diff options
author | Pierre Schmitz <pierre@archlinux.de> | 2010-07-28 11:52:48 +0200 |
---|---|---|
committer | Pierre Schmitz <pierre@archlinux.de> | 2010-07-28 11:52:48 +0200 |
commit | 222b01f5169f1c7e69762e0e8904c24f78f71882 (patch) | |
tree | 8e932e12546bb991357ec48eb1638d1770be7a35 /includes/parser | |
parent | 00ab76a6b686e98a914afc1975812d2b1aaa7016 (diff) |
update to MediaWiki 1.16.0
Diffstat (limited to 'includes/parser')
-rw-r--r-- | includes/parser/CoreParserFunctions.php | 148 | ||||
-rw-r--r-- | includes/parser/CoreTagHooks.php | 49 | ||||
-rw-r--r-- | includes/parser/DateFormatter.php | 10 | ||||
-rw-r--r-- | includes/parser/LinkHolderArray.php | 3 | ||||
-rw-r--r-- | includes/parser/Parser.php | 1121 | ||||
-rw-r--r-- | includes/parser/ParserCache.php | 75 | ||||
-rw-r--r-- | includes/parser/ParserOptions.php | 27 | ||||
-rw-r--r-- | includes/parser/ParserOutput.php | 31 | ||||
-rw-r--r-- | includes/parser/Preprocessor.php | 15 | ||||
-rw-r--r-- | includes/parser/Preprocessor_DOM.php | 31 | ||||
-rw-r--r-- | includes/parser/Preprocessor_Hash.php | 18 |
11 files changed, 966 insertions, 562 deletions
diff --git a/includes/parser/CoreParserFunctions.php b/includes/parser/CoreParserFunctions.php index 774e96a7..8abcc04f 100644 --- a/includes/parser/CoreParserFunctions.php +++ b/includes/parser/CoreParserFunctions.php @@ -16,6 +16,7 @@ class CoreParserFunctions { $parser->setFunctionHook( 'int', array( __CLASS__, 'intFunction' ), SFH_NO_HASH ); $parser->setFunctionHook( 'ns', array( __CLASS__, 'ns' ), SFH_NO_HASH ); + $parser->setFunctionHook( 'nse', array( __CLASS__, 'nse' ), SFH_NO_HASH ); $parser->setFunctionHook( 'urlencode', array( __CLASS__, 'urlencode' ), SFH_NO_HASH ); $parser->setFunctionHook( 'lcfirst', array( __CLASS__, 'lcfirst' ), SFH_NO_HASH ); $parser->setFunctionHook( 'ucfirst', array( __CLASS__, 'ucfirst' ), SFH_NO_HASH ); @@ -67,7 +68,7 @@ class CoreParserFunctions { $parser->setFunctionHook( 'subjectpagename', array( __CLASS__, 'subjectpagename' ), SFH_NO_HASH ); $parser->setFunctionHook( 'subjectpagenamee', array( __CLASS__, 'subjectpagenamee' ), SFH_NO_HASH ); $parser->setFunctionHook( 'tag', array( __CLASS__, 'tagObj' ), SFH_OBJECT_ARGS ); - $parser->setFunctionHook( 'formatdate', array( __CLASS__, 'formatDate' ) ); + $parser->setFunctionHook( 'formatdate', array( __CLASS__, 'formatDate' ) ); if ( $wgAllowDisplayTitle ) { $parser->setFunctionHook( 'displaytitle', array( __CLASS__, 'displaytitle' ), SFH_NO_HASH ); @@ -88,20 +89,20 @@ class CoreParserFunctions { return array( 'found' => false ); } } - + static function formatDate( $parser, $date, $defaultPref = null ) { $df = DateFormatter::getInstance(); - - $date = trim($date); - + + $date = trim( $date ); + $pref = $parser->mOptions->getDateFormat(); - + // Specify a different default date format other than the the normal default - // iff the user has 'default' for their setting - if ($pref == 'default' && $defaultPref) + // iff the user has 'default' for their setting + if ( $pref == 'default' && $defaultPref ) $pref = $defaultPref; - - $date = $df->reformat( $pref, $date, array('match-whole') ); + + $date = $df->reformat( $pref, $date, array( 'match-whole' ) ); return $date; } @@ -119,6 +120,10 @@ class CoreParserFunctions { } } + static function nse( $parser, $part1 = '' ) { + return wfUrlencode( str_replace( ' ', '_', self::ns( $parser, $part1 ) ) ); + } + static function urlencode( $parser, $s = '' ) { return urlencode( $s ); } @@ -163,11 +168,11 @@ class CoreParserFunctions { # and the variable will fail. If we can't get a decent title from the first # attempt, url-decode and try for a second. if( is_null( $title ) ) - $title = Title::newFromUrl( urldecode( $s ) ); + $title = Title::newFromURL( urldecode( $s ) ); if( !is_null( $title ) ) { # Convert NS_MEDIA -> NS_FILE if( $title->getNamespace() == NS_MEDIA ) { - $title = Title::makeTitle( NS_FILE, $title->getDBKey() ); + $title = Title::makeTitle( NS_FILE, $title->getDBkey() ); } if( !is_null( $arg ) ) { $text = $title->$func( $arg ); @@ -193,15 +198,16 @@ class CoreParserFunctions { } static function gender( $parser, $user ) { + wfProfileIn( __METHOD__ ); $forms = array_slice( func_get_args(), 2); // default $gender = User::getDefaultOption( 'gender' ); - + // allow prefix. $title = Title::newFromText( $user ); - - if (is_object( $title ) && $title->getNamespace() == NS_USER) + + if ( is_object( $title ) && $title->getNamespace() == NS_USER ) $user = $title->getText(); // check parameter, or use $wgUser if in interface message @@ -212,10 +218,12 @@ class CoreParserFunctions { global $wgUser; $gender = $wgUser->getOption( 'gender' ); } - return $parser->getFunctionLang()->gender( $gender, $forms ); + $ret = $parser->getFunctionLang()->gender( $gender, $forms ); + wfProfileOut( __METHOD__ ); + return $ret; } - static function plural( $parser, $text = '') { - $forms = array_slice( func_get_args(), 2); + static function plural( $parser, $text = '' ) { + $forms = array_slice( func_get_args(), 2 ); $text = $parser->getFunctionLang()->parseFormattedNumber( $text ); return $parser->getFunctionLang()->convertPlural( $text, $forms ); } @@ -224,21 +232,39 @@ class CoreParserFunctions { * Override the title of the page when viewed, provided we've been given a * title which will normalise to the canonical title * - * @param Parser $parser Parent parser - * @param string $text Desired title text - * @return string + * @param $parser Parser: parent parser + * @param $text String: desired title text + * @return String */ static function displaytitle( $parser, $text = '' ) { global $wgRestrictDisplayTitle; - $text = trim( Sanitizer::decodeCharReferences( $text ) ); - if ( !$wgRestrictDisplayTitle ) { + #parse a limited subset of wiki markup (just the single quote items) + $text = $parser->doQuotes( $text ); + + #remove stripped text (e.g. the UNIQ-QINU stuff) that was generated by tag extensions/whatever + $text = preg_replace( '/' . preg_quote( $parser->uniqPrefix(), '/' ) . '.*?' + . preg_quote( Parser::MARKER_SUFFIX, '/' ) . '/', '', $text ); + + #list of disallowed tags for DISPLAYTITLE + #these will be escaped even though they are allowed in normal wiki text + $bad = array( 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'div', 'blockquote', 'ol', 'ul', 'li', 'hr', + 'table', 'tr', 'th', 'td', 'dl', 'dd', 'caption', 'p', 'ruby', 'rb', 'rt', 'rp', 'br' ); + + #only requested titles that normalize to the actual title are allowed through + #if $wgRestrictDisplayTitle is true (it is by default) + #mimic the escaping process that occurs in OutputPage::setPageTitle + $text = Sanitizer::normalizeCharReferences( Sanitizer::removeHTMLtags( $text, null, array(), array(), $bad ) ); + $title = Title::newFromText( Sanitizer::stripAllTags( $text ) ); + + if( !$wgRestrictDisplayTitle ) { $parser->mOutput->setDisplayTitle( $text ); } else { - $title = Title::newFromText( $text ); - if( $title instanceof Title && $title->getFragment() == '' && $title->equals( $parser->mTitle ) ) + if ( $title instanceof Title && $title->getFragment() == '' && $title->equals( $parser->mTitle ) ) { $parser->mOutput->setDisplayTitle( $text ); + } } + return ''; } @@ -291,9 +317,9 @@ class CoreParserFunctions { } static function numberingroup( $parser, $name = '', $raw = null) { return self::formatRaw( SiteStats::numberingroup( strtolower( $name ) ), $raw ); - } + } + - /** * Given a title, return the namespace name that would be given by the * corresponding magic word @@ -302,37 +328,37 @@ class CoreParserFunctions { */ static function mwnamespace( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return str_replace( '_', ' ', $t->getNsText() ); } static function namespacee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return wfUrlencode( $t->getNsText() ); } static function talkspace( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) || !$t->canTalk() ) + if ( is_null( $t ) || !$t->canTalk() ) return ''; return str_replace( '_', ' ', $t->getTalkNsText() ); } static function talkspacee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) || !$t->canTalk() ) + if ( is_null( $t ) || !$t->canTalk() ) return ''; return wfUrlencode( $t->getTalkNsText() ); } static function subjectspace( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return str_replace( '_', ' ', $t->getSubjectNsText() ); } static function subjectspacee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return wfUrlencode( $t->getSubjectNsText() ); } @@ -342,77 +368,77 @@ class CoreParserFunctions { */ static function pagename( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return wfEscapeWikiText( $t->getText() ); } static function pagenamee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return $t->getPartialURL(); } static function fullpagename( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) || !$t->canTalk() ) + if ( is_null( $t ) || !$t->canTalk() ) return ''; return wfEscapeWikiText( $t->getPrefixedText() ); } static function fullpagenamee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) || !$t->canTalk() ) + if ( is_null( $t ) || !$t->canTalk() ) return ''; return $t->getPrefixedURL(); } static function subpagename( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return $t->getSubpageText(); } static function subpagenamee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return $t->getSubpageUrlForm(); } static function basepagename( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return $t->getBaseText(); } static function basepagenamee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return wfUrlEncode( str_replace( ' ', '_', $t->getBaseText() ) ); - } + } static function talkpagename( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) || !$t->canTalk() ) + if ( is_null( $t ) || !$t->canTalk() ) return ''; return wfEscapeWikiText( $t->getTalkPage()->getPrefixedText() ); } static function talkpagenamee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) || !$t->canTalk() ) + if ( is_null( $t ) || !