diff options
author | Pierre Schmitz <pierre@archlinux.de> | 2011-06-22 11:28:20 +0200 |
---|---|---|
committer | Pierre Schmitz <pierre@archlinux.de> | 2011-06-22 11:28:20 +0200 |
commit | 9db190c7e736ec8d063187d4241b59feaf7dc2d1 (patch) | |
tree | 46d1a0dee7febef5c2d57a9f7b972be16a163b3d /includes/parser | |
parent | 78677c7bbdcc9739f6c10c75935898a20e1acd9e (diff) |
update to MediaWiki 1.17.0
Diffstat (limited to 'includes/parser')
-rw-r--r-- | includes/parser/CoreLinkFunctions.php | 5 | ||||
-rw-r--r-- | includes/parser/CoreParserFunctions.php | 64 | ||||
-rw-r--r-- | includes/parser/CoreTagHooks.php | 11 | ||||
-rw-r--r-- | includes/parser/DateFormatter.php | 10 | ||||
-rw-r--r-- | includes/parser/LinkHolderArray.php | 108 | ||||
-rw-r--r-- | includes/parser/Parser.php | 2105 | ||||
-rw-r--r-- | includes/parser/ParserCache.php | 172 | ||||
-rw-r--r-- | includes/parser/ParserOptions.php | 185 | ||||
-rw-r--r-- | includes/parser/ParserOutput.php | 178 | ||||
-rw-r--r-- | includes/parser/Parser_DiffTest.php | 16 | ||||
-rw-r--r-- | includes/parser/Parser_LinkHooks.php | 26 | ||||
-rw-r--r-- | includes/parser/Preprocessor.php | 8 | ||||
-rw-r--r-- | includes/parser/Preprocessor_DOM.php | 90 | ||||
-rw-r--r-- | includes/parser/Preprocessor_Hash.php | 72 | ||||
-rw-r--r-- | includes/parser/Tidy.php | 28 |
15 files changed, 1852 insertions, 1226 deletions
diff --git a/includes/parser/CoreLinkFunctions.php b/includes/parser/CoreLinkFunctions.php index d6d11880..913ec22b 100644 --- a/includes/parser/CoreLinkFunctions.php +++ b/includes/parser/CoreLinkFunctions.php @@ -1,4 +1,9 @@ <?php +/** + * Link functions provided by MediaWiki core; experimental + * + * @file + */ /** * Various core link functions, registered in Parser::firstCallInit() diff --git a/includes/parser/CoreParserFunctions.php b/includes/parser/CoreParserFunctions.php index 8abcc04f..94949221 100644 --- a/includes/parser/CoreParserFunctions.php +++ b/includes/parser/CoreParserFunctions.php @@ -1,4 +1,9 @@ <?php +/** + * Parser functions provided by MediaWiki core + * + * @file + */ /** * Various core parser functions, registered in Parser::firstCallInit() @@ -11,7 +16,7 @@ class CoreParserFunctions { # 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:...} + # optional SFH_NO_HASH to omit the hash from calls (e.g. {{int:...}} # instead of {{#int:...}}) $parser->setFunctionHook( 'int', array( __CLASS__, 'intFunction' ), SFH_NO_HASH ); @@ -81,7 +86,7 @@ class CoreParserFunctions { static function intFunction( $parser, $part1 = '' /*, ... */ ) { if ( strval( $part1 ) !== '' ) { $args = array_slice( func_get_args(), 2 ); - $message = wfMsgGetKey( $part1, true, false, false ); + $message = wfMsgGetKey( $part1, true, $parser->getOptions()->getUserLang(), false ); $message = wfMsgReplaceArgs( $message, $args ); $message = $parser->replaceVariables( $message ); // like $wgMessageCache->transform() return $message; @@ -95,7 +100,7 @@ class CoreParserFunctions { $date = trim( $date ); - $pref = $parser->mOptions->getDateFormat(); + $pref = $parser->getOptions()->getDateFormat(); // Specify a different default date format other than the the normal default // iff the user has 'default' for their setting @@ -124,8 +129,37 @@ class CoreParserFunctions { return wfUrlencode( str_replace( ' ', '_', self::ns( $parser, $part1 ) ) ); } - static function urlencode( $parser, $s = '' ) { - return urlencode( $s ); + /** + * urlencodes a string according to one of three patterns: (bug 22474) + * + * By default (for HTTP "query" strings), spaces are encoded as '+'. + * Or to encode a value for the HTTP "path", spaces are encoded as '%20'. + * For links to "wiki"s, or similar software, spaces are encoded as '_', + * + * @param $parser Parser object + * @param $s String: The text to encode. + * @param $arg String (optional): The type of encoding. + */ + static function urlencode( $parser, $s = '', $arg = null ) { + static $magicWords = null; + if ( is_null( $magicWords ) ) { + $magicWords = new MagicWordArray( array( 'url_path', 'url_query', 'url_wiki' ) ); + } + switch( $magicWords->matchStartToEnd( $arg ) ) { + + // Encode as though it's a wiki page, '_' for ' '. + case 'url_wiki': + return wfUrlencode( str_replace( ' ', '_', $s ) ); + + // Encode for an HTTP Path, '%20' for ' '. + case 'url_path': + return rawurlencode( $s ); + + // Encode for HTTP query, '+' for ' '. + case 'url_query': + default: + return urlencode( $s ); + } } static function lcfirst( $parser, $s = '' ) { @@ -214,7 +248,7 @@ class CoreParserFunctions { $user = User::newFromName( $user ); if ( $user ) { $gender = $user->getOption( 'gender' ); - } elseif ( $parser->mOptions->getInterfaceMessage() ) { + } elseif ( $parser->getOptions()->getInterfaceMessage() ) { global $wgUser; $gender = $wgUser->getOption( 'gender' ); } @@ -553,17 +587,14 @@ class CoreParserFunctions { } static function anchorencode( $parser, $text ) { - $a = urlencode( $text ); - $a = strtr( $a, array( '%' => '.', '+' => '_' ) ); - # leave colons alone, however - $a = str_replace( '.3A', ':', $a ); - return $a; + return substr( $parser->guessSectionNameFromWikiText( $text ), 1); } static function special( $parser, $text ) { - $title = SpecialPage::getTitleForAlias( $text ); - if ( $title ) { - return $title->getPrefixedText(); + list( $page, $subpage ) = SpecialPage::resolveAliasWithSubpage( $text ); + if ( $page ) { + $title = SpecialPage::getTitleFor( $page, $subpage ); + return $title; } else { return wfMsgForContent( 'nosuchspecialpage' ); } @@ -579,7 +610,7 @@ class CoreParserFunctions { return ''; else return( '<span class="error">' . - wfMsg( 'duplicate-defaultsort', + wfMsgForContent( 'duplicate-defaultsort', htmlspecialchars( $old ), htmlspecialchars( $text ) ) . '</span>' ); @@ -602,7 +633,6 @@ class CoreParserFunctions { * Parser function to extension tag adaptor */ public static function tagObj( $parser, $frame, $args ) { - $xpath = false; if ( !count( $args ) ) { return ''; } @@ -617,7 +647,7 @@ class CoreParserFunctions { $stripList = $parser->getStripList(); if ( !in_array( $tagName, $stripList ) ) { return '<span class="error">' . - wfMsg( 'unknown_extension_tag', $tagName ) . + wfMsgForContent( 'unknown_extension_tag', $tagName ) . '</span>'; } diff --git a/includes/parser/CoreTagHooks.php b/includes/parser/CoreTagHooks.php index 7cc8260e..33f3c824 100644 --- a/includes/parser/CoreTagHooks.php +++ b/includes/parser/CoreTagHooks.php @@ -1,5 +1,14 @@ <?php +/** + * Tag hooks provided by MediaWiki core + * + * @file + */ +/** + * Various tag hooks, registered in Parser::firstCallInit() + * @ingroup Parser + */ class CoreTagHooks { static function register( $parser ) { global $wgRawHtml, $wgUseTeX; @@ -40,7 +49,7 @@ class CoreTagHooks { static function math( $content, $attributes, $parser ) { global $wgContLang; - return $wgContLang->armourMath( MathRenderer::renderMath( $content, $attributes ) ); + return $wgContLang->armourMath( MathRenderer::renderMath( $content, $attributes, $parser->getOptions() ) ); } static function gallery( $content, $attributes, $parser ) { diff --git a/includes/parser/DateFormatter.php b/includes/parser/DateFormatter.php index 602bcff3..cf510171 100644 --- a/includes/parser/DateFormatter.php +++ b/includes/parser/DateFormatter.php @@ -1,4 +1,9 @@ <?php +/** + * Date formatter + * + * @file + */ /** * Date formatter, recognises dates in plain text and formats them accoding to user preferences. @@ -29,7 +34,7 @@ class DateFormatter /** * @todo document */ - function DateFormatter() { + function __construct() { global $wgContLang; $this->monthNames = $this->getMonthRegex(); @@ -116,6 +121,7 @@ class DateFormatter /** * @param $preference String: User preference * @param $text String: Text to reformat + * @param $options Array: can contain 'linked' and/or 'match-whole' */ function reformat( $preference, $text, $options = array('linked') ) { @@ -265,7 +271,7 @@ class DateFormatter $isoBits[] = $bits['y']; $isoBits[] = $bits['m']; $isoBits[] = $bits['d']; - $isoDate = implode( '-', $isoBits );; + $isoDate = implode( '-', $isoBits ); // Output is not strictly HTML (it's wikitext), but <span> is whitelisted. $text = Html::rawElement( 'span', diff --git a/includes/parser/LinkHolderArray.php b/includes/parser/LinkHolderArray.php index 4f382a4f..19313b80 100644 --- a/includes/parser/LinkHolderArray.php +++ b/includes/parser/LinkHolderArray.php @@ -1,5 +1,13 @@ <?php - +/** + * Holder of replacement pairs for wiki links + * + * @file + */ + +/** + * @ingroup Parser + */ class LinkHolderArray { var $internals = array(), $interwikis = array(); var $size = 0; @@ -99,7 +107,7 @@ class LinkHolderArray { function getStubThreshold() { global $wgUser; if ( !isset( $this->stubThreshold ) ) { - $this->stubThreshold = $wgUser->getOption('stubthreshold'); + $this->stubThreshold = $wgUser->getStubThreshold(); } return $this->stubThreshold; } @@ -132,7 +140,7 @@ class LinkHolderArray { global $wgContLang; $colours = array(); - $sk = $this->parent->getOptions()->getSkin(); + $sk = $this->parent->getOptions()->getSkin( $this->parent->mTitle ); $linkCache = LinkCache::singleton(); $output = $this->parent->getOutput(); @@ -144,12 +152,13 @@ class LinkHolderArray { # Sort by namespace ksort( $this->internals ); + $linkcolour_ids = array(); + # Generate query $query = false; $current = null; foreach ( $this->internals as $ns => $entries ) { - foreach ( $entries as $index => $entry ) { - $key = "$ns:$index"; + foreach ( $entries as $entry ) { $title = $entry['title']; $pdbk = $entry['pdbk']; @@ -162,16 +171,19 @@ class LinkHolderArray { # Check if it's a static known link, e.g. interwiki if ( $title->isAlwaysKnown() ) { $colours[$pdbk] = ''; + } elseif ( $ns == NS_SPECIAL ) { + $colours[$pdbk] = 'new'; } elseif ( ( $id = $linkCache->getGoodLinkID( $pdbk ) ) != 0 ) { $colours[$pdbk] = $sk->getLinkColour( $title, $threshold ); $output->addLink( $title, $id ); + $linkcolour_ids[$id] = $pdbk; } elseif ( $linkCache->isBadLink( $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, page_len"; + $query = "SELECT page_id, page_namespace, page_title, page_is_redirect, page_len, page_latest"; $query .= " FROM $page WHERE (page_namespace=$ns AND page_title IN("; } elseif ( $current != $ns ) { $current = $ns; @@ -191,11 +203,10 @@ class LinkHolderArray { # Fetch data and form into an associative array # non-existent = broken - $linkcolour_ids = array(); - while ( $s = $dbr->fetchObject($res) ) { + foreach ( $res as $s ) { $title = Title::makeTitle( $s->page_namespace, $s->page_title ); $pdbk = $title->getPrefixedDBkey(); - $linkCache->addGoodLinkObj( $s->page_id, $title, $s->page_len, $s->page_is_redirect ); + $linkCache->addGoodLinkObj( $s->page_id, $title, $s->page_len, $s->page_is_redirect, $s->page_latest ); $output->addLink( $title, $s->page_id ); # FIXME: convoluted data flow # The redirect status and length is passed to getLinkColour via the LinkCache @@ -205,6 +216,8 @@ class LinkHolderArray { $linkcolour_ids[$s->page_id] = $pdbk; } unset( $res ); + } + if ( count($linkcolour_ids) ) { //pass an array of page_ids to an extension wfRunHooks( 'GetLinkColours', array( $linkcolour_ids, &$colours ) ); } @@ -265,10 +278,12 @@ class LinkHolderArray { wfProfileIn( __METHOD__ ); # Make interwiki link HTML - $sk = $this->parent->getOptions()->getSkin(); + $sk = $this->parent->getOptions()->getSkin( $this->parent->mTitle ); + $output = $this->parent->getOutput(); $replacePairs = array(); foreach( $this->interwikis as $key => $link ) { $replacePairs[$key] = $sk->link( $link['title'], $link['text'] ); + $output->addInterwikiLink( $link['title'] ); } $replacer = new HashtableReplacer( $replacePairs, 1 ); @@ -288,30 +303,52 @@ class LinkHolderArray { $variantMap = array(); // maps $pdbkey_Variant => $keys (of link holders) $output = $this->parent->getOutput(); $linkCache = LinkCache::singleton(); - $sk = $this->parent->getOptions()->getSkin(); + $sk = $this->parent->getOptions()->getSkin( $this->parent->mTitle ); $threshold = $this->getStubThreshold(); - - // Add variants of links to link batch + $titlesToBeConverted = ''; + $titlesAttrs = array(); + + // Concatenate titles to a single string, thus we only need auto convert the + // single string to all variants. This would improve parser's performance + // significantly. foreach ( $this->internals as $ns => $entries ) { foreach ( $entries as $index => $entry ) { - $key = "$ns:$index"; $pdbk = $entry['pdbk']; - $title = $entry['title']; - $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; - } + // we only deal with new links (in its first query) + if ( !isset( $colours[$pdbk] ) ) { + $title = $entry['title']; + $titleText = $title->getText(); + $titlesAttrs[] = array( + 'ns' => $ns, + 'key' => "$ns:$index", + 'titleText' => $titleText, + ); + // separate titles with \0 because it would never appears + // in a valid title + $titlesToBeConverted .= $titleText . "\0"; + } + } + } + + // Now do the conversion and explode string to text of titles + $titlesAllVariants = $wgContLang->autoConvertToAllVariants( $titlesToBeConverted ); + $allVariantsName = array_keys( $titlesAllVariants ); + foreach ( $titlesAllVariants as &$titlesVariant ) { + $titlesVariant = explode( "\0", $titlesVariant ); + } + $l = count( $titlesAttrs ); + // Then add variants of links to link batch + for ( $i = 0; $i < $l; $i ++ ) { + foreach ( $allVariantsName as $variantName ) { + $textVariant = $titlesAllVariants[$variantName][$i]; + extract( $titlesAttrs[$i] ); + if($textVariant != $titleText){ + $variantTitle = Title::makeTitle( $ns, $textVariant ); + if( is_null( $variantTitle ) ) { + continue; } + $linkBatch->addObj( $variantTitle ); + $variantMap[$variantTitle->getPrefixedDBkey()][] = $titlesAttrs[$i]['key']; } } } @@ -320,7 +357,7 @@ class LinkHolderArray { $categoryMap = array(); // maps $category_variant => $category (dbkeys) $varCategories = array(); // category replacements oldDBkey => newDBkey foreach( $output->getCategoryLinks() as $category ){ - $variants = $wgContLang->convertLinkToAllVariants($category); + $variants = $wgContLang->autoConvertToAllVariants( $category ); foreach($variants as $variant){ if($variant != $category){ $variantTitle = Title::newFromDBkey( Title::makeName(NS_CATEGORY,$variant) ); @@ -343,7 +380,7 @@ class LinkHolderArray { $linkcolour_ids = array(); // for each found variants, figure out link holders and replace - while ( $s = $dbr->fetchObject($varRes) ) { + foreach ( $varRes as $s ) { $variantTitle = Title::makeTitle( $s->page_namespace, $s->page_title ); $varPdbk = $variantTitle->getPrefixedDBkey(); @@ -403,8 +440,9 @@ class LinkHolderArray { /** * Replace <!--LINK--> link placeholders with plain text of links * (not HTML-formatted). - * @param string $text - * @return string + * + * @param $text String + * @return String */ function replaceText( $text ) { wfProfileIn( __METHOD__ ); @@ -419,7 +457,9 @@ class LinkHolderArray { } /** - * @param array $matches + * Callback for replaceText() + * + * @param $matches Array * @return string * @private */ diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php index 4f672f5b..4a3aa03b 100644 --- a/includes/parser/Parser.php +++ b/includes/parser/Parser.php @@ -17,15 +17,17 @@ * <pre> * There are five main entry points into the Parser class: * parse() - * produces HTML output + * produces HTML output * preSaveTransform(). - * produces altered wiki markup. + * produces altered wiki markup. * preprocess() - * removes HTML comments and expands templates - * cleanSig() - * Cleans a signature before saving it to preferences + * removes HTML comments and expands templates + * cleanSig() / cleanSigInSig() + * Cleans a signature before saving it to preferences * extractSections() - * Extracts sections from an article for section editing + * Extracts sections from an article for section editing + * getPreloadText() + * Removes <noinclude> sections, and <includeonly> tags. * * Globals used: * objects: $wgLang, $wgContLang @@ -43,8 +45,7 @@ * * @ingroup Parser */ -class Parser -{ +class Parser { /** * Update this version number when the ParserOutput format * changes in an incompatible way, so the parser cache @@ -63,7 +64,7 @@ class Parser const EXT_IMAGE_REGEX = '/^(http:\/\/|https:\/\/)([^][<>"\\x00-\\x20\\x7F]+) \\/([A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]+)\\.((?i)gif|png|jpg|jpeg)$/Sx'; - // State constants for the definition list colon extraction + # State constants for the definition list colon extraction const COLON_STATE_TEXT = 0; const COLON_STATE_TAG = 1; const COLON_STATE_TAGSTART = 2; @@ -73,27 +74,25 @@ class Parser const COLON_STATE_COMMENTDASH = 6; const COLON_STATE_COMMENTDASHDASH = 7; - // Flags for preprocessToDom + # Flags for preprocessToDom const PTD_FOR_INCLUSION = 1; - // Allowed values for $this->mOutputType - // Parameter to startExternalParse(). - const OT_HTML = 1; - const OT_WIKI = 2; - const OT_PREPROCESS = 3; + # Allowed values for $this->mOutputType + # Parameter to startExternalParse(). + const OT_HTML = 1; # like parse() + const OT_WIKI = 2; # like preSaveTransform() + const OT_PREPROCESS = 3; # like preprocess() const OT_MSG = 3; + const OT_PLAIN = 4; # like extractSections() - portions of the original are returned unchanged. - // Marker Suffix needs to be accessible staticly. + # Marker Suffix needs to be accessible staticly. const MARKER_SUFFIX = "-QINU\x7f"; - /**#@+ - * @private - */ # Persistent: - var $mTagHooks, $mTransparentTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables, - $mSubstWords, $mImageParams, $mImageParamsMagicArray, $mStripList, $mMarkerIndex, - $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols, $mDefaultStripList, - $mVarCache, $mConf, $mFunctionTagHooks; + var $mTagHooks, $mTransparentTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables; + var $mSubstWords, $mImageParams, $mImageParamsMagicArray, $mStripList, $mMarkerIndex; + var $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols, $mDefaultStripList; + var $mVarCache, $mConf, $mFunctionTagHooks; # Cleared with clearState(): @@ -101,21 +100,19 @@ class Parser var $mIncludeCount, $mArgStack, $mLastSection, $mInPre; var $mLinkHolders, $mLinkID; var $mIncludeSizes, $mPPNodeCount, $mDefaultSort; - var $mTplExpandCache; // empty-frame expansion cache + var $mTplExpandCache; # empty-frame expansion cache var $mTplRedirCache, $mTplDomCache, $mHeadings, $mDoubleUnderscores; - var $mExpensiveFunctionCount; // number of expensive parser function calls + var $mExpensiveFunctionCount; # number of expensive parser function calls # 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 - - /**#@-*/ + var $mOptions; # ParserOptions object + var $mTitle; # Title context, used for self-link rendering and similar things + var $mOutputType; # Output type, one of the OT_xxx constants + var $ot; # Shortcut alias, see setOutputType() + var $mRevisionId; # ID to display in {{REVISIONID}} tags + var $mRevisionTimestamp; # The timestamp of the specified revision ID + var $mRevIdForTs; # The revision ID which was used to fetch the timestamp /** * Constructor @@ -132,12 +129,12 @@ class Parser $this->mDefaultStripList = $this->mStripList = array(); $this->mUrlProtocols = wfUrlProtocols(); $this->mExtLinkBracketedRegex = '/\[(\b(' . wfUrlProtocols() . ')'. - '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x0a\\x0d]*?)\]/S'; + '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x00-\\x08\\x0a-\\x1F]*?)\]/S'; $this->mVarCache = array(); if ( isset( $conf['preprocessorClass'] ) ) { $this->mPreprocessorClass = $conf['preprocessorClass']; } elseif ( extension_loaded( 'domxml' ) ) { - // PECL extension that conflicts with the core DOM extension (bug 13770) + # PECL extension that conflicts with the core DOM extension (bug 13770) wfDebug( "Warning: you have the obsolete domxml extension for PHP. Please remove it!\n" ); $this->mPreprocessorClass = 'Preprocessor_Hash'; } elseif ( extension_loaded( 'dom' ) ) { @@ -191,6 +188,7 @@ class Parser $this->firstCallInit(); } $this->mOutput = new ParserOutput; + $this->mOptions->registerWatcher( array( $this->mOutput, 'recordOption' ) ); $this->mAutonumber = 0; $this->mLastSection = ''; $this->mDTopen = false; @@ -213,7 +211,7 @@ class Parser * Must not consist of all title characters, or else it will change * the behaviour of <nowiki> in a link. */ - #$this->mUniqPrefix = "\x07UNIQ" . Parser::getRandomString(); + # $this->mUniqPrefix = "\x07UNIQ" . Parser::getRandomString(); # Changed to \x7f to allow XML double-parsing -- TS $this->mUniqPrefix = "\x7fUNIQ" . self::getRandomString(); @@ -242,51 +240,6 @@ class Parser wfProfileOut( __METHOD__ ); } - function setOutputType( $ot ) { - $this->mOutputType = $ot; - // Shortcut alias - $this->ot = array( - 'html' => $ot == self::OT_HTML, - 'wiki' => $ot == self::OT_WIKI, - 'pre' => $ot == self::OT_PREPROCESS, - ); - } - - /** - * Set the context title - */ - function setTitle( $t ) { - if ( !$t || $t instanceof FakeTitle ) { - $t = Title::newFromText( 'NO TITLE' ); - } - - if ( strval( $t->getFragment() ) !== '' ) { - # Strip the fragment to avoid various odd effects - $this->mTitle = clone $t; - $this->mTitle->setFragment( '' ); - } else { - $this->mTitle = $t; - } - } - - /** - * Accessor for mUniqPrefix. - * - * @public - */ - function uniqPrefix() { - if( !isset( $this->mUniqPrefix ) ) { - // @todo Fixme: this is probably *horribly wrong* - // LanguageConverter seems to want $wgParser's uniqPrefix, however - // if this is called for a parser cache hit, the parser may not - // have ever been initialized in the first place. - // Not really sure what the heck is supposed to be going on here. - return ''; - //throw new MWException( "Accessing uninitialized mUniqPrefix" ); - } - return $this->mUniqPrefix; - } - /** * Convert wikitext to HTML * Do not call this function recursively. @@ -310,16 +263,16 @@ class Parser wfProfileIn( __METHOD__ ); wfProfileIn( $fname ); + $this->mOptions = $options; if ( $clearState ) { $this->clearState(); } - $this->mOptions = $options; - $this->setTitle( $title ); // Page title has to be set for the pre-processor + $this->setTitle( $title ); # Page title has to be set for the pre-processor $oldRevisionId = $this->mRevisionId; $oldRevisionTimestamp = $this->mRevisionTimestamp; - if( $revid !== null ) { + if ( $revid !== null ) { $this->mRevisionId = $revid; $this->mRevisionTimestamp = null; } @@ -335,12 +288,12 @@ class Parser $fixtags = array( # french spaces, last one Guillemet-left # only if there is something before the space - '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1 \\2', + '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1 \\2', # french spaces, Guillemet-right - '/(\\302\\253) /' => '\\1 ', - '/ (!\s*important)/' => ' \\1', #Beware of CSS magic word !important, bug #11874. + '/(\\302\\253) /' => '\\1 ', + '/ (!