - $this->getServerVersion(); - $wgDBversion = $this->numeric_version; - } - - if ( !is_array( $options ) ) - $options = array( $options ); - - if ( isset( $args[0] ) && is_array( $args[0] ) ) { - $multi = true; - $keys = array_keys( $args[0] ); - } - else { - $multi = false; - $keys = array_keys( $args ); - } - - $ignore = in_array( 'IGNORE', $options ) ? 1 : 0; - if ( $ignore ) - $olde = error_reporting( 0 ); - - $sql = "INSERT INTO $table (" . implode( ',', $keys ) . ') VALUES '; - - if ( $multi ) { - if ( $wgDBversion >= 8.2 ) { - $first = true; - foreach ( $args as $row ) { - if ( $first ) { - $first = false; - } else { - $sql .= ','; - } - $sql .= '(' . $this->makeList( $row ) . ')'; - } - $res = (bool)$this->query( $sql, $fname, $ignore ); - } - else { - $res = true; - $origsql = $sql; - foreach ( $args as $row ) { - $tempsql = $origsql; - $tempsql .= '(' . $this->makeList( $row ) . ')'; - $tempres = (bool)$this->query( $tempsql, $fname, $ignore ); - if (! $tempres) - $res = false; - } - } - } - else { - $sql .= '(' . $this->makeList( $args ) . ')'; - $res = (bool)$this->query( $sql, $fname, $ignore ); - } - - if ( $ignore ) { - $olde = error_reporting( $olde ); - return true; - } - - return $res; - - } - - function tableName( $name ) { - # Replace reserved words with better ones - switch( $name ) { - case 'user': - return 'mwuser'; - case 'text': - return 'pagecontent'; - default: - return $name; - } - } - - /** - * Return the next in a sequence, save the value for retrieval via insertId() - */ - function nextSequenceValue( $seqName ) { - $safeseq = preg_replace( "/'/", "''", $seqName ); - $res = $this->query( "SELECT nextval('$safeseq')" ); - $row = $this->fetchRow( $res ); - $this->mInsertId = $row[0]; - $this->freeResult( $res ); - return $this->mInsertId; - } - - /** - * Return the current value of a sequence. Assumes it has ben nextval'ed in this session. - */ - function currentSequenceValue( $seqName ) { - $safeseq = preg_replace( "/'/", "''", $seqName ); - $res = $this->query( "SELECT currval('$safeseq')" ); - $row = $this->fetchRow( $res ); - $currval = $row[0]; - $this->freeResult( $res ); - return $currval; - } - - /** - * Postgres does not have a "USE INDEX" clause, so return an empty string - */ - function useIndexClause( $index ) { - return ''; - } - - # REPLACE query wrapper - # Postgres simulates this with a DELETE followed by INSERT - # $row is the row to insert, an associative array - # $uniqueIndexes is an array of indexes. Each element may be either a - # field name or an array of field names - # - # It may be more efficient to leave off unique indexes which are unlikely to collide. - # However if you do this, you run the risk of encountering errors which wouldn't have - # occurred in MySQL - function replace( $table, $uniqueIndexes, $rows, $fname = 'Database::replace' ) { - $table = $this->tableName( $table ); - - if (count($rows)==0) { - return; - } - - # Single row case - if ( !is_array( reset( $rows ) ) ) { - $rows = array( $rows ); - } - - foreach( $rows as $row ) { - # Delete rows which collide - if ( $uniqueIndexes ) { - $sql = "DELETE FROM $table WHERE "; - $first = true; - foreach ( $uniqueIndexes as $index ) { - if ( $first ) { - $first = false; - $sql .= "("; - } else { - $sql .= ') OR ('; - } - if ( is_array( $index ) ) { - $first2 = true; - foreach ( $index as $col ) { - if ( $first2 ) { - $first2 = false; - } else { - $sql .= ' AND '; - } - $sql .= $col.'=' . $this->addQuotes( $row[$col] ); - } - } else { - $sql .= $index.'=' . $this->addQuotes( $row[$index] ); - } - } - $sql .= ')'; - $this->query( $sql, $fname ); - } - - # Now insert the row - $sql = "INSERT INTO $table (" . $this->makeList( array_keys( $row ), LIST_NAMES ) .') VALUES (' . - $this->makeList( $row, LIST_COMMA ) . ')'; - $this->query( $sql, $fname ); - } - } - - # DELETE where the condition is a join - function deleteJoin( $delTable, $joinTable, $delVar, $joinVar, $conds, $fname = "Database::deleteJoin" ) { - if ( !$conds ) { - throw new DBUnexpectedError($this, 'Database::deleteJoin() called with empty $conds' ); - } - - $delTable = $this->tableName( $delTable ); - $joinTable = $this->tableName( $joinTable ); - $sql = "DELETE FROM $delTable WHERE $delVar IN (SELECT $joinVar FROM $joinTable "; - if ( $conds != '*' ) { - $sql .= 'WHERE ' . $this->makeList( $conds, LIST_AND ); - } - $sql .= ')'; - - $this->query( $sql, $fname ); - } - - # Returns the size of a text field, or -1 for "unlimited" - function textFieldSize( $table, $field ) { - $table = $this->tableName( $table ); - $sql = "SELECT t.typname as ftype,a.atttypmod as size - FROM pg_class c, pg_attribute a, pg_type t - WHERE relname='$table' AND a.attrelid=c.oid AND - a.atttypid=t.oid and a.attname='$field'"; - $res =$this->query($sql); - $row=$this->fetchObject($res); - if ($row->ftype=="varchar") { - $size=$row->size-4; - } else { - $size=$row->size; - } - $this->freeResult( $res ); - return $size; - } - - function lowPriorityOption() { - return ''; - } - - function limitResult($sql, $limit,$offset=false) { - return "$sql LIMIT $limit ".(is_numeric($offset)?" OFFSET {$offset} ":""); - } - - /** - * Returns an SQL expression for a simple conditional. - * Uses CASE on Postgres - * - * @param string $cond SQL expression which will result in a boolean value - * @param string $trueVal SQL expression to return if true - * @param string $falseVal SQL expression to return if false - * @return string SQL fragment - */ - function conditional( $cond, $trueVal, $falseVal ) { - return " (CASE WHEN $cond THEN $trueVal ELSE $falseVal END) "; - } - - function wasDeadlock() { - return $this->lastErrno() == '40P01'; - } - - function timestamp( $ts=0 ) { - return wfTimestamp(TS_POSTGRES,$ts); - } - - /** - * Return aggregated value function call - */ - function aggregateValue ($valuedata,$valuename='value') { - return $valuedata; - } - - - function reportQueryError( $error, $errno, $sql, $fname, $tempIgnore = false ) { - // Ignore errors during error handling to avoid infinite recursion - $ignore = $this->ignoreErrors( true ); - $this->mErrorCount++; - - if ($ignore || $tempIgnore) { - wfDebug("SQL ERROR (ignored): $error\n"); - $this->ignoreErrors( $ignore ); - } - else { - $message = "A database error has occurred\n" . - "Query: $sql\n" . - "Function: $fname\n" . - "Error: $errno $error\n"; - throw new DBUnexpectedError($this, $message); - } - } - - /** - * @return string wikitext of a link to the server software's web site - */ - function getSoftwareLink() { - return "[http://www.postgresql.org/ PostgreSQL]"; - } - - /** - * @return string Version information from the database - */ - function getServerVersion() { - $version = pg_fetch_result($this->doQuery("SELECT version()"),0,0); - $thisver = array(); - if (!preg_match('/PostgreSQL (\d+\.\d+)(\S+)/', $version, $thisver)) { - die("Could not determine the numeric version from $version!"); - } - $this->numeric_version = $thisver[1]; - return $version; - } - - - /** - * Query whether a given relation exists (in the given schema, or the - * default mw one if not given) - */ - function relationExists( $table, $types, $schema = false ) { - global $wgDBmwschema; - if (!is_array($types)) - $types = array($types); - if (! $schema ) - $schema = $wgDBmwschema; - $etable = $this->addQuotes($table); - $eschema = $this->addQuotes($schema); - $SQL = "SELECT 1 FROM pg_catalog.pg_class c, pg_catalog.pg_namespace n " - . "WHERE c.relnamespace = n.oid AND c.relname = $etable AND n.nspname = $eschema " - . "AND c.relkind IN ('" . implode("','", $types) . "')"; - $res = $this->query( $SQL ); - $count = $res ? $res->numRows() : 0; - if ($res) - $this->freeResult( $res ); - return $count ? true : false; - } - - /* - * For backward compatibility, this function checks both tables and - * views. - */ - function tableExists ($table, $schema = false) { - return $this->relationExists($table, array('r', 'v'), $schema); - } - - function sequenceExists ($sequence, $schema = false) { - return $this->relationExists($sequence, 'S', $schema); - } - - function triggerExists($table, $trigger) { - global $wgDBmwschema; - - $q = <<query(sprintf($q, - $this->addQuotes($wgDBmwschema), - $this->addQuotes($table), - $this->addQuotes($trigger))); - if (!$res) - return NULL; - $rows = $res->numRows(); - $this->freeResult($res); - return $rows; - } - - function ruleExists($table, $rule) { - global $wgDBmwschema; - $exists = $this->selectField("pg_rules", "rulename", - array( "rulename" => $rule, - "tablename" => $table, - "schemaname" => $wgDBmwschema)); - return $exists === $rule; - } - - function constraintExists($table, $constraint) { - global $wgDBmwschema; - $SQL = sprintf("SELECT 1 FROM information_schema.table_constraints ". - "WHERE constraint_schema = %s AND table_name = %s AND constraint_name = %s", - $this->addQuotes($wgDBmwschema), - $this->addQuotes($table), - $this->addQuotes($constraint)); - $res = $this->query($SQL); - if (!$res) - return NULL; - $rows = $res->numRows(); - $this->freeResult($res); - return $rows; - } - - /** - * Query whether a given schema exists. Returns the name of the owner - */ - function schemaExists( $schema ) { - $eschema = preg_replace("/'/", "''", $schema); - $SQL = "SELECT rolname FROM pg_catalog.pg_namespace n, pg_catalog.pg_roles r " - ."WHERE n.nspowner=r.oid AND n.nspname = '$eschema'"; - $res = $this->query( $SQL ); - if ( $res && $res->numRows() ) { - $row = $res->fetchObject(); - $owner = $row->rolname; - } else { - $owner = false; - } - if ($res) - $this->freeResult($res); - return $owner; - } - - /** - * Query whether a given column exists in the mediawiki schema - */ - function fieldExists( $table, $field, $fname = 'DatabasePostgres::fieldExists' ) { - global $wgDBmwschema; - $etable = preg_replace("/'/", "''", $table); - $eschema = preg_replace("/'/", "''", $wgDBmwschema); - $ecol = preg_replace("/'/", "''", $field); - $SQL = "SELECT 1 FROM pg_catalog.pg_class c, pg_catalog.pg_namespace n, pg_catalog.pg_attribute a " - . "WHERE c.relnamespace = n.oid AND c.relname = '$etable' AND n.nspname = '$eschema' " - . "AND a.attrelid = c.oid AND a.attname = '$ecol'"; - $res = $this->query( $SQL, $fname ); - $count = $res ? $res->numRows() : 0; - if ($res) - $this->freeResult( $res ); - return $count; - } - - function fieldInfo( $table, $field ) { - return PostgresField::fromText($this, $table, $field); - } - - function begin( $fname = 'DatabasePostgres::begin' ) { - $this->query( 'BEGIN', $fname ); - $this->mTrxLevel = 1; - } - function immediateCommit( $fname = 'DatabasePostgres::immediateCommit' ) { - return true; - } - function commit( $fname = 'DatabasePostgres::commit' ) { - $this->query( 'COMMIT', $fname ); - $this->mTrxLevel = 0; - } - - /* Not even sure why this is used in the main codebase... */ - function limitResultForUpdate($sql, $num) { - return $sql; - } - - function setup_database() { - global $wgVersion, $wgDBmwschema, $wgDBts2schema, $wgDBport, $wgDBuser; - - // Make sure that we can write to the correct schema - // If not, Postgres will happily and silently go to the next search_path item - $ctest = "mediawiki_test_table"; - $safeschema = $this->quote_ident($wgDBmwschema); - if ($this->tableExists($ctest, $wgDBmwschema)) { - $this->doQuery("DROP TABLE $safeschema.$ctest"); - } - $SQL = "CREATE TABLE $safeschema.$ctest(a int)"; - $olde = error_reporting( 0 ); - $res = $this->doQuery($SQL); - error_reporting( $olde ); - if (!$res) { - print "FAILED. Make sure that the user \"$wgDBuser\" can write to the schema \"$wgDBmwschema\"\n"; - dieout(""); - } - $this->doQuery("DROP TABLE $safeschema.$ctest"); - - $res = dbsource( "../maintenance/postgres/tables.sql", $this); - - ## Update version information - $mwv = $this->addQuotes($wgVersion); - $pgv = $this->addQuotes($this->getServerVersion()); - $pgu = $this->addQuotes($this->mUser); - $mws = $this->addQuotes($wgDBmwschema); - $tss = $this->addQuotes($wgDBts2schema); - $pgp = $this->addQuotes($wgDBport); - $dbn = $this->addQuotes($this->mDBname); - $ctype = pg_fetch_result($this->doQuery("SHOW lc_ctype"),0,0); - - $SQL = "UPDATE mediawiki_version SET mw_version=$mwv, pg_version=$pgv, pg_user=$pgu, ". - "mw_schema = $mws, ts2_schema = $tss, pg_port=$pgp, pg_dbname=$dbn, ". - "ctype = '$ctype' ". - "WHERE type = 'Creation'"; - $this->query($SQL); - - ## Avoid the non-standard "REPLACE INTO" syntax - $f = fopen( "../maintenance/interwiki.sql", 'r' ); - if ($f == false ) { - dieout( "
  • Could not find the interwiki.sql file"); - } - ## We simply assume it is already empty as we have just created it - $SQL = "INSERT INTO interwiki(iw_prefix,iw_url,iw_local) VALUES "; - while ( ! feof( $f ) ) { - $line = fgets($f,1024); - $matches = array(); - if (!preg_match('/^\s*(\(.+?),(\d)\)/', $line, $matches)) { - continue; - } - $this->query("$SQL $matches[1],$matches[2])"); - } - print " (table interwiki successfully populated)...\n"; - - $this->doQuery("COMMIT"); - } - - function encodeBlob( $b ) { - return new Blob ( pg_escape_bytea( $b ) ) ; - } - - function decodeBlob( $b ) { - if ($b instanceof Blob) { - $b = $b->fetch(); - } - return pg_unescape_bytea( $b ); - } - - function strencode( $s ) { ## Should not be called by us - return pg_escape_string( $s ); - } - - function addQuotes( $s ) { - if ( is_null( $s ) ) { - return 'NULL'; - } else if ($s instanceof Blob) { - return "'".$s->fetch($s)."'"; - } - return "'" . pg_escape_string($s) . "'"; - } - - function quote_ident( $s ) { - return '"' . preg_replace( '/"/', '""', $s) . '"'; - } - - /* For now, does nothing */ - function selectDB( $db ) { - return true; - } - - /** - * Postgres specific version of replaceVars. - * Calls the parent version in Database.php - * - * @private - * - * @param string $com SQL string, read from a stream (usually tables.sql) - * - * @return string SQL string - */ - protected function replaceVars( $ins ) { - - $ins = parent::replaceVars( $ins ); - - if ($this->numeric_version >= 8.3) { - // Thanks for not providing backwards-compatibility, 8.3 - $ins = preg_replace( "/to_tsvector\s*\(\s*'default'\s*,/", 'to_tsvector(', $ins ); - } - - if ($this->numeric_version <= 8.1) { // Our minimum version - $ins = str_replace( 'USING gin', 'USING gist', $ins ); - } - - return $ins; - } - - /** - * Various select options - * - * @private - * - * @param array $options an associative array of options to be turned into - * an SQL query, valid keys are listed in the function. - * @return array - */ - function makeSelectOptions( $options ) { - $preLimitTail = $postLimitTail = ''; - $startOpts = $useIndex = ''; - - $noKeyOptions = array(); - foreach ( $options as $key => $option ) { - if ( is_numeric( $key ) ) { - $noKeyOptions[$option] = true; - } - } - - if ( isset( $options['GROUP BY'] ) ) $preLimitTail .= " GROUP BY " . $options['GROUP BY']; - if ( isset( $options['HAVING'] ) ) $preLimitTail .= " HAVING {$options['HAVING']}"; - if ( isset( $options['ORDER BY'] ) ) $preLimitTail .= " ORDER BY " . $options['ORDER BY']; - - //if (isset($options['LIMIT'])) { - // $tailOpts .= $this->limitResult('', $options['LIMIT'], - // isset($options['OFFSET']) ? $options['OFFSET'] - // : false); - //} - - if ( isset( $noKeyOptions['FOR UPDATE'] ) ) $postLimitTail .= ' FOR UPDATE'; - if ( isset( $noKeyOptions['LOCK IN SHARE MODE'] ) ) $postLimitTail .= ' LOCK IN SHARE MODE'; - if ( isset( $noKeyOptions['DISTINCT'] ) || isset( $noKeyOptions['DISTINCTROW'] ) ) $startOpts .= 'DISTINCT'; - - return array( $startOpts, $useIndex, $preLimitTail, $postLimitTail ); - } - - public function setTimeout( $timeout ) { - // @todo fixme no-op - } - - function ping() { - wfDebug( "Function ping() not written for DatabasePostgres.php yet"); - return true; - } - - /** - * How lagged is this slave? - * - */ - public function getLag() { - # Not implemented for PostgreSQL - return false; - } - - function buildConcat( $stringList ) { - return implode( ' || ', $stringList ); - } - -} // end DatabasePostgres class - - diff --git a/includes/DateFormatter.php b/includes/DateFormatter.php deleted file mode 100644 index bbad6d15..00000000 --- a/includes/DateFormatter.php +++ /dev/null @@ -1,285 +0,0 @@ -monthNames = $this->getMonthRegex(); - for ( $i=1; $i<=12; $i++ ) { - $this->xMonths[$wgContLang->lc( $wgContLang->getMonthName( $i ) )] = $i; - $this->xMonths[$wgContLang->lc( $wgContLang->getMonthAbbreviation( $i ) )] = $i; - } - - $this->regexTrail = '(?![a-z])/iu'; - - # Partial regular expressions - $this->prxDM = '\[\[(\d{1,2})[ _](' . $this->monthNames . ')]]'; - $this->prxMD = '\[\[(' . $this->monthNames . ')[ _](\d{1,2})]]'; - $this->prxY = '\[\[(\d{1,4}([ _]BC|))]]'; - $this->prxISO1 = '\[\[(-?\d{4})]]-\[\[(\d{2})-(\d{2})]]'; - $this->prxISO2 = '\[\[(-?\d{4})-(\d{2})-(\d{2})]]'; - - # Real regular expressions - $this->regexes[self::DMY] = "/{$this->prxDM} *,? *{$this->prxY}{$this->regexTrail}"; - $this->regexes[self::YDM] = "/{$this->prxY} *,? *{$this->prxDM}{$this->regexTrail}"; - $this->regexes[self::MDY] = "/{$this->prxMD} *,? *{$this->prxY}{$this->regexTrail}"; - $this->regexes[self::YMD] = "/{$this->prxY} *,? *{$this->prxMD}{$this->regexTrail}"; - $this->regexes[self::DM] = "/{$this->prxDM}{$this->regexTrail}"; - $this->regexes[self::MD] = "/{$this->prxMD}{$this->regexTrail}"; - $this->regexes[self::ISO1] = "/{$this->prxISO1}{$this->regexTrail}"; - $this->regexes[self::ISO2] = "/{$this->prxISO2}{$this->regexTrail}"; - - # Extraction keys - # See the comments in replace() for the meaning of the letters - $this->keys[self::DMY] = 'jFY'; - $this->keys[self::YDM] = 'Y jF'; - $this->keys[self::MDY] = 'FjY'; - $this->keys[self::YMD] = 'Y Fj'; - $this->keys[self::DM] = 'jF'; - $this->keys[self::MD] = 'Fj'; - $this->keys[self::ISO1] = 'ymd'; # y means ISO year - $this->keys[self::ISO2] = 'ymd'; - - # Target date formats - $this->targets[self::DMY] = '[[F j|j F]] [[Y]]'; - $this->targets[self::YDM] = '[[Y]], [[F j|j F]]'; - $this->targets[self::MDY] = '[[F j]], [[Y]]'; - $this->targets[self::YMD] = '[[Y]] [[F j]]'; - $this->targets[self::DM] = '[[F j|j F]]'; - $this->targets[self::MD] = '[[F j]]'; - $this->targets[self::ISO1] = '[[Y|y]]-[[F j|m-d]]'; - $this->targets[self::ISO2] = '[[y-m-d]]'; - - # Rules - # pref source target - $this->rules[self::DMY][self::MD] = self::DM; - $this->rules[self::ALL][self::MD] = self::MD; - $this->rules[self::MDY][self::DM] = self::MD; - $this->rules[self::ALL][self::DM] = self::DM; - $this->rules[self::NONE][self::ISO2] = self::ISO1; - - $this->preferences = array( - 'default' => self::NONE, - 'dmy' => self::DMY, - 'mdy' => self::MDY, - 'ymd' => self::YMD, - 'ISO 8601' => self::ISO1, - ); - } - - /** - * @static - */ - function &getInstance() { - global $wgMemc; - static $dateFormatter = false; - if ( !$dateFormatter ) { - $dateFormatter = $wgMemc->get( wfMemcKey( 'dateformatter' ) ); - if ( !$dateFormatter ) { - $dateFormatter = new DateFormatter; - $wgMemc->set( wfMemcKey( 'dateformatter' ), $dateFormatter, 3600 ); - } - } - return $dateFormatter; - } - - /** - * @param string $preference User preference - * @param string $text Text to reformat - */ - function reformat( $preference, $text ) { - if ( isset( $this->preferences[$preference] ) ) { - $preference = $this->preferences[$preference]; - } else { - $preference = self::NONE; - } - for ( $i=1; $i<=self::LAST; $i++ ) { - $this->mSource = $i; - if ( isset ( $this->rules[$preference][$i] ) ) { - # Specific rules - $this->mTarget = $this->rules[$preference][$i]; - } elseif ( isset ( $this->rules[self::ALL][$i] ) ) { - # General rules - $this->mTarget = $this->rules[self::ALL][$i]; - } elseif ( $preference ) { - # User preference - $this->mTarget = $preference; - } else { - # Default - $this->mTarget = $i; - } - $text = preg_replace_callback( $this->regexes[$i], array( &$this, 'replace' ), $text ); - } - return $text; - } - - /** - * @param $matches - */ - function replace( $matches ) { - # Extract information from $matches - $bits = array(); - $key = $this->keys[$this->mSource]; - for ( $p=0; $p < strlen($key); $p++ ) { - if ( $key{$p} != ' ' ) { - $bits[$key{$p}] = $matches[$p+1]; - } - } - - $format = $this->targets[$this->mTarget]; - - # Construct new date - $text = ''; - $fail = false; - - for ( $p=0; $p < strlen( $format ); $p++ ) { - $char = $format{$p}; - switch ( $char ) { - case 'd': # ISO day of month - if ( !isset($bits['d']) ) { - $text .= sprintf( '%02d', $bits['j'] ); - } else { - $text .= $bits['d']; - } - break; - case 'm': # ISO month - if ( !isset($bits['m']) ) { - $m = $this->makeIsoMonth( $bits['F'] ); - if ( !$m || $m == '00' ) { - $fail = true; - } else { - $text .= $m; - } - } else { - $text .= $bits['m']; - } - break; - case 'y': # ISO year - if ( !isset( $bits['y'] ) ) { - $text .= $this->makeIsoYear( $bits['Y'] ); - } else { - $text .= $bits['y']; - } - break; - case 'j': # ordinary day of month - if ( !isset($bits['j']) ) { - $text .= intval( $bits['d'] ); - } else { - $text .= $bits['j']; - } - break; - case 'F': # long month - if ( !isset( $bits['F'] ) ) { - $m = intval($bits['m']); - if ( $m > 12 || $m < 1 ) { - $fail = true; - } else { - global $wgContLang; - $text .= $wgContLang->getMonthName( $m ); - } - } else { - $text .= ucfirst( $bits['F'] ); - } - break; - case 'Y': # ordinary (optional BC) year - if ( !isset( $bits['Y'] ) ) { - $text .= $this->makeNormalYear( $bits['y'] ); - } else { - $text .= $bits['Y']; - } - break; - default: - $text .= $char; - } - } - if ( $fail ) { - $text = $matches[0]; - } - return $text; - } - - /** - * @todo document - */ - function getMonthRegex() { - global $wgContLang; - $names = array(); - for( $i = 1; $i <= 12; $i++ ) { - $names[] = $wgContLang->getMonthName( $i ); - $names[] = $wgContLang->getMonthAbbreviation( $i ); - } - return implode( '|', $names ); - } - - /** - * Makes an ISO month, e.g. 02, from a month name - * @param $monthName String: month name - * @return string ISO month name - */ - function makeIsoMonth( $monthName ) { - global $wgContLang; - - $n = $this->xMonths[$wgContLang->lc( $monthName )]; - return sprintf( '%02d', $n ); - } - - /** - * @todo document - * @param $year String: Year name - * @return string ISO year name - */ - function makeIsoYear( $year ) { - # Assumes the year is in a nice format, as enforced by the regex - if ( substr( $year, -2 ) == 'BC' ) { - $num = intval(substr( $year, 0, -3 )) - 1; - # PHP bug note: sprintf( "%04d", -1 ) fails poorly - $text = sprintf( '-%04d', $num ); - - } else { - $text = sprintf( '%04d', $year ); - } - return $text; - } - - /** - * @todo document - */ - function makeNormalYear( $iso ) { - if ( $iso{0} == '-' ) { - $text = (intval( substr( $iso, 1 ) ) + 1) . ' BC'; - } else { - $text = intval( $iso ); - } - return $text; - } -} - - diff --git a/includes/DefaultSettings.php b/includes/DefaultSettings.php index d1d04a45..cb8bb001 100644 --- a/includes/DefaultSettings.php +++ b/includes/DefaultSettings.php @@ -31,7 +31,7 @@ require_once( "$IP/includes/SiteConfiguration.php" ); $wgConf = new SiteConfiguration; /** MediaWiki version number */ -$wgVersion = '1.13.2'; +$wgVersion = '1.13.3'; /** Name of the site. It must be changed in LocalSettings.php */ $wgSitename = 'MediaWiki'; @@ -1794,6 +1794,8 @@ $wgMimeTypeBlacklist= array( 'application/x-php', 'text/x-php', # Other types that may be interpreted by some servers 'text/x-python', 'text/x-perl', 'text/x-bash', 'text/x-sh', 'text/x-csh', + # Client-side hazards on Internet Explorer + 'text/scriptlet', 'application/x-msdownload', # Windows metafile, client-side vulnerability on some systems 'application/x-msmetafile' ); @@ -2288,7 +2290,7 @@ $wgAutoloadClasses = array(); * $wgExtensionCredits[$type][] = array( * 'name' => 'Example extension', * 'version' => 1.9, - * 'svn-revision' => '$LastChangedRevision: 41545 $', + * 'svn-revision' => '$LastChangedRevision: 44568 $', * 'author' => 'Foo Barstein', * 'url' => 'http://wwww.example.com/Example%20Extension/', * 'description' => 'An example extension', diff --git a/includes/Exception.php b/includes/Exception.php index 74820204..ab25f0b8 100644 --- a/includes/Exception.php +++ b/includes/Exception.php @@ -274,7 +274,16 @@ function wfReportException( Exception $e ) { } } } else { - echo $e->__toString(); + $message = "Unexpected non-MediaWiki exception encountered, of type \"" . get_class( $e ) . "\"\n" . + $e->__toString() . "\n"; + if ( $GLOBALS['wgShowExceptionDetails'] ) { + $message .= "\n" . $e->getTraceAsString() ."\n"; + } + if ( !empty( $GLOBALS['wgCommandLineMode'] ) ) { + wfPrintError( $message ); + } else { + echo nl2br( htmlspecialchars( $message ) ). "\n"; + } } } diff --git a/includes/HTMLForm.php b/includes/HTMLForm.php deleted file mode 100644 index 69ec1007..00000000 --- a/includes/HTMLForm.php +++ /dev/null @@ -1,107 +0,0 @@ -mRequest = $request; - } - - /** - * @private - * @param $name String: name of the fieldset. - * @param $content String: HTML content to put in. - * @return string HTML fieldset - */ - function fieldset( $name, $content ) { - return "
    ".wfMsg($this->mName.'-'.$name)."\n" . - $content . "\n
    \n"; - } - - /** - * @private - * @param $varname String: name of the checkbox. - * @param $checked Boolean: set true to check the box (default False). - */ - function checkbox( $varname, $checked=false ) { - if ( $this->mRequest->wasPosted() && !is_null( $this->mRequest->getVal( $varname ) ) ) { - $checked = $this->mRequest->getCheck( $varname ); - } - return "
    \n"; - } - - /** - * @private - * @param $varname String: name of the textbox. - * @param $value String: optional value (default empty) - * @param $size Integer: optional size of the textbox (default 20) - */ - function textbox( $varname, $value='', $size=20 ) { - if ( $this->mRequest->wasPosted() ) { - $value = $this->mRequest->getText( $varname, $value ); - } - $value = htmlspecialchars( $value ); - return "
    \n"; - } - - /** - * @private - * @param $varname String: name of the radiobox. - * @param $fields Array: Various fields. - */ - function radiobox( $varname, $fields ) { - foreach ( $fields as $value => $checked ) { - $s .= "
    \n"; - } - return $this->fieldset( $varname, $s ); - } - - /** - * @private - * @param $varname String: name of the textareabox. - * @param $value String: optional value (default empty) - * @param $size Integer: optional size of the textarea (default 20) - */ - function textareabox ( $varname, $value='', $size=20 ) { - if ( $this->mRequest->wasPosted() ) { - $value = $this->mRequest->getText( $varname, $value ); - } - $value = htmlspecialchars( $value ); - return '
    \n"; - } - - /** - * @private - * @param $varname String: name of the arraybox. - * @param $size Integer: Optional size of the textarea (default 20) - */ - function arraybox( $varname , $size=20 ) { - $s = ''; - if ( $this->mRequest->wasPosted() ) { - $arr = $this->mRequest->getArray( $varname ); - if ( is_array( $arr ) ) { - foreach ( $_POST[$varname] as $element ) { - $s .= htmlspecialchars( $element )."\n"; - } - } - } - return "
  • '; } - else if ( ':' == $char ) { - if ( $this->mDTopen ) { - $this->mDTopen = false; - $text = ''; - } else { - $text = ''; - } - } - else { return ''; } - return $text."\n"; - } - /**#@-*/ - - /** - * Make lists from lines starting with ':', '*', '#', etc. - * - * @private - * @return string the lists rendered as HTML - */ - function doBlockLevels( $text, $linestart ) { - $fname = 'Parser::doBlockLevels'; - wfProfileIn( $fname ); - - # Parsing through the text line by line. The main thing - # happening here is handling of block-level elements p, pre, - # and making lists from lines starting with * # : etc. - # - $textLines = explode( "\n", $text ); - - $lastPrefix = $output = ''; - $this->mDTopen = $inBlockElem = false; - $prefixLength = 0; - $paragraphStack = false; - - if ( !$linestart ) { - $output .= array_shift( $textLines ); - } - foreach ( $textLines as $oLine ) { - $lastPrefixLength = strlen( $lastPrefix ); - $preCloseMatch = preg_match('/<\\/pre/i', $oLine ); - $preOpenMatch = preg_match('/
    mInPre ) {
    -				# Multiple prefixes may abut each other for nested lists.
    -				$prefixLength = strspn( $oLine, '*#:;' );
    -				$pref = substr( $oLine, 0, $prefixLength );
    -
    -				# eh?
    -				$pref2 = str_replace( ';', ':', $pref );
    -				$t = substr( $oLine, $prefixLength );
    -				$this->mInPre = !empty($preOpenMatch);
    -			} else {
    -				# Don't interpret any other prefixes in preformatted text
    -				$prefixLength = 0;
    -				$pref = $pref2 = '';
    -				$t = $oLine;
    -			}
    -
    -			# List generation
    -			if( $prefixLength && 0 == strcmp( $lastPrefix, $pref2 ) ) {
    -				# Same as the last item, so no need to deal with nesting or opening stuff
    -				$output .= $this->nextItem( substr( $pref, -1 ) );
    -				$paragraphStack = false;
    -
    -				if ( substr( $pref, -1 ) == ';') {
    -					# The one nasty exception: definition lists work like this:
    -					# ; title : definition text
    -					# So we check for : in the remainder text to split up the
    -					# title and definition, without b0rking links.
    -					$term = $t2 = '';
    -					if ($this->findColonNoLinks($t, $term, $t2) !== false) {
    -						$t = $t2;
    -						$output .= $term . $this->nextItem( ':' );
    -					}
    -				}
    -			} elseif( $prefixLength || $lastPrefixLength ) {
    -				# Either open or close a level...
    -				$commonPrefixLength = $this->getCommon( $pref, $lastPrefix );
    -				$paragraphStack = false;
    -
    -				while( $commonPrefixLength < $lastPrefixLength ) {
    -					$output .= $this->closeList( $lastPrefix{$lastPrefixLength-1} );
    -					--$lastPrefixLength;
    -				}
    -				if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) {
    -					$output .= $this->nextItem( $pref{$commonPrefixLength-1} );
    -				}
    -				while ( $prefixLength > $commonPrefixLength ) {
    -					$char = substr( $pref, $commonPrefixLength, 1 );
    -					$output .= $this->openList( $char );
    -
    -					if ( ';' == $char ) {
    -						# FIXME: This is dupe of code above
    -						if ($this->findColonNoLinks($t, $term, $t2) !== false) {
    -							$t = $t2;
    -							$output .= $term . $this->nextItem( ':' );
    -						}
    -					}
    -					++$commonPrefixLength;
    -				}
    -				$lastPrefix = $pref2;
    -			}
    -			if( 0 == $prefixLength ) {
    -				wfProfileIn( "$fname-paragraph" );
    -				# No prefix (not in list)--go to paragraph mode
    -				// XXX: use a stack for nestable elements like span, table and div
    -				$openmatch = preg_match('/(?:mUniqPrefix.'-pre|<\\/li|<\\/ul|<\\/ol|<\\/?center)/iS', $t );
    -				if ( $openmatch or $closematch ) {
    -					$paragraphStack = false;
    -					# TODO bug 5718: paragraph closed
    -					$output .= $this->closeParagraph();
    -					if ( $preOpenMatch and !$preCloseMatch ) {
    -						$this->mInPre = true;
    -					}
    -					if ( $closematch ) {
    -						$inBlockElem = false;
    -					} else {
    -						$inBlockElem = true;
    -					}
    -				} else if ( !$inBlockElem && !$this->mInPre ) {
    -					if ( ' ' == $t{0} and ( $this->mLastSection == 'pre' or trim($t) != '' ) ) {
    -						// pre
    -						if ($this->mLastSection != 'pre') {
    -							$paragraphStack = false;
    -							$output .= $this->closeParagraph().'
    ';
    -							$this->mLastSection = 'pre';
    -						}
    -						$t = substr( $t, 1 );
    -					} else {
    -						// paragraph
    -						if ( '' == trim($t) ) {
    -							if ( $paragraphStack ) {
    -								$output .= $paragraphStack.'
    '; - $paragraphStack = false; - $this->mLastSection = 'p'; - } else { - if ($this->mLastSection != 'p' ) { - $output .= $this->closeParagraph(); - $this->mLastSection = ''; - $paragraphStack = '

    '; - } else { - $paragraphStack = '

    '; - } - } - } else { - if ( $paragraphStack ) { - $output .= $paragraphStack; - $paragraphStack = false; - $this->mLastSection = 'p'; - } else if ($this->mLastSection != 'p') { - $output .= $this->closeParagraph().'