$t->canTalk() ) return ''; return $t->getTalkPage()->getPrefixedUrl(); } static function subjectpagename( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return wfEscapeWikiText( $t->getSubjectPage()->getPrefixedText() ); } static function subjectpagenamee( $parser, $title = null ) { $t = Title::newFromText( $title ); - if ( is_null($t) ) + if ( is_null( $t ) ) return ''; return $t->getSubjectPage()->getPrefixedUrl(); } - + /** * Return the number of pages in the given category, or 0 if it's nonexis- * tent. This is an expensive parser function and can't be called too many @@ -443,16 +469,16 @@ class CoreParserFunctions { * Return the size of the given page, or 0 if it's nonexistent. This is an * expensive parser function and can't be called too many times per page. * - * @FIXME This doesn't work correctly on preview for getting the size of - * the current page. - * @FIXME Title::getLength() documentation claims that it adds things to - * the link cache, so the local cache here should be unnecessary, but in - * fact calling getLength() repeatedly for the same $page does seem to + * @todo Fixme: This doesn't work correctly on preview for getting the size + * of the current page. + * @todo Fixme: Title::getLength() documentation claims that it adds things + * to the link cache, so the local cache here should be unnecessary, but + * in fact calling getLength() repeatedly for the same $page does seem to * run one query for each call? */ static function pagesize( $parser, $page = '', $raw = null ) { static $cache = array(); - $title = Title::newFromText($page); + $title = Title::newFromText( $page ); if( !is_object( $title ) ) { $cache[$page] = 0; @@ -466,16 +492,16 @@ class CoreParserFunctions { if( isset( $cache[$page] ) ) { $length = $cache[$page]; } elseif( $parser->incrementExpensiveFunctionCount() ) { - $rev = Revision::newFromTitle($title); + $rev = Revision::newFromTitle( $title ); $id = $rev ? $rev->getPage() : 0; $length = $cache[$page] = $rev ? $rev->getSize() : 0; - + // Register dependency in templatelinks $parser->mOutput->addTemplate( $title, $id, $rev ? $rev->getId() : 0 ); - } + } return self::formatRaw( $length, $raw ); } - + /** * Returns the requested protection level for the current page */ @@ -496,12 +522,12 @@ class CoreParserFunctions { * Unicode-safe str_pad with the restriction that $length is forced to be <= 500 */ static function pad( $string, $length, $padding = '0', $direction = STR_PAD_RIGHT ) { - $lengthOfPadding = mb_strlen( $padding ); + $lengthOfPadding = mb_strlen( $padding ); if ( $lengthOfPadding == 0 ) return $string; - + # The remaining length to add counts down to 0 as padding is added $length = min( $length, 500 ) - mb_strlen( $string ); - # $finalPadding is just $padding repeated enough times so that + # $finalPadding is just $padding repeated enough times so that # mb_strlen( $string ) + mb_strlen( $finalPadding ) == $length $finalPadding = ''; while ( $length > 0 ) { @@ -510,7 +536,7 @@ class CoreParserFunctions { $finalPadding .= mb_substr( $padding, 0, $length ); $length -= $lengthOfPadding; } - + if ( $direction == STR_PAD_LEFT ) { return $finalPadding . $string; } else { diff --git a/includes/parser/CoreTagHooks.php b/includes/parser/CoreTagHooks.php new file mode 100644 index 00000000..7cc8260e --- /dev/null +++ b/includes/parser/CoreTagHooks.php @@ -0,0 +1,49 @@ +<?php + +class CoreTagHooks { + static function register( $parser ) { + global $wgRawHtml, $wgUseTeX; + $parser->setHook( 'pre', array( __CLASS__, 'pre' ) ); + $parser->setHook( 'nowiki', array( __CLASS__, 'nowiki' ) ); + $parser->setHook( 'gallery', array( __CLASS__, 'gallery' ) ); + if ( $wgRawHtml ) { + $parser->setHook( 'html', array( __CLASS__, 'html' ) ); + } + if ( $wgUseTeX ) { + $parser->setHook( 'math', array( __CLASS__, 'math' ) ); + } + } + + static function pre( $text, $attribs, $parser ) { + // Backwards-compatibility hack + $content = StringUtils::delimiterReplace( '<nowiki>', '</nowiki>', '$1', $text, 'i' ); + + $attribs = Sanitizer::validateTagAttributes( $attribs, 'pre' ); + return Xml::openElement( 'pre', $attribs ) . + Xml::escapeTagsOnly( $content ) . + '</pre>'; + } + + static function html( $content, $attributes, $parser ) { + global $wgRawHtml; + if( $wgRawHtml ) { + return array( $content, 'markerType' => 'nowiki' ); + } else { + throw new MWException( '<html> extension tag encountered unexpectedly' ); + } + } + + static function nowiki( $content, $attributes, $parser ) { + $content = strtr( $content, array( '-{' => '-{', '}-' => '}-' ) ); + return array( Xml::escapeTagsOnly( $content ), 'markerType' => 'nowiki' ); + } + + static function math( $content, $attributes, $parser ) { + global $wgContLang; + return $wgContLang->armourMath( MathRenderer::renderMath( $content, $attributes ) ); + } + + static function gallery( $content, $attributes, $parser ) { + return $parser->renderImageGallery( $content, $attributes ); + } +} diff --git a/includes/parser/DateFormatter.php b/includes/parser/DateFormatter.php index aa6415e4..602bcff3 100644 --- a/includes/parser/DateFormatter.php +++ b/includes/parser/DateFormatter.php @@ -48,10 +48,10 @@ class DateFormatter $this->prxISO2 = '\[\[(-?\d{4})-(\d{2})-(\d{2})\]\]'; # Real regular expressions - $this->regexes[self::DMY] = "/{$this->prxDM} *,? *{$this->prxY}{$this->regexTrail}"; - $this->regexes[self::YDM] = "/{$this->prxY} *,? *{$this->prxDM}{$this->regexTrail}"; - $this->regexes[self::MDY] = "/{$this->prxMD} *,? *{$this->prxY}{$this->regexTrail}"; - $this->regexes[self::YMD] = "/{$this->prxY} *,? *{$this->prxMD}{$this->regexTrail}"; + $this->regexes[self::DMY] = "/{$this->prxDM}(?: *, *| +){$this->prxY}{$this->regexTrail}"; + $this->regexes[self::YDM] = "/{$this->prxY}(?: *, *| +){$this->prxDM}{$this->regexTrail}"; + $this->regexes[self::MDY] = "/{$this->prxMD}(?: *, *| +){$this->prxY}{$this->regexTrail}"; + $this->regexes[self::YMD] = "/{$this->prxY}(?: *, *| +){$this->prxMD}{$this->regexTrail}"; $this->regexes[self::DM] = "/{$this->prxDM}{$this->regexTrail}"; $this->regexes[self::MD] = "/{$this->prxMD}{$this->regexTrail}"; $this->regexes[self::ISO1] = "/{$this->prxISO1}{$this->regexTrail}"; @@ -268,7 +268,7 @@ class DateFormatter $isoDate = implode( '-', $isoBits );; // Output is not strictly HTML (it's wikitext), but <span> is whitelisted. - $text = Xml::tags( 'span', + $text = Html::rawElement( 'span', array( 'class' => 'mw-formatted-date', 'title' => $isoDate ), $text ); return $text; diff --git a/includes/parser/LinkHolderArray.php b/includes/parser/LinkHolderArray.php index 35b672b9..4f382a4f 100644 --- a/includes/parser/LinkHolderArray.php +++ b/includes/parser/LinkHolderArray.php @@ -105,6 +105,7 @@ class LinkHolderArray { } /** + * FIXME: update documentation. makeLinkObj() is deprecated. * Replace <!--LINK--> link placeholders with actual links, in the buffer * Placeholders created in Skin::makeLinkObj() * Returns an array of link CSS classes, indexed by PDBK. @@ -228,10 +229,12 @@ class LinkHolderArray { $linkCache->addBadLinkObj( $title ); $colours[$pdbk] = 'new'; $output->addLink( $title, 0 ); + // FIXME: replace deprecated makeBrokenLinkObj() by link() $replacePairs[$searchkey] = $sk->makeBrokenLinkObj( $title, $entry['text'], $query ); } else { + // FIXME: replace deprecated makeColouredLinkObj() by link() $replacePairs[$searchkey] = $sk->makeColouredLinkObj( $title, $colours[$pdbk], $entry['text'], $query ); diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php index e6a68782..4f672f5b 100644 --- a/includes/parser/Parser.php +++ b/includes/parser/Parser.php @@ -91,8 +91,9 @@ class Parser */ # Persistent: var $mTagHooks, $mTransparentTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables, - $mImageParams, $mImageParamsMagicArray, $mStripList, $mMarkerIndex, $mPreprocessor, - $mExtLinkBracketedRegex, $mUrlProtocols, $mDefaultStripList, $mVarCache, $mConf; + $mSubstWords, $mImageParams, $mImageParamsMagicArray, $mStripList, $mMarkerIndex, + $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols, $mDefaultStripList, + $mVarCache, $mConf, $mFunctionTagHooks; # Cleared with clearState(): @@ -103,7 +104,6 @@ class Parser var $mTplExpandCache; // empty-frame expansion cache var $mTplRedirCache, $mTplDomCache, $mHeadings, $mDoubleUnderscores; var $mExpensiveFunctionCount; // number of expensive parser function calls - var $mFileCache; # Temporary # These are variables reset at least once per parse regardless of $clearState @@ -127,8 +127,9 @@ class Parser $this->mTagHooks = array(); $this->mTransparentTagHooks = array(); $this->mFunctionHooks = array(); + $this->mFunctionTagHooks = array(); $this->mFunctionSynonyms = array( 0 => array(), 1 => array() ); - $this->mDefaultStripList = $this->mStripList = array( 'nowiki', 'gallery' ); + $this->mDefaultStripList = $this->mStripList = array(); $this->mUrlProtocols = wfUrlProtocols(); $this->mExtLinkBracketedRegex = '/\[(\b(' . wfUrlProtocols() . ')'. '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x0a\\x0d]*?)\]/S'; @@ -171,8 +172,8 @@ class Parser wfProfileIn( __METHOD__ ); - $this->setHook( 'pre', array( $this, 'renderPreTag' ) ); CoreParserFunctions::register( $this ); + CoreTagHooks::register( $this ); $this->initialiseVariables(); wfRunHooks( 'ParserFirstCallInit', array( &$this ) ); @@ -200,6 +201,7 @@ class Parser $this->mLinkHolders = new LinkHolderArray( $this ); $this->mLinkID = 0; $this->mRevisionTimestamp = $this->mRevisionId = null; + $this->mVarCache = array(); /** * Prefix for temporary replacement strings for the multipass parser. @@ -230,7 +232,6 @@ class Parser $this->mHeadings = array(); $this->mDoubleUnderscores = array(); $this->mExpensiveFunctionCount = 0; - $this->mFileCache = array(); # Fix cloning if ( isset( $this->mPreprocessor ) && $this->mPreprocessor->parser !== $this ) { @@ -255,9 +256,10 @@ class Parser * Set the context title */ function setTitle( $t ) { - if ( !$t || $t instanceof FakeTitle ) { - $t = Title::newFromText( 'NO TITLE' ); - } + if ( !$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; @@ -274,7 +276,7 @@ class Parser */ function uniqPrefix() { if( !isset( $this->mUniqPrefix ) ) { - // @fixme this is probably *horribly wrong* + // @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. @@ -303,7 +305,7 @@ class Parser * to internalParse() which does all the real work. */ - global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang; + global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang, $wgDisableLangConversion, $wgDisableTitleConversion; $fname = __METHOD__.'-' . wfGetCaller(); wfProfileIn( __METHOD__ ); wfProfileIn( $fname ); @@ -313,7 +315,8 @@ class Parser } $this->mOptions = $options; - $this->setTitle( $title ); + $this->setTitle( $title ); // Page title has to be set for the pre-processor + $oldRevisionId = $this->mRevisionId; $oldRevisionTimestamp = $this->mRevisionTimestamp; if( $revid !== null ) { @@ -325,6 +328,7 @@ class Parser # No more strip! wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) ); $text = $this->internalParse( $text ); + $text = $this->mStripState->unstripGeneral( $text ); # Clean up special characters, only run once, next-to-last before doBlockLevels @@ -342,11 +346,51 @@ class Parser $this->replaceLinkHolders( $text ); - # the position of the parserConvert() call should not be changed. it - # assumes that the links are all replaced and the only thing left - # is the <nowiki> mark. - # Side-effects: this calls $this->mOutput->setTitleText() - $text = $wgContLang->parserConvert( $text, $this ); + /** + * The page doesn't get language converted if + * a) It's disabled + * b) Content isn't converted + * c) It's a conversion table + */ + if ( !( $wgDisableLangConversion + || isset( $this->mDoubleUnderscores['nocontentconvert'] ) + || $this->mTitle->isConversionTable() ) ) { + + # The position of the convert() call should not be changed. it + # assumes that the links are all replaced and the only thing left + # is the <nowiki> mark. + + $text = $wgContLang->convert( $text ); + } + + /** + * 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. + */ + if ( !( $wgDisableLangConversion + || $wgDisableTitleConversion + || isset( $this->mDoubleUnderscores['nocontentconvert'] ) + || isset( $this->mDoubleUnderscores['notitleconvert'] ) + || $this->mOutput->getDisplayTitle() !== false ) ) + { + $convruletitle = $wgContLang->getConvRuleTitle(); + if ( $convruletitle ) { + $this->mOutput->setTitleText( $convruletitle ); + } else { + $titleText = $wgContLang->convertTitle( $title ); + $this->mOutput->setTitleText( $titleText ); + } + } $text = $this->mStripState->unstripNoWiki( $text ); @@ -412,7 +456,6 @@ class Parser # Information on include size limits, for the benefit of users who try to skirt them if ( $this->mOptions->getEnableLimitReport() ) { - global $wgExpensiveParserFunctionLimit; $max = $this->mOptions->getMaxIncludeSize(); $PFreport = "Expensive parser function count: {$this->mExpensiveFunctionCount}/$wgExpensiveParserFunctionLimit\n"; $limitReport = @@ -425,6 +468,7 @@ class Parser $text .