X-Git-Url: http://git.heureux-cyclage.org/?a=blobdiff_plain;f=includes%2Fparser%2FParser.php;h=29775c6134a836c019e924527be715549c04e6f6;hb=6d20e79faf22ba192704e9afb2e1bc4b619d8899;hp=34b31ce52e91a463797cb44d4370e65c0f6998cb;hpb=bec172cb54790b0951825d67ac22c6bad9007ee4;p=lhc%2Fweb%2Fwiklou.git diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php index 34b31ce52e..29775c6134 100644 --- a/includes/parser/Parser.php +++ b/includes/parser/Parser.php @@ -129,7 +129,7 @@ class Parser { $this->mDefaultStripList = $this->mStripList = array(); $this->mUrlProtocols = wfUrlProtocols(); $this->mExtLinkBracketedRegex = '/\[(\b(' . wfUrlProtocols() . ')'. - '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x0a\\x0d]*?)\]/S'; + '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x00-\\x08\\x0a-\\x1F]*?)\]/S'; $this->mVarCache = array(); if ( isset( $conf['preprocessorClass'] ) ) { $this->mPreprocessorClass = $conf['preprocessorClass']; @@ -239,52 +239,6 @@ class Parser { wfProfileOut( __METHOD__ ); } - function setOutputType( $ot ) { - $this->mOutputType = $ot; - # Shortcut alias - $this->ot = array( - 'html' => $ot == self::OT_HTML, - 'wiki' => $ot == self::OT_WIKI, - 'pre' => $ot == self::OT_PREPROCESS, - 'plain' => $ot == self::OT_PLAIN, - ); - } - - /** - * Set the context title - */ - function setTitle( $t ) { - if ( !$t || $t instanceof FakeTitle ) { - $t = Title::newFromText( 'NO TITLE' ); - } - - if ( strval( $t->getFragment() ) !== '' ) { - # Strip the fragment to avoid various odd effects - $this->mTitle = clone $t; - $this->mTitle->setFragment( '' ); - } else { - $this->mTitle = $t; - } - } - - /** - * Accessor for mUniqPrefix. - * - * @public - */ - function uniqPrefix() { - if ( !isset( $this->mUniqPrefix ) ) { - # @todo Fixme: this is probably *horribly wrong* - # LanguageConverter seems to want $wgParser's uniqPrefix, however - # if this is called for a parser cache hit, the parser may not - # have ever been initialized in the first place. - # Not really sure what the heck is supposed to be going on here. - return ''; - # throw new MWException( "Accessing uninitialized mUniqPrefix" ); - } - return $this->mUniqPrefix; - } - /** * Convert wikitext to HTML * Do not call this function recursively. @@ -303,7 +257,7 @@ class Parser { * to internalParse() which does all the real work. */ - global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang, $wgDisableLangConversion; + global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang, $wgDisableLangConversion, $wgDisableTitleConversion; $fname = __METHOD__.'-' . wfGetCaller(); wfProfileIn( __METHOD__ ); wfProfileIn( $fname ); @@ -333,10 +287,10 @@ class Parser { $fixtags = array( # french spaces, last one Guillemet-left # only if there is something before the space - '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1 \\2', + '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1 \\2', # french spaces, Guillemet-right - '/(\\302\\253) /' => '\\1 ', - '/ (!\s*important)/' => ' \\1', # Beware of CSS magic word !important, bug #11874. + '/(\\302\\253) /' => '\\1 ', + '/ (!\s*important)/' => ' \\1', # Beware of CSS magic word !important, bug #11874. ); $text = preg_replace( array_keys( $fixtags ), array_values( $fixtags ), $text ); @@ -344,22 +298,50 @@ class Parser { $this->replaceLinkHolders( $text ); - # 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 mark. + /** + * 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->isTalkPage() || $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 mark. + $text = $wgContLang->convert( $text ); } - # A title may have been set in a conversion rule. - # 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 ( $wgContLang->getConvRuleTitle() ) { - $this->mOutput->setTitleText( $wgContLang->getConvRuleTitle() ); + /** + * 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 ); @@ -387,7 +369,7 @@ class Parser { $text = Sanitizer::normalizeCharReferences( $text ); - if ( ( $wgUseTidy && $this->mOptions->mTidy ) || $wgAlwaysUseTidy ) { + if ( ( $wgUseTidy && $this->mOptions->getTidy() ) || $wgAlwaysUseTidy ) { $text = MWTidy::tidy( $text ); } else { # attempt to sanitize at least some nesting problems @@ -429,7 +411,7 @@ class Parser { $PFreport = "Expensive parser function count: {$this->mExpensiveFunctionCount}/$wgExpensiveParserFunctionLimit\n"; $limitReport = "NewPP limit report\n" . - "Preprocessor node count: {$this->mPPNodeCount}/{$this->mOptions->mMaxPPNodeCount}\n" . + "Preprocessor node count: {$this->mPPNodeCount}/{$this->mOptions->getMaxPPNodeCount()}\n" . "Post-expand include size: {$this->mIncludeSizes['post-expand']}/$max bytes\n" . "Template argument size: {$this->mIncludeSizes['arg']}/$max bytes\n". $PFreport; @@ -453,7 +435,7 @@ class Parser { * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded * * @param $text String: text extension wants to have parsed - * @param PPFrame $frame: The frame to use for expanding any template variables + * @param $frame PPFrame: The frame to use for expanding any template variables */ function recursiveTagParse( $text, $frame=false ) { wfProfileIn( __METHOD__ ); @@ -488,7 +470,7 @@ class Parser { /** * Process the wikitext for the ?preload= feature. (bug 5210) * - * , etc. are parsed as for template transclusion, + * , etc. are parsed as for template transclusion, * comments, templates, arguments, tags hooks and parser functions are untouched. */ public function getPreloadText( $text, $title, $options ) { @@ -496,7 +478,7 @@ class Parser { $this->clearState(); $this->setOutputType( self::OT_PLAIN ); $this->mOptions = $options; - $this->setTitle( $title ); + $this->setTitle( $title ); $flags = PPFrame::NO_ARGS | PPFrame::NO_TEMPLATES; $dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION ); @@ -509,15 +491,121 @@ class Parser { * @private * @static */ - function getRandomString() { + static function getRandomString() { return dechex( mt_rand( 0, 0x7fffffff ) ) . dechex( mt_rand( 0, 0x7fffffff ) ); } - function &getTitle() { return $this->mTitle; } - function getOptions() { return $this->mOptions; } - function getRevisionId() { return $this->mRevisionId; } - function getOutput() { return $this->mOutput; } - function nextLinkID() { return $this->mLinkID++; } + /** + * Accessor for mUniqPrefix. + * + * @return String + */ + public function uniqPrefix() { + if ( !isset( $this->mUniqPrefix ) ) { + # @todo Fixme: this is probably *horribly wrong* + # LanguageConverter seems to want $wgParser's uniqPrefix, however + # if this is called for a parser cache hit, the parser may not + # have ever been initialized in the first place. + # Not really sure what the heck is supposed to be going on here. + return ''; + # throw new MWException( "Accessing uninitialized mUniqPrefix" ); + } + return $this->mUniqPrefix; + } + + /** + * Set the context title + */ + function setTitle( $t ) { + if ( !