\s*important)/' => ' \\1', # Beware of CSS magic word !important, bug #11874. ); - $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text ); + $text = preg_replace( array_keys( $fixtags ), array_values( $fixtags ), $text ); $text = $this->doBlockLevels( $text, $linestart ); @@ -364,18 +317,11 @@ class Parser } /** - * A page get its title converted except: - * a) Language conversion is globally disabled - * b) Title convert is globally disabled - * c) The page is a redirect page - * d) User request with a "linkconvert" set to "no" - * e) A "nocontentconvert" magic word has been set - * f) A "notitleconvert" magic word has been set - * g) User sets "noconvertlink" in his/her preference - * - * Note that if a user tries to set a title in a conversion - * rule but content conversion was not done, then the parser - * won't pick it up. This is probably expected behavior. + * A converted title will be provided in the output object if title and + * content conversion are enabled, the article text does not contain + * a conversion-suppressing double-underscore tag, and no + * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over + * automatic link conversion. */ if ( !( $wgDisableLangConversion || $wgDisableTitleConversion @@ -401,14 +347,13 @@ class Parser $uniq_prefix = $this->mUniqPrefix; $matches = array(); $elements = array_keys( $this->mTransparentTagHooks ); - $text = self::extractTagsAndParams( $elements, $text, $matches, $uniq_prefix ); + $text = $this->extractTagsAndParams( $elements, $text, $matches, $uniq_prefix ); - foreach( $matches as $marker => $data ) { + 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 ) ); + if ( isset( $this->mTransparentTagHooks[$tagName] ) ) { + $output = call_user_func_array( $this->mTransparentTagHooks[$tagName], array( $content, $params, $this ) ); } else { $output = $tag; } @@ -418,7 +363,7 @@ class Parser $text = Sanitizer::normalizeCharReferences( $text ); - if ( ( $wgUseTidy && $this->mOptions->mTidy ) || $wgAlwaysUseTidy ) { + if ( ( $wgUseTidy && $this->mOptions->getTidy() ) || $wgAlwaysUseTidy ) { $text = MWTidy::tidy( $text ); } else { # attempt to sanitize at least some nesting problems @@ -460,7 +405,7 @@ class Parser $PFreport = "Expensive parser function count: {$this->mExpensiveFunctionCount}/$wgExpensiveParserFunctionLimit\n"; $limitReport = "NewPP limit report\n" . - "Preprocessor node count: {$this->mPPNodeCount}/{$this->mOptions->mMaxPPNodeCount}\n" . + "Preprocessor node count: {$this->mPPNodeCount}/{$this->mOptions->getMaxPPNodeCount()}\n" . "Post-expand include size: {$this->mIncludeSizes['post-expand']}/$max bytes\n" . "Template argument size: {$this->mIncludeSizes['arg']}/$max bytes\n". $PFreport; @@ -484,7 +429,7 @@ class Parser * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded * * @param $text String: text extension wants to have parsed - * @param PPFrame $frame: The frame to use for expanding any template variables + * @param $frame PPFrame: The frame to use for expanding any template variables */ function recursiveTagParse( $text, $frame=false ) { wfProfileIn( __METHOD__ ); @@ -499,13 +444,13 @@ class Parser * Expand templates and variables in the text, producing valid, static wikitext. * Also removes comments. */ - function preprocess( $text, $title, $options, $revid = null ) { + function preprocess( $text, Title $title, ParserOptions $options, $revid = null ) { wfProfileIn( __METHOD__ ); + $this->mOptions = $options; $this->clearState(); $this->setOutputType( self::OT_PREPROCESS ); - $this->mOptions = $options; $this->setTitle( $title ); - if( $revid !== null ) { + if ( $revid !== null ) { $this->mRevisionId = $revid; } wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) ); @@ -517,20 +462,146 @@ class Parser } /** + * Process the wikitext for the ?preload= feature. (bug 5210) + * + * <noinclude>, <includeonly> etc. are parsed as for template transclusion, + * comments, templates, arguments, tags hooks and parser functions are untouched. + */ + public function getPreloadText( $text, Title $title, ParserOptions $options ) { + # Parser (re)initialisation + $this->mOptions = $options; + $this->clearState(); + $this->setOutputType( self::OT_PLAIN ); + $this->setTitle( $title ); + + $flags = PPFrame::NO_ARGS | PPFrame::NO_TEMPLATES; + $dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION ); + $text = $this->getPreprocessor()->newFrame()->expand( $dom, $flags ); + $text = $this->mStripState->unstripBoth( $text ); + return $text; + } + + /** * Get a random string * * @private * @static */ - function getRandomString() { - return dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff)); + static private function getRandomString() { + return dechex( mt_rand( 0, 0x7fffffff ) ) . dechex( mt_rand( 0, 0x7fffffff ) ); + } + + /** + * Accessor for mUniqPrefix. + * + * @return String + */ + public function uniqPrefix() { + if ( !isset( $this->mUniqPrefix ) ) { + # @todo Fixme: this is probably *horribly wrong* + # LanguageConverter seems to want $wgParser's uniqPrefix, however + # if this is called for a parser cache hit, the parser may not + # have ever been initialized in the first place. + # Not really sure what the heck is supposed to be going on here. + return ''; + # throw new MWException( "Accessing uninitialized mUniqPrefix" ); + } + return $this->mUniqPrefix; + } + + /** + * Set the context title + */ + function setTitle( $t ) { + if ( !$t || $t instanceof FakeTitle ) { + $t = Title::newFromText( 'NO TITLE' ); + } + + if ( strval( $t->getFragment() ) !== '' ) { + # Strip the fragment to avoid various odd effects + $this->mTitle = clone $t; + $this->mTitle->setFragment( '' ); + } else { + $this->mTitle = $t; + } + } + + /** + * Accessor for the Title object + * + * @return Title object + */ + function getTitle() { + return $this->mTitle; + } + + /** + * Accessor/mutator for the Title object + * + * @param $x New Title object or null to just get the current one + * @return Title object + */ + function Title( $x = null ) { + return wfSetVar( $this->mTitle, $x ); + } + + /** + * Set the output type + * + * @param $ot Integer: new value + */ + function setOutputType( $ot ) { + $this->mOutputType = $ot; + # Shortcut alias + $this->ot = array( + 'html' => $ot == self::OT_HTML, + 'wiki' => $ot == self::OT_WIKI, + 'pre' => $ot == self::OT_PREPROCESS, + 'plain' => $ot == self::OT_PLAIN, + ); } - function &getTitle() { return $this->mTitle; } - function getOptions() { return $this->mOptions; } - function getRevisionId() { return $this->mRevisionId; } - function getOutput() { return $this->mOutput; } - function nextLinkID() { return $this->mLinkID++; } + /** + * Accessor/mutator for the output type + * + * @param $x New value or null to just get the current one + * @return Integer + */ + function OutputType( $x = null ) { + return wfSetVar( $this->mOutputType, $x ); + } + + /** + * Get the ParserOutput object + * + * @return ParserOutput object + */ + function getOutput() { + return $this->mOutput; + } + + /** + * Get the ParserOptions object + * + * @return ParserOptions object + */ + function getOptions() { + return $this->mOptions; + } + + /** + * Accessor/mutator for the ParserOptions object + * + * @param $x New value or null to just get the current one + * @return Current ParserOptions object + */ + function Options( $x = null ) { + return wfSetVar( $this->mOptions, $x ); + } + + function nextLinkID() { + return $this->mLinkID++; + } function getFunctionLang() { global $wgLang, $wgContLang; @@ -545,6 +616,8 @@ class Parser /** * Get a preprocessor object + * + * @return Preprocessor instance */ function getPreprocessor() { if ( !isset( $this->mPreprocessor ) ) { @@ -567,12 +640,13 @@ class Parser * * @param $elements list of element names. Comments are always extracted. * @param $text Source text string. + * @param $matches Out parameter, Array: extracted tags * @param $uniq_prefix + * @return String: stripped text * - * @public * @static */ - function extractTagsAndParams($elements, $text, &$matches, $uniq_prefix = ''){ + public function extractTagsAndParams( $elements, $text, &$matches, $uniq_prefix = '' ) { static $n = 1; $stripped = ''; $matches = array(); @@ -583,40 +657,40 @@ class Parser while ( $text != '' ) { $p = preg_split( $start, $text, 2, PREG_SPLIT_DELIM_CAPTURE ); $stripped .= $p[0]; - if( count( $p ) < 5 ) { + if ( count( $p ) < 5 ) { break; } - if( count( $p ) > 5 ) { - // comment + if ( count( $p ) > 5 ) { + # comment $element = $p[4]; $attributes = ''; $close = ''; $inside = $p[5]; } else { - // tag + # tag $element = $p[1]; $attributes = $p[2]; $close = $p[3]; $inside = $p[4]; } - $marker = "$uniq_prefix-$element-" . sprintf('%08X', $n++) . self::MARKER_SUFFIX; + $marker = "$uniq_prefix-$element-" . sprintf( '%08X', $n++ ) . self::MARKER_SUFFIX; $stripped .= $marker; if ( $close === '/>' ) { - // Empty element tag, <tag /> + # Empty element tag, <tag /> $content = null; $text = $inside; $tail = null; } else { - if( $element === '!--' ) { + 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 ) { + if ( count( $q ) < 3 ) { # No end tag -- let it run out to the end of the text. $tail = ''; $text = ''; @@ -644,7 +718,7 @@ class Parser /** * @deprecated use replaceVariables */ - function strip( $text, $state, $stripcomments = false , $dontstrip = array () ) { + function strip( $text, $state, $stripcomments = false , $dontstrip = array() ) { return $text; } @@ -704,34 +778,35 @@ class Parser * * @private */ - function doTableStuff ( $text ) { + function doTableStuff( $text ) { wfProfileIn( __METHOD__ ); - + $lines = StringUtils::explode( "\n", $text ); $out = ''; - $td_history = array (); // Is currently a td tag open? - $last_tag_history = array (); // Save history of last lag activated (td, th or caption) - $tr_history = array (); // Is currently a tr tag open? - $tr_attributes = array (); // history of tr attributes - $has_opened_tr = array(); // Did this table open a <tr> element? - $indent_level = 0; // indent level of the table + $td_history = array(); # Is currently a td tag open? + $last_tag_history = array(); # Save history of last lag activated (td, th or caption) + $tr_history = array(); # Is currently a tr tag open? + $tr_attributes = array(); # history of tr attributes + $has_opened_tr = array(); # Did this table open a <tr> element? + $indent_level = 0; # indent level of the table foreach ( $lines as $outLine ) { $line = trim( $outLine ); - if( $line == '' ) { // empty line, go to next line + if ( $line === '' ) { # empty line, go to next line $out .= $outLine."\n"; continue; } + $first_character = $line[0]; $matches = array(); if ( preg_match( '/^(:*)\{\|(.*)$/', $line , $matches ) ) { - // First check if we are starting a new table + # 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' ); + $attributes = Sanitizer::fixTagAttributes( $attributes , 'table' ); $outLine = str_repeat( '<dl><dd>' , $indent_level ) . "<table{$attributes}>"; array_push( $td_history , false ); @@ -739,156 +814,152 @@ class Parser 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 + } elseif ( count( $td_history ) == 0 ) { + # Don't do any of the following $out .= $outLine."\n"; continue; - } else if ( substr ( $line , 0 , 2 ) === '|}' ) { - // We are ending a table - $line = '</table>' . substr ( $line , 2 ); - $last_tag = array_pop ( $last_tag_history ); + } elseif ( substr( $line , 0 , 2 ) === '|}' ) { + # We are ending a table + $line = '</table>' . substr( $line , 2 ); + $last_tag = array_pop( $last_tag_history ); - if ( !array_pop ( $has_opened_tr ) ) { + if ( !array_pop( $has_opened_tr ) ) { $line = "<tr><td></td></tr>{$line}"; } - if ( array_pop ( $tr_history ) ) { + if ( array_pop( $tr_history ) ) { $line = "</tr>{$line}"; } - if ( array_pop ( $td_history ) ) { + if ( array_pop( $td_history ) ) { $line = "</{$last_tag}>{$line}"; } - array_pop ( $tr_attributes ); + array_pop( $tr_attributes ); $outLine = $line . str_repeat( '</dd></dl>' , $indent_level ); - } else if ( substr ( $line , 0 , 2 ) === '|-' ) { - // Now we have a table row + } elseif ( substr( $line , 0 , 2 ) === '|-' ) { + # Now we have a table row $line = preg_replace( '#^\|-+#', '', $line ); - // Whats after the tag is now only attributes + # 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 ); + $last_tag = array_pop( $last_tag_history ); + array_pop( $has_opened_tr ); + array_push( $has_opened_tr , true ); - if ( array_pop ( $tr_history ) ) { + if ( array_pop( $tr_history ) ) { $line = '</tr>'; } - if ( array_pop ( $td_history ) ) { + if ( array_pop( $td_history ) ) { $line = "</{$last_tag}>{$line}"; } $outLine = $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 ) === '|+' ) { + array_push( $tr_history , false ); + array_push( $td_history , false ); + array_push( $last_tag_history , '' ); + } elseif ( $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 ); } - $line = substr ( $line , 1 ); + $line = substr( $line , 1 ); if ( $first_character === '!' ) { - $line = str_replace ( '!!' , '||' , $line ); + $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 "||". + # 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 ); $outLine = ''; - // Loop through each table cell - foreach ( $cells as $cell ) - { + # Loop through each table cell + foreach ( $cells as $cell ) { $previous = ''; - if ( $first_character !== '+' ) - { - $tr_after = array_pop ( $tr_attributes ); - if ( !array_pop ( $tr_history ) ) { + if ( $first_character !== '+' ) { + $tr_after = array_pop( $tr_attributes ); + if ( !array_pop( $tr_history ) ) { $previous = "<tr{$tr_after}>\n"; } - array_push ( $tr_history , true ); - array_push ( $tr_attributes , '' ); - array_pop ( $has_opened_tr ); - array_push ( $has_opened_tr , true ); + array_push( $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 ); + $last_tag = array_pop( $last_tag_history ); - if ( array_pop ( $td_history ) ) { - $previous = "</{$last_tag}>{$previous}"; + if ( array_pop( $td_history ) ) { + $previous = "</{$last_tag}>\n{$previous}"; } if ( $first_character === '|' ) { $last_tag = 'td'; - } else if ( $first_character === '!' ) { + } elseif ( $first_character === '!' ) { $last_tag = 'th'; - } else if ( $first_character === '+' ) { + } elseif ( $first_character === '+' ) { $last_tag = 'caption'; } else { $last_tag = ''; } - array_push ( $last_tag_history , $last_tag ); + array_push( $last_tag_history , $last_tag ); - // A cell could contain both parameters and data - $cell_data = explode ( '|' , $cell , 2 ); + # 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 + # 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 ) + } elseif ( count( $cell_data ) == 1 ) { $cell = "{$previous}<{$last_tag}>{$cell_data[0]}"; - else { + } else { $attributes = $this->mStripState->unstripBoth( $cell_data[0] ); $attributes = Sanitizer::fixTagAttributes( $attributes , $last_tag ); $cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[1]}"; } $outLine .= $cell; - array_push ( $td_history , true ); + array_push( $td_history , true ); } } $out .= $outLine . "\n"; } - // Closing open td, tr && table - while ( count ( $td_history ) > 0 ) - { - if ( array_pop ( $td_history ) ) { + # Closing open td, tr && table + while ( count( $td_history ) > 0 ) { + if ( array_pop( $td_history ) ) { $out .= "</td>\n"; } - if ( array_pop ( $tr_history ) ) { + if ( array_pop( $tr_history ) ) { $out .= "</tr>\n"; } - if ( !array_pop ( $has_opened_tr ) ) { + if ( !array_pop( $has_opened_tr ) ) { $out .= "<tr><td></td></tr>\n" ; } $out .= "</table>\n"; } - // Remove trailing line-ending (b/c) + # Remove trailing line-ending (b/c) if ( substr( $out, -1 ) === "\n" ) { $out = substr( $out, 0, -1 ); } - // special case: don't return empty table - if( $out === "<table>\n<tr><td></td></tr>\n</table>" ) { + # special case: don't return empty table + if ( $out === "<table>\n<tr><td></td></tr>\n</table>" ) { $out = ''; } @@ -914,29 +985,29 @@ class Parser return $text ; } - // if $frame is provided, then use $frame for replacing any variables - if ($frame) { - // use frame depth to infer how include/noinclude tags should be handled - // depth=0 means this is the top-level document; otherwise it's an included document - if( !$frame->depth ) + # if $frame is provided, then use $frame for replacing any variables + if ( $frame ) { + # use frame depth to infer how include/noinclude tags should be handled + # depth=0 means this is the top-level document; otherwise it's an included document + if ( !$frame->depth ) { $flag = 0; - else + } else { $flag = Parser::PTD_FOR_INCLUSION; + } $dom = $this->preprocessToDom( $text, $flag ); $text = $frame->expand( $dom ); - } - // if $frame is not provided, then use old-style replaceVariables - else { + } else { + # if $frame is not provided, then use old-style replaceVariables $text = $this->replaceVariables( $text ); } $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), false, array_keys( $this->mTransparentTagHooks ) ); wfRunHooks( 'InternalParseBeforeLinks', array( &$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. + # 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<hr />', $text ); @@ -944,17 +1015,17 @@ class Parser $text = $this->doDoubleUnderscore( $text ); $text = $this->doHeadings( $text ); - if( $this->mOptions->getUseDynamicDates() ) { + if ( $this->mOptions->getUseDynamicDates() ) { $df = DateFormatter::getInstance(); $text = $df->reformat( $this->mOptions->getDateFormat(), $text ); } - $text = $this->doAllQuotes( $text ); $text = $this->replaceInternalLinks( $text ); + $text = $this->doAllQuotes( $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 = str_replace( $this->mUniqPrefix.'NOPARSE', '', $text ); $text = $this->doMagicLinks( $text ); $text = $this->formatHeadings( $text, $origText, $isMain ); @@ -976,7 +1047,7 @@ class Parser $urlChar = self::EXT_LINK_URL_CLASS; $text = preg_replace_callback( '!(?: # Start cases - (<a.*?</a>) | # m[1]: Skip link text + (<a[ \t\r\n>].*?</a>) | # m[1]: Skip link text (<.*?>) | # m[2]: Skip stuff inside HTML elements' . " (\\b(?:$prots)$urlChar+) | # m[3]: Free external links" . ' (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number @@ -1002,7 +1073,6 @@ class Parser return $this->makeFreeExternalLink( $m[0] ); } elseif ( isset( $m[4] ) && $m[4] !== '' ) { # RFC or PMID - $CssClass = ''; if ( substr( $m[0], 0, 3 ) === 'RFC' ) { $keyword = 'RFC'; $urlmsg = 'rfcurl'; @@ -1015,10 +1085,10 @@ class Parser $id = $m[4]; } else { throw new MWException( __METHOD__.': unrecognised match type "' . - substr($m[0], 0, 20 ) . '"' ); + substr( $m[0], 0, 20 ) . '"' ); } - $url = wfMsg( $urlmsg, $id); - $sk = $this->mOptions->getSkin(); + $url = wfMsgForContent( $urlmsg, $id); + $sk = $this->mOptions->getSkin( $this->mTitle ); $la = $sk->getExternalLinkAttributes( "external $CssClass" ); return "<a href=\"{$url}\"{$la}>{$keyword} {$id}</a>"; } elseif ( isset( $m[5] ) && $m[5] !== '' ) { @@ -1047,16 +1117,16 @@ class Parser global $wgContLang; wfProfileIn( __METHOD__ ); - $sk = $this->mOptions->getSkin(); + $sk = $this->mOptions->getSkin( $this->mTitle ); $trail = ''; # 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]); + 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 @@ -1118,7 +1188,7 @@ class Parser foreach ( $lines as $line ) { $outtext .= $this->doQuotes( $line ) . "\n"; } - $outtext = substr($outtext, 0,-1); + $outtext = substr( $outtext, 0,-1 ); wfProfileOut( __METHOD__ ); return $outtext; } @@ -1128,89 +1198,84 @@ class Parser */ public function doQuotes( $text ) { $arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE ); - if ( count( $arr ) == 1 ) + if ( count( $arr ) == 1 ) { return $text; - else - { + } 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 ) - { + for ( $i = 0; $i < count( $arr ); $i++ ) { + if ( ( $i % 2 ) == 1 ) { # If there are ever four apostrophes, assume the first is supposed to # be text, and the remaining three constitute mark-up for bold text. - if ( strlen( $arr[$i] ) == 4 ) - { + if ( strlen( $arr[$i] ) == 4 ) { $arr[$i-1] .= "'"; $arr[$i] = "'''"; - } - # If there are more than 5 apostrophes in a row, assume they're all - # text except for the last 5. - else if ( strlen( $arr[$i] ) > 5 ) - { + } elseif ( strlen( $arr[$i] ) > 5 ) { + # If there are more than 5 apostrophes in a row, assume they're all + # text except for the last 5. $arr[$i-1] .= str_repeat( "'", strlen( $arr[$i] ) - 5 ); $arr[$i] = "'''''"; } # Count the number of occurrences of bold and italics mark-ups. # We are not counting sequences of five apostrophes. - if ( strlen( $arr[$i] ) == 2 ) { $numitalics++; } - else if ( strlen( $arr[$i] ) == 3 ) { $numbold++; } - else if ( strlen( $arr[$i] ) == 5 ) { $numitalics++; $numbold++; } + if ( strlen( $arr[$i] ) == 2 ) { + $numitalics++; + } elseif ( strlen( $arr[$i] ) == 3 ) { + $numbold++; + } elseif ( 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 ) ) - { + 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; + 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; + } + } elseif ( $x2 === ' ') { + if ( $firstsingleletterword == -1 ) { + $firstsingleletterword = $i; + } } else { - if ($firstmultiletterword == -1) $firstmultiletterword = $i; + 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 ] .= "'"; + if ( $firstsingleletterword > -1 ) { + $arr[$firstsingleletterword] = "''"; + $arr[$firstsingleletterword-1] .= "'"; + } elseif ( $firstmultiletterword > -1 ) { + # If not, but there's a multi-letter word, use that one. + $arr[$firstmultiletterword] = "''"; + $arr[$firstmultiletterword-1] .= "'"; + } elseif ( $firstspace > -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) + $arr[$firstspace] = "''"; + $arr[$firstspace-1] .= "'"; } } @@ -1219,71 +1284,70 @@ class Parser $buffer = ''; $state = ''; $i = 0; - foreach ($arr as $r) - { - if (($i % 2) == 0) - { - if ($state === 'both') + foreach ( $arr as $r ) { + if ( ( $i % 2 ) == 0 ) { + if ( $state === 'both' ) { $buffer .= $r; - else + } else { $output .= $r; - } - else - { - if (strlen ($r) == 2) - { - if ($state === 'i') - { $output .= '</i>'; $state = ''; } - else if ($state === 'bi') - { $output .= '</i>'; $state = 'b'; } - else if ($state === 'ib') - { $output .= '</b></i><b>'; $state = 'b'; } - else if ($state === 'both') - { $output .= '<b><i>'.$buffer.'</i>'; $state = 'b'; } - else # $state can be 'b' or '' - { $output .= '<i>'; $state .= 'i'; } - } - else if (strlen ($r) == 3) - { - if ($state === 'b') - { $output .= '</b>'; $state = ''; } - else if ($state === 'bi') - { $output .= '</i></b><i>'; $state = 'i'; } - else if ($state === 'ib') - { $output .= '</b>'; $state = 'i'; } - else if ($state === 'both') - { $output .= '<i><b>'.$buffer.'</b>'; $state = 'i'; } - else # $state can be 'i' or '' - { $output .= '<b>'; $state .= 'b'; } } - else if (strlen ($r) == 5) - { - if ($state === 'b') - { $output .= '</b><i>'; $state = 'i'; } - else if ($state === 'i') - { $output .