    '; - $this->mLastSection = 'p'; - } - } - } - } - wfProfileOut( "$fname-paragraph" ); - } - // somewhere above we forget to get out of pre block (bug 785) - if($preCloseMatch && $this->mInPre) { - $this->mInPre = false; - } - if ($paragraphStack === false) { - $output .= $t."\n"; - } - } - while ( $prefixLength ) { - $output .= $this->closeList( $pref2{$prefixLength-1} ); - --$prefixLength; - } - if ( '' != $this->mLastSection ) { - $output .= 'mLastSection . '>'; - $this->mLastSection = ''; - } - - wfProfileOut( $fname ); - return $output; - } - - /** - * Split up a string on ':', ignoring any occurences inside tags - * to prevent illegal overlapping. - * @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) { - $fname = 'Parser::findColonNoLinks'; - wfProfileIn( $fname ); - - $pos = strpos( $str, ':' ); - if( $pos === false ) { - // Nothing to find! - wfProfileOut( $fname ); - return false; - } - - $lt = strpos( $str, '<' ); - if( $lt === false || $lt > $pos ) { - // Easy; no tag nesting to worry about - $before = substr( $str, 0, $pos ); - $after = substr( $str, $pos+1 ); - wfProfileOut( $fname ); - return $pos; - } - - // Ugly state machine to walk through avoiding tags. - $state = self::COLON_STATE_TEXT; - $stack = 0; - $len = strlen( $str ); - for( $i = 0; $i < $len; $i++ ) { - $c = $str{$i}; - - switch( $state ) { - // (Using the number is a performance hack for common cases) - case 0: // self::COLON_STATE_TEXT: - switch( $c ) { - case "<": - // Could be either a tag or an tag - $state = self::COLON_STATE_TAGSTART; - break; - case ":": - if( $stack == 0 ) { - // We found it! - $before = substr( $str, 0, $i ); - $after = substr( $str, $i + 1 ); - wfProfileOut( $fname ); - return $i; - } - // Embedded in a tag; don't break it. - break; - default: - // Skip ahead looking for something interesting - $colon = strpos( $str, ':', $i ); - if( $colon === false ) { - // Nothing else interesting - wfProfileOut( $fname ); - return false; - } - $lt = strpos( $str, '<', $i ); - if( $stack === 0 ) { - if( $lt === false || $colon < $lt ) { - // We found it! - $before = substr( $str, 0, $colon ); - $after = substr( $str, $colon + 1 ); - wfProfileOut( $fname ); - return $i; - } - } - if( $lt === false ) { - // Nothing else interesting to find; abort! - // We're nested, but there's no close tags left. Abort! - break 2; - } - // Skip ahead to next tag start - $i = $lt; - $state = self::COLON_STATE_TAGSTART; - } - break; - case 1: // self::COLON_STATE_TAG: - // In a - switch( $c ) { - case ">": - $stack++; - $state = self::COLON_STATE_TEXT; - break; - case "/": - // Slash may be followed by >? - $state = self::COLON_STATE_TAGSLASH; - break; - default: - // ignore - } - break; - case 2: // self::COLON_STATE_TAGSTART: - switch( $c ) { - case "/": - $state = self::COLON_STATE_CLOSETAG; - break; - case "!": - $state = self::COLON_STATE_COMMENT; - break; - case ">": - // Illegal early close? This shouldn't happen D: - $state = self::COLON_STATE_TEXT; - break; - default: - $state = self::COLON_STATE_TAG; - } - break; - case 3: // self::COLON_STATE_CLOSETAG: - // In a - if( $c == ">" ) { - $stack--; - if( $stack < 0 ) { - wfDebug( "Invalid input in $fname; too many close tags\n" ); - wfProfileOut( $fname ); - return false; - } - $state = self::COLON_STATE_TEXT; - } - break; - case self::COLON_STATE_TAGSLASH: - if( $c == ">" ) { - // Yes, a self-closed tag - $state = self::COLON_STATE_TEXT; - } else { - // Probably we're jumping the gun, and this is an attribute - $state = self::COLON_STATE_TAG; - } - break; - case 5: // self::COLON_STATE_COMMENT: - if( $c == "-" ) { - $state = self::COLON_STATE_COMMENTDASH; - } - break; - case self::COLON_STATE_COMMENTDASH: - if( $c == "-" ) { - $state = self::COLON_STATE_COMMENTDASHDASH; - } else { - $state = self::COLON_STATE_COMMENT; - } - break; - case self::COLON_STATE_COMMENTDASHDASH: - if( $c == ">" ) { - $state = self::COLON_STATE_TEXT; - } else { - $state = self::COLON_STATE_COMMENT; - } - break; - default: - throw new MWException( "State machine error in $fname" ); - } - } - if( $stack > 0 ) { - wfDebug( "Invalid input in $fname; not enough close tags (stack $stack, state $state)\n" ); - return false; - } - wfProfileOut( $fname ); - return false; - } - - /** - * Return value of a magic variable (like PAGENAME) - * - * @private - */ - function getVariableValue( $index ) { - global $wgContLang, $wgSitename, $wgServer, $wgServerName, $wgScriptPath; - - /** - * Some of these require message or data lookups and can be - * expensive to check many times. - */ - if ( wfRunHooks( 'ParserGetVariableValueVarCache', array( &$this, &$this->mVarCache ) ) ) { - if ( isset( $this->mVarCache[$index] ) ) { - return $this->mVarCache[$index]; - } - } - - $ts = wfTimestamp( TS_UNIX, $this->mOptions->getTimestamp() ); - wfRunHooks( 'ParserGetVariableValueTs', array( &$this, &$ts ) ); - - # Use the time zone - global $wgLocaltimezone; - if ( isset( $wgLocaltimezone ) ) { - $oldtz = getenv( 'TZ' ); - putenv( 'TZ='.$wgLocaltimezone ); - } - - wfSuppressWarnings(); // E_STRICT system time bitching - $localTimestamp = date( 'YmdHis', $ts ); - $localMonth = date( 'm', $ts ); - $localMonthName = date( 'n', $ts ); - $localDay = date( 'j', $ts ); - $localDay2 = date( 'd', $ts ); - $localDayOfWeek = date( 'w', $ts ); - $localWeek = date( 'W', $ts ); - $localYear = date( 'Y', $ts ); - $localHour = date( 'H', $ts ); - if ( isset( $wgLocaltimezone ) ) { - putenv( 'TZ='.$oldtz ); - } - wfRestoreWarnings(); - - switch ( $index ) { - case 'currentmonth': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'm', $ts ) ); - case 'currentmonthname': - return $this->mVarCache[$index] = $wgContLang->getMonthName( gmdate( 'n', $ts ) ); - case 'currentmonthnamegen': - return $this->mVarCache[$index] = $wgContLang->getMonthNameGen( gmdate( 'n', $ts ) ); - case 'currentmonthabbrev': - return $this->mVarCache[$index] = $wgContLang->getMonthAbbreviation( gmdate( 'n', $ts ) ); - case 'currentday': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'j', $ts ) ); - case 'currentday2': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'd', $ts ) ); - case 'localmonth': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localMonth ); - case 'localmonthname': - return $this->mVarCache[$index] = $wgContLang->getMonthName( $localMonthName ); - case 'localmonthnamegen': - return $this->mVarCache[$index] = $wgContLang->getMonthNameGen( $localMonthName ); - case 'localmonthabbrev': - return $this->mVarCache[$index] = $wgContLang->getMonthAbbreviation( $localMonthName ); - case 'localday': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localDay ); - case 'localday2': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localDay2 ); - case 'pagename': - return wfEscapeWikiText( $this->mTitle->getText() ); - case 'pagenamee': - return $this->mTitle->getPartialURL(); - case 'fullpagename': - return wfEscapeWikiText( $this->mTitle->getPrefixedText() ); - case 'fullpagenamee': - return $this->mTitle->getPrefixedURL(); - case 'subpagename': - return wfEscapeWikiText( $this->mTitle->getSubpageText() ); - case 'subpagenamee': - return $this->mTitle->getSubpageUrlForm(); - case 'basepagename': - return wfEscapeWikiText( $this->mTitle->getBaseText() ); - case 'basepagenamee': - return wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ); - case 'talkpagename': - if( $this->mTitle->canTalk() ) { - $talkPage = $this->mTitle->getTalkPage(); - return wfEscapeWikiText( $talkPage->getPrefixedText() ); - } else { - return ''; - } - case 'talkpagenamee': - if( $this->mTitle->canTalk() ) { - $talkPage = $this->mTitle->getTalkPage(); - return $talkPage->getPrefixedUrl(); - } else { - return ''; - } - case 'subjectpagename': - $subjPage = $this->mTitle->getSubjectPage(); - return wfEscapeWikiText( $subjPage->getPrefixedText() ); - case 'subjectpagenamee': - $subjPage = $this->mTitle->getSubjectPage(); - return $subjPage->getPrefixedUrl(); - case 'revisionid': - // Let the edit saving system know we should parse the page - // *after* a revision ID has been assigned. - $this->mOutput->setFlag( 'vary-revision' ); - wfDebug( __METHOD__ . ": {{REVISIONID}} used, setting vary-revision...\n" ); - return $this->mRevisionId; - case 'revisionday': - // Let the edit saving system know we should parse the page - // *after* a revision ID has been assigned. This is for null edits. - $this->mOutput->setFlag( 'vary-revision' ); - wfDebug( __METHOD__ . ": {{REVISIONDAY}} used, setting vary-revision...\n" ); - return intval( substr( $this->getRevisionTimestamp(), 6, 2 ) ); - case 'revisionday2': - // Let the edit saving system know we should parse the page - // *after* a revision ID has been assigned. This is for null edits. - $this->mOutput->setFlag( 'vary-revision' ); - wfDebug( __METHOD__ . ": {{REVISIONDAY2}} used, setting vary-revision...\n" ); - return substr( $this->getRevisionTimestamp(), 6, 2 ); - case 'revisionmonth': - // Let the edit saving system know we should parse the page - // *after* a revision ID has been assigned. This is for null edits. - $this->mOutput->setFlag( 'vary-revision' ); - wfDebug( __METHOD__ . ": {{REVISIONMONTH}} used, setting vary-revision...\n" ); - return intval( substr( $this->getRevisionTimestamp(), 4, 2 ) ); - case 'revisionyear': - // Let the edit saving system know we should parse the page - // *after* a revision ID has been assigned. This is for null edits. - $this->mOutput->setFlag( 'vary-revision' ); - wfDebug( __METHOD__ . ": {{REVISIONYEAR}} used, setting vary-revision...\n" ); - return substr( $this->getRevisionTimestamp(), 0, 4 ); - case 'revisiontimestamp': - // Let the edit saving system know we should parse the page - // *after* a revision ID has been assigned. This is for null edits. - $this->mOutput->setFlag( 'vary-revision' ); - wfDebug( __METHOD__ . ": {{REVISIONTIMESTAMP}} used, setting vary-revision...\n" ); - return $this->getRevisionTimestamp(); - case 'namespace': - return str_replace('_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) ); - case 'namespacee': - return wfUrlencode( $wgContLang->getNsText( $this->mTitle->getNamespace() ) ); - case 'talkspace': - return $this->mTitle->canTalk() ? str_replace('_',' ',$this->mTitle->getTalkNsText()) : ''; - case 'talkspacee': - return $this->mTitle->canTalk() ? wfUrlencode( $this->mTitle->getTalkNsText() ) : ''; - case 'subjectspace': - return $this->mTitle->getSubjectNsText(); - case 'subjectspacee': - return( wfUrlencode( $this->mTitle->getSubjectNsText() ) ); - case 'currentdayname': - return $this->mVarCache[$index] = $wgContLang->getWeekdayName( gmdate( 'w', $ts ) + 1 ); - case 'currentyear': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'Y', $ts ), true ); - case 'currenttime': - return $this->mVarCache[$index] = $wgContLang->time( wfTimestamp( TS_MW, $ts ), false, false ); - case 'currenthour': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'H', $ts ), true ); - case 'currentweek': - // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to - // int to remove the padding - return $this->mVarCache[$index] = $wgContLang->formatNum( (int)gmdate( 'W', $ts ) ); - case 'currentdow': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'w', $ts ) ); - case 'localdayname': - return $this->mVarCache[$index] = $wgContLang->getWeekdayName( $localDayOfWeek + 1 ); - case 'localyear': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localYear, true ); - case 'localtime': - return $this->mVarCache[$index] = $wgContLang->time( $localTimestamp, false, false ); - case 'localhour': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localHour, true ); - case 'localweek': - // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to - // int to remove the padding - return $this->mVarCache[$index] = $wgContLang->formatNum( (int)$localWeek ); - case 'localdow': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localDayOfWeek ); - case 'numberofarticles': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::articles() ); - case 'numberoffiles': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::images() ); - case 'numberofusers': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::users() ); - case 'numberofpages': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::pages() ); - case 'numberofadmins': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::admins() ); - case 'numberofedits': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::edits() ); - case 'currenttimestamp': - return $this->mVarCache[$index] = wfTimestamp( TS_MW, $ts ); - case 'localtimestamp': - return $this->mVarCache[$index] = $localTimestamp; - case 'currentversion': - return $this->mVarCache[$index] = SpecialVersion::getVersion(); - case 'sitename': - return $wgSitename; - case 'server': - return $wgServer; - case 'servername': - return $wgServerName; - case 'scriptpath': - return $wgScriptPath; - case 'directionmark': - return $wgContLang->getDirMark(); - case 'contentlanguage': - global $wgContLanguageCode; - return $wgContLanguageCode; - default: - $ret = null; - if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret ) ) ) - return $ret; - else - return null; - } - } - - /** - * initialise the magic variables (like CURRENTMONTHNAME) - * - * @private - */ - function initialiseVariables() { - $fname = 'Parser::initialiseVariables'; - wfProfileIn( $fname ); - $variableIDs = MagicWord::getVariableIDs(); - - $this->mVariables = new MagicWordArray( $variableIDs ); - wfProfileOut( $fname ); - } - - /** - * Preprocess some wikitext and return the document tree. - * This is the ghost of replace_variables(). - * - * @param string $text The text to parse - * @param integer flags Bitwise combination of: - * self::PTD_FOR_INCLUSION Handle / as if the text is being - * included. Default is to assume a direct page view. - * - * The generated DOM tree must depend only on the input text and the flags. - * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899. - * - * Any flag added to the $flags parameter here, or any other parameter liable to cause a - * change in the DOM tree for a given text, must be passed through the section identifier - * in the section edit link and thus back to extractSections(). - * - * The output of this function is currently only cached in process memory, but a persistent - * cache may be implemented at a later date which takes further advantage of these strict - * dependency requirements. - * - * @private - */ - function preprocessToDom ( $text, $flags = 0 ) { - $dom = $this->getPreprocessor()->preprocessToObj( $text, $flags ); - return $dom; - } - - /* - * Return a three-element array: leading whitespace, string contents, trailing whitespace - */ - public static function splitWhitespace( $s ) { - $ltrimmed = ltrim( $s ); - $w1 = substr( $s, 0, strlen( $s ) - strlen( $ltrimmed ) ); - $trimmed = rtrim( $ltrimmed ); - $diff = strlen( $ltrimmed ) - strlen( $trimmed ); - if ( $diff > 0 ) { - $w2 = substr( $ltrimmed, -$diff ); - } else { - $w2 = ''; - } - return array( $w1, $trimmed, $w2 ); - } - - /** - * Replace magic variables, templates, and template arguments - * with the appropriate text. Templates are substituted recursively, - * taking care to avoid infinite loops. - * - * Note that the substitution depends on value of $mOutputType: - * self::OT_WIKI: only {{subst:}} templates - * self::OT_PREPROCESS: templates but not extension tags - * self::OT_HTML: all templates and extension tags - * - * @param string $tex The text to transform - * @param PPFrame $frame Object describing the arguments passed to the template - * @param bool $argsOnly Only do argument (triple-brace) expansion, not double-brace expansion - * @private - */ - function replaceVariables( $text, $frame = false, $argsOnly = false ) { - # Prevent too big inclusions - if( strlen( $text ) > $this->mOptions->getMaxIncludeSize() ) { - return $text; - } - - $fname = __METHOD__; - wfProfileIn( $fname ); - - if ( $frame === false ) { - $frame = $this->getPreprocessor()->newFrame(); - } elseif ( !( $frame instanceof PPFrame ) ) { - throw new MWException( __METHOD__ . ' called using the old argument format' ); - } - - $dom = $this->preprocessToDom( $text ); - $flags = $argsOnly ? PPFrame::NO_TEMPLATES : 0; - $text = $frame->expand( $dom, $flags ); - - wfProfileOut( $fname ); - return $text; - } - - /// Clean up argument array - refactored in 1.9 so parserfunctions can use it, too. - static function createAssocArgs( $args ) { - $assocArgs = array(); - $index = 1; - foreach( $args as $arg ) { - $eqpos = strpos( $arg, '=' ); - if ( $eqpos === false ) { - $assocArgs[$index++] = $arg; - } else { - $name = trim( substr( $arg, 0, $eqpos ) ); - $value = trim( substr( $arg, $eqpos+1 ) ); - if ( $value === false ) { - $value = ''; - } - if ( $name !== false ) { - $assocArgs[$name] = $value; - } - } - } - - return $assocArgs; - } - - /** - * Return the text of a template, after recursively - * replacing any variables or templates within the template. - * - * @param array $piece The parts of the template - * $piece['title']: the title, i.e. the part before the | - * $piece['parts']: the parameter array - * $piece['lineStart']: whether the brace was at the start of a line - * @param PPFrame The current frame, contains template arguments - * @return string the text of the template - * @private - */ - function braceSubstitution( $piece, $frame ) { - global $wgContLang, $wgLang, $wgAllowDisplayTitle, $wgNonincludableNamespaces; - $fname = __METHOD__; - wfProfileIn( $fname ); - wfProfileIn( __METHOD__.'-setup' ); - - # Flags - $found = false; # $text has been filled - $nowiki = false; # wiki markup in $text should be escaped - $isHTML = false; # $text is HTML, armour it against wikitext transformation - $forceRawInterwiki = false; # Force interwiki transclusion to be done in raw mode not rendered - $isChildObj = false; # $text is a DOM node needing expansion in a child frame - $isLocalObj = false; # $text is a DOM node needing expansion in the current frame - - # Title object, where $text came from - $title = NULL; - - # $part1 is the bit before the first |, and must contain only title characters. - # Various prefixes will be stripped from it later. - $titleWithSpaces = $frame->expand( $piece['title'] ); - $part1 = trim( $titleWithSpaces ); - $titleText = false; - - # Original title text preserved for various purposes - $originalTitle = $part1; - - # $args is a list of argument nodes, starting from index 0, not including $part1 - $args = (null == $piece['parts']) ? array() : $piece['parts']; - wfProfileOut( __METHOD__.'-setup' ); - - # SUBST - wfProfileIn( __METHOD__.'-modifiers' ); - if ( !$found ) { - $mwSubst =& MagicWord::get( 'subst' ); - if ( $mwSubst->matchStartAndRemove( $part1 ) xor $this->ot['wiki'] ) { - # One of two possibilities is true: - # 1) Found SUBST but not in the PST phase - # 2) Didn't find SUBST and in the PST phase - # In either case, return without further processing - $text = $frame->virtualBracketedImplode( '{{', '|', '}}', $titleWithSpaces, $args ); - $isLocalObj = true; - $found = true; - } - } - - # Variables - if ( !$found && $args->getLength() == 0 ) { - $id = $this->mVariables->matchStartToEnd( $part1 ); - if ( $id !== false ) { - $text = $this->getVariableValue( $id ); - if (MagicWord::getCacheTTL($id)>-1) - $this->mOutput->mContainsOldMagic = true; - $found = true; - } - } - - # MSG, MSGNW and RAW - if ( !$found ) { - # Check for MSGNW: - $mwMsgnw =& MagicWord::get( 'msgnw' ); - if ( $mwMsgnw->matchStartAndRemove( $part1 ) ) { - $nowiki = true; - } else { - # Remove obsolete MSG: - $mwMsg =& MagicWord::get( 'msg' ); - $mwMsg->matchStartAndRemove( $part1 ); - } - - # Check for RAW: - $mwRaw =& MagicWord::get( 'raw' ); - if ( $mwRaw->matchStartAndRemove( $part1 ) ) { - $forceRawInterwiki = true; - } - } - wfProfileOut( __METHOD__.'-modifiers' ); - - # Parser functions - if ( !$found ) { - wfProfileIn( __METHOD__ . '-pfunc' ); - - $colonPos = strpos( $part1, ':' ); - if ( $colonPos !== false ) { - # Case sensitive functions - $function = substr( $part1, 0, $colonPos ); - if ( isset( $this->mFunctionSynonyms[1][$function] ) ) { - $function = $this->mFunctionSynonyms[1][$function]; - } else { - # Case insensitive functions - $function = strtolower( $function ); - if ( isset( $this->mFunctionSynonyms[0][$function] ) ) { - $function = $this->mFunctionSynonyms[0][$function]; - } else { - $function = false; - } - } - if ( $function ) { - list( $callback, $flags ) = $this->mFunctionHooks[$function]; - $initialArgs = array( &$this ); - $funcArgs = array( trim( substr( $part1, $colonPos + 1 ) ) ); - if ( $flags & SFH_OBJECT_ARGS ) { - # Add a frame parameter, and pass the arguments as an array - $allArgs = $initialArgs; - $allArgs[] = $frame; - for ( $i = 0; $i < $args->getLength(); $i++ ) { - $funcArgs[] = $args->item( $i ); - } - $allArgs[] = $funcArgs; - } else { - # Convert arguments to plain text - for ( $i = 0; $i < $args->getLength(); $i++ ) { - $funcArgs[] = trim( $frame->expand( $args->item( $i ) ) ); - } - $allArgs = array_merge( $initialArgs, $funcArgs ); - } - - # Workaround for PHP bug 35229 and similar - if ( !is_callable( $callback ) ) { - throw new MWException( "Tag hook for $name is not callable\n" ); - } - $result = call_user_func_array( $callback, $allArgs ); - $found = true; - - if ( is_array( $result ) ) { - if ( isset( $result[0] ) ) { - $text = $result[0]; - unset( $result[0] ); - } - - // Extract flags into the local scope - // This allows callers to set flags such as nowiki, found, etc. - extract( $result ); - } else { - $text = $result; - } - } - } - wfProfileOut( __METHOD__ . '-pfunc' ); - } - - # Finish mangling title and then check for loops. - # Set $title to a Title object and $titleText to the PDBK - if ( !$found ) { - $ns = NS_TEMPLATE; - # Split the title into page and subpage - $subpage = ''; - $part1 = $this->maybeDoSubpageLink( $part1, $subpage ); - if ($subpage !== '') { - $ns = $this->mTitle->getNamespace(); - } - $title = Title::newFromText( $part1, $ns ); - if ( $title ) { - $titleText = $title->getPrefixedText(); - # Check for language variants if the template is not found - if($wgContLang->hasVariants() && $title->getArticleID() == 0){ - $wgContLang->findVariantLink($part1, $title); - } - # Do infinite loop check - if ( !$frame->loopCheck( $title ) ) { - $found = true; - $text = "Template loop detected: [[$titleText]]"; - wfDebug( __METHOD__.": template loop broken at '$titleText'\n" ); - } - # Do recursion depth check - $limit = $this->mOptions->getMaxTemplateDepth(); - if ( $frame->depth >= $limit ) { - $found = true; - $text = "Template recursion depth limit exceeded ($limit)"; - } - } - } - - # Load from database - if ( !$found && $title ) { - wfProfileIn( __METHOD__ . '-loadtpl' ); - if ( !$title->isExternal() ) { - if ( $title->getNamespace() == NS_SPECIAL && $this->mOptions->getAllowSpecialInclusion() && $this->ot['html'] ) { - $text = SpecialPage::capturePath( $title ); - if ( is_string( $text ) ) { - $found = true; - $isHTML = true; - $this->disableCache(); - } - } else if ( $wgNonincludableNamespaces && in_array( $title->getNamespace(), $wgNonincludableNamespaces ) ) { - $found = false; //access denied - wfDebug( "$fname: template inclusion denied for " . $title->getPrefixedDBkey() ); - } else { - list( $text, $title ) = $this->getTemplateDom( $title ); - if ( $text !== false ) { - $found = true; - $isChildObj = true; - } - } - - # If the title is valid but undisplayable, make a link to it - if ( !$found && ( $this->ot['html'] || $this->ot['pre'] ) ) { - $text = "[[:$titleText]]"; - $found = true; - } - } elseif ( $title->isTrans() ) { - // Interwiki transclusion - if ( $this->ot['html'] && !$forceRawInterwiki ) { - $text = $this->interwikiTransclude( $title, 'render' ); - $isHTML = true; - } else { - $text = $this->interwikiTransclude( $title, 'raw' ); - // Preprocess it like a template - $text = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION ); - $isChildObj = true; - } - $found = true; - } - wfProfileOut( __METHOD__ . '-loadtpl' ); - } - - # If we haven't found text to substitute by now, we're done - # Recover the source wikitext and return it - if ( !$found ) { - $text = $frame->virtualBracketedImplode( '{{', '|', '}}', $titleWithSpaces, $args ); - wfProfileOut( $fname ); - return array( 'object' => $text ); - } - - # Expand DOM-style return values in a child frame - if ( $isChildObj ) { - # Clean up argument array - $newFrame = $frame->newChild( $args, $title ); - - if ( $nowiki ) { - $text = $newFrame->expand( $text, PPFrame::RECOVER_ORIG ); - } elseif ( $titleText !== false && $newFrame->isEmpty() ) { - # Expansion is eligible for the empty-frame cache - if ( isset( $this->mTplExpandCache[$titleText] ) ) { - $text = $this->mTplExpandCache[$titleText]; - } else { - $text = $newFrame->expand( $text ); - $this->mTplExpandCache[$titleText] = $text; - } - } else { - # Uncached expansion - $text = $newFrame->expand( $text ); - } - } - if ( $isLocalObj && $nowiki ) { - $text = $frame->expand( $text, PPFrame::RECOVER_ORIG ); - $isLocalObj = false; - } - - # Replace raw HTML by a placeholder - # Add a blank line preceding, to prevent it from mucking up - # immediately preceding headings - if ( $isHTML ) { - $text = "\n\n" . $this->insertStripItem( $text ); - } - # Escape nowiki-style return values - elseif ( $nowiki && ( $this->ot['html'] || $this->ot['pre'] ) ) { - $text = wfEscapeWikiText( $text ); - } - # Bug 529: if the template begins with a table or block-level - # element, it should be treated as beginning a new line. - # This behaviour is somewhat controversial. - elseif ( is_string( $text ) && !$piece['lineStart'] && preg_match('/^(?:{\\||:|;|#|\*)/', $text)) /*}*/{ - $text = "\n" . $text; - } - - if ( is_string( $text ) && !$this->incrementIncludeSize( 'post-expand', strlen( $text ) ) ) { - # Error, oversize inclusion - $text = "[[$originalTitle]]" . - $this->insertStripItem( '' ); - } - - if ( $isLocalObj ) { - $ret = array( 'object' => $text ); - } else { - $ret = array( 'text' => $text ); - } - - wfProfileOut( $fname ); - return $ret; - } - - /** - * Get the semi-parsed DOM representation of a template with a given title, - * and its redirect destination title. Cached. - */ - function getTemplateDom( $title ) { - $cacheTitle = $title; - $titleText = $title->getPrefixedDBkey(); - - if ( isset( $this->mTplRedirCache[$titleText] ) ) { - list( $ns, $dbk ) = $this->mTplRedirCache[$titleText]; - $title = Title::makeTitle( $ns, $dbk ); - $titleText = $title->getPrefixedDBkey(); - } - if ( isset( $this->mTplDomCache[$titleText] ) ) { - return array( $this->mTplDomCache[$titleText], $title ); - } - - // Cache miss, go to the database - list( $text, $title ) = $this->fetchTemplateAndTitle( $title ); - - if ( $text === false ) { - $this->mTplDomCache[$titleText] = false; - return array( false, $title ); - } - - $dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION ); - $this->mTplDomCache[ $titleText ] = $dom; - - if (! $title->equals($cacheTitle)) { - $this->mTplRedirCache[$cacheTitle->getPrefixedDBkey()] = - array( $title->getNamespace(),$cdb = $title->getDBkey() ); - } - - return array( $dom, $title ); - } - - /** - * Fetch the unparsed text of a template and register a reference to it. - */ - function fetchTemplateAndTitle( $title ) { - $templateCb = $this->mOptions->getTemplateCallback(); - $stuff = call_user_func( $templateCb, $title ); - $text = $stuff['text']; - $finalTitle = isset( $stuff['finalTitle'] ) ? $stuff['finalTitle'] : $title; - if ( isset( $stuff['deps'] ) ) { - foreach ( $stuff['deps'] as $dep ) { - $this->mOutput->addTemplate( $dep['title'], $dep['page_id'], $dep['rev_id'] ); - } - } - return array($text,$finalTitle); - } - - function fetchTemplate( $title ) { - $rv = $this->fetchTemplateAndTitle($title); - return $rv[0]; - } - - /** - * Static function to get a template - * Can be overridden via ParserOptions::setTemplateCallback(). - */ - static function statelessFetchTemplate( $title ) { - $text = $skip = false; - $finalTitle = $title; - $deps = array(); - - // Loop to fetch the article, with up to 1 redirect - for ( $i = 0; $i < 2 && is_object( $title ); $i++ ) { - # Give extensions a chance to select the revision instead - $id = false; // Assume current - wfRunHooks( 'BeforeParserFetchTemplateAndtitle', array( false, &$title, &$skip, &$id ) ); - - if( $skip ) { - $text = false; - $deps[] = array( - 'title' => $title, - 'page_id' => $title->getArticleID(), - 'rev_id' => null ); - break; - } - $rev = $id ? Revision::newFromId( $id ) : Revision::newFromTitle( $title ); - $rev_id = $rev ? $rev->getId() : 0; - - $deps[] = array( - 'title' => $title, - 'page_id' => $title->getArticleID(), - 'rev_id' => $rev_id ); - - if( $rev ) { - $text = $rev->getText(); - } elseif( $title->getNamespace() == NS_MEDIAWIKI ) { - global $wgLang; - $message = $wgLang->lcfirst( $title->getText() ); - $text = wfMsgForContentNoTrans( $message ); - if( wfEmptyMsg( $message, $text ) ) { - $text = false; - break; - } - } else { - break; - } - if ( $text === false ) { - break; - } - // Redirect? - $finalTitle = $title; - $title = Title::newFromRedirect( $text ); - } - return array( - 'text' => $text, - 'finalTitle' => $finalTitle, - 'deps' => $deps ); - } - - /** - * Transclude an interwiki link. - */ - function interwikiTransclude( $title, $action ) { - global $wgEnableScaryTranscluding; - - if (!$wgEnableScaryTranscluding) - return wfMsg('scarytranscludedisabled'); - - $url = $title->getFullUrl( "action=$action" ); - - if (strlen($url) > 255) - return wfMsg('scarytranscludetoolong'); - return $this->fetchScaryTemplateMaybeFromCache($url); - } - - function fetchScaryTemplateMaybeFromCache($url) { - global $wgTranscludeCacheExpiry; - $dbr = wfGetDB(DB_SLAVE); - $obj = $dbr->selectRow('transcache', array('tc_time', 'tc_contents'), - array('tc_url' => $url)); - if ($obj) { - $time = $obj->tc_time; - $text = $obj->tc_contents; - if ($time && time() < $time + $wgTranscludeCacheExpiry ) { - return $text; - } - } - - $text = Http::get($url); - if (!$text) - return wfMsg('scarytranscludefailed', $url); - - $dbw = wfGetDB(DB_MASTER); - $dbw->replace('transcache', array('tc_url'), array( - 'tc_url' => $url, - 'tc_time' => time(), - 'tc_contents' => $text)); - return $text; - } - - - /** - * Triple brace replacement -- used for template arguments - * @private - */ - function argSubstitution( $piece, $frame ) { - wfProfileIn( __METHOD__ ); - - $error = false; - $parts = $piece['parts']; - $nameWithSpaces = $frame->expand( $piece['title'] ); - $argName = trim( $nameWithSpaces ); - $object = false; - $text = $frame->getArgument( $argName ); - if ( $text === false && $parts->getLength() > 0 - && ( - $this->ot['html'] - || $this->ot['pre'] - || ( $this->ot['wiki'] && $frame->isTemplate() ) - ) - ) { - # No match in frame, use the supplied default - $object = $parts->item( 0 )->getChildren(); - } - if ( !$this->incrementIncludeSize( 'arg', strlen( $text ) ) ) { - $error = ''; - } - - if ( $text === false && $object === false ) { - # No match anywhere - $object = $frame->virtualBracketedImplode( '{{{', '|', '}}}', $nameWithSpaces, $parts ); - } - if ( $error !== false ) { - $text .= $error; - } - if ( $object !== false ) { - $ret = array( 'object' => $object ); - } else { - $ret = array( 'text' => $text ); - } - - wfProfileOut( __METHOD__ ); - return $ret; - } - - /** - * Return the text to be used for a given extension tag. - * This is the ghost of strip(). - * - * @param array $params Associative array of parameters: - * name PPNode for the tag name - * attr PPNode for unparsed text where tag attributes are thought to be - * attributes Optional associative array of parsed attributes - * inner Contents of extension element - * noClose Original text did not have a close tag - * @param PPFrame $frame - */ - function extensionSubstitution( $params, $frame ) { - global $wgRawHtml, $wgContLang; - - $name = $frame->expand( $params['name'] ); - $attrText = !isset( $params['attr'] ) ? null : $frame->expand( $params['attr'] ); - $content = !isset( $params['inner'] ) ? null : $frame->expand( $params['inner'] ); - - $marker = "{$this->mUniqPrefix}-$name-" . sprintf('%08X', $this->mMarkerIndex++) . $this->mMarkerSuffix; - - if ( $this->ot['html'] ) { - $name = strtolower( $name ); - - $attributes = Sanitizer::decodeTagAttributes( $attrText ); - if ( isset( $params['attributes'] ) ) { - $attributes = $attributes + $params['attributes']; - } - switch ( $name ) { - case 'html': - if( $wgRawHtml ) { - $output = $content; - break; - } else { - throw new MWException( ' extension tag encountered unexpectedly' ); - } - case 'nowiki': - $output = Xml::escapeTagsOnly( $content ); - break; - case 'math': - $output = $wgContLang->armourMath( - MathRenderer::renderMath( $content, $attributes ) ); - break; - case 'gallery': - $output = $this->renderImageGallery( $content, $attributes ); - break; - default: - if( isset( $this->mTagHooks[$name] ) ) { - # Workaround for PHP bug 35229 and similar - if ( !is_callable( $this->mTagHooks[$name] ) ) { - throw new MWException( "Tag hook for $name is not callable\n" ); - } - $output = call_user_func_array( $this->mTagHooks[$name], - array( $content, $attributes, $this ) ); - } else { - throw new MWException( "Invalid call hook $name" ); - } - } - } else { - if ( is_null( $attrText ) ) { - $attrText = ''; - } - if ( isset( $params['attributes'] ) ) { - foreach ( $params['attributes'] as $attrName => $attrValue ) { - $attrText .= ' ' . htmlspecialchars( $attrName ) . '="' . - htmlspecialchars( $attrValue ) . '"'; - } - } - if ( $content === null ) { - $output = "<$name$attrText/>"; - } else { - $close = is_null( $params['close'] ) ? '' : $frame->expand( $params['close'] ); - $output = "<$name$attrText>$content$close"; - } - } - - if ( $name == 'html' || $name == 'nowiki' ) { - $this->mStripState->nowiki->setPair( $marker, $output ); - } else { - $this->mStripState->general->setPair( $marker, $output ); - } - return $marker; - } - - /** - * Increment an include size counter - * - * @param string $type The type of expansion - * @param integer $size The size of the text - * @return boolean False if this inclusion would take it over the maximum, true otherwise - */ - function incrementIncludeSize( $type, $size ) { - if ( $this->mIncludeSizes[$type] + $size > $this->mOptions->getMaxIncludeSize( $type ) ) { - return false; - } else { - $this->mIncludeSizes[$type] += $size; - return true; - } - } - - /** - * Detect __NOGALLERY__ magic word and set a placeholder - */ - function stripNoGallery( &$text ) { - # if the string __NOGALLERY__ (not case-sensitive) occurs in the HTML, - # do not add TOC - $mw = MagicWord::get( 'nogallery' ); - $this->mOutput->mNoGallery = $mw->matchAndRemove( $text ) ; - } - - /** - * Find the first __TOC__ magic word and set a - * placeholder that will then be replaced by the real TOC in - * ->formatHeadings, this works because at this points real - * comments will have already been discarded by the sanitizer. - * - * Any additional __TOC__ magic words left over will be discarded - * as there can only be one TOC on the page. - */ - function stripToc( $text ) { - # if the string __NOTOC__ (not case-sensitive) occurs in the HTML, - # do not add TOC - $mw = MagicWord::get( 'notoc' ); - if( $mw->matchAndRemove( $text ) ) { - $this->mShowToc = false; - } - - $mw = MagicWord::get( 'toc' ); - if( $mw->match( $text ) ) { - $this->mShowToc = true; - $this->mForceTocPosition = true; - - // Set a placeholder. At the end we'll fill it in with the TOC. - $text = $mw->replace( '', $text, 1 ); - - // Only keep the first one. - $text = $mw->replace( '', $text ); - } - return $text; - } - - /** - * This function accomplishes several tasks: - * 1) Auto-number headings if that option is enabled - * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page - * 3) Add a Table of contents on the top for users who have enabled the option - * 4) Auto-anchor headings - * - * It loops through all headlines, collects the necessary data, then splits up the - * string and re-inserts the newly formatted headlines. - * - * @param string $text - * @param boolean $isMain - * @private - */ - function formatHeadings( $text, $isMain=true ) { - global $wgMaxTocLevel, $wgContLang; - - $doNumberHeadings = $this->mOptions->getNumberHeadings(); - if( !$this->mTitle->quickUserCan( 'edit' ) ) { - $showEditLink = 0; - } else { - $showEditLink = $this->mOptions->getEditSection(); - } - - # Inhibit editsection links if requested in the page - $esw =& MagicWord::get( 'noeditsection' ); - if( $esw->matchAndRemove( $text ) ) { - $showEditLink = 0; - } - - # Get all headlines for numbering them and adding funky stuff like [edit] - # links - this is for later, but we need the number of headlines right now - $matches = array(); - $numMatches = preg_match_all( '/[1-6])(?P.*?'.'>)(?P

    .*?)<\/H[1-6] *>/i', $text, $matches ); - - # if there are fewer than 4 headlines in the article, do not show TOC - # unless it's been explicitly enabled. - $enoughToc = $this->mShowToc && - (($numMatches >= 4) || $this->mForceTocPosition); - - # Allow user to stipulate that a page should have a "new section" - # link added via __NEWSECTIONLINK__ - $mw =& MagicWord::get( 'newsectionlink' ); - if( $mw->matchAndRemove( $text ) ) - $this->mOutput->setNewSection( true ); - - # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML, - # override above conditions and always show TOC above first header - $mw =& MagicWord::get( 'forcetoc' ); - if ($mw->matchAndRemove( $text ) ) { - $this->mShowToc = true; - $enoughToc = true; - } - - # We need this to perform operations on the HTML - $sk = $this->mOptions->getSkin(); - - # headline counter - $headlineCount = 0; - $numVisible = 0; - - # Ugh .. the TOC should have neat indentation levels which can be - # passed to the skin functions. These are determined here - $toc = ''; - $full = ''; - $head = array(); - $sublevelCount = array(); - $levelCount = array(); - $toclevel = 0; - $level = 0; - $prevlevel = 0; - $toclevel = 0; - $prevtoclevel = 0; - $markerRegex = "{$this->mUniqPrefix}-h-(\d+)-{$this->mMarkerSuffix}"; - $baseTitleText = $this->mTitle->getPrefixedDBkey(); - $tocraw = array(); - - foreach( $matches[3] as $headline ) { - $isTemplate = false; - $titleText = false; - $sectionIndex = false; - $numbering = ''; - $markerMatches = array(); - if (preg_match("/^$markerRegex/", $headline, $markerMatches)) { - $serial = $markerMatches[1]; - list( $titleText, $sectionIndex ) = $this->mHeadings[$serial]; - $isTemplate = ($titleText != $baseTitleText); - $headline = preg_replace("/^$markerRegex/", "", $headline); - } - - if( $toclevel ) { - $prevlevel = $level; - $prevtoclevel = $toclevel; - } - $level = $matches[1][$headlineCount]; - - if( $doNumberHeadings || $enoughToc ) { - - if ( $level > $prevlevel ) { - # Increase TOC level - $toclevel++; - $sublevelCount[$toclevel] = 0; - if( $toclevel<$wgMaxTocLevel ) { - $prevtoclevel = $toclevel; - $toc .= $sk->tocIndent(); - $numVisible++; - } - } - elseif ( $level < $prevlevel && $toclevel > 1 ) { - # Decrease TOC level, find level to jump to - - if ( $toclevel == 2 && $level <= $levelCount[1] ) { - # Can only go down to level 1 - $toclevel = 1; - } else { - for ($i = $toclevel; $i > 0; $i--) { - if ( $levelCount[$i] == $level ) { - # Found last matching level - $toclevel = $i; - break; - } - elseif ( $levelCount[$i] < $level ) { - # Found first matching level below current level - $toclevel = $i + 1; - break; - } - } - } - if( $toclevel<$wgMaxTocLevel ) { - if($prevtoclevel < $wgMaxTocLevel) { - # Unindent only if the previous toc level was shown :p - $toc .= $sk->tocUnindent( $prevtoclevel - $toclevel ); - } else { - $toc .= $sk->tocLineEnd(); - } - } - } - else { - # No change in level, end TOC line - if( $toclevel<$wgMaxTocLevel ) { - $toc .= $sk->tocLineEnd(); - } - } - - $levelCount[$toclevel] = $level; - - # count number of headlines for each level - @$sublevelCount[$toclevel]++; - $dot = 0; - for( $i = 1; $i <= $toclevel; $i++ ) { - if( !empty( $sublevelCount[$i] ) ) { - if( $dot ) { - $numbering .= '.'; - } - $numbering .= $wgContLang->formatNum( $sublevelCount[$i] ); - $dot = 1; - } - } - } - - # The safe header is a version of the header text safe to use for links - # Avoid insertion of weird stuff like by expanding the relevant sections - $safeHeadline = $this->mStripState->unstripBoth( $headline ); - - # Remove link placeholders by the link text. - # - # turns into - # link text with suffix - $safeHeadline = preg_replace( '//e', - "\$this->mLinkHolders['texts'][\$1]", - $safeHeadline ); - $safeHeadline = preg_replace( '//e', - "\$this->mInterwikiLinkHolders['texts'][\$1]", - $safeHeadline ); - - # Strip out HTML (other than plain and : bug 8393) - $tocline = preg_replace( - array( '#<(?!/?(sup|sub)).*?'.'>#', '#<(/?(sup|sub)).*?'.'>#' ), - array( '', '<$1>'), - $safeHeadline - ); - $tocline = trim( $tocline ); - - # For the anchor, strip out HTML-y stuff period - $safeHeadline = preg_replace( '/<.*?'.'>/', '', $safeHeadline ); - $safeHeadline = trim( $safeHeadline ); - - # Save headline for section edit hint before it's escaped - $headlineHint = $safeHeadline; - $safeHeadline = Sanitizer::escapeId( $safeHeadline ); - $refers[$headlineCount] = $safeHeadline; - - # count how many in assoc. array so we can track dupes in anchors - isset( $refers[$safeHeadline] ) ? $refers[$safeHeadline]++ : $refers[$safeHeadline] = 1; - $refcount[$headlineCount] = $refers[$safeHeadline]; - - # Don't number the heading if it is the only one (looks silly) - if( $doNumberHeadings && count( $matches[3] ) > 1) { - # the two are different if the line contains a link - $headline=$numbering . ' ' . $headline; - } - - # Create the anchor for linking from the TOC to the section - $anchor = $safeHeadline; - if($refcount[$headlineCount] > 1 ) { - $anchor .= '_' . $refcount[$headlineCount]; - } - if( $enoughToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) { - $toc .= $sk->tocLine($anchor, $tocline, $numbering, $toclevel); - $tocraw[] = array( 'toclevel' => $toclevel, 'level' => $level, 'line' => $tocline, 'number' => $numbering ); - } - # give headline the correct tag - if( $showEditLink && $sectionIndex !== false ) { - if( $isTemplate ) { - # Put a T flag in the section identifier, to indicate to extractSections() - # that sections inside should be counted. - $editlink = $sk->editSectionLinkForOther($titleText, "T-$sectionIndex"); - } else { - $editlink = $sk->editSectionLink($this->mTitle, $sectionIndex, $headlineHint); - } - } else { - $editlink = ''; - } - $head[$headlineCount] = $sk->makeHeadline( $level, $matches['attrib'][$headlineCount], $anchor, $headline, $editlink ); - - $headlineCount++; - } - - $this->mOutput->setSections( $tocraw ); - - # Never ever show TOC if no headers - if( $numVisible < 1 ) { - $enoughToc = false; - } - - if( $enoughToc ) { - if( $prevtoclevel > 0 && $prevtoclevel < $wgMaxTocLevel ) { - $toc .= $sk->tocUnindent( $prevtoclevel - 1 ); - } - $toc = $sk->tocList( $toc ); - } - - # split up and insert constructed headlines - - $blocks = preg_split( '/.*?<\/H[1-6]>/i', $text ); - $i = 0; - - foreach( $blocks as $block ) { - if( $showEditLink && $headlineCount > 0 && $i == 0 && $block != "\n" ) { - # This is the [edit] link that appears for the top block of text when - # section editing is enabled - - # Disabled because it broke block formatting - # For example, a bullet point in the top line - # $full .= $sk->editSectionLink(0); - } - $full .= $block; - if( $enoughToc && !$i && $isMain && !$this->mForceTocPosition ) { - # Top anchor now in skin - $full = $full.$toc; - } - - if( !empty( $head[$i] ) ) { - $full .= $head[$i]; - } - $i++; - } - if( $this->mForceTocPosition ) { - return str_replace( '', $toc, $full ); - } else { - return $full; - } - } - - /** - * Transform wiki markup when saving a page by doing \r\n -> \n - * conversion, substitting signatures, {{subst:}} templates, etc. - * - * @param string $text the text to transform - * @param Title &$title the Title object for the current article - * @param User &$user the User object describing the current user - * @param ParserOptions $options parsing options - * @param bool $clearState whether to clear the parser state first - * @return string the altered wiki markup - * @public - */ - function preSaveTransform( $text, &$title, $user, $options, $clearState = true ) { - $this->mOptions = $options; - $this->setTitle( $title ); - $this->setOutputType( self::OT_WIKI ); - - if ( $clearState ) { - $this->clearState(); - } - - $pairs = array( - "\r\n" => "\n", - ); - $text = str_replace( array_keys( $pairs ), array_values( $pairs ), $text ); - $text = $this->pstPass2( $text, $user ); - $text = $this->mStripState->unstripBoth( $text ); - return $text; - } - - /** - * Pre-save transform helper function - * @private - */ - function pstPass2( $text, $user ) { - global $wgContLang, $wgLocaltimezone; - - /* Note: This is the timestamp saved as hardcoded wikitext to - * the database, we use $wgContLang here in order to give - * everyone the same signature and use the default one rather - * than the one selected in each user's preferences. - * - * (see also bug 12815) - */ - $ts = $this->mOptions->getTimestamp(); - $tz = 'UTC'; - if ( isset( $wgLocaltimezone ) ) { - $unixts = wfTimestamp( TS_UNIX, $ts ); - $oldtz = getenv( 'TZ' ); - putenv( 'TZ='.$wgLocaltimezone ); - $ts = date( 'YmdHis', $unixts ); - $tz = date( 'T', $unixts ); # might vary on DST changeover! - putenv( 'TZ='.$oldtz ); - } - $d = $wgContLang->timeanddate( $ts, false, false ) . " ($tz)"; - - # Variable replacement - # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags - $text = $this->replaceVariables( $text ); - - # Signatures - $sigText = $this->getUserSig( $user ); - $text = strtr( $text, array( - '~~~~~' => $d, - '~~~~' => "$sigText $d", - '~~~' => $sigText - ) ); - - # Context links: [[|name]] and [[name (context)|]] - # - global $wgLegalTitleChars; - $tc = "[$wgLegalTitleChars]"; - $nc = '[ _0-9A-Za-z\x80-\xff]'; # Namespaces can use non-ascii! - - $p1 = "/\[\[(:?$nc+:|:|)($tc+?)( \\($tc+\\))\\|]]/"; # [[ns:page (context)|]] - $p3 = "/\[\[(:?$nc+:|:|)($tc+?)( \\($tc+\\)|)(, $tc+|)\\|]]/"; # [[ns:page (context), context|]] - $p2 = "/\[\[\\|($tc+)]]/"; # [[|page]] - - # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]" - $text = preg_replace( $p1, '[[\\1\\2\\3|\\2]]', $text ); - $text = preg_replace( $p3, '[[\\1\\2\\3\\4|\\2]]', $text ); - - $t = $this->mTitle->getText(); - $m = array(); - if ( preg_match( "/^($nc+:|)$tc+?( \\($tc+\\))$/", $t, $m ) ) { - $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text ); - } elseif ( preg_match( "/^($nc+:|)$tc+?(, $tc+|)$/", $t, $m ) && '' != "$m[1]$m[2]" ) { - $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text ); - } else { - # if there's no context, don't bother duplicating the title - $text = preg_replace( $p2, '[[\\1]]', $text ); - } - - # Trim trailing whitespace - $text = rtrim( $text ); - - return $text; - } - - /** - * Fetch the user's signature text, if any, and normalize to - * validated, ready-to-insert wikitext. - * - * @param User $user - * @return string - * @private - */ - function getUserSig( &$user ) { - global $wgMaxSigChars; - - $username = $user->getName(); - $nickname = $user->getOption( 'nickname' ); - $nickname = $nickname === '' ? $username : $nickname; - - if( mb_strlen( $nickname ) > $wgMaxSigChars ) { - $nickname = $username; - wfDebug( __METHOD__ . ": $username has overlong signature.\n" ); - } elseif( $user->getBoolOption( 'fancysig' ) !== false ) { - # Sig. might contain markup; validate this - if( $this->validateSig( $nickname ) !== false ) { - # Validated; clean up (if needed) and return it - return $this->cleanSig( $nickname, true ); - } else { - # Failed to validate; fall back to the default - $nickname = $username; - wfDebug( "Parser::getUserSig: $username has bad XML tags in signature.\n" ); - } - } - - // Make sure nickname doesnt get a sig in a sig - $nickname = $this->cleanSigInSig( $nickname ); - - # If we're still here, make it a link to the user page - $userText = wfEscapeWikiText( $username ); - $nickText = wfEscapeWikiText( $nickname ); - if ( $user->isAnon() ) { - return wfMsgExt( 'signature-anon', array( 'content', 'parsemag' ), $userText, $nickText ); - } else { - return wfMsgExt( 'signature', array( 'content', 'parsemag' ), $userText, $nickText ); - } - } - - /** - * Check that the user's signature contains no bad XML - * - * @param string $text - * @return mixed An expanded string, or false if invalid. - */ - function validateSig( $text ) { - return( wfIsWellFormedXmlFragment( $text ) ? $text : false ); - } - - /** - * Clean up signature text - * - * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig - * 2) Substitute all transclusions - * - * @param string $text - * @param $parsing Whether we're cleaning (preferences save) or parsing - * @return string Signature text - */ - function cleanSig( $text, $parsing = false ) { - if ( !$parsing ) { - global $wgTitle; - $this->clearState(); - $this->setTitle( $wgTitle ); - $this->mOptions = new ParserOptions; - $this->setOutputType = self::OT_PREPROCESS; - } - - # FIXME: regex doesn't respect extension tags or nowiki - # => Move this logic to braceSubstitution() - $substWord = MagicWord::get( 'subst' ); - $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase(); - $substText = '{{' . $substWord->getSynonym( 0 ); - - $text = preg_replace( $substRegex, $substText, $text ); - $text = $this->cleanSigInSig( $text ); - $dom = $this->preprocessToDom( $text ); - $frame = $this->getPreprocessor()->newFrame(); - $text = $frame->expand( $dom ); - - if ( !$parsing ) { - $text = $this->mStripState->unstripBoth( $text ); - } - - return $text; - } - - /** - * Strip ~~~, ~~~~ and ~~~~~ out of signatures - * @param string $text - * @return string Signature text with /~{3,5}/ removed - */ - function cleanSigInSig( $text ) { - $text = preg_replace( '/~{3,5}/', '', $text ); - return $text; - } - - /** - * Set up some variables which are usually set up in parse() - * so that an external function can call some class members with confidence - * @public - */ - function startExternalParse( &$title, $options, $outputType, $clearState = true ) { - $this->setTitle( $title ); - $this->mOptions = $options; - $this->setOutputType( $outputType ); - if ( $clearState ) { - $this->clearState(); - } - } - - /** - * Wrapper for preprocess() - * - * @param string $text the text to preprocess - * @param ParserOptions $options options - * @return string - * @public - */ - function transformMsg( $text, $options ) { - global $wgTitle; - static $executing = false; - - $fname = "Parser::transformMsg"; - - # Guard against infinite recursion - if ( $executing ) { - return $text; - } - $executing = true; - - wfProfileIn($fname); - $text = $this->preprocess( $text, $wgTitle, $options ); - - $executing = false; - wfProfileOut($fname); - return $text; - } - - /** - * Create an HTML-style tag, e.g. special text - * The callback should have the following form: - * function myParserHook( $text, $params, &$parser ) { ... } - * - * Transform and return $text. Use $parser for any required context, e.g. use - * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions - * - * @public - * - * @param mixed $tag The tag to use, e.g. 'hook' for - * @param mixed $callback The callback function (and object) to use for the tag - * - * @return The old value of the mTagHooks array associated with the hook - */ - function setHook( $tag, $callback ) { - $tag = strtolower( $tag ); - $oldVal = isset( $this->mTagHooks[$tag] ) ? $this->mTagHooks[$tag] : null; - $this->mTagHooks[$tag] = $callback; - $this->mStripList[] = $tag; - - return $oldVal; - } - - function setTransparentTagHook( $tag, $callback ) { - $tag = strtolower( $tag ); - $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null; - $this->mTransparentTagHooks[$tag] = $callback; - - return $oldVal; - } - - /** - * Remove all tag hooks - */ - function clearTagHooks() { - $this->mTagHooks = array(); - $this->mStripList = $this->mDefaultStripList; - } - - /** - * Create a function, e.g. {{sum:1|2|3}} - * The callback function should have the form: - * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... } - * - * The callback may either return the text result of the function, or an array with the text - * in element 0, and a number of flags in the other elements. The names of the flags are - * specified in the keys. Valid flags are: - * found The text returned is valid, stop processing the template. This - * is on by default. - * nowiki Wiki markup in the return value should be escaped - * isHTML The returned text is HTML, armour it against wikitext transformation - * - * @public - * - * @param string $id The magic word ID - * @param mixed $callback The callback function (and object) to use - * @param integer $flags a combination of the following flags: - * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}} - * - * @return The old callback function for this name, if any - */ - function setFunctionHook( $id, $callback, $flags = 0 ) { - $oldVal = isset( $this->mFunctionHooks[$id] ) ? $this->mFunctionHooks[$id][0] : null; - $this->mFunctionHooks[$id] = array( $callback, $flags ); - - # Add to function cache - $mw = MagicWord::get( $id ); - if( !$mw ) - throw new MWException( 'Parser::setFunctionHook() expecting a magic word identifier.' ); - - $synonyms = $mw->getSynonyms(); - $sensitive = intval( $mw->isCaseSensitive() ); - - foreach ( $synonyms as $syn ) { - # Case - if ( !$sensitive ) { - $syn = strtolower( $syn ); - } - # Add leading hash - if ( !( $flags & SFH_NO_HASH ) ) { - $syn = '#' . $syn; - } - # Remove trailing colon - if ( substr( $syn, -1, 1 ) == ':' ) { - $syn = substr( $syn, 0, -1 ); - } - $this->mFunctionSynonyms[$sensitive][$syn] = $id; - } - return $oldVal; - } - - /** - * Get all registered function hook identifiers - * - * @return array - */ - function getFunctionHooks() { - return array_keys( $this->mFunctionHooks ); - } - - /** - * Replace link placeholders with actual links, in the buffer - * Placeholders created in Skin::makeLinkObj() - * Returns an array of link CSS classes, indexed by PDBK. - * $options is a bit field, RLH_FOR_UPDATE to select for update - */ - function replaceLinkHolders( &$text, $options = 0 ) { - global $wgUser; - global $wgContLang; - - $fname = 'Parser::replaceLinkHolders'; - wfProfileIn( $fname ); - - $pdbks = array(); - $colours = array(); - $linkcolour_ids = array(); - $sk = $this->mOptions->getSkin(); - $linkCache =& LinkCache::singleton(); - - if ( !empty( $this->mLinkHolders['namespaces'] ) ) { - wfProfileIn( $fname.'-check' ); - $dbr = wfGetDB( DB_SLAVE ); - $page = $dbr->tableName( 'page' ); - $threshold = $wgUser->getOption('stubthreshold'); - - # Sort by namespace - asort( $this->mLinkHolders['namespaces'] ); - - # Generate query - $query = false; - $current = null; - foreach ( $this->mLinkHolders['namespaces'] as $key => $ns ) { - # Make title object - $title = $this->mLinkHolders['titles'][$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 a static known link, e.g. interwiki - if ( $title->isAlwaysKnown() ) { - $colours[$pdbk] = ''; - } elseif ( ( $id = $linkCache->getGoodLinkID( $pdbk ) ) != 0 ) { - $colours[$pdbk] = ''; - $this->mOutput->addLink( $title, $id ); - } elseif ( $linkCache->isBadLink( $pdbk ) ) { - $colours[$pdbk] = 'new'; - } elseif ( $title->getNamespace() == NS_SPECIAL && !SpecialPage::exists( $pdbk ) ) { - $colours[$pdbk] = 'new'; - } else { - # Not in the link cache, add it to the query - if ( !isset( $current ) ) { - $current = $ns; - $query = "SELECT page_id, page_namespace, page_title, page_is_redirect"; - if ( $threshold > 0 ) { - $query .= ', page_len'; - } - $query .= " FROM $page WHERE (page_namespace=$ns AND page_title IN("; - } elseif ( $current != $ns ) { - $current = $ns; - $query .= ")) OR (page_namespace=$ns AND page_title IN("; - } else { - $query .= ', '; - } - - $query .= $dbr->addQuotes( $this->mLinkHolders['dbkeys'][$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 - while ( $s = $dbr->fetchObject($res) ) { - $title = Title::makeTitle( $s->page_namespace, $s->page_title ); - $pdbk = $title->getPrefixedDBkey(); - $linkCache->addGoodLinkObj( $s->page_id, $title ); - $this->mOutput->addLink( $title, $s->page_id ); - $colours[$pdbk] = $sk->getLinkColour( $s, $threshold ); - //add id to the extension todolist - $linkcolour_ids[$s->page_id] = $pdbk; - } - //pass an array of page_ids to an extension - wfRunHooks( 'GetLinkColours', array( $linkcolour_ids, &$colours ) ); - } - wfProfileOut( $fname.'-check' ); - - # Do a second query for different language variants of links and categories - if($wgContLang->hasVariants()){ - $linkBatch = new LinkBatch(); - $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 ) { - $title = $this->mLinkHolders['titles'][$key]; - if ( is_null( $title ) ) - continue; - - $pdbk = $title->getPrefixedDBkey(); - $titleText = $title->getText(); - - // generate all variants of the link title text - $allTextVariants = $wgContLang->convertLinkToAllVariants($titleText); - - // if link was not found (in first query), add all variants to query - if ( !isset($colours[$pdbk]) ){ - foreach($allTextVariants as $textVariant){ - if($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 $variant){ - if($variant != $category){ - $variantTitle = Title::newFromDBkey( Title::makeName(NS_CATEGORY,$variant) ); - if(is_null($variantTitle)) continue; - $linkBatch->addObj( $variantTitle ); - $categoryMap[$variant] = $category; - } - } - } - - - if(!$linkBatch->isEmpty()){ - // construct query - $titleClause = $linkBatch->constructSet('page', $dbr); - - $variantQuery = "SELECT page_id, page_namespace, page_title, page_is_redirect"; - if ( $threshold > 0 ) { - $variantQuery .= ', page_len'; - } - - $variantQuery .= " FROM $page WHERE $titleClause"; - if ( $options & RLH_FOR_UPDATE ) { - $variantQuery .= ' FOR UPDATE'; - } - - $varRes = $dbr->query( $variantQuery, $fname ); - - // for each found variants, figure out link holders and replace - while ( $s = $dbr->fetchObject($varRes) ) { - - $variantTitle = Title::makeTitle( $s->page_namespace, $s->page_title ); - $varPdbk = $variantTitle->getPrefixedDBkey(); - $vardbk = $variantTitle->getDBkey(); - - $holderKeys = array(); - if(isset($variantMap[$varPdbk])){ - $holderKeys = $variantMap[$varPdbk]; - $linkCache->addGoodLinkObj( $s->page_id, $variantTitle ); - $this->mOutput->addLink( $variantTitle, $s->page_id ); - } - - // loop over link holders - foreach($holderKeys as $key){ - $title = $this->mLinkHolders['titles'][$key]; - if ( is_null( $title ) ) continue; - - $pdbk = $title->getPrefixedDBkey(); - - if(!isset($colours[$pdbk])){ - // found link in some of the variants, replace the link holder data - $this->mLinkHolders['titles'][$key] = $variantTitle; - $this->mLinkHolders['dbkeys'][$key] = $variantTitle->getDBkey(); - - // set pdbk and colour - $pdbks[$key] = $varPdbk; - $colours[$varPdbk] = $sk->getLinkColour( $s, $threshold ); - $linkcolour_ids[$s->page_id] = $pdbk; - } - wfRunHooks( 'GetLinkColours', array( $linkcolour_ids, &$colours ) ); - } - - // 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' ); - $replacePairs = array(); - foreach ( $this->mLinkHolders['namespaces'] as $key => $ns ) { - $pdbk = $pdbks[$key]; - $searchkey = ""; - $title = $this->mLinkHolders['titles'][$key]; - if ( !isset( $colours[$pdbk] ) || $colours[$pdbk] == 'new' ) { - $linkCache->addBadLinkObj( $title ); - $colours[$pdbk] = 'new'; - $this->mOutput->addLink( $title, 0 ); - $replacePairs[$searchkey] = $sk->makeBrokenLinkObj( $title, - $this->mLinkHolders['texts'][$key], - $this->mLinkHolders['queries'][$key] ); - } else { - $replacePairs[$searchkey] = $sk->makeColouredLinkObj( $title, $colours[$pdbk], - $this->mLinkHolders['texts'][$key], - $this->mLinkHolders['queries'][$key] ); - } - } - $replacer = new HashtableReplacer( $replacePairs, 1 ); - wfProfileOut( $fname.'-construct' ); - - # Do the thing - wfProfileIn( $fname.'-replace' ); - $text = preg_replace_callback( - '/()/', - $replacer->cb(), - $text); - - wfProfileOut( $fname.'-replace' ); - } - - # Now process interwiki link holders - # This is quite a bit simpler than internal links - if ( !empty( $this->mInterwikiLinkHolders['texts'] ) ) { - wfProfileIn( $fname.'-interwiki' ); - # Make interwiki link HTML - $replacePairs = array(); - foreach( $this->mInterwikiLinkHolders['texts'] as $key => $link ) { - $title = $this->mInterwikiLinkHolders['titles'][$key]; - $replacePairs[$key] = $sk->makeLinkObj( $title, $link ); - } - $replacer = new HashtableReplacer( $replacePairs, 1 ); - - $text = preg_replace_callback( - '//', - $replacer->cb(), - $text ); - wfProfileOut( $fname.'-interwiki' ); - } - - wfProfileOut( $fname ); - return $colours; - } - - /** - * Replace link placeholders with plain text of links - * (not HTML-formatted). - * @param string $text - * @return string - */ - function replaceLinkHoldersText( $text ) { - $fname = 'Parser::replaceLinkHoldersText'; - wfProfileIn( $fname ); - - $text = preg_replace_callback( - '//', - array( &$this, 'replaceLinkHoldersTextCallback' ), - $text ); - - wfProfileOut( $fname ); - return $text; - } - - /** - * @param array $matches - * @return string - * @private - */ - function replaceLinkHoldersTextCallback( $matches ) { - $type = $matches[1]; - $key = $matches[2]; - if( $type == 'LINK' ) { - if( isset( $this->mLinkHolders['texts'][$key] ) ) { - return $this->mLinkHolders['texts'][$key]; - } - } elseif( $type == 'IWLINK' ) { - if( isset( $this->mInterwikiLinkHolders['texts'][$key] ) ) { - return $this->mInterwikiLinkHolders['texts'][$key]; - } - } - return $matches[0]; - } - - /** - * Tag hook handler for 'pre'. - */ - function renderPreTag( $text, $attribs ) { - // Backwards-compatibility hack - $content = StringUtils::delimiterReplace( '', '', '$1', $text, 'i' ); - - $attribs = Sanitizer::validateTagAttributes( $attribs, 'pre' ); - return wfOpenElement( 'pre', $attribs ) . - Xml::escapeTagsOnly( $content ) . - '
