X-Git-Url: https://git.heureux-cyclage.org/?a=blobdiff_plain;f=includes%2Fparser%2FParser.php;h=b723bea43399f936229110411f3804fd3dcf8638;hb=969561ae3c9051eb408626e9b6debb6c0f210b71;hp=cd32813e5a35cf798a57fe714784a280588fc865;hpb=3b150034a08872e858684c604d37ca92b58b920a;p=lhc%2Fweb%2Fwiklou.git diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php index cd32813e5a..b723bea433 100644 --- a/includes/parser/Parser.php +++ b/includes/parser/Parser.php @@ -22,10 +22,12 @@ * produces altered wiki markup. * preprocess() * removes HTML comments and expands templates - * cleanSig() + * cleanSig() / cleanSigInSig() * Cleans a signature before saving it to preferences - * extractSections() - * Extracts sections from an article for section editing + * getSection() + * Return the content of a section from an article for section editing + * replaceSection() + * Replaces a section by number inside an article * getPreloadText() * Removes sections, and tags. * @@ -89,10 +91,20 @@ class Parser { const MARKER_SUFFIX = "-QINU\x7f"; # Persistent: - var $mTagHooks, $mTransparentTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables; - var $mSubstWords, $mImageParams, $mImageParamsMagicArray, $mStripList, $mMarkerIndex; - var $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols, $mDefaultStripList; - var $mVarCache, $mConf, $mFunctionTagHooks; + var $mTagHooks = array(); + var $mTransparentTagHooks = array(); + var $mFunctionHooks = array(); + var $mFunctionSynonyms = array( 0 => array(), 1 => array() ); + var $mFunctionTagHooks = array(); + var $mStripList = array(); + var $mDefaultStripList = array(); + var $mVarCache = array(); + var $mImageParams = array(); + var $mImageParamsMagicArray = array(); + var $mMarkerIndex = 0; + var $mFirstCall = true; + var $mVariables, $mSubstWords; # Initialised by initialiseVariables() + var $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols; # Initialised in constructor # Cleared with clearState(): @@ -103,6 +115,7 @@ class Parser { var $mTplExpandCache; # empty-frame expansion cache var $mTplRedirCache, $mTplDomCache, $mHeadings, $mDoubleUnderscores; var $mExpensiveFunctionCount; # number of expensive parser function calls + var $mUser; # User object; only used when doing pre-save transform # Temporary # These are variables reset at least once per parse regardless of $clearState @@ -110,8 +123,10 @@ class Parser { 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 $mRevisionObject; # The revision object of the specified revision ID var $mRevisionId; # ID to display in {{REVISIONID}} tags var $mRevisionTimestamp; # The timestamp of the specified revision ID + var $mRevisionUser; # Userto display in {{REVISIONUSER}} tag var $mRevIdForTs; # The revision ID which was used to fetch the timestamp /** @@ -121,16 +136,9 @@ class Parser { */ function __construct( $conf = array() ) { $this->mConf = $conf; - $this->mTagHooks = array(); - $this->mTransparentTagHooks = array(); - $this->mFunctionHooks = array(); - $this->mFunctionTagHooks = array(); - $this->mFunctionSynonyms = array( 0 => array(), 1 => array() ); - $this->mDefaultStripList = $this->mStripList = array(); $this->mUrlProtocols = wfUrlProtocols(); $this->mExtLinkBracketedRegex = '/\[(\b(' . wfUrlProtocols() . ')'. '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x00-\\x08\\x0a-\\x1F]*?)\]/S'; - $this->mVarCache = array(); if ( isset( $conf['preprocessorClass'] ) ) { $this->mPreprocessorClass = $conf['preprocessorClass']; } elseif ( extension_loaded( 'domxml' ) ) { @@ -142,8 +150,6 @@ class Parser { } else { $this->mPreprocessorClass = 'Preprocessor_Hash'; } - $this->mMarkerIndex = 0; - $this->mFirstCall = true; } /** @@ -188,6 +194,7 @@ class Parser { $this->firstCallInit(); } $this->mOutput = new ParserOutput; + $this->mOptions->registerWatcher( array( $this->mOutput, 'recordOption' ) ); $this->mAutonumber = 0; $this->mLastSection = ''; $this->mDTopen = false; @@ -197,8 +204,10 @@ class Parser { $this->mInPre = false; $this->mLinkHolders = new LinkHolderArray( $this ); $this->mLinkID = 0; - $this->mRevisionTimestamp = $this->mRevisionId = null; + $this->mRevisionObject = $this->mRevisionTimestamp = + $this->mRevisionId = $this->mRevisionUser = null; $this->mVarCache = array(); + $this->mUser = null; /** * Prefix for temporary replacement strings for the multipass parser. @@ -239,52 +248,6 @@ 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, - 'plain' => $ot == self::OT_PLAIN, - ); - } - - /** - * 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. @@ -308,18 +271,22 @@ 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 $oldRevisionId = $this->mRevisionId; + $oldRevisionObject = $this->mRevisionObject; $oldRevisionTimestamp = $this->mRevisionTimestamp; + $oldRevisionUser = $this->mRevisionUser; if ( $revid !