= '</i><b>'; $state = 'b'; } - else if ($state === 'bi') - { $output .= '</i></b>'; $state = ''; } - else if ($state === 'ib') - { $output .= '</b></i>'; $state = ''; } - else if ($state === 'both') - { $output .= '<i><b>'.$buffer.'</b></i>'; $state = ''; } - else # ($state == '') - { $buffer = ''; $state = 'both'; } + } else { + if ( strlen( $r ) == 2 ) { + if ( $state === 'i' ) { + $output .= '</i>'; $state = ''; + } elseif ( $state === 'bi' ) { + $output .= '</i>'; $state = 'b'; + } elseif ( $state === 'ib' ) { + $output .= '</b></i><b>'; $state = 'b'; + } elseif ( $state === 'both' ) { + $output .= '<b><i>'.$buffer.'</i>'; $state = 'b'; + } else { # $state can be 'b' or '' + $output .= '<i>'; $state .= 'i'; + } + } elseif ( strlen( $r ) == 3 ) { + if ( $state === 'b' ) { + $output .= '</b>'; $state = ''; + } elseif ( $state === 'bi' ) { + $output .= '</i></b><i>'; $state = 'i'; + } elseif ( $state === 'ib' ) { + $output .= '</b>'; $state = 'i'; + } elseif ( $state === 'both' ) { + $output .= '<i><b>'.$buffer.'</b>'; $state = 'i'; + } else { # $state can be 'i' or '' + $output .= '<b>'; $state .= 'b'; + } + } elseif ( strlen( $r ) == 5 ) { + if ( $state === 'b' ) { + $output .= '</b><i>'; $state = 'i'; + } elseif ( $state === 'i' ) { + $output .= '</i><b>'; $state = 'b'; + } elseif ( $state === 'bi' ) { + $output .= '</i></b>'; $state = ''; + } elseif ( $state === 'ib' ) { + $output .= '</b></i>'; $state = ''; + } elseif ( $state === 'both' ) { + $output .= '<i><b>'.$buffer.'</b></i>'; $state = ''; + } else { # ($state == '') + $buffer = ''; $state = 'both'; + } } } $i++; } # Now close all remaining tags. Notice that the order is important. - if ($state === 'b' || $state === 'ib') + if ( $state === 'b' || $state === 'ib' ) { $output .= '</b>'; - if ($state === 'i' || $state === 'bi' || $state === 'ib') + } + if ( $state === 'i' || $state === 'bi' || $state === 'ib' ) { $output .= '</i>'; - if ($state === 'bi') + } + if ( $state === 'bi' ) { $output .= '</b>'; + } # There might be lonely ''''', so make sure we have a buffer - if ($state === 'both' && $buffer) + if ( $state === 'both' && $buffer ) { $output .= '<b><i>'.$buffer.'</i></b>'; + } return $output; } } @@ -1300,7 +1364,7 @@ class Parser global $wgContLang; wfProfileIn( __METHOD__ ); - $sk = $this->mOptions->getSkin(); + $sk = $this->mOptions->getSkin( $this->mTitle ); $bits = preg_split( $this->mExtLinkBracketedRegex, $text, -1, PREG_SPLIT_DELIM_CAPTURE ); $s = array_shift( $bits ); @@ -1316,9 +1380,9 @@ class Parser # 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 ( preg_match( '/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE ) ) { + $text = substr( $url, $m2[0][1] ) . ' ' . $text; + $url = substr( $url, 0, $m2[0][1] ); } # If the link text is an image URL, replace it with an <img> tag @@ -1331,12 +1395,12 @@ class Parser $dtrail = ''; # Set linktype for CSS - if URL==text, link is essentially free - $linktype = ($text === $url) ? 'free' : 'text'; + $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 ) { + if ( strpos( wfUrlProtocols(), substr( $protocol, 0, strpos( $protocol, ':' ) ) ) !== false ) { $langObj = $this->getFunctionLang(); $text = '[' . $langObj->formatNum( ++$this->mAutonumber ) . ']'; $linktype = 'autonumber'; @@ -1351,13 +1415,13 @@ class Parser list( $dtrail, $trail ) = Linker::splitTrail( $trail ); } - $text = $wgContLang->markNoConversion($text); + $text = $wgContLang->markNoConversion( $text ); $url = Sanitizer::cleanUrl( $url ); # 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 + # Funny characters like ö aren't valid in URLs anyway # This was changed in August 2004 $s .= $sk->makeExternalLink( $url, $text, false, $linktype, $this->getExternalLinkAttribs( $url ) ) . $dtrail . $trail; @@ -1379,15 +1443,15 @@ class Parser * (depending on configuration, namespace, and the URL's domain) and/or a * target attribute (depending on configuration). * - * @param string $url Optional URL, to extract the domain from for rel => + * @param $url String: optional URL, to extract the domain from for rel => * nofollow if appropriate - * @return array Associative array of HTML attributes + * @return Array: associative array of HTML attributes */ function getExternalLinkAttribs( $url = false ) { $attribs = array(); global $wgNoFollowLinks, $wgNoFollowNsExceptions; $ns = $this->mTitle->getNamespace(); - if( $wgNoFollowLinks && !in_array($ns, $wgNoFollowNsExceptions) ) { + if ( $wgNoFollowLinks && !in_array( $ns, $wgNoFollowNsExceptions ) ) { $attribs['rel'] = 'nofollow'; global $wgNoFollowDomainExceptions; @@ -1395,8 +1459,7 @@ class Parser $bits = wfParseUrl( $url ); if ( is_array( $bits ) && isset( $bits['host'] ) ) { foreach ( $wgNoFollowDomainExceptions as $domain ) { - if( substr( $bits['host'], -strlen( $domain ) ) - == $domain ) { + if ( substr( $bits['host'], -strlen( $domain ) ) == $domain ) { unset( $attribs['rel'] ); break; } @@ -1413,9 +1476,10 @@ class Parser /** * Replace unusual URL escape codes with their equivalent characters - * @param string - * @return string - * @static + * + * @param $url String + * @return String + * * @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 @@ -1429,18 +1493,16 @@ class Parser /** * 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? + # 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 + # No, shouldn't be escaped return $char; } else { - // Yes, leave it escaped + # Yes, leave it escaped return $matches[0]; } } @@ -1451,21 +1513,21 @@ class Parser * @private */ function maybeMakeExternalImage( $url ) { - $sk = $this->mOptions->getSkin(); + $sk = $this->mOptions->getSkin( $this->mTitle ); $imagesfrom = $this->mOptions->getAllowExternalImagesFrom(); - $imagesexception = !empty($imagesfrom); + $imagesexception = !empty( $imagesfrom ); $text = false; # $imagesfrom could be either a single string or an array of strings, parse out the latter - if( $imagesexception && is_array( $imagesfrom ) ) { + if ( $imagesexception && is_array( $imagesfrom ) ) { $imagematch = false; - foreach( $imagesfrom as $match ) { - if( strpos( $url, $match ) === 0 ) { + foreach ( $imagesfrom as $match ) { + if ( strpos( $url, $match ) === 0 ) { $imagematch = true; break; } } - } elseif( $imagesexception ) { - $imagematch = (strpos( $url, $imagesfrom ) === 0); + } elseif ( $imagesexception ) { + $imagematch = ( strpos( $url, $imagesfrom ) === 0 ); } else { $imagematch = false; } @@ -1476,14 +1538,15 @@ class Parser $text = $sk->makeExternalImage( $url ); } } - if( !$text && $this->mOptions->getEnableImageWhitelist() + if ( !$text && $this->mOptions->getEnableImageWhitelist() && preg_match( self::EXT_IMAGE_REGEX, $url ) ) { $whitelist = explode( "\n", wfMsgForContent( 'external_image_whitelist' ) ); - foreach( $whitelist as $entry ) { + foreach ( $whitelist as $entry ) { # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments - if( strpos( $entry, '#' ) === 0 || $entry === '' ) + if ( strpos( $entry, '#' ) === 0 || $entry === '' ) { continue; - if( preg_match( '/' . str_replace( '/', '\\/', $entry ) . '/i', $url ) ) { + } + if ( preg_match( '/' . str_replace( '/', '\\/', $entry ) . '/i', $url ) ) { # Image matches a whitelist entry $text = $sk->makeExternalImage( $url ); break; @@ -1495,7 +1558,7 @@ class Parser /** * Process [[ ]] wikilinks - * @return processed text + * @return String: processed text * * @private */ @@ -1526,12 +1589,12 @@ class Parser $e1_img = "/^([{$tc}]+)\\|(.*)\$/sD"; } - $sk = $this->mOptions->getSkin(); + $sk = $this->mOptions->getSkin( $this->mTitle ); $holders = new LinkHolderArray( $this ); - #split the entire text string on occurences of [[ + # split the entire text string on occurences of [[ $a = StringUtils::explode( '[[', ' ' . $s ); - #get the first element (all text up to first [[), and remove the space we added + # get the first element (all text up to first [[), and remove the space we added $s = $a->current(); $a->next(); $line = $a->current(); # Workaround for broken ArrayIterator::next() that returns "void" @@ -1545,7 +1608,7 @@ class Parser $e2 = wfMsgForContent( 'linkprefix' ); } - if( is_null( $this->mTitle ) ) { + if ( is_null( $this->mTitle ) ) { wfProfileOut( __METHOD__.'-setup' ); wfProfileOut( __METHOD__ ); throw new MWException( __METHOD__.": \$this->mTitle is null\n" ); @@ -1563,10 +1626,10 @@ class Parser $prefix = ''; } - if($wgContLang->hasVariants()) { - $selflink = $wgContLang->convertLinkToAllVariants($this->mTitle->getPrefixedText()); + if ( $wgContLang->hasVariants() ) { + $selflink = $wgContLang->autoConvertToAllVariants( $this->mTitle->getPrefixedText() ); } else { - $selflink = array($this->mTitle->getPrefixedText()); + $selflink = array( $this->mTitle->getPrefixedText() ); } $useSubpages = $this->areSubpagesAllowed(); wfProfileOut( __METHOD__.'-setup' ); @@ -1590,7 +1653,7 @@ class Parser $prefix=''; } # first link - if($first_prefix) { + if ( $first_prefix ) { $prefix = $first_prefix; $first_prefix = false; } @@ -1610,25 +1673,25 @@ class Parser # Still some problems for cases where the ] is meant to be outside punctuation, # and no image is in sight. See bug 2095. # - if( $text !== '' && + if ( $text !== '' && substr( $m[3], 0, 1 ) === ']' && - strpos($text, '[') !== false + 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 ) { + if ( strpos( $m[1], '%' ) !== false ) { # Should anchors '#' also be rejected? - $m[1] = str_replace( array('<', '>'), array('<', '>'), urldecode($m[1]) ); + $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 + } 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]); + $m[1] = urldecode( $m[1] ); } $trail = ""; } else { # Invalid form; output directly @@ -1655,8 +1718,8 @@ class Parser $link = $m[1]; } - $noforce = (substr( $m[1], 0, 1 ) !== ':'); - if (!$noforce) { + $noforce = ( substr( $m[1], 0, 1 ) !== ':' ); + if ( !$noforce ) { # Strip off leading ':' $link = substr( $link, 1 ); } @@ -1676,10 +1739,10 @@ class Parser if ( $might_be_img ) { # if this is actually an invalid link wfProfileIn( __METHOD__."-might_be_img" ); - if ( $ns == NS_FILE && $noforce ) { #but might be an image + if ( $ns == NS_FILE && $noforce ) { # but might be an image $found = false; while ( true ) { - #look at the next 'line' to see if we can close it there + # look at the next 'line' to see if we can close it there $a->next(); $next_line = $a->current(); if ( $next_line === false || $next_line === null ) { @@ -1693,24 +1756,24 @@ class Parser $trail = $m[2]; break; } elseif ( count( $m ) == 2 ) { - #if there's exactly one ]] that's fine, we'll keep looking + # 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 + # 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 + # but don't ignore what might be perfectly normal links in the text we've examined $holders->merge( $this->replaceInternalLinks2( $text ) ); $s .= "{$prefix}[[$link|$text"; # note: no $trail, because without an end, there *is* no trail wfProfileOut( __METHOD__."-might_be_img" ); continue; } - } else { #it's not an image, so output it raw + } 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( __METHOD__."-might_be_img" ); @@ -1720,7 +1783,14 @@ class Parser } $wasblank = ( $text == '' ); - if ( $wasblank ) $text = $link; + if ( $wasblank ) { + $text = $link; + } else { + # Bug 4598 madness. Handle the quotes only if they come from the alternate part + # [[Lista d''e paise d''o munno]] -> <a href="">Lista d''e paise d''o munno</a> + # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']] -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a> + $text = $this->doQuotes($text); + } # Link not escaped by : , create the various objects if ( $noforce ) { @@ -1729,8 +1799,8 @@ class Parser wfProfileIn( __METHOD__."-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; + $s = rtrim( $s . $prefix ); + $s .= trim( $trail, "\n" ) == '' ? '': $prefix . $trail; wfProfileOut( __METHOD__."-interwiki" ); continue; } @@ -1749,7 +1819,7 @@ class Parser # 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->replaceExternalLinks( $text ); $holders->merge( $this->replaceInternalLinks2( $text ) ); } # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them @@ -1765,7 +1835,7 @@ class Parser if ( $ns == NS_CATEGORY ) { wfProfileIn( __METHOD__."-category" ); - $s = rtrim($s . "\n"); # bug 87 + $s = rtrim( $s . "\n" ); # bug 87 if ( $wasblank ) { $sortkey = $this->getDefaultSort(); @@ -1781,7 +1851,7 @@ class Parser * Strip the whitespace Category links produce, see bug 87 * @todo We might want to use trim($tmp, "\n") here. */ - $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail; + $s .= trim( $prefix . $trail, "\n" ) == '' ? '': $prefix . $trail; wfProfileOut( __METHOD__."-category" ); continue; @@ -1789,8 +1859,8 @@ class Parser } # Self-link checking - if( $nt->getFragment() === '' && $ns != NS_SPECIAL ) { - if( in_array( $nt->getPrefixedText(), $selflink, true ) ) { + if ( $nt->getFragment() === '' && $ns != NS_SPECIAL ) { + if ( in_array( $nt->getPrefixedText(), $selflink, true ) ) { $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail ); continue; } @@ -1798,7 +1868,7 @@ class Parser # NS_MEDIA is a pseudo-namespace for linking directly to a file # FIXME: Should do batch file existence checks, see comment below - if( $ns == NS_MEDIA ) { + if ( $ns == NS_MEDIA ) { wfProfileIn( __METHOD__."-media" ); # Give extensions a chance to select the file revision for us $skip = $time = false; @@ -1821,7 +1891,7 @@ class Parser # # FIXME: isAlwaysKnown() can be expensive for file links; we should really do # batch file existence checks for NS_FILE and NS_MEDIA - if( $iw == '' && $nt->isAlwaysKnown() ) { + if ( $iw == '' && $nt->isAlwaysKnown() ) { $this->mOutput->addLink( $nt ); $s .= $this->makeKnownLinkHolder( $nt, $text, '', $trail, $prefix ); } else { @@ -1853,17 +1923,17 @@ class Parser * 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 + * @param $nt Title + * @param $text String + * @param $query String + * @param $trail String + * @param $prefix String + * @return String: HTML-wikitext mix oh yuck */ function makeKnownLinkHolder( $nt, $text = '', $query = '', $trail = '', $prefix = '' ) { list( $inside, $trail ) = Linker::splitTrail( $trail ); - $sk = $this->mOptions->getSkin(); - // FIXME: use link() instead of deprecated makeKnownLinkObj() + $sk = $this->mOptions->getSkin( $this->mTitle ); + # FIXME: use link() instead of deprecated makeKnownLinkObj() $link = $sk->makeKnownLinkObj( $nt, $text, $query, $inside, $prefix ); return $this->armorLinks( $link ) . $trail; } @@ -1875,8 +1945,8 @@ class Parser * Not needed quite as much as it used to be since free links are a bit * more sensible these days. But bracketed links are still an issue. * - * @param string more-or-less HTML - * @return string less-or-more HTML with NOPARSE bits + * @param $text String: more-or-less HTML + * @return String: less-or-more HTML with NOPARSE bits */ function armorLinks( $text ) { return preg_replace( '/\b(' . wfUrlProtocols() . ')/', @@ -1885,7 +1955,7 @@ class Parser /** * Return true if subpage links should be expanded on this page. - * @return bool + * @return Boolean */ function areSubpagesAllowed() { # Some namespaces don't allow subpages @@ -1894,12 +1964,13 @@ class Parser /** * Handle link to subpage if necessary - * @param string $target the source of the link - * @param string &$text the link text, modified as necessary + * + * @param $target String: the source of the link + * @param &$text String: the link text, modified as necessary * @return string the full name of the link * @private */ - function maybeDoSubpageLink($target, &$text) { + function maybeDoSubpageLink( $target, &$text ) { return Linker::normalizeSubpageLink( $this->mTitle, $target, $text ); } @@ -1907,7 +1978,7 @@ class Parser * Used by doBlockLevels() * @private */ - /* private */ function closeParagraph() { + function closeParagraph() { $result = ''; if ( $this->mLastSection != '' ) { $result = '</' . $this->mLastSection . ">\n"; @@ -1916,42 +1987,64 @@ class Parser $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 ) { + + /** + * 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; } + if ( $fl < $shorter ) { + $shorter = $fl; + } for ( $i = 0; $i < $shorter; ++$i ) { - if ( $st1{$i} != $st2{$i} ) { break; } + 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 ) { + + /** + * 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 .= '<ul><li>'; } - elseif ( '#' === $char ) { $result .= '<ol><li>'; } - elseif ( ':' === $char ) { $result .= '<dl><dd>'; } - elseif ( ';' === $char ) { + if ( '*' === $char ) { + $result .= '<ul><li>'; + } elseif ( '#' === $char ) { + $result .= '<ol><li>'; + } elseif ( ':' === $char ) { + $result .= '<dl><dd>'; + } elseif ( ';' === $char ) { $result .= '<dl><dt>'; $this->mDTopen = true; + } else { + $result = '<!-- ERR 1 -->'; } - else { $result = '<!-- ERR 1 -->'; } return $result; } - /* private */ function nextItem( $char ) { - if ( '*' === $char || '#' === $char ) { return '</li><li>'; } - elseif ( ':' === $char || ';' === $char ) { + /** + * TODO: document + * @param $char String + * @private + */ + function nextItem( $char ) { + if ( '*' === $char || '#' === $char ) { + return '</li><li>'; + } elseif ( ':' === $char || ';' === $char ) { $close = '</dd>'; - if ( $this->mDTopen ) { $close = '</dt>'; } + if ( $this->mDTopen ) { + $close = '</dt>'; + } if ( ';' === $char ) { $this->mDTopen = true; return $close . '<dt>'; @@ -1963,18 +2056,26 @@ class Parser return '<!-- ERR 2 -->'; } - /* private */ function closeList( $char ) { - if ( '*' === $char ) { $text = '</li></ul>'; } - elseif ( '#' === $char ) { $text = '</li></ol>'; } - elseif ( ':' === $char ) { + /** + * TODO: document + * @param $char String + * @private + */ + function closeList( $char ) { + if ( '*' === $char ) { + $text = '</li></ul>'; + } elseif ( '#' === $char ) { + $text = '</li></ol>'; + } elseif ( ':' === $char ) { if ( $this->mDTopen ) { $this->mDTopen = false; $text = '</dt></dl>'; } else { $text = '</dd></dl>'; } + } else { + return '<!-- ERR 3 -->'; } - else { return '<!-- ERR 3 -->'; } return $text."\n"; } /**#@-*/ @@ -1982,7 +2083,8 @@ class Parser /** * Make lists from lines starting with ':', '*', '#', etc. (DBL) * - * @param $linestart bool whether or not this is at the start of a line. + * @param $text String + * @param $linestart Boolean: whether or not this is at the start of a line. * @private * @return string the lists rendered as HTML */ @@ -2007,24 +2109,24 @@ class Parser $linestart = true; continue; } - // * = ul - // # = ol - // ; = dt - // : = dd + # * = ul + # # = ol + # ; = dt + # : = dd $lastPrefixLength = strlen( $lastPrefix ); - $preCloseMatch = preg_match('/<\\/pre/i', $oLine ); - $preOpenMatch = preg_match('/<pre/i', $oLine ); - // If not in a <pre> element, scan for and figure out what prefixes are there. + $preCloseMatch = preg_match( '/<\\/pre/i', $oLine ); + $preOpenMatch = preg_match( '/<pre/i', $oLine ); + # If not in a <pre> element, scan for and figure out what prefixes are there. if ( !$this->mInPre ) { # Multiple prefixes may abut each other for nested lists. $prefixLength = strspn( $oLine, '*#:;' ); $prefix = substr( $oLine, 0, $prefixLength ); # eh? - // ; and : are both from definition-lists, so they're equivalent - // for the purposes of determining whether or not we need to open/close - // elements. + # ; and : are both from definition-lists, so they're equivalent + # for the purposes of determining whether or not we need to open/close + # elements. $prefix2 = str_replace( ';', ':', $prefix ); $t = substr( $oLine, $prefixLength ); $this->mInPre = (bool)$preOpenMatch; @@ -2036,7 +2138,7 @@ class Parser } # List generation - if( $prefixLength && $lastPrefix === $prefix2 ) { + if ( $prefixLength && $lastPrefix === $prefix2 ) { # Same as the last item, so no need to deal with nesting or opening stuff $output .= $this->nextItem( substr( $prefix, -1 ) ); $paragraphStack = false; @@ -2047,37 +2149,37 @@ class Parser # 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) { + if ( $this->findColonNoLinks( $t, $term, $t2 ) !== false ) { $t = $t2; $output .= $term . $this->nextItem( ':' ); } } - } elseif( $prefixLength || $lastPrefixLength ) { - // We need to open or close prefixes, or both. + } elseif ( $prefixLength || $lastPrefixLength ) { + # We need to open or close prefixes, or both. # Either open or close a level... $commonPrefixLength = $this->getCommon( $prefix, $lastPrefix ); $paragraphStack = false; - // Close all the prefixes which aren't shared. - while( $commonPrefixLength < $lastPrefixLength ) { + # Close all the prefixes which aren't shared. + while ( $commonPrefixLength < $lastPrefixLength ) { $output .= $this->closeList( $lastPrefix[$lastPrefixLength-1] ); --$lastPrefixLength; } - // Continue the current prefix if appropriate. + # Continue the current prefix if appropriate. if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) { $output .= $this->nextItem( $prefix[$commonPrefixLength-1] ); } - // Open prefixes where appropriate. + # Open prefixes where appropriate. while ( $prefixLength > $commonPrefixLength ) { $char = substr( $prefix, $commonPrefixLength, 1 ); $output .= $this->openList( $char ); if ( ';' === $char ) { # FIXME: This is dupe of code above - if ($this->findColonNoLinks($t, $term, $t2) !== false) { + if ( $this->findColonNoLinks( $t, $term, $t2 ) !== false ) { $t = $t2; $output .= $term . $this->nextItem( ':' ); } @@ -2087,11 +2189,11 @@ class Parser $lastPrefix = $prefix2; } - // If we have no prefixes, go to paragraph mode. - if( 0 == $prefixLength ) { + # If we have no prefixes, go to paragraph mode. + if ( 0 == $prefixLength ) { wfProfileIn( __METHOD__."-paragraph" ); # No prefix (not in list)--go to paragraph mode - // XXX: use a stack for nestable elements like span, table and div + # XXX: use a stack for nestable elements like span, table and div $openmatch = preg_match('/(?:<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<li|<\\/tr|<\\/td|<\\/th)/iS', $t ); $closematch = preg_match( '/(?:<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|'. @@ -2103,29 +2205,25 @@ class Parser if ( $preOpenMatch and !$preCloseMatch ) { $this->mInPre = true; } - if ( $closematch ) { - $inBlockElem = false; - } else { - $inBlockElem = true; - } - } else if ( !$inBlockElem && !$this->mInPre ) { - if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' or trim($t) != '' ) ) { - // pre - if ($this->mLastSection !== 'pre') { + $inBlockElem = !$closematch; + } elseif ( !$inBlockElem && !$this->mInPre ) { + if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' || trim( $t ) != '' ) ) { + # pre + if ( $this->mLastSection !== 'pre' ) { $paragraphStack = false; $output .= $this->closeParagraph().'<pre>'; $this->mLastSection = 'pre'; } $t = substr( $t, 1 ); } else { - // paragraph - if ( trim($t) == '' ) { + # paragraph + if ( trim( $t ) === '' ) { if ( $paragraphStack ) { $output .