= "\n<!-- \n$limitReport-->\n"; } $this->mOutput->setText( $text ); + $this->mRevisionId = $oldRevisionId; $this->mRevisionTimestamp = $oldRevisionTimestamp; wfProfileOut( $fname ); @@ -436,12 +480,17 @@ class Parser /** * Recursive parser entry point that can be called from an extension tag * hook. + * + * 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 */ - function recursiveTagParse( $text ) { + function recursiveTagParse( $text, $frame=false ) { wfProfileIn( __METHOD__ ); wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) ); wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) ); - $text = $this->internalParse( $text ); + $text = $this->internalParse( $text, false, $frame ); wfProfileOut( __METHOD__ ); return $text; } @@ -529,9 +578,9 @@ class Parser $matches = array(); $taglist = implode( '|', $elements ); - $start = "/<($taglist)(\\s+[^>]*?|\\s*?)(\/?>)|<(!--)/i"; + $start = "/<($taglist)(\\s+[^>]*?|\\s*?)(\/?" . ">)|<(!--)/i"; - while ( '' != $text ) { + while ( $text != '' ) { $p = preg_split( $start, $text, 2, PREG_SPLIT_DELIM_CAPTURE ); $stripped .= $p[0]; if( count( $p ) < 5 ) { @@ -589,15 +638,7 @@ class Parser * Get a list of strippable XML-like elements */ function getStripList() { - global $wgRawHtml; - $elements = $this->mStripList; - if( $wgRawHtml ) { - $elements[] = 'html'; - } - if( $this->mOptions->getUseTeX() ) { - $elements[] = 'math'; - } - return $elements; + return $this->mStripList; } /** @@ -648,14 +689,14 @@ class Parser $this->mStripState->general->setPair( $rnd, $text ); return $rnd; } - + /** * Interface with html tidy * @deprecated Use MWTidy::tidy() */ public static function tidy( $text ) { wfDeprecated( __METHOD__ ); - return MWTidy::tidy( $text ); + return MWTidy::tidy( $text ); } /** @@ -693,11 +734,11 @@ class Parser $attributes = Sanitizer::fixTagAttributes ( $attributes , 'table' ); $outLine = str_repeat( '<dl><dd>' , $indent_level ) . "<table{$attributes}>"; - array_push ( $td_history , false ); - array_push ( $last_tag_history , '' ); - array_push ( $tr_history , false ); - array_push ( $tr_attributes , '' ); - array_push ( $has_opened_tr , false ); + array_push( $td_history , false ); + array_push( $last_tag_history , '' ); + array_push( $tr_history , false ); + array_push( $tr_attributes , '' ); + array_push( $has_opened_tr , false ); } else if ( count ( $td_history ) == 0 ) { // Don't do any of the following $out .= $outLine."\n"; @@ -726,9 +767,9 @@ class Parser // 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 ); + $attributes = Sanitizer::fixTagAttributes( $attributes, 'tr' ); + array_pop( $tr_attributes ); + array_push( $tr_attributes, $attributes ); $line = ''; $last_tag = array_pop ( $last_tag_history ); @@ -862,17 +903,33 @@ class Parser * * @private */ - function internalParse( $text ) { - $isMain = true; + function internalParse( $text, $isMain = true, $frame=false ) { wfProfileIn( __METHOD__ ); + $origText = $text; + # Hook to suspend the parser in this state if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) { wfProfileOut( __METHOD__ ); return $text ; } - $text = $this->replaceVariables( $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 ) + $flag = 0; + 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 { + $text = $this->replaceVariables( $text ); + } + $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), false, array_keys( $this->mTransparentTagHooks ) ); wfRunHooks( 'InternalParseBeforeLinks', array( &$this, &$text, &$this->mStripState ) ); @@ -885,6 +942,7 @@ class Parser $text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text ); $text = $this->doDoubleUnderscore( $text ); + $text = $this->doHeadings( $text ); if( $this->mOptions->getUseDynamicDates() ) { $df = DateFormatter::getInstance(); @@ -899,7 +957,7 @@ class Parser $text = str_replace($this->mUniqPrefix.'NOPARSE', '', $text); $text = $this->doMagicLinks( $text ); - $text = $this->formatHeadings( $text, $isMain ); + $text = $this->formatHeadings( $text, $origText, $isMain ); wfProfileOut( __METHOD__ ); return $text; @@ -908,7 +966,7 @@ class Parser /** * Replace special strings like "ISBN xxx" and "RFC xxx" with * magic external links. - * + * * DML * @private */ @@ -918,7 +976,7 @@ class Parser $urlChar = self::EXT_LINK_URL_CLASS; $text = preg_replace_callback( '!(?: # Start cases - (<a.*?</a>) | # m[1]: Skip link text + (<a.*?</a>) | # m[1]: Skip link text (<.*?>) | # m[2]: Skip stuff inside HTML elements' . " (\\b(?:$prots)$urlChar+) | # m[3]: Free external links" . ' (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number @@ -944,13 +1002,16 @@ 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'; + $CssClass = 'mw-magiclink-rfc'; $id = $m[4]; } elseif ( substr( $m[0], 0, 4 ) === 'PMID' ) { $keyword = 'PMID'; $urlmsg = 'pubmedurl'; + $CssClass = 'mw-magiclink-pmid'; $id = $m[4]; } else { throw new MWException( __METHOD__.': unrecognised match type "' . @@ -958,7 +1019,7 @@ class Parser } $url = wfMsg( $urlmsg, $id); $sk = $this->mOptions->getSkin(); - $la = $sk->getExternalLinkAttributes( $url, $keyword.$id ); + $la = $sk->getExternalLinkAttributes( "external $CssClass" ); return "<a href=\"{$url}\"{$la}>{$keyword} {$id}</a>"; } elseif ( isset( $m[5] ) && $m[5] !== '' ) { # ISBN @@ -971,7 +1032,7 @@ class Parser $titleObj = SpecialPage::getTitleFor( 'Booksources', $num ); return'<a href="' . $titleObj->escapeLocalUrl() . - "\" class=\"internal\">ISBN $isbn</a>"; + "\" class=\"internal mw-magiclink-isbn\">ISBN $isbn</a>"; } else { return $m[0]; } @@ -1017,7 +1078,7 @@ class Parser $text = $this->maybeMakeExternalImage( $url ); if ( $text === false ) { # Not an image, make a link - $text = $sk->makeExternalLink( $url, $wgContLang->markNoConversion($url), true, 'free', + $text = $sk->makeExternalLink( $url, $wgContLang->markNoConversion($url), true, 'free', $this->getExternalLinkAttribs( $url ) ); # Register it in the output object... # Replace unnecessary URL escape codes with their equivalent characters @@ -1457,7 +1518,7 @@ class Parser wfProfileIn( __METHOD__.'-setup' ); static $tc = FALSE, $e1, $e1_img; # the % is needed to support urlencoded titles as well - if ( !$tc ) { + if ( !$tc ) { $tc = Title::legalChars() . '#%'; # Match a link having the form [[namespace:link|alternate]]trail $e1 = "/^([{$tc}]+)(?:\\|(.+?))?]](.*)\$/sD"; @@ -1581,29 +1642,29 @@ class Parser # Don't allow internal links to pages containing # PROTO: where PROTO is a valid URL protocol; these # should be external links. - if (preg_match('/^\b(?:' . wfUrlProtocols() . ')/', $m[1])) { + if ( preg_match( '/^\b(?:' . wfUrlProtocols() . ')/', $m[1] ) ) { $s .= $prefix . '[[' . $line ; wfProfileOut( __METHOD__."-misc" ); continue; } # Make subpage if necessary - if( $useSubpages ) { + if ( $useSubpages ) { $link = $this->maybeDoSubpageLink( $m[1], $text ); } else { $link = $m[1]; } - $noforce = (substr($m[1], 0, 1) !== ':'); + $noforce = (substr( $m[1], 0, 1 ) !== ':'); if (!$noforce) { # Strip off leading ':' - $link = substr($link, 1); + $link = substr( $link, 1 ); } wfProfileOut( __METHOD__."-misc" ); wfProfileIn( __METHOD__."-title" ); - $nt = Title::newFromText( $this->mStripState->unstripNoWiki($link) ); - if( $nt === NULL ) { + $nt = Title::newFromText( $this->mStripState->unstripNoWiki( $link ) ); + if ( $nt === null ) { $s .= $prefix . '[[' . $line; wfProfileOut( __METHOD__."-title" ); continue; @@ -1613,9 +1674,9 @@ class Parser $iw = $nt->getInterWiki(); wfProfileOut( __METHOD__."-title" ); - if ($might_be_img) { # if this is actually an invalid link + 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 @@ -1658,15 +1719,15 @@ class Parser wfProfileOut( __METHOD__."-might_be_img" ); } - $wasblank = ( '' == $text ); - if( $wasblank ) $text = $link; + $wasblank = ( $text == '' ); + if ( $wasblank ) $text = $link; # Link not escaped by : , create the various objects - if( $noforce ) { + if ( $noforce ) { # Interwikis wfProfileIn( __METHOD__."-interwiki" ); - if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) { + if ( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) { $this->mOutput->addLanguageLink( $nt->getFullText() ); $s = rtrim($s . $prefix); $s .= trim($trail, "\n") == '' ? '': $prefix . $trail; @@ -1678,14 +1739,23 @@ class Parser if ( $ns == NS_FILE ) { wfProfileIn( __METHOD__."-image" ); if ( !wfIsBadImage( $nt->getDBkey(), $this->mTitle ) ) { - # recursively parse links inside the image caption - # actually, this will parse them in any other parameters, too, - # but it might be hard to fix that, and it doesn't matter ATM - $text = $this->replaceExternalLinks($text); - $holders->merge( $this->replaceInternalLinks2( $text ) ); - + if ( $wasblank ) { + # if no parameters were passed, $text + # becomes something like "File:Foo.png", + # which we don't want to pass on to the + # image generator + $text = ''; + } else { + # 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); + $holders->merge( $this->replaceInternalLinks2( $text ) ); + } # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them $s .= $prefix . $this->armorLinks( $this->makeImage( $nt, $text, $holders ) ) . $trail; + } else { + $s .= $prefix . $trail; } $this->mOutput->addImage( $nt->getDBkey() ); wfProfileOut( __METHOD__."-image" ); @@ -1793,6 +1863,7 @@ class Parser function makeKnownLinkHolder( $nt, $text = '', $query = '', $trail = '', $prefix = '' ) { list( $inside, $trail ) = Linker::splitTrail( $trail ); $sk = $this->mOptions->getSkin(); + // FIXME: use link() instead of deprecated makeKnownLinkObj() $link = $sk->makeKnownLinkObj( $nt, $text, $query, $inside, $prefix ); return $this->armorLinks( $link ) . $trail; } @@ -1829,75 +1900,7 @@ class Parser * @private */ function maybeDoSubpageLink($target, &$text) { - # Valid link forms: - # Foobar -- normal - # :Foobar -- override special treatment of prefix (images, language links) - # /Foobar -- convert to CurrentPage/Foobar - # /Foobar/ -- convert to CurrentPage/Foobar, strip the initial / from text - # ../ -- convert to CurrentPage, from CurrentPage/CurrentSubPage - # ../Foobar -- convert to CurrentPage/Foobar, from CurrentPage/CurrentSubPage - - wfProfileIn( __METHOD__ ); - $ret = $target; # default return value is no change - - # Some namespaces don't allow subpages, - # so only perform processing if subpages are allowed - if( $this->areSubpagesAllowed() ) { - $hash = strpos( $target, '#' ); - if( $hash !