    '; - } - - /** - * 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, $params ) { - $ig = new ImageGallery(); - $ig->setContextTitle( $this->mTitle ); - $ig->setShowBytes( false ); - $ig->setShowFilename( false ); - $ig->setParser( $this ); - $ig->setHideBadImages(); - $ig->setAttributes( Sanitizer::validateTagAttributes( $params, 'table' ) ); - $ig->useSkin( $this->mOptions->getSkin() ); - $ig->mRevisionId = $this->mRevisionId; - - if( isset( $params['caption'] ) ) { - $caption = $params['caption']; - $caption = htmlspecialchars( $caption ); - $caption = $this->replaceInternalLinks( $caption ); - $ig->setCaptionHtml( $caption ); - } - if( isset( $params['perrow'] ) ) { - $ig->setPerRow( $params['perrow'] ); - } - if( isset( $params['widths'] ) ) { - $ig->setWidths( $params['widths'] ); - } - if( isset( $params['heights'] ) ) { - $ig->setHeights( $params['heights'] ); - } - - wfRunHooks( 'BeforeParserrenderImageGallery', array( &$this, &$ig ) ); - - $lines = explode( "\n", $text ); - foreach ( $lines as $line ) { - # match lines like these: - # Image:someimage.jpg|This is some image - $matches = array(); - preg_match( "/^([^|]+)(\\|(.*))?$/", $line, $matches ); - # Skip empty lines - if ( count( $matches ) == 0 ) { - continue; - } - $tp = Title::newFromText( $matches[1] ); - $nt =& $tp; - if( is_null( $nt ) ) { - # Bogus title. Ignore these so we don't bomb out later. - continue; - } - if ( isset( $matches[3] ) ) { - $label = $matches[3]; - } else { - $label = ''; - } - - $html = $this->recursiveTagParse( trim( $label ) ); - - $ig->add( $nt, $html ); - - # Only add real images (bug #5586) - if ( $nt->getNamespace() == NS_IMAGE ) { - $this->mOutput->addImage( $nt->getDBkey() ); - } - } - return $ig->toHTML(); - } - - function getImageParams( $handler ) { - if ( $handler ) { - $handlerClass = get_class( $handler ); - } else { - $handlerClass = ''; - } - if ( !isset( $this->mImageParams[$handlerClass] ) ) { - // Initialise static lists - static $internalParamNames = array( - 'horizAlign' => array( 'left', 'right', 'center', 'none' ), - 'vertAlign' => array( 'baseline', 'sub', 'super', 'top', 'text-top', 'middle', - 'bottom', 'text-bottom' ), - 'frame' => array( 'thumbnail', 'manualthumb', 'framed', 'frameless', - 'upright', 'border' ), - ); - static $internalParamMap; - if ( !$internalParamMap ) { - $internalParamMap = array(); - foreach ( $internalParamNames as $type => $names ) { - foreach ( $names as $name ) { - $magicName = str_replace( '-', '_', "img_$name" ); - $internalParamMap[$magicName] = array( $type, $name ); - } - } - } - - // Add handler params - $paramMap = $internalParamMap; - if ( $handler ) { - $handlerParamMap = $handler->getParamMap(); - foreach ( $handlerParamMap as $magic => $paramName ) { - $paramMap[$magic] = array( 'handler', $paramName ); - } - } - $this->mImageParams[$handlerClass] = $paramMap; - $this->mImageParamsMagicArray[$handlerClass] = new MagicWordArray( array_keys( $paramMap ) ); - } - return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] ); - } - - /** - * Parse image options text and use it to make an image - */ - function makeImage( $title, $options ) { - # @TODO: let the MediaHandler specify its transform parameters - # - # Check if the options text is of the form "options|alt text" - # Options are: - # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang - # * left no resizing, just left align. label is used for alt= only - # * right same, but right aligned - # * none same, but not aligned - # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox - # * center center the image - # * framed Keep original image size, no magnify-button. - # * frameless like 'thumb' but without a frame. Keeps user preferences for width - # * upright reduce width for upright images, rounded to full __0 px - # * border draw a 1px border around the image - # vertical-align values (no % or length right now): - # * baseline - # * sub - # * super - # * top - # * text-top - # * middle - # * bottom - # * text-bottom - - $parts = array_map( 'trim', explode( '|', $options) ); - $sk = $this->mOptions->getSkin(); - - # Give extensions a chance to select the file revision for us - $skip = $time = false; - wfRunHooks( 'BeforeParserMakeImageLinkObj', array( &$this, &$title, &$skip, &$time ) ); - - if ( $skip ) { - return $sk->makeLinkObj( $title ); - } - - # Get parameter map - $file = wfFindFile( $title, $time ); - $handler = $file ? $file->getHandler() : false; - - list( $paramMap, $mwArray ) = $this->getImageParams( $handler ); - - # Process the input parameters - $caption = ''; - $params = array( 'frame' => array(), 'handler' => array(), - 'horizAlign' => array(), 'vertAlign' => array() ); - foreach( $parts as $part ) { - list( $magicName, $value ) = $mwArray->matchVariableStartToEnd( $part ); - if ( isset( $paramMap[$magicName] ) ) { - list( $type, $paramName ) = $paramMap[$magicName]; - $params[$type][$paramName] = $value; - - // Special case; width and height come in one variable together - if( $type == 'handler' && $paramName == 'width' ) { - $m = array(); - if ( preg_match( '/^([0-9]*)x([0-9]*)$/', $value, $m ) ) { - $params[$type]['width'] = intval( $m[1] ); - $params[$type]['height'] = intval( $m[2] ); - } else { - $params[$type]['width'] = intval( $value ); - } - } - } else { - $caption = $part; - } - } - - # Process alignment parameters - if ( $params['horizAlign'] ) { - $params['frame']['align'] = key( $params['horizAlign'] ); - } - if ( $params['vertAlign'] ) { - $params['frame']['valign'] = key( $params['vertAlign'] ); - } - - # Validate the handler parameters - if ( $handler ) { - foreach ( $params['handler'] as $name => $value ) { - if ( !$handler->validateParam( $name, $value ) ) { - unset( $params['handler'][$name] ); - } - } - } - - # Strip bad stuff out of the alt text - $alt = $this->replaceLinkHoldersText( $caption ); - - # 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->mStripState->unstripBoth( $alt ); - $alt = Sanitizer::stripAllTags( $alt ); - - $params['frame']['alt'] = $alt; - $params['frame']['caption'] = $caption; - - # Linker does the rest - $ret = $sk->makeImageLink2( $title, $file, $params['frame'], $params['handler'] ); - - # Give the handler a chance to modify the parser object - if ( $handler ) { - $handler->parserTransformHook( $this, $file ); - } - - return $ret; - } - - /** - * Set a flag in the output object indicating that the content is dynamic and - * shouldn't be cached. - */ - function disableCache() { - wfDebug( "Parser output marked as uncacheable.\n" ); - $this->mOutput->mCacheTime = -1; - } - - /**#@+ - * Callback from the Sanitizer for expanding items found in HTML attribute - * values, so they can be safely tested and escaped. - * @param string $text - * @param PPFrame $frame - * @return string - * @private - */ - function attributeStripCallback( &$text, $frame = false ) { - $text = $this->replaceVariables( $text, $frame ); - $text = $this->mStripState->unstripBoth( $text ); - return $text; - } - - /**#@-*/ - - /**#@+ - * Accessor/mutator - */ - function Title( $x = NULL ) { return wfSetVar( $this->mTitle, $x ); } - function Options( $x = NULL ) { return wfSetVar( $this->mOptions, $x ); } - function OutputType( $x = NULL ) { return wfSetVar( $this->mOutputType, $x ); } - /**#@-*/ - - /**#@+ - * Accessor - */ - function getTags() { return array_merge( array_keys($this->mTransparentTagHooks), array_keys( $this->mTagHooks ) ); } - /**#@-*/ - - - /** - * Break wikitext input into sections, and either pull or replace - * some particular section's text. - * - * External callers should use the getSection and replaceSection methods. - * - * @param string $text Page wikitext - * @param string $section A section identifier string of the form: - * - - ... -
    - * - * Currently the only recognised flag is "T", which means the target section number - * was derived during a template inclusion parse, in other words this is a template - * section edit link. If no flags are given, it was an ordinary section edit link. - * This flag is required to avoid a section numbering mismatch when a section is - * enclosed by (bug 6563). - * - * The section number 0 pulls the text before the first heading; other numbers will - * pull the given section along with its lower-level subsections. If the section is - * not found, $mode=get will return $newtext, and $mode=replace will return $text. - * - * @param string $mode One of "get" or "replace" - * @param string $newText Replacement text for section data. - * @return string for "get", the extracted section text. - * for "replace", the whole page with the section replaced. - */ - private function extractSections( $text, $section, $mode, $newText='' ) { - global $wgTitle; - $this->clearState(); - $this->setTitle( $wgTitle ); // not generally used but removes an ugly failure mode - $this->mOptions = new ParserOptions; - $this->setOutputType( self::OT_WIKI ); - $outText = ''; - $frame = $this->getPreprocessor()->newFrame(); - - // Process section extraction flags - $flags = 0; - $sectionParts = explode( '-', $section ); - $sectionIndex = array_pop( $sectionParts ); - foreach ( $sectionParts as $part ) { - if ( $part == 'T' ) { - $flags |= self::PTD_FOR_INCLUSION; - } - } - // Preprocess the text - $root = $this->preprocessToDom( $text, $flags ); - - // nodes indicate section breaks - // They can only occur at the top level, so we can find them by iterating the root's children - $node = $root->getFirstChild(); - - // Find the target section - if ( $sectionIndex == 0 ) { - // Section zero doesn't nest, level=big - $targetLevel = 1000; - } else { - while ( $node ) { - if ( $node->getName() == 'h' ) { - $bits = $node->splitHeading(); - if ( $bits['i'] == $sectionIndex ) { - $targetLevel = $bits['level']; - break; - } - } - if ( $mode == 'replace' ) { - $outText .= $frame->expand( $node, PPFrame::RECOVER_ORIG ); - } - $node = $node->getNextSibling(); - } - } - - if ( !$node ) { - // Not found - if ( $mode == 'get' ) { - return $newText; - } else { - return $text; - } - } - - // Find the end of the section, including nested sections - do { - if ( $node->getName() == 'h' ) { - $bits = $node->splitHeading(); - $curLevel = $bits['level']; - if ( $bits['i'] != $sectionIndex && $curLevel <= $targetLevel ) { - break; - } - } - if ( $mode == 'get' ) { - $outText .= $frame->expand( $node, PPFrame::RECOVER_ORIG ); - } - $node = $node->getNextSibling(); - } while ( $node ); - - // Write out the remainder (in replace mode only) - if ( $mode == 'replace' ) { - // Output the replacement text - // Add two newlines on -- trailing whitespace in $newText is conventionally - // stripped by the editor, so we need both newlines to restore the paragraph gap - $outText .= $newText . "\n\n"; - while ( $node ) { - $outText .= $frame->expand( $node, PPFrame::RECOVER_ORIG ); - $node = $node->getNextSibling(); - } - } - - if ( is_string( $outText ) ) { - // Re-insert stripped tags - $outText = trim( $this->mStripState->unstripBoth( $outText ) ); - } - - return $outText; - } - - /** - * This function returns the text of a section, specified by a number ($section). - * A section is text under a heading like == Heading == or \Heading\, or - * the first section before any such heading (section 0). - * - * If a section contains subsections, these are also returned. - * - * @param string $text text to look in - * @param string $section section identifier - * @param string $deftext default to return if section is not found - * @return string text of the requested section - */ - public function getSection( $text, $section, $deftext='' ) { - return $this->extractSections( $text, $section, "get", $deftext ); - } - - public function replaceSection( $oldtext, $section, $text ) { - 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; - } - - /** - * Mutator for $mDefaultSort - * - * @param $sort New value - */ - public function setDefaultSort( $sort ) { - $this->mDefaultSort = $sort; - } - - /** - * Accessor for $mDefaultSort - * Will use the title/prefixed title if none is set - * - * @return string - */ - public function getDefaultSort() { - if( $this->mDefaultSort !== false ) { - return $this->mDefaultSort; - } else { - return $this->mTitle->getNamespace() == NS_CATEGORY - ? $this->mTitle->getText() - : $this->mTitle->getPrefixedText(); - } - } - - /** - * Try to guess the section anchor name based on a wikitext fragment - * presumably extracted from a heading, for example "Header" from - * "== Header ==". - */ - public function guessSectionNameFromWikiText( $text ) { - # Strip out wikitext links(they break the anchor) - $text = $this->stripSectionName( $text ); - $headline = Sanitizer::decodeCharReferences( $text ); - # strip out HTML - $headline = StringUtils::delimiterReplace( '<', '>', '', $headline ); - $headline = trim( $headline ); - $sectionanchor = '#' . urlencode( str_replace( ' ', '_', $headline ) ); - $replacearray = array( - '%3A' => ':', - '%' => '.' - ); - return str_replace( - array_keys( $replacearray ), - array_values( $replacearray ), - $sectionanchor ); - } - - /** - * Strips a text string of wikitext for use in a section anchor - * - * Accepts a text string and then removes all wikitext from the - * string and leaves only the resultant text (i.e. the result of - * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of - * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended - * to create valid section anchors by mimicing the output of the - * parser when headings are parsed. - * - * @param $text string Text string to be stripped of wikitext - * for use in a Section anchor - * @return Filtered text string - */ - public function stripSectionName( $text ) { - # Strip internal link markup - $text = preg_replace('/\[\[:?([^[|]+)\|([^[]+)\]\]/','$2',$text); - $text = preg_replace('/\[\[:?([^[]+)\|?\]\]/','$1',$text); - - # Strip external link markup (FIXME: Not Tolerant to blank link text - # I.E. [http://www.mediawiki.org] will render as [1] or something depending - # on how many empty links there are on the page - need to figure that out. - $text = preg_replace('/\[(?:' . wfUrlProtocols() . ')([^ ]+?) ([^[]+)\]/','$2',$text); - - # Parse wikitext quotes (italics & bold) - $text = $this->doQuotes($text); - - # Strip HTML tags - $text = StringUtils::delimiterReplace( '<', '>', '', $text ); - return $text; - } - - function srvus( $text ) { - return $this->testSrvus( $text, $this->mOutputType ); - } - - /** - * strip/replaceVariables/unstrip for preprocessor regression testing - */ - function testSrvus( $text, $title, $options, $outputType = self::OT_HTML ) { - $this->clearState(); - if ( ! ( $title instanceof Title ) ) { - $title = Title::newFromText( $title ); - } - $this->mTitle = $title; - $this->mOptions = $options; - $this->setOutputType( $outputType ); - $text = $this->replaceVariables( $text ); - $text = $this->mStripState->unstripBoth( $text ); - $text = Sanitizer::removeHTMLtags( $text ); - return $text; - } - - function testPst( $text, $title, $options ) { - global $wgUser; - if ( ! ( $title instanceof Title ) ) { - $title = Title::newFromText( $title ); - } - return $this->preSaveTransform( $text, $title, $wgUser, $options ); - } - - function testPreprocess( $text, $title, $options ) { - if ( ! ( $title instanceof Title ) ) { - $title = Title::newFromText( $title ); - } - return $this->testSrvus( $text, $title, $options, self::OT_PREPROCESS ); - } - - function markerSkipCallback( $s, $callback ) { - $i = 0; - $out = ''; - while ( $i < strlen( $s ) ) { - $markerStart = strpos( $s, $this->mUniqPrefix, $i ); - if ( $markerStart === false ) { - $out .= call_user_func( $callback, substr( $s, $i ) ); - break; - } else { - $out .= call_user_func( $callback, substr( $s, $i, $markerStart - $i ) ); - $markerEnd = strpos( $s, $this->mMarkerSuffix, $markerStart ); - if ( $markerEnd === false ) { - $out .= substr( $s, $markerStart ); - break; - } else { - $markerEnd += strlen( $this->mMarkerSuffix ); - $out .= substr( $s, $markerStart, $markerEnd - $markerStart ); - $i = $markerEnd; - } - } - } - return $out; - } -} - -/** - * @todo document, briefly. - * @addtogroup Parser - */ -class StripState { - var $general, $nowiki; - - function __construct() { - $this->general = new ReplacementArray; - $this->nowiki = new ReplacementArray; - } - - function unstripGeneral( $text ) { - wfProfileIn( __METHOD__ ); - do { - $oldText = $text; - $text = $this->general->replace( $text ); - } while ( $text != $oldText ); - wfProfileOut( __METHOD__ ); - return $text; - } - - function unstripNoWiki( $text ) { - wfProfileIn( __METHOD__ ); - do { - $oldText = $text; - $text = $this->nowiki->replace( $text ); - } while ( $text != $oldText ); - wfProfileOut( __METHOD__ ); - return $text; - } - - function unstripBoth( $text ) { - wfProfileIn( __METHOD__ ); - do { - $oldText = $text; - $text = $this->general->replace( $text ); - $text = $this->nowiki->replace( $text ); - } while ( $text != $oldText ); - wfProfileOut( __METHOD__ ); - return $text; - } -} - -/** - * @todo document, briefly. - * @addtogroup Parser - */ -class OnlyIncludeReplacer { - var $output = ''; - - function replace( $matches ) { - if ( substr( $matches[1], -1 ) == "\n" ) { - $this->output .= substr( $matches[1], 0, -1 ); - } else { - $this->output .= $matches[1]; - } - } -} - diff --git a/includes/ParserCache.php b/includes/ParserCache.php deleted file mode 100644 index 129b7132..00000000 --- a/includes/ParserCache.php +++ /dev/null @@ -1,119 +0,0 @@ -mMemc =& $memCached; - } - - function getKey( &$article, &$user ) { - global $action; - $hash = $user->getPageRenderingHash(); - if( !$article->mTitle->quickUserCan( 'edit' ) ) { - // section edit links are suppressed even if the user has them on - $edit = '!edit=0'; - } else { - $edit = ''; - } - $pageid = intval( $article->getID() ); - $renderkey = (int)($action == 'render'); - $key = wfMemcKey( 'pcache', 'idhash', "$pageid-$renderkey!$hash$edit" ); - return $key; - } - - function getETag( &$article, &$user ) { - return 'W/"' . $this->getKey($article, $user) . "--" . $article->mTouched. '"'; - } - - function get( &$article, &$user ) { - global $wgCacheEpoch; - $fname = 'ParserCache::get'; - wfProfileIn( $fname ); - - $key = $this->getKey( $article, $user ); - - wfDebug( "Trying parser cache $key\n" ); - $value = $this->mMemc->get( $key ); - if ( is_object( $value ) ) { - wfDebug( "Found.\n" ); - # Delete if article has changed since the cache was made - $canCache = $article->checkTouched(); - $cacheTime = $value->getCacheTime(); - $touched = $article->mTouched; - if ( !$canCache || $value->expired( $touched ) ) { - if ( !$canCache ) { - wfIncrStats( "pcache_miss_invalid" ); - wfDebug( "Invalid cached redirect, touched $touched, epoch $wgCacheEpoch, cached $cacheTime\n" ); - } else { - wfIncrStats( "pcache_miss_expired" ); - wfDebug( "Key expired, touched $touched, epoch $wgCacheEpoch, cached $cacheTime\n" ); - } - $this->mMemc->delete( $key ); - $value = false; - } else { - if ( isset( $value->mTimestamp ) ) { - $article->mTimestamp = $value->mTimestamp; - } - wfIncrStats( "pcache_hit" ); - } - } else { - wfDebug( "Parser cache miss.\n" ); - wfIncrStats( "pcache_miss_absent" ); - $value = false; - } - - wfProfileOut( $fname ); - return $value; - } - - function save( $parserOutput, &$article, &$user ){ - global $wgParserCacheExpireTime; - $key = $this->getKey( $article, $user ); - - if( $parserOutput->getCacheTime() != -1 ) { - - $now = wfTimestampNow(); - $parserOutput->setCacheTime( $now ); - - // Save the timestamp so that we don't have to load the revision row on view - $parserOutput->mTimestamp = $article->getTimestamp(); - - $parserOutput->mText .= "\n\n"; - wfDebug( "Saved in parser cache with key $key and timestamp $now\n" ); - - if( $parserOutput->containsOldMagic() ){ - $expire = 3600; # 1 hour - } else { - $expire = $wgParserCacheExpireTime; - } - $this->mMemc->set( $key, $parserOutput, $expire ); - - } else { - wfDebug( "Parser output was marked as uncacheable and has not been saved.\n" ); - } - } - -} - - diff --git a/includes/ParserOptions.php b/includes/ParserOptions.php deleted file mode 100644 index 996bba21..00000000 --- a/includes/ParserOptions.php +++ /dev/null @@ -1,145 +0,0 @@ - tags - var $mUseDynamicDates; # Use DateFormatter to format dates - var $mInterwikiMagic; # Interlanguage links are removed and returned in an array - var $mAllowExternalImages; # Allow external images inline - var $mAllowExternalImagesFrom; # If not, any exception? - var $mSkin; # Reference to the preferred skin - var $mDateFormat; # Date format index - var $mEditSection; # Create "edit section" links - var $mNumberHeadings; # Automatically number headings - var $mAllowSpecialInclusion; # Allow inclusion of special pages - var $mTidy; # Ask for tidy cleanup - var $mInterfaceMessage; # Which lang to call for PLURAL and GRAMMAR - var $mMaxIncludeSize; # Maximum size of template expansions, in bytes - var $mMaxPPNodeCount; # Maximum number of nodes touched by PPFrame::expand() - var $mMaxTemplateDepth; # Maximum recursion depth for templates within templates - var $mRemoveComments; # Remove HTML comments. ONLY APPLIES TO PREPROCESS OPERATIONS - var $mTemplateCallback; # Callback for template fetching - var $mEnableLimitReport; # Enable limit report in an HTML comment on output - var $mTimestamp; # Timestamp used for {{CURRENTDAY}} etc. - - var $mUser; # Stored user object, just used to initialise the skin - - function getUseTeX() { return $this->mUseTeX; } - function getUseDynamicDates() { return $this->mUseDynamicDates; } - function getInterwikiMagic() { return $this->mInterwikiMagic; } - function getAllowExternalImages() { return $this->mAllowExternalImages; } - function getAllowExternalImagesFrom() { return $this->mAllowExternalImagesFrom; } - function getEditSection() { return $this->mEditSection; } - function getNumberHeadings() { return $this->mNumberHeadings; } - function getAllowSpecialInclusion() { return $this->mAllowSpecialInclusion; } - function getTidy() { return $this->mTidy; } - function getInterfaceMessage() { return $this->mInterfaceMessage; } - function getMaxIncludeSize() { return $this->mMaxIncludeSize; } - function getMaxPPNodeCount() { return $this->mMaxPPNodeCount; } - function getMaxTemplateDepth() { return $this->mMaxTemplateDepth; } - function getRemoveComments() { return $this->mRemoveComments; } - function getTemplateCallback() { return $this->mTemplateCallback; } - function getEnableLimitReport() { return $this->mEnableLimitReport; } - - function getSkin() { - if ( !isset( $this->mSkin ) ) { - $this->mSkin = $this->mUser->getSkin(); - } - return $this->mSkin; - } - - function getDateFormat() { - if ( !isset( $this->mDateFormat ) ) { - $this->mDateFormat = $this->mUser->getDatePreference(); - } - return $this->mDateFormat; - } - - function getTimestamp() { - if ( !isset( $this->mTimestamp ) ) { - $this->mTimestamp = wfTimestampNow(); - } - return $this->mTimestamp; - } - - function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX, $x ); } - function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates, $x ); } - function setInterwikiMagic( $x ) { return wfSetVar( $this->mInterwikiMagic, $x ); } - function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages, $x ); } - function setAllowExternalImagesFrom( $x ) { return wfSetVar( $this->mAllowExternalImagesFrom, $x ); } - function setDateFormat( $x ) { return wfSetVar( $this->mDateFormat, $x ); } - function setEditSection( $x ) { return wfSetVar( $this->mEditSection, $x ); } - function setNumberHeadings( $x ) { return wfSetVar( $this->mNumberHeadings, $x ); } - function setAllowSpecialInclusion( $x ) { return wfSetVar( $this->mAllowSpecialInclusion, $x ); } - function setTidy( $x ) { return wfSetVar( $this->mTidy, $x); } - function setSkin( $x ) { $this->mSkin = $x; } - function setInterfaceMessage( $x ) { return wfSetVar( $this->mInterfaceMessage, $x); } - function setMaxIncludeSize( $x ) { return wfSetVar( $this->mMaxIncludeSize, $x ); } - function setMaxPPNodeCount( $x ) { return wfSetVar( $this->mMaxPPNodeCount, $x ); } - function setMaxTemplateDepth( $x ) { return wfSetVar( $this->mMaxTemplateDepth, $x ); } - function setRemoveComments( $x ) { return wfSetVar( $this->mRemoveComments, $x ); } - function setTemplateCallback( $x ) { return wfSetVar( $this->mTemplateCallback, $x ); } - function enableLimitReport( $x = true ) { return wfSetVar( $this->mEnableLimitReport, $x ); } - function setTimestamp( $x ) { return wfSetVar( $this->mTimestamp, $x ); } - - function __construct( $user = null ) { - $this->initialiseFromUser( $user ); - } - - /** - * Get parser options - * @static - */ - static function newFromUser( $user ) { - return new ParserOptions( $user ); - } - - /** Get user options */ - function initialiseFromUser( $userInput ) { - global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages; - global $wgAllowExternalImagesFrom, $wgAllowSpecialInclusion, $wgMaxArticleSize; - global $wgMaxPPNodeCount, $wgMaxTemplateDepth; - $fname = 'ParserOptions::initialiseFromUser'; - wfProfileIn( $fname ); - if ( !$userInput ) { - global $wgUser; - if ( isset( $wgUser ) ) { - $user = $wgUser; - } else { - $user = new User; - } - } else { - $user =& $userInput; - } - - $this->mUser = $user; - - $this->mUseTeX = $wgUseTeX; - $this->mUseDynamicDates = $wgUseDynamicDates; - $this->mInterwikiMagic = $wgInterwikiMagic; - $this->mAllowExternalImages = $wgAllowExternalImages; - $this->mAllowExternalImagesFrom = $wgAllowExternalImagesFrom; - $this->mSkin = null; # Deferred - $this->mDateFormat = null; # Deferred - $this->mEditSection = true; - $this->mNumberHeadings = $user->getOption( 'numberheadings' ); - $this->mAllowSpecialInclusion = $wgAllowSpecialInclusion; - $this->mTidy = false; - $this->mInterfaceMessage = false; - $this->mMaxIncludeSize = $wgMaxArticleSize * 1024; - $this->mMaxPPNodeCount = $wgMaxPPNodeCount; - $this->mMaxTemplateDepth = $wgMaxTemplateDepth; - $this->mRemoveComments = true; - $this->mTemplateCallback = array( 'Parser', 'statelessFetchTemplate' ); - $this->mEnableLimitReport = false; - wfProfileOut( $fname ); - } -} - - diff --git a/includes/ParserOutput.php b/includes/ParserOutput.php deleted file mode 100644 index 9b3c12c1..00000000 --- a/includes/ParserOutput.php +++ /dev/null @@ -1,189 +0,0 @@ - section - $mOutputHooks, # Hook tags as per $wgParserOutputHooks - $mWarnings, # Warning text to be returned to the user. Wikitext formatted. - $mSections; # Table of contents - - /** - * Overridden title for display - */ - private $displayTitle = false; - - function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(), - $containsOldMagic = false, $titletext = '' ) - { - $this->mText = $text; - $this->mLanguageLinks = $languageLinks; - $this->mCategories = $categoryLinks; - $this->mContainsOldMagic = $containsOldMagic; - $this->mCacheTime = ''; - $this->mVersion = Parser::VERSION; - $this->mTitleText = $titletext; - $this->mSections = array(); - $this->mLinks = array(); - $this->mTemplates = array(); - $this->mImages = array(); - $this->mExternalLinks = array(); - $this->mNewSection = false; - $this->mNoGallery = false; - $this->mHeadItems = array(); - $this->mTemplateIds = array(); - $this->mOutputHooks = array(); - $this->mWarnings = array(); - } - - function getText() { return $this->mText; } - function &getLanguageLinks() { return $this->mLanguageLinks; } - function getCategoryLinks() { return array_keys( $this->mCategories ); } - function &getCategories() { return $this->mCategories; } - function getCacheTime() { return $this->mCacheTime; } - function getTitleText() { return $this->mTitleText; } - function getSections() { return $this->mSections; } - function &getLinks() { return $this->mLinks; } - function &getTemplates() { return $this->mTemplates; } - function &getImages() { return $this->mImages; } - function &getExternalLinks() { return $this->mExternalLinks; } - function getNoGallery() { return $this->mNoGallery; } - function getSubtitle() { return $this->mSubtitle; } - function getOutputHooks() { return (array)$this->mOutputHooks; } - function getWarnings() { return isset( $this->mWarnings ) ? $this->mWarnings : array(); } - - 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->mCategories, $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 setSections( $toc ) { return wfSetVar( $this->mSections, $toc ); } - - function addCategory( $c, $sort ) { $this->mCategories[$c] = $sort; } - function addLanguageLink( $t ) { $this->mLanguageLinks[] = $t; } - function addExternalLink( $url ) { $this->mExternalLinks[$url] = 1; } - function addWarning( $s ) { $this->mWarnings[] = $s; } - - function addOutputHook( $hook, $data = false ) { - $this->mOutputHooks[] = array( $hook, $data ); - } - - function setNewSection( $value ) { - $this->mNewSection = (bool)$value; - } - function getNewSection() { - return (bool)$this->mNewSection; - } - - function addLink( $title, $id = null ) { - $ns = $title->getNamespace(); - $dbk = $title->getDBkey(); - if ( !isset( $this->mLinks[$ns] ) ) { - $this->mLinks[$ns] = array(); - } - if ( is_null( $id ) ) { - $id = $title->getArticleID(); - } - $this->mLinks[$ns][$dbk] = $id; - } - - function addImage( $name ) { - $this->mImages[$name] = 1; - } - - function addTemplate( $title, $page_id, $rev_id ) { - $ns = $title->getNamespace(); - $dbk = $title->getDBkey(); - if ( !isset( $this->mTemplates[$ns] ) ) { - $this->mTemplates[$ns] = array(); - } - $this->mTemplates[$ns][$dbk] = $page_id; - if ( !isset( $this->mTemplateIds[$ns] ) ) { - $this->mTemplateIds[$ns] = array(); - } - $this->mTemplateIds[$ns][$dbk] = $rev_id; // For versioning - } - - /** - * 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 - * @public - */ - function expired( $touched ) { - global $wgCacheEpoch; - return $this->getCacheTime() == -1 || // parser says it's uncacheable - $this->getCacheTime() < $touched || - $this->getCacheTime() <= $wgCacheEpoch || - !isset( $this->mVersion ) || - version_compare( $this->mVersion, Parser::VERSION, "lt" ); - } - - /** - * Add some text to the . - * If $tag is set, the section with that tag will only be included once - * in a given page. - */ - function addHeadItem( $section, $tag = false ) { - if ( $tag !== false ) { - $this->mHeadItems[$tag] = $section; - } else { - $this->mHeadItems[] = $section; - } - } - - /** - * Override the title to be used for display - * -- this is assumed to have been validated - * (check equal normalisation, etc.) - * - * @param string $text Desired title text - */ - public function setDisplayTitle( $text ) { - $this->displayTitle = $text; - } - - /** - * Get the title to be used for display - * - * @return string - */ - public function getDisplayTitle() { - return $this->displayTitle; - } - - /** - * Fairly generic flag setter thingy. - */ - public function setFlag( $flag ) { - $this->mFlags[$flag] = true; - } - - public function getFlag( $flag ) { - return isset( $this->mFlags[$flag] ); - } - -} - - diff --git a/includes/ParserXML.php b/includes/ParserXML.php deleted file mode 100644 index e7b64f6e..00000000 --- a/includes/ParserXML.php +++ /dev/null @@ -1,643 +0,0 @@ -(X)HTML parser - * Based on work by Jan Hidders and Magnus Manske - * To use, set - * $wgUseXMLparser = true ; - * $wgEnableParserCache = false ; - * $wgWiki2xml to the path and executable of the command line version (cli) - * in LocalSettings.php - * @package MediaWiki - * @subpackage Experimental - */ - -/** - * the base class for an element - * @package MediaWiki - * @subpackage Experimental - */ -class element { - var $name = ''; - var $attrs = array (); - var $children = array (); - - /** - * This finds the ATTRS element and returns the ATTR sub-children as a single string - * @todo FIXME $parser always empty when calling makeXHTML() - */ - function getSourceAttrs() { - $ret = ''; - foreach ($this->children as $child) { - if (!is_string($child) AND $child->name == 'ATTRS') { - $ret = $child->makeXHTML($parser); - } - } - return $ret; - } - - /** - * This collects the ATTR thingies for getSourceAttrs() - */ - function getTheseAttrs() { - $ret = array (); - foreach ($this->children as $child) { - if (!is_string($child) AND $child->name == 'ATTR') { - $ret[] = $child->attrs["NAME"]."='".$child->children[0]."'"; - } - } - return implode(' ', $ret); - } - - function fixLinkTails(& $parser, $key) { - $k2 = $key +1; - if (!isset ($this->children[$k2])) - return; - if (!is_string($this->children[$k2])) - return; - if (is_string($this->children[$key])) - return; - if ($this->children[$key]->name != "LINK") - return; - - $n = $this->children[$k2]; - $s = ''; - while ($n != '' AND (($n[0] >= 'a' AND $n[0] <= 'z') OR $n[0] == 'ä' OR $n[0] == 'ö' OR $n[0] == 'ü' OR $n[0] == 'ß')) { - $s .= $n[0]; - $n = substr($n, 1); - } - $this->children[$k2] = $n; - - if (count($this->children[$key]->children) > 1) { - $kl = array_keys($this->children[$key]->children); - $kl = array_pop($kl); - $this->children[$key]->children[$kl]->children[] = $s; - } else { - $e = new element; - $e->name = "LINKOPTION"; - $t = $this->children[$key]->sub_makeXHTML($parser); - $e->children[] = trim($t).$s; - $this->children[$key]->children[] = $e; - } - } - - /** - * This function generates the XHTML for the entire subtree - */ - function sub_makeXHTML(& $parser, $tag = '', $attr = '') { - $ret = ''; - - $attr2 = $this->getSourceAttrs(); - if ($attr != '' AND $attr2 != '') - $attr .= ' '; - $attr .= $attr2; - - if ($tag != '') { - $ret .= '<'.$tag; - if ($attr != '') - $ret .= ' '.$attr; - $ret .= '>'; - } - - # THIS SHOULD BE DONE IN THE WIKI2XML-PARSER INSTEAD - # foreach ( array_keys ( $this->children ) AS $x ) - # $this->fixLinkTails ( $parser , $x ) ; - - foreach ($this->children as $child) { - if (is_string($child)) { - $ret .= $child; - } elseif ($child->name != 'ATTRS') { - $ret .= $child->makeXHTML($parser); - } - } - if ($tag != '') - $ret .= '\n"; - return $ret; - } - - /** - * Link functions - */ - function createInternalLink(& $parser, $target, $display_title, $options) { - global $wgUser; - $skin = $wgUser->getSkin(); - $tp = explode(':', $target); # tp = target parts - $title = ''; # The plain title - $language = ''; # The language/meta/etc. part - $namespace = ''; # The namespace, if any - $subtarget = ''; # The '#' thingy - - $nt = Title :: newFromText($target); - $fl = strtoupper($this->attrs['FORCEDLINK']) == 'YES'; - - if ($fl || count($tp) == 1) { - # Plain and simple case - $title = $target; - } else { - # There's stuff missing here... - if ($nt->getNamespace() == NS_IMAGE) { - $options[] = $display_title; - return $parser->makeImage($nt, implode('|', $options)); - } else { - # Default - $title = $target; - } - } - - if ($language != '') { - # External link within the WikiMedia project - return "{language link}"; - } else { - if ($namespace != '') { - # Link to another namespace, check for image/media stuff - return "{namespace link}"; - } else { - return $skin->makeLink($target, $display_title); - } - } - } - - /** @todo document */ - function makeInternalLink(& $parser) { - $target = ''; - $option = array (); - foreach ($this->children as $child) { - if (is_string($child)) { - # This shouldn't be the case! - } else { - if ($child->name == 'LINKTARGET') { - $target = trim($child->makeXHTML($parser)); - } else { - $option[] = trim($child->makeXHTML($parser)); - } - } - } - - if (count($option) == 0) - $option[] = $target; # Create dummy display title - $display_title = array_pop($option); - return $this->createInternalLink($parser, $target, $display_title, $option); - } - - /** @todo document */ - function getTemplateXHTML($title, $parts, & $parser) { - global $wgLang, $wgUser; - $skin = $wgUser->getSkin(); - $ot = $title; # Original title - if (count(explode(':', $title)) == 1) - $title = $wgLang->getNsText(NS_TEMPLATE).":".$title; - $nt = Title :: newFromText($title); - $id = $nt->getArticleID(); - if ($id == 0) { - # No/non-existing page - return $skin->makeBrokenLink($title, $ot); - } - - $a = 0; - $tv = array (); # Template variables - foreach ($parts AS $part) { - $a ++; - $x = explode('=', $part, 2); - if (count($x) == 1) - $key = "{$a}"; - else - $key = $x[0]; - $value = array_pop($x); - $tv[$key] = $value; - } - $art = new Article($nt); - $text = $art->getContent(false); - $parser->plain_parse($text, true, $tv); - - return $text; - } - - /** - * This function actually converts wikiXML into XHTML tags - * @todo use switch() ! - */ - function makeXHTML(& $parser) { - $ret = ''; - $n = $this->name; # Shortcut - - if ($n == 'EXTENSION') { - # Fix allowed HTML - $old_n = $n; - $ext = strtoupper($this->attrs['NAME']); - - switch($ext) { - case 'B': - case 'STRONG': - $n = 'BOLD'; - break; - case 'I': - case 'EM': - $n = 'ITALICS'; - break; - case 'U': - $n = 'UNDERLINED'; # Hey, virtual wiki tag! ;-) - break; - case 'S': - $n = 'STRIKE'; - break; - case 'P': - $n = 'PARAGRAPH'; - break; - case 'TABLE': - $n = 'TABLE'; - break; - case 'TR': - $n = 'TABLEROW'; - break; - case 'TD': - $n = 'TABLECELL'; - break; - case 'TH': - $n = 'TABLEHEAD'; - break; - case 'CAPTION': - $n = 'CAPTION'; - break; - case 'NOWIKI': - $n = 'NOWIKI'; - break; - } - if ($n != $old_n) { - unset ($this->attrs['NAME']); # Cleanup - } elseif ($parser->nowiki > 0) { - # No 'real' wiki tags allowed in nowiki section - $n = ''; - } - } // $n = 'EXTENSION' - - switch($n) { - case 'ARTICLE': - $ret .= $this->sub_makeXHTML($parser); - break; - case 'HEADING': - $ret .= $this->sub_makeXHTML($parser, 'h'.$this->attrs['LEVEL']); - break; - case 'PARAGRAPH': - $ret .= $this->sub_makeXHTML($parser, 'p'); - break; - case 'BOLD': - $ret .= $this->sub_makeXHTML($parser, 'strong'); - break; - case 'ITALICS': - $ret .= $this->sub_makeXHTML($parser, 'em'); - break; - - # These don't exist as wiki markup - case 'UNDERLINED': - $ret .= $this->sub_makeXHTML($parser, 'u'); - break; - case 'STRIKE': - $ret .= $this->sub_makeXHTML($parser, 'strike'); - break; - - # HTML comment - case 'COMMENT': - # Comments are parsed out - $ret .= ''; - break; - - - # Links - case 'LINK': - $ret .= $this->makeInternalLink($parser); - break; - case 'LINKTARGET': - case 'LINKOPTION': - $ret .= $this->sub_makeXHTML($parser); - break; - - case 'TEMPLATE': - $parts = $this->sub_makeXHTML($parser); - $parts = explode('|', $parts); - $title = array_shift($parts); - $ret .= $this->getTemplateXHTML($title, $parts, & $parser); - break; - - case 'TEMPLATEVAR': - $x = $this->sub_makeXHTML($parser); - if (isset ($parser->mCurrentTemplateOptions["{$x}"])) - $ret .= $parser->mCurrentTemplateOptions["{$x}"]; - break; - - # Internal use, not generated by wiki2xml parser - case 'IGNORE': - $ret .= $this->sub_makeXHTML($parser); - - case 'NOWIKI': - $parser->nowiki++; - $ret .= $this->sub_makeXHTML($parser, ''); - $parser->nowiki--; - - - # Unknown HTML extension - case 'EXTENSION': # This is currently a dummy!!! - $ext = $this->attrs['NAME']; - - $ret .= '<'.$ext.'>'; - $ret .= $this->sub_makeXHTML($parser); - $ret .= '</'.$ext.'> '; - break; - - - # Table stuff - - case 'TABLE': - $ret .= $this->sub_makeXHTML($parser, 'table'); - break; - case 'TABLEROW': - $ret .= $this->sub_makeXHTML($parser, 'tr'); - break; - case 'TABLECELL': - $ret .= $this->sub_makeXHTML($parser, 'td'); - break; - case 'TABLEHEAD': - $ret .= $this->sub_makeXHTML($parser, 'th'); - break; - case 'CAPTION': - $ret .= $this->sub_makeXHTML($parser, 'caption'); - break; - case 'ATTRS': # SPECIAL CASE : returning attributes - return $this->getTheseAttrs(); - - - # Lists stuff - case 'LISTITEM': - if ($parser->mListType == 'dl') - $ret .= $this->sub_makeXHTML($parser, 'dd'); - else - $ret .= $this->sub_makeXHTML($parser, 'li'); - break; - case 'LIST': - $type = 'ol'; # Default - if ($this->attrs['TYPE'] == 'bullet') - $type = 'ul'; - else - if ($this->attrs['TYPE'] == 'indent') - $type = 'dl'; - $oldtype = $parser->mListType; - $parser->mListType = $type; - $ret .= $this->sub_makeXHTML($parser, $type); - $parser->mListType = $oldtype; - break; - - # Something else entirely - default: - $ret .= '<'.$n.'>'; - $ret .= $this->sub_makeXHTML($parser); - $ret .= '</'.$n.'> '; - } // switch($n) - - $ret = "\n{$ret}\n"; - $ret = str_replace("\n\n", "\n", $ret); - return $ret; - } - - /** - * A function for additional debugging output - */ - function myPrint() { - $ret = "
      \n"; - $ret .= "
    • Name: $this->name
    • \n"; - // print attributes - $ret .= '
    • Attributes: '; - foreach ($this->attrs as $name => $value) { - $ret .= "$name => $value; "; - } - $ret .= "
    • \n"; - // print children - foreach ($this->children as $child) { - if (is_string($child)) { - $ret .= "
    • $child
    • \n"; - } else { - $ret .= $child->myPrint(); - } - } - $ret .= "
    \n"; - return $ret; - } -} - -$ancStack = array (); // the stack with ancestral elements - -// START Three global functions needed for parsing, sorry guys -/** @todo document */ -function wgXMLstartElement($parser, $name, $attrs) { - global $ancStack; - - $newElem = new element; - $newElem->name = $name; - $newElem->attrs = $attrs; - - array_push($ancStack, $newElem); -} - -/** @todo document */ -function wgXMLendElement($parser, $name) { - global $ancStack, $rootElem; - // pop element off stack - $elem = array_pop($ancStack); - if (count($ancStack) == 0) - $rootElem = $elem; - else - // add it to its parent - array_push($ancStack[count($ancStack) - 1]->children, $elem); -} - -/** @todo document */ -function wgXMLcharacterData($parser, $data) { - global $ancStack; - $data = trim($data); // Don't add blank lines, they're no use... - // add to parent if parent exists - if ($ancStack && $data != "") { - array_push($ancStack[count($ancStack) - 1]->children, $data); - } -} -// END Three global functions needed for parsing, sorry guys - -/** - * Here's the class that generates a nice tree - * @package MediaWiki - * @subpackage Experimental - */ -class xml2php { - - /** @todo document */ - function & scanFile($filename) { - global $ancStack, $rootElem; - $ancStack = array (); - - $xml_parser = xml_parser_create(); - xml_set_element_handler($xml_parser, 'wgXMLstartElement', 'wgXMLendElement'); - xml_set_character_data_handler($xml_parser, 'wgXMLcharacterData'); - if (!($fp = fopen($filename, 'r'))) { - die('could not open XML input'); - } - while ($data = fread($fp, 4096)) { - if (!xml_parse($xml_parser, $data, feof($fp))) { - die(sprintf("XML error: %s at line %d", xml_error_string(xml_get_error_code($xml_parser)), xml_get_current_line_number($xml_parser))); - } - } - xml_parser_free($xml_parser); - - // return the remaining root element we copied in the beginning - return $rootElem; - } - - /** @todo document */ - function scanString($input) { - global $ancStack, $rootElem; - $ancStack = array (); - - $xml_parser = xml_parser_create(); - xml_set_element_handler($xml_parser, 'wgXMLstartElement', 'wgXMLendElement'); - xml_set_character_data_handler($xml_parser, 'wgXMLcharacterData'); - - if (!xml_parse($xml_parser, $input, true)) { - die(sprintf("XML error: %s at line %d", xml_error_string(xml_get_error_code($xml_parser)), xml_get_current_line_number($xml_parser))); - } - xml_parser_free($xml_parser); - - // return the remaining root element we copied in the beginning - return $rootElem; - } - -} - -/** - * @todo document - * @package MediaWiki - * @subpackage Experimental - */ -class ParserXML extends Parser { - /**#@+ - * @private - */ - # Persistent: - var $mTagHooks, $mListType; - - # Cleared with clearState(): - var $mOutput, $mAutonumber, $mDTopen, $mStripState = array (); - var $mVariables, $mIncludeCount, $mArgStack, $mLastSection, $mInPre; - - # Temporary: - var $mOptions, $mTitle, $mOutputType, $mTemplates, // cache of already loaded templates, avoids - // multiple SQL queries for the same string - $mTemplatePath; // stores an unsorted hash of all the templates already loaded - // in this path. Used for loop detection. - - var $nowikicount, $mCurrentTemplateOptions; - - /**#@-*/ - - /** - * Constructor - * - * @public - */ - function ParserXML() { - $this->mTemplates = array (); - $this->mTemplatePath = array (); - $this->mTagHooks = array (); - $this->clearState(); - } - - /** - * Clear Parser state - * - * @private - */ - function clearState() { - $this->mOutput = new ParserOutput; - $this->mAutonumber = 0; - $this->mLastSection = ""; - $this->mDTopen = false; - $this->mVariables = false; - $this->mIncludeCount = array (); - $this->mStripState = array (); - $this->mArgStack = array (); - $this->mInPre = false; - } - - /** - * Turns the wikitext into XML by calling the external parser - * - */ - function html2xml(& $text) { - global $wgWiki2xml; - - # generating html2xml command path - $a = $wgWiki2xml; - $a = explode('/', $a); - array_pop($a); - $a[] = 'html2xml'; - $html2xml = implode('/', $a); - $a = array (); - - $tmpfname = tempnam( wfTempDir(), 'FOO' ); - $handle = fopen($tmpfname, 'w'); - fwrite($handle, utf8_encode($text)); - fclose($handle); - exec($html2xml.' < '.$tmpfname, $a); - $text = utf8_decode(implode("\n", $a)); - unlink($tmpfname); - } - - /** @todo document */ - function runXMLparser(& $text) { - global $wgWiki2xml; - - $this->html2xml($text); - - $tmpfname = tempnam( wfTempDir(), 'FOO'); - $handle = fopen($tmpfname, 'w'); - fwrite($handle, $text); - fclose($handle); - exec($wgWiki2xml.' < '.$tmpfname, $a); - $text = utf8_decode(implode("\n", $a)); - unlink($tmpfname); - } - - /** @todo document */ - function plain_parse(& $text, $inline = false, $templateOptions = array ()) { - $this->runXMLparser($text); - $nowikicount = 0; - $w = new xml2php; - $result = $w->scanString($text); - - $oldTemplateOptions = $this->mCurrentTemplateOptions; - $this->mCurrentTemplateOptions = $templateOptions; - - if ($inline) { # Inline rendering off for templates - if (count($result->children) == 1) - $result->children[0]->name = 'IGNORE'; - } - - if (1) - $text = $result->makeXHTML($this); # No debugging info - else - $text = $result->makeXHTML($this).'