= $paragraphStack.'<br />'; $paragraphStack = false; $this->mLastSection = 'p'; } else { - if ($this->mLastSection !== 'p' ) { + if ( $this->mLastSection !== 'p' ) { $output .= $this->closeParagraph(); $this->mLastSection = ''; $paragraphStack = '<p>'; @@ -2138,7 +2236,7 @@ class Parser $output .= $paragraphStack; $paragraphStack = false; $this->mLastSection = 'p'; - } else if ($this->mLastSection !== 'p') { + } elseif ( $this->mLastSection !== 'p' ) { $output .= $this->closeParagraph().'<p>'; $this->mLastSection = 'p'; } @@ -2147,11 +2245,11 @@ class Parser } wfProfileOut( __METHOD__."-paragraph" ); } - // somewhere above we forget to get out of pre block (bug 785) - if($preCloseMatch && $this->mInPre) { + # somewhere above we forget to get out of pre block (bug 785) + if ( $preCloseMatch && $this->mInPre ) { $this->mInPre = false; } - if ($paragraphStack === false) { + if ( $paragraphStack === false ) { $output .= $t."\n"; } } @@ -2171,31 +2269,32 @@ class Parser /** * 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 + * + * @param $str String: the string to split + * @param &$before String: set to everything before the ':' + * @param &$after String: set to everything after the ':' + * return String: the position of the ':', or false if none found */ - function findColonNoLinks($str, &$before, &$after) { + function findColonNoLinks( $str, &$before, &$after ) { wfProfileIn( __METHOD__ ); $pos = strpos( $str, ':' ); - if( $pos === false ) { - // Nothing to find! + if ( $pos === false ) { + # Nothing to find! wfProfileOut( __METHOD__ ); return false; } $lt = strpos( $str, '<' ); - if( $lt === false || $lt > $pos ) { - // Easy; no tag nesting to worry about + if ( $lt === false || $lt > $pos ) { + # Easy; no tag nesting to worry about $before = substr( $str, 0, $pos ); $after = substr( $str, $pos+1 ); wfProfileOut( __METHOD__ ); return $pos; } - // Ugly state machine to walk through avoiding tags. + # Ugly state machine to walk through avoiding tags. $state = self::COLON_STATE_TEXT; $stack = 0; $len = strlen( $str ); @@ -2203,67 +2302,67 @@ class Parser $c = $str{$i}; switch( $state ) { - // (Using the number is a performance hack for common cases) - case 0: // self::COLON_STATE_TEXT: + # (Using the number is a performance hack for common cases) + case 0: # self::COLON_STATE_TEXT: switch( $c ) { case "<": - // Could be either a <start> tag or an </end> tag + # Could be either a <start> tag or an </end> tag $state = self::COLON_STATE_TAGSTART; break; case ":": - if( $stack == 0 ) { - // We found it! + if ( $stack == 0 ) { + # We found it! $before = substr( $str, 0, $i ); $after = substr( $str, $i + 1 ); wfProfileOut( __METHOD__ ); return $i; } - // Embedded in a tag; don't break it. + # Embedded in a tag; don't break it. break; default: - // Skip ahead looking for something interesting + # Skip ahead looking for something interesting $colon = strpos( $str, ':', $i ); - if( $colon === false ) { - // Nothing else interesting + if ( $colon === false ) { + # Nothing else interesting wfProfileOut( __METHOD__ ); return false; } $lt = strpos( $str, '<', $i ); - if( $stack === 0 ) { - if( $lt === false || $colon < $lt ) { - // We found it! + if ( $stack === 0 ) { + if ( $lt === false || $colon < $lt ) { + # We found it! $before = substr( $str, 0, $colon ); $after = substr( $str, $colon + 1 ); wfProfileOut( __METHOD__ ); return $i; } } - if( $lt === false ) { - // Nothing else interesting to find; abort! - // We're nested, but there's no close tags left. Abort! + 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 + # Skip ahead to next tag start $i = $lt; $state = self::COLON_STATE_TAGSTART; } break; - case 1: // self::COLON_STATE_TAG: - // In a <tag> + case 1: # self::COLON_STATE_TAG: + # In a <tag> switch( $c ) { case ">": $stack++; $state = self::COLON_STATE_TEXT; break; case "/": - // Slash may be followed by >? + # Slash may be followed by >? $state = self::COLON_STATE_TAGSLASH; break; default: - // ignore + # ignore } break; - case 2: // self::COLON_STATE_TAGSTART: + case 2: # self::COLON_STATE_TAGSTART: switch( $c ) { case "/": $state = self::COLON_STATE_CLOSETAG; @@ -2272,18 +2371,18 @@ class Parser $state = self::COLON_STATE_COMMENT; break; case ">": - // Illegal early close? This shouldn't happen D: + # 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 </tag> - if( $c === ">" ) { + case 3: # self::COLON_STATE_CLOSETAG: + # In a </tag> + if ( $c === ">" ) { $stack--; - if( $stack < 0 ) { + if ( $stack < 0 ) { wfDebug( __METHOD__.": Invalid input; too many close tags\n" ); wfProfileOut( __METHOD__ ); return false; @@ -2292,28 +2391,28 @@ class Parser } break; case self::COLON_STATE_TAGSLASH: - if( $c === ">" ) { - // Yes, a self-closed tag <blah/> + if ( $c === ">" ) { + # Yes, a self-closed tag <blah/> $state = self::COLON_STATE_TEXT; } else { - // Probably we're jumping the gun, and this is an attribute + # 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 === "-" ) { + case 5: # self::COLON_STATE_COMMENT: + if ( $c === "-" ) { $state = self::COLON_STATE_COMMENTDASH; } break; case self::COLON_STATE_COMMENTDASH: - if( $c === "-" ) { + if ( $c === "-" ) { $state = self::COLON_STATE_COMMENTDASHDASH; } else { $state = self::COLON_STATE_COMMENT; } break; case self::COLON_STATE_COMMENTDASHDASH: - if( $c === ">" ) { + if ( $c === ">" ) { $state = self::COLON_STATE_TEXT; } else { $state = self::COLON_STATE_COMMENT; @@ -2323,8 +2422,9 @@ class Parser throw new MWException( "State machine error in " . __METHOD__ ); } } - if( $stack > 0 ) { + if ( $stack > 0 ) { wfDebug( __METHOD__.": Invalid input; not enough close tags (stack $stack, state $state)\n" ); + wfProfileOut( __METHOD__ ); return false; } wfProfileOut( __METHOD__ ); @@ -2337,8 +2437,8 @@ class Parser * @private */ function getVariableValue( $index, $frame=false ) { - global $wgContLang, $wgSitename, $wgServer, $wgServerName; - global $wgScriptPath, $wgStylePath; + global $wgContLang, $wgSitename, $wgServer; + global $wgArticlePath, $wgScriptPath, $wgStylePath; /** * Some of these require message or data lookups and can be @@ -2442,7 +2542,7 @@ class Parser $value = wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ); break; case 'talkpagename': - if( $this->mTitle->canTalk() ) { + if ( $this->mTitle->canTalk() ) { $talkPage = $this->mTitle->getTalkPage(); $value = wfEscapeWikiText( $talkPage->getPrefixedText() ); } else { @@ -2450,7 +2550,7 @@ class Parser } break; case 'talkpagenamee': - if( $this->mTitle->canTalk() ) { + if ( $this->mTitle->canTalk() ) { $talkPage = $this->mTitle->getTalkPage(); $value = $talkPage->getPrefixedUrl(); } else { @@ -2466,62 +2566,69 @@ class Parser $value = $subjPage->getPrefixedUrl(); break; case 'revisionid': - // Let the edit saving system know we should parse the page - // *after* a revision ID has been assigned. + # 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" ); $value = $this->mRevisionId; break; 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. + # 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" ); $value = intval( substr( $this->getRevisionTimestamp(), 6, 2 ) ); break; 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. + # 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" ); $value = substr( $this->getRevisionTimestamp(), 6, 2 ); break; 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. + # 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" ); + $value = substr( $this->getRevisionTimestamp(), 4, 2 ); + break; + case 'revisionmonth1': + # 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__ . ": {{REVISIONMONTH1}} used, setting vary-revision...\n" ); $value = intval( substr( $this->getRevisionTimestamp(), 4, 2 ) ); break; 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. + # 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" ); $value = substr( $this->getRevisionTimestamp(), 0, 4 ); break; 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. + # 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" ); $value = $this->getRevisionTimestamp(); break; case 'revisionuser': - // Let the edit saving system know we should parse the page - // *after* a revision ID has been assigned. This is for null edits. + # 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__ . ": {{REVISIONUSER}} used, setting vary-revision...\n" ); $value = $this->getRevisionUser(); break; case 'namespace': - $value = str_replace('_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) ); + $value = str_replace( '_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) ); break; case 'namespacee': $value = wfUrlencode( $wgContLang->getNsText( $this->mTitle->getNamespace() ) ); break; case 'talkspace': - $value = $this->mTitle->canTalk() ? str_replace('_',' ',$this->mTitle->getTalkNsText()) : ''; + $value = $this->mTitle->canTalk() ? str_replace( '_',' ',$this->mTitle->getTalkNsText() ) : ''; break; case 'talkspacee': $value = $this->mTitle->canTalk() ? wfUrlencode( $this->mTitle->getTalkNsText() ) : ''; @@ -2545,8 +2652,8 @@ class Parser $value = $wgContLang->formatNum( gmdate( 'H', $ts ), true ); break; case 'currentweek': - // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to - // int to remove the padding + # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to + # int to remove the padding $value = $wgContLang->formatNum( (int)gmdate( 'W', $ts ) ); break; case 'currentdow': @@ -2565,8 +2672,8 @@ class Parser $value = $wgContLang->formatNum( $localHour, true ); break; case 'localweek': - // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to - // int to remove the padding + # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to + # int to remove the padding $value = $wgContLang->formatNum( (int)$localWeek ); break; case 'localdow': @@ -2588,7 +2695,7 @@ class Parser $value = $wgContLang->formatNum( SiteStats::pages() ); break; case 'numberofadmins': - $value = $wgContLang->formatNum( SiteStats::numberingroup('sysop') ); + $value = $wgContLang->formatNum( SiteStats::numberingroup( 'sysop' ) ); break; case 'numberofedits': $value = $wgContLang->formatNum( SiteStats::edits() ); @@ -2605,12 +2712,17 @@ class Parser case 'currentversion': $value = SpecialVersion::getVersion(); break; + case 'articlepath': + return $wgArticlePath; case 'sitename': return $wgSitename; case 'server': return $wgServer; case 'servername': - return $wgServerName; + wfSuppressWarnings(); # May give an E_WARNING in PHP < 5.3.3 + $serverName = parse_url( $wgServer, PHP_URL_HOST ); + wfRestoreWarnings(); + return $serverName ? $serverName : $wgServer; case 'scriptpath': return $wgScriptPath; case 'stylepath': @@ -2618,14 +2730,15 @@ class Parser case 'directionmark': return $wgContLang->getDirMark(); case 'contentlanguage': - global $wgContLanguageCode; - return $wgContLanguageCode; + global $wgLanguageCode; + return $wgLanguageCode; default: $ret = null; - if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) + if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) { return $ret; - else + } else { return null; + } } if ( $index ) @@ -2635,7 +2748,7 @@ class Parser } /** - * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers + * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers * * @private */ @@ -2653,8 +2766,8 @@ class Parser * 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: + * @param $text String: The text to parse + * @param $flags Integer: bitwise combination of: * self::PTD_FOR_INCLUSION Handle <noinclude>/<includeonly> as if the text is being * included. Default is to assume a direct page view. * @@ -2671,12 +2784,12 @@ class Parser * * @private */ - function preprocessToDom ( $text, $flags = 0 ) { + 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 ) { @@ -2702,11 +2815,11 @@ class Parser * 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 $text String: the text to transform + * @param $frame PPFrame Object describing the arguments passed to the template. * Arguments may also be provided as an associative array, as was the usual case before MW1.12. * Providing arguments this way may be useful for extensions wishing to perform variable replacement explicitly. - * @param bool $argsOnly Only do argument (triple-brace) expansion, not double-brace expansion + * @param $argsOnly Boolean: only do argument (triple-brace) expansion, not double-brace expansion * @private */ function replaceVariables( $text, $frame = false, $argsOnly = false ) { @@ -2720,7 +2833,7 @@ class Parser $frame = $this->getPreprocessor()->newFrame(); } elseif ( !( $frame instanceof PPFrame ) ) { wfDebug( __METHOD__." called using plain parameters instead of a PPFrame instance. Creating custom frame.\n" ); - $frame = $this->getPreprocessor()->newCustomFrame($frame); + $frame = $this->getPreprocessor()->newCustomFrame( $frame ); } $dom = $this->preprocessToDom( $text ); @@ -2731,11 +2844,11 @@ class Parser return $text; } - /// Clean up argument array - refactored in 1.9 so parserfunctions can use it, too. + # 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 ) { + foreach ( $args as $arg ) { $eqpos = strpos( $arg, '=' ); if ( $eqpos === false ) { $assocArgs[$index++] = $arg; @@ -2758,15 +2871,22 @@ class Parser * Warn the user when a parser limitation is reached * Will warn at most once the user per limitation type * - * @param string $limitationType, should be one of: - * 'expensive-parserfunction' (corresponding messages: 'expensive-parserfunction-warning', 'expensive-parserfunction-category') - * 'post-expand-template-argument' (corresponding messages: 'post-expand-template-argument-warning', 'post-expand-template-argument-category') - * 'post-expand-template-inclusion' (corresponding messages: 'post-expand-template-inclusion-warning', 'post-expand-template-inclusion-category') - * @params int $current, $max When an explicit limit has been + * @param $limitationType String: should be one of: + * 'expensive-parserfunction' (corresponding messages: + * 'expensive-parserfunction-warning', + * 'expensive-parserfunction-category') + * 'post-expand-template-argument' (corresponding messages: + * 'post-expand-template-argument-warning', + * 'post-expand-template-argument-category') + * 'post-expand-template-inclusion' (corresponding messages: + * 'post-expand-template-inclusion-warning', + * 'post-expand-template-inclusion-category') + * @param $current Current value + * @param $max Maximum allowed, when an explicit limit has been * exceeded, provide the values (optional) */ function limitationWarn( $limitationType, $current=null, $max=null) { - //does no harm if $current and $max are present but are unnecessary for the message + # does no harm if $current and $max are present but are unnecessary for the message $warning = wfMsgExt( "$limitationType-warning", array( 'parsemag', 'escape' ), $current, $max ); $this->mOutput->addWarning( $warning ); $this->addTrackingCategory( "$limitationType-category" ); @@ -2776,12 +2896,12 @@ class Parser * Return the text of a template, after recursively * replacing any variables or templates within the template. * - * @param array $piece The parts of the template + * @param $piece Array: 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 + * @param $frame PPFrame The current frame, contains template arguments + * @return String: the text of the template * @private */ function braceSubstitution( $piece, $frame ) { @@ -2810,7 +2930,7 @@ class Parser $originalTitle = $part1; # $args is a list of argument nodes, starting from index 0, not including $part1 - $args = (null == $piece['parts']) ? array() : $piece['parts']; + $args = ( null == $piece['parts'] ) ? array() : $piece['parts']; wfProfileOut( __METHOD__.'-setup' ); # SUBST @@ -2846,8 +2966,9 @@ class Parser $id = $this->mVariables->matchStartToEnd( $part1 ); if ( $id !== false ) { $text = $this->getVariableValue( $id, $frame ); - if (MagicWord::getCacheTTL($id)>-1) - $this->mOutput->mContainsOldMagic = true; + if ( MagicWord::getCacheTTL( $id ) > -1 ) { + $this->mOutput->updateCacheExpiry( MagicWord::getCacheTTL( $id ) ); + } $found = true; } } @@ -2928,8 +3049,8 @@ class Parser unset( $result[0] ); } - // Extract flags into the local scope - // This allows callers to set flags such as nowiki, found, etc. + # Extract flags into the local scope + # This allows callers to set flags such as nowiki, found, etc. extract( $result ); } else { $text = $result; @@ -2950,21 +3071,23 @@ class Parser # Split the title into page and subpage $subpage = ''; $part1 = $this->maybeDoSubpageLink( $part1, $subpage ); - if ($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){ + if ( $wgContLang->hasVariants() && $title->getArticleID() == 0 ) { $wgContLang->findVariantLink( $part1, $title, true ); } # Do recursion depth check $limit = $this->mOptions->getMaxTemplateDepth(); if ( $frame->depth >= $limit ) { $found = true; - $text = '<span class="error">' . wfMsgForContent( 'parser-template-recursion-depth-warning', $limit ) . '</span>'; + $text = '<span class="error">' + . wfMsgForContent( 'parser-template-recursion-depth-warning', $limit ) + . '</span>'; } } } @@ -2973,15 +3096,18 @@ class Parser if ( !$found && $title ) { wfProfileIn( __METHOD__ . '-loadtpl' ); if ( !$title->isExternal() ) { - if ( $title->getNamespace() == NS_SPECIAL && $this->mOptions->getAllowSpecialInclusion() && $this->ot['html'] ) { + 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 + } elseif ( $wgNonincludableNamespaces && in_array( $title->getNamespace(), $wgNonincludableNamespaces ) ) { + $found = false; # access denied wfDebug( __METHOD__.": template inclusion denied for " . $title->getPrefixedDBkey() ); } else { list( $text, $title ) = $this->getTemplateDom( $title ); @@ -2997,13 +3123,13 @@ class Parser $found = true; } } elseif ( $title->isTrans() ) { - // Interwiki transclusion + # 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 + # Preprocess it like a template $text = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION ); $isChildObj = true; } @@ -3058,22 +3184,31 @@ class Parser # immediately preceding headings if ( $isHTML ) { $text = "\n\n" . $this->insertStripItem( $text ); - } - # Escape nowiki-style return values - elseif ( $nowiki && ( $this->ot['html'] || $this->ot['pre'] ) ) { + } elseif ( $nowiki && ( $this->ot['html'] || $this->ot['pre'] ) ) { + # Escape nowiki-style return values $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)) /*}*/{ + } elseif ( is_string( $text ) + && !$piece['lineStart'] + && preg_match( '/^(?:{\\||:|;|#|\*)/', $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. $text = "\n" . $text; } if ( is_string( $text ) && !$this->incrementIncludeSize( 'post-expand', strlen( $text ) ) ) { # Error, oversize inclusion - $text = "[[$originalTitle]]" . - $this->insertStripItem( '<!-- WARNING: template omitted, post-expand include size too large -->' ); + if ( $titleText !== false ) { + # Make a working, properly escaped link if possible (bug 23588) + $text = "[[:$titleText]]"; + } else { + # This will probably not be a working link, but at least it may + # provide some hint of where the problem is + preg_replace( '/^:/', '', $originalTitle ); + $text = "[[:$originalTitle]]"; + } + $text .= $this->insertStripItem( '<!-- WARNING: template omitted, post-expand include size too large -->' ); $this->limitationWarn( 'post-expand-template-inclusion' ); } @@ -3104,7 +3239,7 @@ class Parser return array( $this->mTplDomCache[$titleText], $title ); } - // Cache miss, go to the database + # Cache miss, go to the database list( $text, $title ) = $this->fetchTemplateAndTitle( $title ); if ( $text === false ) { @@ -3115,9 +3250,9 @@ class Parser $dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION ); $this->mTplDomCache[ $titleText ] = $dom; - if (! $title->equals($cacheTitle)) { + if ( !$title->equals( $cacheTitle ) ) { $this->mTplRedirCache[$cacheTitle->getPrefixedDBkey()] = - array( $title->getNamespace(),$cdb = $title->getDBkey() ); + array( $title->getNamespace(), $cdb = $title->getDBkey() ); } return array( $dom, $title ); @@ -3127,7 +3262,7 @@ class Parser * Fetch the unparsed text of a template and register a reference to it. */ function fetchTemplateAndTitle( $title ) { - $templateCb = $this->mOptions->getTemplateCallback(); + $templateCb = $this->mOptions->getTemplateCallback(); # Defaults to Parser::statelessFetchTemplate() $stuff = call_user_func( $templateCb, $title, $this ); $text = $stuff['text']; $finalTitle = isset( $stuff['finalTitle'] ) ? $stuff['finalTitle'] : $title; @@ -3136,11 +3271,11 @@ class Parser $this->mOutput->addTemplate( $dep['title'], $dep['page_id'], $dep['rev_id'] ); } } - return array($text,$finalTitle); + return array( $text, $finalTitle ); } function fetchTemplate( $title ) { - $rv = $this->fetchTemplateAndTitle($title); + $rv = $this->fetchTemplateAndTitle( $title ); return $rv[0]; } @@ -3153,13 +3288,13 @@ class Parser $finalTitle = $title; $deps = array(); - // Loop to fetch the article, with up to 1 redirect + # 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 + $id = false; # Assume current wfRunHooks( 'BeforeParserFetchTemplateAndtitle', array( $parser, &$title, &$skip, &$id ) ); - if( $skip ) { + if ( $skip ) { $text = false; $deps[] = array( 'title' => $title, @@ -3169,8 +3304,8 @@ class Parser } $rev = $id ? Revision::newFromId( $id ) : Revision::newFromTitle( $title ); $rev_id = $rev ? $rev->getId() : 0; - // If there is no current revision, there is no page - if( $id === false && !$rev ) { + # If there is no current revision, there is no page + if ( $id === false && !$rev ) { $linkCache = LinkCache::singleton(); $linkCache->addBadLinkObj( $title ); } @@ -3180,13 +3315,13 @@ class Parser 'page_id' => $title->getArticleID(), 'rev_id' => $rev_id ); - if( $rev ) { + if ( $rev ) { $text = $rev->getText(); - } elseif( $title->getNamespace() == NS_MEDIAWIKI ) { + } elseif ( $title->getNamespace() == NS_MEDIAWIKI ) { global $wgContLang; $message = $wgContLang->lcfirst( $title->getText() ); $text = wfMsgForContentNoTrans( $message ); - if( wfEmptyMsg( $message, $text ) ) { + if ( wfEmptyMsg( $message, $text ) ) { $text = false; break; } @@ -3196,7 +3331,7 @@ class Parser if ( $text === false ) { break; } - // Redirect? + # Redirect? $finalTitle = $title; $title = Title::newFromRedirect( $text ); } @@ -3212,35 +3347,39 @@ class Parser function interwikiTransclude( $title, $action ) { global $wgEnableScaryTranscluding; - if (!$wgEnableScaryTranscluding) - return wfMsg('scarytranscludedisabled'); + if ( !