$t || $t instanceof FakeTitle ) { + $t = Title::newFromText( 'NO TITLE' ); + } + + if ( strval( $t->getFragment() ) !== '' ) { + # Strip the fragment to avoid various odd effects + $this->mTitle = clone $t; + $this->mTitle->setFragment( '' ); + } else { + $this->mTitle = $t; + } + } + + /** + * Accessor for the Title object + * + * @return Title object + */ + function &getTitle() { + return $this->mTitle; + } + + /** + * Accessor/mutator for the Title object + * + * @param $x New Title object or null to just get the current one + * @return Title object + */ + function Title( $x = null ) { + return wfSetVar( $this->mTitle, $x ); + } + + /** + * Set the output type + * + * @param $ot Integer: new value + */ + function setOutputType( $ot ) { + $this->mOutputType = $ot; + # Shortcut alias + $this->ot = array( + 'html' => $ot == self::OT_HTML, + 'wiki' => $ot == self::OT_WIKI, + 'pre' => $ot == self::OT_PREPROCESS, + 'plain' => $ot == self::OT_PLAIN, + ); + } + + /** + * Accessor/mutator for the output type + * + * @param $x New value or null to just get the current one + * @return Integer + */ + function OutputType( $x = null ) { + return wfSetVar( $this->mOutputType, $x ); + } + + /** + * Get the ParserOutput object + * + * @return ParserOutput object + */ + function getOutput() { + return $this->mOutput; + } + + /** + * Get the ParserOptions object + * + * @return ParserOptions object + */ + function getOptions() { + return $this->mOptions; + } + + /** + * Accessor/mutator for the ParserOptions object + * + * @param $x New value or null to just get the current one + * @return Current ParserOptions object + */ + function Options( $x = null ) { + return wfSetVar( $this->mOptions, $x ); + } + + function nextLinkID() { + return $this->mLinkID++; + } function getFunctionLang() { global $wgLang, $wgContLang; @@ -532,6 +620,8 @@ class Parser { /** * Get a preprocessor object + * + * @return Preprocessor instance */ function getPreprocessor() { if ( !isset( $this->mPreprocessor ) ) { @@ -554,12 +644,13 @@ class Parser { * * @param $elements list of element names. Comments are always extracted. * @param $text Source text string. + * @param $matches Out parameter, Array: extarcted tags * @param $uniq_prefix + * @return String: stripped text * - * @public * @static */ - function extractTagsAndParams( $elements, $text, &$matches, $uniq_prefix = '' ) { + public function extractTagsAndParams( $elements, $text, &$matches, $uniq_prefix = '' ) { static $n = 1; $stripped = ''; $matches = array(); @@ -693,8 +784,9 @@ class Parser { */ function doTableStuff( $text ) { wfProfileIn( __METHOD__ ); - + $lines = StringUtils::explode( "\n", $text ); + $text = null; $out = ''; $td_history = array(); # Is currently a td tag open? $last_tag_history = array(); # Save history of last lag activated (td, th or caption) @@ -703,13 +795,26 @@ class Parser { $has_opened_tr = array(); # Did this table open a element? $indent_level = 0; # indent level of the table + $table_tag = 'table'; + $tr_tag = 'tr'; + $th_tag = 'th'; + $td_tag = 'td'; + $caption_tag = 'caption'; + + $extra_table_attribs = null; + $extra_tr_attribs = null; + $extra_td_attribs = null; + + $convert_style = false; + foreach ( $lines as $outLine ) { $line = trim( $outLine ); - if ( $line == '' ) { # empty line, go to next line + if ( $line === '' ) { # empty line, go to next line $out .= $outLine."\n"; continue; } + $first_character = $line[0]; $matches = array(); @@ -718,9 +823,31 @@ class Parser { $indent_level = strlen( $matches[1] ); $attributes = $this->mStripState->unstripBoth( $matches[2] ); - $attributes = Sanitizer::fixTagAttributes( $attributes , 'table' ); - $outLine = str_repeat( '
' , $indent_level ) . ""; + $attr = Sanitizer::decodeTagAttributes( $attributes ); + + $mode = @$attr['mode']; + if ( !$mode ) $mode = 'data'; + + if ( $mode == 'grid' || $mode == 'layout' ) { + $table_tag = 'div'; + $tr_tag = 'div'; + $th_tag = 'div'; + $td_tag = 'div'; + $caption_tag = 'div'; + + $extra_table_attribs = array( 'class' => 'grid-table' ); + $extra_tr_attribs = array( 'class' => 'grid-row' ); + $extra_td_attribs = array( 'class' => 'grid-cell' ); + + $convert_style = true; + } + + if ($convert_style) $attr['style'] = Sanitizer::styleFromAttributes( $attr ); + $attr = Sanitizer::validateTagAttributes( $attr, $table_tag ); + $attributes = Sanitizer::collapseTagAttributes( $attr, $extra_table_attribs ); + + $outLine = str_repeat( '
' , $indent_level ) . "<$table_tag{$attributes}>"; array_push( $td_history , false ); array_push( $last_tag_history , '' ); array_push( $tr_history , false ); @@ -732,15 +859,15 @@ class Parser { continue; } elseif ( substr( $line , 0 , 2 ) === '|}' ) { # We are ending a table - $line = '' . substr( $line , 2 ); + $line = "" . substr( $line , 2 ); $last_tag = array_pop( $last_tag_history ); if ( !array_pop( $has_opened_tr ) ) { - $line = "{$line}"; + $line = "<$tr_tag><$td_tag>{$line}"; } if ( array_pop( $tr_history ) ) { - $line = "{$line}"; + $line = "{$line}"; } if ( array_pop( $td_history ) ) { @@ -754,7 +881,12 @@ class Parser { # Whats after the tag is now only attributes $attributes = $this->mStripState->unstripBoth( $line ); - $attributes = Sanitizer::fixTagAttributes( $attributes, 'tr' ); + + $attr = Sanitizer::decodeTagAttributes( $attributes ); + if ($convert_style) $attr['style'] = Sanitizer::styleFromAttributes( $attr ); + $attr = Sanitizer::validateTagAttributes( $attr, $tr_tag ); + $attributes = Sanitizer::collapseTagAttributes( $attr, $extra_tr_attribs ); + array_pop( $tr_attributes ); array_push( $tr_attributes, $attributes ); @@ -764,7 +896,7 @@ class Parser { array_push( $has_opened_tr , true ); if ( array_pop( $tr_history ) ) { - $line = ''; + $line = ""; } if ( array_pop( $td_history ) ) { @@ -802,7 +934,7 @@ class Parser { if ( $first_character !== '+' ) { $tr_after = array_pop( $tr_attributes ); if ( !array_pop( $tr_history ) ) { - $previous = "\n"; + $previous = "<$tr_tag{$tr_after}>\n"; } array_push( $tr_history , true ); array_push( $tr_attributes , '' ); @@ -813,15 +945,15 @@ class Parser { $last_tag = array_pop( $last_tag_history ); if ( array_pop( $td_history ) ) { - $previous = "{$previous}"; + $previous = "\n{$previous}"; } if ( $first_character === '|' ) { - $last_tag = 'td'; + $last_tag = $td_tag; } elseif ( $first_character === '!' ) { - $last_tag = 'th'; + $last_tag = $th_tag; } elseif ( $first_character === '+' ) { - $last_tag = 'caption'; + $last_tag = $caption_tag; } else { $last_tag = ''; } @@ -831,15 +963,24 @@ class Parser { # A cell could contain both parameters and data $cell_data = explode( '|' , $cell , 2 ); + $attributes = ''; + # Bug 553: Note that a '|' inside an invalid link should not # be mistaken as delimiting cell parameters if ( strpos( $cell_data[0], '[[' ) !