== null ) { $this->mRevisionId = $revid; + $this->mRevisionObject = null; $this->mRevisionTimestamp = null; + $this->mRevisionUser = null; } $this->setOutputType( self::OT_HTML ); wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) ); @@ -333,10 +300,10 @@ 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 ); @@ -379,7 +346,7 @@ class Parser { || $wgDisableTitleConversion || isset( $this->mDoubleUnderscores['nocontentconvert'] ) || isset( $this->mDoubleUnderscores['notitleconvert'] ) - || $this->mOutput->getDisplayTitle() !== false ) ) + || $this->mOutput->getDisplayTitle() !== false ) ) { $convruletitle = $wgContLang->getConvRuleTitle(); if ( $convruletitle ) { @@ -399,7 +366,7 @@ 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 ) { list( $element, $content, $params, $tag ) = $data; @@ -415,7 +382,7 @@ 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 @@ -457,7 +424,7 @@ 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; @@ -467,7 +434,9 @@ class Parser { $this->mOutput->setText( $text ); $this->mRevisionId = $oldRevisionId; + $this->mRevisionObject = $oldRevisionObject; $this->mRevisionTimestamp = $oldRevisionTimestamp; + $this->mRevisionUser = $oldRevisionUser; wfProfileOut( $fname ); wfProfileOut( __METHOD__ ); @@ -481,7 +450,7 @@ 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__ ); @@ -498,9 +467,9 @@ class Parser { */ function preprocess( $text, $title, $options, $revid = null ) { wfProfileIn( __METHOD__ ); + $this->mOptions = $options; $this->clearState(); $this->setOutputType( self::OT_PREPROCESS ); - $this->mOptions = $options; $this->setTitle( $title ); if ( $revid !== null ) { $this->mRevisionId = $revid; @@ -521,9 +490,9 @@ class Parser { */ public function getPreloadText( $text, $title, $options ) { # Parser (re)initialisation + $this->mOptions = $options; $this->clearState(); $this->setOutputType( self::OT_PLAIN ); - $this->mOptions = $options; $this->setTitle( $title ); $flags = PPFrame::NO_ARGS | PPFrame::NO_TEMPLATES; @@ -537,15 +506,131 @@ class Parser { * @private * @static */ - function getRandomString() { + static private function getRandomString() { return dechex( mt_rand( 0, 0x7fffffff ) ) . dechex( mt_rand( 0, 0x7fffffff ) ); } - 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++; } + /** + * Set the current user. + * Should only be used when doing pre-save transform. + * + * @param $user Mixed: User object or null (to reset) + */ + function setUser( $user ) { + $this->mUser = $user; + } + + /** + * 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, + ); + } + + /** + * 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; @@ -558,8 +643,23 @@ class Parser { } } + /** + * Get a User object either from $this->mUser, if set, or from the + * ParserOptions object otherwise + * + * @return User object + */ + function getUser() { + if ( !is_null( $this->mUser ) ) { + return $this->mUser; + } + return $this->mOptions->getUser(); + } + /** * Get a preprocessor object + * + * @return Preprocessor instance */ function getPreprocessor() { if ( !isset( $this->mPreprocessor ) ) { @@ -582,12 +682,13 @@ 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(); @@ -721,9 +822,8 @@ class Parser { */ function doTableStuff( $text ) { wfProfileIn( __METHOD__ ); - + $lines = StringUtils::explode( "\n", $text ); - $text = null; $out = ''; $td_history = array(); # Is currently a td tag open? $last_tag_history = array(); # Save history of last lag activated (td, th or caption) @@ -735,7 +835,7 @@ class Parser { 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; } @@ -843,7 +943,7 @@ class Parser { $last_tag = array_pop( $last_tag_history ); if ( array_pop( $td_history ) ) { - $previous = "{$previous}"; + $previous = "\n{$previous}"; } if ( $first_character === '|' ) { @@ -961,8 +1061,8 @@ class Parser { $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( $text ); # replaceInternalLinks may sometimes leave behind @@ -994,10 +1094,10 @@ class Parser { (\\b(?:$prots)$urlChar+) | # m[3]: Free external links" . ' (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number ISBN\s+(\b # m[5]: ISBN, capture number - (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix - (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters - [0-9Xx] # check digit - \b) + (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix + (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters + [0-9Xx] # check digit + \b) )!x', array( &$this, 'magicLinkCallback' ), $text ); wfProfileOut( __METHOD__ ); return $text; @@ -1015,7 +1115,6 @@ 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'; @@ -1030,8 +1129,8 @@ class Parser { throw new MWException( __METHOD__.': unrecognised match type "' . 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 $CssClass" ); return "{$keyword} {$id}"; } elseif ( isset( $m[5] ) && $m[5] !== '' ) { @@ -1060,7 +1159,7 @@ class Parser { global $wgContLang; wfProfileIn( __METHOD__ ); - $sk = $this->mOptions->getSkin(); + $sk = $this->mOptions->getSkin( $this->mTitle ); $trail = ''; # The characters '<' and '>' (which were escaped by @@ -1147,10 +1246,9 @@ class Parser { # 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 ) { + 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. @@ -1174,7 +1272,6 @@ class Parser { $numbold++; } } - $i++; } # If there is an odd number of both bold and italics, it is likely @@ -1300,7 +1397,7 @@ class Parser { /** * Replace external links (REL) * - * Note: this is all very hackish and the order of execution matters a lot. + * Note: this is all very hackish and the order of execution matters a lot. * Make sure to run maintenance/parserTests.php if you change this code. * * @private @@ -1309,7 +1406,7 @@ 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 ); @@ -1366,7 +1463,7 @@ class Parser { # 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; @@ -1388,9 +1485,9 @@ 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(); @@ -1421,9 +1518,10 @@ 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 @@ -1437,8 +1535,6 @@ 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] ); @@ -1459,7 +1555,7 @@ class Parser { * @private */ function maybeMakeExternalImage( $url ) { - $sk = $this->mOptions->getSkin(); + $sk = $this->mOptions->getSkin( $this->mTitle ); $imagesfrom = $this->mOptions->getAllowExternalImagesFrom(); $imagesexception = !empty( $imagesfrom ); $text = false; @@ -1478,7 +1574,7 @@ class Parser { $imagematch = false; } if ( $this->mOptions->getAllowExternalImages() - || ( $imagesexception && $imagematch ) ) { + || ( $imagesexception && $imagematch ) ) { if ( preg_match( self::EXT_IMAGE_REGEX, $url ) ) { # Image found $text = $sk->makeExternalImage( $url ); @@ -1504,7 +1600,7 @@ class Parser { /** * Process [[ ]] wikilinks - * @return processed text + * @return String: processed text * * @private */ @@ -1535,10 +1631,10 @@ 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 $s = $a->current(); @@ -1573,7 +1669,7 @@ class Parser { } if ( $wgContLang->hasVariants() ) { - $selflink = $wgContLang->convertLinkToAllVariants( $this->mTitle->getPrefixedText() ); + $selflink = $wgContLang->autoConvertToAllVariants( $this->mTitle->getPrefixedText() ); } else { $selflink = array( $this->mTitle->getPrefixedText() ); } @@ -1630,14 +1726,14 @@ class Parser { # fix up urlencoded title texts 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('<', '>'), rawurldecode( $m[1] ) ); } $trail = $m[3]; } elseif ( preg_match( $e1_img, $line, $m ) ) { # Invalid, but might be an image with a link in its caption $might_be_img = true; $text = $m[2]; if ( strpos( $m[1], '%' ) !== false ) { - $m[1] = urldecode( $m[1] ); + $m[1] = rawurldecode( $m[1] ); } $trail = ""; } else { # Invalid form; output directly @@ -1731,6 +1827,12 @@ class Parser { $wasblank = ( $text == '' ); 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]] -> Lista d''e paise d''o munno + # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']] + # -> Criticism of Harry Potter + $text = $this->doQuotes( $text ); } # Link not escaped by : , create the various objects @@ -1792,7 +1894,7 @@ class Parser { * Strip the whitespace Category links produce, see bug 87 * @todo We might want to use trim($tmp, "\n") here. */ - $s .