$wgEnableScaryTranscluding ) { + return wfMsgForContent('scarytranscludedisabled'); + } $url = $title->getFullUrl( "action=$action" ); - if (strlen($url) > 255) - return wfMsg('scarytranscludetoolong'); - return $this->fetchScaryTemplateMaybeFromCache($url); + if ( strlen( $url ) > 255 ) { + return wfMsgForContent( 'scarytranscludetoolong' ); + } + return $this->fetchScaryTemplateMaybeFromCache( $url ); } - function fetchScaryTemplateMaybeFromCache($url) { + function fetchScaryTemplateMaybeFromCache( $url ) { global $wgTranscludeCacheExpiry; - $dbr = wfGetDB(DB_SLAVE); + $dbr = wfGetDB( DB_SLAVE ); $tsCond = $dbr->timestamp( time() - $wgTranscludeCacheExpiry ); - $obj = $dbr->selectRow('transcache', array('tc_time', 'tc_contents'), - array('tc_url' => $url, "tc_time >= " . $dbr->addQuotes( $tsCond ) ) ); - if ($obj) { + $obj = $dbr->selectRow( 'transcache', array('tc_time', 'tc_contents' ), + array( 'tc_url' => $url, "tc_time >= " . $dbr->addQuotes( $tsCond ) ) ); + if ( $obj ) { return $obj->tc_contents; } - $text = Http::get($url); - if (!$text) - return wfMsg('scarytranscludefailed', $url); + $text = Http::get( $url ); + if ( !$text ) { + return wfMsgForContent( 'scarytranscludefailed', $url ); + } - $dbw = wfGetDB(DB_MASTER); - $dbw->replace('transcache', array('tc_url'), array( + $dbw = wfGetDB( DB_MASTER ); + $dbw->replace( 'transcache', array('tc_url'), array( 'tc_url' => $url, 'tc_time' => $dbw->timestamp( time() ), - 'tc_contents' => $text)); + 'tc_contents' => $text) + ); return $text; } @@ -3294,21 +3433,19 @@ class Parser * Return the text to be used for a given extension tag. * This is the ghost of strip(). * - * @param array $params Associative array of parameters: + * @param $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 + * @param $frame PPFrame */ 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++) . self::MARKER_SUFFIX; + $marker = "{$this->mUniqPrefix}-$name-" . sprintf( '%08X', $this->mMarkerIndex++ ) . self::MARKER_SUFFIX; $isFunctionTag = isset( $this->mFunctionTagHooks[strtolower($name)] ) && ( $this->ot['html'] || $this->ot['pre'] ); @@ -3324,27 +3461,27 @@ class Parser $attributes = $attributes + $params['attributes']; } - if( isset( $this->mTagHooks[$name] ) ) { + 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, $frame ) ); - } elseif( isset( $this->mFunctionTagHooks[$name] ) ) { + } elseif ( isset( $this->mFunctionTagHooks[$name] ) ) { list( $callback, $flags ) = $this->mFunctionTagHooks[$name]; - if( !is_callable( $callback ) ) + if ( !is_callable( $callback ) ) { throw new MWException( "Tag hook for $name is not callable\n" ); + } - $output = call_user_func_array( $callback, - array( &$this, $frame, $content, $attributes ) ); + $output = call_user_func_array( $callback, array( &$this, $frame, $content, $attributes ) ); } else { $output = '<span class="error">Invalid tag extension name: ' . htmlspecialchars( $name ) . '</span>'; } if ( is_array( $output ) ) { - // Extract flags to local scope (to override $markerType) + # Extract flags to local scope (to override $markerType) $flags = $output; $output = $flags[0]; unset( $flags[0] ); @@ -3368,7 +3505,7 @@ class Parser } } - if( $markerType === 'none' ) { + if ( $markerType === 'none' ) { return $output; } elseif ( $markerType === 'nowiki' ) { $this->mStripState->nowiki->setPair( $marker, $output ); @@ -3383,9 +3520,9 @@ class Parser /** * 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 + * @param $type String: the type of expansion + * @param $size Integer: 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 ) ) { @@ -3399,12 +3536,12 @@ class Parser /** * Increment the expensive function count * - * @return boolean False if the limit has been exceeded + * @return Boolean: false if the limit has been exceeded */ function incrementExpensiveFunctionCount() { global $wgExpensiveParserFunctionLimit; $this->mExpensiveFunctionCount++; - if($this->mExpensiveFunctionCount <= $wgExpensiveParserFunctionLimit) { + if ( $this->mExpensiveFunctionCount <= $wgExpensiveParserFunctionLimit ) { return true; } return false; @@ -3417,20 +3554,20 @@ class Parser function doDoubleUnderscore( $text ) { wfProfileIn( __METHOD__ ); - // The position of __TOC__ needs to be recorded + # The position of __TOC__ needs to be recorded $mw = MagicWord::get( 'toc' ); - if( $mw->match( $text ) ) { + if ( $mw->match( $text ) ) { $this->mShowToc = true; $this->mForceTocPosition = true; - // Set a placeholder. At the end we'll fill it in with the TOC. + # Set a placeholder. At the end we'll fill it in with the TOC. $text = $mw->replace( '<!--MWTOC-->', $text, 1 ); - // Only keep the first one. + # Only keep the first one. $text = $mw->replace( '', $text ); } - // Now match and remove the rest of them + # Now match and remove the rest of them $mwa = MagicWord::getDoubleUnderscoreArray(); $this->mDoubleUnderscores = $mwa->matchAndRemove( $text ); @@ -3441,21 +3578,25 @@ class Parser $this->mShowToc = false; } if ( isset( $this->mDoubleUnderscores['hiddencat'] ) && $this->mTitle->getNamespace() == NS_CATEGORY ) { - $this->mOutput->setProperty( 'hiddencat', 'y' ); $this->addTrackingCategory( 'hidden-category-category' ); } # (bug 8068) Allow control over whether robots index a page. # # FIXME (bug 14899): __INDEX__ always overrides __NOINDEX__ here! This # is not desirable, the last one on the page should win. - if( isset( $this->mDoubleUnderscores['noindex'] ) && $this->mTitle->canUseNoindex() ) { + if ( isset( $this->mDoubleUnderscores['noindex'] ) && $this->mTitle->canUseNoindex() ) { $this->mOutput->setIndexPolicy( 'noindex' ); $this->addTrackingCategory( 'noindex-category' ); } - if( isset( $this->mDoubleUnderscores['index'] ) && $this->mTitle->canUseNoindex() ){ + if ( isset( $this->mDoubleUnderscores['index'] ) && $this->mTitle->canUseNoindex() ) { $this->mOutput->setIndexPolicy( 'index' ); $this->addTrackingCategory( 'index-category' ); } + + # Cache all double underscores in the database + foreach ( $this->mDoubleUnderscores as $key => $val ) { + $this->mOutput->setProperty( $key, '' ); + } wfProfileOut( __METHOD__ ); return $text; @@ -3464,14 +3605,17 @@ class Parser /** * Add a tracking category, getting the title from a system message, * or print a debug message if the title is invalid. - * @param $msg String message key - * @return Bool whether the addition was successful + * + * @param $msg String: message key + * @return Boolean: whether the addition was successful */ - protected function addTrackingCategory( $msg ){ + protected function addTrackingCategory( $msg ) { $cat = wfMsgForContent( $msg ); # Allow tracking categories to be disabled by setting them to "-" - if( $cat === '-' ) return false; + if ( $cat === '-' ) { + return false; + } $containerCategory = Title::makeTitleSafe( NS_CATEGORY, $cat ); if ( $containerCategory ) { @@ -3493,25 +3637,21 @@ class Parser * 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 string $origText Original, untouched wikitext - * @param boolean $isMain + * @param $text String + * @param $origText String: original, untouched wikitext + * @param $isMain Boolean * @private */ function formatHeadings( $text, $origText, $isMain=true ) { global $wgMaxTocLevel, $wgContLang, $wgHtml5, $wgExperimentalHtmlIds; $doNumberHeadings = $this->mOptions->getNumberHeadings(); - $showEditLink = $this->mOptions->getEditSection(); - - // Do not call quickUserCan unless necessary - if( $showEditLink && !$this->mTitle->quickUserCan( 'edit' ) ) { - $showEditLink = 0; - } - + # Inhibit editsection links if requested in the page - if ( isset( $this->mDoubleUnderscores['noeditsection'] ) || $this->mOptions->getIsPrintable() ) { + if ( isset( $this->mDoubleUnderscores['noeditsection'] ) ) { $showEditLink = 0; + } else { + $showEditLink = $this->mOptions->getEditSection(); } # Get all headlines for numbering them and adding funky stuff like [edit] @@ -3522,7 +3662,7 @@ class Parser # 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); + ( ( $numMatches >= 4 ) || $this->mForceTocPosition ); # Allow user to stipulate that a page should have a "new section" # link added via __NEWSECTIONLINK__ @@ -3544,7 +3684,7 @@ class Parser } # We need this to perform operations on the HTML - $sk = $this->mOptions->getSkin(); + $sk = $this->mOptions->getSkin( $this->mTitle ); # headline counter $headlineCount = 0; @@ -3557,7 +3697,6 @@ class Parser $head = array(); $sublevelCount = array(); $levelCount = array(); - $toclevel = 0; $level = 0; $prevlevel = 0; $toclevel = 0; @@ -3571,23 +3710,23 @@ class Parser $node = $root->getFirstChild(); $byteOffset = 0; $tocraw = array(); + $refers = array(); - foreach( $matches[3] as $headline ) { + foreach ( $matches[3] as $headline ) { $isTemplate = false; $titleText = false; $sectionIndex = false; $numbering = ''; $markerMatches = array(); - if (preg_match("/^$markerRegex/", $headline, $markerMatches)) { + if ( preg_match("/^$markerRegex/", $headline, $markerMatches ) ) { $serial = $markerMatches[1]; list( $titleText, $sectionIndex ) = $this->mHeadings[$serial]; - $isTemplate = ($titleText != $baseTitleText); - $headline = preg_replace("/^$markerRegex/", "", $headline); + $isTemplate = ( $titleText != $baseTitleText ); + $headline = preg_replace( "/^$markerRegex/", "", $headline ); } - if( $toclevel ) { + if ( $toclevel ) { $prevlevel = $level; - $prevtoclevel = $toclevel; } $level = $matches[1][$headlineCount]; @@ -3595,30 +3734,30 @@ class Parser # Increase TOC level $toclevel++; $sublevelCount[$toclevel] = 0; - if( $toclevel<$wgMaxTocLevel ) { + if ( $toclevel<$wgMaxTocLevel ) { $prevtoclevel = $toclevel; $toc .= $sk->tocIndent(); $numVisible++; } - } - elseif ( $level < $prevlevel && $toclevel > 1 ) { + } elseif ( $level < $prevlevel && $toclevel > 1 ) { # Decrease TOC level, find level to jump to - for ($i = $toclevel; $i > 0; $i--) { + for ( $i = $toclevel; $i > 0; $i-- ) { if ( $levelCount[$i] == $level ) { # Found last matching level $toclevel = $i; break; - } - elseif ( $levelCount[$i] < $level ) { + } elseif ( $levelCount[$i] < $level ) { # Found first matching level below current level $toclevel = $i + 1; break; } } - if( $i == 0 ) $toclevel = 1; - if( $toclevel<$wgMaxTocLevel ) { - if($prevtoclevel < $wgMaxTocLevel) { + if ( $i == 0 ) { + $toclevel = 1; + } + if ( $toclevel<$wgMaxTocLevel ) { + if ( $prevtoclevel < $wgMaxTocLevel ) { # Unindent only if the previous toc level was shown :p $toc .= $sk->tocUnindent( $prevtoclevel - $toclevel ); $prevtoclevel = $toclevel; @@ -3626,10 +3765,9 @@ class Parser $toc .= $sk->tocLineEnd(); } } - } - else { + } else { # No change in level, end TOC line - if( $toclevel<$wgMaxTocLevel ) { + if ( $toclevel<$wgMaxTocLevel ) { $toc .= $sk->tocLineEnd(); } } @@ -3640,8 +3778,8 @@ class Parser @$sublevelCount[$toclevel]++; $dot = 0; for( $i = 1; $i <= $toclevel; $i++ ) { - if( !empty( $sublevelCount[$i] ) ) { - if( $dot ) { + if ( !empty( $sublevelCount[$i] ) ) { + if ( $dot ) { $numbering .= '.'; } $numbering .= $wgContLang->formatNum( $sublevelCount[$i] ); @@ -3662,15 +3800,14 @@ class Parser # Strip out HTML (other than plain <sup> and <sub>: bug 8393) $tocline = preg_replace( array( '#<(?!/?(sup|sub)).*?'.'>#', '#<(/?(sup|sub)).*?'.'>#' ), - array( '', '<$1>'), + array( '', '<$1>' ), $safeHeadline ); $tocline = trim( $tocline ); # For the anchor, strip out HTML-y stuff period $safeHeadline = preg_replace( '/<.*?'.'>/', '', $safeHeadline ); - $safeHeadline = preg_replace( '/[ _]+/', ' ', $safeHeadline ); - $safeHeadline = trim( $safeHeadline ); + $safeHeadline = Sanitizer::normalizeSectionNameWhitespace( $safeHeadline ); # Save headline for section edit hint before it's escaped $headlineHint = $safeHeadline; @@ -3700,9 +3837,10 @@ class Parser 'noninitial' ); } - # HTML names must be case-insensitively unique (bug 10721). FIXME: - # Does this apply to Unicode characters? Because we aren't - # handling those here. + # HTML names must be case-insensitively unique (bug 10721). + # This does not apply to Unicode characters per + # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison + # FIXME: We may be changing them depending on the current locale. $arrayKey = strtolower( $safeHeadline ); if ( $legacyHeadline === false ) { $legacyArrayKey = false; @@ -3723,7 +3861,7 @@ class Parser } # Don't number the heading if it is the only one (looks silly) - if( $doNumberHeadings && count( $matches[3] ) > 1) { + if ( $doNumberHeadings && count( $matches[3] ) > 1) { # the two are different if the line contains a link $headline = $numbering . ' ' . $headline; } @@ -3737,9 +3875,9 @@ class Parser if ( $legacyHeadline !== false && $refers[$legacyArrayKey] > 1 ) { $legacyAnchor .= '_' . $refers[$legacyArrayKey]; } - if( $enoughToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) { - $toc .= $sk->tocLine($anchor, $tocline, - $numbering, $toclevel, ($isTemplate ? false : $sectionIndex)); + if ( $enoughToc && ( !isset( $wgMaxTocLevel ) || $toclevel < $wgMaxTocLevel ) ) { + $toc .= $sk->tocLine( $anchor, $tocline, + $numbering, $toclevel, ( $isTemplate ? false : $sectionIndex ) ); } # Add the section to the section tree @@ -3747,8 +3885,9 @@ class Parser while ( $node && !$isTemplate ) { if ( $node->getName() === 'h' ) { $bits = $node->splitHeading(); - if ( $bits['i'] == $sectionIndex ) + if ( $bits['i'] == $sectionIndex ) { break; + } } $byteOffset += mb_strlen( $this->mStripState->unstripBoth( $frame->expand( $node, PPFrame::RECOVER_ORIG ) ) ); @@ -3759,20 +3898,20 @@ class Parser 'level' => $level, 'line' => $tocline, 'number' => $numbering, - 'index' => ($isTemplate ? 'T-' : '' ) . $sectionIndex, + 'index' => ( $isTemplate ? 'T-' : '' ) . $sectionIndex, 'fromtitle' => $titleText, 'byteoffset' => ( $isTemplate ? null : $byteOffset ), 'anchor' => $anchor, ); # give headline the correct <h#> tag - if( $showEditLink && $sectionIndex !== false ) { - if( $isTemplate ) { + if ( $showEditLink && $sectionIndex !== false ) { + if ( $isTemplate ) { # Put a T flag in the section identifier, to indicate to extractSections() # that sections inside <includeonly> should be counted. - $editlink = $sk->doEditSectionLink(Title::newFromText( $titleText ), "T-$sectionIndex"); + $editlink = $sk->doEditSectionLink( Title::newFromText( $titleText ), "T-$sectionIndex", null, $this->mOptions->getUserLang() ); } else { - $editlink = $sk->doEditSectionLink($this->mTitle, $sectionIndex, $headlineHint); + $editlink = $sk->doEditSectionLink( $this->mTitle, $sectionIndex, $headlineHint, $this->mOptions->getUserLang() ); } } else { $editlink = ''; @@ -3787,15 +3926,15 @@ class Parser $this->setOutputType( $oldType ); # Never ever show TOC if no headers - if( $numVisible < 1 ) { + if ( $numVisible < 1 ) { $enoughToc = false; } - if( $enoughToc ) { - if( $prevtoclevel > 0 && $prevtoclevel < $wgMaxTocLevel ) { + if ( $enoughToc ) { + if ( $prevtoclevel > 0 && $prevtoclevel < $wgMaxTocLevel ) { $toc .= $sk->tocUnindent( $prevtoclevel - 1 ); } - $toc = $sk->tocList( $toc ); + $toc = $sk->tocList( $toc, $this->mOptions->getUserLang() ); $this->mOutput->setTOCHTML( $toc ); } @@ -3808,8 +3947,8 @@ class Parser $blocks = preg_split( '/<H[1-6].*?' . '>.*?<\/H[1-6]>/i', $text ); $i = 0; - foreach( $blocks as $block ) { - if( $showEditLink && $headlineCount > 0 && $i == 0 && $block !== "\n" ) { + 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 @@ -3818,17 +3957,17 @@ class Parser # $full .= $sk->editSectionLink(0); } $full .= $block; - if( $enoughToc && !$i && $isMain && !$this->mForceTocPosition ) { + if ( $enoughToc && !$i && $isMain && !$this->mForceTocPosition ) { # Top anchor now in skin $full = $full.$toc; } - if( !empty( $head[$i] ) ) { + if ( !empty( $head[$i] ) ) { $full .= $head[$i]; } $i++; } - if( $this->mForceTocPosition ) { + if ( $this->mForceTocPosition ) { return str_replace( '<!--MWTOC-->', $toc, $full ); } else { return $full; @@ -3836,108 +3975,17 @@ class Parser } /** - * Merge $tree2 into $tree1 by replacing the section with index - * $section in $tree1 and its descendants with the sections in $tree2. - * Note that in the returned section tree, only the 'index' and - * 'byteoffset' fields are guaranteed to be correct. - * @param $tree1 array Section tree from ParserOutput::getSectons() - * @param $tree2 array Section tree - * @param $section int Section index - * @param $title Title Title both section trees come from - * @param $len2 int Length of the original wikitext for $tree2 - * @return array Merged section tree - */ - public static function mergeSectionTrees( $tree1, $tree2, $section, $title, $len2 ) { - global $wgContLang; - $newTree = array(); - $targetLevel = false; - $merged = false; - $lastLevel = 1; - $nextIndex = 1; - $numbering = array( 0 ); - $titletext = $title->getPrefixedDBkey(); - foreach ( $tree1 as $s ) { - if ( $targetLevel !== false ) { - if ( $s['level'] <= $targetLevel ) - // We've skipped enough - $targetLevel = false; - else - continue; - } - if ( $s['index'] != $section || - $s['fromtitle'] != $titletext ) { - self::incrementNumbering( $numbering, - $s['toclevel'], $lastLevel ); - - // Rewrite index, byteoffset and number - if ( $s['fromtitle'] == $titletext ) { - $s['index'] = $nextIndex++; - if ( $merged ) - $s['byteoffset'] += $len2; - } - $s['number'] = implode( '.', array_map( - array( $wgContLang, 'formatnum' ), - $numbering ) ); - $lastLevel = $s['toclevel']; - $newTree[] = $s; - } else { - // We're at $section - // Insert sections from $tree2 here - foreach ( $tree2 as $s2 ) { - // Rewrite the fields in $s2 - // before inserting it - $s2['toclevel'] += $s['toclevel'] - 1; - $s2['level'] += $s['level'] - 1; - $s2['index'] = $nextIndex++; - $s2['byteoffset'] += $s['byteoffset']; - - self::incrementNumbering( $numbering, - $s2['toclevel'], $lastLevel ); - $s2['number'] = implode( '.', array_map( - array( $wgContLang, 'formatnum' ), - $numbering ) ); - $lastLevel = $s2['toclevel']; - $newTree[] = $s2; - } - // Skip all descendants of $section in $tree1 - $targetLevel = $s['level']; - $merged = true; - } - } - return $newTree; - } - - /** - * Increment a section number. Helper function for mergeSectionTrees() - * @param $number array Array representing a section number - * @param $level int Current TOC level (depth) - * @param $lastLevel int Level of previous TOC entry - */ - private static function incrementNumbering( &$number, $level, $lastLevel ) { - if ( $level > $lastLevel ) - $number[$level - 1] = 1; - else if ( $level < $lastLevel ) { - foreach ( $number as $key => $unused ) - if ( $key >= $level ) - unset( $number[$key] ); - $number[$level - 1]++; - } else - $number[$level - 1]++; - } - - /** * 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 + * @param $text String: 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 $options ParserOptions: parsing options + * @param $clearState Boolean: whether to clear the parser state first + * @return String: the altered wiki markup */ - function preSaveTransform( $text, Title $title, $user, $options, $clearState = true ) { + public function preSaveTransform( $text, Title $title, $user, $options, $clearState = true ) { $this->mOptions = $options; $this->setTitle( $title ); $this->setOutputType( self::OT_WIKI ); @@ -3962,13 +4010,11 @@ class Parser 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) - */ + # 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(); if ( isset( $wgLocaltimezone ) ) { $tz = $wgLocaltimezone; @@ -3982,13 +4028,14 @@ class Parser $ts = date( 'YmdHis', $unixts ); $tzMsg = date( 'T', $unixts ); # might vary on DST changeover! - /* Allow translation of timezones trough wiki. date() can return - * whatever crap the system uses, localised or not, so we cannot - * ship premade translations. - */ + # Allow translation of timezones through wiki. date() can return + # whatever crap the system uses, localised or not, so we cannot + # ship premade translations. $key = 'timezone-' . strtolower( trim( $tzMsg ) ); $value = wfMsgForContent( $key ); - if ( !wfEmptyMsg( $key, $value ) ) $tzMsg = $value; + if ( !wfEmptyMsg( $key, $value ) ) { + $tzMsg = $value; + } date_default_timezone_set( $oldtz ); @@ -3998,6 +4045,9 @@ class Parser # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags $text = $this->replaceVariables( $text ); + # This works almost by chance, as the replaceVariables are done before the getUserSig(), + # which may corrupt this parser instance via its wfMsgExt( parsemag ) call- + # Signatures $sigText = $this->getUserSig( $user ); $text = strtr( $text, array( @@ -4007,7 +4057,6 @@ class Parser ) ); # Context links: [[|name]] and [[name (context)|]] - # global $wgLegalTitleChars; $tc = "[$wgLegalTitleChars]"; $nc = '[ _0-9A-Za-z\x80-\xff-]'; # Namespaces can use non-ascii! @@ -4045,7 +4094,10 @@ class Parser * If you have pre-fetched the nickname or the fancySig option, you can * specify them here to save a database query. * - * @param User $user + * @param $user User + * @param $nickname String: nickname to use or false to use user's default nickname + * @param $fancySig Boolean: whether the nicknname is the complete signature + * or null to use default value * @return string */ function getUserSig( &$user, $nickname = false, $fancySig = null ) { @@ -4053,21 +4105,22 @@ class Parser $username = $user->getName(); - // If not given, retrieve from the user object. + # If not given, retrieve from the user object. if ( $nickname === false ) $nickname = $user->getOption( 'nickname' ); - if ( is_null( $fancySig ) ) + if ( is_null( $fancySig ) ) { $fancySig = $user->getBoolOption( 'fancysig' ); + } $nickname = $nickname == null ? $username : $nickname; - if( mb_strlen( $nickname ) > $wgMaxSigChars ) { + if ( mb_strlen( $nickname ) > $wgMaxSigChars ) { $nickname = $username; wfDebug( __METHOD__ . ": $username has overlong signature.\n" ); - } elseif( $fancySig !== false ) { + } elseif ( $fancySig !== false ) { # Sig. might contain markup; validate this - if( $this->validateSig( $nickname ) !== false ) { + if ( $this->validateSig( $nickname ) !== false ) { # Validated; clean up (if needed) and return it return $this->cleanSig( $nickname, true ); } else { @@ -4077,7 +4130,7 @@ class Parser } } - // Make sure nickname doesnt get a sig in a sig + # 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 @@ -4093,7 +4146,7 @@ class Parser /** * Check that the user's signature contains no bad XML * - * @param string $text + * @param $text String * @return mixed An expanded string, or false if invalid. */ function validateSig( $text ) { @@ -4106,16 +4159,16 @@ class Parser * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig * 2) Substitute all transclusions * - * @param string $text + * @param $text String * @param $parsing Whether we're cleaning (preferences save) or parsing - * @return string Signature text + * @return String: signature text */ function cleanSig( $text, $parsing = false ) { if ( !