== false ) { - $suffix = substr( $target, $hash ); - $target = substr( $target, 0, $hash ); - } else { - $suffix = ''; - } - # bug 7425 - $target = trim( $target ); - # Look at the first character - if( $target != '' && $target{0} === '/' ) { - # / at end means we don't want the slash to be shown - $m = array(); - $trailingSlashes = preg_match_all( '%(/+)$%', $target, $m ); - if( $trailingSlashes ) { - $noslash = $target = substr( $target, 1, -strlen($m[0][0]) ); - } else { - $noslash = substr( $target, 1 ); - } - - $ret = $this->mTitle->getPrefixedText(). '/' . trim($noslash) . $suffix; - if( '' === $text ) { - $text = $target . $suffix; - } # this might be changed for ugliness reasons - } else { - # check for .. subpage backlinks - $dotdotcount = 0; - $nodotdot = $target; - while( strncmp( $nodotdot, "../", 3 ) == 0 ) { - ++$dotdotcount; - $nodotdot = substr( $nodotdot, 3 ); - } - if($dotdotcount > 0) { - $exploded = explode( '/', $this->mTitle->GetPrefixedText() ); - if( count( $exploded ) > $dotdotcount ) { # not allowed to go below top level page - $ret = implode( '/', array_slice( $exploded, 0, -$dotdotcount ) ); - # / at the end means don't show full path - if( substr( $nodotdot, -1, 1 ) === '/' ) { - $nodotdot = substr( $nodotdot, 0, -1 ); - if( '' === $text ) { - $text = $nodotdot . $suffix; - } - } - $nodotdot = trim( $nodotdot ); - if( $nodotdot != '' ) { - $ret .= '/' . $nodotdot; - } - $ret .= $suffix; - } - } - } - } - - wfProfileOut( __METHOD__ ); - return $ret; + return Linker::normalizeSubpageLink( $this->mTitle, $target, $text ); } /**#@+ @@ -1906,7 +1909,7 @@ class Parser */ /* private */ function closeParagraph() { $result = ''; - if ( '' != $this->mLastSection ) { + if ( $this->mLastSection != '' ) { $result = '</' . $this->mLastSection . ">\n"; } $this->mInPre = false; @@ -1933,9 +1936,9 @@ class Parser $result = $this->closeParagraph(); if ( '*' === $char ) { $result .= '<ul><li>'; } - else if ( '#' === $char ) { $result .= '<ol><li>'; } - else if ( ':' === $char ) { $result .= '<dl><dd>'; } - else if ( ';' === $char ) { + elseif ( '#' === $char ) { $result .= '<ol><li>'; } + elseif ( ':' === $char ) { $result .= '<dl><dd>'; } + elseif ( ';' === $char ) { $result .= '<dl><dt>'; $this->mDTopen = true; } @@ -1946,7 +1949,7 @@ class Parser /* private */ function nextItem( $char ) { if ( '*' === $char || '#' === $char ) { return '</li><li>'; } - else if ( ':' === $char || ';' === $char ) { + elseif ( ':' === $char || ';' === $char ) { $close = '</dd>'; if ( $this->mDTopen ) { $close = '</dt>'; } if ( ';' === $char ) { @@ -1962,8 +1965,8 @@ class Parser /* private */ function closeList( $char ) { if ( '*' === $char ) { $text = '</li></ul>'; } - else if ( '#' === $char ) { $text = '</li></ol>'; } - else if ( ':' === $char ) { + elseif ( '#' === $char ) { $text = '</li></ol>'; } + elseif ( ':' === $char ) { if ( $this->mDTopen ) { $this->mDTopen = false; $text = '</dt></dl>'; @@ -1979,6 +1982,7 @@ class Parser /** * Make lists from lines starting with ':', '*', '#', etc. (DBL) * + * @param $linestart bool whether or not this is at the start of a line. * @private * @return string the lists rendered as HTML */ @@ -2003,16 +2007,24 @@ class Parser $linestart = true; continue; } + // * = ul + // # = ol + // ; = dt + // : = dd $lastPrefixLength = strlen( $lastPrefix ); $preCloseMatch = preg_match('/<\\/pre/i', $oLine ); $preOpenMatch = preg_match('/<pre/i', $oLine ); + // If not in a <pre> element, scan for and figure out what prefixes are there. if ( !$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. $prefix2 = str_replace( ';', ':', $prefix ); $t = substr( $oLine, $prefixLength ); $this->mInPre = (bool)$preOpenMatch; @@ -2041,17 +2053,24 @@ class Parser } } } 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 ) { $output .= $this->closeList( $lastPrefix[$lastPrefixLength-1] ); --$lastPrefixLength; } + + // Continue the current prefix if appropriate. if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) { $output .= $this->nextItem( $prefix[$commonPrefixLength-1] ); } + + // Open prefixes where appropriate. while ( $prefixLength > $commonPrefixLength ) { $char = substr( $prefix, $commonPrefixLength, 1 ); $output .= $this->openList( $char ); @@ -2067,6 +2086,8 @@ class Parser } $lastPrefix = $prefix2; } + + // If we have no prefixes, go to paragraph mode. if( 0 == $prefixLength ) { wfProfileIn( __METHOD__."-paragraph" ); # No prefix (not in list)--go to paragraph mode @@ -2098,7 +2119,7 @@ class Parser $t = substr( $t, 1 ); } else { // paragraph - if ( '' == trim($t) ) { + if ( trim($t) == '' ) { if ( $paragraphStack ) { $output .= $paragraphStack.'<br />'; $paragraphStack = false; @@ -2138,7 +2159,7 @@ class Parser $output .= $this->closeList( $prefix2[$prefixLength-1] ); --$prefixLength; } - if ( '' != $this->mLastSection ) { + if ( $this->mLastSection != '' ) { $output .= '</' . $this->mLastSection . '>'; $this->mLastSection = ''; } @@ -2315,8 +2336,9 @@ class Parser * * @private */ - function getVariableValue( $index ) { - global $wgContLang, $wgSitename, $wgServer, $wgServerName, $wgScriptPath; + function getVariableValue( $index, $frame=false ) { + global $wgContLang, $wgSitename, $wgServer, $wgServerName; + global $wgScriptPath, $wgStylePath; /** * Some of these require message or data lookups and can be @@ -2334,13 +2356,13 @@ class Parser # Use the time zone global $wgLocaltimezone; if ( isset( $wgLocaltimezone ) ) { - $oldtz = getenv( 'TZ' ); - putenv( 'TZ='.$wgLocaltimezone ); + $oldtz = date_default_timezone_get(); + date_default_timezone_set( $wgLocaltimezone ); } - wfSuppressWarnings(); // E_STRICT system time bitching $localTimestamp = date( 'YmdHis', $ts ); $localMonth = date( 'm', $ts ); + $localMonth1 = date( 'n', $ts ); $localMonthName = date( 'n', $ts ); $localDay = date( 'j', $ts ); $localDay2 = date( 'd', $ts ); @@ -2349,175 +2371,240 @@ class Parser $localYear = date( 'Y', $ts ); $localHour = date( 'H', $ts ); if ( isset( $wgLocaltimezone ) ) { - putenv( 'TZ='.$oldtz ); + date_default_timezone_set( $oldtz ); } - wfRestoreWarnings(); switch ( $index ) { case 'currentmonth': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'm', $ts ) ); + $value = $wgContLang->formatNum( gmdate( 'm', $ts ) ); + break; + case 'currentmonth1': + $value = $wgContLang->formatNum( gmdate( 'n', $ts ) ); + break; case 'currentmonthname': - return $this->mVarCache[$index] = $wgContLang->getMonthName( gmdate( 'n', $ts ) ); + $value = $wgContLang->getMonthName( gmdate( 'n', $ts ) ); + break; case 'currentmonthnamegen': - return $this->mVarCache[$index] = $wgContLang->getMonthNameGen( gmdate( 'n', $ts ) ); + $value = $wgContLang->getMonthNameGen( gmdate( 'n', $ts ) ); + break; case 'currentmonthabbrev': - return $this->mVarCache[$index] = $wgContLang->getMonthAbbreviation( gmdate( 'n', $ts ) ); + $value = $wgContLang->getMonthAbbreviation( gmdate( 'n', $ts ) ); + break; case 'currentday': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'j', $ts ) ); + $value = $wgContLang->formatNum( gmdate( 'j', $ts ) ); + break; case 'currentday2': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'd', $ts ) ); + $value = $wgContLang->formatNum( gmdate( 'd', $ts ) ); + break; case 'localmonth': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localMonth ); + $value = $wgContLang->formatNum( $localMonth ); + break; + case 'localmonth1': + $value = $wgContLang->formatNum( $localMonth1 ); + break; case 'localmonthname': - return $this->mVarCache[$index] = $wgContLang->getMonthName( $localMonthName ); + $value = $wgContLang->getMonthName( $localMonthName ); + break; case 'localmonthnamegen': - return $this->mVarCache[$index] = $wgContLang->getMonthNameGen( $localMonthName ); + $value = $wgContLang->getMonthNameGen( $localMonthName ); + break; case 'localmonthabbrev': - return $this->mVarCache[$index] = $wgContLang->getMonthAbbreviation( $localMonthName ); + $value = $wgContLang->getMonthAbbreviation( $localMonthName ); + break; case 'localday': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localDay ); + $value = $wgContLang->formatNum( $localDay ); + break; case 'localday2': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localDay2 ); + $value = $wgContLang->formatNum( $localDay2 ); + break; case 'pagename': - return wfEscapeWikiText( $this->mTitle->getText() ); + $value = wfEscapeWikiText( $this->mTitle->getText() ); + break; case 'pagenamee': - return $this->mTitle->getPartialURL(); + $value = $this->mTitle->getPartialURL(); + break; case 'fullpagename': - return wfEscapeWikiText( $this->mTitle->getPrefixedText() ); + $value = wfEscapeWikiText( $this->mTitle->getPrefixedText() ); + break; case 'fullpagenamee': - return $this->mTitle->getPrefixedURL(); + $value = $this->mTitle->getPrefixedURL(); + break; case 'subpagename': - return wfEscapeWikiText( $this->mTitle->getSubpageText() ); + $value = wfEscapeWikiText( $this->mTitle->getSubpageText() ); + break; case 'subpagenamee': - return $this->mTitle->getSubpageUrlForm(); + $value = $this->mTitle->getSubpageUrlForm(); + break; case 'basepagename': - return wfEscapeWikiText( $this->mTitle->getBaseText() ); + $value = wfEscapeWikiText( $this->mTitle->getBaseText() ); + break; case 'basepagenamee': - return wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ); + $value = wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ); + break; case 'talkpagename': if( $this->mTitle->canTalk() ) { $talkPage = $this->mTitle->getTalkPage(); - return wfEscapeWikiText( $talkPage->getPrefixedText() ); + $value = wfEscapeWikiText( $talkPage->getPrefixedText() ); } else { - return ''; + $value = ''; } + break; case 'talkpagenamee': if( $this->mTitle->canTalk() ) { $talkPage = $this->mTitle->getTalkPage(); - return $talkPage->getPrefixedUrl(); + $value = $talkPage->getPrefixedUrl(); } else { - return ''; + $value = ''; } + break; case 'subjectpagename': $subjPage = $this->mTitle->getSubjectPage(); - return wfEscapeWikiText( $subjPage->getPrefixedText() ); + $value = wfEscapeWikiText( $subjPage->getPrefixedText() ); + break; case 'subjectpagenamee': $subjPage = $this->mTitle->getSubjectPage(); - return $subjPage->getPrefixedUrl(); + $value = $subjPage->getPrefixedUrl(); + break; case 'revisionid': // Let the edit saving system know we should parse the page // *after* a revision ID has been assigned. $this->mOutput->setFlag( 'vary-revision' ); wfDebug( __METHOD__ . ": {{REVISIONID}} used, setting vary-revision...\n" ); - return $this->mRevisionId; + $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. $this->mOutput->setFlag( 'vary-revision' ); wfDebug( __METHOD__ . ": {{REVISIONDAY}} used, setting vary-revision...\n" ); - return intval( substr( $this->getRevisionTimestamp(), 6, 2 ) ); + $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. $this->mOutput->setFlag( 'vary-revision' ); wfDebug( __METHOD__ . ": {{REVISIONDAY2}} used, setting vary-revision...\n" ); - return substr( $this->getRevisionTimestamp(), 6, 2 ); + $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. $this->mOutput->setFlag( 'vary-revision' ); wfDebug( __METHOD__ . ": {{REVISIONMONTH}} used, setting vary-revision...\n" ); - return intval( substr( $this->getRevisionTimestamp(), 4, 2 ) ); + $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. $this->mOutput->setFlag( 'vary-revision' ); wfDebug( __METHOD__ . ": {{REVISIONYEAR}} used, setting vary-revision...\n" ); - return substr( $this->getRevisionTimestamp(), 0, 4 ); + $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. $this->mOutput->setFlag( 'vary-revision' ); wfDebug( __METHOD__ . ": {{REVISIONTIMESTAMP}} used, setting vary-revision...\n" ); - return $this->getRevisionTimestamp(); + $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. $this->mOutput->setFlag( 'vary-revision' ); wfDebug( __METHOD__ . ": {{REVISIONUSER}} used, setting vary-revision...