= trim( $prefix . $trail, "\n" ) == '' ? '': $prefix . $trail; + $s .= trim( $prefix . $trail, "\n" ) == '' ? '' : $prefix . $trail; wfProfileOut( __METHOD__."-category" ); continue; @@ -1864,16 +1966,16 @@ 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(); + $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; @@ -1886,8 +1988,8 @@ 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() . ')/', @@ -1896,7 +1998,7 @@ 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 @@ -1905,8 +2007,9 @@ 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 */ @@ -1927,6 +2030,7 @@ class Parser { $this->mLastSection = ''; return $result; } + /** * getCommon() returns the length of the longest common substring * of both arguments, starting at the beginning of both. @@ -1946,6 +2050,7 @@ class Parser { } return $i; } + /** * These next three functions open, continue, and close the list * element appropriate to the prefix character passed into them. @@ -2021,7 +2126,8 @@ 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 */ @@ -2142,11 +2248,7 @@ class Parser { if ( $preOpenMatch and !$preCloseMatch ) { $this->mInPre = true; } - if ( $closematch ) { - $inBlockElem = false; - } else { - $inBlockElem = true; - } + $inBlockElem = !$closematch; } elseif ( !$inBlockElem && !$this->mInPre ) { if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' || trim( $t ) != '' ) ) { # pre @@ -2210,10 +2312,11 @@ 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 ) { wfProfileIn( __METHOD__ ); @@ -2376,8 +2479,8 @@ 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 @@ -2651,12 +2754,17 @@ 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': @@ -2664,8 +2772,8 @@ 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 ) ) ) { @@ -2700,8 +2808,8 @@ 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 / as if the text is being * included. Default is to assume a direct page view. * @@ -2749,11 +2857,11 @@ 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 ) { @@ -2805,7 +2913,7 @@ 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: + * @param $limitationType String: should be one of: * 'expensive-parserfunction' (corresponding messages: * 'expensive-parserfunction-warning', * 'expensive-parserfunction-category') @@ -2815,7 +2923,8 @@ class Parser { * '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 $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) { @@ -2829,12 +2938,12 @@ 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 ) { @@ -2863,6 +2972,7 @@ class Parser { $originalTitle = $part1; # $args is a list of argument nodes, starting from index 0, not including $part1 + # *** FIXME if piece['parts'] is null then the call to getLength() below won't work b/c this $args isn't an object $args = ( null == $piece['parts'] ) ? array() : $piece['parts']; wfProfileOut( __METHOD__.'-setup' ); @@ -2900,7 +3010,7 @@ class Parser { if ( $id !== false ) { $text = $this->getVariableValue( $id, $frame ); if ( MagicWord::getCacheTTL( $id ) > -1 ) { - $this->mOutput->mContainsOldMagic = true; + $this->mOutput->updateCacheExpiry( MagicWord::getCacheTTL( $id ) ); } $found = true; } @@ -3185,7 +3295,7 @@ class Parser { if ( !$title->equals( $cacheTitle ) ) { $this->mTplRedirCache[$cacheTitle->getPrefixedDBkey()] = - array( $title->getNamespace(),$cdb = $title->getDBkey() ); + array( $title->getNamespace(), $cdb = $title->getDBkey() ); } return array( $dom, $title ); @@ -3195,7 +3305,7 @@ 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; @@ -3281,13 +3391,13 @@ class Parser { global $wgEnableScaryTranscluding; if ( !$wgEnableScaryTranscluding ) { - return wfMsg('scarytranscludedisabled'); + return wfMsgForContent('scarytranscludedisabled'); } $url = $title->getFullUrl( "action=$action" ); if ( strlen( $url ) > 255 ) { - return wfMsg( 'scarytranscludetoolong' ); + return wfMsgForContent( 'scarytranscludetoolong' ); } return $this->fetchScaryTemplateMaybeFromCache( $url ); } @@ -3304,7 +3414,7 @@ class Parser { $text = Http::get( $url ); if ( !