== false ) { - $cell = "{$previous}<{$last_tag}>{$cell}"; + if ($extra_td_attribs) $attributes = Sanitizer::collapseTagAttributes( $extra_td_attribs ); + $cell = "{$previous}<{$last_tag}{$attributes}>{$cell}"; } elseif ( count( $cell_data ) == 1 ) { - $cell = "{$previous}<{$last_tag}>{$cell_data[0]}"; + if ($extra_td_attribs) $attributes = Sanitizer::collapseTagAttributes( $extra_td_attribs ); + $cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[0]}"; } else { $attributes = $this->mStripState->unstripBoth( $cell_data[0] ); - $attributes = Sanitizer::fixTagAttributes( $attributes , $last_tag ); + + $attr = Sanitizer::decodeTagAttributes( $attributes ); + if ($convert_style) $attr['style'] = Sanitizer::styleFromAttributes( $attr ); + $attr = Sanitizer::validateTagAttributes( $attr, $last_tag ); + $attributes = Sanitizer::collapseTagAttributes( $attr, $extra_td_attribs ); + $cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[1]}"; } @@ -853,16 +994,16 @@ class Parser { # Closing open td, tr && table while ( count( $td_history ) > 0 ) { if ( array_pop( $td_history ) ) { - $out .= "\n"; + $out .= "\n"; } if ( array_pop( $tr_history ) ) { - $out .= "\n"; + $out .= "\n"; } if ( !array_pop( $has_opened_tr ) ) { - $out .= "\n" ; + $out .= "<$tr_tag><$td_tag>\n" ; } - $out .= "\n"; + $out .= "\n"; } # Remove trailing line-ending (b/c) @@ -871,7 +1012,7 @@ class Parser { } # special case: don't return empty table - if ( $out === "\n\n
" ) { + if ( $out === "<$table_tag>\n<$tr_tag><$td_tag>\n" ) { $out = ''; } @@ -905,9 +1046,9 @@ class Parser { $flag = 0; } else { $flag = Parser::PTD_FOR_INCLUSION; + } $dom = $this->preprocessToDom( $text, $flag ); $text = $frame->expand( $dom ); - } } else { # if $frame is not provided, then use old-style replaceVariables $text = $this->replaceVariables( $text ); @@ -931,8 +1072,8 @@ class Parser { $df = DateFormatter::getInstance(); $text = $df->reformat( $this->mOptions->getDateFormat(), $text ); } - $text = $this->doAllQuotes( $text ); $text = $this->replaceInternalLinks( $text ); + $text = $this->doAllQuotes( $text ); $text = $this->replaceExternalLinks( $text ); # replaceInternalLinks may sometimes leave behind @@ -1135,12 +1276,12 @@ class Parser { } # Count the number of occurrences of bold and italics mark-ups. # We are not counting sequences of five apostrophes. - if ( strlen( $arr[$i] ) == 2 ) { + if ( strlen( $arr[$i] ) == 2 ) { $numitalics++; - } elseif ( strlen( $arr[$i] ) == 3 ) { + } elseif ( strlen( $arr[$i] ) == 3 ) { $numbold++; - } elseif ( strlen( $arr[$i] ) == 5 ) { - $numitalics++; + } elseif ( strlen( $arr[$i] ) == 5 ) { + $numitalics++; $numbold++; } } @@ -1165,7 +1306,7 @@ class Parser { $firstspace = $i; } } elseif ( $x2 === ' ') { - if ( $firstsingleletterword == -1 ) { + if ( $firstsingleletterword == -1 ) { $firstsingleletterword = $i; } } else { @@ -1336,7 +1477,7 @@ class Parser { # Use the encoded URL # This means that users can paste URLs directly into the text - # Funny characters like ö aren't valid in URLs anyway + # Funny characters like ö aren't valid in URLs anyway # This was changed in August 2004 $s .= $sk->makeExternalLink( $url, $text, false, $linktype, $this->getExternalLinkAttribs( $url ) ) . $dtrail . $trail; @@ -1358,9 +1499,9 @@ class Parser { * (depending on configuration, namespace, and the URL's domain) and/or a * target attribute (depending on configuration). * - * @param string $url Optional URL, to extract the domain from for rel => + * @param $url String: optional URL, to extract the domain from for rel => * nofollow if appropriate - * @return array Associative array of HTML attributes + * @return Array: associative array of HTML attributes */ function getExternalLinkAttribs( $url = false ) { $attribs = array(); @@ -1391,9 +1532,10 @@ class Parser { /** * Replace unusual URL escape codes with their equivalent characters - * @param string - * @return string - * @static + * + * @param $url String + * @return String + * * @todo This can merge genuinely required bits in the path or query string, * breaking legit URLs. A proper fix would treat the various parts of * the URL differently; as a workaround, just use the output for @@ -1407,8 +1549,6 @@ class Parser { /** * Callback function used in replaceUnusualEscapes(). * Replaces unusual URL escape codes with their equivalent character - * @static - * @private */ private static function replaceUnusualEscapesCallback( $matches ) { $char = urldecode( $matches[0] ); @@ -1585,10 +1725,10 @@ class Parser { # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up, # the real problem is with the $e1 regex # See bug 1300. - # + # # Still some problems for cases where the ] is meant to be outside punctuation, # and no image is in sight. See bug 2095. - # + # if ( $text !== '' && substr( $m[3], 0, 1 ) === ']' && strpos( $text, '[' ) !== false @@ -1701,6 +1841,11 @@ class Parser { $wasblank = ( $text == '' ); if ( $wasblank ) { $text = $link; + } else { + # Bug 4598 madness. Handle the quotes only if they come from the alternate part + # [[Lista d''e paise d''o munno]] -> Lista d''e paise d''o munno + # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']] -> Criticism of Harry Potter + $text = $this->doQuotes($text); } # Link not escaped by : , create the various objects @@ -1799,7 +1944,7 @@ class Parser { wfProfileIn( __METHOD__."-always_known" ); # Some titles, such as valid special pages or files in foreign repos, should # be shown as bluelinks even though they're not included in the page table - # + # # FIXME: isAlwaysKnown() can be expensive for file links; we should really do # batch file existence checks for NS_FILE and NS_MEDIA if ( $iw == '' && $nt->isAlwaysKnown() ) { @@ -1834,12 +1979,12 @@ class Parser { * breaking URLs in the following text without breaking trails on the * wiki links, it's been made into a horrible function. * - * @param Title $nt - * @param string $text - * @param string $query - * @param string $trail - * @param string $prefix - * @return string HTML-wikitext mix oh yuck + * @param $nt Title + * @param $text String + * @param $query String + * @param $trail String + * @param $prefix String + * @return String: HTML-wikitext mix oh yuck */ function makeKnownLinkHolder( $nt, $text = '', $query = '', $trail = '', $prefix = '' ) { list( $inside, $trail ) = Linker::splitTrail( $trail ); @@ -1856,8 +2001,8 @@ class Parser { * Not needed quite as much as it used to be since free links are a bit * more sensible these days. But bracketed links are still an issue. * - * @param string more-or-less HTML - * @return string less-or-more HTML with NOPARSE bits + * @param $text String: more-or-less HTML + * @return String: less-or-more HTML with NOPARSE bits */ function armorLinks( $text ) { return preg_replace( '/\b(' . wfUrlProtocols() . ')/', @@ -1866,7 +2011,7 @@ class Parser { /** * Return true if subpage links should be expanded on this