\n" ); - return $this->getRevisionUser(); + $value = $this->getRevisionUser(); + break; case 'namespace': - return str_replace('_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) ); + $value = str_replace('_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) ); + break; case 'namespacee': - return wfUrlencode( $wgContLang->getNsText( $this->mTitle->getNamespace() ) ); + $value = wfUrlencode( $wgContLang->getNsText( $this->mTitle->getNamespace() ) ); + break; case 'talkspace': - return $this->mTitle->canTalk() ? str_replace('_',' ',$this->mTitle->getTalkNsText()) : ''; + $value = $this->mTitle->canTalk() ? str_replace('_',' ',$this->mTitle->getTalkNsText()) : ''; + break; case 'talkspacee': - return $this->mTitle->canTalk() ? wfUrlencode( $this->mTitle->getTalkNsText() ) : ''; + $value = $this->mTitle->canTalk() ? wfUrlencode( $this->mTitle->getTalkNsText() ) : ''; + break; case 'subjectspace': - return $this->mTitle->getSubjectNsText(); + $value = $this->mTitle->getSubjectNsText(); + break; case 'subjectspacee': - return( wfUrlencode( $this->mTitle->getSubjectNsText() ) ); + $value = ( wfUrlencode( $this->mTitle->getSubjectNsText() ) ); + break; case 'currentdayname': - return $this->mVarCache[$index] = $wgContLang->getWeekdayName( gmdate( 'w', $ts ) + 1 ); + $value = $wgContLang->getWeekdayName( gmdate( 'w', $ts ) + 1 ); + break; case 'currentyear': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'Y', $ts ), true ); + $value = $wgContLang->formatNum( gmdate( 'Y', $ts ), true ); + break; case 'currenttime': - return $this->mVarCache[$index] = $wgContLang->time( wfTimestamp( TS_MW, $ts ), false, false ); + $value = $wgContLang->time( wfTimestamp( TS_MW, $ts ), false, false ); + break; case 'currenthour': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'H', $ts ), true ); + $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 - return $this->mVarCache[$index] = $wgContLang->formatNum( (int)gmdate( 'W', $ts ) ); + $value = $wgContLang->formatNum( (int)gmdate( 'W', $ts ) ); + break; case 'currentdow': - return $this->mVarCache[$index] = $wgContLang->formatNum( gmdate( 'w', $ts ) ); + $value = $wgContLang->formatNum( gmdate( 'w', $ts ) ); + break; case 'localdayname': - return $this->mVarCache[$index] = $wgContLang->getWeekdayName( $localDayOfWeek + 1 ); + $value = $wgContLang->getWeekdayName( $localDayOfWeek + 1 ); + break; case 'localyear': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localYear, true ); + $value = $wgContLang->formatNum( $localYear, true ); + break; case 'localtime': - return $this->mVarCache[$index] = $wgContLang->time( $localTimestamp, false, false ); + $value = $wgContLang->time( $localTimestamp, false, false ); + break; case 'localhour': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localHour, true ); + $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 - return $this->mVarCache[$index] = $wgContLang->formatNum( (int)$localWeek ); + $value = $wgContLang->formatNum( (int)$localWeek ); + break; case 'localdow': - return $this->mVarCache[$index] = $wgContLang->formatNum( $localDayOfWeek ); + $value = $wgContLang->formatNum( $localDayOfWeek ); + break; case 'numberofarticles': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::articles() ); + $value = $wgContLang->formatNum( SiteStats::articles() ); + break; case 'numberoffiles': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::images() ); + $value = $wgContLang->formatNum( SiteStats::images() ); + break; case 'numberofusers': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::users() ); + $value = $wgContLang->formatNum( SiteStats::users() ); + break; case 'numberofactiveusers': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::activeUsers() ); + $value = $wgContLang->formatNum( SiteStats::activeUsers() ); + break; case 'numberofpages': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::pages() ); + $value = $wgContLang->formatNum( SiteStats::pages() ); + break; case 'numberofadmins': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::numberingroup('sysop') ); + $value = $wgContLang->formatNum( SiteStats::numberingroup('sysop') ); + break; case 'numberofedits': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::edits() ); + $value = $wgContLang->formatNum( SiteStats::edits() ); + break; case 'numberofviews': - return $this->mVarCache[$index] = $wgContLang->formatNum( SiteStats::views() ); + $value = $wgContLang->formatNum( SiteStats::views() ); + break; case 'currenttimestamp': - return $this->mVarCache[$index] = wfTimestamp( TS_MW, $ts ); + $value = wfTimestamp( TS_MW, $ts ); + break; case 'localtimestamp': - return $this->mVarCache[$index] = $localTimestamp; + $value = $localTimestamp; + break; case 'currentversion': - return $this->mVarCache[$index] = SpecialVersion::getVersion(); + $value = SpecialVersion::getVersion(); + break; case 'sitename': return $wgSitename; case 'server': @@ -2526,6 +2613,8 @@ class Parser return $wgServerName; case 'scriptpath': return $wgScriptPath; + case 'stylepath': + return $wgStylePath; case 'directionmark': return $wgContLang->getDirMark(); case 'contentlanguage': @@ -2533,23 +2622,30 @@ class Parser return $wgContLanguageCode; default: $ret = null; - if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret ) ) ) + if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) return $ret; else return null; } + + if ( $index ) + $this->mVarCache[$index] = $value; + + return $value; } /** - * initialise the magic variables (like CURRENTMONTHNAME) + * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers * * @private */ function initialiseVariables() { wfProfileIn( __METHOD__ ); $variableIDs = MagicWord::getVariableIDs(); + $substIDs = MagicWord::getSubstIDs(); $this->mVariables = new MagicWordArray( $variableIDs ); + $this->mSubstWords = new MagicWordArray( $substIDs ); wfProfileOut( __METHOD__ ); } @@ -2607,7 +2703,7 @@ class Parser * 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 PPFrame $frame 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 @@ -2670,14 +2766,10 @@ class Parser * exceeded, provide the values (optional) */ function limitationWarn( $limitationType, $current=null, $max=null) { - $msgName = $limitationType . '-warning'; //does no harm if $current and $max are present but are unnecessary for the message - $warning = wfMsgExt( $msgName, array( 'parsemag', 'escape' ), $current, $max ); + $warning = wfMsgExt( "$limitationType-warning", array( 'parsemag', 'escape' ), $current, $max ); $this->mOutput->addWarning( $warning ); - $cat = Title::makeTitleSafe( NS_CATEGORY, wfMsgForContent( $limitationType . '-category' ) ); - if ( $cat ) { - $this->mOutput->addCategory( $cat->getDBkey(), $this->getDefaultSort() ); - } + $this->addTrackingCategory( "$limitationType-category" ); } /** @@ -2706,7 +2798,7 @@ class Parser $isLocalObj = false; # $text is a DOM node needing expansion in the current frame # Title object, where $text came from - $title = NULL; + $title = null; # $part1 is the bit before the first |, and must contain only title characters. # Various prefixes will be stripped from it later. @@ -2724,12 +2816,25 @@ class Parser # SUBST wfProfileIn( __METHOD__.'-modifiers' ); if ( !$found ) { - $mwSubst = MagicWord::get( 'subst' ); - if ( $mwSubst->matchStartAndRemove( $part1 ) xor $this->ot['wiki'] ) { - # One of two possibilities is true: - # 1) Found SUBST but not in the PST phase - # 2) Didn't find SUBST and in the PST phase - # In either case, return without further processing + + $substMatch = $this->mSubstWords->matchStartAndRemove( $part1 ); + + # Possibilities for substMatch: "subst", "safesubst" or FALSE + # Decide whether to expand template or keep wikitext as-is. + if ( $this->ot['wiki'] ) { + if ( $substMatch === false ) { + $literal = true; # literal when in PST with no prefix + } else { + $literal = false; # expand when in PST with subst: or safesubst: + } + } else { + if ( $substMatch == 'subst' ) { + $literal = true; # literal when not in PST with plain subst: + } else { + $literal = false; # expand when not in PST with safesubst: or no prefix + } + } + if ( $literal ) { $text = $frame->virtualBracketedImplode( '{{', '|', '}}', $titleWithSpaces, $args ); $isLocalObj = true; $found = true; @@ -2740,7 +2845,7 @@ class Parser if ( !$found && $args->getLength() == 0 ) { $id = $this->mVariables->matchStartToEnd( $part1 ); if ( $id !== false ) { - $text = $this->getVariableValue( $id ); + $text = $this->getVariableValue( $id, $frame ); if (MagicWord::getCacheTTL($id)>-1) $this->mOutput->mContainsOldMagic = true; $found = true; @@ -2779,7 +2884,7 @@ class Parser $function = $this->mFunctionSynonyms[1][$function]; } else { # Case insensitive functions - $function = strtolower( $function ); + $function = $wgContLang->lc( $function ); if ( isset( $this->mFunctionSynonyms[0][$function] ) ) { $function = $this->mFunctionSynonyms[0][$function]; } else { @@ -2808,13 +2913,15 @@ class Parser # Workaround for PHP bug 35229 and similar if ( !is_callable( $callback ) ) { + wfProfileOut( __METHOD__ . '-pfunc' ); + wfProfileOut( __METHOD__ ); throw new MWException( "Tag hook for $function is not callable\n" ); } $result = call_user_func_array( $callback, $allArgs ); $found = true; $noparse = true; $preprocessFlags = 0; - + if ( is_array( $result ) ) { if ( isset( $result[0] ) ) { $text = $result[0]; @@ -3118,14 +3225,11 @@ class Parser function fetchScaryTemplateMaybeFromCache($url) { global $wgTranscludeCacheExpiry; $dbr = wfGetDB(DB_SLAVE); + $tsCond = $dbr->timestamp( time() - $wgTranscludeCacheExpiry ); $obj = $dbr->selectRow('transcache', array('tc_time', 'tc_contents'), - array('tc_url' => $url)); + array('tc_url' => $url, "tc_time >= " . $dbr->addQuotes( $tsCond ) ) ); if ($obj) { - $time = $obj->tc_time; - $text = $obj->tc_contents; - if ($time && time() < $time + $wgTranscludeCacheExpiry ) { - return $text; - } + return $obj->tc_contents; } $text = Http::get($url); @@ -3135,7 +3239,7 @@ class Parser $dbw = wfGetDB(DB_MASTER); $dbw->replace('transcache', array('tc_url'), array( 'tc_url' => $url, - 'tc_time' => time(), + 'tc_time' => $dbw->timestamp( time() ), 'tc_contents' => $text)); return $text; } @@ -3204,47 +3308,47 @@ class Parser $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; - if ( $this->ot['html'] ) { + $isFunctionTag = isset( $this->mFunctionTagHooks[strtolower($name)] ) && + ( $this->ot['html'] || $this->ot['pre'] ); + if ( $isFunctionTag ) { + $markerType = 'none'; + } else { + $markerType = 'general'; + } + if ( $this->ot['html'] || $isFunctionTag ) { $name = strtolower( $name ); - $attributes = Sanitizer::decodeTagAttributes( $attrText ); if ( isset( $params['attributes'] ) ) { $attributes = $attributes + $params['attributes']; } - switch ( $name ) { - case 'html': - if( $wgRawHtml ) { - $output = $content; - break; - } else { - throw new MWException( '<html> extension tag encountered unexpectedly' ); - } - case 'nowiki': - $content = strtr($content, array('-{' => '-{', '}-' => '}-')); - $output = Xml::escapeTagsOnly( $content ); - break; - case 'math': - $output = $wgContLang->armourMath( - MathRenderer::renderMath( $content, $attributes ) ); - break; - case 'gallery': - $output = $this->renderImageGallery( $content, $attributes ); - break; - default: - if( isset( $this->mTagHooks[$name] ) ) { - # Workaround for PHP bug 35229 and similar - if ( !is_callable( $this->mTagHooks[$name] ) ) { - throw new MWException( "Tag hook for $name is not callable\n" ); - } - $output = call_user_func_array( $this->mTagHooks[$name], - array( $content, $attributes, $this ) ); - } else { - $output = '<span class="error">Invalid tag extension name: ' . - htmlspecialchars( $name ) . '</span>'; - } + + 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] ) ) { + list( $callback, $flags ) = $this->mFunctionTagHooks[$name]; + 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 ) ); + } else { + $output = '<span class="error">Invalid tag extension name: ' . + htmlspecialchars( $name ) . '</span>'; + } + + if ( is_array( $output ) ) { + // Extract flags to local scope (to override $markerType) + $flags = $output; + $output = $flags[0]; + unset( $flags[0] ); + extract( $flags ); } } else { if ( is_null( $attrText ) ) { @@ -3264,10 +3368,14 @@ class Parser } } - if ( $name === 'html' || $name === 'nowiki' ) { + if( $markerType === 'none' ) { + return $output; + } elseif ( $markerType === 'nowiki' ) { $this->mStripState->nowiki->setPair( $marker, $output ); - } else { + } elseif ( $markerType === 'general' ) { $this->mStripState->general->setPair( $marker, $output ); + } else { + throw new MWException( __METHOD__.': invalid marker type' ); } return $marker; } @@ -3308,6 +3416,7 @@ class Parser */ function doDoubleUnderscore( $text ) { wfProfileIn( __METHOD__ ); + // The position of __TOC__ needs to be recorded $mw = MagicWord::get( 'toc' ); if( $mw->match( $text ) ) { @@ -3333,28 +3442,48 @@ class Parser } if ( isset( $this->mDoubleUnderscores['hiddencat'] ) && $this->mTitle->getNamespace() == NS_CATEGORY ) { $this->mOutput->setProperty( 'hiddencat', 'y' ); - - $containerCategory = Title::makeTitleSafe( NS_CATEGORY, wfMsgForContent( 'hidden-category-category' ) ); - if ( $containerCategory ) { - $this->mOutput->addCategory( $containerCategory->getDBkey(), $this->getDefaultSort() ); - } else { - wfDebug( __METHOD__.": [[MediaWiki:hidden-category-category]] is not a valid title!\n" ); - } + $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'] ) ) { + if( isset( $this->mDoubleUnderscores['noindex'] ) && $this->mTitle->canUseNoindex() ) { $this->mOutput->setIndexPolicy( 'noindex' ); - } elseif( isset( $this->mDoubleUnderscores['index'] ) ) { + $this->addTrackingCategory( 'noindex-category' ); + } + if( isset( $this->mDoubleUnderscores['index'] ) && $this->mTitle->canUseNoindex() ){ $this->mOutput->setIndexPolicy( 'index' ); + $this->addTrackingCategory( 'index-category' ); } + wfProfileOut( __METHOD__ ); return $text; } /** + * 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 + */ + protected function addTrackingCategory( $msg ){ + $cat = wfMsgForContent( $msg ); + + # Allow tracking categories to be disabled by setting them to "-" + if( $cat === '-' ) return false; + + $containerCategory = Title::makeTitleSafe( NS_CATEGORY, $cat ); + if ( $containerCategory ) { + $this->mOutput->addCategory( $containerCategory->getDBkey(), $this->getDefaultSort() ); + return true; + } else { + wfDebug( __METHOD__.": [[MediaWiki:$msg]] is not a valid title!\n" ); + return false; + } + } + + /** * This function accomplishes several tasks: * 1) Auto-number headings if that option is enabled * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page @@ -3365,11 +3494,12 @@ class Parser * string and re-inserts the newly formatted headlines. * * @param string $text + * @param string $origText Original, untouched wikitext * @param boolean $isMain * @private */ - function formatHeadings( $text, $isMain=true ) { - global $wgMaxTocLevel, $wgContLang, $wgEnforceHtmlIds; + function formatHeadings( $text, $origText, $isMain=true ) { + global $wgMaxTocLevel, $wgContLang, $wgHtml5, $wgExperimentalHtmlIds; $doNumberHeadings = $this->mOptions->getNumberHeadings(); $showEditLink = $this->mOptions->getEditSection(); @@ -3434,6 +3564,12 @@ class Parser $prevtoclevel = 0; $markerRegex = "{$this->mUniqPrefix}-h-(\d+)-" . self::MARKER_SUFFIX; $baseTitleText = $this->mTitle->getPrefixedDBkey(); + $oldType = $this->mOutputType; + $this->setOutputType( self::OT_WIKI ); + $frame = $this->getPreprocessor()->newFrame(); + $root = $this->preprocessToDom( $origText ); + $node = $root->getFirstChild(); + $byteOffset = 0; $tocraw = array(); foreach( $matches[3] as $headline ) { @@ -3455,68 +3591,61 @@ class Parser } $level = $matches[1][$headlineCount]; - if( $doNumberHeadings || $enoughToc ) { - - if ( $level > $prevlevel ) { - # Increase TOC level - $toclevel++; - $sublevelCount[$toclevel] = 0; - if( $toclevel<$wgMaxTocLevel ) { - $prevtoclevel = $toclevel; - $toc .= $sk->tocIndent(); - $numVisible++; - } + if ( $level > $prevlevel ) { + # Increase TOC level + $toclevel++; + $sublevelCount[$toclevel] = 0; + if( $toclevel<$wgMaxTocLevel ) { + $prevtoclevel = $toclevel; + $toc .= $sk->tocIndent(); + $numVisible++; } - elseif ( $level < $prevlevel && $toclevel > 1 ) { - # Decrease TOC level, find level to jump to + } + elseif ( $level < $prevlevel && $toclevel > 1 ) { + # Decrease TOC level, find level to jump to - if ( $toclevel == 2 && $level <= $levelCount[1] ) { - # Can only go down to level 1 - $toclevel = 1; - } else { - for ($i = $toclevel; $i > 0; $i--) { - if ( $levelCount[$i] == $level ) { - # Found last matching level - $toclevel = $i; - break; - } - elseif ( $levelCount[$i] < $level ) { - # Found first matching level below current level - $toclevel = $i + 1; - break; - } - } + for ($i = $toclevel; $i > 0; $i--) { + if ( $levelCount[$i] == $level ) { + # Found last matching level + $toclevel = $i; + break; } - if( $toclevel<$wgMaxTocLevel ) { - if($prevtoclevel < $wgMaxTocLevel) { - # Unindent only if the previous toc level was shown :p - $toc .= $sk->tocUnindent( $prevtoclevel - $toclevel ); - $prevtoclevel = $toclevel; - } else { - $toc .= $sk->tocLineEnd(); - } + elseif ( $levelCount[$i] < $level ) { + # Found first matching level below current level + $toclevel = $i + 1; + break; } } - else { - # No change in level, end TOC line - if( $toclevel<$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; + } else { $toc .= $sk->tocLineEnd(); } } + } + else { + # No change in level, end TOC line + if( $toclevel<$wgMaxTocLevel ) { + $toc .= $sk->tocLineEnd(); + } + } - $levelCount[$toclevel] = $level; + $levelCount[$toclevel] = $level; - # count number of headlines for each level - @$sublevelCount[$toclevel]++; - $dot = 0; - for( $i = 1; $i <= $toclevel; $i++ ) { - if( !empty( $sublevelCount[$i] ) ) { - if( $dot ) { - $numbering .= '.'; - } - $numbering .= $wgContLang->formatNum( $sublevelCount[$i] ); - $dot = 1; + # count number of headlines for each level + @$sublevelCount[$toclevel]++; + $dot = 0; + for( $i = 1; $i <= $toclevel; $i++ ) { + if( !empty( $sublevelCount[$i] ) ) { + if( $dot ) { + $numbering .= '.'; } + $numbering .= $wgContLang->formatNum( $sublevelCount[$i] ); + $dot = 1; } } @@ -3540,16 +3669,13 @@ class Parser # For the anchor, strip out HTML-y stuff period $safeHeadline = preg_replace( '/<.*?'.'>/', '', $safeHeadline ); + $safeHeadline = preg_replace( '/[ _]+/', ' ', $safeHeadline ); $safeHeadline = trim( $safeHeadline ); # Save headline for section edit hint before it's escaped $headlineHint = $safeHeadline; - if ( $wgEnforceHtmlIds ) { - $legacyHeadline = false; - $safeHeadline = Sanitizer::escapeId( $safeHeadline, - 'noninitial' ); - } else { + if ( $wgHtml5 && $wgExperimentalHtmlIds ) { # For reverse compatibility, provide an id that's # HTML4-compatible, like we used to. # @@ -3561,20 +3687,17 @@ class Parser # to type in section names like "abc_.D7.93.D7.90.D7.A4" # manually, so let's not bother worrying about it. $legacyHeadline = Sanitizer::escapeId( $safeHeadline, - 'noninitial' ); - $safeHeadline = Sanitizer::escapeId( $safeHeadline, 'xml' ); + array( 'noninitial', 'legacy' ) ); + $safeHeadline = Sanitizer::escapeId( $safeHeadline ); if ( $legacyHeadline == $safeHeadline ) { # No reason to have both (in fact, we can't) $legacyHeadline = false; - } elseif ( $legacyHeadline != Sanitizer::escapeId( - $legacyHeadline, 'xml' ) ) { - # The legacy id is invalid XML. We used to allow this, but - # there's no reason to do so anymore. Backward - # compatibility will fail slightly in this case, but it's - # no big deal. - $legacyHeadline = false; } + } else { + $legacyHeadline = false; + $safeHeadline = Sanitizer::escapeId( $safeHeadline, + 'noninitial' ); } # HTML names must be case-insensitively unique (bug 10721). FIXME: @@ -3602,7 +3725,7 @@ class Parser # Don't number the heading if it is the only one (looks silly) if( $doNumberHeadings && count( $matches[3] ) > 1) { # the two are different if the line contains a link - $headline=$numbering . ' ' . $headline; + $headline = $numbering . ' ' . $headline; } # Create the anchor for linking from the TOC to the section @@ -3615,9 +3738,33 @@ class Parser $legacyAnchor .= '_' . $refers[$legacyArrayKey]; } if( $enoughToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) { - $toc .= $sk->tocLine($anchor, $tocline, $numbering, $toclevel); - $tocraw[] = array( 'toclevel' => $toclevel, 'level' => $level, 'line' => $tocline, 'number' => $numbering ); + $toc .= $sk->tocLine($anchor, $tocline, + $numbering, $toclevel, ($isTemplate ? false : $sectionIndex)); + } + + # Add the section to the section tree + # Find the DOM node for this header + while ( $node && !$isTemplate ) { + if ( $node->getName() === 'h' ) { + $bits = $node->splitHeading(); + if ( $bits['i'] == $sectionIndex ) + break; + } + $byteOffset += mb_strlen( $this->mStripState->unstripBoth( + $frame->expand( $node, PPFrame::RECOVER_ORIG ) ) ); + $node = $node->getNextSibling(); } + $tocraw[] = array( + 'toclevel' => $toclevel, + 'level' => $level, + 'line' => $tocline, + 'number' => $numbering, + 'index' => ($isTemplate ? 'T-' : '' ) . $sectionIndex, + 'fromtitle' => $titleText, + 'byteoffset' => ( $isTemplate ? null : $byteOffset ), + 'anchor' => $anchor, + ); + # give headline the correct <h#> tag if( $showEditLink && $sectionIndex !== false ) { if( $isTemplate ) { @@ -3637,7 +3784,7 @@ class Parser $headlineCount++; } - $this->mOutput->setSections( $tocraw ); + $this->setOutputType( $oldType ); # Never ever show TOC if no headers if( $numVisible < 1 ) { @@ -3649,6 +3796,11 @@ class Parser $toc .= $sk->tocUnindent( $prevtoclevel - 1 ); } $toc = $sk->tocList( $toc ); + $this->mOutput->setTOCHTML( $toc ); + } + + if ( $isMain ) { + $this->mOutput->setSections( $tocraw ); } # split up and insert constructed headlines @@ -3684,6 +3836,96 @@ 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. * @@ -3728,26 +3970,29 @@ class Parser * (see also bug 12815) */ $ts = $this->mOptions->getTimestamp(); - $tz = wfMsgForContent( 'timezone-utc' ); if ( isset( $wgLocaltimezone ) ) { - $unixts = wfTimestamp( TS_UNIX, $ts ); - $oldtz = getenv( 'TZ' ); - putenv( 'TZ='.$wgLocaltimezone ); - $ts = date( 'YmdHis', $unixts ); - $tz = date( 'T', $unixts ); # might vary on DST changeover! + $tz = $wgLocaltimezone; + } else { + $tz = date_default_timezone_get(); + } - /* Allow translation of timezones trough wiki. date() can return - * whatever crap the system uses, localised or not, so we cannot - * ship premade translations. - */ - $key = 'timezone-' . strtolower( trim( $tz ) ); - $value = wfMsgForContent( $key ); - if ( !wfEmptyMsg( $key, $value ) ) $tz = $value; + $unixts = wfTimestamp( TS_UNIX, $ts ); + $oldtz = date_default_timezone_get(); + date_default_timezone_set( $tz ); + $ts = date( 'YmdHis', $unixts ); + $tzMsg = date( 'T', $unixts ); # might vary on DST changeover! - putenv( 'TZ='.