* <pre>
* There are five main entry points into the Parser class:
* parse()
- * produces HTML output
+ * produces HTML output
* preSaveTransform().
- * produces altered wiki markup.
+ * produces altered wiki markup.
* preprocess()
- * removes HTML comments and expands templates
+ * removes HTML comments and expands templates
* cleanSig()
- * Cleans a signature before saving it to preferences
+ * Cleans a signature before saving it to preferences
* extractSections()
- * Extracts sections from an article for section editing
+ * Extracts sections from an article for section editing
+ * getPreloadText()
+ * Removes <noinclude> sections, and <includeonly> tags.
*
* Globals used:
* objects: $wgLang, $wgContLang
*
* @ingroup Parser
*/
-class Parser
-{
+class Parser {
/**
* Update this version number when the ParserOutput format
* changes in an incompatible way, so the parser cache
const EXT_IMAGE_REGEX = '/^(http:\/\/|https:\/\/)([^][<>"\\x00-\\x20\\x7F]+)
\\/([A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]+)\\.((?i)gif|png|jpg|jpeg)$/Sx';
- // State constants for the definition list colon extraction
+ # State constants for the definition list colon extraction
const COLON_STATE_TEXT = 0;
const COLON_STATE_TAG = 1;
const COLON_STATE_TAGSTART = 2;
const COLON_STATE_COMMENTDASH = 6;
const COLON_STATE_COMMENTDASHDASH = 7;
- // Flags for preprocessToDom
+ # Flags for preprocessToDom
const PTD_FOR_INCLUSION = 1;
- // Allowed values for $this->mOutputType
- // Parameter to startExternalParse().
- const OT_HTML = 1;
- const OT_WIKI = 2;
- const OT_PREPROCESS = 3;
+ # Allowed values for $this->mOutputType
+ # Parameter to startExternalParse().
+ const OT_HTML = 1; # like parse()
+ const OT_WIKI = 2; # like preSaveTransform()
+ const OT_PREPROCESS = 3; # like preprocess()
const OT_MSG = 3;
+ const OT_PLAIN = 4; # like extractSections() - portions of the original are returned unchanged.
- // Marker Suffix needs to be accessible staticly.
+ # Marker Suffix needs to be accessible staticly.
const MARKER_SUFFIX = "-QINU\x7f";
- /**#@+
- * @private
- */
# Persistent:
- var $mTagHooks, $mTransparentTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables,
- $mImageParams, $mImageParamsMagicArray, $mStripList, $mMarkerIndex, $mPreprocessor,
- $mExtLinkBracketedRegex, $mUrlProtocols, $mDefaultStripList, $mVarCache, $mConf,
- $mFunctionTagHooks;
+ var $mTagHooks, $mTransparentTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables;
+ var $mSubstWords, $mImageParams, $mImageParamsMagicArray, $mStripList, $mMarkerIndex;
+ var $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols, $mDefaultStripList;
+ var $mVarCache, $mConf, $mFunctionTagHooks;
# Cleared with clearState():
var $mIncludeCount, $mArgStack, $mLastSection, $mInPre;
var $mLinkHolders, $mLinkID;
var $mIncludeSizes, $mPPNodeCount, $mDefaultSort;
- var $mTplExpandCache; // empty-frame expansion cache
+ var $mTplExpandCache; # empty-frame expansion cache
var $mTplRedirCache, $mTplDomCache, $mHeadings, $mDoubleUnderscores;
- var $mExpensiveFunctionCount; // number of expensive parser function calls
+ var $mExpensiveFunctionCount; # number of expensive parser function calls
# Temporary
# These are variables reset at least once per parse regardless of $clearState
- var $mOptions, // ParserOptions object
- $mTitle, // Title context, used for self-link rendering and similar things