page. - * @return bool + * @return Boolean */ function areSubpagesAllowed() { # Some namespaces don't allow subpages @@ -1875,8 +2020,9 @@ class Parser { /** * Handle link to subpage if necessary - * @param string $target the source of the link - * @param string &$text the link text, modified as necessary + * + * @param $target String: the source of the link + * @param &$text String: the link text, modified as necessary * @return string the full name of the link * @private */ @@ -1897,44 +2043,46 @@ class Parser { $this->mLastSection = ''; return $result; } + /** * getCommon() returns the length of the longest common substring * of both arguments, starting at the beginning of both. - * @private - */ + * @private + */ function getCommon( $st1, $st2 ) { $fl = strlen( $st1 ); $shorter = strlen( $st2 ); - if ( $fl < $shorter ) { - $shorter = $fl; + if ( $fl < $shorter ) { + $shorter = $fl; } for ( $i = 0; $i < $shorter; ++$i ) { - if ( $st1{$i} != $st2{$i} ) { - break; + if ( $st1{$i} != $st2{$i} ) { + break; } } return $i; } + /** * These next three functions open, continue, and close the list * element appropriate to the prefix character passed into them. - * @private + * @private */ function openList( $char ) { $result = $this->closeParagraph(); - if ( '*' === $char ) { - $result .= '
  • '; - } elseif ( '#' === $char ) { - $result .= '
    1. '; - } elseif ( ':' === $char ) { - $result .= '
      '; + if ( '*' === $char ) { + $result .= '
      • '; + } elseif ( '#' === $char ) { + $result .= '
        1. '; + } elseif ( ':' === $char ) { + $result .= '
          '; } elseif ( ';' === $char ) { $result .= '
          '; $this->mDTopen = true; - } else { - $result = ''; + } else { + $result = ''; } return $result; @@ -1946,12 +2094,12 @@ class Parser { * @private */ function nextItem( $char ) { - if ( '*' === $char || '#' === $char ) { - return '
        2. '; + if ( '*' === $char || '#' === $char ) { + return '
        3. '; } elseif ( ':' === $char || ';' === $char ) { $close = '
      '; - if ( $this->mDTopen ) { - $close = ''; + if ( $this->mDTopen ) { + $close = ''; } if ( ';' === $char ) { $this->mDTopen = true; @@ -1970,10 +2118,10 @@ class Parser { * @private */ function closeList( $char ) { - if ( '*' === $char ) { - $text = '
'; - } elseif ( '#' === $char ) { - $text = ''; + if ( '*' === $char ) { + $text = ''; + } elseif ( '#' === $char ) { + $text = ''; } elseif ( ':' === $char ) { if ( $this->mDTopen ) { $this->mDTopen = false; @@ -1981,8 +2129,8 @@ class Parser { } else { $text = '
'; } - } else { - return ''; + } else { + return ''; } return $text."\n"; } @@ -1991,7 +2139,8 @@ class Parser { /** * Make lists from lines starting with ':', '*', '#', etc. (DBL) * - * @param $linestart bool whether or not this is at the start of a line. + * @param $text String + * @param $linestart Boolean: whether or not this is at the start of a line. * @private * @return string the lists rendered as HTML */ @@ -2001,7 +2150,7 @@ class Parser { # Parsing through the text line by line. The main thing # happening here is handling of block-level elements p, pre, # and making lists from lines starting with * # : etc. - # + # $textLines = StringUtils::explode( "\n", $text ); $lastPrefix = $output = ''; @@ -2128,7 +2277,7 @@ class Parser { $t = substr( $t, 1 ); } else { # paragraph - if ( trim( $t ) == '' ) { + if ( trim( $t ) === '' ) { if ( $paragraphStack ) { $output .= $paragraphStack.'
'; $paragraphStack = false; @@ -2180,10 +2329,11 @@ class Parser { /** * Split up a string on ':', ignoring any occurences inside tags * to prevent illegal overlapping. - * @param string $str the string to split - * @param string &$before set to everything before the ':' - * @param string &$after set to everything after the ':' - * return string the position of the ':', or false if none found + * + * @param $str String: the string to split + * @param &$before String: set to everything before the ':' + * @param &$after String: set to everything after the ':' + * return String: the position of the ':', or false if none found */ function findColonNoLinks( $str, &$before, &$after ) { wfProfileIn( __METHOD__ ); @@ -2347,7 +2497,7 @@ class Parser { */ function getVariableValue( $index, $frame=false ) { global $wgContLang, $wgSitename, $wgServer, $wgServerName; - global $wgScriptPath, $wgStylePath; + global $wgArticlePath, $wgScriptPath, $wgStylePath; /** * Some of these require message or data lookups and can be @@ -2500,6 +2650,13 @@ class Parser { # *after* a revision ID has been assigned. This is for null edits. $this->mOutput->setFlag( 'vary-revision' ); wfDebug( __METHOD__ . ": {{REVISIONMONTH}} used, setting vary-revision...\n" ); + $value = substr( $this->getRevisionTimestamp(), 4, 2 ); + break; + case 'revisionmonth1': + # Let the edit saving system know we should parse the page + # *after* a revision ID has been assigned. This is for null edits. + $this->mOutput->setFlag( 'vary-revision' ); + wfDebug( __METHOD__ . ": {{REVISIONMONTH1}} used, setting vary-revision...\n" ); $value = intval( substr( $this->getRevisionTimestamp(), 4, 2 ) ); break; case 'revisionyear': @@ -2614,6 +2771,8 @@ class Parser { case 'currentversion': $value = SpecialVersion::getVersion(); break; + case 'articlepath': + return $wgArticlePath; case 'sitename': return $wgSitename; case 'server': @@ -2645,7 +2804,7 @@ class Parser { } /** - * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers + * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers * * @private */ @@ -2663,8 +2822,8 @@ class Parser { * Preprocess some wikitext and return the document tree. * This is the ghost of replace_variables(). * - * @param string $text The text to parse - * @param integer flags Bitwise combination of: + * @param $text String: The text to parse + * @param $flags Integer: bitwise combination of: * self::PTD_FOR_INCLUSION Handle / as if the text is being * included. Default is to assume a direct page view. * @@ -2712,11 +2871,11 @@ class Parser { * self::OT_PREPROCESS: templates but not extension tags * self::OT_HTML: all templates and extension tags * - * @param string $tex The text to transform - * @param PPFrame $frame Object describing the arguments passed to the template. + * @param $text String: the text to transform + * @param $frame PPFrame Object describing the arguments passed to the template. * Arguments may also be provided as an associative array, as was the usual case before MW1.12. * Providing arguments this way may be useful for extensions wishing to perform variable replacement explicitly. - * @param bool $argsOnly Only do argument (triple-brace) expansion, not double-brace expansion + * @param $argsOnly Boolean: only do argument (triple-brace) expansion, not double-brace expansion * @private */ function replaceVariables( $text, $frame = false, $argsOnly = false ) { @@ -2768,17 +2927,18 @@ class Parser { * Warn the user when a parser limitation is reached * Will warn at most once the user per limitation type * - * @param string $limitationType, should be one of: - * 'expensive-parserfunction' (corresponding messages: - * 'expensive-parserfunction-warning', + * @param $limitationType String: should be one of: + * 'expensive-parserfunction' (corresponding messages: + * 'expensive-parserfunction-warning', * 'expensive-parserfunction-category') - * 'post-expand-template-argument' (corresponding messages: - * 'post-expand-template-argument-warning', + * 'post-expand-template-argument' (corresponding messages: + * 'post-expand-template-argument-warning', * 'post-expand-template-argument-category') - * 'post-expand-template-inclusion' (corresponding messages: - * 'post-expand-template-inclusion-warning', + * 'post-expand-template-inclusion' (corresponding messages: + * 'post-expand-template-inclusion-warning', * 'post-expand-template-inclusion-category') - * @params int $current, $max When an explicit limit has been + * @param $current Current value + * @param $max Maximum allowed, when an explicit limit has been * exceeded, provide the values (optional) */ function limitationWarn( $limitationType, $current=null, $max=null) { @@ -2792,12 +2952,12 @@ class Parser { * Return the text of a template, after recursively * replacing any variables or templates within the template. * - * @param array $piece The parts of the template + * @param $piece Array: the parts of the template * $piece['title']: the title, i.e. the part before the | * $piece['parts']: the parameter array * $piece['lineStart']: whether the brace was at the start of a line - * @param PPFrame The current frame, contains template arguments - * @return string the text of the template + * @param $frame PPFrame The current frame, contains template arguments + * @return String: the text of the template * @private */ function braceSubstitution( $piece, $frame ) { @@ -2863,7 +3023,7 @@ class Parser { if ( $id !== false ) { $text = $this->getVariableValue( $id, $frame ); if ( MagicWord::getCacheTTL( $id ) > -1 ) { - $this->mOutput->mContainsOldMagic = true; + $this->mOutput->updateCacheExpiry( MagicWord::getCacheTTL( $id ) ); } $found = true; } @@ -2981,8 +3141,8 @@ class Parser { $limit = $this->mOptions->getMaxTemplateDepth(); if ( $frame->depth >= $limit ) { $found = true; - $text = '' - . wfMsgForContent( 'parser-template-recursion-depth-warning', $limit ) + $text = '' + . wfMsgForContent( 'parser-template-recursion-depth-warning', $limit ) . ''; } } @@ -2992,9 +3152,9 @@ class Parser { if ( !$found && $title ) { wfProfileIn( __METHOD__ . '-loadtpl' ); if ( !$title->isExternal() ) { - if ( $title->getNamespace() == NS_SPECIAL - && $this->mOptions->getAllowSpecialInclusion() - && $this->ot['html'] ) + if ( $title->getNamespace() == NS_SPECIAL + && $this->mOptions->getAllowSpecialInclusion() + && $this->ot['html'] ) { $text = SpecialPage::capturePath( $title ); if ( is_string( $text ) ) { @@ -3084,7 +3244,7 @@ class Parser { # Escape nowiki-style return values $text = wfEscapeWikiText( $text ); } elseif ( is_string( $text ) - && !$piece['lineStart'] + && !$piece['lineStart'] && preg_match( '/^(?:{\\||:|;|#|\*)/', $text ) ) { # Bug 529: if the template begins with a table or block-level @@ -3095,8 +3255,16 @@ class Parser { if ( is_string( $text ) && !$this->incrementIncludeSize( 'post-expand', strlen( $text ) ) ) { # Error, oversize inclusion - $text = "[[$originalTitle]]" . - $this->insertStripItem( '' ); + if ( $titleText !== false ) { + # Make a working, properly escaped link if possible (bug 23588) + $text = "[[:$titleText]]"; + } else { + # This will probably not be a working link, but at least it may + # provide some hint of where the problem is + preg_replace( '/^:/', '', $originalTitle ); + $text = "[[:$originalTitle]]"; + } + $text .= $this->insertStripItem( '' ); $this->limitationWarn( 'post-expand-template-inclusion' ); } @@ -3150,7 +3318,7 @@ class Parser { * Fetch the unparsed text of a template and register a reference to it. */ function fetchTemplateAndTitle( $title ) { - $templateCb = $this->mOptions->getTemplateCallback(); + $templateCb = $this->mOptions->getTemplateCallback(); # Defaults to Parser::statelessFetchTemplate() $stuff = call_user_func( $templateCb, $title, $this ); $text = $stuff['text']; $finalTitle = isset( $stuff['finalTitle'] ) ? $stuff['finalTitle'] : $title; @@ -3321,17 +3489,15 @@ class Parser { * Return the text to be used for a given extension tag. * This is the ghost of strip(). * - * @param array $params Associative array of parameters: + * @param $params Associative array of parameters: * name PPNode for the tag name * attr PPNode for unparsed text where tag attributes are thought to be * attributes Optional associative array of parsed attributes * inner Contents of extension element * noClose Original text did not have a close tag - * @param PPFrame $frame + * @param $frame PPFrame */ function extensionSubstitution( $params, $frame ) { - global $wgRawHtml, $wgContLang; - $name = $frame->expand( $params['name'] ); $attrText = !isset( $params['attr'] ) ? null : $frame->expand( $params['attr'] ); $content = !isset( $params['inner'] ) ? null : $frame->expand( $params['inner'] ); @@ -3410,9 +3576,9 @@ class Parser { /** * Increment an include size counter * - * @param string $type The type of expansion - * @param integer $size The size of the text - * @return boolean False if this inclusion would take it over the maximum, true otherwise + * @param $type String: the type of expansion + * @param $size Integer: the size of the text + * @return Boolean: false if this inclusion would take it over the maximum, true otherwise */ function incrementIncludeSize( $type, $size ) { if ( $this->mIncludeSizes[$type] + $size > $this->mOptions->getMaxIncludeSize( $type ) ) { @@ -3426,7 +3592,7 @@ class Parser { /** * Increment the expensive function count * - * @return boolean False if the limit has been exceeded + * @return Boolean: false if the limit has been exceeded */ function incrementExpensiveFunctionCount() { global $wgExpensiveParserFunctionLimit; @@ -3468,11 +3634,10 @@ class Parser { $this->mShowToc = false; } if ( isset( $this->mDoubleUnderscores['hiddencat'] ) && $this->mTitle->getNamespace() == NS_CATEGORY ) { - $this->mOutput->setProperty( 'hiddencat', 'y' ); $this->addTrackingCategory( 'hidden-category-category' ); } # (bug 8068) Allow control over whether robots index a page. - # + # # FIXME (bug 14899): __INDEX__ always overrides __NOINDEX__ here! This # is not desirable, the last one on the page should win. if ( isset( $this->mDoubleUnderscores['noindex'] ) && $this->mTitle->canUseNoindex() ) { @@ -3483,6 +3648,11 @@ class Parser { $this->mOutput->setIndexPolicy( 'index' ); $this->addTrackingCategory( 