$oldtz ); - } + /* Allow translation of timezones trough 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; + + date_default_timezone_set( $oldtz ); - $d = $wgContLang->timeanddate( $ts, false, false ) . " ($tz)"; + $d = $wgContLang->timeanddate( $ts, false, false ) . " ($tzMsg)"; # Variable replacement # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags @@ -3781,7 +4026,7 @@ class Parser $m = array(); if ( preg_match( "/^($nc+:|)$tc+?( \\($tc+\\))$/", $t, $m ) ) { $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text ); - } elseif ( preg_match( "/^($nc+:|)$tc+?(, $tc+|)$/", $t, $m ) && '' != "$m[1]$m[2]" ) { + } elseif ( preg_match( "/^($nc+:|)$tc+?(, $tc+|)$/", $t, $m ) && "$m[1]$m[2]" != '' ) { $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text ); } else { # if there's no context, don't bother duplicating the title @@ -3797,22 +4042,30 @@ class Parser /** * Fetch the user's signature text, if any, and normalize to * validated, ready-to-insert wikitext. + * If you have pre-fetched the nickname or the fancySig option, you can + * specify them here to save a database query. * * @param User $user * @return string - * @private */ - function getUserSig( &$user ) { + function getUserSig( &$user, $nickname = false, $fancySig = null ) { global $wgMaxSigChars; $username = $user->getName(); - $nickname = $user->getOption( 'nickname' ); - $nickname = $nickname === '' ? $username : $nickname; + + // If not given, retrieve from the user object. + if ( $nickname === false ) + $nickname = $user->getOption( 'nickname' ); + + if ( is_null( $fancySig ) ) + $fancySig = $user->getBoolOption( 'fancysig' ); + + $nickname = $nickname == null ? $username : $nickname; if( mb_strlen( $nickname ) > $wgMaxSigChars ) { $nickname = $username; wfDebug( __METHOD__ . ": $username has overlong signature.\n" ); - } elseif( $user->getBoolOption( 'fancysig' ) !== false ) { + } elseif( $fancySig !== false ) { # Sig. might contain markup; validate this if( $this->validateSig( $nickname ) !== false ) { # Validated; clean up (if needed) and return it @@ -4005,28 +4258,30 @@ class Parser * @param integer $flags 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 + * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This * allows for conditional expansion of the parse tree, allowing you to eliminate dead - * branches and thus speed up parsing. It is also possible to analyse the parse tree of + * branches and thus speed up parsing. It is also possible to analyse the parse tree of * the arguments, and to control the way they are expanded. * * The $frame parameter is a PPFrame. This can be used to produce expanded text from the * arguments, for instance: * $text = isset( $args[0] ) ? $frame->expand( $args[0] ) : ''; * - * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in + * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in * future versions. Please call $frame->expand() on it anyway so that your code keeps * working if/when this is changed. * * If you want whitespace to be trimmed from $args, you need to do it yourself, post- * expansion. * - * Please read the documentation in includes/parser/Preprocessor.php for more information + * Please read the documentation in includes/parser/Preprocessor.php for more information * about the methods available in PPFrame and PPNode. * * @return The old callback function for this name, if any */ function setFunctionHook( $id, $callback, $flags = 0 ) { + global $wgContLang; + $oldVal = isset( $this->mFunctionHooks[$id] ) ? $this->mFunctionHooks[$id][0] : null; $this->mFunctionHooks[$id] = array( $callback, $flags ); @@ -4041,7 +4296,7 @@ class Parser foreach ( $synonyms as $syn ) { # Case if ( !$sensitive ) { - $syn = strtolower( $syn ); + $syn = $wgContLang->lc( $syn ); } # Add leading hash if ( !( $flags & SFH_NO_HASH ) ) { @@ -4066,6 +4321,25 @@ class Parser } /** + * Create a tag function, e.g. <test>some stuff</test>. + * Unlike tag hooks, tag functions are parsed at preprocessor level. + * Unlike parser functions, their content is not preprocessed. + */ + function setFunctionTagHook( $tag, $callback, $flags ) { + $tag = strtolower( $tag ); + $old = isset( $this->mFunctionTagHooks[$tag] ) ? + $this->mFunctionTagHooks[$tag] : null; + $this->mFunctionTagHooks[$tag] = array( $callback, $flags ); + + if( !in_array( $tag, $this->mStripList ) ) { + $this->mStripList[] = $tag; + } + + return $old; + } + + /** + * FIXME: update documentation. makeLinkObj() is deprecated. * Replace <!--LINK--> link placeholders with actual links, in the buffer * Placeholders created in Skin::makeLinkObj() * Returns an array of link CSS classes, indexed by PDBK. @@ -4085,19 +4359,6 @@ class Parser } /** - * Tag hook handler for 'pre'. - */ - function renderPreTag( $text, $attribs ) { - // Backwards-compatibility hack - $content = StringUtils::delimiterReplace( '<nowiki>', '</nowiki>', '$1', $text, 'i' ); - - $attribs = Sanitizer::validateTagAttributes( $attribs, 'pre' ); - return Xml::openElement( 'pre', $attribs ) . - Xml::escapeTagsOnly( $content ) . - '</pre>'; - } - - /** * Renders an image gallery from a text with one line per image. * text labels may be given by using |-style alternative text. E.g. * Image:one.jpg|The number "1" @@ -4145,7 +4406,7 @@ class Parser if ( count( $matches ) == 0 ) { continue; } - + if ( strpos( $matches[0], '%' ) !== false ) $matches[1] = urldecode( $matches[1] ); $tp = Title::newFromText( $matches[1]/*, NS_FILE*/ ); @@ -4227,11 +4488,13 @@ class Parser # * none same, but not aligned # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox # * center center the image - # * framed Keep original image size, no magnify-button. + # * frame Keep original image size, no magnify-button. + # * framed Same as "frame" # * frameless like 'thumb' but without a frame. Keeps user preferences for width # * upright reduce width for upright images, rounded to full __0 px # * border draw a 1px border around the image # * alt Text for HTML alt attribute (defaults to empty) + # * link Set the target of the image link. Can be external, interwiki, or local # vertical-align values (no % or length right now): # * baseline # * sub @@ -4255,15 +4518,7 @@ class Parser # Get the file $imagename = $title->getDBkey(); - if ( isset( $this->mFileCache[$imagename][$time] ) ) { - $file = $this->mFileCache[$imagename][$time]; - } else { - $file = wfFindFile( $title, $time ); - if ( count( $this->mFileCache ) > 1000 ) { - $this->mFileCache = array(); - } - $this->mFileCache[$imagename][$time] = $file; - } + $file = wfFindFile( $title, array( 'time' => $time ) ); # Get parameter map $handler = $file ? $file->getHandler() : false; @@ -4312,7 +4567,7 @@ class Parser switch( $paramName ) { case 'manualthumb': case 'alt': - // @fixme - possibly check validity here for + // @todo Fixme: possibly check validity here for // manualthumb? downstream behavior seems odd with // missing manual thumbs. $validated = true; @@ -4367,7 +4622,11 @@ class Parser $params['frame']['caption'] = $caption; - $params['frame']['title'] = $this->stripAltText( $caption, $holders ); + # 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'] ); # 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 @@ -4385,11 +4644,27 @@ class Parser # named parameter entirely for images without a caption; adding an ex- # plicit caption= parameter and preserving the old magic unnamed para- # meter for BC; ... - if( $caption !== '' && !isset( $params['frame']['alt'] ) - && !isset( $params['frame']['framed'] ) - && !isset( $params['frame']['thumbnail'] ) - && !isset( $params['frame']['manualthumb'] ) ) { - $params['frame']['alt'] = $params['frame']['title']; + if ( $imageIsFramed ) { # Framed image + if ( $caption === '' && !isset( $params['frame']['alt'] ) ) { + # No caption or alt text, add the filename as the alt text so + # that screen readers at least get some description of the image + $params['frame']['alt'] = $title->getText(); + } + # Do not set $params['frame']['title'] because tooltips don't make sense + # for framed images + } else { # Inline image + if ( !isset( $params['frame']['alt'] ) ) { + # No alt text, use the "caption" for the alt text + if ( $caption !== '') { + $params['frame']['alt'] = $this->stripAltText( $caption, $holders ); + } else { + # No caption, fall back to using the filename for the + # alt text + $params['frame']['alt'] = $title->getText(); + } + } + # Use the "caption" for the tooltip text + $params['frame']['title'] = $this->stripAltText( $caption, $holders ); } wfRunHooks( 'ParserMakeImageParams', array( $title, $file, &$params ) ); @@ -4404,7 +4679,7 @@ class Parser return $ret; } - + protected function stripAltText( $caption, $holders ) { # Strip bad stuff out of the title (tooltip). We can't just use # replaceLinkHoldersText() here, because if this function is called @@ -4420,7 +4695,7 @@ class Parser # remove the tags $tooltip = $this->mStripState->unstripBoth( $tooltip ); $tooltip = Sanitizer::stripAllTags( $tooltip ); - + return $tooltip; } @@ -4452,9 +4727,9 @@ class Parser /**#@+ * 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 ); } + 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 ); } /**#@-*/ /**#@+ @@ -4856,7 +5131,7 @@ class Parser $links['interwiki'][] = $this->mLinkHolders->interwiki[$key]; $pos = $start_pos + strlen( "<!--IWLINK $key-->" ); } - + $data['linkholder'] = $links; return $data; @@ -4865,7 +5140,7 @@ class Parser function unserialiseHalfParsedText( $data, $intPrefix = null /* Unique identifying prefix */ ) { if (!$intPrefix) $intPrefix = $this->getRandomString(); - + // First, extract the strip state. $stripState = $data['stripstate']; $this->mStripState->general->merge( $stripState->general ); diff --git a/includes/parser/ParserCache.php b/includes/parser/ParserCache.php index d17214c3..524d6be5 100644 --- a/includes/parser/ParserCache.php +++ b/includes/parser/ParserCache.php @@ -7,7 +7,7 @@ class ParserCache { /** * Get an instance of this object */ - public static function &singleton() { + public static function singleton() { static $instance; if ( !isset( $instance ) ) { global $parserMemc; @@ -22,11 +22,11 @@ class ParserCache { * * @param object $memCached */ - function __construct( &$memCached ) { - $this->mMemc =& $memCached; + function __construct( $memCached ) { + $this->mMemc = $memCached; } - function getKey( &$article, $popts ) { + function getKey( $article, $popts ) { global $wgRequest; if( $popts instanceof User ) // It used to be getKey( &$article, &$user ) @@ -47,52 +47,55 @@ class ParserCache { return $key; } - function getETag( &$article, $popts ) { + function getETag( $article, $popts ) { return 'W/"' . $this->getKey($article, $popts) . "--" . $article->mTouched. '"'; } - function get( &$article, $popts ) { - global $wgCacheEpoch; - $fname = 'ParserCache::get'; - wfProfileIn( $fname ); - + function getDirty( $article, $popts ) { $key = $this->getKey( $article, $popts ); - wfDebug( "Trying parser cache $key\n" ); $value = $this->mMemc->get( $key ); - if ( is_object( $value ) ) { - wfDebug( "Found.\n" ); - # Delete if article has changed since the cache was made - $canCache = $article->checkTouched(); - $cacheTime = $value->getCacheTime(); - $touched = $article->mTouched; - if ( !