    '.$text.'
    '.$result->myPrint(); - $this->mCurrentTemplateOptions = $oldTemplateOptions; - } - - /** @todo document */ - function parse($text, & $title, $options, $linestart = true, $clearState = true) { - $this->plain_parse($text); - $this->mOutput->setText($text); - return $this->mOutput; - } - -} -?> diff --git a/includes/Parser_DiffTest.php b/includes/Parser_DiffTest.php deleted file mode 100644 index d88709f0..00000000 --- a/includes/Parser_DiffTest.php +++ /dev/null @@ -1,85 +0,0 @@ -conf = $conf; - $this->dtUniqPrefix = "\x7fUNIQ" . Parser::getRandomString(); - } - - function init() { - if ( !is_null( $this->parsers ) ) { - return; - } - - global $wgHooks; - static $doneHook = false; - if ( !$doneHook ) { - $doneHook = true; - $wgHooks['ParserClearState'][] = array( $this, 'onClearState' ); - } - - foreach ( $this->conf['parsers'] as $i => $parserConf ) { - if ( !is_array( $parserConf ) ) { - $class = $parserConf; - $parserConf = array( 'class' => $parserConf ); - } else { - $class = $parserConf['class']; - } - $this->parsers[$i] = new $class( $parserConf ); - } - } - - function __call( $name, $args ) { - $this->init(); - $results = array(); - $mismatch = false; - $lastResult = null; - $first = true; - foreach ( $this->parsers as $i => $parser ) { - $currentResult = call_user_func_array( array( &$this->parsers[$i], $name ), $args ); - if ( $first ) { - $first = false; - } else { - if ( is_object( $lastResult ) ) { - if ( $lastResult != $currentResult ) { - $mismatch = true; - } - } else { - if ( $lastResult !== $currentResult ) { - $mismatch = true; - } - } - } - $results[$i] = $currentResult; - $lastResult = $currentResult; - } - if ( $mismatch ) { - throw new MWException( "Parser_DiffTest: results mismatch on call to $name\n" . - 'Arguments: ' . var_export( $args, true ) . "\n" . - 'Results: ' . var_export( $results, true ) . "\n" ); - } - return $lastResult; - } - - function setFunctionHook( $id, $callback, $flags = 0 ) { - $this->init(); - foreach ( $this->parsers as $i => $parser ) { - $parser->setFunctionHook( $id, $callback, $flags ); - } - } - - function onClearState( &$parser ) { - // hack marker prefixes to get identical output - $parser->mUniqPrefix = $this->dtUniqPrefix; - return true; - } -} - diff --git a/includes/Parser_OldPP.php b/includes/Parser_OldPP.php deleted file mode 100644 index c10de257..00000000 --- a/includes/Parser_OldPP.php +++ /dev/null @@ -1,4942 +0,0 @@ -"\\x00-\\x20\\x7F]'; - const EXT_IMAGE_REGEX = '/^(http:\/\/|https:\/\/)([^][<>"\\x00-\\x20\\x7F]+)\\/([A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]+)\\.((?i)gif|png|jpg|jpeg)$/S'; - - // State constants for the definition list colon extraction - const COLON_STATE_TEXT = 0; - const COLON_STATE_TAG = 1; - const COLON_STATE_TAGSTART = 2; - const COLON_STATE_CLOSETAG = 3; - const COLON_STATE_TAGSLASH = 4; - const COLON_STATE_COMMENT = 5; - const COLON_STATE_COMMENTDASH = 6; - const COLON_STATE_COMMENTDASHDASH = 7; - - // Allowed values for $this->mOutputType - // Parameter to startExternalParse(). - const OT_HTML = 1; - const OT_WIKI = 2; - const OT_PREPROCESS = 3; - const OT_MSG = 4; - - /**#@+ - * @private - */ - # Persistent: - var $mTagHooks, $mTransparentTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables, - $mImageParams, $mImageParamsMagicArray, $mExtLinkBracketedRegex; - - # Cleared with clearState(): - var $mOutput, $mAutonumber, $mDTopen, $mStripState; - var $mIncludeCount, $mArgStack, $mLastSection, $mInPre; - var $mInterwikiLinkHolders, $mLinkHolders, $mUniqPrefix; - var $mIncludeSizes, $mDefaultSort; - var $mTemplates, // cache of already loaded templates, avoids - // multiple SQL queries for the same string - $mTemplatePath; // stores an unsorted hash of all the templates already loaded - // in this path. Used for loop detection. - - # Temporary - # These are variables reset at least once per parse regardless of $clearState - var $mOptions, // ParserOptions object - $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 - $mRevisionTimestamp, // The timestamp of the specified revision ID - $mRevIdForTs; // The revision ID which was used to fetch the timestamp - - /**#@-*/ - - /** - * Constructor - * - * @public - */ - function __construct( $conf = array() ) { - $this->mTagHooks = array(); - $this->mTransparentTagHooks = array(); - $this->mFunctionHooks = array(); - $this->mFunctionSynonyms = array( 0 => array(), 1 => array() ); - $this->mFirstCall = true; - $this->mExtLinkBracketedRegex = '/\[(\b(' . wfUrlProtocols() . ')'. - '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x0a\\x0d]*?)\]/S'; - } - - /** - * Do various kinds of initialisation on the first call of the parser - */ - function firstCallInit() { - if ( !$this->mFirstCall ) { - return; - } - $this->mFirstCall = false; - - wfProfileIn( __METHOD__ ); - global $wgAllowDisplayTitle, $wgAllowSlowParserFunctions; - - $this->setHook( 'pre', array( $this, 'renderPreTag' ) ); - - # Syntax for arguments (see self::setFunctionHook): - # "name for lookup in localized magic words array", - # function callback, - # optional SFH_NO_HASH to omit the hash from calls (e.g. {{int:...} - # instead of {{#int:...}}) - $this->setFunctionHook( 'int', array( 'CoreParserFunctions', 'intFunction' ), SFH_NO_HASH ); - $this->setFunctionHook( 'ns', array( 'CoreParserFunctions', 'ns' ), SFH_NO_HASH ); - $this->setFunctionHook( 'urlencode', array( 'CoreParserFunctions', 'urlencode' ), SFH_NO_HASH ); - $this->setFunctionHook( 'lcfirst', array( 'CoreParserFunctions', 'lcfirst' ), SFH_NO_HASH ); - $this->setFunctionHook( 'ucfirst', array( 'CoreParserFunctions', 'ucfirst' ), SFH_NO_HASH ); - $this->setFunctionHook( 'lc', array( 'CoreParserFunctions', 'lc' ), SFH_NO_HASH ); - $this->setFunctionHook( 'uc', array( 'CoreParserFunctions', 'uc' ), SFH_NO_HASH ); - $this->setFunctionHook( 'localurl', array( 'CoreParserFunctions', 'localurl' ), SFH_NO_HASH ); - $this->setFunctionHook( 'localurle', array( 'CoreParserFunctions', 'localurle' ), SFH_NO_HASH ); - $this->setFunctionHook( 'fullurl', array( 'CoreParserFunctions', 'fullurl' ), SFH_NO_HASH ); - $this->setFunctionHook( 'fullurle', array( 'CoreParserFunctions', 'fullurle' ), SFH_NO_HASH ); - $this->setFunctionHook( 'formatnum', array( 'CoreParserFunctions', 'formatnum' ), SFH_NO_HASH ); - $this->setFunctionHook( 'grammar', array( 'CoreParserFunctions', 'grammar' ), SFH_NO_HASH ); - $this->setFunctionHook( 'plural', array( 'CoreParserFunctions', 'plural' ), SFH_NO_HASH ); - $this->setFunctionHook( 'numberofpages', array( 'CoreParserFunctions', 'numberofpages' ), SFH_NO_HASH ); - $this->setFunctionHook( 'numberofusers', array( 'CoreParserFunctions', 'numberofusers' ), SFH_NO_HASH ); - $this->setFunctionHook( 'numberofarticles', array( 'CoreParserFunctions', 'numberofarticles' ), SFH_NO_HASH ); - $this->setFunctionHook( 'numberoffiles', array( 'CoreParserFunctions', 'numberoffiles' ), SFH_NO_HASH ); - $this->setFunctionHook( 'numberofadmins', array( 'CoreParserFunctions', 'numberofadmins' ), SFH_NO_HASH ); - $this->setFunctionHook( 'numberofedits', array( 'CoreParserFunctions', 'numberofedits' ), SFH_NO_HASH ); - $this->setFunctionHook( 'language', array( 'CoreParserFunctions', 'language' ), SFH_NO_HASH ); - $this->setFunctionHook( 'padleft', array( 'CoreParserFunctions', 'padleft' ), SFH_NO_HASH ); - $this->setFunctionHook( 'padright', array( 'CoreParserFunctions', 'padright' ), SFH_NO_HASH ); - $this->setFunctionHook( 'anchorencode', array( 'CoreParserFunctions', 'anchorencode' ), SFH_NO_HASH ); - $this->setFunctionHook( 'special', array( 'CoreParserFunctions', 'special' ) ); - $this->setFunctionHook( 'defaultsort', array( 'CoreParserFunctions', 'defaultsort' ), SFH_NO_HASH ); - $this->setFunctionHook( 'filepath', array( 'CoreParserFunctions', 'filepath' ), SFH_NO_HASH ); - - if ( $wgAllowDisplayTitle ) { - $this->setFunctionHook( 'displaytitle', array( 'CoreParserFunctions', 'displaytitle' ), SFH_NO_HASH ); - } - if ( $wgAllowSlowParserFunctions ) { - $this->setFunctionHook( 'pagesinnamespace', array( 'CoreParserFunctions', 'pagesinnamespace' ), SFH_NO_HASH ); - } - - $this->initialiseVariables(); - - wfRunHooks( 'ParserFirstCallInit', array( &$this ) ); - wfProfileOut( __METHOD__ ); - } - - /** - * Clear Parser state - * - * @private - */ - function clearState() { - wfProfileIn( __METHOD__ ); - if ( $this->mFirstCall ) { - $this->firstCallInit(); - } - $this->mOutput = new ParserOutput; - $this->mAutonumber = 0; - $this->mLastSection = ''; - $this->mDTopen = false; - $this->mIncludeCount = array(); - $this->mStripState = new StripState; - $this->mArgStack = array(); - $this->mInPre = false; - $this->mInterwikiLinkHolders = array( - 'texts' => array(), - 'titles' => array() - ); - $this->mLinkHolders = array( - 'namespaces' => array(), - 'dbkeys' => array(), - 'queries' => array(), - 'texts' => array(), - 'titles' => array() - ); - $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. - * Using it at the front also gives us a little extra robustness - * since it shouldn't match when butted up against identifier-like - * string constructs. - */ - $this->mUniqPrefix = "\x07UNIQ" . self::getRandomString(); - - # Clear these on every parse, bug 4549 - $this->mTemplates = array(); - $this->mTemplatePath = array(); - - $this->mShowToc = true; - $this->mForceTocPosition = false; - $this->mIncludeSizes = array( - 'pre-expand' => 0, - 'post-expand' => 0, - 'arg' => 0 - ); - $this->mDefaultSort = false; - - wfRunHooks( 'ParserClearState', array( &$this ) ); - wfProfileOut( __METHOD__ ); - } - - function setOutputType( $ot ) { - $this->mOutputType = $ot; - // Shortcut alias - $this->ot = array( - 'html' => $ot == self::OT_HTML, - 'wiki' => $ot == self::OT_WIKI, - 'msg' => $ot == self::OT_MSG, - 'pre' => $ot == self::OT_PREPROCESS, - ); - } - - /** - * Accessor for mUniqPrefix. - * - * @public - */ - function uniqPrefix() { - return $this->mUniqPrefix; - } - - /** - * Convert wikitext to HTML - * Do not call this function recursively. - * - * @param string $text Text we want to parse - * @param Title &$title A title object - * @param array $options - * @param boolean $linestart - * @param boolean $clearState - * @param int $revid number to pass in {{REVISIONID}} - * @return ParserOutput a ParserOutput - */ - public function parse( $text, &$title, $options, $linestart = true, $clearState = true, $revid = null ) { - /** - * First pass--just handle sections, pass the rest off - * to internalParse() which does all the real work. - */ - - global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang; - $fname = 'Parser::parse-' . wfGetCaller(); - wfProfileIn( __METHOD__ ); - wfProfileIn( $fname ); - - if ( $clearState ) { - $this->clearState(); - } - - $this->mOptions = $options; - $this->mTitle =& $title; - $oldRevisionId = $this->mRevisionId; - $oldRevisionTimestamp = $this->mRevisionTimestamp; - if( $revid !== null ) { - $this->mRevisionId = $revid; - $this->mRevisionTimestamp = null; - } - $this->setOutputType( self::OT_HTML ); - wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) ); - $text = $this->strip( $text, $this->mStripState ); - wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) ); - $text = $this->internalParse( $text ); - $text = $this->mStripState->unstripGeneral( $text ); - - # Clean up special characters, only run once, next-to-last before doBlockLevels - $fixtags = array( - # french spaces, last one Guillemet-left - # only if there is something before the space - '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1 \\2', - # french spaces, Guillemet-right - '/(\\302\\253) /' => '\\1 ', - ); - $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text ); - - # only once and last - $text = $this->doBlockLevels( $text, $linestart ); - - $this->replaceLinkHolders( $text ); - - # the position of the parserConvert() call should not be changed. it - # assumes that the links are all replaced and the only thing left - # is the mark. - # Side-effects: this calls $this->mOutput->setTitleText() - $text = $wgContLang->parserConvert( $text, $this ); - - $text = $this->mStripState->unstripNoWiki( $text ); - - wfRunHooks( 'ParserBeforeTidy', array( &$this, &$text ) ); - -//!JF Move to its own function - - $uniq_prefix = $this->mUniqPrefix; - $matches = array(); - $elements = array_keys( $this->mTransparentTagHooks ); - $text = self::extractTagsAndParams( $elements, $text, $matches, $uniq_prefix ); - - foreach( $matches as $marker => $data ) { - list( $element, $content, $params, $tag ) = $data; - $tagName = strtolower( $element ); - if( isset( $this->mTransparentTagHooks[$tagName] ) ) { - $output = call_user_func_array( $this->mTransparentTagHooks[$tagName], - array( $content, $params, $this ) ); - } else { - $output = $tag; - } - $this->mStripState->general->setPair( $marker, $output ); - } - $text = $this->mStripState->unstripGeneral( $text ); - - $text = Sanitizer::normalizeCharReferences( $text ); - - if (($wgUseTidy and $this->mOptions->mTidy) or $wgAlwaysUseTidy) { - $text = self::tidy($text); - } else { - # attempt to sanitize at least some nesting problems - # (bug #2702 and quite a few others) - $tidyregs = array( - # ''Something [http://www.cool.com cool''] --> - # Somethingcool> - '/(<([bi])>)(<([bi])>)?([^<]*)(<\/?a[^<]*>)([^<]*)(<\/\\4>)?(<\/\\2>)/' => - '\\1\\3\\5\\8\\9\\6\\1\\3\\7\\8\\9', - # fix up an anchor inside another anchor, only - # at least for a single single nested link (bug 3695) - '/(]+>)([^<]*)(]+>[^<]*)<\/a>(.*)<\/a>/' => - '\\1\\2\\3\\1\\4', - # fix div inside inline elements- doBlockLevels won't wrap a line which - # contains a div, so fix it up here; replace - # div with escaped text - '/(<([aib]) [^>]+>)([^<]*)(]*)>)(.*)(<\/div>)([^<]*)(<\/\\2>)/' => - '\\1\\3<div\\5>\\6</div>\\8\\9', - # remove empty italic or bold tag pairs, some - # introduced by rules above - '/<([bi])><\/\\1>/' => '', - ); - - $text = preg_replace( - array_keys( $tidyregs ), - array_values( $tidyregs ), - $text ); - } - - wfRunHooks( 'ParserAfterTidy', array( &$this, &$text ) ); - - # Information on include size limits, for the benefit of users who try to skirt them - if ( $this->mOptions->getEnableLimitReport() ) { - $max = $this->mOptions->getMaxIncludeSize(); - $limitReport = - "Pre-expand include size: {$this->mIncludeSizes['pre-expand']}/$max bytes\n" . - "Post-expand include size: {$this->mIncludeSizes['post-expand']}/$max bytes\n" . - "Template argument size: {$this->mIncludeSizes['arg']}/$max bytes\n"; - wfRunHooks( 'ParserLimitReport', array( $this, &$limitReport ) ); - $text .= "\n"; - } - $this->mOutput->setText( $text ); - $this->mRevisionId = $oldRevisionId; - $this->mRevisionTimestamp = $oldRevisionTimestamp; - wfProfileOut( $fname ); - wfProfileOut( __METHOD__ ); - - return $this->mOutput; - } - - /** - * Recursive parser entry point that can be called from an extension tag - * hook. - */ - function recursiveTagParse( $text ) { - wfProfileIn( __METHOD__ ); - wfRunHooks( 'ParserBeforeStrip', array( &$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; - } - - /** - * Expand templates and variables in the text, producing valid, static wikitext. - * Also removes comments. - */ - function preprocess( $text, $title, $options, $revid = null ) { - wfProfileIn( __METHOD__ ); - $this->clearState(); - $this->setOutputType( self::OT_PREPROCESS ); - $this->mOptions = $options; - $this->mTitle = $title; - if( $revid !== null ) { - $this->mRevisionId = $revid; - } - wfRunHooks( 'ParserBeforeStrip', array( &$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->mStripState->unstripBoth( $text ); - wfProfileOut( __METHOD__ ); - return $text; - } - - /** - * Get a random string - * - * @private - * @static - */ - function getRandomString() { - return dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff)); - } - - function &getTitle() { return $this->mTitle; } - function getOptions() { return $this->mOptions; } - - function getFunctionLang() { - global $wgLang, $wgContLang; - return $this->mOptions->getInterfaceMessage() ? $wgLang : $wgContLang; - } - - /** - * Replaces all occurrences of HTML-style comments and the given tags - * in the text with a random marker and returns teh next text. The output - * parameter $matches will be an associative array filled with data in - * the form: - * 'UNIQ-xxxxx' => array( - * 'element', - * 'tag content', - * array( 'param' => 'x' ), - * 'tag content' ) ) - * - * @param $elements list of element names. Comments are always extracted. - * @param $text Source text string. - * @param $uniq_prefix - * - * @public - * @static - */ - function extractTagsAndParams($elements, $text, &$matches, $uniq_prefix = ''){ - static $n = 1; - $stripped = ''; - $matches = array(); - - $taglist = implode( '|', $elements ); - $start = "/<($taglist)(\\s+[^>]*?|\\s*?)(\/?>)|<(!--)/i"; - - while ( '' != $text ) { - $p = preg_split( $start, $text, 2, PREG_SPLIT_DELIM_CAPTURE ); - $stripped .= $p[0]; - if( count( $p ) < 5 ) { - break; - } - if( count( $p ) > 5 ) { - // comment - $element = $p[4]; - $attributes = ''; - $close = ''; - $inside = $p[5]; - } else { - // tag - $element = $p[1]; - $attributes = $p[2]; - $close = $p[3]; - $inside = $p[4]; - } - - $marker = "$uniq_prefix-$element-" . sprintf('%08X', $n++) . "-QINU\x07"; - $stripped .= $marker; - - if ( $close === '/>' ) { - // Empty element tag, - $content = null; - $text = $inside; - $tail = null; - } else { - if( $element == '!--' ) { - $end = '/(-->)/'; - } else { - $end = "/(<\\/$element\\s*>)/i"; - } - $q = preg_split( $end, $inside, 2, PREG_SPLIT_DELIM_CAPTURE ); - $content = $q[0]; - if( count( $q ) < 3 ) { - # No end tag -- let it run out to the end of the text. - $tail = ''; - $text = ''; - } else { - $tail = $q[1]; - $text = $q[2]; - } - } - - $matches[$marker] = array( $element, - $content, - Sanitizer::decodeTagAttributes( $attributes ), - "<$element$attributes$close$content$tail" ); - } - return $stripped; - } - - /** - * 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() - * - * @param StripState $state - * - * @param bool $stripcomments when set, HTML comments - * will be stripped in addition to other tags. This is important - * for section editing, where these comments cause confusion when - * counting the sections in the wikisource - * - * @param array dontstrip contains tags which should not be stripped; - * used to prevent stipping of when saving (fixes bug 2700) - * - * @private - */ - function strip( $text, $state, $stripcomments = false , $dontstrip = array () ) { - global $wgContLang; - wfProfileIn( __METHOD__ ); - $render = ($this->mOutputType == self::OT_HTML); - - $uniq_prefix = $this->mUniqPrefix; - $commentState = new ReplacementArray; - $nowikiItems = array(); - $generalItems = array(); - - $elements = array_merge( - array( 'nowiki', 'gallery' ), - array_keys( $this->mTagHooks ) ); - global $wgRawHtml; - if( $wgRawHtml ) { - $elements[] = 'html'; - } - if( $this->mOptions->getUseTeX() ) { - $elements[] = 'math'; - } - - # Removing $dontstrip tags from $elements list (currently only 'gallery', fixing bug 2700) - foreach ( $elements AS $k => $v ) { - if ( !in_array ( $v , $dontstrip ) ) continue; - unset ( $elements[$k] ); - } - - $matches = array(); - $text = self::extractTagsAndParams( $elements, $text, $matches, $uniq_prefix ); - - foreach( $matches as $marker => $data ) { - list( $element, $content, $params, $tag ) = $data; - if( $render ) { - $tagName = strtolower( $element ); - wfProfileIn( __METHOD__."-render-$tagName" ); - switch( $tagName ) { - case '!--': - // Comment - if( substr( $tag, -3 ) == '-->' ) { - $output = $tag; - } else { - // Unclosed comment in input. - // Close it so later stripping can remove it - $output = "$tag-->"; - } - break; - case 'html': - if( $wgRawHtml ) { - $output = $content; - break; - } - // Shouldn't happen otherwise. :) - case 'nowiki': - $output = Xml::escapeTagsOnly( $content ); - break; - case 'math': - $output = $wgContLang->armourMath( - MathRenderer::renderMath( $content, $params ) ); - break; - case 'gallery': - $output = $this->renderImageGallery( $content, $params ); - break; - default: - if( isset( $this->mTagHooks[$tagName] ) ) { - $output = call_user_func_array( $this->mTagHooks[$tagName], - array( $content, $params, $this ) ); - } else { - throw new MWException( "Invalid call hook $element" ); - } - } - wfProfileOut( __METHOD__."-render-$tagName" ); - } else { - // Just stripping tags; keep the source - $output = $tag; - } - - // Unstrip the output, to support recursive strip() calls - $output = $state->unstripBoth( $output ); - - if( !$stripcomments && $element == '!--' ) { - $commentState->setPair( $marker, $output ); - } elseif ( $element == 'html' || $element == 'nowiki' ) { - $nowikiItems[$marker] = $output; - } else { - $generalItems[$marker] = $output; - } - } - # Add the new items to the state - # We do this after the loop instead of during it to avoid slowing - # down the recursive unstrip - $state->nowiki->mergeArray( $nowikiItems ); - $state->general->mergeArray( $generalItems ); - - # Unstrip comments unless explicitly told otherwise. - # (The comments are always stripped prior to this point, so as to - # not invoke any extension tags / parser hooks contained within - # a comment.) - if ( !$stripcomments ) { - // Put them all back and forget them - $text = $commentState->replace( $text ); - } - - wfProfileOut( __METHOD__ ); - return $text; - } - - /** - * Restores pre, math, and other extensions removed by strip() - * - * always call unstripNoWiki() after this one - * @private - * @deprecated use $this->mStripState->unstrip() - */ - function unstrip( $text, $state ) { - return $state->unstripGeneral( $text ); - } - - /** - * Always call this after unstrip() to preserve the order - * - * @private - * @deprecated use $this->mStripState->unstrip() - */ - function unstripNoWiki( $text, $state ) { - return $state->unstripNoWiki( $text ); - } - - /** - * @deprecated use $this->mStripState->unstripBoth() - */ - function unstripForHTML( $text ) { - return $this->mStripState->unstripBoth( $text ); - } - - /** - * Add an item to the strip state - * Returns the unique tag which must be inserted into the stripped text - * The tag will be replaced with the original text in unstrip() - * - * @private - */ - function insertStripItem( $text, &$state ) { - $rnd = $this->mUniqPrefix . '-item' . self::getRandomString(); - $state->general->setPair( $rnd, $text ); - return $rnd; - } - - /** - * Interface with html tidy, used if $wgUseTidy = true. - * If tidy isn't able to correct the markup, the original will be - * returned in all its glory with a warning comment appended. - * - * Either the external tidy program or the in-process tidy extension - * will be used depending on availability. Override the default - * $wgTidyInternal setting to disable the internal if it's not working. - * - * @param string $text Hideous HTML input - * @return string Corrected HTML output - * @public - * @static - */ - function tidy( $text ) { - global $wgTidyInternal; - $wrappedtext = ''. -'test'.$text.''; - if( $wgTidyInternal ) { - $correctedtext = self::internalTidy( $wrappedtext ); - } else { - $correctedtext = self::externalTidy( $wrappedtext ); - } - if( is_null( $correctedtext ) ) { - wfDebug( "Tidy error detected!\n" ); - return $text . "\n\n"; - } - return $correctedtext; - } - - /** - * Spawn an external HTML tidy process and get corrected markup back from it. - * - * @private - * @static - */ - function externalTidy( $text ) { - global $wgTidyConf, $wgTidyBin, $wgTidyOpts; - $fname = 'Parser::externalTidy'; - wfProfileIn( $fname ); - - $cleansource = ''; - $opts = ' -utf8'; - - $descriptorspec = array( - 0 => array('pipe', 'r'), - 1 => array('pipe', 'w'), - 2 => array('file', wfGetNull(), 'a') - ); - $pipes = array(); - $process = proc_open("$wgTidyBin -config $wgTidyConf $wgTidyOpts$opts", $descriptorspec, $pipes); - if (is_resource($process)) { - // Theoretically, this style of communication could cause a deadlock - // here. If the stdout buffer fills up, then writes to stdin could - // block. This doesn't appear to happen with tidy, because tidy only - // writes to stdout after it's finished reading from stdin. Search - // for tidyParseStdin and tidySaveStdout in console/tidy.c - fwrite($pipes[0], $text); - fclose($pipes[0]); - while (!feof($pipes[1])) { - $cleansource .= fgets($pipes[1], 1024); - } - fclose($pipes[1]); - proc_close($process); - } - - wfProfileOut( $fname ); - - if( $cleansource == '' && $text != '') { - // Some kind of error happened, so we couldn't get the corrected text. - // Just give up; we'll use the source text and append a warning. - return null; - } else { - return $cleansource; - } - } - - /** - * Use the HTML tidy PECL extension to use the tidy library in-process, - * saving the overhead of spawning a new process. - * - * 'pear install tidy' should be able to compile the extension module. - * - * @private - * @static - */ - function internalTidy( $text ) { - global $wgTidyConf, $IP; - $fname = 'Parser::internalTidy'; - wfProfileIn( $fname ); - - $tidy = new tidy; - $tidy->parseString( $text, $wgTidyConf, 'utf8' ); - $tidy->cleanRepair(); - if( $tidy->getStatus() == 2 ) { - // 2 is magic number for fatal error - // http://www.php.net/manual/en/function.tidy-get-status.php - $cleansource = null; - } else { - $cleansource = tidy_get_output( $tidy ); - } - wfProfileOut( $fname ); - return $cleansource; - } - - /** - * parse the wiki syntax used to render tables - * - * @private - */ - function doTableStuff ( $text ) { - $fname = 'Parser::doTableStuff'; - wfProfileIn( $fname ); - - $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 element? - $indent_level = 0; // indent level of the table - foreach ( $lines as $key => $line ) - { - $line = trim ( $line ); - - if( $line == '' ) { // empty line, go to next line - continue; - } - $first_character = $line{0}; - $matches = array(); - - if ( preg_match( '/^(:*)\{\|(.*)$/' , $line , $matches ) ) { - // First check if we are starting a new table - $indent_level = strlen( $matches[1] ); - - $attributes = $this->mStripState->unstripBoth( $matches[2] ); - $attributes = Sanitizer::fixTagAttributes ( $attributes , 'table' ); - - $lines[$key] = str_repeat( '
    ' , $indent_level ) . ""; - 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 = '' . substr ( $line , 2 ); - $last_tag = array_pop ( $last_tag_history ); - - if ( !array_pop ( $has_opened_tr ) ) { - $line = "{$line}"; - } - - if ( array_pop ( $tr_history ) ) { - $line = "{$line}"; - } - - if ( array_pop ( $td_history ) ) { - $line = "{$line}"; - } - array_pop ( $tr_attributes ); - $lines[$key] = $line . str_repeat( '
    ' , $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_history ) ) { - $line = ''; - } - - if ( array_pop ( $td_history ) ) { - $line = "{$line}"; - } - - $lines[$key] = $line; - array_push ( $tr_history , false ); - array_push ( $td_history , false ); - array_push ( $last_tag_history , '' ); - } - 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 ); - } - - // Split up multiple cells on the same line. - // FIXME : This can result in improper nesting of tags processed - // by earlier parser steps, but should avoid splitting up eg - // attribute values containing literal "||". - $cells = StringUtils::explodeMarkup( '||' , $line ); - - $lines[$key] = ''; - - // Loop through each table cell - foreach ( $cells as $cell ) - { - $previous = ''; - if ( $first_character != '+' ) - { - $tr_after = array_pop ( $tr_attributes ); - if ( !array_pop ( $tr_history ) ) { - $previous = "\n"; - } - array_push ( $tr_history , true ); - array_push ( $tr_attributes , '' ); - array_pop ( $has_opened_tr ); - array_push ( $has_opened_tr , true ); - } - - $last_tag = array_pop ( $last_tag_history ); - - if ( array_pop ( $td_history ) ) { - $previous = "{$previous}"; - } - - if ( $first_character == '|' ) { - $last_tag = 'td'; - } else if ( $first_character == '!' ) { - $last_tag = 'th'; - } else if ( $first_character == '+' ) { - $last_tag = 'caption'; - } else { - $last_tag = ''; - } - - 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->mStripState->unstripBoth( $cell_data[0] ); - $attributes = Sanitizer::fixTagAttributes( $attributes , $last_tag ); - $cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[1]}"; - } - - $lines[$key] .= $cell; - array_push ( $td_history , true ); - } - } - } - - // Closing open td, tr && table - while ( count ( $td_history ) > 0 ) - { - if ( array_pop ( $td_history ) ) { - $lines[] = '' ; - } - if ( array_pop ( $tr_history ) ) { - $lines[] = '' ; - } - if ( !array_pop ( $has_opened_tr ) ) { - $lines[] = "" ; - } - - $lines[] = '' ; - } - - $output = implode ( "\n" , $lines ) ; - - // special case: don't return empty table - if( $output == "\n\n
    " ) { - $output = ''; - } - - wfProfileOut( $fname ); - - return $output; - } - - /** - * Helper function for parse() that transforms wiki markup into - * HTML. Only called for $mOutputType == OT_HTML. - * - * @private - */ - function internalParse( $text ) { - $args = array(); - $isMain = true; - $fname = 'Parser::internalParse'; - wfProfileIn( $fname ); - - # Hook to suspend the parser in this state - if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) { - wfProfileOut( $fname ); - return $text ; - } - - # Remove tags and sections - $text = strtr( $text, array( '' => '' , '' => '' ) ); - $text = strtr( $text, array( '' => '', '' => '') ); - $text = StringUtils::delimiterReplace( '', '', '', $text ); - - $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), array(), array_keys( $this->mTransparentTagHooks ) ); - - $text = $this->replaceVariables( $text, $args ); - wfRunHooks( 'InternalParseBeforeLinks', array( &$this, &$text, &$this->mStripState ) ); - - // Tables need to come after variable replacement for things to work - // properly; putting them before other transformations should keep - // exciting things like link expansions from showing up in surprising - // places. - $text = $this->doTableStuff( $text ); - - $text = preg_replace( '/(^|\n)-----*/', '\\1
    ', $text ); - - $text = $this->stripToc( $text ); - $this->stripNoGallery( $text ); - $text = $this->doHeadings( $text ); - if($this->mOptions->getUseDynamicDates()) { - $df =& DateFormatter::getInstance(); - $text = $df->reformat( $this->mOptions->getDateFormat(), $text ); - } - $text = $this->doAllQuotes( $text ); - $text = $this->replaceInternalLinks( $text ); - $text = $this->replaceExternalLinks( $text ); - - # replaceInternalLinks may sometimes leave behind - # absolute URLs, which have to be masked to hide them from replaceExternalLinks - $text = str_replace($this->mUniqPrefix."NOPARSE", "", $text); - - $text = $this->doMagicLinks( $text ); - $text = $this->formatHeadings( $text, $isMain ); - - wfProfileOut( $fname ); - return $text; - } - - /** - * Replace special strings like "ISBN xxx" and "RFC xxx" with - * magic external links. - * - * @private - */ - function &doMagicLinks( &$text ) { - wfProfileIn( __METHOD__ ); - $text = preg_replace_callback( - '!(?: # Start cases - | # Skip link text - <.*?> | # Skip stuff inside HTML elements - (?:RFC|PMID)\s+([0-9]+) | # RFC or PMID, capture number as m[1] - ISBN\s+(\b # ISBN, capture number as m[2] - (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix - (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters - [0-9Xx] # check digit - \b) - )!x', array( &$this, 'magicLinkCallback' ), $text ); - wfProfileOut( __METHOD__ ); - return $text; - } - - function magicLinkCallback( $m ) { - if ( substr( $m[0], 0, 1 ) == '<' ) { - # Skip HTML element - return $m[0]; - } elseif ( substr( $m[0], 0, 4 ) == 'ISBN' ) { - $isbn = $m[2]; - $num = strtr( $isbn, array( - '-' => '', - ' ' => '', - 'x' => 'X', - )); - $titleObj = SpecialPage::getTitleFor( 'Booksources' ); - $text = 'ISBN $isbn"; - } else { - if ( substr( $m[0], 0, 3 ) == 'RFC' ) { - $keyword = 'RFC'; - $urlmsg = 'rfcurl'; - $id = $m[1]; - } elseif ( substr( $m[0], 0, 4 ) == 'PMID' ) { - $keyword = 'PMID'; - $urlmsg = 'pubmedurl'; - $id = $m[1]; - } else { - throw new MWException( __METHOD__.': unrecognised match type "' . - substr($m[0], 0, 20 ) . '"' ); - } - - $url = wfMsg( $urlmsg, $id); - $sk = $this->mOptions->getSkin(); - $la = $sk->getExternalLinkAttributes( $url, $keyword.$id ); - $text = "{$keyword} {$id}"; - } - return $text; - } - - /** - * Parse headers and return html - * - * @private - */ - function doHeadings( $text ) { - $fname = 'Parser::doHeadings'; - wfProfileIn( $fname ); - for ( $i = 6; $i >= 1; --$i ) { - $h = str_repeat( '=', $i ); - $text = preg_replace( "/^{$h}(.+){$h}\\s*$/m", - "\\1\\2", $text ); - } - wfProfileOut( $fname ); - return $text; - } - - /** - * Replace single quotes with HTML markup - * @private - * @return string the altered text - */ - function doAllQuotes( $text ) { - $fname = 'Parser::doAllQuotes'; - wfProfileIn( $fname ); - $outtext = ''; - $lines = explode( "\n", $text ); - foreach ( $lines as $line ) { - $outtext .= $this->doQuotes ( $line ) . "\n"; - } - $outtext = substr($outtext, 0,-1); - wfProfileOut( $fname ); - return $outtext; - } - - /** - * Helper function for doAllQuotes() - */ - public function doQuotes( $text ) { - $arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE ); - if ( count( $arr ) == 1 ) - return $text; - else - { - # First, do some preliminary work. This may shift some apostrophes from - # being mark-up to being text. It also counts the number of occurrences - # of bold and italics mark-ups. - $i = 0; - $numbold = 0; - $numitalics = 0; - foreach ( $arr as $r ) - { - 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 ) - { - $arr[$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 ) - { - $arr[$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++; } - } - $i++; - } - - # If there is an odd number of both bold and italics, it is likely - # 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 ) ) - { - $i = 0; - $firstsingleletterword = -1; - $firstmultiletterword = -1; - $firstspace = -1; - foreach ( $arr as $r ) - { - if ( ( $i % 2 == 1 ) and ( strlen( $r ) == 3 ) ) - { - $x1 = substr ($arr[$i-1], -1); - $x2 = substr ($arr[$i-1], -2, 1); - if ($x1 == ' ') { - if ($firstspace == -1) $firstspace = $i; - } else if ($x2 == ' ') { - if ($firstsingleletterword == -1) $firstsingleletterword = $i; - } else { - if ($firstmultiletterword == -1) $firstmultiletterword = $i; - } - } - $i++; - } - - # If there is a single-letter word, use it! - if ($firstsingleletterword > -1) - { - $arr [ $firstsingleletterword ] = "''"; - $arr [ $firstsingleletterword-1 ] .= "'"; - } - # If not, but there's a multi-letter word, use that one. - else if ($firstmultiletterword > -1) - { - $arr [ $firstmultiletterword ] = "''"; - $arr [ $firstmultiletterword-1 ] .= "'"; - } - # ... otherwise use the first one that has neither. - # (notice that it is possible for all three to be -1 if, for example, - # there is only one pentuple-apostrophe in the line) - else if ($firstspace > -1) - { - $arr [ $firstspace ] = "''"; - $arr [ $firstspace-1 ] .= "'"; - } - } - - # Now let's actually convert our apostrophic mush to HTML! - $output = ''; - $buffer = ''; - $state = ''; - $i = 0; - foreach ($arr as $r) - { - if (($i % 2) == 0) - { - if ($state == 'both') - $buffer .= $r; - else - $output .= $r; - } - else - { - if (strlen ($r) == 2) - { - if ($state == 'i') - { $output .= ''; $state = ''; } - else if ($state == 'bi') - { $output .= ''; $state = 'b'; } - else if ($state == 'ib') - { $output .= ''; $state = 'b'; } - else if ($state == 'both') - { $output .= ''.$buffer.''; $state = 'b'; } - else # $state can be 'b' or '' - { $output .= ''; $state .= 'i'; } - } - else if (strlen ($r) == 3) - { - if ($state == 'b') - { $output .= ''; $state = ''; } - else if ($state == 'bi') - { $output .= ''; $state = 'i'; } - else if ($state == 'ib') - { $output .= ''; $state = 'i'; } - else if ($state == 'both') - { $output .= ''.$buffer.''; $state = 'i'; } - else # $state can be 'i' or '' - { $output .= ''; $state .= 'b'; } - } - else if (strlen ($r) == 5) - { - if ($state == 'b') - { $output .= ''; $state = 'i'; } - else if ($state == 'i') - { $output .= ''; $state = 'b'; } - else if ($state == 'bi') - { $output .= ''; $state = ''; } - else if ($state == 'ib') - { $output .= ''; $state = ''; } - else if ($state == 'both') - { $output .= ''.$buffer.''; $state = ''; } - else # ($state == '') - { $buffer = ''; $state = 'both'; } - } - } - $i++; - } - # Now close all remaining tags. Notice that the order is important. - if ($state == 'b' || $state == 'ib') - $output .= ''; - if ($state == 'i' || $state == 'bi' || $state == 'ib') - $output .= ''; - if ($state == 'bi') - $output .= ''; - # There might be lonely ''''', so make sure we have a buffer - if ($state == 'both' && $buffer) - $output .= ''.$buffer.''; - return $output; - } - } - - /** - * Replace external links - * - * 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. - * - * @private - */ - function replaceExternalLinks( $text ) { - global $wgContLang; - $fname = 'Parser::replaceExternalLinks'; - wfProfileIn( $fname ); - - $sk = $this->mOptions->getSkin(); - - $bits = preg_split( $this->mExtLinkBracketedRegex, $text, -1, PREG_SPLIT_DELIM_CAPTURE ); - - $s = $this->replaceFreeExternalLinks( array_shift( $bits ) ); - - $i = 0; - while ( $i' (which were escaped by - # removeHTMLtags()) should not be included in - # URLs, per RFC 2396. - $m2 = array(); - 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 tag - # This happened by accident in the original parser, but some people used it extensively - $img = $this->maybeMakeExternalImage( $text ); - if ( $img !== false ) { - $text = $img; - } - - $dtrail = ''; - - # Set linktype for CSS - if URL==text, link is essentially free - $linktype = ($text == $url) ? 'free' : 'text'; - - # No link text, e.g. [http://domain.tld/some.link] - if ( $text == '' ) { - # Autonumber if allowed. See bug #5918 - if ( strpos( wfUrlProtocols(), substr($protocol, 0, strpos($protocol, ':')) ) !== false ) { - $text = '[' . ++$this->mAutonumber . ']'; - $linktype = 'autonumber'; - } else { - # Otherwise just use the URL - $text = htmlspecialchars( $url ); - $linktype = 'free'; - } - } else { - # Have link text, e.g. [http://domain.tld/some.link text]s - # Check for trail - list( $dtrail, $trail ) = Linker::splitTrail( $trail ); - } - - $text = $wgContLang->markNoConversion($text); - - $url = Sanitizer::cleanUrl( $url ); - - # Process the trail (i.e. everything after this link up until start of the next link), - # replacing any non-bracketed links - $trail = $this->replaceFreeExternalLinks( $trail ); - - # 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 .= $sk->makeExternalLink( $url, $text, false, $linktype, $this->mTitle->getNamespace() ) . $dtrail . $trail; - - # Register link in the output object. - # Replace unnecessary URL escape codes with the referenced character - # This prevents spammers from hiding links from the filters - $pasteurized = self::replaceUnusualEscapes( $url ); - $this->mOutput->addExternalLink( $pasteurized ); - } - - wfProfileOut( $fname ); - return $s; - } - - /** - * Replace anything that looks like a URL with a link - * @private - */ - function replaceFreeExternalLinks( $text ) { - global $wgContLang; - $fname = 'Parser::replaceFreeExternalLinks'; - wfProfileIn( $fname ); - - $bits = preg_split( '/(\b(?:' . wfUrlProtocols() . '))/S', $text, -1, PREG_SPLIT_DELIM_CAPTURE ); - $s = array_shift( $bits ); - $i = 0; - - $sk = $this->mOptions->getSkin(); - - while ( $i < count( $bits ) ){ - $protocol = $bits[$i++]; - $remainder = $bits[$i++]; - - $m = array(); - if ( preg_match( '/^('.self::EXT_LINK_URL_CLASS.'+)(.*)$/s', $remainder, $m ) ) { - # Found some characters after the protocol that look promising - $url = $protocol . $m[1]; - $trail = $m[2]; - - # special case: handle urls as url args: - # http://www.example.com/foo?=http://www.example.com/bar - if(strlen($trail) == 0 && - isset($bits[$i]) && - preg_match('/^'. wfUrlProtocols() . '$/S', $bits[$i]) && - preg_match( '/^('.self::EXT_LINK_URL_CLASS.'+)(.*)$/s', $bits[$i + 1], $m )) - { - # add protocol, arg - $url .= $bits[$i] . $m[1]; # protocol, url as arg to previous link - $i += 2; - $trail = $m[2]; - } - - # The characters '<' and '>' (which were escaped by - # removeHTMLtags()) should not be included in - # URLs, per RFC 2396. - $m2 = array(); - 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 - if ( strpos( $url, '(' ) === false ) { - $sep .= ')'; - } - - $numSepChars = strspn( strrev( $url ), $sep ); - if ( $numSepChars ) { - $trail = substr( $url, -$numSepChars ) . $trail; - $url = substr( $url, 0, -$numSepChars ); - } - - $url = Sanitizer::cleanUrl( $url ); - - # Is this an external image? - $text = $this->maybeMakeExternalImage( $url ); - if ( $text === false ) { - # Not an image, make a link - $text = $sk->makeExternalLink( $url, $wgContLang->markNoConversion($url), true, 'free', $this->mTitle->getNamespace() ); - # Register it in the output object... - # Replace unnecessary URL escape codes with their equivalent characters - $pasteurized = self::replaceUnusualEscapes( $url ); - $this->mOutput->addExternalLink( $pasteurized ); - } - $s .= $text . $trail; - } else { - $s .= $protocol . $remainder; - } - } - wfProfileOut( $fname ); - return $s; - } - - /** - * Replace unusual URL escape codes with their equivalent characters - * @param string - * @return string - * @static - * @todo This can merge genuinely required bits in the path or query string, - * breaking legit URLs. A proper fix would treat the various parts of - * the URL differently; as a workaround, just use the output for - * statistical records, not for actual linking/output. - */ - static function replaceUnusualEscapes( $url ) { - return preg_replace_callback( '/%[0-9A-Fa-f]{2}/', - array( __CLASS__, 'replaceUnusualEscapesCallback' ), $url ); - } - - /** - * Callback function used in replaceUnusualEscapes(). - * Replaces unusual URL escape codes with their equivalent character - * @static - * @private - */ - private static function replaceUnusualEscapesCallback( $matches ) { - $char = urldecode( $matches[0] ); - $ord = ord( $char ); - // Is it an unsafe or HTTP reserved character according to RFC 1738? - if ( $ord > 32 && $ord < 127 && strpos( '<>"#{}|\^~[]`;/?', $char ) === false ) { - // No, shouldn't be escaped - return $char; - } else { - // Yes, leave it escaped - return $matches[0]; - } - } - - /** - * make an image if it's allowed, either through the global - * option or through the exception - * @private - */ - function maybeMakeExternalImage( $url ) { - $sk = $this->mOptions->getSkin(); - $imagesfrom = $this->mOptions->getAllowExternalImagesFrom(); - $imagesexception = !empty($imagesfrom); - $text = false; - if ( $this->mOptions->getAllowExternalImages() - || ( $imagesexception && strpos( $url, $imagesfrom ) === 0 ) ) { - if ( preg_match( self::EXT_IMAGE_REGEX, $url ) ) { - # Image found - $text = $sk->makeExternalImage( htmlspecialchars( $url ) ); - } - } - return $text; - } - - /** - * Process [[ ]] wikilinks - * - * @private - */ - function replaceInternalLinks( $s ) { - global $wgContLang; - 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(); - - #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 - $e2 = wfMsgForContent( 'linkprefix' ); - - $useLinkPrefixExtension = $wgContLang->linkPrefixExtension(); - if( is_null( $this->mTitle ) ) { - throw new MWException( __METHOD__.": \$this->mTitle is null\n" ); - } - $nottalk = !$this->mTitle->isTalkPage(); - - if ( $useLinkPrefixExtension ) { - $m = array(); - if ( preg_match( $e2, $s, $m ) ) { - $first_prefix = $m[2]; - } else { - $first_prefix = false; - } - } else { - $prefix = ''; - } - - if($wgContLang->hasVariants()) { - $selflink = $wgContLang->convertLinkToAllVariants($this->mTitle->getPrefixedText()); - } else { - $selflink = array($this->mTitle->getPrefixedText()); - } - $useSubpages = $this->areSubpagesAllowed(); - wfProfileOut( $fname.'-setup' ); - - # Loop for each link - for ($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]; - } else { - $prefix=''; - } - # first link - if($first_prefix) { - $prefix = $first_prefix; - $first_prefix = false; - } - wfProfileOut( $fname.'-prefixhandling' ); - } - - $might_be_img = false; - - wfProfileIn( "$fname-e1" ); - 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. - # - # Still some problems for cases where the ] is meant to be outside punctuation, - # and no image is in sight. See bug 2095. - # - if( $text !== '' && - substr( $m[3], 0, 1 ) === ']' && - strpos($text, '[') !== false - ) - { - $text .= ']'; # so that replaceExternalLinks($text) works later - $m[3] = substr( $m[3], 1 ); - } - # fix up urlencoded title texts - if( strpos( $m[1], '%' ) !== false ) { - # Should anchors '#' also be rejected? - $m[1] = str_replace( array('<', '>'), array('<', '>'), urldecode($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 ( strpos( $m[1], '%' ) !== false ) { - $m[1] = urldecode($m[1]); - } - $trail = ""; - } else { # Invalid form; output directly - $s .= $prefix . '[[' . $line ; - wfProfileOut( "$fname-e1" ); - continue; - } - wfProfileOut( "$fname-e1" ); - wfProfileIn( "$fname-misc" ); - - # Don't allow internal links to pages containing - # PROTO: where PROTO is a valid URL protocol; these - # should be external links. - if (preg_match('/^\b(?:' . wfUrlProtocols() . ')/', $m[1])) { - $s .= $prefix . '[[' . $line ; - continue; - } - - # Make subpage if necessary - if( $useSubpages ) { - $link = $this->maybeDoSubpageLink( $m[1], $text ); - } else { - $link = $m[1]; - } - - $noforce = (substr($m[1], 0, 1) != ':'); - if (!$noforce) { - # Strip off leading ':' - $link = substr($link, 1); - } - - wfProfileOut( "$fname-misc" ); - wfProfileIn( "$fname-title" ); - $nt = Title::newFromText( $this->mStripState->unstripNoWiki($link) ); - if( !