'index-category' ); } + + # Cache all double underscores in the database + foreach ( $this->mDoubleUnderscores as $key => $val ) { + $this->mOutput->setProperty( $key, '' ); + } wfProfileOut( __METHOD__ ); return $text; @@ -3491,8 +3661,9 @@ class Parser { /** * Add a tracking category, getting the title from a system message, * or print a debug message if the title is invalid. - * @param $msg String message key - * @return Bool whether the addition was successful + * + * @param $msg String: message key + * @return Boolean: whether the addition was successful */ protected function addTrackingCategory( $msg ) { $cat = wfMsgForContent( $msg ); @@ -3522,25 +3693,21 @@ class Parser { * It loops through all headlines, collects the necessary data, then splits up the * string and re-inserts the newly formatted headlines. * - * @param string $text - * @param string $origText Original, untouched wikitext - * @param boolean $isMain + * @param $text String + * @param $origText String: original, untouched wikitext + * @param $isMain Boolean * @private */ function formatHeadings( $text, $origText, $isMain=true ) { global $wgMaxTocLevel, $wgContLang, $wgHtml5, $wgExperimentalHtmlIds; $doNumberHeadings = $this->mOptions->getNumberHeadings(); - $showEditLink = $this->mOptions->getEditSection(); - - # Do not call quickUserCan unless necessary - if ( $showEditLink && !$this->mTitle->quickUserCan( 'edit' ) ) { - $showEditLink = 0; - } - + # Inhibit editsection links if requested in the page - if ( isset( $this->mDoubleUnderscores['noeditsection'] ) || $this->mOptions->getIsPrintable() ) { + if ( isset( $this->mDoubleUnderscores['noeditsection'] ) ) { $showEditLink = 0; + } else { + $showEditLink = $this->mOptions->getEditSection(); } # Get all headlines for numbering them and adding funky stuff like [edit] @@ -3697,8 +3864,7 @@ class Parser { # For the anchor, strip out HTML-y stuff period $safeHeadline = preg_replace( '/<.*?'.'>/', '', $safeHeadline ); - $safeHeadline = preg_replace( '/[ _]+/', ' ', $safeHeadline ); - $safeHeadline = trim( $safeHeadline ); + $safeHeadline = Sanitizer::normalizeSectionNameWhitespace( $safeHeadline ); # Save headline for section edit hint before it's escaped $headlineHint = $safeHeadline; @@ -3706,7 +3872,7 @@ class Parser { if ( $wgHtml5 && $wgExperimentalHtmlIds ) { # For reverse compatibility, provide an id that's # HTML4-compatible, like we used to. - # + # # It may be worth noting, academically, that it's possible for # the legacy anchor to conflict with a non-legacy headline # anchor on the page. In this case likely the "correct" thing @@ -3863,113 +4029,18 @@ class Parser { } } - /** - * Merge $tree2 into $tree1 by replacing the section with index - * $section in $tree1 and its descendants with the sections in $tree2. - * Note that in the returned section tree, only the 'index' and - * 'byteoffset' fields are guaranteed to be correct. - * @param $tree1 array Section tree from ParserOutput::getSectons() - * @param $tree2 array Section tree - * @param $section int Section index - * @param $title Title Title both section trees come from - * @param $len2 int Length of the original wikitext for $tree2 - * @return array Merged section tree - */ - public static function mergeSectionTrees( $tree1, $tree2, $section, $title, $len2 ) { - global $wgContLang; - $newTree = array(); - $targetLevel = false; - $merged = false; - $lastLevel = 1; - $nextIndex = 1; - $numbering = array( 0 ); - $titletext = $title->getPrefixedDBkey(); - foreach ( $tree1 as $s ) { - if ( $targetLevel !== false ) { - if ( $s['level'] <= $targetLevel ) { - # We've skipped enough - $targetLevel = false; - } else { - continue; - } - } - if ( $s['index'] != $section || - $s['fromtitle'] != $titletext ) { - self::incrementNumbering( $numbering, - $s['toclevel'], $lastLevel ); - - # Rewrite index, byteoffset and number - if ( $s['fromtitle'] == $titletext ) { - $s['index'] = $nextIndex++; - if ( $merged ) { - $s['byteoffset'] += $len2; - } - } - $s['number'] = implode( '.', array_map( - array( $wgContLang, 'formatnum' ), - $numbering ) ); - $lastLevel = $s['toclevel']; - $newTree[] = $s; - } else { - # We're at $section - # Insert sections from $tree2 here - foreach ( $tree2 as $s2 ) { - # Rewrite the fields in $s2 - # before inserting it - $s2['toclevel'] += $s['toclevel'] - 1; - $s2['level'] += $s['level'] - 1; - $s2['index'] = $nextIndex++; - $s2['byteoffset'] += $s['byteoffset']; - - self::incrementNumbering( $numbering, - $s2['toclevel'], $lastLevel ); - $s2['number'] = implode( '.', array_map( - array( $wgContLang, 'formatnum' ), - $numbering ) ); - $lastLevel = $s2['toclevel']; - $newTree[] = $s2; - } - # Skip all descendants of $section in $tree1 - $targetLevel = $s['level']; - $merged = true; - } - } - return $newTree; - } - - /** - * Increment a section number. Helper function for mergeSectionTrees() - * @param $number array Array representing a section number - * @param $level int Current TOC level (depth) - * @param $lastLevel int Level of previous TOC entry - */ - private static function incrementNumbering( &$number, $level, $lastLevel ) { - if ( $level > $lastLevel ) { - $number[$level - 1] = 1; - } elseif ( $level < $lastLevel ) { - foreach ( $number as $key => $unused ) - if ( $key >= $level ) { - unset( $number[$key] ); - } - $number[$level - 1]++; - } else { - $number[$level - 1]++; - } - } - /** * Transform wiki markup when saving a page by doing \r\n -> \n * conversion, substitting signatures, {{subst:}} templates, etc. * - * @param string $text the text to transform - * @param Title &$title the Title object for the current article - * @param User $user the User object describing the current user - * @param ParserOptions $options parsing options - * @param bool $clearState whether to clear the parser state first - * @return string the altered wiki markup - * @public + * @param $text String: the text to transform + * @param &$title Title: the Title object for the current article + * @param $user User: the User object describing the current user + * @param $options ParserOptions: parsing options + * @param $clearState Boolean: whether to clear the parser state first + * @return String: the altered wiki markup */ - function preSaveTransform( $text, Title $title, $user, $options, $clearState = true ) { + public function preSaveTransform( $text, Title $title, $user, $options, $clearState = true ) { $this->mOptions = $options; $this->setTitle( $title ); $this->setOutputType( self::OT_WIKI ); @@ -4012,7 +4083,7 @@ class Parser { $ts = date( 'YmdHis', $unixts ); $tzMsg = date( 'T', $unixts ); # might vary on DST changeover! - # Allow translation of timezones trough wiki. date() can return + # Allow translation of timezones through wiki. date() can return # whatever crap the system uses, localised or not, so we cannot # ship premade translations. $key = 