$text ) { - return wfMsg( 'scarytranscludefailed', $url ); + return wfMsgForContent( 'scarytranscludefailed', $url ); } $dbw = wfGetDB( DB_MASTER ); @@ -3332,9 +3442,9 @@ class Parser { $text = $frame->getArgument( $argName ); if ( $text === false && $parts->getLength() > 0 && ( - $this->ot['html'] - || $this->ot['pre'] - || ( $this->ot['wiki'] && $frame->isTemplate() ) + $this->ot['html'] + || $this->ot['pre'] + || ( $this->ot['wiki'] && $frame->isTemplate() ) ) ) { # No match in frame, use the supplied default @@ -3366,17 +3476,15 @@ 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'] ); @@ -3455,12 +3563,12 @@ 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 ) ) { + if ( $this->mIncludeSizes[$type] + $size > $this->mOptions->getMaxIncludeSize() ) { return false; } else { $this->mIncludeSizes[$type] += $size; @@ -3471,7 +3579,7 @@ 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; @@ -3513,7 +3621,6 @@ 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. @@ -3529,6 +3636,11 @@ class Parser { $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; } @@ -3536,8 +3648,9 @@ 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 ) { $cat = wfMsgForContent( $msg ); @@ -3567,25 +3680,27 @@ 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(); + } + if ( $showEditLink ) { + $editLinkAsToken = $this->mOptions->getEditSectionTokens(); + if ( $editLinkAsToken ) { + $this->mOutput->setEditSectionTokens( "{$this->mUniqPrefix}-editsection-", self::MARKER_SUFFIX ); + } } # Get all headlines for numbering them and adding funky stuff like [edit] @@ -3618,7 +3733,7 @@ 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; @@ -3631,7 +3746,6 @@ class Parser { $head = array(); $sublevelCount = array(); $levelCount = array(); - $toclevel = 0; $level = 0; $prevlevel = 0; $toclevel = 0; @@ -3645,6 +3759,7 @@ class Parser { $node = $root->getFirstChild(); $byteOffset = 0; $tocraw = array(); + $refers = array(); foreach ( $matches[3] as $headline ) { $isTemplate = false; @@ -3661,7 +3776,6 @@ class Parser { if ( $toclevel ) { $prevlevel = $level; - $prevtoclevel = $toclevel; } $level = $matches[1][$headlineCount]; @@ -3742,8 +3856,7 @@ class Parser { # 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; @@ -3773,9 +3886,10 @@ 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; @@ -3820,8 +3934,9 @@ 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 ) ) ); @@ -3840,12 +3955,28 @@ class Parser { # give headline the correct tag if ( $showEditLink && $sectionIndex !== false ) { - if ( $isTemplate ) { - # Put a T flag in the section identifier, to indicate to extractSections() - # that sections inside should be counted. - $editlink = $sk->doEditSectionLink( Title::newFromText( $titleText ), "T-$sectionIndex" ); + if ( $editLinkAsToken ) { + // Output edit section links as markers with styles that can be customized by skins + if ( $isTemplate ) { + # Put a T flag in the section identifier, to indicate to extractSections() + # that sections inside should be counted. + $editlinkArgs = array( $titleText, "T-$sectionIndex", null ); + } else { + $editlinkArgs = array( $this->mTitle->getPrefixedText(), $sectionIndex, $headlineHint ); + } + // We use nearly the same structure as uniqPrefix and the marker stuffix (besides there being nothing random) + // However the this is output into the parser output itself not replaced early, so we hardcode this in case + // the constants change in a different version of MediaWiki, which would break this code. + $editlink = "{$this->mUniqPrefix}-editsection-" . serialize($editlinkArgs) . self::MARKER_SUFFIX; } else { - $editlink = $sk->doEditSectionLink( $this->mTitle, $sectionIndex, $headlineHint ); + // Output edit section links directly as markup like we used to + if ( $isTemplate ) { + # Put a T flag in the section identifier, to indicate to extractSections() + # that sections inside should be counted. + $editlink = $sk->doEditSectionLink( Title::newFromText( $titleText ), "T-$sectionIndex", null, $this->mOptions->getUserLang() ); + } else { + $editlink = $sk->doEditSectionLink( $this->mTitle, $sectionIndex, $headlineHint, $this->mOptions->getUserLang() ); + } } } else { $editlink = ''; @@ -3908,115 +4039,21 @@ 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; - } elseif ( $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 $user, ParserOptions $options, $clearState = true ) { $this->mOptions = $options; $this->setTitle( $title ); + $this->setUser( $user ); $this->setOutputType( self::OT_WIKI ); if ( $clearState ) { @@ -4029,6 +4066,9 @@ class Parser { $text = str_replace( array_keys( $pairs ), array_values( $pairs ), $text ); $text = $this->pstPass2( $text, $user ); $text = $this->mStripState->unstripBoth( $text ); + + $this->setUser( null ); #Reset + return $text; } @@ -4120,7 +4160,10 @@ 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 ) { @@ -4169,7 +4212,7 @@ 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 ) { @@ -4182,16 +4225,16 @@ 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; } @@ -4221,8 +4264,9 @@ 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}/', '', $text ); @@ -4232,10 +4276,10 @@ class Parser { /** * Set up some variables which are usually set up in parse() * so that an external function can call some class members with confidence - * @public */ - function startExternalParse( &$title, $options, $outputType, $clearState = true ) { + public function startExternalParse( &$title, $options, $outputType, $clearState = true ) { $this->setTitle( $title ); + $options->resetUsage(); $this->mOptions = $options; $this->setOutputType( $outputType ); if ( $clearState ) { @@ -4246,12 +4290,11 @@ 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; @@ -4277,14 +4320,11 @@ class Parser { * Transform and return $text. Use $parser for any required context, e.g. use * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions * - * @public - * - * @param mixed $tag The tag to use, e.g. 'hook' for - * @param mixed $callback The callback function (and object) to use for the tag - * + * @param $tag Mixed: the tag to use, e.g. 'hook' for + * @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; @@ -4327,11 +4367,9 @@ 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 @@ -4355,7 +4393,7 @@ 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; @@ -4390,7 +4428,7 @@ class Parser { /** * Get all registered function hook identifiers * - * @return array + * @return Array */ function getFunctionHooks() { return array_keys( $this->mFunctionHooks ); @@ -4427,8 +4465,9 @@ class Parser { /** * Replace 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 ); @@ -4451,7 +4490,7 @@ 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['showfilename'] ) ) { @@ -4489,7 +4528,7 @@ class Parser { } if ( strpos( $matches[0], '%' ) !== false ) { - $matches[1] = urldecode( $matches[1] ); + $matches[1] = rawurldecode( $matches[1] ); } $tp = Title::newFromText( $matches[1] ); $nt =& $tp; @@ -4557,9 +4596,10 @@ 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" @@ -4588,7 +4628,7 @@ 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; @@ -4599,7 +4639,6 @@ class Parser { } # Get the file - $imagename = $title->getDBkey(); $file = wfFindFile( $title, array( 'time' => $time ) ); # Get parameter map $handler = $file ? $file->getHandler() : false; @@ -4666,6 +4705,9 @@ class Parser { if ( preg_match( "/^($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 { @@ -4706,9 +4748,9 @@ class Parser { # Will the image be presented in a frame, with the caption below? $imageIsFramed = isset( $params['frame']['frame'] ) || - isset( $params['frame']['framed'] ) || - isset( $params['frame']['thumbnail'] ) || - isset( $params['frame']['manualthumb'] ); + isset( $params['frame']['framed'] ) || + isset( $params['frame']['thumbnail'] ) || + isset( $params['frame']['manualthumb'] ); # In the old days, [[Image:Foo|text...]] would set alt text. Later it # came to also set the caption, ordinary text after the image -- which @@ -4752,7 +4794,7 @@ 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 ) { @@ -4787,15 +4829,17 @@ 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 ) { @@ -4804,24 +4848,12 @@ 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 ) ); } - /**#@-*/ - /** * Break wikitext input into sections, and either pull or replace @@ -4829,8 +4861,8 @@ 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: * - - ... -