$nt ) { - $s .= $prefix . '[[' . $line; - wfProfileOut( "$fname-title" ); - continue; - } - - $ns = $nt->getNamespace(); - $iw = $nt->getInterWiki(); - wfProfileOut( "$fname-title" ); - - if ($might_be_img) { # if this is actually an invalid link - wfProfileIn( "$fname-might_be_img" ); - 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 - $spliced = array_splice( $a, $k + 1, 1 ); - $next_line = array_shift( $spliced ); - $m = explode( ']]', $next_line, 3 ); - if ( count( $m ) == 3 ) { - # the first ]] closes the inner link, the second the image - $found = true; - $text .= "[[{$m[0]}]]{$m[1]}"; - $trail = $m[2]; - break; - } elseif ( count( $m ) == 2 ) { - #if there's exactly one ]] that's fine, we'll keep looking - $text .= "[[{$m[0]}]]{$m[1]}"; - } 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 $trail, because without an end, there *is* no trail - wfProfileOut( "$fname-might_be_img" ); - continue; - } - } else { #it's not an image, so output it raw - $s .= "{$prefix}[[$link|$text"; - # note: no $trail, because without an end, there *is* no trail - wfProfileOut( "$fname-might_be_img" ); - continue; - } - wfProfileOut( "$fname-might_be_img" ); - } - - $wasblank = ( '' == $text ); - if( $wasblank ) $text = $link; - - # Link not escaped by : , create the various objects - if( $noforce ) { - - # Interwikis - wfProfileIn( "$fname-interwiki" ); - if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) { - $this->mOutput->addLanguageLink( $nt->getFullText() ); - $s = rtrim($s . $prefix); - $s .= trim($trail, "\n") == '' ? '': $prefix . $trail; - wfProfileOut( "$fname-interwiki" ); - continue; - } - wfProfileOut( "$fname-interwiki" ); - - if ( $ns == NS_IMAGE ) { - wfProfileIn( "$fname-image" ); - if ( !wfIsBadImage( $nt->getDBkey(), $this->mTitle ) ) { - # 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 . $this->armorLinks( $this->makeImage( $nt, $text ) ) . $trail; - $this->mOutput->addImage( $nt->getDBkey() ); - - wfProfileOut( "$fname-image" ); - continue; - } else { - # We still need to record the image's presence on the page - $this->mOutput->addImage( $nt->getDBkey() ); - } - wfProfileOut( "$fname-image" ); - - } - - if ( $ns == NS_CATEGORY ) { - wfProfileIn( "$fname-category" ); - $s = rtrim($s . "\n"); # bug 87 - - if ( $wasblank ) { - $sortkey = $this->getDefaultSort(); - } else { - $sortkey = $text; - } - $sortkey = Sanitizer::decodeCharReferences( $sortkey ); - $sortkey = str_replace( "\n", '', $sortkey ); - $sortkey = $wgContLang->convertCategoryKey( $sortkey ); - $this->mOutput->addCategory( $nt->getDBkey(), $sortkey ); - - /** - * Strip the whitespace Category links produce, see bug 87 - * @todo We might want to use trim($tmp, "\n") here. - */ - $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail; - - wfProfileOut( "$fname-category" ); - continue; - } - } - - # Self-link checking - if( $nt->getFragment() === '' ) { - if( in_array( $nt->getPrefixedText(), $selflink, true ) ) { - $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail ); - continue; - } - } - - # Special and Media are pseudo-namespaces; no pages actually exist in them - if( $ns == NS_MEDIA ) { - $link = $sk->makeMediaLinkObj( $nt, $text ); - # Cloak with NOPARSE to avoid replacement in replaceExternalLinks - $s .= $prefix . $this->armorLinks( $link ) . $trail; - $this->mOutput->addImage( $nt->getDBkey() ); - continue; - } elseif( $ns == NS_SPECIAL ) { - if( SpecialPage::exists( $nt->getDBkey() ) ) { - $s .= $this->makeKnownLinkHolder( $nt, $text, '', $trail, $prefix ); - } else { - $s .= $this->makeLinkHolder( $nt, $text, '', $trail, $prefix ); - } - continue; - } elseif( $ns == NS_IMAGE ) { - $img = wfFindFile( $nt ); - if( $img ) { - // Force a blue link if the file exists; may be a remote - // upload on the shared repository, and we want to see its - // auto-generated page. - $s .= $this->makeKnownLinkHolder( $nt, $text, '', $trail, $prefix ); - $this->mOutput->addLink( $nt ); - continue; - } - } - $s .= $this->makeLinkHolder( $nt, $text, '', $trail, $prefix ); - } - wfProfileOut( $fname ); - return $s; - } - - /** - * Make a link placeholder. The text returned can be later resolved to a real link with - * replaceLinkHolders(). This is done for two reasons: firstly to avoid further - * parsing of interwiki links, and secondly to allow all existence checks and - * article length checks (for stub links) to be bundled into a single query. - * - */ - function makeLinkHolder( &$nt, $text = '', $query = '', $trail = '', $prefix = '' ) { - wfProfileIn( __METHOD__ ); - if ( ! is_object($nt) ) { - # Fail gracefully - $retVal = "{$prefix}{$text}{$trail}"; - } else { - # Separate the link trail from the rest of the link - list( $inside, $trail ) = Linker::splitTrail( $trail ); - - if ( $nt->isExternal() ) { - $nr = array_push( $this->mInterwikiLinkHolders['texts'], $prefix.$text.$inside ); - $this->mInterwikiLinkHolders['titles'][] = $nt; - $retVal = '{$trail}"; - } else { - $nr = array_push( $this->mLinkHolders['namespaces'], $nt->getNamespace() ); - $this->mLinkHolders['dbkeys'][] = $nt->getDBkey(); - $this->mLinkHolders['queries'][] = $query; - $this->mLinkHolders['texts'][] = $prefix.$text.$inside; - $this->mLinkHolders['titles'][] = $nt; - - $retVal = '{$trail}"; - } - } - wfProfileOut( __METHOD__ ); - return $retVal; - } - - /** - * Render a forced-blue link inline; protect against double expansion of - * URLs if we're in a mode that prepends full URL prefixes to internal links. - * Since this little disaster has to split off the trail text to avoid - * breaking URLs in the following text without breaking trails on the - * wiki links, it's been made into a horrible function. - * - * @param Title $nt - * @param string $text - * @param string $query - * @param string $trail - * @param string $prefix - * @return string HTML-wikitext mix oh yuck - */ - function makeKnownLinkHolder( $nt, $text = '', $query = '', $trail = '', $prefix = '' ) { - list( $inside, $trail ) = Linker::splitTrail( $trail ); - $sk = $this->mOptions->getSkin(); - $link = $sk->makeKnownLinkObj( $nt, $text, $query, $inside, $prefix ); - return $this->armorLinks( $link ) . $trail; - } - - /** - * Insert a NOPARSE hacky thing into any inline links in a chunk that's - * going to go through further parsing steps before inline URL expansion. - * - * In particular this is important when using action=render, which causes - * full URLs to be included. - * - * Oh man I hate our multi-layer parser! - * - * @param string more-or-less HTML - * @return string less-or-more HTML with NOPARSE bits - */ - function armorLinks( $text ) { - return preg_replace( '/\b(' . wfUrlProtocols() . ')/', - "{$this->mUniqPrefix}NOPARSE$1", $text ); - } - - /** - * Return true if subpage links should be expanded on this page. - * @return bool - */ - function areSubpagesAllowed() { - # Some namespaces don't allow subpages - global $wgNamespacesWithSubpages; - return !empty($wgNamespacesWithSubpages[$this->mTitle->getNamespace()]); - } - - /** - * Handle link to subpage if necessary - * @param string $target the source of the link - * @param string &$text the link text, modified as necessary - * @return string the full name of the link - * @private - */ - function maybeDoSubpageLink($target, &$text) { - # 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 - # ../ -- convert to CurrentPage, from CurrentPage/CurrentSubPage - # ../Foobar -- convert to CurrentPage/Foobar, from CurrentPage/CurrentSubPage - - $fname = 'Parser::maybeDoSubpageLink'; - wfProfileIn( $fname ); - $ret = $target; # default return value is no change - - # Some namespaces don't allow subpages, - # so only perform processing if subpages are allowed - if( $this->areSubpagesAllowed() ) { - $hash = strpos( $target, '#' ); - if( $hash !== false ) { - $suffix = substr( $target, $hash ); - $target = substr( $target, 0, $hash ); - } else { - $suffix = ''; - } - # bug 7425 - $target = trim( $target ); - # Look at the first character - if( $target != '' && $target{0} == '/' ) { - # / at end means we don't want the slash to be shown - $m = array(); - $trailingSlashes = preg_match_all( '%(/+)$%', $target, $m ); - if( $trailingSlashes ) { - $noslash = $target = substr( $target, 1, -strlen($m[0][0]) ); - } else { - $noslash = substr( $target, 1 ); - } - - $ret = $this->mTitle->getPrefixedText(). '/' . trim($noslash) . $suffix; - if( '' === $text ) { - $text = $target . $suffix; - } # this might be changed for ugliness reasons - } else { - # check for .. subpage backlinks - $dotdotcount = 0; - $nodotdot = $target; - while( strncmp( $nodotdot, "../", 3 ) == 0 ) { - ++$dotdotcount; - $nodotdot = substr( $nodotdot, 3 ); - } - if($dotdotcount > 0) { - $exploded = explode( '/', $this->mTitle->GetPrefixedText() ); - if( count( $exploded ) > $dotdotcount ) { # not allowed to go below top level page - $ret = implode( '/', array_slice( $exploded, 0, -$dotdotcount ) ); - # / at the end means don't show full path - if( substr( $nodotdot, -1, 1 ) == '/' ) { - $nodotdot = substr( $nodotdot, 0, -1 ); - if( '' === $text ) { - $text = $nodotdot . $suffix; - } - } - $nodotdot = trim( $nodotdot ); - if( $nodotdot != '' ) { - $ret .= '/' . $nodotdot; - } - $ret .= $suffix; - } - } - } - } - - wfProfileOut( $fname ); - return $ret; - } - - /**#@+ - * Used by doBlockLevels() - * @private - */ - /* private */ function closeParagraph() { - $result = ''; - if ( '' != $this->mLastSection ) { - $result = 'mLastSection . ">\n"; - } - $this->mInPre = false; - $this->mLastSection = ''; - return $result; - } - # getCommon() returns the length of the longest common substring - # of both arguments, starting at the beginning of both. - # - /* private */ function getCommon( $st1, $st2 ) { - $fl = strlen( $st1 ); - $shorter = strlen( $st2 ); - if ( $fl < $shorter ) { $shorter = $fl; } - - for ( $i = 0; $i < $shorter; ++$i ) { - if ( $st1{$i} != $st2{$i} ) { break; } - } - return $i; - } - # These next three functions open, continue, and close the list - # element appropriate to the prefix character passed into them. - # - /* private */ function openList( $char ) { - $result = $this->closeParagraph(); - - if ( '*' == $char ) { $result .= '
    • '; } - else if ( '#' == $char ) { $result .= '
      1. '; } - else if ( ':' == $char ) { $result .= '
        '; } - else if ( ';' == $char ) { - $result .= '
        '; - $this->mDTopen = true; - } - else { $result = ''; } - - return $result; - } - - /* private */ function nextItem( $char ) { - if ( '*' == $char || '#' == $char ) { return '
      2. '; } - else if ( ':' == $char || ';' == $char ) { - $close = ''; - if ( $this->mDTopen ) { $close = ''; } - if ( ';' == $char ) { - $this->mDTopen = true; - return $close . '
        '; - } else { - $this->mDTopen = false; - return $close . '
        '; - } - } - return ''; - } - - /* private */ function closeList( $char ) { - if ( '*' == $char ) { $text = '
    '; } - else if ( '#' == $char ) { $text = ''; } - else if ( ':' == $char ) { - if ( $this->mDTopen ) { - $this->mDTopen = false; - $text = ''; - } else { - $text = ''; - } - } - else { return ''; } - return $text."\n"; - } - /**#@-*/ - - /** - * Make lists from lines starting with ':', '*', '#', etc. - * - * @private - * @return string the lists rendered as HTML - */ - function doBlockLevels( $text, $linestart ) { - $fname = 'Parser::doBlockLevels'; - wfProfileIn( $fname ); - - # Parsing through the text line by line. The main thing - # happening here is handling of block-level elements p, pre, - # and making lists from lines starting with * # : etc. - # - $textLines = explode( "\n", $text ); - - $lastPrefix = $output = ''; - $this->mDTopen = $inBlockElem = false; - $prefixLength = 0; - $paragraphStack = false; - - if ( !$linestart ) { - $output .= array_shift( $textLines ); - } - foreach ( $textLines as $oLine ) { - $lastPrefixLength = strlen( $lastPrefix ); - $preCloseMatch = preg_match('/<\\/pre/i', $oLine ); - $preOpenMatch = preg_match('/
    mInPre ) {
    -				# Multiple prefixes may abut each other for nested lists.
    -				$prefixLength = strspn( $oLine, '*#:;' );
    -				$pref = substr( $oLine, 0, $prefixLength );
    -
    -				# eh?
    -				$pref2 = str_replace( ';', ':', $pref );
    -				$t = substr( $oLine, $prefixLength );
    -				$this->mInPre = !empty($preOpenMatch);
    -			} else {
    -				# Don't interpret any other prefixes in preformatted text
    -				$prefixLength = 0;
    -				$pref = $pref2 = '';
    -				$t = $oLine;
    -			}
    -
    -			# List generation
    -			if( $prefixLength && 0 == strcmp( $lastPrefix, $pref2 ) ) {
    -				# Same as the last item, so no need to deal with nesting or opening stuff
    -				$output .= $this->nextItem( substr( $pref, -1 ) );
    -				$paragraphStack = false;
    -
    -				if ( substr( $pref, -1 ) == ';') {
    -					# The one nasty exception: definition lists work like this:
    -					# ; title : definition text
    -					# So we check for : in the remainder text to split up the
    -					# title and definition, without b0rking links.
    -					$term = $t2 = '';
    -					if ($this->findColonNoLinks($t, $term, $t2) !== false) {
    -						$t = $t2;
    -						$output .= $term . $this->nextItem( ':' );
    -					}
    -				}
    -			} elseif( $prefixLength || $lastPrefixLength ) {
    -				# Either open or close a level...
    -				$commonPrefixLength = $this->getCommon( $pref, $lastPrefix );
    -				$paragraphStack = false;
    -
    -				while( $commonPrefixLength < $lastPrefixLength ) {
    -					$output .= $this->closeList( $lastPrefix{$lastPrefixLength-1} );
    -					--$lastPrefixLength;
    -				}
    -				if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) {
    -					$output .= $this->nextItem( $pref{$commonPrefixLength-1} );
    -				}
    -				while ( $prefixLength > $commonPrefixLength ) {
    -					$char = substr( $pref, $commonPrefixLength, 1 );
    -					$output .= $this->openList( $char );
    -
    -					if ( ';' == $char ) {
    -						# FIXME: This is dupe of code above
    -						if ($this->findColonNoLinks($t, $term, $t2) !== false) {
    -							$t = $t2;
    -							$output .= $term . $this->nextItem( ':' );
    -						}
    -					}
    -					++$commonPrefixLength;
    -				}
    -				$lastPrefix = $pref2;
    -			}
    -			if( 0 == $prefixLength ) {
    -				wfProfileIn( "$fname-paragraph" );
    -				# No prefix (not in list)--go to paragraph mode
    -				// XXX: use a stack for nestable elements like span, table and div
    -				$openmatch = preg_match('/(?:mUniqPrefix.'-pre|<\\/li|<\\/ul|<\\/ol|<\\/?center)/iS', $t );
    -				if ( $openmatch or $closematch ) {
    -					$paragraphStack = false;
    -					# TODO bug 5718: paragraph closed
    -					$output .= $this->closeParagraph();
    -					if ( $preOpenMatch and !$preCloseMatch ) {
    -						$this->mInPre = true;
    -					}
    -					if ( $closematch ) {
    -						$inBlockElem = false;
    -					} else {
    -						$inBlockElem = true;
    -					}
    -				} else if ( !$inBlockElem && !$this->mInPre ) {
    -					if ( ' ' == $t{0} and ( $this->mLastSection == 'pre' or trim($t) != '' ) ) {
    -						// pre
    -						if ($this->mLastSection != 'pre') {
    -							$paragraphStack = false;
    -							$output .= $this->closeParagraph().'
    ';
    -							$this->mLastSection = 'pre';
    -						}
    -						$t = substr( $t, 1 );
    -					} else {
    -						// paragraph
    -						if ( '' == trim($t) ) {
    -							if ( $paragraphStack ) {
    -								$output .= $paragraphStack.'
    '; - $paragraphStack = false; - $this->mLastSection = 'p'; - } else { - if ($this->mLastSection != 'p' ) { - $output .= $this->closeParagraph(); - $this->mLastSection = ''; - $paragraphStack = '

    '; - } else { - $paragraphStack = '

    '; - } - } - } else { - if ( $paragraphStack ) { - $output .= $paragraphStack; - $paragraphStack = false; - $this->mLastSection = 'p'; - } else if ($this->mLastSection != 'p') { - $output .= $this->closeParagraph().'

    '; - $this->mLastSection = 'p'; - } - } - } - } - wfProfileOut( "$fname-paragraph" ); - } - // somewhere above we forget to get out of pre block (bug 785) - if($preCloseMatch && $this->mInPre) { - $this->mInPre = false; - } - if ($paragraphStack === false) { - $output .= $t."\n"; - } - } - while ( $prefixLength ) { - $output .= $this->closeList( $pref2{$prefixLength-1} ); - --$prefixLength; - } - if ( '' != $this->mLastSection ) { - $output .= 'mLastSection . '>'; - $this->mLastSection = ''; - } - - wfProfileOut( $fname ); - return $output; - } - - /** - * Split up a string on ':', ignoring any occurences inside tags - * to prevent illegal overlapping. - * @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) { - $fname = 'Parser::findColonNoLinks'; - wfProfileIn( $fname ); - - $pos = strpos( $str, ':' ); - if( $pos === false ) { - // Nothing to find! - wfProfileOut( $fname ); - return false; - } - - $lt = strpos( $str, '<' ); - if( $lt === false || $lt > $pos ) { - // Easy; no tag nesting to worry about - $before = substr( $str, 0, $pos ); - $after = substr( $str, $pos+1 ); - wfProfileOut( $fname ); - return $pos; - } - - // Ugly state machine to walk through avoiding tags. - $state = self::COLON_STATE_TEXT; - $stack = 0; - $len = strlen( $str ); - for( $i = 0; $i < $len; $i++ ) { - $c = $str{$i}; - - switch( $state ) { - // (Using the number is a performance hack for common cases) - case 0: // self::COLON_STATE_TEXT: - switch( $c ) { - case "<": - // Could be either a tag or an tag - $state = self::COLON_STATE_TAGSTART; - break; - case ":": - if( $stack == 0 ) { - // We found it! - $before = substr( $str, 0, $i ); - $after = substr( $str, $i + 1 ); - wfProfileOut( $fname ); - return $i; - } - // Embedded in a tag; don't break it. - break; - default: - // Skip ahead looking for something interesting - $colon = strpos( $str, ':', $i ); - if( $colon === false ) { - // Nothing else interesting - wfProfileOut( $fname ); - return false; - } - $lt = strpos( $str, '<', $i ); - if( $stack === 0 ) { - if( $lt === false || $colon < $lt ) { - // We found it! - $before = substr( $str, 0, $colon ); - $after = substr( $str, $colon + 1 ); - wfProfileOut( $fname ); - return $i; - } - } - if( $lt === false ) { - // Nothing else interesting to find; abort! - // We're nested, but there's no close tags left. Abort! - break 2; - } - // Skip ahead to next tag start - $i = $lt; - $state = self::COLON_STATE_TAGSTART; - } - break; - case 1: // self::COLON_STATE_TAG: - // In a - switch( $c ) { - case ">": - $stack++; - $state = self::COLON_STATE_TEXT; - break; - case "/": - // Slash may be followed by >? - $state = self::COLON_STATE_TAGSLASH; - break; - default: - // ignore - } - break; - case 2: // self::COLON_STATE_TAGSTART: - switch( $c ) { - case "/": - $state = self::COLON_STATE_CLOSETAG; - break; - case "!": - $state = self::COLON_STATE_COMMENT; - break; - case ">": - // Illegal early close? This shouldn't happen D: - $state = self::COLON_STATE_TEXT; - break; - default: - $state = self::COLON_STATE_TAG; - } - break; - case 3: // self::COLON_STATE_CLOSETAG: - // In a - if( $c == ">" ) { - $stack--; - if( $stack < 0 ) { - wfDebug( "Invalid input in $fname; too many close tags\n" ); - wfProfileOut( $fname ); - return false; - } - $state = self::COLON_STATE_TEXT; - } - break; - case self::COLON_STATE_TAGSLASH: - if( $c == ">" ) { - // Yes, a self-closed tag - $state = self::COLON_STATE_TEXT; - } else { - // Probably we're jumping the gun, and this is an attribute - $state = self::COLON_STATE_TAG; - } - break; - case 5: // self::COLON_STATE_COMMENT: - if( $c == "-" ) { - $state = self::COLON_STATE_COMMENTDASH; - } - break; - case self::COLON_STATE_COMMENTDASH: - if( $c == "-" ) { - $state = self::COLON_STATE_COMMENTDASHDASH; - } else { - $state = self::COLON_STATE_COMMENT; - } - break; - case self::COLON_STATE_COMMENTDASHDASH: - if( $c == ">" ) { - $state = self::COLON_STATE_TEXT; - } else { - $state = self::COLON_STATE_COMMENT; - } - break; - default: - throw new MWException( "State machine error in $fname" ); - } - } - if( $stack > 0 ) { - wfDebug( "Invalid input in $fname; not enough close tags (stack $stack, state $state)\n" ); - return false; - } - wfProfileOut( $fname ); - return false; - } - - /** - * Return value of a magic variable (like PAGENAME) - * - * @private - */ - function getVariableValue( $index ) { - global $wgContLang, $wgSitename, $wgServer, $wgServerName, $wgScriptPath; - - /** - * Some of these require message or data lookups and can be - * expensive to check many times. - */ - static $varCache = array(); - if ( wfRunHooks( 'ParserGetVariableValueVarCache', array( &$this, &$varCache ) ) ) { - if ( isset( $varCache[$index] ) ) { - return $varCache[$index]; - } - } - - $ts = time(); - wfRunHooks( 'ParserGetVariableValueTs', array( &$this, &$ts ) ); - - # Use the time zone - global $wgLocaltimezone; - if ( isset( $wgLocaltimezone ) ) { - $oldtz = getenv( 'TZ' ); - putenv( 'TZ='.$wgLocaltimezone ); - } - - wfSuppressWarnings(); // E_STRICT system time bitching - $localTimestamp = date( 'YmdHis', $ts ); - $localMonth = date( 'm', $ts ); - $localMonthName = date( 'n', $ts ); - $localDay = date( 'j', $ts ); - $localDay2 = date( 'd', $ts ); - $localDayOfWeek = date( 'w', $ts ); - $localWeek = date( 'W', $ts ); - $localYear = date( 'Y', $ts ); - $localHour = date( 'H', $ts ); - if ( isset( $wgLocaltimezone ) ) { - putenv( 'TZ='.$oldtz ); - } - wfRestoreWarnings(); - - switch ( $index ) { - case 'currentmonth': - return $varCache[$index] = $wgContLang->formatNum( gmdate( 'm', $ts ) ); - case 'currentmonthname': - return $varCache[$index] = $wgContLang->getMonthName( gmdate( 'n', $ts ) ); - case 'currentmonthnamegen': - return $varCache[$index] = $wgContLang->getMonthNameGen( gmdate( 'n', $ts ) ); - case 'currentmonthabbrev': - return $varCache[$index] = $wgContLang->getMonthAbbreviation( gmdate( 'n', $ts ) ); - case 'currentday': - return $varCache[$index] = $wgContLang->formatNum( gmdate( 'j', $ts ) ); - case 'currentday2': - return $varCache[$index] = $wgContLang->formatNum( gmdate( 'd', $ts ) ); - case 'localmonth': - return $varCache[$index] = $wgContLang->formatNum( $localMonth ); - case 'localmonthname': - return $varCache[$index] = $wgContLang->getMonthName( $localMonthName ); - case 'localmonthnamegen': - return $varCache[$index] = $wgContLang->getMonthNameGen( $localMonthName ); - case 'localmonthabbrev': - return $varCache[$index] = $wgContLang->getMonthAbbreviation( $localMonthName ); - case 'localday': - return $varCache[$index] = $wgContLang->formatNum( $localDay ); - case 'localday2': - return $varCache[$index] = $wgContLang->formatNum( $localDay2 ); - case 'pagename': - return wfEscapeWikiText( $this->mTitle->getText() ); - case 'pagenamee': - return $this->mTitle->getPartialURL(); - case 'fullpagename': - return wfEscapeWikiText( $this->mTitle->getPrefixedText() ); - case 'fullpagenamee': - return $this->mTitle->getPrefixedURL(); - case 'subpagename': - return wfEscapeWikiText( $this->mTitle->getSubpageText() ); - case 'subpagenamee': - return $this->mTitle->getSubpageUrlForm(); - case 'basepagename': - return wfEscapeWikiText( $this->mTitle->getBaseText() ); - case 'basepagenamee': - return wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ); - case 'talkpagename': - if( $this->mTitle->canTalk() ) { - $talkPage = $this->mTitle->getTalkPage(); - return wfEscapeWikiText( $talkPage->getPrefixedText() ); - } else { - return ''; - } - case 'talkpagenamee': - if( $this->mTitle->canTalk() ) { - $talkPage = $this->mTitle->getTalkPage(); - return $talkPage->getPrefixedUrl(); - } else { - return ''; - } - case 'subjectpagename': - $subjPage = $this->mTitle->getSubjectPage(); - return wfEscapeWikiText( $subjPage->getPrefixedText() ); - case 'subjectpagenamee': - $subjPage = $this->mTitle->getSubjectPage(); - return $subjPage->getPrefixedUrl(); - case 'revisionid': - return $this->mRevisionId; - case 'revisionday': - return intval( substr( $this->getRevisionTimestamp(), 6, 2 ) ); - case 'revisionday2': - return substr( $this->getRevisionTimestamp(), 6, 2 ); - case 'revisionmonth': - return intval( substr( $this->getRevisionTimestamp(), 4, 2 ) ); - case 'revisionyear': - return substr( $this->getRevisionTimestamp(), 0, 4 ); - case 'revisiontimestamp': - return $this->getRevisionTimestamp(); - case 'namespace': - return str_replace('_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) ); - case 'namespacee': - return wfUrlencode( $wgContLang->getNsText( $this->mTitle->getNamespace() ) ); - case 'talkspace': - return $this->mTitle->canTalk() ? str_replace('_',' ',$this->mTitle->getTalkNsText()) : ''; - case 'talkspacee': - return $this->mTitle->canTalk() ? wfUrlencode( $this->mTitle->getTalkNsText() ) : ''; - case 'subjectspace': - return $this->mTitle->getSubjectNsText(); - case 'subjectspacee': - return( wfUrlencode( $this->mTitle->getSubjectNsText() ) ); - case 'currentdayname': - return $varCache[$index] = $wgContLang->getWeekdayName( gmdate( 'w', $ts ) + 1 ); - case 'currentyear': - return $varCache[$index] = $wgContLang->formatNum( gmdate( 'Y', $ts ), true ); - case 'currenttime': - return $varCache[$index] = $wgContLang->time( wfTimestamp( TS_MW, $ts ), false, false ); - case 'currenthour': - return $varCache[$index] = $wgContLang->formatNum( gmdate( 'H', $ts ), true ); - case 'currentweek': - // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to - // int to remove the padding - return $varCache[$index] = $wgContLang->formatNum( (int)gmdate( 'W', $ts ) ); - case 'currentdow': - return $varCache[$index] = $wgContLang->formatNum( gmdate( 'w', $ts ) ); - case 'localdayname': - return $varCache[$index] = $wgContLang->getWeekdayName( $localDayOfWeek + 1 ); - case 'localyear': - return $varCache[$index] = $wgContLang->formatNum( $localYear, true ); - case 'localtime': - return $varCache[$index] = $wgContLang->time( $localTimestamp, false, false ); - case 'localhour': - return $varCache[$index] = $wgContLang->formatNum( $localHour, true ); - case 'localweek': - // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to - // int to remove the padding - return $varCache[$index] = $wgContLang->formatNum( (int)$localWeek ); - case 'localdow': - return $varCache[$index] = $wgContLang->formatNum( $localDayOfWeek ); - case 'numberofarticles': - return $varCache[$index] = $wgContLang->formatNum( SiteStats::articles() ); - case 'numberoffiles': - return $varCache[$index] = $wgContLang->formatNum( SiteStats::images() ); - case 'numberofusers': - return $varCache[$index] = $wgContLang->formatNum( SiteStats::users() ); - case 'numberofpages': - return $varCache[$index] = $wgContLang->formatNum( SiteStats::pages() ); - case 'numberofadmins': - return $varCache[$index] = $wgContLang->formatNum( SiteStats::admins() ); - case 'numberofedits': - return $varCache[$index] = $wgContLang->formatNum( SiteStats::edits() ); - case 'currenttimestamp': - return $varCache[$index] = wfTimestampNow(); - case 'localtimestamp': - return $varCache[$index] = $localTimestamp; - case 'currentversion': - return $varCache[$index] = SpecialVersion::getVersion(); - case 'sitename': - return $wgSitename; - case 'server': - return $wgServer; - case 'servername': - return $wgServerName; - case 'scriptpath': - return $wgScriptPath; - case 'directionmark': - return $wgContLang->getDirMark(); - case 'contentlanguage': - global $wgContLanguageCode; - return $wgContLanguageCode; - default: - $ret = null; - if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$varCache, &$index, &$ret ) ) ) - return $ret; - else - return null; - } - } - - /** - * initialise the magic variables (like CURRENTMONTHNAME) - * - * @private - */ - function initialiseVariables() { - $fname = 'Parser::initialiseVariables'; - wfProfileIn( $fname ); - $variableIDs = MagicWord::getVariableIDs(); - - $this->mVariables = array(); - foreach ( $variableIDs as $id ) { - $mw =& MagicWord::get( $id ); - $mw->addToArray( $this->mVariables, $id ); - } - wfProfileOut( $fname ); - } - - /** - * parse any parentheses in format ((title|part|part)) - * and call callbacks to get a replacement text for any found piece - * - * @param string $text The text to parse - * @param array $callbacks rules in form: - * '{' => array( # opening parentheses - * 'end' => '}', # closing parentheses - * 'cb' => array(2 => callback, # replacement callback to call if {{..}} is found - * 3 => callback # replacement callback to call if {{{..}}} is found - * ) - * ) - * 'min' => 2, # Minimum parenthesis count in cb - * 'max' => 3, # Maximum parenthesis count in cb - * @private - */ - function replace_callback ($text, $callbacks) { - wfProfileIn( __METHOD__ ); - $openingBraceStack = array(); # this array will hold a stack of parentheses which are not closed yet - $lastOpeningBrace = -1; # last not closed parentheses - - $validOpeningBraces = implode( '', array_keys( $callbacks ) ); - - $i = 0; - while ( $i < strlen( $text ) ) { - # Find next opening brace, closing brace or pipe - if ( $lastOpeningBrace == -1 ) { - $currentClosing = ''; - $search = $validOpeningBraces; - } else { - $currentClosing = $openingBraceStack[$lastOpeningBrace]['braceEnd']; - $search = $validOpeningBraces . '|' . $currentClosing; - } - $rule = null; - $i += strcspn( $text, $search, $i ); - if ( $i < strlen( $text ) ) { - if ( $text[$i] == '|' ) { - $found = 'pipe'; - } elseif ( $text[$i] == $currentClosing ) { - $found = 'close'; - } elseif ( isset( $callbacks[$text[$i]] ) ) { - $found = 'open'; - $rule = $callbacks[$text[$i]]; - } else { - # Some versions of PHP have a strcspn which stops on null characters - # Ignore and continue - ++$i; - continue; - } - } else { - # All done - break; - } - - if ( $found == 'open' ) { - # found opening brace, let's add it to parentheses stack - $piece = array('brace' => $text[$i], - 'braceEnd' => $rule['end'], - 'title' => '', - 'parts' => null); - - # count opening brace characters - $piece['count'] = strspn( $text, $piece['brace'], $i ); - $piece['startAt'] = $piece['partStart'] = $i + $piece['count']; - $i += $piece['count']; - - # we need to add to stack only if opening brace count is enough for one of the rules - if ( $piece['count'] >= $rule['min'] ) { - $lastOpeningBrace ++; - $openingBraceStack[$lastOpeningBrace] = $piece; - } - } elseif ( $found == 'close' ) { - # lets check if it is enough characters for closing brace - $maxCount = $openingBraceStack[$lastOpeningBrace]['count']; - $count = strspn( $text, $text[$i], $i, $maxCount ); - - # check for maximum matching characters (if there are 5 closing - # characters, we will probably need only 3 - depending on the rules) - $matchingCount = 0; - $matchingCallback = null; - $cbType = $callbacks[$openingBraceStack[$lastOpeningBrace]['brace']]; - if ( $count > $cbType['max'] ) { - # The specified maximum exists in the callback array, unless the caller - # has made an error - $matchingCount = $cbType['max']; - } else { - # Count is less than the maximum - # Skip any gaps in the callback array to find the true largest match - # Need to use array_key_exists not isset because the callback can be null - $matchingCount = $count; - while ( $matchingCount > 0 && !array_key_exists( $matchingCount, $cbType['cb'] ) ) { - --$matchingCount; - } - } - - if ($matchingCount <= 0) { - $i += $count; - continue; - } - $matchingCallback = $cbType['cb'][$matchingCount]; - - # let's set a title or last part (if '|' was found) - if (null === $openingBraceStack[$lastOpeningBrace]['parts']) { - $openingBraceStack[$lastOpeningBrace]['title'] = - substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], - $i - $openingBraceStack[$lastOpeningBrace]['partStart']); - } else { - $openingBraceStack[$lastOpeningBrace]['parts'][] = - substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], - $i - $openingBraceStack[$lastOpeningBrace]['partStart']); - } - - $pieceStart = $openingBraceStack[$lastOpeningBrace]['startAt'] - $matchingCount; - $pieceEnd = $i + $matchingCount; - - if( is_callable( $matchingCallback ) ) { - $cbArgs = array ( - 'text' => substr($text, $pieceStart, $pieceEnd - $pieceStart), - 'title' => trim($openingBraceStack[$lastOpeningBrace]['title']), - 'parts' => $openingBraceStack[$lastOpeningBrace]['parts'], - 'lineStart' => (($pieceStart > 0) && ($text[$pieceStart-1] == "\n")), - ); - # finally we can call a user callback and replace piece of text - $replaceWith = call_user_func( $matchingCallback, $cbArgs ); - $text = substr($text, 0, $pieceStart) . $replaceWith . substr($text, $pieceEnd); - $i = $pieceStart + strlen($replaceWith); - } else { - # null value for callback means that parentheses should be parsed, but not replaced - $i += $matchingCount; - } - - # reset last opening parentheses, but keep it in case there are unused characters - $piece = array('brace' => $openingBraceStack[$lastOpeningBrace]['brace'], - 'braceEnd' => $openingBraceStack[$lastOpeningBrace]['braceEnd'], - 'count' => $openingBraceStack[$lastOpeningBrace]['count'], - 'title' => '', - 'parts' => null, - 'startAt' => $openingBraceStack[$lastOpeningBrace]['startAt']); - $openingBraceStack[$lastOpeningBrace--] = null; - - if ($matchingCount < $piece['count']) { - $piece['count'] -= $matchingCount; - $piece['startAt'] -= $matchingCount; - $piece['partStart'] = $piece['startAt']; - # do we still qualify for any callback with remaining count? - $currentCbList = $callbacks[$piece['brace']]['cb']; - while ( $piece['count'] ) { - if ( array_key_exists( $piece['count'], $currentCbList ) ) { - $lastOpeningBrace++; - $openingBraceStack[$lastOpeningBrace] = $piece; - break; - } - --$piece['count']; - } - } - } elseif ( $found == 'pipe' ) { - # lets set a title if it is a first separator, or next part otherwise - if (null === $openingBraceStack[$lastOpeningBrace]['parts']) { - $openingBraceStack[$lastOpeningBrace]['title'] = - substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], - $i - $openingBraceStack[$lastOpeningBrace]['partStart']); - $openingBraceStack[$lastOpeningBrace]['parts'] = array(); - } else { - $openingBraceStack[$lastOpeningBrace]['parts'][] = - substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], - $i - $openingBraceStack[$lastOpeningBrace]['partStart']); - } - $openingBraceStack[$lastOpeningBrace]['partStart'] = ++$i; - } - } - - wfProfileOut( __METHOD__ ); - return $text; - } - - /** - * Replace magic variables, templates, and template arguments - * with the appropriate text. Templates are substituted recursively, - * taking care to avoid infinite loops. - * - * Note that the substitution depends on value of $mOutputType: - * self::OT_WIKI: only {{subst:}} templates - * self::OT_MSG: only magic variables - * self::OT_HTML: all templates and magic variables - * - * @param string $tex The text to transform - * @param array $args Key-value pairs representing template parameters to substitute - * @param bool $argsOnly Only do argument (triple-brace) expansion, not double-brace expansion - * @private - */ - function replaceVariables( $text, $args = array(), $argsOnly = false ) { - # Prevent too big inclusions - if( strlen( $text ) > $this->mOptions->getMaxIncludeSize() ) { - return $text; - } - - $fname = __METHOD__ /*. '-L' . count( $this->mArgStack )*/; - wfProfileIn( $fname ); - - # This function is called recursively. To keep track of arguments we need a stack: - array_push( $this->mArgStack, $args ); - - $braceCallbacks = array(); - if ( !$argsOnly ) { - $braceCallbacks[2] = array( &$this, 'braceSubstitution' ); - } - if ( $this->mOutputType != self::OT_MSG ) { - $braceCallbacks[3] = array( &$this, 'argSubstitution' ); - } - if ( $braceCallbacks ) { - $callbacks = array( - '{' => array( - 'end' => '}', - 'cb' => $braceCallbacks, - 'min' => $argsOnly ? 3 : 2, - 'max' => isset( $braceCallbacks[3] ) ? 3 : 2, - ), - '[' => array( - 'end' => ']', - 'cb' => array(2=>null), - 'min' => 2, - 'max' => 2, - ) - ); - $text = $this->replace_callback ($text, $callbacks); - - array_pop( $this->mArgStack ); - } - wfProfileOut( $fname ); - return $text; - } - - /** - * Replace magic variables - * @private - */ - function variableSubstitution( $matches ) { - global $wgContLang; - $fname = 'Parser::variableSubstitution'; - $varname = $wgContLang->lc($matches[1]); - wfProfileIn( $fname ); - $skip = false; - if ( $this->mOutputType == self::OT_WIKI ) { - # Do only magic variables prefixed by SUBST - $mwSubst =& MagicWord::get( 'subst' ); - if (!$mwSubst->matchStartAndRemove( $varname )) - $skip = true; - # Note that if we don't substitute the variable below, - # we don't remove the {{subst:}} magic word, in case - # it is a template rather than a magic variable. - } - if ( !$skip && array_key_exists( $varname, $this->mVariables ) ) { - $id = $this->mVariables[$varname]; - # Now check if we did really match, case sensitive or not - $mw =& MagicWord::get( $id ); - if ($mw->match($matches[1])) { - $text = $this->getVariableValue( $id ); - if (MagicWord::getCacheTTL($id)>-1) - $this->mOutput->mContainsOldMagic = true; - } else { - $text = $matches[0]; - } - } else { - $text = $matches[0]; - } - wfProfileOut( $fname ); - return $text; - } - - - /// Clean up argument array - refactored in 1.9 so parserfunctions can use it, too. - static function createAssocArgs( $args ) { - $assocArgs = array(); - $index = 1; - foreach( $args as $arg ) { - $eqpos = strpos( $arg, '=' ); - if ( $eqpos === false ) { - $assocArgs[$index++] = $arg; - } else { - $name = trim( substr( $arg, 0, $eqpos ) ); - $value = trim( substr( $arg, $eqpos+1 ) ); - if ( $value === false ) { - $value = ''; - } - if ( $name !== false ) { - $assocArgs[$name] = $value; - } - } - } - - return $assocArgs; - } - - /** - * Return the text of a template, after recursively - * replacing any variables or templates within the template. - * - * @param array $piece The parts of the template - * $piece['text']: matched text - * $piece['title']: the title, i.e. the part before the | - * $piece['parts']: the parameter array - * @return string the text of the template - * @private - */ - function braceSubstitution( $piece ) { - global $wgContLang, $wgLang, $wgAllowDisplayTitle, $wgNonincludableNamespaces; - $fname = __METHOD__ /*. '-L' . count( $this->mArgStack )*/; - wfProfileIn( $fname ); - wfProfileIn( __METHOD__.'-setup' ); - - # Flags - $found = false; # $text has been filled - $nowiki = false; # wiki markup in $text should be escaped - $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 - - # Title object, where $text came from - $title = NULL; - - $linestart = ''; - - - # $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 - - $titleText = $part1 = $piece['title']; - # If the third subpattern matched anything, it will start with | - - if (null == $piece['parts']) { - $replaceWith = $this->variableSubstitution (array ($piece['text'], $piece['title'])); - if ($replaceWith != $piece['text']) { - $text = $replaceWith; - $found = true; - $noparse = true; - $noargs = true; - } - } - - $args = (null == $piece['parts']) ? array() : $piece['parts']; - wfProfileOut( __METHOD__.'-setup' ); - - # SUBST - wfProfileIn( __METHOD__.'-modifiers' ); - if ( !$found ) { - $mwSubst =& MagicWord::get( 'subst' ); - if ( $mwSubst->matchStartAndRemove( $part1 ) xor $this->ot['wiki'] ) { - # One of two possibilities is true: - # 1) Found SUBST but not in the PST phase - # 2) Didn't find SUBST and in the PST phase - # In either case, return without further processing - $text = $piece['text']; - $found = true; - $noparse = true; - $noargs = true; - } - } - - # MSG, MSGNW and RAW - if ( !$found ) { - # Check for MSGNW: - $mwMsgnw =& MagicWord::get( 'msgnw' ); - if ( $mwMsgnw->matchStartAndRemove( $part1 ) ) { - $nowiki = true; - } else { - # Remove obsolete MSG: - $mwMsg =& MagicWord::get( 'msg' ); - $mwMsg->matchStartAndRemove( $part1 ); - } - - # Check for RAW: - $mwRaw =& MagicWord::get( 'raw' ); - if ( $mwRaw->matchStartAndRemove( $part1 ) ) { - $forceRawInterwiki = true; - } - } - wfProfileOut( __METHOD__.'-modifiers' ); - - //save path level before recursing into functions & templates. - $lastPathLevel = $this->mTemplatePath; - - # Parser functions - if ( !$found ) { - wfProfileIn( __METHOD__ . '-pfunc' ); - - $colonPos = strpos( $part1, ':' ); - if ( $colonPos !== false ) { - # Case sensitive functions - $function = substr( $part1, 0, $colonPos ); - if ( isset( $this->mFunctionSynonyms[1][$function] ) ) { - $function = $this->mFunctionSynonyms[1][$function]; - } else { - # Case insensitive functions - $function = strtolower( $function ); - if ( isset( $this->mFunctionSynonyms[0][$function] ) ) { - $function = $this->mFunctionSynonyms[0][$function]; - } else { - $function = false; - } - } - if ( $function ) { - $funcArgs = array_map( 'trim', $args ); - $funcArgs = array_merge( array( &$this, trim( substr( $part1, $colonPos + 1 ) ) ), $funcArgs ); - $result = call_user_func_array( $this->mFunctionHooks[$function], $funcArgs ); - $found = true; - - // The text is usually already parsed, doesn't need triple-brace tags expanded, etc. - //$noargs = true; - //$noparse = true; - - if ( is_array( $result ) ) { - if ( isset( $result[0] ) ) { - $text = $linestart . $result[0]; - unset( $result[0] ); - } - - // Extract flags into the local scope - // This allows callers to set flags such as nowiki, noparse, found, etc. - extract( $result ); - } else { - $text = $linestart . $result; - } - } - } - wfProfileOut( __METHOD__ . '-pfunc' ); - } - - # Template table test - - # Did we encounter this template already? If yes, it is in the cache - # and we need to check for loops. - if ( !$found && isset( $this->mTemplates[$piece['title']] ) ) { - $found = true; - - # Infinite loop test - if ( isset( $this->mTemplatePath[$part1] ) ) { - $noparse = true; - $noargs = true; - $found = true; - $text = $linestart . - "[[$part1]]"; - wfDebug( __METHOD__.": template loop broken at '$part1'\n" ); - } else { - # set $text to cached message. - $text = $linestart . $this->mTemplates[$piece['title']]; - #treat title for cached page the same as others - $ns = NS_TEMPLATE; - $subpage = ''; - $part1 = $this->maybeDoSubpageLink( $part1, $subpage ); - if ($subpage !== '') { - $ns = $this->mTitle->getNamespace(); - } - $title = Title::newFromText( $part1, $ns ); - //used by include size checking - $titleText = $title->getPrefixedText(); - //used by edit section links - $replaceHeadings = true; - - } - } - - # Load from database - if ( !$found ) { - wfProfileIn( __METHOD__ . '-loadtpl' ); - $ns = NS_TEMPLATE; - # declaring $subpage directly in the function call - # does not work correctly with references and breaks - # {{/subpage}}-style inclusions - $subpage = ''; - $part1 = $this->maybeDoSubpageLink( $part1, $subpage ); - if ($subpage !== '') { - $ns = $this->mTitle->getNamespace(); - } - $title = Title::newFromText( $part1, $ns ); - - - if ( !is_null( $title ) ) { - $titleText = $title->getPrefixedText(); - # Check for language variants if the template is not found - if($wgContLang->hasVariants() && $title->getArticleID() == 0){ - $wgContLang->findVariantLink($part1, $title); - } - - if ( !$title->isExternal() ) { - if ( $title->getNamespace() == NS_SPECIAL && $this->mOptions->getAllowSpecialInclusion() && $this->ot['html'] ) { - $text = SpecialPage::capturePath( $title ); - if ( is_string( $text ) ) { - $found = true; - $noparse = true; - $noargs = true; - $isHTML = true; - $this->disableCache(); - } - } else if ( $wgNonincludableNamespaces && in_array( $title->getNamespace(), $wgNonincludableNamespaces ) ) { - $found = false; //access denied - wfDebug( "$fname: template inclusion denied for " . $title->getPrefixedDBkey() ); - } else { - list($articleContent,$title) = $this->fetchTemplateAndtitle( $title ); - if ( $articleContent !== false ) { - $found = true; - $text = $articleContent; - $replaceHeadings = true; - } - } - - # If the title is valid but undisplayable, make a link to it - if ( !$found && ( $this->ot['html'] || $this->ot['pre'] ) ) { - $text = "[[:$titleText]]"; - $found = true; - } - } elseif ( $title->isTrans() ) { - // Interwiki transclusion - if ( $this->ot['html'] && !$forceRawInterwiki ) { - $text = $this->interwikiTransclude( $title, 'render' ); - $isHTML = true; - $noparse = true; - } else { - $text = $this->interwikiTransclude( $title, 'raw' ); - $replaceHeadings = true; - } - $found = true; - } - - # Template cache array insertion - # Use the original $piece['title'] not the mangled $part1, so that - # modifiers such as RAW: produce separate cache entries - if( $found ) { - if( $isHTML ) { - // A special page; don't store it in the template cache. - } else { - $this->mTemplates[$piece['title']] = $text; - } - $text = $linestart . $text; - } - } - wfProfileOut( __METHOD__ . '-loadtpl' ); - } - - if ( $found && !$this->incrementIncludeSize( 'pre-expand', strlen( $text ) ) ) { - # Error, oversize inclusion - $text = $linestart . - "[[$titleText]]"; - $noparse = true; - $noargs = true; - } - - # Recursive parsing, escaping and link table handling - # Only for HTML output - if ( $nowiki && $found && ( $this->ot['html'] || $this->ot['pre'] ) ) { - $text = wfEscapeWikiText( $text ); - } elseif ( !$this->ot['msg'] && $found ) { - if ( $noargs ) { - $assocArgs = array(); - } else { - # Clean up argument array - $assocArgs = self::createAssocArgs($args); - # Add a new element to the templace recursion path - $this->mTemplatePath[$part1] = 1; - } - - if ( !$noparse ) { - # If there are any tags, only include them - if ( in_string( '', $text ) && in_string( '', $text ) ) { - $replacer = new OnlyIncludeReplacer; - StringUtils::delimiterReplaceCallback( '', '', - array( &$replacer, 'replace' ), $text ); - $text = $replacer->output; - } - # Remove sections and tags - $text = StringUtils::delimiterReplace( '', '', '', $text ); - $text = strtr( $text, array( '' => '' , '' => '' ) ); - - if( $this->ot['html'] || $this->ot['pre'] ) { - # Strip ,

    , etc.