'timezone-' . strtolower( trim( $tzMsg ) ); @@ -4075,7 +4146,10 @@ class Parser { * If you have pre-fetched the nickname or the fancySig option, you can * specify them here to save a database query. * - * @param User $user + * @param $user User + * @param $nickname String: nickname to use or false to use user's default nickname + * @param $fancySig Boolean: whether the nicknname is the complete signature + * or null to use default value * @return string */ function getUserSig( &$user, $nickname = false, $fancySig = null ) { @@ -4124,7 +4198,7 @@ class Parser { /** * Check that the user's signature contains no bad XML * - * @param string $text + * @param $text String * @return mixed An expanded string, or false if invalid. */ function validateSig( $text ) { @@ -4137,9 +4211,9 @@ class Parser { * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig * 2) Substitute all transclusions * - * @param string $text + * @param $text String * @param $parsing Whether we're cleaning (preferences save) or parsing - * @return string Signature text + * @return String: signature text */ function cleanSig( $text, $parsing = false ) { if ( !$parsing ) { @@ -4176,8 +4250,9 @@ class Parser { /** * Strip ~~~, ~~~~ and ~~~~~ out of signatures - * @param string $text - * @return string Signature text with /~{3,5}/ removed + * + * @param $text String + * @return String: signature text with /~{3,5}/ removed */ function cleanSigInSig( $text ) { $text = preg_replace( '/~{3,5}/', '', $text ); @@ -4187,9 +4262,8 @@ class Parser { /** * Set up some variables which are usually set up in parse() * so that an external function can call some class members with confidence - * @public */ - function startExternalParse( &$title, $options, $outputType, $clearState = true ) { + public function startExternalParse( &$title, $options, $outputType, $clearState = true ) { $this->setTitle( $title ); $this->mOptions = $options; $this->setOutputType( $outputType ); @@ -4201,12 +4275,11 @@ class Parser { /** * Wrapper for preprocess() * - * @param string $text the text to preprocess - * @param ParserOptions $options options - * @return string - * @public + * @param $text String: the text to preprocess + * @param $options ParserOptions: options + * @return String */ - function transformMsg( $text, $options ) { + public function transformMsg( $text, $options ) { global $wgTitle; static $executing = false; @@ -4227,19 +4300,16 @@ class Parser { /** * Create an HTML-style tag, e.g. special text * The callback should have the following form: - * function myParserHook( $text, $params, &$parser ) { ... } + * function myParserHook( $text, $params, $parser ) { ... } * * Transform and return $text. Use $parser for any required context, e.g. use * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions * - * @public - * - * @param mixed $tag The tag to use, e.g. 'hook' for - * @param mixed $callback The callback function (and object) to use for the tag - * + * @param $tag Mixed: the tag to use, e.g. 'hook' for + * @param $callback Mixed: the callback function (and object) to use for the tag * @return The old value of the mTagHooks array associated with the hook */ - function setHook( $tag, $callback ) { + public function setHook( $tag, $callback ) { $tag = strtolower( $tag ); $oldVal = isset( $this->mTagHooks[$tag] ) ? $this->mTagHooks[$tag] : null; $this->mTagHooks[$tag] = $callback; @@ -4282,11 +4352,9 @@ class Parser { * nowiki Wiki markup in the return value should be escaped * isHTML The returned text is HTML, armour it against wikitext transformation * - * @public - * - * @param string $id The magic word ID - * @param mixed $callback The callback function (and object) to use - * @param integer $flags a combination of the following flags: + * @param $id String: The magic word ID + * @param $callback Mixed: the callback function (and object) to use + * @param $flags Integer: a combination of the following flags: * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}} * * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This @@ -4310,7 +4378,7 @@ class Parser { * * @return The old callback function for this name, if any */ - function setFunctionHook( $id, $callback, $flags = 0 ) { + public function setFunctionHook( $id, $callback, $flags = 0 ) { global $wgContLang; $oldVal = isset( $this->mFunctionHooks[$id] ) ? $this->mFunctionHooks[$id][0] : null; @@ -4345,7 +4413,7 @@ class Parser { /** * Get all registered function hook identifiers * - * @return array + * @return Array */ function getFunctionHooks() { return array_keys( $this->mFunctionHooks ); @@ -4382,8 +4450,9 @@ class Parser { /** * Replace link placeholders with plain text of links * (not HTML-formatted). - * @param string $text - * @return string + * + * @param $text String + * @return String */ function replaceLinkHoldersText( $text ) { return $this->mLinkHolders->replaceText( $text ); @@ -4512,9 +4581,10 @@ class Parser { /** * Parse image options text and use it to make an image - * @param Title $title - * @param string $options - * @param LinkHolderArray $holders + * + * @param $title Title + * @param $options String + * @param $holders LinkHolderArray */ function makeImage( $title, $options, $holders = false ) { # Check if the options text is of the form "options|alt text" @@ -4669,13 +4739,13 @@ class Parser { # came to also set the caption, ordinary text after the image -- which # makes no sense, because that just repeats the text multiple times in # screen readers. It *also* came to set the title attribute. - # + # # Now that we have an alt attribute, we should not set the alt text to # equal the caption: that's worse than useless, it just repeats the # text. This is the framed/thumbnail case. If there's no caption, we # use the unnamed parameter for alt text as well, just for the time be- # ing, if the unnamed param is set and the alt param is not. - # + # # For the future, we need to figure out if we want to tweak this more, # e.g., introducing a title= parameter for the title; ignoring the un- # named parameter entirely for images without a caption; adding an ex- @@ -4707,7 +4777,7 @@ class Parser { wfRunHooks( 'ParserMakeImageParams', array( $title, $file, &$params ) ); # Linker does the rest - $ret = $sk->makeImageLink2( $title, $file, $params['frame'], $params['handler'], $time, $descQuery ); + $ret = $sk->makeImageLink2( $title, $file, $params['frame'], $params['handler'], $time, $descQuery, $this->mOptions->getThumbSize() ); # Give the handler a chance to modify the parser object if ( $handler ) { @@ -4742,15 +4812,17 @@ class Parser { */ function disableCache() { wfDebug( "Parser output marked as uncacheable.