* * Currently the only recognised flag is "T", which means the target section number @@ -4843,16 +4875,16 @@ 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->mOptions = new ParserOptions; $this->clearState(); $this->setTitle( $wgTitle ); # not generally used but removes an ugly failure mode - $this->mOptions = new ParserOptions; $this->setOutputType( self::OT_PLAIN ); $outText = ''; $frame = $this->getPreprocessor()->newFrame(); @@ -4878,11 +4910,11 @@ class Parser { # 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; } } @@ -4948,19 +4980,54 @@ 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 revision object for $this->mRevisionId + * + * @return either a Revision object or null + */ + protected function getRevisionObject() { + if ( !is_null( $this->mRevisionObject ) ) { + return $this->mRevisionObject; + } + if ( is_null( $this->mRevisionId ) ) { + return null; + } + + $this->mRevisionObject = Revision::newFromId( $this->mRevisionId ); + return $this->mRevisionObject; + } + /** * Get the timestamp associated with the current revision, adjusted for * the default server-local timestamp @@ -4968,24 +5035,21 @@ class Parser { function getRevisionTimestamp() { if ( is_null( $this->mRevisionTimestamp ) ) { wfProfileIn( __METHOD__ ); - global $wgContLang; - $dbr = wfGetDB( DB_SLAVE ); - $timestamp = $dbr->selectField( 'revision', 'rev_timestamp', - array( 'rev_id' => $this->mRevisionId ), __METHOD__ ); - - # Normalize timestamp to internal MW format for timezone processing. - # This has the added side-effect of replacing a null value with - # the current time, which gives us more sensible behavior for - # previews. - $timestamp = wfTimestamp( TS_MW, $timestamp ); - - # The cryptic '' timezone parameter tells to use the site-default - # timezone offset instead of the user settings. - # - # Since this value will be saved into the parser cache, served - # to other users, and potentially even used inside links and such, - # it needs to be consistent for all visitors. - $this->mRevisionTimestamp = $wgContLang->userAdjust( $timestamp, '' ); + + $revObject = $this->getRevisionObject(); + $timestamp = $revObject ? $revObject->getTimestamp() : false; + + if( $timestamp !== false ) { + global $wgContLang; + + # 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__ ); } @@ -4994,18 +5058,22 @@ 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 ) { - $revision = Revision::newFromId( $this->mRevisionId ); - $revuser = $revision->getUserText(); - } else { - global $wgUser; - $revuser = $wgUser->getName(); + if( is_null( $this->mRevisionUser ) ) { + $revObject = $this->getRevisionObject(); + + # if this template is subst: the revision id will be blank, + # so just use the current user's name + if( $revObject ) { + $this->mRevisionUser = $revObject->getUserText(); + } elseif( $this->ot['wiki'] || $this->mOptions->getIsPreview() ) { + $this->mRevisionUser = $this->getUser()->getName(); + } } - return $revuser; + return $this->mRevisionUser; } /** @@ -5015,6 +5083,7 @@ class Parser { */ public function setDefaultSort( $sort ) { $this->mDefaultSort = $sort; + $this->mOutput->setProperty( 'defaultsort', $sort ); } /** @@ -5024,15 +5093,10 @@ class Parser { * @return string */ public function getDefaultSort() { - global $wgCategoryPrefixedDefaultSortkey; if ( $this->mDefaultSort !== false ) { return $this->mDefaultSort; - } elseif ( $this->mTitle->getNamespace() == NS_CATEGORY || - !$wgCategoryPrefixedDefaultSortkey ) - { - return $this->mTitle->getText(); } else { - return $this->mTitle->getPrefixedText(); + return $this->mTitle->getCategorySortkey(); } } @@ -5054,19 +5118,23 @@ 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' ) ); } /** @@ -5079,7 +5147,7 @@ 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 */ @@ -5101,20 +5169,17 @@ class Parser { return $text; } - function srvus( $text ) { - return $this->testSrvus( $text, $this->mOutputType ); - } - /** * strip/replaceVariables/unstrip for preprocessor regression testing */ function testSrvus( $text, $title, $options, $outputType = self::OT_HTML ) { + $this->mOptions = $options; $this->clearState(); if ( !$title instanceof Title ) { $title = Title::newFromText( $title ); } $this->mTitle = $title; - $this->mOptions = $options; + $options->resetUsage(); $this->setOutputType( $outputType ); $text = $this->replaceVariables( $text ); $text = $this->mStripState->unstripBoth( $text ); @@ -5232,7 +5297,7 @@ class Parser { */ function unserialiseHalfParsedText( $data, $intPrefix = null ) { if ( !$intPrefix ) { - $intPrefix = $this->getRandomString(); + $intPrefix = self::getRandomString(); } # First, extract the strip state.