    -					$text = $this->strip( $text, $this->mStripState );
    -					if ( $this->ot['html'] ) {
    -						$text = Sanitizer::removeHTMLtags( $text, array( &$this, 'replaceVariables' ), $assocArgs );
    -					} elseif ( $this->ot['pre'] && $this->mOptions->getRemoveComments() ) {
    -						$text = Sanitizer::removeHTMLcomments( $text );
    -					}
    -				}
    -				$text = $this->replaceVariables( $text, $assocArgs );
    -
    -				# If the template begins with a table or block-level
    -				# element, it should be treated as beginning a new line.
    -				if (!$piece['lineStart'] && preg_match('/^(?:{\\||:|;|#|\*)/', $text)) /*}*/{
    -					$text = "\n" . $text;
    -				}
    -			} elseif ( !$noargs ) {
    -				# $noparse and !$noargs
    -				# Just replace the arguments, not any double-brace items
    -				# This is used for rendered interwiki transclusion
    -				$text = $this->replaceVariables( $text, $assocArgs, true );
    -			}
    -		}
    -		# Prune lower levels off the recursion check path
    -		$this->mTemplatePath = $lastPathLevel;
    -
    -		if ( $found && !$this->incrementIncludeSize( 'post-expand', strlen( $text ) ) ) {
    -			# Error, oversize inclusion
    -			$text = $linestart .
    -				"[[$titleText]]";
    -			$noparse = true;
    -			$noargs = true;
    -		}
    -
    -		if ( !$found ) {
    -			wfProfileOut( $fname );
    -			return $piece['text'];
    -		} else {
    -			wfProfileIn( __METHOD__ . '-placeholders' );
    -			if ( $isHTML ) {
    -				# Replace raw HTML by a placeholder
    -				# Add a blank line preceding, to prevent it from mucking up
    -				# immediately preceding headings
    -				$text = "\n\n" . $this->insertStripItem( $text, $this->mStripState );
    -			} else {
    -				# replace ==section headers==
    -				# XXX this needs to go away once we have a better parser.
    -				if ( !$this->ot['wiki'] && !$this->ot['pre'] && $replaceHeadings ) {
    -					if( !is_null( $title ) )
    -						$encodedname = base64_encode($title->getPrefixedDBkey());
    -					else
    -						$encodedname = base64_encode("");
    -					$m = preg_split('/(^={1,6}.*?={1,6}\s*?$)/m', $text, -1,
    -						PREG_SPLIT_DELIM_CAPTURE);
    -					$text = '';
    -					$nsec = $headingOffset;
    -
    -					for( $i = 0; $i < count($m); $i += 2 ) {
    -						$text .= $m[$i];
    -						if (!isset($m[$i + 1]) || $m[$i + 1] == "") continue;
    -						$hl = $m[$i + 1];
    -						if( strstr($hl, "" . $m2[3];
    -
    -						$nsec++;
    -					}
    -				}
    -			}
    -			wfProfileOut( __METHOD__ . '-placeholders' );
    -		}
    -
    -		# Prune lower levels off the recursion check path
    -		$this->mTemplatePath = $lastPathLevel;
    -
    -		if ( !$found ) {
    -			wfProfileOut( $fname );
    -			return $piece['text'];
    -		} else {
    -			wfProfileOut( $fname );
    -			return $text;
    -		}
    -	}
    -
    -	/**
    -	 * Fetch the unparsed text of a template and register a reference to it.
    -	 */
    -	function fetchTemplateAndTitle( $title ) {
    -		$templateCb = $this->mOptions->getTemplateCallback();
    -		$stuff = call_user_func( $templateCb, $title );
    -		$text = $stuff['text'];
    -		$finalTitle = isset( $stuff['finalTitle'] ) ? $stuff['finalTitle'] : $title;
    -		if ( isset( $stuff['deps'] ) ) {
    -			foreach ( $stuff['deps'] as $dep ) {
    -				$this->mOutput->addTemplate( $dep['title'], $dep['page_id'], $dep['rev_id'] );
    -			}
    -		}
    -		return array($text,$finalTitle);
    -	}
    -
    -	function fetchTemplate( $title ) {
    -		$rv = $this->fetchTemplateAndtitle($title);
    -		return $rv[0];
    -	}
    -
    -	/**
    -	 * Static function to get a template
    -	 * Can be overridden via ParserOptions::setTemplateCallback().
    -	 *
    -	 * Returns an associative array:
    -	 *    text          The unparsed template text
    -	 *    finalTitle    (Optional) The title after following redirects
    -	 *    deps          (Optional) An array of associative array dependencies:
    -	 *                       title:    The dependency title, to be registered in templatelinks
    -	 *                       page_id:  The page_id of the title
    -	 *                       rev_id:   The revision ID loaded
    -	 */
    -	static function statelessFetchTemplate( $title ) {
    -		$text = $skip = false;
    -		$finalTitle = $title;
    -		$deps = array();
    -		
    -		// Loop to fetch the article, with up to 1 redirect
    -		for ( $i = 0; $i < 2 && is_object( $title ); $i++ ) {
    -			# Give extensions a chance to select the revision instead
    -			$id = false; // Assume current
    -			wfRunHooks( 'BeforeParserFetchTemplateAndtitle', array( false, &$title, &$skip, &$id ) );
    -			
    -			if( $skip ) {
    -				$text = false;
    -				$deps[] = array(
    -					'title' => $title,
    -					'page_id' => $title->getArticleID(),
    -					'rev_id' => null );
    -				break;
    -			}
    -			$rev = $id ? Revision::newFromId( $id ) : Revision::newFromTitle( $title );
    -			$rev_id = $rev ? $rev->getId() : 0;
    -
    -			$deps[] = array( 
    -				'title' => $title, 
    -				'page_id' => $title->getArticleID(), 
    -				'rev_id' => $rev_id );
    -
    -			if( $rev ) {
    -				$text = $rev->getText();
    -			} elseif( $title->getNamespace() == NS_MEDIAWIKI ) {
    -				global $wgLang;
    -				$message = $wgLang->lcfirst( $title->getText() );
    -				$text = wfMsgForContentNoTrans( $message );
    -				if( wfEmptyMsg( $message, $text ) ) {
    -					$text = false;
    -					break;
    -				}
    -			} else {
    -				break;
    -			}
    -			if ( $text === false ) {
    -				break;
    -			}
    -			// Redirect?
    -			$finalTitle = $title;
    -			$title = Title::newFromRedirect( $text );
    -		}
    -		return array(
    -			'text' => $text,
    -			'finalTitle' => $finalTitle,
    -			'deps' => $deps );
    -	}
    -
    -	/**
    -	 * Transclude an interwiki link.
    -	 */
    -	function interwikiTransclude( $title, $action ) {
    -		global $wgEnableScaryTranscluding;
    -
    -		if (!$wgEnableScaryTranscluding)
    -			return wfMsg('scarytranscludedisabled');
    -
    -		$url = $title->getFullUrl( "action=$action" );
    -
    -		if (strlen($url) > 255)
    -			return wfMsg('scarytranscludetoolong');
    -		return $this->fetchScaryTemplateMaybeFromCache($url);
    -	}
    -
    -	function fetchScaryTemplateMaybeFromCache($url) {
    -		global $wgTranscludeCacheExpiry;
    -		$dbr = wfGetDB(DB_SLAVE);
    -		$obj = $dbr->selectRow('transcache', array('tc_time', 'tc_contents'),
    -				array('tc_url' => $url));
    -		if ($obj) {
    -			$time = $obj->tc_time;
    -			$text = $obj->tc_contents;
    -			if ($time && time() < $time + $wgTranscludeCacheExpiry ) {
    -				return $text;
    -			}
    -		}
    -
    -		$text = Http::get($url);
    -		if (!$text)
    -			return wfMsg('scarytranscludefailed', $url);
    -
    -		$dbw = wfGetDB(DB_MASTER);
    -		$dbw->replace('transcache', array('tc_url'), array(
    -			'tc_url' => $url,
    -			'tc_time' => time(),
    -			'tc_contents' => $text));
    -		return $text;
    -	}
    -
    -
    -	/**
    -	 * Triple brace replacement -- used for template arguments
    -	 * @private
    -	 */
    -	function argSubstitution( $matches ) {
    -		$arg = trim( $matches['title'] );
    -		$text = $matches['text'];
    -		$inputArgs = end( $this->mArgStack );
    -
    -		if ( array_key_exists( $arg, $inputArgs ) ) {
    -			$text = $inputArgs[$arg];
    -		} else if (($this->mOutputType == self::OT_HTML || $this->mOutputType == self::OT_PREPROCESS ) &&
    -		null != $matches['parts'] && count($matches['parts']) > 0) {
    -			$text = $matches['parts'][0];
    -		}
    -		if ( !$this->incrementIncludeSize( 'arg', strlen( $text ) ) ) {
    -			$text = $matches['text'] .
    -				'';
    -		}
    -
    -		return $text;
    -	}
    -
    -	/**
    -	 * Increment an include size counter
    -	 *
    -	 * @param string $type The type of expansion
    -	 * @param integer $size The size of the text
    -	 * @return boolean False if this inclusion would take it over the maximum, true otherwise
    -	 */
    -	function incrementIncludeSize( $type, $size ) {
    -		if ( $this->mIncludeSizes[$type] + $size > $this->mOptions->getMaxIncludeSize() ) {
    -			return false;
    -		} else {
    -			$this->mIncludeSizes[$type] += $size;
    -			return true;
    -		}
    -	}
    -
    -	/**
    -	 * Detect __NOGALLERY__ magic word and set a placeholder
    -	 */
    -	function stripNoGallery( &$text ) {
    -		# if the string __NOGALLERY__ (not case-sensitive) occurs in the HTML,
    -		# do not add TOC
    -		$mw = MagicWord::get( 'nogallery' );
    -		$this->mOutput->mNoGallery = $mw->matchAndRemove( $text ) ;
    -	}
    -
    -	/**
    -	 * Find the first __TOC__ magic word and set a 
    -	 * placeholder that will then be replaced by the real TOC in
    -	 * ->formatHeadings, this works because at this points real
    -	 * comments will have already been discarded by the sanitizer.
    -	 *
    -	 * Any additional __TOC__ magic words left over will be discarded
    -	 * as there can only be one TOC on the page.
    -	 */
    -	function stripToc( $text ) {
    -		# if the string __NOTOC__ (not case-sensitive) occurs in the HTML,
    -		# do not add TOC
    -		$mw = MagicWord::get( 'notoc' );
    -		if( $mw->matchAndRemove( $text ) ) {
    -			$this->mShowToc = false;
    -		}
    -
    -		$mw = MagicWord::get( 'toc' );
    -		if( $mw->match( $text ) ) {
    -			$this->mShowToc = true;
    -			$this->mForceTocPosition = true;
    -
    -			// Set a placeholder. At the end we'll fill it in with the TOC.
    -			$text = $mw->replace( '', $text, 1 );
    -
    -			// Only keep the first one.
    -			$text = $mw->replace( '', $text );
    -		}
    -		return $text;
    -	}
    -
    -	/**
    -	 * This function accomplishes several tasks:
    -	 * 1) Auto-number headings if that option is enabled
    -	 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
    -	 * 3) Add a Table of contents on the top for users who have enabled the option
    -	 * 4) Auto-anchor headings
    -	 *
    -	 * It loops through all headlines, collects the necessary data, then splits up the
    -	 * string and re-inserts the newly formatted headlines.
    -	 *
    -	 * @param string $text
    -	 * @param boolean $isMain
    -	 * @private
    -	 */
    -	function formatHeadings( $text, $isMain=true ) {
    -		global $wgMaxTocLevel, $wgContLang;
    -
    -		$doNumberHeadings = $this->mOptions->getNumberHeadings();
    -		if( !$this->mTitle->quickUserCan( 'edit' ) ) {
    -			$showEditLink = 0;
    -		} else {
    -			$showEditLink = $this->mOptions->getEditSection();
    -		}
    -
    -		# Inhibit editsection links if requested in the page
    -		$esw =& MagicWord::get( 'noeditsection' );
    -		if( $esw->matchAndRemove( $text ) ) {
    -			$showEditLink = 0;
    -		}
    -
    -		# Get all headlines for numbering them and adding funky stuff like [edit]
    -		# links - this is for later, but we need the number of headlines right now
    -		$matches = array();
    -		$numMatches = preg_match_all( '/[1-6])(?P.*?'.'>)(?P
    .*?)<\/H[1-6] *>/i', $text, $matches ); - - # if there are fewer than 4 headlines in the article, do not show TOC - # unless it's been explicitly enabled. - $enoughToc = $this->mShowToc && - (($numMatches >= 4) || $this->mForceTocPosition); - - # Allow user to stipulate that a page should have a "new section" - # link added via __NEWSECTIONLINK__ - $mw =& MagicWord::get( 'newsectionlink' ); - if( $mw->matchAndRemove( $text ) ) - $this->mOutput->setNewSection( true ); - - # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML, - # override above conditions and always show TOC above first header - $mw =& MagicWord::get( 'forcetoc' ); - if ($mw->matchAndRemove( $text ) ) { - $this->mShowToc = true; - $enoughToc = true; - } - - # We need this to perform operations on the HTML - $sk = $this->mOptions->getSkin(); - - # headline counter - $headlineCount = 0; - $sectionCount = 0; # headlineCount excluding template sections - $numVisible = 0; - - # Ugh .. the TOC should have neat indentation levels which can be - # passed to the skin functions. These are determined here - $toc = ''; - $full = ''; - $head = array(); - $sublevelCount = array(); - $levelCount = array(); - $toclevel = 0; - $level = 0; - $prevlevel = 0; - $toclevel = 0; - $prevtoclevel = 0; - $tocraw = array(); - - foreach( $matches[3] as $headline ) { - $istemplate = 0; - $templatetitle = ''; - $templatesection = 0; - $numbering = ''; - $mat = array(); - if (preg_match("//", $headline, $mat)) { - $istemplate = 1; - $templatetitle = base64_decode($mat[1]); - $templatesection = 1 + (int)base64_decode($mat[2]); - $headline = preg_replace("//", "", $headline); - } - - if( $toclevel ) { - $prevlevel = $level; - $prevtoclevel = $toclevel; - } - $level = $matches[1][$headlineCount]; - - if( $doNumberHeadings || $enoughToc ) { - - if ( $level > $prevlevel ) { - # Increase TOC level - $toclevel++; - $sublevelCount[$toclevel] = 0; - if( $toclevel<$wgMaxTocLevel ) { - $prevtoclevel = $toclevel; - $toc .= $sk->tocIndent(); - $numVisible++; - } - } - elseif ( $level < $prevlevel && $toclevel > 1 ) { - # Decrease TOC level, find level to jump to - - if ( $toclevel == 2 && $level <= $levelCount[1] ) { - # Can only go down to level 1 - $toclevel = 1; - } else { - for ($i = $toclevel; $i > 0; $i--) { - if ( $levelCount[$i] == $level ) { - # Found last matching level - $toclevel = $i; - break; - } - elseif ( $levelCount[$i] < $level ) { - # Found first matching level below current level - $toclevel = $i + 1; - break; - } - } - } - if( $toclevel<$wgMaxTocLevel ) { - if($prevtoclevel < $wgMaxTocLevel) { - # Unindent only if the previous toc level was shown :p - $toc .= $sk->tocUnindent( $prevtoclevel - $toclevel ); - } else { - $toc .= $sk->tocLineEnd(); - } - } - } - else { - # No change in level, end TOC line - if( $toclevel<$wgMaxTocLevel ) { - $toc .= $sk->tocLineEnd(); - } - } - - $levelCount[$toclevel] = $level; - - # count number of headlines for each level - @$sublevelCount[$toclevel]++; - $dot = 0; - for( $i = 1; $i <= $toclevel; $i++ ) { - if( !empty( $sublevelCount[$i] ) ) { - if( $dot ) { - $numbering .= '.'; - } - $numbering .= $wgContLang->formatNum( $sublevelCount[$i] ); - $dot = 1; - } - } - } - - # The canonized header is a version of the header text safe to use for links - # Avoid insertion of weird stuff like by expanding the relevant sections - $canonized_headline = $this->mStripState->unstripBoth( $headline ); - - # Remove link placeholders by the link text. - # - # turns into - # link text with suffix - $canonized_headline = preg_replace( '//e', - "\$this->mLinkHolders['texts'][\$1]", - $canonized_headline ); - $canonized_headline = preg_replace( '//e', - "\$this->mInterwikiLinkHolders['texts'][\$1]", - $canonized_headline ); - - # Strip out HTML (other than plain and : bug 8393) - $tocline = preg_replace( - array( '#<(?!/?(sup|sub)).*?'.'>#', '#<(/?(sup|sub)).*?'.'>#' ), - array( '', '<$1>'), - $canonized_headline - ); - $tocline = trim( $tocline ); - - # For the anchor, strip out HTML-y stuff period - $canonized_headline = preg_replace( '/<.*?'.'>/', '', $canonized_headline ); - $canonized_headline = trim( $canonized_headline ); - - # Save headline for section edit hint before it's escaped - $headline_hint = $canonized_headline; - $canonized_headline = Sanitizer::escapeId( $canonized_headline ); - $refers[$headlineCount] = $canonized_headline; - - # count how many in assoc. array so we can track dupes in anchors - 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) - if( $doNumberHeadings && count( $matches[3] ) > 1) { - # the two are different if the line contains a link - $headline=$numbering . ' ' . $headline; - } - - # Create the anchor for linking from the TOC to the section - $anchor = $canonized_headline; - if($refcount[$headlineCount] > 1 ) { - $anchor .= '_' . $refcount[$headlineCount]; - } - if( $enoughToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) { - $toc .= $sk->tocLine($anchor, $tocline, $numbering, $toclevel); - $tocraw[] = array( 'toclevel' => $toclevel, 'level' => $level, 'line' => $tocline, 'number' => $numbering ); - } - # give headline the correct tag - if( $showEditLink && ( !$istemplate || $templatetitle !== "" ) ) { - if( $istemplate ) - $editlink = $sk->editSectionLinkForOther($templatetitle, $templatesection); - else - $editlink = $sk->editSectionLink($this->mTitle, $sectionCount+1, $headline_hint); - } else { - $editlink = ''; - } - $head[$headlineCount] = $sk->makeHeadline( $level, $matches['attrib'][$headlineCount], $anchor, $headline, $editlink ); - - $headlineCount++; - if( !$istemplate ) - $sectionCount++; - } - - $this->mOutput->setSections( $tocraw ); - - # Never ever show TOC if no headers - if( $numVisible < 1 ) { - $enoughToc = false; - } - - if( $enoughToc ) { - if( $prevtoclevel > 0 && $prevtoclevel < $wgMaxTocLevel ) { - $toc .= $sk->tocUnindent( $prevtoclevel - 1 ); - } - $toc = $sk->tocList( $toc ); - } - - # split up and insert constructed headlines - - $blocks = preg_split( '/.*?<\/H[1-6]>/i', $text ); - $i = 0; - - foreach( $blocks as $block ) { - if( $showEditLink && $headlineCount > 0 && $i == 0 && $block != "\n" ) { - # This is the [edit] link that appears for the top block of text when - # section editing is enabled - - # Disabled because it broke block formatting - # For example, a bullet point in the top line - # $full .= $sk->editSectionLink(0); - } - $full .= $block; - if( $enoughToc && !$i && $isMain && !$this->mForceTocPosition ) { - # Top anchor now in skin - $full = $full.$toc; - } - - if( !empty( $head[$i] ) ) { - $full .= $head[$i]; - } - $i++; - } - if( $this->mForceTocPosition ) { - return str_replace( '', $toc, $full ); - } else { - return $full; - } - } - - /** - * Transform wiki markup when saving a page by doing \r\n -> \n - * conversion, substitting signatures, {{subst:}} templates, etc. - * - * @param string $text the text to transform - * @param Title &$title the Title object for the current article - * @param User &$user the User object describing the current user - * @param ParserOptions $options parsing options - * @param bool $clearState whether to clear the parser state first - * @return string the altered wiki markup - * @public - */ - function preSaveTransform( $text, &$title, $user, $options, $clearState = true ) { - $this->mOptions = $options; - $this->mTitle =& $title; - $this->setOutputType( self::OT_WIKI ); - - if ( $clearState ) { - $this->clearState(); - } - - $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 = $stripState->unstripBoth( $text ); - return $text; - } - - /** - * Pre-save transform helper function - * @private - */ - function pstPass2( $text, &$stripState, $user ) { - global $wgContLang, $wgLocaltimezone; - - /* Note: This is the timestamp saved as hardcoded wikitext to - * the database, we use $wgContLang here in order to give - * everyone the same signature and use the default one rather - * than the one selected in each user's preferences. - */ - if ( isset( $wgLocaltimezone ) ) { - $oldtz = getenv( 'TZ' ); - putenv( 'TZ='.$wgLocaltimezone ); - } - $d = $wgContLang->timeanddate( date( 'YmdHis' ), false, false) . - ' (' . date( 'T' ) . ')'; - if ( isset( $wgLocaltimezone ) ) { - putenv( 'TZ='.$oldtz ); - } - - # Variable replacement - # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags - $text = $this->replaceVariables( $text ); - - # Strip out etc. added via replaceVariables - $text = $this->strip( $text, $stripState, false, array( 'gallery' ) ); - - # Signatures - $sigText = $this->getUserSig( $user ); - $text = strtr( $text, array( - '~~~~~' => $d, - '~~~~' => "$sigText $d", - '~~~' => $sigText - ) ); - - # Context links: [[|name]] and [[name (context)|]] - # - global $wgLegalTitleChars; - $tc = "[$wgLegalTitleChars]"; - $nc = '[ _0-9A-Za-z\x80-\xff]'; # Namespaces can use non-ascii! - - $p1 = "/\[\[(:?$nc+:|:|)($tc+?)( \\($tc+\\))\\|]]/"; # [[ns:page (context)|]] - $p3 = "/\[\[(:?$nc+:|:|)($tc+?)( \\($tc+\\)|)(, $tc+|)\\|]]/"; # [[ns:page (context), context|]] - $p2 = "/\[\[\\|($tc+)]]/"; # [[|page]] - - # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]" - $text = preg_replace( $p1, '[[\\1\\2\\3|\\2]]', $text ); - $text = preg_replace( $p3, '[[\\1\\2\\3\\4|\\2]]', $text ); - - $t = $this->mTitle->getText(); - $m = array(); - if ( preg_match( "/^($nc+:|)$tc+?( \\($tc+\\))$/", $t, $m ) ) { - $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text ); - } elseif ( preg_match( "/^($nc+:|)$tc+?(, $tc+|)$/", $t, $m ) && '' != "$m[1]$m[2]" ) { - $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text ); - } else { - # if there's no context, don't bother duplicating the title - $text = preg_replace( $p2, '[[\\1]]', $text ); - } - - # Trim trailing whitespace - $text = rtrim( $text ); - - return $text; - } - - /** - * Fetch the user's signature text, if any, and normalize to - * validated, ready-to-insert wikitext. - * - * @param User $user - * @return string - * @private - */ - function getUserSig( &$user ) { - global $wgMaxSigChars; - - $username = $user->getName(); - $nickname = $user->getOption( 'nickname' ); - $nickname = $nickname === '' ? $username : $nickname; - - if( mb_strlen( $nickname ) > $wgMaxSigChars ) { - $nickname = $username; - wfDebug( __METHOD__ . ": $username has overlong signature.\n" ); - } elseif( $user->getBoolOption( 'fancysig' ) !== false ) { - # Sig. might contain markup; validate this - if( $this->validateSig( $nickname ) !== false ) { - # Validated; clean up (if needed) and return it - return $this->cleanSig( $nickname, true ); - } else { - # Failed to validate; fall back to the default - $nickname = $username; - wfDebug( "Parser::getUserSig: $username has bad XML tags in signature.\n" ); - } - } - - // Make sure nickname doesnt get a sig in a sig - $nickname = $this->cleanSigInSig( $nickname ); - - # If we're still here, make it a link to the user page - $userText = wfEscapeWikiText( $username ); - $nickText = wfEscapeWikiText( $nickname ); - if ( $user->isAnon() ) { - return wfMsgExt( 'signature-anon', array( 'content', 'parsemag' ), $userText, $nickText ); - } else { - return wfMsgExt( 'signature', array( 'content', 'parsemag' ), $userText, $nickText ); - } - } - - /** - * Check that the user's signature contains no bad XML - * - * @param string $text - * @return mixed An expanded string, or false if invalid. - */ - function validateSig( $text ) { - return( wfIsWellFormedXmlFragment( $text ) ? $text : false ); - } - - /** - * Clean up signature text - * - * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig - * 2) Substitute all transclusions - * - * @param string $text - * @param $parsing Whether we're cleaning (preferences save) or parsing - * @return string Signature text - */ - function cleanSig( $text, $parsing = false ) { - global $wgTitle; - $this->startExternalParse( $this->mTitle, new ParserOptions(), $parsing ? self::OT_WIKI : self::OT_MSG ); - - $substWord = MagicWord::get( 'subst' ); - $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase(); - $substText = '{{' . $substWord->getSynonym( 0 ); - - $text = preg_replace( $substRegex, $substText, $text ); - $text = $this->cleanSigInSig( $text ); - $text = $this->replaceVariables( $text ); - - $this->clearState(); - return $text; - } - - /** - * Strip ~~~, ~~~~ and ~~~~~ out of signatures - * @param string $text - * @return string Signature text with /~{3,5}/ removed - */ - function cleanSigInSig( $text ) { - $text = preg_replace( '/~{3,5}/', '', $text ); - return $text; - } - - /** - * Set up some variables which are usually set up in parse() - * so that an external function can call some class members with confidence - * @public - */ - function startExternalParse( &$title, $options, $outputType, $clearState = true ) { - $this->mTitle =& $title; - $this->mOptions = $options; - $this->setOutputType( $outputType ); - if ( $clearState ) { - $this->clearState(); - } - } - - /** - * Transform a MediaWiki message by replacing magic variables. - * - * @param string $text the text to transform - * @param ParserOptions $options options - * @return string the text with variables substituted - * @public - */ - function transformMsg( $text, $options ) { - global $wgTitle; - static $executing = false; - - $fname = "Parser::transformMsg"; - - # Guard against infinite recursion - if ( $executing ) { - return $text; - } - $executing = true; - - wfProfileIn($fname); - - if ( $wgTitle && !( $wgTitle instanceof FakeTitle ) ) { - $this->mTitle = $wgTitle; - } else { - $this->mTitle = Title::newFromText('msg'); - } - $this->mOptions = $options; - $this->setOutputType( self::OT_MSG ); - $this->clearState(); - $text = $this->replaceVariables( $text ); - - $executing = false; - wfProfileOut($fname); - return $text; - } - - /** - * Create an HTML-style tag, e.g. special text - * The callback should have the following form: - * function myParserHook( $text, $params, &$parser ) { ... } - * - * Transform and return $text. Use $parser for any required context, e.g. use - * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions - * - * @public - * - * @param mixed $tag The tag to use, e.g. 'hook' for - * @param mixed $callback The callback function (and object) to use for the tag - * - * @return The old value of the mTagHooks array associated with the hook - */ - function setHook( $tag, $callback ) { - $tag = strtolower( $tag ); - $oldVal = isset( $this->mTagHooks[$tag] ) ? $this->mTagHooks[$tag] : null; - $this->mTagHooks[$tag] = $callback; - - return $oldVal; - } - - function setTransparentTagHook( $tag, $callback ) { - $tag = strtolower( $tag ); - $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null; - $this->mTransparentTagHooks[$tag] = $callback; - - return $oldVal; - } - - /** - * Create a function, e.g. {{sum:1|2|3}} - * The callback function should have the form: - * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... } - * - * The callback may either return the text result of the function, or an array with the text - * in element 0, and a number of flags in the other elements. The names of the flags are - * specified in the keys. Valid flags are: - * found The text returned is valid, stop processing the template. This - * is on by default. - * nowiki Wiki markup in the return value should be escaped - * noparse Unsafe HTML tags should not be stripped, etc. - * noargs Don't replace triple-brace arguments in the return value - * isHTML The returned text is HTML, armour it against wikitext transformation - * - * @public - * - * @param string $id The magic word ID - * @param mixed $callback The callback function (and object) to use - * @param integer $flags a combination of the following flags: - * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}} - * - * @return The old callback function for this name, if any - */ - function setFunctionHook( $id, $callback, $flags = 0 ) { - $oldVal = isset( $this->mFunctionHooks[$id] ) ? $this->mFunctionHooks[$id] : null; - $this->mFunctionHooks[$id] = $callback; - - # Add to function cache - $mw = MagicWord::get( $id ); - if( !$mw ) - throw new MWException( 'Parser::setFunctionHook() expecting a magic word identifier.' ); - - $synonyms = $mw->getSynonyms(); - $sensitive = intval( $mw->isCaseSensitive() ); - - foreach ( $synonyms as $syn ) { - # Case - if ( !$sensitive ) { - $syn = strtolower( $syn ); - } - # Add leading hash - if ( !( $flags & SFH_NO_HASH ) ) { - $syn = '#' . $syn; - } - # Remove trailing colon - if ( substr( $syn, -1, 1 ) == ':' ) { - $syn = substr( $syn, 0, -1 ); - } - $this->mFunctionSynonyms[$sensitive][$syn] = $id; - } - return $oldVal; - } - - /** - * Get all registered function hook identifiers - * - * @return array - */ - function getFunctionHooks() { - return array_keys( $this->mFunctionHooks ); - } - - /** - * Replace 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; - global $wgContLang; - - $fname = 'Parser::replaceLinkHolders'; - wfProfileIn( $fname ); - - $pdbks = array(); - $colours = array(); - $sk = $this->mOptions->getSkin(); - $linkCache =& LinkCache::singleton(); - - if ( !empty( $this->mLinkHolders['namespaces'] ) ) { - wfProfileIn( $fname.'-check' ); - $dbr = wfGetDB( DB_SLAVE ); - $page = $dbr->tableName( 'page' ); - $threshold = $wgUser->getOption('stubthreshold'); - - # Sort by namespace - asort( $this->mLinkHolders['namespaces'] ); - - # Generate query - $query = false; - $current = null; - foreach ( $this->mLinkHolders['namespaces'] as $key => $ns ) { - # Make title object - $title = $this->mLinkHolders['titles'][$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 a static known link, e.g. interwiki - if ( $title->isAlwaysKnown() ) { - $colours[$pdbk] = 1; - } elseif ( ( $id = $linkCache->getGoodLinkID( $pdbk ) ) != 0 ) { - $colours[$pdbk] = 1; - $this->mOutput->addLink( $title, $id ); - } elseif ( $linkCache->isBadLink( $pdbk ) ) { - $colours[$pdbk] = 0; - } elseif ( $title->getNamespace() == NS_SPECIAL && !SpecialPage::exists( $pdbk ) ) { - $colours[$pdbk] = 0; - } else { - # Not in the link cache, add it to the query - if ( !isset( $current ) ) { - $current = $ns; - $query = "SELECT page_id, page_namespace, page_title"; - if ( $threshold > 0 ) { - $query .= ', page_len, page_is_redirect'; - } - $query .= " FROM $page WHERE (page_namespace=$ns AND page_title IN("; - } elseif ( $current != $ns ) { - $current = $ns; - $query .= ")) OR (page_namespace=$ns AND page_title IN("; - } else { - $query .= ', '; - } - - $query .= $dbr->addQuotes( $this->mLinkHolders['dbkeys'][$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->page_namespace, $s->page_title ); - $pdbk = $title->getPrefixedDBkey(); - $linkCache->addGoodLinkObj( $s->page_id, $title ); - $this->mOutput->addLink( $title, $s->page_id ); - - $colours[$pdbk] = ( $threshold == 0 || ( - $s->page_len >= $threshold || # always true if $threshold <= 0 - $s->page_is_redirect || - !Namespace::isContent( $s->page_namespace ) ) - ? 1 : 2 ); - } - } - wfProfileOut( $fname.'-check' ); - - # Do a second query for different language variants of links and categories - if($wgContLang->hasVariants()){ - $linkBatch = new LinkBatch(); - $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 ) { - $title = $this->mLinkHolders['titles'][$key]; - if ( is_null( $title ) ) - continue; - - $pdbk = $title->getPrefixedDBkey(); - $titleText = $title->getText(); - - // generate all variants of the link title text - $allTextVariants = $wgContLang->convertLinkToAllVariants($titleText); - - // if link was not found (in first query), add all variants to query - if ( !isset($colours[$pdbk]) ){ - foreach($allTextVariants as $textVariant){ - if($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 $variant){ - if($variant != $category){ - $variantTitle = Title::newFromDBkey( Title::makeName(NS_CATEGORY,$variant) ); - if(is_null($variantTitle)) continue; - $linkBatch->addObj( $variantTitle ); - $categoryMap[$variant] = $category; - } - } - } - - - if(!$linkBatch->isEmpty()){ - // construct query - $titleClause = $linkBatch->constructSet('page', $dbr); - - $variantQuery = "SELECT page_id, page_namespace, page_title"; - if ( $threshold > 0 ) { - $variantQuery .= ', page_len, page_is_redirect'; - } - - $variantQuery .= " FROM $page WHERE $titleClause"; - if ( $options & RLH_FOR_UPDATE ) { - $variantQuery .= ' FOR UPDATE'; - } - - $varRes = $dbr->query( $variantQuery, $fname ); - - // for each found variants, figure out link holders and replace - while ( $s = $dbr->fetchObject($varRes) ) { - - $variantTitle = Title::makeTitle( $s->page_namespace, $s->page_title ); - $varPdbk = $variantTitle->getPrefixedDBkey(); - $vardbk = $variantTitle->getDBkey(); - - $holderKeys = array(); - if(isset($variantMap[$varPdbk])){ - $holderKeys = $variantMap[$varPdbk]; - $linkCache->addGoodLinkObj( $s->page_id, $variantTitle ); - $this->mOutput->addLink( $variantTitle, $s->page_id ); - } - - // loop over link holders - foreach($holderKeys as $key){ - $title = $this->mLinkHolders['titles'][$key]; - if ( is_null( $title ) ) continue; - - $pdbk = $title->getPrefixedDBkey(); - - if(!isset($colours[$pdbk])){ - // found link in some of the variants, replace the link holder data - $this->mLinkHolders['titles'][$key] = $variantTitle; - $this->mLinkHolders['dbkeys'][$key] = $variantTitle->getDBkey(); - - // set pdbk and colour - $pdbks[$key] = $varPdbk; - if ( $threshold > 0 ) { - $size = $s->page_len; - if ( $s->page_is_redirect || $s->page_namespace != 0 || $size >= $threshold ) { - $colours[$varPdbk] = 1; - } else { - $colours[$varPdbk] = 2; - } - } - else { - $colours[$varPdbk] = 1; - } - } - } - - // 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' ); - $replacePairs = array(); - foreach ( $this->mLinkHolders['namespaces'] as $key => $ns ) { - $pdbk = $pdbks[$key]; - $searchkey = ""; - $title = $this->mLinkHolders['titles'][$key]; - if ( empty( $colours[$pdbk] ) ) { - $linkCache->addBadLinkObj( $title ); - $colours[$pdbk] = 0; - $this->mOutput->addLink( $title, 0 ); - $replacePairs[$searchkey] = $sk->makeBrokenLinkObj( $title, - $this->mLinkHolders['texts'][$key], - $this->mLinkHolders['queries'][$key] ); - } elseif ( $colours[$pdbk] == 1 ) { - $replacePairs[$searchkey] = $sk->makeKnownLinkObj( $title, - $this->mLinkHolders['texts'][$key], - $this->mLinkHolders['queries'][$key] ); - } elseif ( $colours[$pdbk] == 2 ) { - $replacePairs[$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( - '/()/', - $replacer->cb(), - $text); - - wfProfileOut( $fname.'-replace' ); - } - - # Now process interwiki link holders - # This is quite a bit simpler than internal links - if ( !empty( $this->mInterwikiLinkHolders['texts'] ) ) { - wfProfileIn( $fname.'-interwiki' ); - # Make interwiki link HTML - $replacePairs = array(); - foreach( $this->mInterwikiLinkHolders['texts'] as $key => $link ) { - $title = $this->mInterwikiLinkHolders['titles'][$key]; - $replacePairs[$key] = $sk->makeLinkObj( $title, $link ); - } - $replacer = new HashtableReplacer( $replacePairs, 1 ); - - $text = preg_replace_callback( - '//', - $replacer->cb(), - $text ); - wfProfileOut( $fname.'-interwiki' ); - } - - wfProfileOut( $fname ); - return $colours; - } - - /** - * Replace link placeholders with plain text of links - * (not HTML-formatted). - * @param string $text - * @return string - */ - function replaceLinkHoldersText( $text ) { - $fname = 'Parser::replaceLinkHoldersText'; - wfProfileIn( $fname ); - - $text = preg_replace_callback( - '//', - array( &$this, 'replaceLinkHoldersTextCallback' ), - $text ); - - wfProfileOut( $fname ); - return $text; - } - - /** - * @param array $matches - * @return string - * @private - */ - function replaceLinkHoldersTextCallback( $matches ) { - $type = $matches[1]; - $key = $matches[2]; - if( $type == 'LINK' ) { - if( isset( $this->mLinkHolders['texts'][$key] ) ) { - return $this->mLinkHolders['texts'][$key]; - } - } elseif( $type == 'IWLINK' ) { - if( isset( $this->mInterwikiLinkHolders['texts'][$key] ) ) { - return $this->mInterwikiLinkHolders['texts'][$key]; - } - } - return $matches[0]; - } - - /** - * Tag hook handler for 'pre'. - */ - function renderPreTag( $text, $attribs ) { - // Backwards-compatibility hack - $content = StringUtils::delimiterReplace( '', '', '$1', $text, 'i' ); - - $attribs = Sanitizer::validateTagAttributes( $attribs, 'pre' ); - return wfOpenElement( 'pre', $attribs ) . - Xml::escapeTagsOnly( $content ) . - '
    '; - } - - /** - * 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, $params ) { - $ig = new ImageGallery(); - $ig->setContextTitle( $this->mTitle ); - $ig->setShowBytes( false ); - $ig->setShowFilename( false ); - $ig->setParser( $this ); - $ig->setHideBadImages(); - $ig->setAttributes( Sanitizer::validateTagAttributes( $params, 'table' ) ); - $ig->useSkin( $this->mOptions->getSkin() ); - $ig->mRevisionId = $this->mRevisionId; - - if( isset( $params['caption'] ) ) { - $caption = $params['caption']; - $caption = htmlspecialchars( $caption ); - $caption = $this->replaceInternalLinks( $caption ); - $ig->setCaptionHtml( $caption ); - } - if( isset( $params['perrow'] ) ) { - $ig->setPerRow( $params['perrow'] ); - } - if( isset( $params['widths'] ) ) { - $ig->setWidths( $params['widths'] ); - } - if( isset( $params['heights'] ) ) { - $ig->setHeights( $params['heights'] ); - } - - wfRunHooks( 'BeforeParserrenderImageGallery', array( &$this, &$ig ) ); - - $lines = explode( "\n", $text ); - foreach ( $lines as $line ) { - # match lines like these: - # Image:someimage.jpg|This is some image - $matches = array(); - preg_match( "/^([^|]+)(\\|(.*))?$/", $line, $matches ); - # Skip empty lines - if ( count( $matches ) == 0 ) { - continue; - } - $tp = Title::newFromText( $matches[1] ); - $nt =& $tp; - if( is_null( $nt ) ) { - # Bogus title. Ignore these so we don't bomb out later. - continue; - } - if ( isset( $matches[3] ) ) { - $label = $matches[3]; - } else { - $label = ''; - } - - $pout = $this->parse( $label, - $this->mTitle, - $this->mOptions, - false, // Strip whitespace...? - false // Don't clear state! - ); - $html = $pout->getText(); - - $ig->add( $nt, $html ); - - # Only add real images (bug #5586) - if ( $nt->getNamespace() == NS_IMAGE ) { - $this->mOutput->addImage( $nt->getDBkey() ); - } - } - return $ig->toHTML(); - } - - function getImageParams( $handler ) { - if ( $handler ) { - $handlerClass = get_class( $handler ); - } else { - $handlerClass = ''; - } - if ( !isset( $this->mImageParams[$handlerClass] ) ) { - // Initialise static lists - static $internalParamNames = array( - 'horizAlign' => array( 'left', 'right', 'center', 'none' ), - 'vertAlign' => array( 'baseline', 'sub', 'super', 'top', 'text-top', 'middle', - 'bottom', 'text-bottom' ), - 'frame' => array( 'thumbnail', 'manualthumb', 'framed', 'frameless', - 'upright', 'border' ), - ); - static $internalParamMap; - if ( !$internalParamMap ) { - $internalParamMap = array(); - foreach ( $internalParamNames as $type => $names ) { - foreach ( $names as $name ) { - $magicName = str_replace( '-', '_', "img_$name" ); - $internalParamMap[$magicName] = array( $type, $name ); - } - } - } - - // Add handler params - $paramMap = $internalParamMap; - if ( $handler ) { - $handlerParamMap = $handler->getParamMap(); - foreach ( $handlerParamMap as $magic => $paramName ) { - $paramMap[$magic] = array( 'handler', $paramName ); - } - } - $this->mImageParams[$handlerClass] = $paramMap; - $this->mImageParamsMagicArray[$handlerClass] = new MagicWordArray( array_keys( $paramMap ) ); - } - return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] ); - } - - /** - * Parse image options text and use it to make an image - */ - function makeImage( $title, $options ) { - # @TODO: let the MediaHandler specify its transform parameters - # - # Check if the options text is of the form "options|alt text" - # Options are: - # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang - # * left no resizing, just left align. label is used for alt= only - # * right same, but right aligned - # * none same, but not aligned - # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox - # * center center the image - # * framed Keep original image size, no magnify-button. - # * frameless like 'thumb' but without a frame. Keeps user preferences for width - # * upright reduce width for upright images, rounded to full __0 px - # * border draw a 1px border around the image - # vertical-align values (no % or length right now): - # * baseline - # * sub - # * super - # * top - # * text-top - # * middle - # * bottom - # * text-bottom - - $parts = array_map( 'trim', explode( '|', $options) ); - $sk = $this->mOptions->getSkin(); - - # Give extensions a chance to select the file revision for us - $skip = $time = false; - wfRunHooks( 'BeforeParserMakeImageLinkObj', array( &$this, &$title, &$skip, &$time ) ); - - if ( $skip ) { - return $sk->makeLinkObj( $title ); - } - - # Get parameter map - $file = wfFindFile( $title, $time ); - $handler = $file ? $file->getHandler() : false; - - list( $paramMap, $mwArray ) = $this->getImageParams( $handler ); - - # Process the input parameters - $caption = ''; - $params = array( 'frame' => array(), 'handler' => array(), - 'horizAlign' => array(), 'vertAlign' => array() ); - foreach( $parts as $part ) { - list( $magicName, $value ) = $mwArray->matchVariableStartToEnd( $part ); - if ( isset( $paramMap[$magicName] ) ) { - list( $type, $paramName ) = $paramMap[$magicName]; - $params[$type][$paramName] = $value; - - // Special case; width and height come in one variable together - if( $type == 'handler' && $paramName == 'width' ) { - $m = array(); - if ( preg_match( '/^([0-9]*)x([0-9]*)$/', $value, $m ) ) { - $params[$type]['width'] = intval( $m[1] ); - $params[$type]['height'] = intval( $m[2] ); - } else { - $params[$type]['width'] = intval( $value ); - } - } - } else { - $caption = $part; - } - } - - # Process alignment parameters - if ( $params['horizAlign'] ) { - $params['frame']['align'] = key( $params['horizAlign'] ); - } - if ( $params['vertAlign'] ) { - $params['frame']['valign'] = key( $params['vertAlign'] ); - } - - # Validate the handler parameters - if ( $handler ) { - foreach ( $params['handler'] as $name => $value ) { - if ( !$handler->validateParam( $name, $value ) ) { - unset( $params['handler'][$name] ); - } - } - } - - # Strip bad stuff out of the alt text - $alt = $this->replaceLinkHoldersText( $caption ); - - # 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->mStripState->unstripBoth( $alt ); - $alt = Sanitizer::stripAllTags( $alt ); - - $params['frame']['alt'] = $alt; - $params['frame']['caption'] = $caption; - - # Linker does the rest - $ret = $sk->makeImageLink2( $title, $file, $params['frame'], $params['handler'] ); - - # Give the handler a chance to modify the parser object - if ( $handler ) { - $handler->parserTransformHook( $this, $file ); - } - - return $ret; - } - - /** - * Set a flag in the output object indicating that the content is dynamic and - * shouldn't be cached. - */ - function disableCache() { - wfDebug( "Parser output marked as uncacheable.