$parsing ) { global $wgTitle; + $this->mOptions = new ParserOptions; $this->clearState(); $this->setTitle( $wgTitle ); - $this->mOptions = new ParserOptions; $this->setOutputType = self::OT_PREPROCESS; } @@ -4145,8 +4198,9 @@ class Parser /** * Strip ~~~, ~~~~ and ~~~~~ out of signatures - * @param string $text - * @return string Signature text with /~{3,5}/ removed + * + * @param $text String + * @return String: signature text with /~{3,5}/ removed */ function cleanSigInSig( $text ) { $text = preg_replace( '/~{3,5}/', '', $text ); @@ -4156,9 +4210,8 @@ class Parser /** * Set up some variables which are usually set up in parse() * so that an external function can call some class members with confidence - * @public */ - function startExternalParse( &$title, $options, $outputType, $clearState = true ) { + public function startExternalParse( Title $title = null, ParserOptions $options, $outputType, $clearState = true ) { $this->setTitle( $title ); $this->mOptions = $options; $this->setOutputType( $outputType ); @@ -4170,12 +4223,11 @@ class Parser /** * Wrapper for preprocess() * - * @param string $text the text to preprocess - * @param ParserOptions $options options - * @return string - * @public + * @param $text String: the text to preprocess + * @param $options ParserOptions: options + * @return String */ - function transformMsg( $text, $options ) { + public function transformMsg( $text, $options ) { global $wgTitle; static $executing = false; @@ -4185,34 +4237,37 @@ class Parser } $executing = true; - wfProfileIn(__METHOD__); - $text = $this->preprocess( $text, $wgTitle, $options ); + wfProfileIn( __METHOD__ ); + $title = $wgTitle; + if ( !$title ) { + # It's not uncommon having a null $wgTitle in scripts. See r80898 + # Create a ghost title in such case + $title = Title::newFromText( 'Dwimmerlaik' ); + } + $text = $this->preprocess( $text, $title, $options ); $executing = false; - wfProfileOut(__METHOD__); + wfProfileOut( __METHOD__ ); return $text; } /** * Create an HTML-style tag, e.g. <yourtag>special text</yourtag> * The callback should have the following form: - * function myParserHook( $text, $params, &$parser ) { ... } + * 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 <hook> - * @param mixed $callback The callback function (and object) to use for the tag - * + * @param $tag Mixed: the tag to use, e.g. 'hook' for <hook> + * @param $callback Mixed: 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 ) { + public function setHook( $tag, $callback ) { $tag = strtolower( $tag ); $oldVal = isset( $this->mTagHooks[$tag] ) ? $this->mTagHooks[$tag] : null; $this->mTagHooks[$tag] = $callback; - if( !in_array( $tag, $this->mStripList ) ) { + if ( !in_array( $tag, $this->mStripList ) ) { $this->mStripList[] = $tag; } @@ -4251,11 +4306,9 @@ class Parser * 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: + * @param $id String: The magic word ID + * @param $callback Mixed: the callback function (and object) to use + * @param $flags Integer: a combination of the following flags: * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}} * * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This @@ -4279,7 +4332,7 @@ class Parser * * @return The old callback function for this name, if any */ - function setFunctionHook( $id, $callback, $flags = 0 ) { + public function setFunctionHook( $id, $callback, $flags = 0 ) { global $wgContLang; $oldVal = isset( $this->mFunctionHooks[$id] ) ? $this->mFunctionHooks[$id][0] : null; @@ -4287,7 +4340,7 @@ class Parser # Add to function cache $mw = MagicWord::get( $id ); - if( !$mw ) + if ( !$mw ) throw new MWException( __METHOD__.'() expecting a magic word identifier.' ); $synonyms = $mw->getSynonyms(); @@ -4314,7 +4367,7 @@ class Parser /** * Get all registered function hook identifiers * - * @return array + * @return Array */ function getFunctionHooks() { return array_keys( $this->mFunctionHooks ); @@ -4331,7 +4384,7 @@ class Parser $this->mFunctionTagHooks[$tag] : null; $this->mFunctionTagHooks[$tag] = array( $callback, $flags ); - if( !in_array( $tag, $this->mStripList ) ) { + if ( !in_array( $tag, $this->mStripList ) ) { $this->mStripList[] = $tag; } @@ -4351,8 +4404,9 @@ class Parser /** * Replace <!--LINK--> link placeholders with plain text of links * (not HTML-formatted). - * @param string $text - * @return string + * + * @param $text String + * @return String */ function replaceLinkHoldersText( $text ) { return $this->mLinkHolders->replaceText( $text ); @@ -4375,22 +4429,27 @@ class Parser $ig->setParser( $this ); $ig->setHideBadImages(); $ig->setAttributes( Sanitizer::validateTagAttributes( $params, 'table' ) ); - $ig->useSkin( $this->mOptions->getSkin() ); + $ig->useSkin( $this->mOptions->getSkin( $this->mTitle ) ); $ig->mRevisionId = $this->mRevisionId; - if( isset( $params['caption'] ) ) { + if ( isset( $params['showfilename'] ) ) { + $ig->setShowFilename( true ); + } else { + $ig->setShowFilename( false ); + } + if ( isset( $params['caption'] ) ) { $caption = $params['caption']; $caption = htmlspecialchars( $caption ); $caption = $this->replaceInternalLinks( $caption ); $ig->setCaptionHtml( $caption ); } - if( isset( $params['perrow'] ) ) { + if ( isset( $params['perrow'] ) ) { $ig->setPerRow( $params['perrow'] ); } - if( isset( $params['widths'] ) ) { + if ( isset( $params['widths'] ) ) { $ig->setWidths( $params['widths'] ); } - if( isset( $params['heights'] ) ) { + if ( isset( $params['heights'] ) ) { $ig->setHeights( $params['heights'] ); } @@ -4407,11 +4466,12 @@ class Parser continue; } - if ( strpos( $matches[0], '%' ) !== false ) + if ( strpos( $matches[0], '%' ) !== false ) { $matches[1] = urldecode( $matches[1] ); - $tp = Title::newFromText( $matches[1]/*, NS_FILE*/ ); + } + $tp = Title::newFromText( $matches[1] ); $nt =& $tp; - if( is_null( $nt ) ) { + if ( is_null( $nt ) ) { # Bogus title. Ignore these so we don't bomb out later. continue; } @@ -4440,7 +4500,7 @@ class Parser $handlerClass = ''; } if ( !isset( $this->mImageParams[$handlerClass] ) ) { - // Initialise static lists + # Initialise static lists static $internalParamNames = array( 'horizAlign' => array( 'left', 'right', 'center', 'none' ), 'vertAlign' => array( 'baseline', 'sub', 'super', 'top', 'text-top', 'middle', @@ -4459,7 +4519,7 @@ class Parser } } - // Add handler params + # Add handler params $paramMap = $internalParamMap; if ( $handler ) { $handlerParamMap = $handler->getParamMap(); @@ -4475,9 +4535,10 @@ class Parser /** * Parse image options text and use it to make an image - * @param Title $title - * @param string $options - * @param LinkHolderArray $holders + * + * @param $title Title + * @param $options String + * @param $holders LinkHolderArray */ function makeImage( $title, $options, $holders = false ) { # Check if the options text is of the form "options|alt text" @@ -4506,7 +4567,7 @@ class Parser # * text-bottom $parts = StringUtils::explode( "|", $options ); - $sk = $this->mOptions->getSkin(); + $sk = $this->mOptions->getSkin( $this->mTitle ); # Give extensions a chance to select the file revision for us $skip = $time = $descQuery = false; @@ -4517,7 +4578,6 @@ class Parser } # Get the file - $imagename = $title->getDBkey(); $file = wfFindFile( $title, array( 'time' => $time ) ); # Get parameter map $handler = $file ? $file->getHandler() : false; @@ -4528,15 +4588,15 @@ class Parser $caption = ''; $params = array( 'frame' => array(), 'handler' => array(), 'horizAlign' => array(), 'vertAlign' => array() ); - foreach( $parts as $part ) { + foreach ( $parts as $part ) { $part = trim( $part ); list( $magicName, $value ) = $mwArray->matchVariableStartToEnd( $part ); $validated = false; - if( isset( $paramMap[$magicName] ) ) { + if ( isset( $paramMap[$magicName] ) ) { list( $type, $paramName ) = $paramMap[$magicName]; - // Special case; width and height come in one variable together - if( $type === 'handler' && $paramName === 'width' ) { + # Special case; width and height come in one variable together + if ( $type === 'handler' && $paramName === 'width' ) { $m = array(); # (bug 13500) In both cases (width/height and width only), # permit trailing "px" for backward compatibility. @@ -4557,7 +4617,7 @@ class Parser $params[$type]['width'] = $width; $validated = true; } - } // else no validation -- bug 13436 + } # else no validation -- bug 13436 } else { if ( $type === 'handler' ) { # Validate handler parameter @@ -4567,9 +4627,9 @@ class Parser switch( $paramName ) { case 'manualthumb': case 'alt': - // @todo Fixme: possibly check validity here for - // manualthumb? downstream behavior seems odd with - // missing manual thumbs. + # @todo Fixme: possibly check validity here for + # manualthumb? downstream behavior seems odd with + # missing manual thumbs. $validated = true; $value = $this->stripAltText( $value, $holders ); break; @@ -4584,6 +4644,9 @@ class Parser if ( preg_match( "/^($prots)$chars+$/", $value, $m ) ) { $paramName = 'link-url'; $this->mOutput->addExternalLink( $value ); + if ( $this->mOptions->getExternalLinkTarget() ) { + $params[$type]['link-target'] = $this->mOptions->getExternalLinkTarget(); + } $validated = true; } } else { @@ -4597,7 +4660,7 @@ class Parser } break; default: - // Most other things appear to be empty or numeric... + # Most other things appear to be empty or numeric... $validated = ( $value === false || is_numeric( trim( $value ) ) ); } } @@ -4670,7 +4733,7 @@ class Parser wfRunHooks( 'ParserMakeImageParams', array( $title, $file, &$params ) ); # Linker does the rest - $ret = $sk->makeImageLink2( $title, $file, $params['frame'], $params['handler'], $time, $descQuery ); + $ret = $sk->makeImageLink2( $title, $file, $params['frame'], $params['handler'], $time, $descQuery, $this->mOptions->getThumbSize() ); # Give the handler a chance to modify the parser object if ( $handler ) { @@ -4705,15 +4768,17 @@ class Parser */ function disableCache() { wfDebug( "Parser output marked as uncacheable.\n" ); - $this->mOutput->mCacheTime = -1; + $this->mOutput->setCacheTime( -1 ); // old style, for compatibility + $this->mOutput->updateCacheExpiry( 0 ); // new style, for consistency } - /**#@+ + /** * 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 + * + * @param $text String + * @param $frame PPFrame + * @return String * @private */ function attributeStripCallback( &$text, $frame = false ) { @@ -4722,22 +4787,12 @@ class Parser 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 ) ); } - /**#@-*/ - + function getTags() { + return array_merge( array_keys( $this->mTransparentTagHooks ), array_keys( $this->mTagHooks ) ); + } /** * Break wikitext input into sections, and either pull or replace @@ -4745,8 +4800,8 @@ class Parser * * External callers should use the getSection and replaceSection methods. * - * @param string $text Page wikitext - * @param string $section A section identifier string of the form: + * @param $text String: Page wikitext + * @param $section String: a section identifier string of the form: * <flag1> - <flag2> - ... - <section number> * * Currently the only recognised flag is "T", which means the target section number @@ -4759,21 +4814,21 @@ class Parser * 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. + * @param $mode String: one of "get" or "replace" + * @param $newText String: 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 ); + $this->clearState(); + $this->setTitle( $wgTitle ); # not generally used but removes an ugly failure mode + $this->setOutputType( self::OT_PLAIN ); $outText = ''; $frame = $this->getPreprocessor()->newFrame(); - // Process section extraction flags + # Process section extraction flags $flags = 0; $sectionParts = explode( '-', $section ); $sectionIndex = array_pop( $sectionParts ); @@ -4782,23 +4837,23 @@ class Parser $flags |= self::PTD_FOR_INCLUSION; } } - // Preprocess the text + # Preprocess the text $root = $this->preprocessToDom( $text, $flags ); - // <h> nodes indicate section breaks - // They can only occur at the top level, so we can find them by iterating the root's children + # <h> 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 + # Find the target section if ( $sectionIndex == 0 ) { - // Section zero doesn't nest, level=big + # Section zero doesn't nest, level=big $targetLevel = 1000; } else { - while ( $node ) { - if ( $node->getName() === 'h' ) { - $bits = $node->splitHeading(); + while ( $node ) { + if ( $node->getName() === 'h' ) { + $bits = $node->splitHeading(); if ( $bits['i'] == $sectionIndex ) { - $targetLevel = $bits['level']; + $targetLevel = $bits['level']; break; } } @@ -4810,7 +4865,7 @@ class Parser } if ( !$node ) { - // Not found + # Not found if ( $mode === 'get' ) { return $newText; } else { @@ -4818,7 +4873,7 @@ class Parser } } - // Find the end of the section, including nested sections + # Find the end of the section, including nested sections do { if ( $node->getName() === 'h' ) { $bits = $node->splitHeading(); @@ -4833,13 +4888,13 @@ class Parser $node = $node->getNextSibling(); } while ( $node ); - // Write out the remainder (in replace mode only) + # Write out the remainder (in replace mode only) if ( $mode === 'replace' ) { - // Output the replacement text - // Add two newlines on -- trailing whitespace in $newText is conventionally - // stripped by the editor, so we need both newlines to restore the paragraph gap - // Only add trailing whitespace if there is newText - if($newText != "") { + # 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 + # Only add trailing whitespace if there is newText + if ( $newText != "" ) { $outText .= $newText . "\n\n"; } @@ -4850,7 +4905,7 @@ class Parser } if ( is_string( $outText ) ) { - // Re-insert stripped tags + # Re-insert stripped tags $outText = rtrim( $this->mStripState->unstripBoth( $outText ) ); } @@ -4864,20 +4919,38 @@ class Parser * * 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 + * @param $text String: text to look in + * @param $section String: section identifier + * @param $deftext String: 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 ); } + /** + * This function returns $oldtext after the content of the section + * specified by $section has been replaced with $text. + * + * @param $text String: former text of the article + * @param $section Numeric: section identifier + * @param $text String: replacing text + * #return String: modified text + */ public function replaceSection( $oldtext, $section, $text ) { return $this->extractSections( $oldtext, $section, "replace", $text ); } /** + * Get the ID of the revision we are parsing + * + * @return Mixed: integer or null + */ + function getRevisionId() { + return $this->mRevisionId; + } + + /** * Get the timestamp associated with the current revision, adjusted for * the default server-local timestamp */ @@ -4889,18 +4962,18 @@ class Parser $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. + # 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. + # 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__ ); @@ -4910,11 +4983,13 @@ class Parser /** * Get the name of the user that edited the last revision + * + * @return String: user name */ function getRevisionUser() { - // if this template is subst: the revision id will be blank, - // so just use the current user's name - if( $this->mRevisionId ) { + # if this template is subst: the revision id will be blank, + # so just use the current user's name + if ( $this->mRevisionId ) { $revision = Revision::newFromId( $this->mRevisionId ); $revuser = $revision->getUserText(); } else { @@ -4931,23 +5006,24 @@ class Parser */ public function setDefaultSort( $sort ) { $this->mDefaultSort = $sort; + $this->mOutput->setProperty( 'defaultsort', $sort ); } /** * Accessor for $mDefaultSort - * Will use the title/prefixed title if none is set + * Will use the empty string if none is set. + * + * This value is treated as a prefix, so the + * empty string is equivalent to sorting by + * page name. * * @return string */ public function getDefaultSort() { - global $wgCategoryPrefixedDefaultSortkey; - if( $this->mDefaultSort !== false ) { + if ( $this->mDefaultSort !== false ) { return $this->mDefaultSort; - } elseif ($this->mTitle->getNamespace() == NS_CATEGORY || - !$wgCategoryPrefixedDefaultSortkey) { - return $this->mTitle->getText(); } else { - return $this->mTitle->getPrefixedText(); + return ''; } } @@ -4969,19 +5045,23 @@ class Parser 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 ); + $text = Sanitizer::normalizeSectionNameWhitespace( $text ); + return '#' . Sanitizer::escapeId( $text, 'noninitial' ); + } + + /** + * Same as guessSectionNameFromWikiText(), but produces legacy anchors + * instead. For use in redirects, since IE6 interprets Redirect: headers + * as something other than UTF-8 (apparently?), resulting in breakage. + * + * @param $text String: The section name + * @return string An anchor + */ + public function guessLegacySectionNameFromWikiText( $text ) { + # Strip out wikitext links(they break the anchor) + $text = $this->stripSectionName( $text ); + $text = Sanitizer::normalizeSectionNameWhitespace( $text ); + return '#' . Sanitizer::escapeId( $text, array( 'noninitial', 'legacy' ) ); } /** @@ -4994,42 +5074,38 @@ class Parser * 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 + * @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); + $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); + $text = preg_replace( '/\[(?:' . wfUrlProtocols() . ')([^ ]+?) ([^[]+)\]/', '$2', $text ); # Parse wikitext quotes (italics & bold) - $text = $this->doQuotes($text); + $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 ) { + function testSrvus( $text, $title, ParserOptions $options, $outputType = self::OT_HTML ) { + $this->mOptions = $options; $this->clearState(); - if ( ! ( $title instanceof Title ) ) { + if ( !$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 ); @@ -5039,14 +5115,14 @@ class Parser function testPst( $text, $title, $options ) { global $wgUser; - if ( ! ( $title instanceof Title ) ) { + if ( !$title instanceof Title ) { $title = Title::newFromText( $title ); } return $this->preSaveTransform( $text, $title, $wgUser, $options ); } function testPreprocess( $text, $title, $options ) { - if ( ! ( $title instanceof Title ) ) { + if ( !$title instanceof Title ) { $title = Title::newFromText( $title ); } return $this->testSrvus( $text, $title, $options, self::OT_PREPROCESS ); @@ -5080,11 +5156,13 @@ class Parser $data = array(); $data['text'] = $text; - // First, find all strip markers, and store their - // data in an array. + # First, find all strip markers, and store their + # data in an array. $stripState = new StripState; $pos = 0; - while( ( $start_pos = strpos( $text, $this->mUniqPrefix, $pos ) ) && ( $end_pos = strpos( $text, self::MARKER_SUFFIX, $pos ) ) ) { + while ( ( $start_pos = strpos( $text, $this->mUniqPrefix, $pos ) ) + && ( $end_pos = strpos( $text, self::MARKER_SUFFIX, $pos ) ) ) + { $end_pos += strlen( self::MARKER_SUFFIX ); $marker = substr( $text, $start_pos, $end_pos-$start_pos ); @@ -5103,17 +5181,17 @@ class Parser } $data['stripstate'] = $stripState; - // Now, find all of our links, and store THEIR - // data in an array! :) + # Now, find all of our links, and store THEIR + # data in an array! :) $links = array( 'internal' => array(), 'interwiki' => array() ); $pos = 0; - // Internal links - while( ( $start_pos = strpos( $text, '<!--LINK ', $pos ) ) ) { + # Internal links + while ( ( $start_pos = strpos( $text, '<!--LINK ', $pos ) ) ) { list( $ns, $trail ) = explode( ':', substr( $text, $start_pos + strlen( '<!--LINK ' ) ), 2 ); - $ns = trim($ns); - if (empty( $links['internal'][$ns] )) { + $ns = trim( $ns ); + if ( empty( $links['internal'][$ns] ) ) { $links['internal'][$ns] = array(); } @@ -5124,8 +5202,8 @@ class Parser $pos = 0; - // Interwiki links - while( ( $start_pos = strpos( $text, '<!--IWLINK ', $pos ) ) ) { + # Interwiki links + while ( ( $start_pos = strpos( $text, '<!--IWLINK ', $pos ) ) ) { $data = substr( $text, $start_pos ); $key = trim( substr( $data, 0, strpos( $data, '-->' ) ) ); $links['interwiki'][] = $this->mLinkHolders->interwiki[$key]; @@ -5137,22 +5215,29 @@ class Parser return $data; } - function unserialiseHalfParsedText( $data, $intPrefix = null /* Unique identifying prefix */ ) { - if (!$intPrefix) - $intPrefix = $this->getRandomString(); + /** + * TODO: document + * @param $data Array + * @param $intPrefix String unique identifying prefix + * @return String + */ + function unserialiseHalfParsedText( $data, $intPrefix = null ) { + if ( !$intPrefix ) { + $intPrefix = self::getRandomString(); + } - // First, extract the strip state. + # First, extract the strip state. $stripState = $data['stripstate']; $this->mStripState->general->merge( $stripState->general ); $this->mStripState->nowiki->merge( $stripState->nowiki ); - // Now, extract the text, and renumber links + # Now, extract the text, and renumber links $text = $data['text']; $links = $data['linkholder']; - // Internal... - foreach( $links['internal'] as $ns => $nsLinks ) { - foreach( $nsLinks as $key => $entry ) { + # Internal... + foreach ( $links['internal'] as $ns => $nsLinks ) { + foreach ( $nsLinks as $key => $entry ) { $newKey = $intPrefix . '-' . $key; $this->mLinkHolders->internals[$ns][$newKey] = $entry; @@ -5160,15 +5245,15 @@ class Parser } } - // Interwiki... - foreach( $links['interwiki'] as $key => $entry ) { + # Interwiki... + foreach ( $links['interwiki'] as $key => $entry ) { $newKey = "$intPrefix-$key"; $this->mLinkHolders->interwikis[$newKey] = $entry; $text = str_replace( "<!--IWLINK $key-->", "<!--IWLINK $newKey-->", $text ); } - // Should be good to go. + # Should be good to go. return $text; } } diff --git a/includes/parser/ParserCache.php b/includes/parser/ParserCache.php index 524d6be5..1e028ae5 100644 --- a/includes/parser/ParserCache.php +++ b/includes/parser/ParserCache.php @@ -1,9 +1,18 @@ <?php /** + * Cache for outputs of the PHP parser + * + * @file + */ + +/** * @ingroup Cache Parser * @todo document */ class ParserCache { + private $mMemc; + const try116cache = false; /* Only useful $wgParserCacheExpireTime after updating to 1.17 */ + /** * Get an instance of this object */ @@ -20,49 +29,119 @@ class ParserCache { * Setup a cache pathway with a given back-end storage mechanism. * May be a memcached client or a BagOStuff derivative. * - * @param object $memCached + * @param $memCached Object */ function __construct( $memCached ) { + if ( !$memCached ) { + throw new MWException( "Tried to create a ParserCache with an invalid memcached" ); + } $this->mMemc = $memCached; } - function getKey( $article, $popts ) { + protected function getParserOutputKey( $article, $hash ) { global $wgRequest; - if( $popts instanceof User ) // It used to be getKey( &$article, &$user ) - $popts = ParserOptions::newFromUser( $popts ); - - $user = $popts->mUser; - $printable = ( $popts->getIsPrintable() ) ? '!printable=1' : ''; - $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 = ''; - } + // idhash seem to mean 'page id' + 'rendering hash' (r3710) $pageid = $article->getID(); $renderkey = (int)($wgRequest->getVal('action') == 'render'); - $key = wfMemcKey( 'pcache', 'idhash', "{$pageid}-{$renderkey}!