$canCache || $value->expired( $touched ) ) { - if ( !$canCache ) { - wfIncrStats( "pcache_miss_invalid" ); - wfDebug( "Invalid cached redirect, touched $touched, epoch $wgCacheEpoch, cached $cacheTime\n" ); - } else { - wfIncrStats( "pcache_miss_expired" ); - wfDebug( "Key expired, touched $touched, epoch $wgCacheEpoch, cached $cacheTime\n" ); - } - $this->mMemc->delete( $key ); - $value = false; - } else { - if ( isset( $value->mTimestamp ) ) { - $article->mTimestamp = $value->mTimestamp; - } - wfIncrStats( "pcache_hit" ); - } - } else { + return is_object( $value ) ? $value : false; + } + + function get( $article, $popts ) { + global $wgCacheEpoch; + wfProfileIn( __METHOD__ ); + + $value = $this->getDirty( $article, $popts ); + if ( !$value ) { wfDebug( "Parser cache miss.\n" ); wfIncrStats( "pcache_miss_absent" ); + wfProfileOut( __METHOD__ ); + return false; + } + + 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" ); + } $value = false; + } else { + if ( isset( $value->mTimestamp ) ) { + $article->mTimestamp = $value->mTimestamp; + } + wfIncrStats( "pcache_hit" ); } - wfProfileOut( $fname ); + wfProfileOut( __METHOD__ ); return $value; } - function save( $parserOutput, &$article, $popts ){ + function save( $parserOutput, $article, $popts ){ global $wgParserCacheExpireTime; $key = $this->getKey( $article, $popts ); diff --git a/includes/parser/ParserOptions.php b/includes/parser/ParserOptions.php index e6a9f3a7..985bba28 100644 --- a/includes/parser/ParserOptions.php +++ b/includes/parser/ParserOptions.php @@ -5,10 +5,8 @@ * @todo document * @ingroup Parser */ -class ParserOptions -{ +class ParserOptions { # All variables are supposed to be private in theory, although in practise this is not the case. - var $mUseTeX; # Use texvc to expand <math> tags var $mUseDynamicDates; # Use DateFormatter to format dates var $mInterwikiMagic; # Interlanguage links are removed and returned in an array var $mAllowExternalImages; # Allow external images inline @@ -35,9 +33,8 @@ class ParserOptions 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 $mIsPrintable; # Parsing the printable version of the page - function getUseTeX() { return $this->mUseTeX; } function getUseDynamicDates() { return $this->mUseDynamicDates; } function getInterwikiMagic() { return $this->mInterwikiMagic; } function getAllowExternalImages() { return $this->mAllowExternalImages; } @@ -59,8 +56,8 @@ class ParserOptions function getExternalLinkTarget() { return $this->mExternalLinkTarget; } function getIsPreview() { return $this->mIsPreview; } function getIsSectionPreview() { return $this->mIsSectionPreview; } - function getIsPrintable() { return $this->mIsPrintable; } - + function getIsPrintable() { return $this->mIsPrintable; } + function getSkin() { if ( !isset( $this->mSkin ) ) { $this->mSkin = $this->mUser->getSkin(); @@ -82,7 +79,6 @@ class ParserOptions return $this->mTimestamp; } - function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX, $x ); } function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates, $x ); } function setInterwikiMagic( $x ) { return wfSetVar( $this->mInterwikiMagic, $x ); } function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages, $x ); } @@ -107,8 +103,8 @@ class ParserOptions function setExternalLinkTarget( $x ) { return wfSetVar( $this->mExternalLinkTarget, $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 ); } - + function setIsPrintable( $x ) { return wfSetVar( $this->mIsPrintable, $x ); } + function __construct( $user = null ) { $this->initialiseFromUser( $user ); } @@ -123,12 +119,13 @@ class ParserOptions /** Get user options */ function initialiseFromUser( $userInput ) { - global $wgUseTeX, $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages; + global $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages; global $wgAllowExternalImagesFrom, $wgEnableImageWhitelist, $wgAllowSpecialInclusion, $wgMaxArticleSize; global $wgMaxPPNodeCount, $wgMaxTemplateDepth, $wgMaxPPExpandDepth, $wgCleanSignatures; global $wgExternalLinkTarget; - $fname = 'ParserOptions::initialiseFromUser'; - wfProfileIn( $fname ); + + wfProfileIn( __METHOD__ ); + if ( !$userInput ) { global $wgUser; if ( isset( $wgUser ) ) { @@ -142,7 +139,6 @@ class ParserOptions $this->mUser = $user; - $this->mUseTeX = $wgUseTeX; $this->mUseDynamicDates = $wgUseDynamicDates; $this->mInterwikiMagic = $wgInterwikiMagic; $this->mAllowExternalImages = $wgAllowExternalImages; @@ -167,6 +163,7 @@ class ParserOptions $this->mExternalLinkTarget = $wgExternalLinkTarget; $this->mIsPreview = false; $this->mIsSectionPreview = false; - wfProfileOut( $fname ); + + wfProfileOut( __METHOD__ ); } } diff --git a/includes/parser/ParserOutput.php b/includes/parser/ParserOutput.php index 22c1dfba..ea5840e6 100644 --- a/includes/parser/ParserOutput.php +++ b/includes/parser/ParserOutput.php @@ -24,14 +24,10 @@ class ParserOutput $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 + $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. - /** - * Overridden title for display - */ - private $displayTitle = false; - function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(), $containsOldMagic = false, $titletext = '' ) { @@ -54,10 +50,12 @@ class ParserOutput function &getImages() { return $this->mImages; } function &getExternalLinks() { return $this->mExternalLinks; } function getNoGallery() { return $this->mNoGallery; } + function getHeadItems() { return $this->mHeadItems; } 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 ); } @@ -68,10 +66,10 @@ class ParserOutput function setTitleText( $t ) { return wfSetVar( $this->mTitleText, $t ); } function setSections( $toc ) { return wfSetVar( $this->mSections, $toc ); } function setIndexPolicy( $policy ) { return wfSetVar( $this->mIndexPolicy, $policy ); } + function setTOCHTML( $tochtml ) { return wfSetVar( $this->mTOCHTML, $tochtml ); } function addCategory( $c, $sort ) { $this->mCategories[$c] = $sort; } function addLanguageLink( $t ) { $this->mLanguageLinks[] = $t; } - function addExternalLink( $url ) { $this->mExternalLinks[$url] = 1; } function addWarning( $s ) { $this->mWarnings[$s] = 1; } function addOutputHook( $hook, $data = false ) { @@ -91,7 +89,18 @@ class ParserOutput return (bool)$this->mNewSection; } + function addExternalLink( $url ) { + # We don't register links pointing to our own server, unless... :-) + global $wgServer, $wgRegisterInternalExternals; + if( $wgRegisterInternalExternals or stripos($url,$wgServer.'/')!==0) + $this->mExternalLinks[$url] = 1; + } + function addLink( $title, $id = null ) { + if ( $title->isExternal() ) { + // Don't record interwikis in pagelinks + return; + } $ns = $title->getNamespace(); $dbk = $title->getDBkey(); if ( $ns == NS_MEDIA ) { @@ -170,7 +179,7 @@ class ParserOutput * @param string $text Desired title text */ public function setDisplayTitle( $text ) { - $this->displayTitle = $text; + $this->setTitleText( $text ); } /** @@ -179,7 +188,11 @@ class ParserOutput * @return string */ public function getDisplayTitle() { - return $this->displayTitle; + $t = $this->getTitleText( ); + if( $t === '' ) { + return false; + } + return $t; } /** diff --git a/includes/parser/Preprocessor.php b/includes/parser/Preprocessor.php index 1a33ac7f..9c417d23 100644 --- a/includes/parser/Preprocessor.php +++ b/includes/parser/Preprocessor.php @@ -66,6 +66,21 @@ interface PPFrame { function isEmpty(); /** + * Returns all arguments of this frame + */ + function getArguments(); + + /** + * Returns all numbered arguments of this frame + */ + function getNumberedArguments(); + + /** + * Returns all named arguments of this frame + */ + function getNamedArguments(); + + /** * Get an argument to this frame by name */ function getArgument( $name ); diff --git a/includes/parser/Preprocessor_DOM.php b/includes/parser/Preprocessor_DOM.php index 2e114545..673ac241 100644 --- a/includes/parser/Preprocessor_DOM.php +++ b/includes/parser/Preprocessor_DOM.php @@ -419,7 +419,8 @@ class Preprocessor_DOM implements Preprocessor { 'count' => $count ); $stack->push( $piece ); $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); + $flags = $stack->getFlags(); + extract( $flags ); $i += $count; } } @@ -470,7 +471,8 @@ class Preprocessor_DOM implements Preprocessor { // Unwind the stack $stack->pop(); $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); + $flags = $stack->getFlags(); + extract( $flags ); // Append the result to the enclosing accumulator $accum .= $element; @@ -497,7 +499,8 @@ class Preprocessor_DOM implements Preprocessor { $stack->push( $piece ); $accum =& $stack->getAccum(); - extract( $stack->getFlags() ); + $flags = $stack->getFlags(); + extract( $flags ); } else { # Add literal brace(s) $accum .= htmlspecialchars( str_repeat( $curChar, $count ) ); @@ -597,8 +600,8 @@ class Preprocessor_DOM implements Preprocessor { } $enclosingAccum .= str_repeat( $piece->open, $skippedBraces ); } - - extract( $stack->getFlags() ); + $flags = $stack->getFlags(); + extract( $flags ); # Add XML element to the enclosing accumulator $accum .= $element; @@ -1189,6 +1192,18 @@ class PPFrame_DOM implements PPFrame { } } + function getArguments() { + return array(); + } + + function getNumberedArguments() { + return array(); + } + + function getNamedArguments() { + return array(); + } + /** * Returns true if there are no arguments in this frame */ @@ -1224,8 +1239,7 @@ class PPTemplateFrame_DOM extends PPFrame_DOM { var $numberedExpansionCache, $namedExpansionCache; function __construct( $preprocessor, $parent = false, $numberedArgs = array(), $namedArgs = array(), $title = false ) { - $this->preprocessor = $preprocessor; - $this->parser = $preprocessor->parser; + PPFrame_DOM::__construct( $preprocessor ); $this->parent = $parent; $this->numberedArgs = $numberedArgs; $this->namedArgs = $namedArgs; @@ -1337,8 +1351,7 @@ class PPCustomFrame_DOM extends PPFrame_DOM { var $args; function __construct( $preprocessor, $args ) { - $this->preprocessor = $preprocessor; - $this->parser = $preprocessor->parser; + PPFrame_DOM::__construct( $preprocessor ); $this->args = $args; } diff --git a/includes/parser/Preprocessor_Hash.php b/includes/parser/Preprocessor_Hash.php index f46ee40c..c5d69685 100644 --- a/includes/parser/Preprocessor_Hash.php +++ b/includes/parser/Preprocessor_Hash.php @@ -1139,6 +1139,18 @@ class PPFrame_Hash implements PPFrame { } } + function getArguments() { + return array(); + } + + function getNumberedArguments() { + return array(); + } + + function getNamedArguments() { + return array(); + } + /** * Returns true if there are no arguments in this frame */ @@ -1174,8 +1186,7 @@ class PPTemplateFrame_Hash extends PPFrame_Hash { var $numberedExpansionCache, $namedExpansionCache; function __construct( $preprocessor, $parent = false, $numberedArgs = array(), $namedArgs = array(), $title = false ) { - $this->preprocessor = $preprocessor; - $this->parser = $preprocessor->parser; + PPFrame_Hash::__construct( $preprocessor ); $this->parent = $parent; $this->numberedArgs = $numberedArgs; $this->namedArgs = $namedArgs; @@ -1287,8 +1298,7 @@ class PPCustomFrame_Hash extends PPFrame_Hash { var $args; function __construct( $preprocessor, $args ) { - $this->preprocessor = $preprocessor; - $this->parser = $preprocessor->parser; + PPFrame_Hash::__construct( $preprocessor ); $this->args = $args; } |