\n" ); - $this->mOutput->mCacheTime = -1; - } - - /**#@+ - * Callback from the Sanitizer for expanding items found in HTML attribute - * values, so they can be safely tested and escaped. - * @param string $text - * @param array $args - * @return string - * @private - */ - function attributeStripCallback( &$text, $args ) { - $text = $this->replaceVariables( $text, $args ); - $text = $this->mStripState->unstripBoth( $text ); - return $text; - } - - /**#@-*/ - - /**#@+ - * Accessor/mutator - */ - function Title( $x = NULL ) { return wfSetVar( $this->mTitle, $x ); } - function Options( $x = NULL ) { return wfSetVar( $this->mOptions, $x ); } - function OutputType( $x = NULL ) { return wfSetVar( $this->mOutputType, $x ); } - /**#@-*/ - - /**#@+ - * Accessor - */ - function getTags() { return array_merge( array_keys($this->mTransparentTagHooks), array_keys( $this->mTagHooks ) ); } - /**#@-*/ - - - /** - * Break wikitext input into sections, and either pull or replace - * some particular section's text. - * - * External callers should use the getSection and replaceSection methods. - * - * @param $text Page wikitext - * @param $section Numbered section. 0 pulls the text before the first - * heading; other numbers will pull the given section - * along with its lower-level subsections. - * @param $mode One of "get" or "replace" - * @param $newtext Replacement text for section data. - * @return string for "get", the extracted section text. - * for "replace", the whole page with the section replaced. - */ - private function extractSections( $text, $section, $mode, $newtext='' ) { - # I.... _hope_ this is right. - # Otherwise, sometimes we don't have things initialized properly. - $this->clearState(); - - # strip NOWIKI etc. to avoid confusion (true-parameter causes HTML - # comments to be stripped as well) - $stripState = new StripState; - - $oldOutputType = $this->mOutputType; - $oldOptions = $this->mOptions; - $this->mOptions = new ParserOptions(); - $this->setOutputType( self::OT_WIKI ); - - $striptext = $this->strip( $text, $stripState, true ); - - $this->setOutputType( $oldOutputType ); - $this->mOptions = $oldOptions; - - # now that we can be sure that no pseudo-sections are in the source, - # split it up by section - $uniq = preg_quote( $this->uniqPrefix(), '/' ); - $comment = "(?:$uniq-!--.*?QINU\x07)"; - $secs = preg_split( - "/ - ( - ^ - (?:$comment|<\/?noinclude>)* # Initial comments will be stripped - (=+) # Should this be limited to 6? - .+? # Section title... - \\2 # Ending = count must match start - (?:$comment|<\/?noinclude>|[ \\t]+)* # Trailing whitespace ok - $ - | - - .*? - <\/h\\3\s*> - ) - /mix", - $striptext, -1, - PREG_SPLIT_DELIM_CAPTURE); - - if( $mode == "get" ) { - if( $section == 0 ) { - // "Section 0" returns the content before any other section. - $rv = $secs[0]; - } else { - //track missing section, will replace if found. - $rv = $newtext; - } - } elseif( $mode == "replace" ) { - if( $section == 0 ) { - $rv = $newtext . "\n\n"; - $remainder = true; - } else { - $rv = $secs[0]; - $remainder = false; - } - } - $count = 0; - $sectionLevel = 0; - for( $index = 1; $index < count( $secs ); ) { - $headerLine = $secs[$index++]; - if( $secs[$index] ) { - // A wiki header - $headerLevel = strlen( $secs[$index++] ); - } else { - // An HTML header - $index++; - $headerLevel = intval( $secs[$index++] ); - } - $content = $secs[$index++]; - - $count++; - if( $mode == "get" ) { - if( $count == $section ) { - $rv = $headerLine . $content; - $sectionLevel = $headerLevel; - } elseif( $count > $section ) { - if( $sectionLevel && $headerLevel > $sectionLevel ) { - $rv .= $headerLine . $content; - } else { - // Broke out to a higher-level section - break; - } - } - } elseif( $mode == "replace" ) { - if( $count < $section ) { - $rv .= $headerLine . $content; - } elseif( $count == $section ) { - $rv .= $newtext . "\n\n"; - $sectionLevel = $headerLevel; - } elseif( $count > $section ) { - if( $headerLevel <= $sectionLevel ) { - // Passed the section's sub-parts. - $remainder = true; - } - if( $remainder ) { - $rv .= $headerLine . $content; - } - } - } - } - if (is_string($rv)) - # reinsert stripped tags - $rv = trim( $stripState->unstripBoth( $rv ) ); - - return $rv; - } - - /** - * This function returns the text of a section, specified by a number ($section). - * A section is text under a heading like == Heading == or \Heading\, or - * the first section before any such heading (section 0). - * - * If a section contains subsections, these are also returned. - * - * @param $text String: text to look in - * @param $section Integer: section number - * @param $deftext: default to return if section is not found - * @return string text of the requested section - */ - public function getSection( $text, $section, $deftext='' ) { - return $this->extractSections( $text, $section, "get", $deftext ); - } - - public function replaceSection( $oldtext, $section, $text ) { - 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; - } - - /** - * Mutator for $mDefaultSort - * - * @param $sort New value - */ - public function setDefaultSort( $sort ) { - $this->mDefaultSort = $sort; - } - - /** - * Accessor for $mDefaultSort - * Will use the title/prefixed title if none is set - * - * @return string - */ - public function getDefaultSort() { - if( $this->mDefaultSort !== false ) { - return $this->mDefaultSort; - } else { - return $this->mTitle->getNamespace() == NS_CATEGORY - ? $this->mTitle->getText() - : $this->mTitle->getPrefixedText(); - } - } - - /** - * Try to guess the section anchor name based on a wikitext fragment - * presumably extracted from a heading, for example "Header" from - * "== Header ==". - */ - public function guessSectionNameFromWikiText( $text ) { - # Strip out wikitext links(they break the anchor) - $text = $this->stripSectionName( $text ); - $headline = Sanitizer::decodeCharReferences( $text ); - # strip out HTML - $headline = StringUtils::delimiterReplace( '<', '>', '', $headline ); - $headline = trim( $headline ); - $sectionanchor = '#' . urlencode( str_replace( ' ', '_', $headline ) ); - $replacearray = array( - '%3A' => ':', - '%' => '.' - ); - return str_replace( - array_keys( $replacearray ), - array_values( $replacearray ), - $sectionanchor ); - } - - /** - * Strips a text string of wikitext for use in a section anchor - * - * Accepts a text string and then removes all wikitext from the - * string and leaves only the resultant text (i.e. the result of - * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of - * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended - * to create valid section anchors by mimicing the output of the - * parser when headings are parsed. - * - * @param $text string Text string to be stripped of wikitext - * for use in a Section anchor - * @return Filtered text string - */ - public function stripSectionName( $text ) { - # Strip internal link markup - $text = preg_replace('/\[\[:?([^[|]+)\|([^[]+)\]\]/','$2',$text); - $text = preg_replace('/\[\[:?([^[]+)\|?\]\]/','$1',$text); - - # Strip external link markup (FIXME: Not Tolerant to blank link text - # I.E. [http://www.mediawiki.org] will render as [1] or something depending - # on how many empty links there are on the page - need to figure that out. - $text = preg_replace('/\[(?:' . wfUrlProtocols() . ')([^ ]+?) ([^[]+)\]/','$2',$text); - - # Parse wikitext quotes (italics & bold) - $text = $this->doQuotes($text); - - # Strip HTML tags - $text = StringUtils::delimiterReplace( '<', '>', '', $text ); - return $text; - } - - /** - * strip/replaceVariables/unstrip for preprocessor regression testing - */ - function srvus( $text ) { - $text = $this->strip( $text, $this->mStripState ); - $text = Sanitizer::removeHTMLtags( $text ); - $text = $this->replaceVariables( $text ); - $text = preg_replace( '//', '', $text ); - $text = $this->mStripState->unstripBoth( $text ); - return $text; - } -} - diff --git a/includes/Preprocessor.php b/includes/Preprocessor.php deleted file mode 100644 index 34bc1e5b..00000000 --- a/includes/Preprocessor.php +++ /dev/null @@ -1,154 +0,0 @@ - node into an associative array containing: - * name PPNode name - * index String index - * value PPNode value - */ - function splitArg(); - - /** - * Split an node into an associative array containing name, attr, inner and close - * All values in the resulting array are PPNodes. Inner and close are optional. - */ - function splitExt(); - - /** - * Split an node - */ - function splitHeading(); -} - diff --git a/includes/Preprocessor_DOM.php b/includes/Preprocessor_DOM.php deleted file mode 100644 index 0e2e9a16..00000000 --- a/includes/Preprocessor_DOM.php +++ /dev/null @@ -1,1356 +0,0 @@ -parser = $parser; - $mem = ini_get( 'memory_limit' ); - $this->memoryLimit = false; - if ( strval( $mem ) !== '' && $mem != -1 ) { - if ( preg_match( '/^\d+$/', $mem ) ) { - $this->memoryLimit = $mem; - } elseif ( preg_match( '/^(\d+)M$/i', $mem, $m ) ) { - $this->memoryLimit = $m[1] * 1048576; - } - } - } - - function newFrame() { - return new PPFrame_DOM( $this ); - } - - function memCheck() { - if ( $this->memoryLimit === false ) { - return; - } - $usage = memory_get_usage(); - if ( $usage > $this->memoryLimit * 0.9 ) { - $limit = intval( $this->memoryLimit * 0.9 / 1048576 + 0.5 ); - throw new MWException( "Preprocessor hit 90% memory limit ($limit MB)" ); - } - return $usage <= $this->memoryLimit * 0.8; - } - - /** - * Preprocess some wikitext and return the document tree. - * This is the ghost of Parser::replace_variables(). - * - * @param string $text The text to parse - * @param integer flags Bitwise combination of: - * Parser::PTD_FOR_INCLUSION Handle / as if the text is being - * included. Default is to assume a direct page view. - * - * The generated DOM tree must depend only on the input text and the flags. - * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899. - * - * Any flag added to the $flags parameter here, or any other parameter liable to cause a - * change in the DOM tree for a given text, must be passed through the section identifier - * in the section edit link and thus back to extractSections(). - * - * The output of this function is currently only cached in process memory, but a persistent - * cache may be implemented at a later date which takes further advantage of these strict - * dependency requirements. - * - * @private - */ - function preprocessToObj( $text, $flags = 0 ) { - wfProfileIn( __METHOD__ ); - wfProfileIn( __METHOD__.'-makexml' ); - - $rules = array( - '{' => array( - 'end' => '}', - 'names' => array( - 2 => 'template', - 3 => 'tplarg', - ), - 'min' => 2, - 'max' => 3, - ), - '[' => array( - 'end' => ']', - 'names' => array( 2 => null ), - 'min' => 2, - 'max' => 2, - ) - ); - - $forInclusion = $flags & Parser::PTD_FOR_INCLUSION; - - $xmlishElements = $this->parser->getStripList(); - $enableOnlyinclude = false; - if ( $forInclusion ) { - $ignoredTags = array( 'includeonly', '/includeonly' ); - $ignoredElements = array( 'noinclude' ); - $xmlishElements[] = 'noinclude'; - if ( strpos( $text, '' ) !== false && strpos( $text, '' ) !== false ) { - $enableOnlyinclude = true; - } - } else { - $ignoredTags = array( 'noinclude', '/noinclude', 'onlyinclude', '/onlyinclude' ); - $ignoredElements = array( 'includeonly' ); - $xmlishElements[] = 'includeonly'; - } - $xmlishRegex = implode( '|', array_merge( $xmlishElements, $ignoredTags ) ); - - // Use "A" modifier (anchored) instead of "^", because ^ doesn't work with an offset - $elementsRegex = "~($xmlishRegex)(?:\s|\/>|>)|(!--)~iA"; - - $stack = new PPDStack; - - $searchBase = "[{<\n"; #} - $revText = strrev( $text ); // For fast reverse searches - - $i = 0; # Input pointer, starts out pointing to a pseudo-newline before the start - $accum =& $stack->getAccum(); # Current accumulator - $accum = ''; - $findEquals = false; # True to find equals signs in arguments - $findPipe = false; # True to take notice of pipe characters - $headingIndex = 1; - $inHeading = false; # True if $i is inside a possible heading - $noMoreGT = false; # True if there are no more greater-than (>) signs right of $i - $findOnlyinclude = $enableOnlyinclude; # True to ignore all input up to the next - $fakeLineStart = true; # Do a line-start run without outputting an LF character - - while ( true ) { - //$this->memCheck(); - - if ( $findOnlyinclude ) { - // Ignore all input up to the next - $startPos = strpos( $text, '', $i ); - if ( $startPos === false ) { - // Ignored section runs to the end - $accum .= '' . htmlspecialchars( substr( $text, $i ) ) . ''; - break; - } - $tagEndPos = $startPos + strlen( '' ); // past-the-end - $accum .= '' . htmlspecialchars( substr( $text, $i, $tagEndPos - $i ) ) . ''; - $i = $tagEndPos; - $findOnlyinclude = false; - } - - if ( $fakeLineStart ) { - $found = 'line-start'; - $curChar = ''; - } else { - # Find next opening brace, closing brace or pipe - $search = $searchBase; - if ( $stack->top === false ) { - $currentClosing = ''; - } else { - $currentClosing = $stack->top->close; - $search .= $currentClosing; - } - if ( $findPipe ) { - $search .= '|'; - } - if ( $findEquals ) { - // First equals will be for the template - $search .= '='; - } - $rule = null; - # Output literal section, advance input counter - $literalLength = strcspn( $text, $search, $i ); - if ( $literalLength > 0 ) { - $accum .= htmlspecialchars( substr( $text, $i, $literalLength ) ); - $i += $literalLength; - } - if ( $i >= strlen( $text ) ) { - if ( $currentClosing == "\n" ) { - // Do a past-the-end run to finish off the heading - $curChar = ''; - $found = 'line-end'; - } else { - # All done - break; - } - } else { - $curChar = $text[$i]; - if ( $curChar == '|' ) { - $found = 'pipe'; - } elseif ( $curChar == '=' ) { - $found = 'equals'; - } elseif ( $curChar == '<' ) { - $found = 'angle'; - } elseif ( $curChar == "\n" ) { - if ( $inHeading ) { - $found = 'line-end'; - } else { - $found = 'line-start'; - } - } elseif ( $curChar == $currentClosing ) { - $found = 'close'; - } elseif ( isset( $rules[$curChar] ) ) { - $found = 'open'; - $rule = $rules[$curChar]; - } else { - # Some versions of PHP have a strcspn which stops on null characters - # Ignore and continue - ++$i; - continue; - } - } - } - - if ( $found == 'angle' ) { - $matches = false; - // Handle - if ( $enableOnlyinclude && substr( $text, $i, strlen( '' ) ) == '' ) { - $findOnlyinclude = true; - continue; - } - - // Determine element name - if ( !preg_match( $elementsRegex, $text, $matches, 0, $i + 1 ) ) { - // Element name missing or not listed - $accum .= '<'; - ++$i; - continue; - } - // Handle comments - if ( isset( $matches[2] ) && $matches[2] == '!--' ) { - // To avoid leaving blank lines, when a comment is both preceded - // and followed by a newline (ignoring spaces), trim leading and - // trailing spaces and one of the newlines. - - // Find the end - $endPos = strpos( $text, '-->', $i + 4 ); - if ( $endPos === false ) { - // Unclosed comment in input, runs to end - $inner = substr( $text, $i ); - $accum .= '' . htmlspecialchars( $inner ) . ''; - $i = strlen( $text ); - } else { - // Search backwards for leading whitespace - $wsStart = $i ? ( $i - strspn( $revText, ' ', strlen( $text ) - $i ) ) : 0; - // Search forwards for trailing whitespace - // $wsEnd will be the position of the last space - $wsEnd = $endPos + 2 + strspn( $text, ' ', $endPos + 3 ); - // Eat the line if possible - // TODO: This could theoretically be done if $wsStart == 0, i.e. for comments at - // the overall start. That's not how Sanitizer::removeHTMLcomments() did it, but - // it's a possible beneficial b/c break. - if ( $wsStart > 0 && substr( $text, $wsStart - 1, 1 ) == "\n" - && substr( $text, $wsEnd + 1, 1 ) == "\n" ) - { - $startPos = $wsStart; - $endPos = $wsEnd + 1; - // Remove leading whitespace from the end of the accumulator - // Sanity check first though - $wsLength = $i - $wsStart; - if ( $wsLength > 0 && substr( $accum, -$wsLength ) === str_repeat( ' ', $wsLength ) ) { - $accum = substr( $accum, 0, -$wsLength ); - } - // Do a line-start run next time to look for headings after the comment - $fakeLineStart = true; - } else { - // No line to eat, just take the comment itself - $startPos = $i; - $endPos += 2; - } - - if ( $stack->top ) { - $part = $stack->top->getCurrentPart(); - if ( isset( $part->commentEnd ) && $part->commentEnd == $wsStart - 1 ) { - // Comments abutting, no change in visual end - $part->commentEnd = $wsEnd; - } else { - $part->visualEnd = $wsStart; - $part->commentEnd = $endPos; - } - } - $i = $endPos + 1; - $inner = substr( $text, $startPos, $endPos - $startPos + 1 ); - $accum .= '' . htmlspecialchars( $inner ) . ''; - } - continue; - } - $name = $matches[1]; - $attrStart = $i + strlen( $name ) + 1; - - // Find end of tag - $tagEndPos = $noMoreGT ? false : strpos( $text, '>', $attrStart ); - if ( $tagEndPos === false ) { - // Infinite backtrack - // Disable tag search to prevent worst-case O(N^2) performance - $noMoreGT = true; - $accum .= '<'; - ++$i; - continue; - } - - // Handle ignored tags - if ( in_array( $name, $ignoredTags ) ) { - $accum .= '' . htmlspecialchars( substr( $text, $i, $tagEndPos - $i + 1 ) ) . ''; - $i = $tagEndPos + 1; - continue; - } - - $tagStartPos = $i; - if ( $text[$tagEndPos-1] == '/' ) { - $attrEnd = $tagEndPos - 1; - $inner = null; - $i = $tagEndPos + 1; - $close = ''; - } else { - $attrEnd = $tagEndPos; - // Find closing tag - if ( preg_match( "/<\/$name\s*>/i", $text, $matches, PREG_OFFSET_CAPTURE, $tagEndPos + 1 ) ) { - $inner = substr( $text, $tagEndPos + 1, $matches[0][1] - $tagEndPos - 1 ); - $i = $matches[0][1] + strlen( $matches[0][0] ); - $close = '' . htmlspecialchars( $matches[0][0] ) . ''; - } else { - // No end tag -- let it run out to the end of the text. - $inner = substr( $text, $tagEndPos + 1 ); - $i = strlen( $text ); - $close = ''; - } - } - // and just become tags - if ( in_array( $name, $ignoredElements ) ) { - $accum .= '' . htmlspecialchars( substr( $text, $tagStartPos, $i - $tagStartPos ) ) - . ''; - continue; - } - - $accum .= ''; - if ( $attrEnd <= $attrStart ) { - $attr = ''; - } else { - $attr = substr( $text, $attrStart, $attrEnd - $attrStart ); - } - $accum .= '' . htmlspecialchars( $name ) . '' . - // Note that the attr element contains the whitespace between name and attribute, - // this is necessary for precise reconstruction during pre-save transform. - '' . htmlspecialchars( $attr ) . ''; - if ( $inner !== null ) { - $accum .= '' . htmlspecialchars( $inner ) . ''; - } - $accum .= $close . ''; - } - - elseif ( $found == 'line-start' ) { - // Is this the start of a heading? - // Line break belongs before the heading element in any case - if ( $fakeLineStart ) { - $fakeLineStart = false; - } else { - $accum .= $curChar; - $i++; - } - - $count = strspn( $text, '=', $i, 6 ); - if ( $count == 1 && $findEquals ) { - // DWIM: This looks kind of like a name/value separator - // Let's let the equals handler have it and break the potential heading - // This is heuristic, but AFAICT the methods for completely correct disambiguation are very complex. - } elseif ( $count > 0 ) { - $piece = array( - 'open' => "\n", - 'close' => "\n", - 'parts' => array( new PPDPart( str_repeat( '=', $count ) ) ), - 'startPos' => $i, - 'count' => $count ); - $stack->push( $piece ); - $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); - $i += $count; - } - } - - elseif ( $found == 'line-end' ) { - $piece = $stack->top; - // A heading must be open, otherwise \n wouldn't have been in the search list - assert( $piece->open == "\n" ); - $part = $piece->getCurrentPart(); - // Search back through the input to see if it has a proper close - // Do this using the reversed string since the other solutions (end anchor, etc.) are inefficient - $wsLength = strspn( $revText, " \t", strlen( $text ) - $i ); - $searchStart = $i - $wsLength; - if ( isset( $part->commentEnd ) && $searchStart - 1 == $part->commentEnd ) { - // Comment found at line end - // Search for equals signs before the comment - $searchStart = $part->visualEnd; - $searchStart -= strspn( $revText, " \t", strlen( $text ) - $searchStart ); - } - $count = $piece->count; - $equalsLength = strspn( $revText, '=', strlen( $text ) - $searchStart ); - if ( $equalsLength > 0 ) { - if ( $i - $equalsLength == $piece->startPos ) { - // This is just a single string of equals signs on its own line - // Replicate the doHeadings behaviour /={count}(.+)={count}/ - // First find out how many equals signs there really are (don't stop at 6) - $count = $equalsLength; - if ( $count < 3 ) { - $count = 0; - } else { - $count = min( 6, intval( ( $count - 1 ) / 2 ) ); - } - } else { - $count = min( $equalsLength, $count ); - } - if ( $count > 0 ) { - // Normal match, output - $element = "$accum"; - $headingIndex++; - } else { - // Single equals sign on its own line, count=0 - $element = $accum; - } - } else { - // No match, no , just pass down the inner text - $element = $accum; - } - // Unwind the stack - $stack->pop(); - $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); - - // Append the result to the enclosing accumulator - $accum .= $element; - // Note that we do NOT increment the input pointer. - // This is because the closing linebreak could be the opening linebreak of - // another heading. Infinite loops are avoided because the next iteration MUST - // hit the heading open case above, which unconditionally increments the - // input pointer. - } - - elseif ( $found == 'open' ) { - # count opening brace characters - $count = strspn( $text, $curChar, $i ); - - # we need to add to stack only if opening brace count is enough for one of the rules - if ( $count >= $rule['min'] ) { - # Add it to the stack - $piece = array( - 'open' => $curChar, - 'close' => $rule['end'], - 'count' => $count, - 'lineStart' => ($i > 0 && $text[$i-1] == "\n"), - ); - - $stack->push( $piece ); - $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); - } else { - # Add literal brace(s) - $accum .= htmlspecialchars( str_repeat( $curChar, $count ) ); - } - $i += $count; - } - - elseif ( $found == 'close' ) { - $piece = $stack->top; - # lets check if there are enough characters for closing brace - $maxCount = $piece->count; - $count = strspn( $text, $curChar, $i, $maxCount ); - - # check for maximum matching characters (if there are 5 closing - # characters, we will probably need only 3 - depending on the rules) - $matchingCount = 0; - $rule = $rules[$piece->open]; - if ( $count > $rule['max'] ) { - # The specified maximum exists in the callback array, unless the caller - # has made an error - $matchingCount = $rule['max']; - } else { - # Count is less than the maximum - # Skip any gaps in the callback array to find the true largest match - # Need to use array_key_exists not isset because the callback can be null - $matchingCount = $count; - while ( $matchingCount > 0 && !array_key_exists( $matchingCount, $rule['names'] ) ) { - --$matchingCount; - } - } - - if ($matchingCount <= 0) { - # No matching element found in callback array - # Output a literal closing brace and continue - $accum .= htmlspecialchars( str_repeat( $curChar, $count ) ); - $i += $count; - continue; - } - $name = $rule['names'][$matchingCount]; - if ( $name === null ) { - // No element, just literal text - $element = $piece->breakSyntax( $matchingCount ) . str_repeat( $rule['end'], $matchingCount ); - } else { - # Create XML element - # Note: $parts is already XML, does not need to be encoded further - $parts = $piece->parts; - $title = $parts[0]->out; - unset( $parts[0] ); - - # The invocation is at the start of the line if lineStart is set in - # the stack, and all opening brackets are used up. - if ( $maxCount == $matchingCount && !empty( $piece->lineStart ) ) { - $attr = ' lineStart="1"'; - } else { - $attr = ''; - } - - $element = "<$name$attr>"; - $element .= "$title"; - $argIndex = 1; - foreach ( $parts as $partIndex => $part ) { - if ( isset( $part->eqpos ) ) { - $argName = substr( $part->out, 0, $part->eqpos ); - $argValue = substr( $part->out, $part->eqpos + 1 ); - $element .= "$argName=$argValue"; - } else { - $element .= "{$part->out}"; - $argIndex++; - } - } - $element .= ""; - } - - # Advance input pointer - $i += $matchingCount; - - # Unwind the stack - $stack->pop(); - $accum =& $stack->getAccum(); - - # Re-add the old stack element if it still has unmatched opening characters remaining - if ($matchingCount < $piece->count) { - $piece->parts = array( new PPDPart ); - $piece->count -= $matchingCount; - # do we still qualify for any callback with remaining count? - $names = $rules[$piece->open]['names']; - $skippedBraces = 0; - $enclosingAccum =& $accum; - while ( $piece->count ) { - if ( array_key_exists( $piece->count, $names ) ) { - $stack->push( $piece ); - $accum =& $stack->getAccum(); - break; - } - --$piece->count; - $skippedBraces ++; - } - $enclosingAccum .= str_repeat( $piece->open, $skippedBraces ); - } - - extract( $stack->getFlags() ); - - # Add XML element to the enclosing accumulator - $accum .= $element; - } - - elseif ( $found == 'pipe' ) { - $findEquals = true; // shortcut for getFlags() - $stack->addPart(); - $accum =& $stack->getAccum(); - ++$i; - } - - elseif ( $found == 'equals' ) { - $findEquals = false; // shortcut for getFlags() - $stack->getCurrentPart()->eqpos = strlen( $accum ); - $accum .= '='; - ++$i; - } - } - - # Output any remaining unclosed brackets - foreach ( $stack->stack as $piece ) { - $stack->rootAccum .= $piece->breakSyntax(); - } - $stack->rootAccum .= ''; - $xml = $stack->rootAccum; - - wfProfileOut( __METHOD__.'-makexml' ); - wfProfileIn( __METHOD__.'-loadXML' ); - $dom = new DOMDocument; - wfSuppressWarnings(); - $result = $dom->loadXML( $xml ); - wfRestoreWarnings(); - if ( !$result ) { - // Try running the XML through UtfNormal to get rid of invalid characters - $xml = UtfNormal::cleanUp( $xml ); - $result = $dom->loadXML( $xml ); - if ( !$result ) { - throw new MWException( __METHOD__.' generated invalid XML' ); - } - } - $obj = new PPNode_DOM( $dom->documentElement ); - wfProfileOut( __METHOD__.'-loadXML' ); - wfProfileOut( __METHOD__ ); - return $obj; - } -} - -/** - * Stack class to help Preprocessor::preprocessToObj() - */ -class PPDStack { - var $stack, $rootAccum, $top; - var $out; - var $elementClass = 'PPDStackElement'; - - static $false = false; - - function __construct() { - $this->stack = array(); - $this->top = false; - $this->rootAccum = ''; - $this->accum =& $this->rootAccum; - } - - function count() { - return count( $this->stack ); - } - - function &getAccum() { - return $this->accum; - } - - function getCurrentPart() { - if ( $this->top === false ) { - return false; - } else { - return $this->top->getCurrentPart(); - } - } - - function push( $data ) { - if ( $data instanceof $this->elementClass ) { - $this->stack[] = $data; - } else { - $class = $this->elementClass; - $this->stack[] = new $class( $data ); - } - $this->top = $this->stack[ count( $this->stack ) - 1 ]; - $this->accum =& $this->top->getAccum(); - } - - function pop() { - if ( !count( $this->stack ) ) { - throw new MWException( __METHOD__.': no elements remaining' ); - } - $temp = array_pop( $this->stack ); - - if ( count( $this->stack ) ) { - $this->top = $this->stack[ count( $this->stack ) - 1 ]; - $this->accum =& $this->top->getAccum(); - } else { - $this->top = self::$false; - $this->accum =& $this->rootAccum; - } - return $temp; - } - - function addPart( $s = '' ) { - $this->top->addPart( $s ); - $this->accum =& $this->top->getAccum(); - } - - function getFlags() { - if ( !count( $this->stack ) ) { - return array( - 'findEquals' => false, - 'findPipe' => false, - 'inHeading' => false, - ); - } else { - return $this->top->getFlags(); - } - } -} - -class PPDStackElement { - var $open, // Opening character (\n for heading) - $close, // Matching closing character - $count, // Number of opening characters found (number of "=" for heading) - $parts, // Array of PPDPart objects describing pipe-separated parts. - $lineStart; // True if the open char appeared at the start of the input line. Not set for headings. - - var $partClass = 'PPDPart'; - - function __construct( $data = array() ) { - $class = $this->partClass; - $this->parts = array( new $class ); - - foreach ( $data as $name => $value ) { - $this->$name = $value; - } - } - - function &getAccum() { - return $this->parts[count($this->parts) - 1]->out; - } - - function addPart( $s = '' ) { - $class = $this->partClass; - $this->parts[] = new $class( $s ); - } - - function getCurrentPart() { - return $this->parts[count($this->parts) - 1]; - } - - function getFlags() { - $partCount = count( $this->parts ); - $findPipe = $this->open != "\n" && $this->open != '['; - return array( - 'findPipe' => $findPipe, - 'findEquals' => $findPipe && $partCount > 1 && !isset( $this->parts[$partCount - 1]->eqpos ), - 'inHeading' => $this->open == "\n", - ); - } - - /** - * Get the output string that would result if the close is not found. - */ - function breakSyntax( $openingCount = false ) { - if ( $this->open == "\n" ) { - $s = $this->parts[0]->out; - } else { - if ( $openingCount === false ) { - $openingCount = $this->count; - } - $s = str_repeat( $this->open, $openingCount ); - $first = true; - foreach ( $this->parts as $part ) { - if ( $first ) { - $first = false; - } else { - $s .= '|'; - } - $s .= $part->out; - } - } - return $s; - } -} - -class PPDPart { - var $out; // Output accumulator string - - // Optional member variables: - // eqpos Position of equals sign in output accumulator - // commentEnd Past-the-end input pointer for the last comment encountered - // visualEnd Past-the-end input pointer for the end of the accumulator minus comments - - function __construct( $out = '' ) { - $this->out = $out; - } -} - -/** - * An expansion frame, used as a context to expand the result of preprocessToObj() - */ -class PPFrame_DOM implements PPFrame { - var $preprocessor, $parser, $title; - var $titleCache; - - /** - * Hashtable listing templates which are disallowed for expansion in this frame, - * having been encountered previously in parent frames. - */ - var $loopCheckHash; - - /** - * Recursion depth of this frame, top = 0 - */ - var $depth; - - - /** - * Construct a new preprocessor frame. - * @param Preprocessor $preprocessor The parent preprocessor - */ - function __construct( $preprocessor ) { - $this->preprocessor = $preprocessor; - $this->parser = $preprocessor->parser; - $this->title = $this->parser->mTitle; - $this->titleCache = array( $this->title ? $this->title->getPrefixedDBkey() : false ); - $this->loopCheckHash = array(); - $this->depth = 0; - } - - /** - * Create a new child frame - * $args is optionally a multi-root PPNode or array containing the template arguments - */ - function newChild( $args = false, $title = false ) { - $namedArgs = array(); - $numberedArgs = array(); - if ( $title === false ) { - $title = $this->title; - } - if ( $args !== false ) { - $xpath = false; - if ( $args instanceof PPNode ) { - $args = $args->node; - } - foreach ( $args as $arg ) { - if ( !$xpath ) { - $xpath = new DOMXPath( $arg->ownerDocument ); - } - - $nameNodes = $xpath->query( 'name', $arg ); - $value = $xpath->query( 'value', $arg ); - if ( $nameNodes->item( 0 )->hasAttributes() ) { - // Numbered parameter - $index = $nameNodes->item( 0 )->attributes->getNamedItem( 'index' )->textContent; - $numberedArgs[$index] = $value->item( 0 ); - unset( $namedArgs[$index] ); - } else { - // Named parameter - $name = trim( $this->expand( $nameNodes->item( 0 ), PPFrame::STRIP_COMMENTS ) ); - $namedArgs[$name] = $value->item( 0 ); - unset( $numberedArgs[$name] ); - } - } - } - return new PPTemplateFrame_DOM( $this->preprocessor, $this, $numberedArgs, $namedArgs, $title ); - } - - function expand( $root, $flags = 0 ) { - if ( is_string( $root ) ) { - return $root; - } - - if ( ++$this->parser->mPPNodeCount > $this->parser->mOptions->mMaxPPNodeCount ) - { - return 'Node-count limit exceeded'; - } - - if ( $root instanceof PPNode_DOM ) { - $root = $root->node; - } - if ( $root instanceof DOMDocument ) { - $root = $root->documentElement; - } - - $outStack = array( '', '' ); - $iteratorStack = array( false, $root ); - $indexStack = array( 0, 0 ); - - while ( count( $iteratorStack ) > 1 ) { - $level = count( $outStack ) - 1; - $iteratorNode =& $iteratorStack[ $level ]; - $out =& $outStack[$level]; - $index =& $indexStack[$level]; - - if ( $iteratorNode instanceof PPNode_DOM ) $iteratorNode = $iteratorNode->node; - - if ( is_array( $iteratorNode ) ) { - if ( $index >= count( $iteratorNode ) ) { - // All done with this iterator - $iteratorStack[$level] = false; - $contextNode = false; - } else { - $contextNode = $iteratorNode[$index]; - $index++; - } - } elseif ( $iteratorNode instanceof DOMNodeList ) { - if ( $index >= $iteratorNode->length ) { - // All done with this iterator - $iteratorStack[$level] = false; - $contextNode = false; - } else { - $contextNode = $iteratorNode->item( $index ); - $index++; - } - } else { - // Copy to $contextNode and then delete from iterator stack, - // because this is not an iterator but we do have to execute it once - $contextNode = $iteratorStack[$level]; - $iteratorStack[$level] = false; - } - - if ( $contextNode instanceof PPNode_DOM ) $contextNode = $contextNode->node; - - $newIterator = false; - - if ( $contextNode === false ) { - // nothing to do - } elseif ( is_string( $contextNode ) ) { - $out .= $contextNode; - } elseif ( is_array( $contextNode ) || $contextNode instanceof DOMNodeList ) { - $newIterator = $contextNode; - } elseif ( $contextNode instanceof DOMNode ) { - if ( $contextNode->nodeType == XML_TEXT_NODE ) { - $out .= $contextNode->nodeValue; - } elseif ( $contextNode->nodeName == 'template' ) { - # Double-brace expansion - $xpath = new DOMXPath( $contextNode->ownerDocument ); - $titles = $xpath->query( 'title', $contextNode ); - $title = $titles->item( 0 ); - $parts = $xpath->query( 'part', $contextNode ); - if ( $flags & self::NO_TEMPLATES ) { - $newIterator = $this->virtualBracketedImplode( '{{', '|', '}}', $title, $parts ); - } else { - $lineStart = $contextNode->getAttribute( 'lineStart' ); - $params = array( - 'title' => new PPNode_DOM( $title ), - 'parts' => new PPNode_DOM( $parts ), - 'lineStart' => $lineStart ); - $ret = $this->parser->braceSubstitution( $params, $this ); - if ( isset( $ret['object'] ) ) { - $newIterator = $ret['object']; - } else { - $out .= $ret['text']; - } - } - } elseif ( $contextNode->nodeName == 'tplarg' ) { - # Triple-brace expansion - $xpath = new DOMXPath( $contextNode->ownerDocument ); - $titles = $xpath->query( 'title', $contextNode ); - $title = $titles->item( 0 ); - $parts = $xpath->query( 'part', $contextNode ); - if ( $flags & self::NO_ARGS ) { - $newIterator = $this->virtualBracketedImplode( '{{{', '|', '}}}', $title, $parts ); - } else { - $params = array( - 'title' => new PPNode_DOM( $title ), - 'parts' => new PPNode_DOM( $parts ) ); - $ret = $this->parser->argSubstitution( $params, $this ); - if ( isset( $ret['object'] ) ) { - $newIterator = $ret['object']; - } else { - $out .= $ret['text']; - } - } - } elseif ( $contextNode->nodeName == 'comment' ) { - # HTML-style comment - # Remove it in HTML, pre+remove and STRIP_COMMENTS modes - if ( $this->parser->ot['html'] - || ( $this->parser->ot['pre'] && $this->parser->mOptions->getRemoveComments() ) - || ( $flags & self::STRIP_COMMENTS ) ) - { - $out .= ''; - } - # Add a strip marker in PST mode so that pstPass2() can run some old-fashioned regexes on the result - # Not in RECOVER_COMMENTS mode (extractSections) though - elseif ( $this->parser->ot['wiki'] && ! ( $flags & self::RECOVER_COMMENTS ) ) { - $out .= $this->parser->insertStripItem( $contextNode->textContent ); - } - # Recover the literal comment in RECOVER_COMMENTS and pre+no-remove - else { - $out .= $contextNode->textContent; - } - } elseif ( $contextNode->nodeName == 'ignore' ) { - # Output suppression used by etc. - # OT_WIKI will only respect in substed templates. - # The other output types respect it unless NO_IGNORE is set. - # extractSections() sets NO_IGNORE and so never respects it. - if ( ( !isset( $this->parent ) && $this->parser->ot['wiki'] ) || ( $flags & self::NO_IGNORE ) ) { - $out .= $contextNode->textContent; - } else { - $out .= ''; - } - } elseif ( $contextNode->nodeName == 'ext' ) { - # Extension tag - $xpath = new DOMXPath( $contextNode->ownerDocument ); - $names = $xpath->query( 'name', $contextNode ); - $attrs = $xpath->query( 'attr', $contextNode ); - $inners = $xpath->query( 'inner', $contextNode ); - $closes = $xpath->query( 'close', $contextNode ); - $params = array( - 'name' => new PPNode_DOM( $names->item( 0 ) ), - 'attr' => $attrs->length > 0 ? new PPNode_DOM( $attrs->item( 0 ) ) : null, - 'inner' => $inners->length > 0 ? new PPNode_DOM( $inners->item( 0 ) ) : null, - 'close' => $closes->length > 0 ? new PPNode_DOM( $closes->item( 0 ) ) : null, - ); - $out .= $this->parser->extensionSubstitution( $params, $this ); - } elseif ( $contextNode->nodeName == 'h' ) { - # Heading - $s = $this->expand( $contextNode->childNodes, $flags ); - - # Insert a heading marker only for children of - # This is to stop extractSections from going over multiple tree levels - if ( $contextNode->parentNode->nodeName == 'root' - && $this->parser->ot['html'] ) - { - # Insert heading index marker - $headingIndex = $contextNode->getAttribute( 'i' ); - $titleText = $this->title->getPrefixedDBkey(); - $this->parser->mHeadings[] = array( $titleText, $headingIndex ); - $serial = count( $this->parser->mHeadings ) - 1; - $marker = "{$this->parser->mUniqPrefix}-h-$serial-{$this->parser->mMarkerSuffix}"; - $count = $contextNode->getAttribute( 'level' ); - $s = substr( $s, 0, $count ) . $marker . substr( $s, $count ); - $this->parser->mStripState->general->setPair( $marker, '' ); - } - $out .= $s; - } else { - # Generic recursive expansion - $newIterator = $contextNode->childNodes; - } - } else { - throw new MWException( __METHOD__.': Invalid parameter type' ); - } - - if ( $newIterator !== false ) { - if ( $newIterator instanceof PPNode_DOM ) { - $newIterator = $newIterator->node; - } - $outStack[] = ''; - $iteratorStack[] = $newIterator; - $indexStack[] = 0; - } elseif ( $iteratorStack[$level] === false ) { - // Return accumulated value to parent - // With tail recursion - while ( $iteratorStack[$level] === false && $level > 0 ) { - $outStack[$level - 1] .= $out; - array_pop( $outStack ); - array_pop( $iteratorStack ); - array_pop( $indexStack ); - $level--; - } - } - } - return $outStack[0]; - } - - function implodeWithFlags( $sep, $flags /*, ... */ ) { - $args = array_slice( func_get_args(), 2 ); - - $first = true; - $s = ''; - foreach ( $args as $root ) { - if ( $root instanceof PPNode_DOM ) $root = $root->node; - if ( !is_array( $root ) && !( $root instanceof DOMNodeList ) ) { - $root = array( $root ); - } - foreach ( $root as $node ) { - if ( $first ) { - $first = false; - } else { - $s .= $sep; - } - $s .= $this->expand( $node, $flags ); - } - } - return $s; - } - - /** - * Implode with no flags specified - * This previously called implodeWithFlags but has now been inlined to reduce stack depth - */ - function implode( $sep /*, ... */ ) { - $args = array_slice( func_get_args(), 1 ); - - $first = true; - $s = ''; - foreach ( $args as $root ) { - if ( $root instanceof PPNode_DOM ) $root = $root->node; - if ( !is_array( $root ) && !( $root instanceof DOMNodeList ) ) { - $root = array( $root ); - } - foreach ( $root as $node ) { - if ( $first ) { - $first = false; - } else { - $s .= $sep; - } - $s .= $this->expand( $node ); - } - } - return $s; - } - - /** - * Makes an object that, when expand()ed, will be the same as one obtained - * with implode() - */ - function virtualImplode( $sep /*, ... */ ) { - $args = array_slice( func_get_args(), 1 ); - $out = array(); - $first = true; - if ( $root instanceof PPNode_DOM ) $root = $root->node; - - foreach ( $args as $root ) { - if ( !is_array( $root ) && !( $root instanceof DOMNodeList ) ) { - $root = array( $root ); - } - foreach ( $root as $node ) { - if ( $first ) { - $first = false; - } else { - $out[] = $sep; - } - $out[] = $node; - } - } - return $out; - } - - /** - * Virtual implode with brackets - */ - function virtualBracketedImplode( $start, $sep, $end /*, ... */ ) { - $args = array_slice( func_get_args(), 3 ); - $out = array( $start ); - $first = true; - - foreach ( $args as $root ) { - if ( $root instanceof PPNode_DOM ) $root = $root->node; - if ( !is_array( $root ) && !( $root instanceof DOMNodeList ) ) { - $root = array( $root ); - } - foreach ( $root as $node ) { - if ( $first ) { - $first = false; - } else { - $out[] = $sep; - } - $out[] = $node; - } - } - $out[] = $end; - return $out; - } - - function __toString() { - return 'frame{}'; - } - - function getPDBK( $level = false ) { - if ( $level === false ) { - return $this->title->getPrefixedDBkey(); - } else { - return isset( $this->titleCache[$level] ) ? $this->titleCache[$level] : false; - } - } - - /** - * Returns true if there are no arguments in this frame - */ - function isEmpty() { - return true; - } - - function getArgument( $name ) { - return false; - } - - /** - * Returns true if the infinite loop check is OK, false if a loop is detected - */ - function loopCheck( $title ) { - return !isset( $this->loopCheckHash[$title->getPrefixedDBkey()] ); - } - - /** - * Return true if the frame is a template frame - */ - function isTemplate() { - return false; - } -} - -/** - * Expansion frame with template arguments - */ -class PPTemplateFrame_DOM extends PPFrame_DOM { - var $numberedArgs, $namedArgs, $parent; - var $numberedExpansionCache, $namedExpansionCache; - - function __construct( $preprocessor, $parent = false, $numberedArgs = array(), $namedArgs = array(), $title = false ) { - $this->preprocessor = $preprocessor; - $this->parser = $preprocessor->parser; - $this->parent = $parent; - $this->numberedArgs = $numberedArgs; - $this->namedArgs = $namedArgs; - $this->title = $title; - $pdbk = $title ? $title->getPrefixedDBkey() : false; - $this->titleCache = $parent->titleCache; - $this->titleCache[] = $pdbk; - $this->loopCheckHash = /*clone*/ $parent->loopCheckHash; - if ( $pdbk !== false ) { - $this->loopCheckHash[$pdbk] = true; - } - $this->depth = $parent->depth + 1; - $this->numberedExpansionCache = $this->namedExpansionCache = array(); - } - - function __toString() { - $s = 'tplframe{'; - $first = true; - $args = $this->numberedArgs + $this->namedArgs; - foreach ( $args as $name => $value ) { - if ( $first ) { - $first = false; - } else { - $s .= ', '; - } - $s .= "\"$name\":\"" . - str_replace( '"', '\\"', $value->ownerDocument->saveXML( $value ) ) . '"'; - } - $s .