{$hash}{$edit}{$printable}" ); + + $key = wfMemcKey( 'pcache', 'idhash', "{$pageid}-{$renderkey}!{$hash}" ); return $key; } + protected function getOptionsKey( $article ) { + $pageid = $article->getID(); + return wfMemcKey( 'pcache', 'idoptions', "{$pageid}" ); + } + + /** + * Provides an E-Tag suitable for the whole page. Note that $article + * is just the main wikitext. The E-Tag has to be unique to the whole + * page, even if the article itself is the same, so it uses the + * complete set of user options. We don't want to use the preference + * of a different user on a message just because it wasn't used in + * $article. For example give a Chinese interface to a user with + * English preferences. That's why we take into account *all* user + * options. (r70809 CR) + */ function getETag( $article, $popts ) { - return 'W/"' . $this->getKey($article, $popts) . "--" . $article->mTouched. '"'; + return 'W/"' . $this->getParserOutputKey( $article, + $popts->optionsHash( ParserOptions::legacyOptions() ) ) . + "--" . $article->getTouched() . '"'; } - function getDirty( $article, $popts ) { - $key = $this->getKey( $article, $popts ); - wfDebug( "Trying parser cache $key\n" ); - $value = $this->mMemc->get( $key ); + /** + * Retrieve the ParserOutput from ParserCache, even if it's outdated. + */ + public function getDirty( $article, $popts ) { + $value = $this->get( $article, $popts, true ); return is_object( $value ) ? $value : false; } - function get( $article, $popts ) { + /** + * Used to provide a unique id for the PoolCounter. + * It would be preferable to have this code in get() + * instead of having Article looking in our internals. + */ + public function getKey( $article, $popts, $useOutdated = true ) { + global $wgCacheEpoch; + + if( $popts instanceof User ) { + wfWarn( "Use of outdated prototype ParserCache::getKey( &\$article, &\$user )\n" ); + $popts = ParserOptions::newFromUser( $popts ); + } + + // Determine the options which affect this article + $optionsKey = $this->mMemc->get( $this->getOptionsKey( $article ) ); + if ( $optionsKey != false ) { + if ( !$useOutdated && $optionsKey->expired( $article->getTouched() ) ) { + wfIncrStats( "pcache_miss_expired" ); + $cacheTime = $optionsKey->getCacheTime(); + wfDebug( "Parser options key expired, touched " . $article->getTouched() . ", epoch $wgCacheEpoch, cached $cacheTime\n" ); + return false; + } + + $usedOptions = $optionsKey->mUsedOptions; + wfDebug( "Parser cache options found.\n" ); + } else { + if ( !$useOutdated && !self::try116cache ) { + return false; + } + $usedOptions = ParserOptions::legacyOptions(); + } + + return $this->getParserOutputKey( $article, $popts->optionsHash( $usedOptions ) ); + } + + /** + * Retrieve the ParserOutput from ParserCache. + * false if not found or outdated. + */ + public function get( $article, $popts, $useOutdated = false ) { global $wgCacheEpoch; wfProfileIn( __METHOD__ ); - $value = $this->getDirty( $article, $popts ); + $canCache = $article->checkTouched(); + if ( !$canCache ) { + // It's a redirect now + wfProfileOut( __METHOD__ ); + return false; + } + + $touched = $article->getTouched(); + + $parserOutputKey = $this->getKey( $article, $popts, $useOutdated ); + if ( $parserOutputKey === false ) { + wfProfileOut( __METHOD__ ); + return false; + } + + $value = $this->mMemc->get( $parserOutputKey ); + if ( self::try116cache && !$value && strpos( $value, '*' ) !== -1 ) { + wfDebug( "New format parser cache miss.\n" ); + $parserOutputKey = $this->getParserOutputKey( $article, $popts->optionsHash( ParserOptions::legacyOptions() ) ); + $value = $this->mMemc->get( $parserOutputKey ); + } if ( !$value ) { wfDebug( "Parser cache miss.\n" ); wfIncrStats( "pcache_miss_absent" ); @@ -71,18 +150,11 @@ class ParserCache { } wfDebug( "Found.\n" ); - # Invalid 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 $cacheTime\n" ); - } + + if ( !$useOutdated && $value->expired( $touched ) ) { + wfIncrStats( "pcache_miss_expired" ); + $cacheTime = $value->getCacheTime(); + wfDebug( "ParserOutput key expired, touched $touched, epoch $wgCacheEpoch, cached $cacheTime\n" ); $value = false; } else { if ( isset( $value->mTimestamp ) ) { @@ -95,31 +167,37 @@ class ParserCache { return $value; } - function save( $parserOutput, $article, $popts ){ - global $wgParserCacheExpireTime; - $key = $this->getKey( $article, $popts ); - if( $parserOutput->getCacheTime() != -1 ) { + public function save( $parserOutput, $article, $popts ) { + $expire = $parserOutput->getCacheExpiry(); + if( $expire > 0 ) { $now = wfTimestampNow(); + + $optionsKey = new CacheTime; + $optionsKey->mUsedOptions = $parserOutput->getUsedOptions(); + $optionsKey->updateCacheExpiry( $expire ); + + $optionsKey->setCacheTime( $now ); $parserOutput->setCacheTime( $now ); + $optionsKey->setContainsOldMagic( $parserOutput->containsOldMagic() ); + + $parserOutputKey = $this->getParserOutputKey( $article, $popts->optionsHash( $optionsKey->mUsedOptions ) ); + // Save the timestamp so that we don't have to load the revision row on view $parserOutput->mTimestamp = $article->getTimestamp(); - $parserOutput->mText .= "\n<!-- Saved in parser cache with key $key and timestamp $now -->\n"; - wfDebug( "Saved in parser cache with key $key and timestamp $now\n" ); + $parserOutput->mText .= "\n<!-- Saved in parser cache with key $parserOutputKey and timestamp $now -->\n"; + wfDebug( "Saved in parser cache with key $parserOutputKey and timestamp $now\n" ); - if( $parserOutput->containsOldMagic() ){ - $expire = 3600; # 1 hour - } else { - $expire = $wgParserCacheExpireTime; - } - $this->mMemc->set( $key, $parserOutput, $expire ); + // Save the parser output + $this->mMemc->set( $parserOutputKey, $parserOutput, $expire ); + // ...and its pointer + $this->mMemc->set( $this->getOptionsKey( $article ), $optionsKey, $expire ); } else { wfDebug( "Parser output was marked as uncacheable and has not been saved.\n" ); } } - } diff --git a/includes/parser/ParserOptions.php b/includes/parser/ParserOptions.php index 985bba28..9d8b3e4f 100644 --- a/includes/parser/ParserOptions.php +++ b/includes/parser/ParserOptions.php @@ -1,5 +1,11 @@ <?php - +/** + * Options for the PHP parser + * + * @file + * @ingroup Parser + */ + /** * Set options of the Parser * @todo document @@ -29,43 +35,61 @@ class ParserOptions { var $mEnableLimitReport; # Enable limit report in an HTML comment on output var $mTimestamp; # Timestamp used for {{CURRENTDAY}} etc. var $mExternalLinkTarget; # Target attribute for external links + var $mMath; # User math preference (as integer) + var $mUserLang; # Language code of the User language. + var $mThumbSize; # Thumb size preferred by the user. + var $mCleanSignatures; # var $mUser; # Stored user object, just used to initialise the skin var $mIsPreview; # Parsing the page for a "preview" operation var $mIsSectionPreview; # Parsing the page for a "preview" operation on a single section var $mIsPrintable; # Parsing the printable version of the page + var $mExtraKey = ''; # Extra key that should be present in the caching key. + + protected $onAccessCallback = null; + function getUseDynamicDates() { return $this->mUseDynamicDates; } function getInterwikiMagic() { return $this->mInterwikiMagic; } function getAllowExternalImages() { return $this->mAllowExternalImages; } function getAllowExternalImagesFrom() { return $this->mAllowExternalImagesFrom; } function getEnableImageWhitelist() { return $this->mEnableImageWhitelist; } - function getEditSection() { return $this->mEditSection; } - function getNumberHeadings() { return $this->mNumberHeadings; } + function getEditSection() { $this->optionUsed('editsection'); + return $this->mEditSection; } + function getNumberHeadings() { $this->optionUsed('numberheadings'); + return $this->mNumberHeadings; } function getAllowSpecialInclusion() { return $this->mAllowSpecialInclusion; } function getTidy() { return $this->mTidy; } function getInterfaceMessage() { return $this->mInterfaceMessage; } function getTargetLanguage() { return $this->mTargetLanguage; } function getMaxIncludeSize() { return $this->mMaxIncludeSize; } function getMaxPPNodeCount() { return $this->mMaxPPNodeCount; } + function getMaxPPExpandDepth() { return $this->mMaxPPExpandDepth; } function getMaxTemplateDepth() { return $this->mMaxTemplateDepth; } function getRemoveComments() { return $this->mRemoveComments; } function getTemplateCallback() { return $this->mTemplateCallback; } function getEnableLimitReport() { return $this->mEnableLimitReport; } function getCleanSignatures() { return $this->mCleanSignatures; } function getExternalLinkTarget() { return $this->mExternalLinkTarget; } + function getMath() { $this->optionUsed('math'); + return $this->mMath; } + function getThumbSize() { $this->optionUsed('thumbsize'); + return $this->mThumbSize; } + function getIsPreview() { return $this->mIsPreview; } function getIsSectionPreview() { return $this->mIsSectionPreview; } - function getIsPrintable() { return $this->mIsPrintable; } + function getIsPrintable() { $this->optionUsed('printable'); + return $this->mIsPrintable; } - function getSkin() { + function getSkin( $title = null ) { if ( !isset( $this->mSkin ) ) { - $this->mSkin = $this->mUser->getSkin(); + $this->mSkin = $this->mUser->getSkin( $title ); } return $this->mSkin; } function getDateFormat() { + $this->optionUsed('dateformat'); if ( !isset( $this->mDateFormat ) ) { $this->mDateFormat = $this->mUser->getDatePreference(); } @@ -79,6 +103,16 @@ class ParserOptions { return $this->mTimestamp; } + /** + * You shouldn't use this. Really. $parser->getFunctionLang() is all you need. + * Using this fragments the cache and is discouraged. Yes, {{int: }} uses this, + * producing inconsistent tables (Bug 14404). + */ + function getUserLang() { + $this->optionUsed('userlang'); + return $this->mUserLang; + } + function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates, $x ); } function setInterwikiMagic( $x ) { return wfSetVar( $this->mInterwikiMagic, $x ); } function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages, $x ); } @@ -101,17 +135,30 @@ class ParserOptions { function setTimestamp( $x ) { return wfSetVar( $this->mTimestamp, $x ); } function setCleanSignatures( $x ) { return wfSetVar( $this->mCleanSignatures, $x ); } function setExternalLinkTarget( $x ) { return wfSetVar( $this->mExternalLinkTarget, $x ); } + function setMath( $x ) { return wfSetVar( $this->mMath, $x ); } + function setUserLang( $x ) { return wfSetVar( $this->mUserLang, $x ); } + function setThumbSize( $x ) { return wfSetVar( $this->mThumbSize, $x ); } + function setIsPreview( $x ) { return wfSetVar( $this->mIsPreview, $x ); } function setIsSectionPreview( $x ) { return wfSetVar( $this->mIsSectionPreview, $x ); } function setIsPrintable( $x ) { return wfSetVar( $this->mIsPrintable, $x ); } + /** + * Extra key that should be present in the parser cache key. + */ + function addExtraKey( $key ) { + $this->mExtraKey .= '!' . $key; + } + function __construct( $user = null ) { $this->initialiseFromUser( $user ); } /** * Get parser options - * @static + * + * @param $user User object + * @return ParserOptions object */ static function newFromUser( $user ) { return new ParserOptions( $user ); @@ -122,7 +169,7 @@ class ParserOptions { global $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages; global $wgAllowExternalImagesFrom, $wgEnableImageWhitelist, $wgAllowSpecialInclusion, $wgMaxArticleSize; global $wgMaxPPNodeCount, $wgMaxTemplateDepth, $wgMaxPPExpandDepth, $wgCleanSignatures; - global $wgExternalLinkTarget; + global $wgExternalLinkTarget, $wgLang; wfProfileIn( __METHOD__ ); @@ -161,9 +208,131 @@ class ParserOptions { $this->mEnableLimitReport = false; $this->mCleanSignatures = $wgCleanSignatures; $this->mExternalLinkTarget = $wgExternalLinkTarget; + $this->mMath = $user->getOption( 'math' ); + $this->mUserLang = $wgLang->getCode(); + $this->mThumbSize = $user->getOption( 'thumbsize' ); + $this->mIsPreview = false; $this->mIsSectionPreview = false; + $this->mIsPrintable = false; wfProfileOut( __METHOD__ ); } + + /** + * Registers a callback for tracking which ParserOptions which are used. + * This is a private API with the parser. + */ + function registerWatcher( $callback ) { + $this->onAccessCallback = $callback; + } + + /** + * Called when an option is accessed. + */ + protected function optionUsed( $optionName ) { + if ( $this->onAccessCallback ) { + call_user_func( $this->onAccessCallback, $optionName ); + } + } + + /** + * Returns the full array of options that would have been used by + * in 1.16. + * Used to get the old parser cache entries when available. + */ + public static function legacyOptions() { + global $wgUseDynamicDates; + $legacyOpts = array( 'math', 'stubthreshold', 'numberheadings', 'userlang', 'thumbsize', 'editsection', 'printable' ); + if ( $wgUseDynamicDates ) { + $legacyOpts[] = 'dateformat'; + } + return $legacyOpts; + } + + /** + * Generate a hash string with the values set on these ParserOptions + * for the keys given in the array. + * This will be used as part of the hash key for the parser cache, + * so users sharign the options with vary for the same page share + * the same cached data safely. + * + * Replaces User::getPageRenderingHash() + * + * Extensions which require it should install 'PageRenderingHash' hook, + * which will give them a chance to modify this key based on their own + * settings. + * + * @since 1.17 + * @return \string Page rendering hash + */ + public function optionsHash( $forOptions ) { + global $wgContLang, $wgRenderHashAppend; + + $confstr = ''; + + if ( in_array( 'math', $forOptions ) ) + $confstr .= $this->mMath; + else + $confstr .= '*'; + + + // Space assigned for the stubthreshold but unused + // since it disables the parser cache, its value will always + // be 0 when this function is called by parsercache. + // The conditional is here to avoid a confusing 0 + if ( true || in_array( 'stubthreshold', $forOptions ) ) + $confstr .= '!0' ; + else + $confstr .= '!*' ; + + if ( in_array( 'dateformat', $forOptions ) ) + $confstr .= '!' . $this->getDateFormat(); + + if ( in_array( 'numberheadings', $forOptions ) ) + $confstr .= '!' . ( $this->mNumberHeadings ? '1' : '' ); + else + $confstr .= '!*'; + + if ( in_array( 'userlang', $forOptions ) ) + $confstr .= '!' . $this->mUserLang; + else + $confstr .= '!*'; + + if ( in_array( 'thumbsize', $forOptions ) ) + $confstr .= '!' . $this->mThumbSize; + else + $confstr .= '!*'; + + // add in language specific options, if any + // FIXME: This is just a way of retrieving the url/user preferred variant + $confstr .= $wgContLang->getExtraHashOptions(); + + // Since the skin could be overloading link(), it should be + // included here but in practice, none of our skins do that. + // $confstr .= "!" . $this->mSkin->getSkinName(); + + $confstr .= $wgRenderHashAppend; + + if ( !in_array( 'editsection', $forOptions ) ) { + $confstr .= '!*'; + } elseif ( !$this->mEditSection ) { + $confstr .= '!edit=0'; + } + + if ( $this->mIsPrintable && in_array( 'printable', $forOptions ) ) + $confstr .= '!printable=1'; + + if ( $this->mExtraKey != '' ) + $confstr .= $this->mExtraKey; + + // Give a chance for extensions to modify the hash, if they have + // extra options or other effects on the parser cache. + wfRunHooks( 'PageRenderingHash', array( &$confstr ) ); + + // Make it a valid memcached key fragment + $confstr = str_replace( ' ', '_', $confstr ); + + return $confstr; + } } diff --git a/includes/parser/ParserOutput.php b/includes/parser/ParserOutput.php index ea5840e6..1e4765db 100644 --- a/includes/parser/ParserOutput.php +++ b/includes/parser/ParserOutput.php @@ -1,34 +1,131 @@ <?php /** + * Output of the PHP parser + * + * @file + * @ingroup Parser + */ + +/** * @todo document * @ingroup Parser */ -class ParserOutput + +class CacheTime { + var $mVersion = Parser::VERSION, # Compatibility check + $mCacheTime = '', # Time when this object was generated, or -1 for uncacheable. Used in ParserCache. + $mCacheExpiry = null, # Seconds after which the object should expire, use 0 for uncachable. Used in ParserCache. + $mContainsOldMagic; # Boolean variable indicating if the input contained variables like {{CURRENTDAY}} + + function getCacheTime() { return $this->mCacheTime; } + + function containsOldMagic() { return $this->mContainsOldMagic; } + function setContainsOldMagic( $com ) { return wfSetVar( $this->mContainsOldMagic, $com ); } + + /** + * setCacheTime() sets the timestamp expressing when the page has been rendered. + * This doesn not control expiry, see updateCacheExpiry() for that! + */ + function setCacheTime( $t ) { return wfSetVar( $this->mCacheTime, $t ); } + + + /** + * Sets the number of seconds after which this object should expire. + * This value is used with the ParserCache. + * If called with a value greater than the value provided at any previous call, + * the new call has no effect. The value returned by getCacheExpiry is smaller + * or equal to the smallest number that was provided as an argument to + * updateCacheExpiry(). + */ + function updateCacheExpiry( $seconds ) { + $seconds = (int)$seconds; + + if ( $this->mCacheExpiry === null || $this->mCacheExpiry > $seconds ) + $this->mCacheExpiry = $seconds; + + // hack: set old-style marker for uncacheable entries. + if ( $this->mCacheExpiry !== null && $this->mCacheExpiry <= 0 ) + $this->mCacheTime = -1; + } + + /** + * Returns the number of seconds after which this object should expire. + * This method is used by ParserCache to determine how long the ParserOutput can be cached. + * The timestamp of expiry can be calculated by adding getCacheExpiry() to getCacheTime(). + * The value returned by getCacheExpiry is smaller or equal to the smallest number + * that was provided to a call of updateCacheExpiry(), and smaller or equal to the + * value of $wgParserCacheExpireTime. + */ + function getCacheExpiry() { + global $wgParserCacheExpireTime; + + if ( $this->mCacheTime < 0 ) return 0; // old-style marker for "not cachable" + + $expire = $this->mCacheExpiry; + + if ( $expire === null ) + $expire = $wgParserCacheExpireTime; + else + $expire = min( $expire, $wgParserCacheExpireTime ); + + if( $this->containsOldMagic() ) { //compatibility hack + $expire = min( $expire, 3600 ); # 1 hour + } + + if ( $expire <= 0 ) return 0; // not cachable + else return $expire; + } + + + function isCacheable() { + return $this->getCacheExpiry() > 0; + } + + /** + * 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 $touched String: the affected article's last touched timestamp + * @return Boolean + */ + public function expired( $touched ) { + global $wgCacheEpoch; + return !$this->isCacheable() || // parser says it's uncacheable + $this->getCacheTime() < $touched || + $this->getCacheTime() <= $wgCacheEpoch || + $this->getCacheTime() < wfTimestamp( TS_MW, time() - $this->getCacheExpiry() ) || // expiry period has passed + !isset( $this->mVersion ) || + version_compare( $this->mVersion, Parser::VERSION, "lt" ); + } +} + +class ParserOutput extends CacheTime { var $mText, # The output text $mLanguageLinks, # List of the full text of language links, in the order they appear $mCategories, # Map of category names to sort keys - $mContainsOldMagic, # Boolean variable indicating if the input contained variables like {{CURRENTDAY}} $mTitleText, # title text of the chosen language variant - $mCacheTime = '', # Time when this object was generated, or -1 for uncacheable. Used in ParserCache. - $mVersion = Parser::VERSION, # Compatibility check $mLinks = array(), # 2-D map of NS/DBK to ID for the links in the document. ID=zero for broken. $mTemplates = array(), # 2-D map of NS/DBK to ID for the template references. ID=zero for broken. $mTemplateIds = array(), # 2-D map of NS/DBK to rev ID for the template references. ID=zero for broken. $mImages = array(), # DB keys of the images used, in the array key only $mExternalLinks = array(), # External link URLs, in the key only + $mInterwikiLinks = array(), # 2-D map of prefix/DBK (in keys only) for the inline interwiki links in the document. $mNewSection = false, # Show a new section link? $mHideNewSection = false, # Hide the new section link? $mNoGallery = false, # No gallery on category page? (__NOGALLERY__) $mHeadItems = array(), # Items to put in the <head> section + $mModules = array(), # Modules to be loaded by the resource loader $mOutputHooks = array(), # Hook tags as per $wgParserOutputHooks $mWarnings = array(), # Warning text to be returned to the user. Wikitext formatted, in the key only $mSections = array(), # Table of contents $mProperties = array(), # Name/value pairs to be cached in the DB $mTOCHTML = ''; # HTML of the TOC private $mIndexPolicy = ''; # 'index' or 'noindex'? Any other value will result in no change. + private $mAccessedOptions = null; # List of ParserOptions (stored in the keys) - function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(), + function __construct( $text = '', $languageLinks = array(), $categoryLinks = array(), $containsOldMagic = false, $titletext = '' ) { $this->mText = $text; @@ -40,9 +137,9 @@ class ParserOutput function getText() { return $this->mText; } function &getLanguageLinks() { return $this->mLanguageLinks; } + function getInterwikiLinks() { return $this->mInterwikiLinks; } 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; } @@ -51,18 +148,17 @@ class ParserOutput function &getExternalLinks() { return $this->mExternalLinks; } function getNoGallery() { return $this->mNoGallery; } function getHeadItems() { return $this->mHeadItems; } + function getModules() { return $this->mModules; } function getSubtitle() { return $this->mSubtitle; } function getOutputHooks() { return (array)$this->mOutputHooks; } function getWarnings() { return array_keys( $this->mWarnings ); } function getIndexPolicy() { return $this->mIndexPolicy; } function getTOCHTML() { return $this->mTOCHTML; } - 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 setIndexPolicy( $policy ) { return wfSetVar( $this->mIndexPolicy, $policy ); } @@ -96,9 +192,16 @@ class ParserOutput $this->mExternalLinks[$url] = 1; } + /** + * Record a local or interwiki inline link for saving in future link tables. + * + * @param $title Title object + * @param $id Mixed: optional known page_id so we can skip the lookup + */ function addLink( $title, $id = null ) { if ( $title->isExternal() ) { // Don't record interwikis in pagelinks + $this->addInterwikiLink( $title ); return; } $ns = $title->getNamespace(); @@ -139,23 +242,20 @@ class ParserOutput } $this->mTemplateIds[$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 + * @param $title Title object, must be an interwiki link + * @throws MWException if given invalid input */ - 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" ); + function