\n" ); - $this->mOutput->mCacheTime = -1; + $this->mOutput->setCacheTime( -1 ); // old style, for compatibility + $this->mOutput->updateCacheExpiry( 0 ); // new style, for consistency } - /**#@+ + /** * Callback from the Sanitizer for expanding items found in HTML attribute * values, so they can be safely tested and escaped. - * @param string $text - * @param PPFrame $frame - * @return string + * + * @param $text String + * @param $frame PPFrame + * @return String * @private */ function attributeStripCallback( &$text, $frame = false ) { @@ -4759,24 +4831,12 @@ class Parser { return $text; } - /**#@-*/ - - /**#@+ - * Accessor/mutator - */ - function Title( $x = null ) { return wfSetVar( $this->mTitle, $x ); } - function Options( $x = null ) { return wfSetVar( $this->mOptions, $x ); } - function OutputType( $x = null ) { return wfSetVar( $this->mOutputType, $x ); } - /**#@-*/ - - /**#@+ + /** * Accessor */ - function getTags() { - return array_merge( array_keys( $this->mTransparentTagHooks ), array_keys( $this->mTagHooks ) ); + function getTags() { + return array_merge( array_keys( $this->mTransparentTagHooks ), array_keys( $this->mTagHooks ) ); } - /**#@-*/ - /** * Break wikitext input into sections, and either pull or replace @@ -4784,8 +4844,8 @@ class Parser { * * External callers should use the getSection and replaceSection methods. * - * @param string $text Page wikitext - * @param string $section A section identifier string of the form: + * @param $text String: Page wikitext + * @param $section String: a section identifier string of the form: * - - ... -
* * Currently the only recognised flag is "T", which means the target section number @@ -4798,10 +4858,10 @@ class Parser { * pull the given section along with its lower-level subsections. If the section is * not found, $mode=get will return $newtext, and $mode=replace will return $text. * - * @param string $mode One of "get" or "replace" - * @param string $newText Replacement text for section data. - * @return string for "get", the extracted section text. - * for "replace", the whole page with the section replaced. + * @param $mode String: one of "get" or "replace" + * @param $newText String: replacement text for section data. + * @return String: for "get", the extracted section text. + * for "replace", the whole page with the section replaced. */ private function extractSections( $text, $section, $mode, $newText='' ) { global $wgTitle; @@ -4833,11 +4893,11 @@ class Parser { # Section zero doesn't nest, level=big $targetLevel = 1000; } else { - while ( $node ) { - if ( $node->getName() === 'h' ) { - $bits = $node->splitHeading(); + while ( $node ) { + if ( $node->getName() === 'h' ) { + $bits = $node->splitHeading(); if ( $bits['i'] == $sectionIndex ) { - $targetLevel = $bits['level']; + $targetLevel = $bits['level']; break; } } @@ -4903,9 +4963,9 @@ class Parser { * * If a section contains subsections, these are also returned. * - * @param string $text text to look in - * @param string $section section identifier - * @param string $deftext default to return if section is not found + * @param $text String: text to look in + * @param $section String: section identifier + * @param $deftext String: default to return if section is not found * @return string text of the requested section */ public function getSection( $text, $section, $deftext='' ) { @@ -4916,6 +4976,15 @@ class Parser { return $this->extractSections( $oldtext, $section, "replace", $text ); } + /** + * Get the ID of the revision we are parsing + * + * @return Mixed: integer or null + */ + function getRevisionId() { + return $this->mRevisionId; + } + /** * Get the timestamp associated with the current revision, adjusted for * the default server-local timestamp @@ -4936,7 +5005,7 @@ class Parser { # The cryptic '' timezone parameter tells to use the site-default # timezone offset instead of the user settings. - # + # # Since this value will be saved into the parser cache, served # to other users, and potentially even used inside links and such, # it needs to be consistent for all visitors. @@ -4949,6 +5018,8 @@ class Parser { /** * Get the name of the user that edited the last revision + * + * @return String: user name */ function getRevisionUser() { # if this template is subst: the revision id will be blank, @@ -4970,6 +5041,7 @@ class Parser { */ public function setDefaultSort( $sort ) { $this->mDefaultSort = $sort; + $this->mOutput->setProperty( 'defaultsort', $sort ); } /** @@ -4979,15 +5051,10 @@ class Parser { * @return string */ public function getDefaultSort() { - global $wgCategoryPrefixedDefaultSortkey; if ( $this->mDefaultSort !== false ) { return $this->mDefaultSort; - } elseif ( $this->mTitle->getNamespace() == NS_CATEGORY || - !$wgCategoryPrefixedDefaultSortkey ) - { - return $this->mTitle->getText(); } else { - return $this->mTitle->getPrefixedText(); + return $this->mTitle->getCategorySortkey(); } } @@ -5009,19 +5076,23 @@ class Parser { public function guessSectionNameFromWikiText( $text ) { # Strip out wikitext links(they break the anchor) $text = $this->stripSectionName( $text ); - $headline = Sanitizer::decodeCharReferences( $text ); - # strip out HTML - $headline = StringUtils::delimiterReplace( '<', '>', '', $headline ); - $headline = trim( $headline ); - $sectionanchor = '#' . urlencode( str_replace( ' ', '_', $headline ) ); - $replacearray = array( - '%3A' => ':', - '%' => '.' - ); - return str_replace( - array_keys( $replacearray ), - array_values( $replacearray ), - $sectionanchor ); + $text = Sanitizer::normalizeSectionNameWhitespace( $text ); + return '#' . Sanitizer::escapeId( $text, 'noninitial' ); + } + + /** + * Same as guessSectionNameFromWikiText(), but produces legacy anchors + * instead. For use in redirects, since IE6 interprets Redirect: headers + * as something other than UTF-8 (apparently?), resulting in breakage. + * + * @param $text String: The section name + * @return string An anchor + */ + public function guessLegacySectionNameFromWikiText( $text ) { + # Strip out wikitext links(they break the anchor) + $text = $this->stripSectionName( $text ); + $text = Sanitizer::normalizeSectionNameWhitespace( $text ); + return '#' . Sanitizer::escapeId( $text, array( 'noninitial', 'legacy' ) ); } /** @@ -5034,7 +5105,7 @@ class Parser { * to create valid section anchors by mimicing the output of the * parser when headings are parsed. * - * @param $text string Text string to be stripped of wikitext + * @param $text String: text string to be stripped of wikitext * for use in a Section anchor * @return Filtered text string */ @@ -5124,8 +5195,8 @@ class Parser { # data in an array. $stripState = new StripState; $pos = 0; - while ( ( $start_pos = strpos( $text, $this->mUniqPrefix, $pos ) ) - && ( $end_pos = strpos( $text, self::MARKER_SUFFIX, $pos ) ) ) + while ( ( $start_pos = strpos( $text, $this->mUniqPrefix, $pos ) ) + && ( $end_pos = strpos( $text, self::MARKER_SUFFIX, $pos ) ) ) { $end_pos += strlen( self::MARKER_SUFFIX ); $marker = substr( $text, $start_pos, $end_pos-$start_pos );