= '}'; - return $s; - } - /** - * Returns true if there are no arguments in this frame - */ - function isEmpty() { - return !count( $this->numberedArgs ) && !count( $this->namedArgs ); - } - - function getNumberedArgument( $index ) { - if ( !isset( $this->numberedArgs[$index] ) ) { - return false; - } - if ( !isset( $this->numberedExpansionCache[$index] ) ) { - # No trimming for unnamed arguments - $this->numberedExpansionCache[$index] = $this->parent->expand( $this->numberedArgs[$index], self::STRIP_COMMENTS ); - } - return $this->numberedExpansionCache[$index]; - } - - function getNamedArgument( $name ) { - if ( !isset( $this->namedArgs[$name] ) ) { - return false; - } - if ( !isset( $this->namedExpansionCache[$name] ) ) { - # Trim named arguments post-expand, for backwards compatibility - $this->namedExpansionCache[$name] = trim( - $this->parent->expand( $this->namedArgs[$name], self::STRIP_COMMENTS ) ); - } - return $this->namedExpansionCache[$name]; - } - - function getArgument( $name ) { - $text = $this->getNumberedArgument( $name ); - if ( $text === false ) { - $text = $this->getNamedArgument( $name ); - } - return $text; - } - - /** - * Return true if the frame is a template frame - */ - function isTemplate() { - return true; - } -} - -class PPNode_DOM implements PPNode { - var $node; - - function __construct( $node, $xpath = false ) { - $this->node = $node; - } - - function __get( $name ) { - if ( $name == 'xpath' ) { - $this->xpath = new DOMXPath( $this->node->ownerDocument ); - } - return $this->xpath; - } - - function __toString() { - if ( $this->node instanceof DOMNodeList ) { - $s = ''; - foreach ( $this->node as $node ) { - $s .= $node->ownerDocument->saveXML( $node ); - } - } else { - $s = $this->node->ownerDocument->saveXML( $this->node ); - } - return $s; - } - - function getChildren() { - return $this->node->childNodes ? new self( $this->node->childNodes ) : false; - } - - function getFirstChild() { - return $this->node->firstChild ? new self( $this->node->firstChild ) : false; - } - - function getNextSibling() { - return $this->node->nextSibling ? new self( $this->node->nextSibling ) : false; - } - - function getChildrenOfType( $type ) { - return new self( $this->xpath->query( $type, $this->node ) ); - } - - function getLength() { - if ( $this->node instanceof DOMNodeList ) { - return $this->node->length; - } else { - return false; - } - } - - function item( $i ) { - $item = $this->node->item( $i ); - return $item ? new self( $item ) : false; - } - - function getName() { - if ( $this->node instanceof DOMNodeList ) { - return '#nodelist'; - } else { - return $this->node->nodeName; - } - } - - /** - * Split a node into an associative array containing: - * name PPNode name - * index String index - * value PPNode value - */ - function splitArg() { - $names = $this->xpath->query( 'name', $this->node ); - $values = $this->xpath->query( 'value', $this->node ); - if ( !$names->length || !$values->length ) { - throw new MWException( 'Invalid brace node passed to ' . __METHOD__ ); - } - $name = $names->item( 0 ); - $index = $name->getAttribute( 'index' ); - return array( - 'name' => new self( $name ), - 'index' => $index, - 'value' => new self( $values->item( 0 ) ) ); - } - - /** - * Split an node into an associative array containing name, attr, inner and close - * All values in the resulting array are PPNodes. Inner and close are optional. - */ - function splitExt() { - $names = $this->xpath->query( 'name', $this->node ); - $attrs = $this->xpath->query( 'attr', $this->node ); - $inners = $this->xpath->query( 'inner', $this->node ); - $closes = $this->xpath->query( 'close', $this->node ); - if ( !$names->length || !$attrs->length ) { - throw new MWException( 'Invalid ext node passed to ' . __METHOD__ ); - } - $parts = array( - 'name' => new self( $names->item( 0 ) ), - 'attr' => new self( $attrs->item( 0 ) ) ); - if ( $inners->length ) { - $parts['inner'] = new self( $inners->item( 0 ) ); - } - if ( $closes->length ) { - $parts['close'] = new self( $closes->item( 0 ) ); - } - return $parts; - } - - /** - * Split a node - */ - function splitHeading() { - if ( !$this->nodeName == 'h' ) { - throw new MWException( 'Invalid h node passed to ' . __METHOD__ ); - } - return array( - 'i' => $this->node->getAttribute( 'i' ), - 'level' => $this->node->getAttribute( 'level' ), - 'contents' => $this->getChildren() - ); - } -} diff --git a/includes/Preprocessor_Hash.php b/includes/Preprocessor_Hash.php deleted file mode 100644 index 2034278d..00000000 --- a/includes/Preprocessor_Hash.php +++ /dev/null @@ -1,1471 +0,0 @@ - nodes that aren't at the top are replaced with - */ - -class Preprocessor_Hash implements Preprocessor { - var $parser; - - function __construct( $parser ) { - $this->parser = $parser; - } - - function newFrame() { - return new PPFrame_Hash( $this ); - } - - /** - * Preprocess some wikitext and return the document tree. - * This is the ghost of Parser::replace_variables(). - * - * @param string $text The text to parse - * @param integer flags Bitwise combination of: - * Parser::PTD_FOR_INCLUSION Handle / as if the text is being - * included. Default is to assume a direct page view. - * - * The generated DOM tree must depend only on the input text and the flags. - * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899. - * - * Any flag added to the $flags parameter here, or any other parameter liable to cause a - * change in the DOM tree for a given text, must be passed through the section identifier - * in the section edit link and thus back to extractSections(). - * - * The output of this function is currently only cached in process memory, but a persistent - * cache may be implemented at a later date which takes further advantage of these strict - * dependency requirements. - * - * @private - */ - function preprocessToObj( $text, $flags = 0 ) { - wfDebug( __METHOD__."\n" . $text . "\n" ); - wfProfileIn( __METHOD__ ); - - $rules = array( - '{' => array( - 'end' => '}', - 'names' => array( - 2 => 'template', - 3 => 'tplarg', - ), - 'min' => 2, - 'max' => 3, - ), - '[' => array( - 'end' => ']', - 'names' => array( 2 => null ), - 'min' => 2, - 'max' => 2, - ) - ); - - $forInclusion = $flags & Parser::PTD_FOR_INCLUSION; - - $xmlishElements = $this->parser->getStripList(); - $enableOnlyinclude = false; - if ( $forInclusion ) { - $ignoredTags = array( 'includeonly', '/includeonly' ); - $ignoredElements = array( 'noinclude' ); - $xmlishElements[] = 'noinclude'; - if ( strpos( $text, '' ) !== false && strpos( $text, '' ) !== false ) { - $enableOnlyinclude = true; - } - } else { - $ignoredTags = array( 'noinclude', '/noinclude', 'onlyinclude', '/onlyinclude' ); - $ignoredElements = array( 'includeonly' ); - $xmlishElements[] = 'includeonly'; - } - $xmlishRegex = implode( '|', array_merge( $xmlishElements, $ignoredTags ) ); - - // Use "A" modifier (anchored) instead of "^", because ^ doesn't work with an offset - $elementsRegex = "~($xmlishRegex)(?:\s|\/>|>)|(!--)~iA"; - - $stack = new PPDStack_Hash; - - $searchBase = "[{<\n"; - $revText = strrev( $text ); // For fast reverse searches - - $i = 0; # Input pointer, starts out pointing to a pseudo-newline before the start - $accum =& $stack->getAccum(); # Current accumulator - $findEquals = false; # True to find equals signs in arguments - $findPipe = false; # True to take notice of pipe characters - $headingIndex = 1; - $inHeading = false; # True if $i is inside a possible heading - $noMoreGT = false; # True if there are no more greater-than (>) signs right of $i - $findOnlyinclude = $enableOnlyinclude; # True to ignore all input up to the next - $fakeLineStart = true; # Do a line-start run without outputting an LF character - - while ( true ) { - //$this->memCheck(); - - if ( $findOnlyinclude ) { - // Ignore all input up to the next - $startPos = strpos( $text, '', $i ); - if ( $startPos === false ) { - // Ignored section runs to the end - $accum->addNodeWithText( 'ignore', substr( $text, $i ) ); - break; - } - $tagEndPos = $startPos + strlen( '' ); // past-the-end - $accum->addNodeWithText( 'ignore', substr( $text, $i, $tagEndPos - $i ) ); - $i = $tagEndPos; - $findOnlyinclude = false; - } - - if ( $fakeLineStart ) { - $found = 'line-start'; - $curChar = ''; - } else { - # Find next opening brace, closing brace or pipe - $search = $searchBase; - if ( $stack->top === false ) { - $currentClosing = ''; - } else { - $currentClosing = $stack->top->close; - $search .= $currentClosing; - } - if ( $findPipe ) { - $search .= '|'; - } - if ( $findEquals ) { - // First equals will be for the template - $search .= '='; - } - $rule = null; - # Output literal section, advance input counter - $literalLength = strcspn( $text, $search, $i ); - if ( $literalLength > 0 ) { - $accum->addLiteral( substr( $text, $i, $literalLength ) ); - $i += $literalLength; - } - if ( $i >= strlen( $text ) ) { - if ( $currentClosing == "\n" ) { - // Do a past-the-end run to finish off the heading - $curChar = ''; - $found = 'line-end'; - } else { - # All done - break; - } - } else { - $curChar = $text[$i]; - if ( $curChar == '|' ) { - $found = 'pipe'; - } elseif ( $curChar == '=' ) { - $found = 'equals'; - } elseif ( $curChar == '<' ) { - $found = 'angle'; - } elseif ( $curChar == "\n" ) { - if ( $inHeading ) { - $found = 'line-end'; - } else { - $found = 'line-start'; - } - } elseif ( $curChar == $currentClosing ) { - $found = 'close'; - } elseif ( isset( $rules[$curChar] ) ) { - $found = 'open'; - $rule = $rules[$curChar]; - } else { - # Some versions of PHP have a strcspn which stops on null characters - # Ignore and continue - ++$i; - continue; - } - } - } - - if ( $found == 'angle' ) { - $matches = false; - // Handle - if ( $enableOnlyinclude && substr( $text, $i, strlen( '' ) ) == '' ) { - $findOnlyinclude = true; - continue; - } - - // Determine element name - if ( !preg_match( $elementsRegex, $text, $matches, 0, $i + 1 ) ) { - // Element name missing or not listed - $accum->addLiteral( '<' ); - ++$i; - continue; - } - // Handle comments - if ( isset( $matches[2] ) && $matches[2] == '!--' ) { - // To avoid leaving blank lines, when a comment is both preceded - // and followed by a newline (ignoring spaces), trim leading and - // trailing spaces and one of the newlines. - - // Find the end - $endPos = strpos( $text, '-->', $i + 4 ); - if ( $endPos === false ) { - // Unclosed comment in input, runs to end - $inner = substr( $text, $i ); - $accum->addNodeWithText( 'comment', $inner ); - $i = strlen( $text ); - } else { - // Search backwards for leading whitespace - $wsStart = $i ? ( $i - strspn( $revText, ' ', strlen( $text ) - $i ) ) : 0; - // Search forwards for trailing whitespace - // $wsEnd will be the position of the last space - $wsEnd = $endPos + 2 + strspn( $text, ' ', $endPos + 3 ); - // Eat the line if possible - // TODO: This could theoretically be done if $wsStart == 0, i.e. for comments at - // the overall start. That's not how Sanitizer::removeHTMLcomments() did it, but - // it's a possible beneficial b/c break. - if ( $wsStart > 0 && substr( $text, $wsStart - 1, 1 ) == "\n" - && substr( $text, $wsEnd + 1, 1 ) == "\n" ) - { - $startPos = $wsStart; - $endPos = $wsEnd + 1; - // Remove leading whitespace from the end of the accumulator - // Sanity check first though - $wsLength = $i - $wsStart; - if ( $wsLength > 0 - && $accum->lastNode instanceof PPNode_Hash_Text - && substr( $accum->lastNode->value, -$wsLength ) === str_repeat( ' ', $wsLength ) ) - { - $accum->lastNode->value = substr( $accum->lastNode->value, 0, -$wsLength ); - } - // Do a line-start run next time to look for headings after the comment - $fakeLineStart = true; - } else { - // No line to eat, just take the comment itself - $startPos = $i; - $endPos += 2; - } - - if ( $stack->top ) { - $part = $stack->top->getCurrentPart(); - if ( isset( $part->commentEnd ) && $part->commentEnd == $wsStart - 1 ) { - // Comments abutting, no change in visual end - $part->commentEnd = $wsEnd; - } else { - $part->visualEnd = $wsStart; - $part->commentEnd = $endPos; - } - } - $i = $endPos + 1; - $inner = substr( $text, $startPos, $endPos - $startPos + 1 ); - $accum->addNodeWithText( 'comment', $inner ); - } - continue; - } - $name = $matches[1]; - $attrStart = $i + strlen( $name ) + 1; - - // Find end of tag - $tagEndPos = $noMoreGT ? false : strpos( $text, '>', $attrStart ); - if ( $tagEndPos === false ) { - // Infinite backtrack - // Disable tag search to prevent worst-case O(N^2) performance - $noMoreGT = true; - $accum->addLiteral( '<' ); - ++$i; - continue; - } - - // Handle ignored tags - if ( in_array( $name, $ignoredTags ) ) { - $accum->addNodeWithText( 'ignore', substr( $text, $i, $tagEndPos - $i + 1 ) ); - $i = $tagEndPos + 1; - continue; - } - - $tagStartPos = $i; - if ( $text[$tagEndPos-1] == '/' ) { - // Short end tag - $attrEnd = $tagEndPos - 1; - $inner = null; - $i = $tagEndPos + 1; - $close = null; - } else { - $attrEnd = $tagEndPos; - // Find closing tag - if ( preg_match( "/<\/$name\s*>/i", $text, $matches, PREG_OFFSET_CAPTURE, $tagEndPos + 1 ) ) { - $inner = substr( $text, $tagEndPos + 1, $matches[0][1] - $tagEndPos - 1 ); - $i = $matches[0][1] + strlen( $matches[0][0] ); - $close = $matches[0][0]; - } else { - // No end tag -- let it run out to the end of the text. - $inner = substr( $text, $tagEndPos + 1 ); - $i = strlen( $text ); - $close = null; - } - } - // and just become tags - if ( in_array( $name, $ignoredElements ) ) { - $accum->addNodeWithText( 'ignore', substr( $text, $tagStartPos, $i - $tagStartPos ) ); - continue; - } - - if ( $attrEnd <= $attrStart ) { - $attr = ''; - } else { - // Note that the attr element contains the whitespace between name and attribute, - // this is necessary for precise reconstruction during pre-save transform. - $attr = substr( $text, $attrStart, $attrEnd - $attrStart ); - } - - $extNode = new PPNode_Hash_Tree( 'ext' ); - $extNode->addChild( PPNode_Hash_Tree::newWithText( 'name', $name ) ); - $extNode->addChild( PPNode_Hash_Tree::newWithText( 'attr', $attr ) ); - if ( $inner !== null ) { - $extNode->addChild( PPNode_Hash_Tree::newWithText( 'inner', $inner ) ); - } - if ( $close !== null ) { - $extNode->addChild( PPNode_Hash_Tree::newWithText( 'close', $close ) ); - } - $accum->addNode( $extNode ); - } - - elseif ( $found == 'line-start' ) { - // Is this the start of a heading? - // Line break belongs before the heading element in any case - if ( $fakeLineStart ) { - $fakeLineStart = false; - } else { - $accum->addLiteral( $curChar ); - $i++; - } - - $count = strspn( $text, '=', $i, 6 ); - if ( $count == 1 && $findEquals ) { - // DWIM: This looks kind of like a name/value separator - // Let's let the equals handler have it and break the potential heading - // This is heuristic, but AFAICT the methods for completely correct disambiguation are very complex. - } elseif ( $count > 0 ) { - $piece = array( - 'open' => "\n", - 'close' => "\n", - 'parts' => array( new PPDPart_Hash( str_repeat( '=', $count ) ) ), - 'startPos' => $i, - 'count' => $count ); - $stack->push( $piece ); - $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); - $i += $count; - } - } - - elseif ( $found == 'line-end' ) { - $piece = $stack->top; - // A heading must be open, otherwise \n wouldn't have been in the search list - assert( $piece->open == "\n" ); - $part = $piece->getCurrentPart(); - // Search back through the input to see if it has a proper close - // Do this using the reversed string since the other solutions (end anchor, etc.) are inefficient - $wsLength = strspn( $revText, " \t", strlen( $text ) - $i ); - $searchStart = $i - $wsLength; - if ( isset( $part->commentEnd ) && $searchStart - 1 == $part->commentEnd ) { - // Comment found at line end - // Search for equals signs before the comment - $searchStart = $part->visualEnd; - $searchStart -= strspn( $revText, " \t", strlen( $text ) - $searchStart ); - } - $count = $piece->count; - $equalsLength = strspn( $revText, '=', strlen( $text ) - $searchStart ); - if ( $equalsLength > 0 ) { - if ( $i - $equalsLength == $piece->startPos ) { - // This is just a single string of equals signs on its own line - // Replicate the doHeadings behaviour /={count}(.+)={count}/ - // First find out how many equals signs there really are (don't stop at 6) - $count = $equalsLength; - if ( $count < 3 ) { - $count = 0; - } else { - $count = min( 6, intval( ( $count - 1 ) / 2 ) ); - } - } else { - $count = min( $equalsLength, $count ); - } - if ( $count > 0 ) { - // Normal match, output - $element = new PPNode_Hash_Tree( 'possible-h' ); - $element->addChild( new PPNode_Hash_Attr( 'level', $count ) ); - $element->addChild( new PPNode_Hash_Attr( 'i', $headingIndex++ ) ); - $element->lastChild->nextSibling = $accum->firstNode; - $element->lastChild = $accum->lastNode; - } else { - // Single equals sign on its own line, count=0 - $element = $accum; - } - } else { - // No match, no , just pass down the inner text - $element = $accum; - } - // Unwind the stack - $stack->pop(); - $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); - - // Append the result to the enclosing accumulator - if ( $element instanceof PPNode ) { - $accum->addNode( $element ); - } else { - $accum->addAccum( $element ); - } - // Note that we do NOT increment the input pointer. - // This is because the closing linebreak could be the opening linebreak of - // another heading. Infinite loops are avoided because the next iteration MUST - // hit the heading open case above, which unconditionally increments the - // input pointer. - } - - elseif ( $found == 'open' ) { - # count opening brace characters - $count = strspn( $text, $curChar, $i ); - - # we need to add to stack only if opening brace count is enough for one of the rules - if ( $count >= $rule['min'] ) { - # Add it to the stack - $piece = array( - 'open' => $curChar, - 'close' => $rule['end'], - 'count' => $count, - 'lineStart' => ($i > 0 && $text[$i-1] == "\n"), - ); - - $stack->push( $piece ); - $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); - } else { - # Add literal brace(s) - $accum->addLiteral( str_repeat( $curChar, $count ) ); - } - $i += $count; - } - - elseif ( $found == 'close' ) { - $piece = $stack->top; - # lets check if there are enough characters for closing brace - $maxCount = $piece->count; - $count = strspn( $text, $curChar, $i, $maxCount ); - - # check for maximum matching characters (if there are 5 closing - # characters, we will probably need only 3 - depending on the rules) - $matchingCount = 0; - $rule = $rules[$piece->open]; - if ( $count > $rule['max'] ) { - # The specified maximum exists in the callback array, unless the caller - # has made an error - $matchingCount = $rule['max']; - } else { - # Count is less than the maximum - # Skip any gaps in the callback array to find the true largest match - # Need to use array_key_exists not isset because the callback can be null - $matchingCount = $count; - while ( $matchingCount > 0 && !array_key_exists( $matchingCount, $rule['names'] ) ) { - --$matchingCount; - } - } - - if ($matchingCount <= 0) { - # No matching element found in callback array - # Output a literal closing brace and continue - $accum->addLiteral( str_repeat( $curChar, $count ) ); - $i += $count; - continue; - } - $name = $rule['names'][$matchingCount]; - if ( $name === null ) { - // No element, just literal text - $element = $piece->breakSyntax( $matchingCount ); - $element->addLiteral( str_repeat( $rule['end'], $matchingCount ) ); - } else { - # Create XML element - # Note: $parts is already XML, does not need to be encoded further - $parts = $piece->parts; - $titleAccum = $parts[0]->out; - unset( $parts[0] ); - - $element = new PPNode_Hash_Tree( $name ); - - # The invocation is at the start of the line if lineStart is set in - # the stack, and all opening brackets are used up. - if ( $maxCount == $matchingCount && !empty( $piece->lineStart ) ) { - $element->addChild( new PPNode_Hash_Attr( 'lineStart', 1 ) ); - } - $titleNode = new PPNode_Hash_Tree( 'title' ); - $titleNode->firstChild = $titleAccum->firstNode; - $titleNode->lastChild = $titleAccum->lastNode; - $element->addChild( $titleNode ); - $argIndex = 1; - foreach ( $parts as $partIndex => $part ) { - if ( isset( $part->eqpos ) ) { - // Find equals - $lastNode = false; - for ( $node = $part->out->firstNode; $node; $node = $node->nextSibling ) { - if ( $node === $part->eqpos ) { - break; - } - $lastNode = $node; - } - if ( !$node ) { - throw new MWException( __METHOD__. ': eqpos not found' ); - } - if ( $node->name !== 'equals' ) { - throw new MWException( __METHOD__ .': eqpos is not equals' ); - } - $equalsNode = $node; - - // Construct name node - $nameNode = new PPNode_Hash_Tree( 'name' ); - if ( $lastNode !== false ) { - $lastNode->nextSibling = false; - $nameNode->firstChild = $part->out->firstNode; - $nameNode->lastChild = $lastNode; - } - - // Construct value node - $valueNode = new PPNode_Hash_Tree( 'value' ); - if ( $equalsNode->nextSibling !== false ) { - $valueNode->firstChild = $equalsNode->nextSibling; - $valueNode->lastChild = $part->out->lastNode; - } - $partNode = new PPNode_Hash_Tree( 'part' ); - $partNode->addChild( $nameNode ); - $partNode->addChild( $equalsNode->firstChild ); - $partNode->addChild( $valueNode ); - $element->addChild( $partNode ); - } else { - $partNode = new PPNode_Hash_Tree( 'part' ); - $nameNode = new PPNode_Hash_Tree( 'name' ); - $nameNode->addChild( new PPNode_Hash_Attr( 'index', $argIndex++ ) ); - $valueNode = new PPNode_Hash_Tree( 'value' ); - $valueNode->firstChild = $part->out->firstNode; - $valueNode->lastChild = $part->out->lastNode; - $partNode->addChild( $nameNode ); - $partNode->addChild( $valueNode ); - $element->addChild( $partNode ); - } - } - } - - # Advance input pointer - $i += $matchingCount; - - # Unwind the stack - $stack->pop(); - $accum =& $stack->getAccum(); - - # Re-add the old stack element if it still has unmatched opening characters remaining - if ($matchingCount < $piece->count) { - $piece->parts = array( new PPDPart_Hash ); - $piece->count -= $matchingCount; - # do we still qualify for any callback with remaining count? - $names = $rules[$piece->open]['names']; - $skippedBraces = 0; - $enclosingAccum =& $accum; - while ( $piece->count ) { - if ( array_key_exists( $piece->count, $names ) ) { - $stack->push( $piece ); - $accum =& $stack->getAccum(); - break; - } - --$piece->count; - $skippedBraces ++; - } - $enclosingAccum->addLiteral( str_repeat( $piece->open, $skippedBraces ) ); - } - - extract( $stack->getFlags() ); - - # Add XML element to the enclosing accumulator - if ( $element instanceof PPNode ) { - $accum->addNode( $element ); - } else { - $accum->addAccum( $element ); - } - } - - elseif ( $found == 'pipe' ) { - $findEquals = true; // shortcut for getFlags() - $stack->addPart(); - $accum =& $stack->getAccum(); - ++$i; - } - - elseif ( $found == 'equals' ) { - $findEquals = false; // shortcut for getFlags() - $accum->addNodeWithText( 'equals', '=' ); - $stack->getCurrentPart()->eqpos = $accum->lastNode; - ++$i; - } - } - - # Output any remaining unclosed brackets - foreach ( $stack->stack as $piece ) { - $stack->rootAccum->addAccum( $piece->breakSyntax() ); - } - - # Enable top-level headings - for ( $node = $stack->rootAccum->firstNode; $node; $node = $node->nextSibling ) { - if ( isset( $node->name ) && $node->name === 'possible-h' ) { - $node->name = 'h'; - } - } - - $rootNode = new PPNode_Hash_Tree( 'root' ); - $rootNode->firstChild = $stack->rootAccum->firstNode; - $rootNode->lastChild = $stack->rootAccum->lastNode; - wfProfileOut( __METHOD__ ); - return $rootNode; - } -} - -/** - * Stack class to help Preprocessor::preprocessToObj() - */ -class PPDStack_Hash extends PPDStack { - function __construct() { - $this->elementClass = 'PPDStackElement_Hash'; - parent::__construct(); - $this->rootAccum = new PPDAccum_Hash; - } -} - -class PPDStackElement_Hash extends PPDStackElement { - function __construct( $data = array() ) { - $this->partClass = 'PPDPart_Hash'; - parent::__construct( $data ); - } - - /** - * Get the accumulator that would result if the close is not found. - */ - function breakSyntax( $openingCount = false ) { - if ( $this->open == "\n" ) { - $accum = $this->parts[0]->out; - } else { - if ( $openingCount === false ) { - $openingCount = $this->count; - } - $accum = new PPDAccum_Hash; - $accum->addLiteral( str_repeat( $this->open, $openingCount ) ); - $first = true; - foreach ( $this->parts as $part ) { - if ( $first ) { - $first = false; - } else { - $accum->addLiteral( '|' ); - } - $accum->addAccum( $part->out ); - } - } - return $accum; - } -} - -class PPDPart_Hash extends PPDPart { - function __construct( $out = '' ) { - $accum = new PPDAccum_Hash; - if ( $out !== '' ) { - $accum->addLiteral( $out ); - } - parent::__construct( $accum ); - } -} - -class PPDAccum_Hash { - var $firstNode, $lastNode; - - function __construct() { - $this->firstNode = $this->lastNode = false; - } - - /** - * Append a string literal - */ - function addLiteral( $s ) { - if ( $this->lastNode === false ) { - $this->firstNode = $this->lastNode = new PPNode_Hash_Text( $s ); - } elseif ( $this->lastNode instanceof PPNode_Hash_Text ) { - $this->lastNode->value .= $s; - } else { - $this->lastNode->nextSibling = new PPNode_Hash_Text( $s ); - $this->lastNode = $this->lastNode->nextSibling; - } - } - - /** - * Append a PPNode - */ - function addNode( PPNode $node ) { - if ( $this->lastNode === false ) { - $this->firstNode = $this->lastNode = $node; - } else { - $this->lastNode->nextSibling = $node; - $this->lastNode = $node; - } - } - - /** - * Append a tree node with text contents - */ - function addNodeWithText( $name, $value ) { - $node = PPNode_Hash_Tree::newWithText( $name, $value ); - $this->addNode( $node ); - } - - /** - * Append a PPAccum_Hash - * Takes over ownership of the nodes in the source argument. These nodes may - * subsequently be modified, especially nextSibling. - */ - function addAccum( $accum ) { - if ( $accum->lastNode === false ) { - // nothing to add - } elseif ( $this->lastNode === false ) { - $this->firstNode = $accum->firstNode; - $this->lastNode = $accum->lastNode; - } else { - $this->lastNode->nextSibling = $accum->firstNode; - $this->lastNode = $accum->lastNode; - } - } -} - -/** - * An expansion frame, used as a context to expand the result of preprocessToObj() - */ -class PPFrame_Hash implements PPFrame { - var $preprocessor, $parser, $title; - var $titleCache; - - /** - * Hashtable listing templates which are disallowed for expansion in this frame, - * having been encountered previously in parent frames. - */ - var $loopCheckHash; - - /** - * Recursion depth of this frame, top = 0 - */ - var $depth; - - - /** - * Construct a new preprocessor frame. - * @param Preprocessor $preprocessor The parent preprocessor - */ - function __construct( $preprocessor ) { - $this->preprocessor = $preprocessor; - $this->parser = $preprocessor->parser; - $this->title = $this->parser->mTitle; - $this->titleCache = array( $this->title ? $this->title->getPrefixedDBkey() : false ); - $this->loopCheckHash = array(); - $this->depth = 0; - } - - /** - * Create a new child frame - * $args is optionally a multi-root PPNode or array containing the template arguments - */ - function newChild( $args = false, $title = false ) { - $namedArgs = array(); - $numberedArgs = array(); - if ( $title === false ) { - $title = $this->title; - } - if ( $args !== false ) { - $xpath = false; - if ( $args instanceof PPNode_Hash_Array ) { - $args = $args->value; - } elseif ( !is_array( $args ) ) { - throw new MWException( __METHOD__ . ': $args must be array or PPNode_Hash_Array' ); - } - foreach ( $args as $arg ) { - $bits = $arg->splitArg(); - if ( $bits['index'] !== '' ) { - // Numbered parameter - $numberedArgs[$bits['index']] = $bits['value']; - unset( $namedArgs[$bits['index']] ); - } else { - // Named parameter - $name = trim( $this->expand( $bits['name'], PPFrame::STRIP_COMMENTS ) ); - $namedArgs[$name] = $bits['value']; - unset( $numberedArgs[$name] ); - } - } - } - return new PPTemplateFrame_Hash( $this->preprocessor, $this, $numberedArgs, $namedArgs, $title ); - } - - function expand( $root, $flags = 0 ) { - if ( is_string( $root ) ) { - return $root; - } - - if ( ++$this->parser->mPPNodeCount > $this->parser->mOptions->mMaxPPNodeCount ) - { - return 'Node-count limit exceeded'; - } - - $outStack = array( '', '' ); - $iteratorStack = array( false, $root ); - $indexStack = array( 0, 0 ); - - while ( count( $iteratorStack ) > 1 ) { - $level = count( $outStack ) - 1; - $iteratorNode =& $iteratorStack[ $level ]; - $out =& $outStack[$level]; - $index =& $indexStack[$level]; - - if ( is_array( $iteratorNode ) ) { - if ( $index >= count( $iteratorNode ) ) { - // All done with this iterator - $iteratorStack[$level] = false; - $contextNode = false; - } else { - $contextNode = $iteratorNode[$index]; - $index++; - } - } elseif ( $iteratorNode instanceof PPNode_Hash_Array ) { - if ( $index >= $iteratorNode->getLength() ) { - // All done with this iterator - $iteratorStack[$level] = false; - $contextNode = false; - } else { - $contextNode = $iteratorNode->item( $index ); - $index++; - } - } else { - // Copy to $contextNode and then delete from iterator stack, - // because this is not an iterator but we do have to execute it once - $contextNode = $iteratorStack[$level]; - $iteratorStack[$level] = false; - } - - $newIterator = false; - - if ( $contextNode === false ) { - // nothing to do - } elseif ( is_string( $contextNode ) ) { - $out .= $contextNode; - } elseif ( is_array( $contextNode ) || $contextNode instanceof PPNode_Hash_Array ) { - $newIterator = $contextNode; - } elseif ( $contextNode instanceof PPNode_Hash_Attr ) { - // No output - } elseif ( $contextNode instanceof PPNode_Hash_Text ) { - $out .= $contextNode->value; - } elseif ( $contextNode instanceof PPNode_Hash_Tree ) { - if ( $contextNode->name == 'template' ) { - # Double-brace expansion - $bits = $contextNode->splitTemplate(); - if ( $flags & self::NO_TEMPLATES ) { - $newIterator = $this->virtualBracketedImplode( '{{', '|', '}}', $bits['title'], $bits['parts'] ); - } else { - $ret = $this->parser->braceSubstitution( $bits, $this ); - if ( isset( $ret['object'] ) ) { - $newIterator = $ret['object']; - } else { - $out .= $ret['text']; - } - } - } elseif ( $contextNode->name == 'tplarg' ) { - # Triple-brace expansion - $bits = $contextNode->splitTemplate(); - if ( $flags & self::NO_ARGS ) { - $newIterator = $this->virtualBracketedImplode( '{{{', '|', '}}}', $bits['title'], $bits['parts'] ); - } else { - $ret = $this->parser->argSubstitution( $bits, $this ); - if ( isset( $ret['object'] ) ) { - $newIterator = $ret['object']; - } else { - $out .= $ret['text']; - } - } - } elseif ( $contextNode->name == 'comment' ) { - # HTML-style comment - # Remove it in HTML, pre+remove and STRIP_COMMENTS modes - if ( $this->parser->ot['html'] - || ( $this->parser->ot['pre'] && $this->parser->mOptions->getRemoveComments() ) - || ( $flags & self::STRIP_COMMENTS ) ) - { - $out .= ''; - } - # Add a strip marker in PST mode so that pstPass2() can run some old-fashioned regexes on the result - # Not in RECOVER_COMMENTS mode (extractSections) though - elseif ( $this->parser->ot['wiki'] && ! ( $flags & self::RECOVER_COMMENTS ) ) { - $out .= $this->parser->insertStripItem( $contextNode->firstChild->value ); - } - # Recover the literal comment in RECOVER_COMMENTS and pre+no-remove - else { - $out .= $contextNode->firstChild->value; - } - } elseif ( $contextNode->name == 'ignore' ) { - # Output suppression used by etc. - # OT_WIKI will only respect in substed templates. - # The other output types respect it unless NO_IGNORE is set. - # extractSections() sets NO_IGNORE and so never respects it. - if ( ( !isset( $this->parent ) && $this->parser->ot['wiki'] ) || ( $flags & self::NO_IGNORE ) ) { - $out .= $contextNode->firstChild->value; - } else { - //$out .= ''; - } - } elseif ( $contextNode->name == 'ext' ) { - # Extension tag - $bits = $contextNode->splitExt() + array( 'attr' => null, 'inner' => null, 'close' => null ); - $out .= $this->parser->extensionSubstitution( $bits, $this ); - } elseif ( $contextNode->name == 'h' ) { - # Heading - if ( $this->parser->ot['html'] ) { - # Expand immediately and insert heading index marker - $s = ''; - for ( $node = $contextNode->firstChild; $node; $node = $node->nextSibling ) { - $s .= $this->expand( $node, $flags ); - } - - $bits = $contextNode->splitHeading(); - $titleText = $this->title->getPrefixedDBkey(); - $this->parser->mHeadings[] = array( $titleText, $bits['i'] ); - $serial = count( $this->parser->mHeadings ) - 1; - $marker = "{$this->parser->mUniqPrefix}-h-$serial-{$this->parser->mMarkerSuffix}"; - $s = substr( $s, 0, $bits['level'] ) . $marker . substr( $s, $bits['level'] ); - $this->parser->mStripState->general->setPair( $marker, '' ); - $out .= $s; - } else { - # Expand in virtual stack - $newIterator = $contextNode->getChildren(); - } - } else { - # Generic recursive expansion - $newIterator = $contextNode->getChildren(); - } - } else { - throw new MWException( __METHOD__.': Invalid parameter type' ); - } - - if ( $newIterator !== false ) { - $outStack[] = ''; - $iteratorStack[] = $newIterator; - $indexStack[] = 0; - } elseif ( $iteratorStack[$level] === false ) { - // Return accumulated value to parent - // With tail recursion - while ( $iteratorStack[$level] === false && $level > 0 ) { - $outStack[$level - 1] .= $out; - array_pop( $outStack ); - array_pop( $iteratorStack ); - array_pop( $indexStack ); - $level--; - } - } - } - return $outStack[0]; - } - - function implodeWithFlags( $sep, $flags /*, ... */ ) { - $args = array_slice( func_get_args(), 2 ); - - $first = true; - $s = ''; - foreach ( $args as $root ) { - if ( $root instanceof PPNode_Hash_Array ) { - $root = $root->value; - } - if ( !is_array( $root ) ) { - $root = array( $root ); - } - foreach ( $root as $node ) { - if ( $first ) { - $first = false; - } else { - $s .= $sep; - } - $s .= $this->expand( $node, $flags ); - } - } - return $s; - } - - /** - * Implode with no flags specified - * This previously called implodeWithFlags but has now been inlined to reduce stack depth - */ - function implode( $sep /*, ... */ ) { - $args = array_slice( func_get_args(), 1 ); - - $first = true; - $s = ''; - foreach ( $args as $root ) { - if ( $root instanceof PPNode_Hash_Array ) { - $root = $root->value; - } - if ( !is_array( $root ) ) { - $root = array( $root ); - } - foreach ( $root as $node ) { - if ( $first ) { - $first = false; - } else { - $s .= $sep; - } - $s .= $this->expand( $node ); - } - } - return $s; - } - - /** - * Makes an object that, when expand()ed, will be the same as one obtained - * with implode() - */ - function virtualImplode( $sep /*, ... */ ) { - $args = array_slice( func_get_args(), 1 ); - $out = array(); - $first = true; - - foreach ( $args as $root ) { - if ( $root instanceof PPNode_Hash_Array ) { - $root = $root->value; - } - if ( !is_array( $root ) ) { - $root = array( $root ); - } - foreach ( $root as $node ) { - if ( $first ) { - $first = false; - } else { - $out[] = $sep; - } - $out[] = $node; - } - } - return new PPNode_Hash_Array( $out ); - } - - /** - * Virtual implode with brackets - */ - function virtualBracketedImplode( $start, $sep, $end /*, ... */ ) { - $args = array_slice( func_get_args(), 3 ); - $out = array( $start ); - $first = true; - - foreach ( $args as $root ) { - if ( $root instanceof PPNode_Hash_Array ) { - $root = $root->value; - } - if ( !is_array( $root ) ) { - $root = array( $root ); - } - foreach ( $root as $node ) { - if ( $first ) { - $first = false; - } else { - $out[] = $sep; - } - $out[] = $node; - } - } - $out[] = $end; - return new PPNode_Hash_Array( $out ); - } - - function __toString() { - return 'frame{}'; - } - - function getPDBK( $level = false ) { - if ( $level === false ) { - return $this->title->getPrefixedDBkey(); - } else { - return isset( $this->titleCache[$level] ) ? $this->titleCache[$level] : false; - } - } - - /** - * Returns true if there are no arguments in this frame - */ - function isEmpty() { - return true; - } - - function getArgument( $name ) { - return false; - } - - /** - * Returns true if the infinite loop check is OK, false if a loop is detected - */ - function loopCheck( $title ) { - return !isset( $this->loopCheckHash[$title->getPrefixedDBkey()] ); - } - - /** - * Return true if the frame is a template frame - */ - function isTemplate() { - return false; - } -} - -/** - * Expansion frame with template arguments - */ -class PPTemplateFrame_Hash extends PPFrame_Hash { - var $numberedArgs, $namedArgs, $parent; - var $numberedExpansionCache, $namedExpansionCache; - - function __construct( $preprocessor, $parent = false, $numberedArgs = array(), $namedArgs = array(), $title = false ) { - $this->preprocessor = $preprocessor; - $this->parser = $preprocessor->parser; - $this->parent = $parent; - $this->numberedArgs = $numberedArgs; - $this->namedArgs = $namedArgs; - $this->title = $title; - $pdbk = $title ? $title->getPrefixedDBkey() : false; - $this->titleCache = $parent->titleCache; - $this->titleCache[] = $pdbk; - $this->loopCheckHash = /*clone*/ $parent->loopCheckHash; - if ( $pdbk !== false ) { - $this->loopCheckHash[$pdbk] = true; - } - $this->depth = $parent->depth + 1; - $this->numberedExpansionCache = $this->namedExpansionCache = array(); - } - - function __toString() { - $s = 'tplframe{'; - $first = true; - $args = $this->numberedArgs + $this->namedArgs; - foreach ( $args as $name => $value ) { - if ( $first ) { - $first = false; - } else { - $s .= ', '; - } - $s .= "\"$name\":\"" . - str_replace( '"', '\\"', $value->__toString() ) . '"'; - } - $s .= '}'; - return $s; - } - /** - * Returns true if there are no arguments in this frame - */ - function isEmpty() { - return !count( $this->numberedArgs ) && !count( $this->namedArgs ); - } - - function getNumberedArgument( $index ) { - if ( !isset( $this->numberedArgs[$index] ) ) { - return false; - } - if ( !isset( $this->numberedExpansionCache[$index] ) ) { - # No trimming for unnamed arguments - $this->numberedExpansionCache[$index] = $this->parent->expand( $this->numberedArgs[$index], self::STRIP_COMMENTS ); - } - return $this->numberedExpansionCache[$index]; - } - - function getNamedArgument( $name ) { - if ( !isset( $this->namedArgs[$name] ) ) { - return false; - } - if ( !isset( $this->namedExpansionCache[$name] ) ) { - # Trim named arguments post-expand, for backwards compatibility - $this->namedExpansionCache[$name] = trim( - $this->parent->expand( $this->namedArgs[$name], self::STRIP_COMMENTS ) ); - } - return $this->namedExpansionCache[$name]; - } - - function getArgument( $name ) { - $text = $this->getNumberedArgument( $name ); - if ( $text === false ) { - $text = $this->getNamedArgument( $name ); - } - return $text; - } - - /** - * Return true if the frame is a template frame - */ - function isTemplate() { - return true; - } -} - -class PPNode_Hash_Tree implements PPNode { - var $name, $firstChild, $lastChild, $nextSibling; - - function __construct( $name ) { - $this->name = $name; - $this->firstChild = $this->lastChild = $this->nextSibling = false; - } - - function __toString() { - $inner = ''; - $attribs = ''; - for ( $node = $this->firstChild; $node; $node = $node->nextSibling ) { - if ( $node instanceof PPNode_Hash_Attr ) { - $attribs .= ' ' . $node->name . '="' . htmlspecialchars( $node->value ) . '"'; - } else { - $inner .= $node->__toString(); - } - } - if ( $inner === '' ) { - return "<{$this->name}$attribs/>"; - } else { - return "<{$this->name}$attribs>$innername}>"; - } - } - - function newWithText( $name, $text ) { - $obj = new self( $name ); - $obj->addChild( new PPNode_Hash_Text( $text ) ); - return $obj; - } - - function addChild( $node ) { - if ( $this->lastChild === false ) { - $this->firstChild = $this->lastChild = $node; - } else { - $this->lastChild->nextSibling = $node; - $this->lastChild = $node; - } - } - - function getChildren() { - $children = array(); - for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) { - $children[] = $child; - } - return new PPNode_Hash_Array( $children ); - } - - function getFirstChild() { - return $this->firstChild; - } - - function getNextSibling() { - return $this->nextSibling; - } - - function getChildrenOfType( $name ) { - $children = array(); - for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) { - if ( isset( $child->name ) && $child->name === $name ) { - $children[] = $name; - } - } - return $children; - } - - function getLength() { return false; } - function item( $i ) { return false; } - - function getName() { - return $this->name; - } - - /** - * Split a node into an associative array containing: - * name PPNode name - * index String index - * value PPNode value - */ - function splitArg() { - $bits = array(); - for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) { - if ( !isset( $child->name ) ) { - continue; - } - if ( $child->name === 'name' ) { - $bits['name'] = $child; - if ( $child->firstChild instanceof PPNode_Hash_Attr - && $child->firstChild->name === 'index' ) - { - $bits['index'] = $child->firstChild->value; - } - } elseif ( $child->name === 'value' ) { - $bits['value'] = $child; - } - } - - if ( !isset( $bits['name'] ) ) { - throw new MWException( 'Invalid brace node passed to ' . __METHOD__ ); - } - if ( !isset( $bits['index'] ) ) { - $bits['index'] = ''; - } - return $bits; - } - - /** - * Split an node into an associative array containing name, attr, inner and close - * All values in the resulting array are PPNodes. Inner and close are optional. - */ - function splitExt() { - $bits = array(); - for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) { - if ( !isset( $child->name ) ) { - continue; - } - if ( $child->name == 'name' ) { - $bits['name'] = $child; - } elseif ( $child->name == 'attr' ) { - $bits['attr'] = $child; - } elseif ( $child->name == 'inner' ) { - $bits['inner'] = $child; - } elseif ( $child->name == 'close' ) { - $bits['close'] = $child; - } - } - if ( !isset( $bits['name'] ) ) { - throw new MWException( 'Invalid ext node passed to ' . __METHOD__ ); - } - return $bits; - } - - /** - * Split an node - */ - function splitHeading() { - if ( $this->name !== 'h' ) { - throw new MWException( 'Invalid h node passed to ' . __METHOD__ ); - } - $bits = array(); - for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) { - if ( !isset( $child->name ) ) { - continue; - } - if ( $child->name == 'i' ) { - $bits['i'] = $child->value; - } elseif ( $child->name == 'level' ) { - $bits['level'] = $child->value; - } - } - if ( !isset( $bits['i'] ) ) { - throw new MWException( 'Invalid h node passed to ' . __METHOD__ ); - } - return $bits; - } - - /** - * Split a