addInterwikiLink( $title ) { + $prefix = $title->getInterwiki(); + if( $prefix == '' ) { + throw new MWException( 'Non-interwiki link passed, internal parser error.' ); + } + if (!isset($this->mInterwikiLinks[$prefix])) { + $this->mInterwikiLinks[$prefix] = array(); + } + $this->mInterwikiLinks[$prefix][$title->getDBkey()] = 1; } /** @@ -170,22 +270,27 @@ class ParserOutput $this->mHeadItems[] = $section; } } + + function addModules( $modules ) { + $this->mModules = array_merge( $this->mModules, (array) $modules ); + } /** * 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 + * @param $text String: desired title text */ public function setDisplayTitle( $text ) { $this->setTitleText( $text ); + $this->setProperty( 'displaytitle', $text ); } /** * Get the title to be used for display * - * @return string + * @return String */ public function getDisplayTitle() { $t = $this->getTitleText( ); @@ -223,4 +328,25 @@ class ParserOutput } return $this->mProperties; } + + + /** + * Returns the options from its ParserOptions which have been taken + * into account to produce this output or false if not available. + * @return mixed Array/false + */ + public function getUsedOptions() { + if ( !isset( $this->mAccessedOptions ) ) { + return false; + } + return array_keys( $this->mAccessedOptions ); + } + + /** + * Callback passed by the Parser to the ParserOptions to keep track of which options are used. + * @access private + */ + function recordOption( $option ) { + $this->mAccessedOptions[$option] = true; + } } diff --git a/includes/parser/Parser_DiffTest.php b/includes/parser/Parser_DiffTest.php index 608c883a..c6dd76e5 100644 --- a/includes/parser/Parser_DiffTest.php +++ b/includes/parser/Parser_DiffTest.php @@ -1,4 +1,9 @@ <?php +/** + * Fake parser that output the difference of two different parsers + * + * @file + */ /** * @ingroup Parser @@ -8,14 +13,13 @@ class Parser_DiffTest var $parsers, $conf; var $shortOutput = false; - var $dfUniqPrefix; + var $dtUniqPrefix; function __construct( $conf ) { if ( !isset( $conf['parsers'] ) ) { throw new MWException( __METHOD__ . ': no parsers specified' ); } $this->conf = $conf; - $this->dtUniqPrefix = "\x7fUNIQ" . Parser::getRandomString(); } function init() { @@ -102,14 +106,18 @@ class Parser_DiffTest function setFunctionHook( $id, $callback, $flags = 0 ) { $this->init(); - foreach ( $this->parsers as $i => $parser ) { + foreach ( $this->parsers as $parser ) { $parser->setFunctionHook( $id, $callback, $flags ); } } function onClearState( &$parser ) { // hack marker prefixes to get identical output - $parser->mUniqPrefix = $this->dtUniqPrefix; + if ( !isset( $this->dtUniqPrefix ) ) { + $this->dtUniqPrefix = $parser->uniqPrefix(); + } else { + $parser->mUniqPrefix = $this->dtUniqPrefix; + } return true; } } diff --git a/includes/parser/Parser_LinkHooks.php b/includes/parser/Parser_LinkHooks.php index 2b306933..7c17ce4e 100644 --- a/includes/parser/Parser_LinkHooks.php +++ b/includes/parser/Parser_LinkHooks.php @@ -1,5 +1,11 @@ <?php /** + * Modified version of the PHP parser with hooks for wiki links; experimental + * + * @file + */ + +/** * Parser with LinkHooks experiment * @ingroup Parser */ @@ -78,9 +84,9 @@ class Parser_LinkHooks extends Parser * * @public * - * @param integer|string $ns The Namespace ID or regex pattern if SLH_PATTERN is set - * @param mixed $callback The callback function (and object) to use - * @param integer $flags a combination of the following flags: + * @param $ns Integer or String: the Namespace ID or regex pattern if SLH_PATTERN is set + * @param $callback Mixed: the callback function (and object) to use + * @param $flags Integer: a combination of the following flags: * SLH_PATTERN Use a regex link pattern rather than a namespace * * @return The old callback function for this name, if any @@ -111,8 +117,6 @@ class Parser_LinkHooks extends Parser * @private */ function replaceInternalLinks2( &$s ) { - global $wgContLang; - wfProfileIn( __METHOD__ ); wfProfileIn( __METHOD__.'-setup' ); @@ -128,7 +132,6 @@ class Parser_LinkHooks extends Parser $titleRegex = "/^([{$tc}]+)$/sD"; } - $sk = $this->mOptions->getSkin(); $holders = new LinkHolderArray( $this ); if( is_null( $this->mTitle ) ) { @@ -136,13 +139,7 @@ class Parser_LinkHooks extends Parser wfProfileOut( __METHOD__.'-setup' ); throw new MWException( __METHOD__.": \$this->mTitle is null\n" ); } - $nottalk = !$this->mTitle->isTalkPage(); - - if($wgContLang->hasVariants()) { - $selflink = $wgContLang->convertLinkToAllVariants($this->mTitle->getPrefixedText()); - } else { - $selflink = array($this->mTitle->getPrefixedText()); - } + wfProfileOut( __METHOD__.'-setup' ); $offset = 0; @@ -268,9 +265,10 @@ class Parser_LinkHooks extends Parser if( $return === false ) { # False (no link) was returned, output plain wikitext # Build it again as the hook is allowed to modify $paramText - return isset($paramText) ? "[[$titleText|$paramText]]" : "[[$titleText]]"; + $return = isset($paramText) ? "[[$titleText|$paramText]]" : "[[$titleText]]"; } # Content was returned, return it + wfProfileOut( __METHOD__ ); return $return; } diff --git a/includes/parser/Preprocessor.php b/includes/parser/Preprocessor.php index 9c417d23..c31f37bf 100644 --- a/includes/parser/Preprocessor.php +++ b/includes/parser/Preprocessor.php @@ -1,4 +1,9 @@ <?php +/** + * Interfaces for preprocessors + * + * @file + */ /** * @ingroup Parser @@ -13,6 +18,9 @@ interface Preprocessor { /** Create a new custom frame for programmatic use of parameter replacement as used in some extensions */ function newCustomFrame( $args ); + /** Create a new custom node for programmatic use of parameter replacement as used in some extensions */ + function newPartNodeArray( $values ); + /** Preprocess text to a PPNode */ function preprocessToObj( $text, $flags = 0 ); } diff --git a/includes/parser/Preprocessor_DOM.php b/includes/parser/Preprocessor_DOM.php index 673ac241..2b635f7c 100644 --- a/includes/parser/Preprocessor_DOM.php +++ b/includes/parser/Preprocessor_DOM.php @@ -1,5 +1,11 @@ <?php - +/** + * Preprocessor using PHP's dom extension + * + * @file + * @ingroup Parser + */ + /** * @ingroup Parser */ @@ -29,6 +35,30 @@ class Preprocessor_DOM implements Preprocessor { return new PPCustomFrame_DOM( $this, $args ); } + function newPartNodeArray( $values ) { + //NOTE: DOM manipulation is slower than building & parsing XML! (or so Tim sais) + $xml = ""; + $xml .= "<list>"; + + foreach ( $values as $k => $val ) { + + if ( is_int( $k ) ) { + $xml .= "<part><name index=\"$k\"/><value>" . htmlspecialchars( $val ) ."</value></part>"; + } else { + $xml .= "<part><name>" . htmlspecialchars( $k ) . "</name>=<value>" . htmlspecialchars( $val ) . "</value></part>"; + } + } + + $xml .= "</list>"; + + $dom = new DOMDocument(); + $dom->loadXML( $xml ); + $root = $dom->documentElement; + + $node = new PPNode_DOM( $root->childNodes ); + return $node; + } + function memCheck() { if ( $this->memoryLimit === false ) { return; @@ -45,8 +75,8 @@ class Preprocessor_DOM implements Preprocessor { * 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: + * @param $text String: the text to parse + * @param $flags Integer: bitwise combination of: * Parser::PTD_FOR_INCLUSION Handle <noinclude>/<includeonly> as if the text is being * included. Default is to assume a direct page view. * @@ -443,7 +473,7 @@ class Preprocessor_DOM implements Preprocessor { $count = $piece->count; $equalsLength = strspn( $revText, '=', strlen( $text ) - $searchStart ); if ( $equalsLength > 0 ) { - if ( $i - $equalsLength == $piece->startPos ) { + if ( $searchStart - $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) @@ -481,9 +511,7 @@ class Preprocessor_DOM implements Preprocessor { // 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' ) { + } elseif ( $found == 'open' ) { # count opening brace characters $count = strspn( $text, $curChar, $i ); @@ -506,9 +534,7 @@ class Preprocessor_DOM implements Preprocessor { $accum .= htmlspecialchars( str_repeat( $curChar, $count ) ); } $i += $count; - } - - elseif ( $found == 'close' ) { + } elseif ( $found == 'close' ) { $piece = $stack->top; # lets check if there are enough characters for closing brace $maxCount = $piece->count; @@ -516,7 +542,6 @@ class Preprocessor_DOM implements Preprocessor { # 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 @@ -561,7 +586,7 @@ class Preprocessor_DOM implements Preprocessor { $element = "<$name$attr>"; $element .= "<title>$title</title>"; $argIndex = 1; - foreach ( $parts as $partIndex => $part ) { + foreach ( $parts as $part ) { if ( isset( $part->eqpos ) ) { $argName = substr( $part->out, 0, $part->eqpos ); $argValue = substr( $part->out, $part->eqpos + 1 ); @@ -822,7 +847,7 @@ class PPFrame_DOM implements PPFrame { /** * Construct a new preprocessor frame. - * @param Preprocessor $preprocessor The parent preprocessor + * @param $preprocessor Preprocessor: The parent preprocessor */ function __construct( $preprocessor ) { $this->preprocessor = $preprocessor; @@ -877,12 +902,12 @@ class PPFrame_DOM implements PPFrame { return $root; } - if ( ++$this->parser->mPPNodeCount > $this->parser->mOptions->mMaxPPNodeCount ) + if ( ++$this->parser->mPPNodeCount > $this->parser->mOptions->getMaxPPNodeCount() ) { return '<span class="error">Node-count limit exceeded</span>'; } - if ( $expansionDepth > $this->parser->mOptions->mMaxPPExpandDepth ) { + if ( $expansionDepth > $this->parser->mOptions->getMaxPPExpandDepth() ) { return '<span class="error">Expansion depth limit exceeded</span>'; } wfProfileIn( __METHOD__ ); @@ -932,7 +957,9 @@ class PPFrame_DOM implements PPFrame { $iteratorStack[$level] = false; } - if ( $contextNode instanceof PPNode_DOM ) $contextNode = $contextNode->node; + if ( $contextNode instanceof PPNode_DOM ) { + $contextNode = $contextNode->node; + } $newIterator = false; @@ -951,7 +978,7 @@ class PPFrame_DOM implements PPFrame { $titles = $xpath->query( 'title', $contextNode ); $title = $titles->item( 0 ); $parts = $xpath->query( 'part', $contextNode ); - if ( $flags & self::NO_TEMPLATES ) { + if ( $flags & PPFrame::NO_TEMPLATES ) { $newIterator = $this->virtualBracketedImplode( '{{', '|', '}}', $title, $parts ); } else { $lineStart = $contextNode->getAttribute( 'lineStart' ); @@ -972,7 +999,7 @@ class PPFrame_DOM implements PPFrame { $titles = $xpath->query( 'title', $contextNode ); $title = $titles->item( 0 ); $parts = $xpath->query( 'part', $contextNode ); - if ( $flags & self::NO_ARGS ) { + if ( $flags & PPFrame::NO_ARGS ) { $newIterator = $this->virtualBracketedImplode( '{{{', '|', '}}}', $title, $parts ); } else { $params = array( @@ -990,13 +1017,13 @@ class PPFrame_DOM implements PPFrame { # 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 ) ) + || ( $flags & PPFrame::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 ) ) { + elseif ( $this->parser->ot['wiki'] && ! ( $flags & PPFrame::RECOVER_COMMENTS ) ) { $out .= $this->parser->insertStripItem( $contextNode->textContent ); } # Recover the literal comment in RECOVER_COMMENTS and pre+no-remove @@ -1008,7 +1035,7 @@ class PPFrame_DOM implements PPFrame { # OT_WIKI will only respect <ignore> 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 ) ) { + if ( ( !isset( $this->parent ) && $this->parser->ot['wiki'] ) || ( $flags & PPFrame::NO_IGNORE ) ) { $out .= $contextNode->textContent; } else { $out .= ''; @@ -1112,7 +1139,9 @@ class PPFrame_DOM implements PPFrame { $first = true; $s = ''; foreach ( $args as $root ) { - if ( $root instanceof PPNode_DOM ) $root = $root->node; + if ( $root instanceof PPNode_DOM ) { + $root = $root->node; + } if ( !is_array( $root ) && !( $root instanceof DOMNodeList ) ) { $root = array( $root ); } @@ -1136,9 +1165,11 @@ class PPFrame_DOM implements PPFrame { $args = array_slice( func_get_args(), 1 ); $out = array(); $first = true; - if ( $root instanceof PPNode_DOM ) $root = $root->node; foreach ( $args as $root ) { + if ( $root instanceof PPNode_DOM ) { + $root = $root->node; + } if ( !is_array( $root ) && !( $root instanceof DOMNodeList ) ) { $root = array( $root ); } @@ -1163,7 +1194,9 @@ class PPFrame_DOM implements PPFrame { $first = true; foreach ( $args as $root ) { - if ( $root instanceof PPNode_DOM ) $root = $root->node; + if ( $root instanceof PPNode_DOM ) { + $root = $root->node; + } if ( !is_array( $root ) && !( $root instanceof DOMNodeList ) ) { $root = array( $root ); } @@ -1239,7 +1272,8 @@ class PPTemplateFrame_DOM extends PPFrame_DOM { var $numberedExpansionCache, $namedExpansionCache; function __construct( $preprocessor, $parent = false, $numberedArgs = array(), $namedArgs = array(), $title = false ) { - PPFrame_DOM::__construct( $preprocessor ); + parent::__construct( $preprocessor ); + $this->parent = $parent; $this->numberedArgs = $numberedArgs; $this->namedArgs = $namedArgs; @@ -1310,7 +1344,7 @@ class PPTemplateFrame_DOM extends PPFrame_DOM { } if ( !isset( $this->numberedExpansionCache[$index] ) ) { # No trimming for unnamed arguments - $this->numberedExpansionCache[$index] = $this->parent->expand( $this->numberedArgs[$index], self::STRIP_COMMENTS ); + $this->numberedExpansionCache[$index] = $this->parent->expand( $this->numberedArgs[$index], PPFrame::STRIP_COMMENTS ); } return $this->numberedExpansionCache[$index]; } @@ -1322,7 +1356,7 @@ class PPTemplateFrame_DOM extends PPFrame_DOM { 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 ) ); + $this->parent->expand( $this->namedArgs[$name], PPFrame::STRIP_COMMENTS ) ); } return $this->namedExpansionCache[$name]; } @@ -1351,7 +1385,7 @@ class PPCustomFrame_DOM extends PPFrame_DOM { var $args; function __construct( $preprocessor, $args ) { - PPFrame_DOM::__construct( $preprocessor ); + parent::__construct( $preprocessor ); $this->args = $args; } diff --git a/includes/parser/Preprocessor_Hash.php b/includes/parser/Preprocessor_Hash.php index c5d69685..6cb2febc 100644 --- a/includes/parser/Preprocessor_Hash.php +++ b/includes/parser/Preprocessor_Hash.php @@ -1,5 +1,11 @@ <?php - +/** + * Preprocessor using PHP arrays + * + * @file + * @ingroup Parser + */ + /** * Differences from DOM schema: * * attribute nodes are children @@ -23,12 +29,39 @@ class Preprocessor_Hash implements Preprocessor { return new PPCustomFrame_Hash( $this, $args ); } + function newPartNodeArray( $values ) { + $list = array(); + + foreach ( $values as $k => $val ) { + $partNode = new PPNode_Hash_Tree( 'part' ); + $nameNode = new PPNode_Hash_Tree( 'name' ); + + if ( is_int( $k ) ) { + $nameNode->addChild( new PPNode_Hash_Attr( 'index', $k ) ); + $partNode->addChild( $nameNode ); + } else { + $nameNode->addChild( new PPNode_Hash_Text( $k ) ); + $partNode->addChild( $nameNode ); + $partNode->addChild( new PPNode_Hash_Text( '=' ) ); + } + + $valueNode = new PPNode_Hash_Tree( 'value' ); + $valueNode->addChild( new PPNode_Hash_Text( $val ) ); + $partNode->addChild( $valueNode ); + + $list[] = $partNode; + } + + $node = new PPNode_Hash_Array( $list ); + return $node; + } + /** * 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: + * @param $text String: the text to parse + * @param $flags Integer: bitwise combination of: * Parser::PTD_FOR_INCLUSION Handle <noinclude>/<includeonly> as if the text is being * included. Default is to assume a direct page view. * @@ -401,7 +434,7 @@ class Preprocessor_Hash implements Preprocessor { $count = $piece->count; $equalsLength = strspn( $revText, '=', strlen( $text ) - $searchStart ); if ( $equalsLength > 0 ) { - if ( $i - $equalsLength == $piece->startPos ) { + if ( $searchStart - $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) @@ -479,7 +512,6 @@ class Preprocessor_Hash implements Preprocessor { # 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 @@ -526,7 +558,7 @@ class Preprocessor_Hash implements Preprocessor { $titleNode->lastChild = $titleAccum->lastNode; $element->addChild( $titleNode ); $argIndex = 1; - foreach ( $parts as $partIndex => $part ) { + foreach ( $parts as $part ) { if ( isset( $part->eqpos ) ) { // Find equals $lastNode = false; @@ -647,7 +679,7 @@ class Preprocessor_Hash implements Preprocessor { // Cache if ($cacheable) { - $cacheValue = sprintf( "%08d", self::CACHE_VERSION ) . serialize( $rootNode );; + $cacheValue = sprintf( "%08d", self::CACHE_VERSION ) . serialize( $rootNode ); $wgMemc->set( $cacheKey, $cacheValue, 86400 ); wfProfileOut( __METHOD__.'-cache-miss' ); wfProfileOut( __METHOD__.'-cacheable' ); @@ -804,7 +836,7 @@ class PPFrame_Hash implements PPFrame { /** * Construct a new preprocessor frame. - * @param Preprocessor $preprocessor The parent preprocessor + * @param $preprocessor Preprocessor: the parent preprocessor */ function __construct( $preprocessor ) { $this->preprocessor = $preprocessor; @@ -826,7 +858,6 @@ class PPFrame_Hash implements PPFrame { $title = $this->title; } if ( $args !== false ) { - $xpath = false; if ( $args instanceof PPNode_Hash_Array ) { $args = $args->value; } elseif ( !is_array( $args ) ) { @@ -855,11 +886,11 @@ class PPFrame_Hash implements PPFrame { return $root; } - if ( ++$this->parser->mPPNodeCount > $this->parser->mOptions->mMaxPPNodeCount ) + if ( ++$this->parser->mPPNodeCount > $this->parser->mOptions->getMaxPPNodeCount() ) { return '<span class="error">Node-count limit exceeded</span>'; } - if ( $expansionDepth > $this->parser->mOptions->mMaxPPExpandDepth ) { + if ( $expansionDepth > $this->parser->mOptions->getMaxPPExpandDepth() ) { return '<span class="error">Expansion depth limit exceeded</span>'; } ++$expansionDepth; @@ -915,7 +946,7 @@ class PPFrame_Hash implements PPFrame { if ( $contextNode->name == 'template' ) { # Double-brace expansion $bits = $contextNode->splitTemplate(); - if ( $flags & self::NO_TEMPLATES ) { + if ( $flags & PPFrame::NO_TEMPLATES ) { $newIterator = $this->virtualBracketedImplode( '{{', '|', '}}', $bits['title'], $bits['parts'] ); } else { $ret = $this->parser->braceSubstitution( $bits, $this ); @@ -928,7 +959,7 @@ class PPFrame_Hash implements PPFrame { } elseif ( $contextNode->name == 'tplarg' ) { # Triple-brace expansion $bits = $contextNode->splitTemplate(); - if ( $flags & self::NO_ARGS ) { + if ( $flags & PPFrame::NO_ARGS ) { $newIterator = $this->virtualBracketedImplode( '{{{', '|', '}}}', $bits['title'], $bits['parts'] ); } else { $ret = $this->parser->argSubstitution( $bits, $this ); @@ -943,13 +974,13 @@ class PPFrame_Hash implements PPFrame { # 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 ) ) + || ( $flags & PPFrame::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 ) ) { + elseif ( $this->parser->ot['wiki'] && ! ( $flags & PPFrame::RECOVER_COMMENTS ) ) { $out .= $this->parser->insertStripItem( $contextNode->firstChild->value ); } # Recover the literal comment in RECOVER_COMMENTS and pre+no-remove @@ -961,7 +992,7 @@ class PPFrame_Hash implements PPFrame { # OT_WIKI will only respect <ignore> 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 ) ) { + if ( ( !isset( $this->parent ) && $this->parser->ot['wiki'] ) || ( $flags & PPFrame::NO_IGNORE ) ) { $out .= $contextNode->firstChild->value; } else { //$out .= ''; @@ -1186,7 +1217,8 @@ class PPTemplateFrame_Hash extends PPFrame_Hash { var $numberedExpansionCache, $namedExpansionCache; function __construct( $preprocessor, $parent = false, $numberedArgs = array(), $namedArgs = array(), $title = false ) { - PPFrame_Hash::__construct( $preprocessor ); + parent::__construct( $preprocessor ); + $this->parent = $parent; $this->numberedArgs = $numberedArgs; $this->namedArgs = $namedArgs; @@ -1257,7 +1289,7 @@ class PPTemplateFrame_Hash extends PPFrame_Hash { } if ( !isset( $this->numberedExpansionCache[$index] ) ) { # No trimming for unnamed arguments - $this->numberedExpansionCache[$index] = $this->parent->expand( $this->numberedArgs[$index], self::STRIP_COMMENTS ); + $this->numberedExpansionCache[$index] = $this->parent->expand( $this->numberedArgs[$index], PPFrame::STRIP_COMMENTS ); } return $this->numberedExpansionCache[$index]; } @@ -1269,7 +1301,7 @@ class PPTemplateFrame_Hash extends PPFrame_Hash { 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 ) ); + $this->parent->expand( $this->namedArgs[$name], PPFrame::STRIP_COMMENTS ) ); } return $this->namedExpansionCache[$name]; } @@ -1298,7 +1330,7 @@ class PPCustomFrame_Hash extends PPFrame_Hash { var $args; function __construct( $preprocessor, $args ) { - PPFrame_Hash::__construct( $preprocessor ); + parent::__construct( $preprocessor ); $this->args = $args; } diff --git a/includes/parser/Tidy.php b/includes/parser/Tidy.php index 95f83621..38f22fd8 100644 --- a/includes/parser/Tidy.php +++ b/includes/parser/Tidy.php @@ -1,4 +1,9 @@ <?php +/** + * HTML validation and correction + * + * @file + */ /** * Class to interact with HTML tidy @@ -16,8 +21,8 @@ class MWTidy { * If tidy isn't able to correct the markup, the original will be * returned in all its glory with a warning comment appended. * - * @param string $text Hideous HTML input - * @return string Corrected HTML output + * @param $text String: hideous HTML input + * @return String: corrected HTML output */ public static function tidy( $text ) { global $wgTidyInternal; @@ -26,9 +31,6 @@ class MWTidy { ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>'. '<head><title>test</title></head><body>'.$text.'</body></html>'; - # Tidy is known to clobber tabs; convert them to entities - $wrappedtext = str_replace( "\t", '	', $wrappedtext ); - if( $wgTidyInternal ) { $correctedtext = self::execInternalTidy( $wrappedtext ); } else { @@ -39,9 +41,6 @@ class MWTidy { return $text . "\n<!-- Tidy found serious XHTML errors -->\n"; } - # Convert the tabs back from entities - $correctedtext = str_replace( '	', "\t", $correctedtext ); - return $correctedtext; } @@ -71,7 +70,7 @@ class MWTidy { * @param $text String: HTML to check * @param $stderr Boolean: Whether to read from STDERR rather than STDOUT * @param &$retval Exit code (-1 on internal error) - * @retrun mixed String or null + * @return mixed String or null */ private static function execExternalTidy( $text, $stderr = false, &$retval = null ) { global $wgTidyConf, $wgTidyBin, $wgTidyOpts; @@ -119,15 +118,13 @@ class MWTidy { $retval = -1; } - wfProfileOut( __METHOD__ ); - if( !$stderr && $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; + $cleansource = null; } + wfProfileOut( __METHOD__ ); + return $cleansource; } /** @@ -137,7 +134,7 @@ class MWTidy { * 'pear install tidy' should be able to compile the extension module. */ private static function execInternalTidy( $text, $stderr = false, &$retval = null ) { - global $wgTidyConf, $IP, $wgDebugTidy; + global $wgTidyConf, $wgDebugTidy; wfProfileIn( __METHOD__ ); $tidy = new tidy; @@ -145,6 +142,7 @@ class MWTidy { if( $stderr ) { $retval = $tidy->getStatus(); + wfProfileOut( __METHOD__ ); return $tidy->errorBuffer; } else { $tidy->cleanRepair(); |