- $mOutputType, // Output type, one of the OT_xxx constants
- $ot, // Shortcut alias, see setOutputType()
- $mRevisionId, // ID to display in {{REVISIONID}} tags
- $mRevisionTimestamp, // The timestamp of the specified revision ID
- $mRevIdForTs; // The revision ID which was used to fetch the timestamp
-
- /**#@-*/
+ var $mOptions; # ParserOptions object
+ var $mTitle; # Title context, used for self-link rendering and similar things
+ var $mOutputType; # Output type, one of the OT_xxx constants
+ var $ot; # Shortcut alias, see setOutputType()
+ var $mRevisionId; # ID to display in {{REVISIONID}} tags
+ var $mRevisionTimestamp; # The timestamp of the specified revision ID
+ var $mRevIdForTs; # The revision ID which was used to fetch the timestamp
/**
* Constructor
$this->mFunctionHooks = array();
$this->mFunctionTagHooks = array();
$this->mFunctionSynonyms = array( 0 => array(), 1 => array() );
- $this->mDefaultStripList = $this->mStripList = array( 'nowiki', 'gallery', 'a' );
+ $this->mDefaultStripList = $this->mStripList = array();
$this->mUrlProtocols = wfUrlProtocols();
$this->mExtLinkBracketedRegex = '/\[(\b(' . wfUrlProtocols() . ')'.
- '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x0a\\x0d]*?)\]/S';
+ '[^][<>"\\x00-\\x20\\x7F]+) *([^\]\\x00-\\x08\\x0a-\\x1F]*?)\]/S';
$this->mVarCache = array();
if ( isset( $conf['preprocessorClass'] ) ) {
$this->mPreprocessorClass = $conf['preprocessorClass'];
} elseif ( extension_loaded( 'domxml' ) ) {
- // PECL extension that conflicts with the core DOM extension (bug 13770)
+ # PECL extension that conflicts with the core DOM extension (bug 13770)
wfDebug( "Warning: you have the obsolete domxml extension for PHP. Please remove it!\n" );
$this->mPreprocessorClass = 'Preprocessor_Hash';
} elseif ( extension_loaded( 'dom' ) ) {
wfProfileIn( __METHOD__ );
- $this->setHook( 'pre', array( $this, 'renderPreTag' ) );
CoreParserFunctions::register( $this );
+ CoreTagHooks::register( $this );
$this->initialiseVariables();
wfRunHooks( 'ParserFirstCallInit', array( &$this ) );
wfProfileOut( __METHOD__ );
}
- function setOutputType( $ot ) {
- $this->mOutputType = $ot;
- // Shortcut alias
- $this->ot = array(
- 'html' => $ot == self::OT_HTML,
- 'wiki' => $ot == self::OT_WIKI,
- 'pre' => $ot == self::OT_PREPROCESS,
- );
- }
-
- /**
- * Set the context title
- */
- function setTitle( $t ) {
- if ( !$t || $t instanceof FakeTitle ) {
- $t = Title::newFromText( 'NO TITLE' );
- }
-
- if ( strval( $t->getFragment() ) !== '' ) {
- # Strip the fragment to avoid various odd effects
- $this->mTitle = clone $t;
- $this->mTitle->setFragment( '' );
- } else {
- $this->mTitle = $t;
- }
- }
-
- /**
- * Accessor for mUniqPrefix.
- *
- * @public
- */
- function uniqPrefix() {
- if( !isset( $this->mUniqPrefix ) ) {
- // @todo Fixme: this is probably *horribly wrong*
- // LanguageConverter seems to want $wgParser's uniqPrefix, however
- // if this is called for a parser cache hit, the parser may not
- // have ever been initialized in the first place.
- // Not really sure what the heck is supposed to be going on here.
- return '';
- //throw new MWException( "Accessing uninitialized mUniqPrefix" );
- }
- return $this->mUniqPrefix;
- }
-
/**
* Convert wikitext to HTML
* Do not call this function recursively.
* 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 );
$this->clearState();
}
+ $options->resetUsage();
$this->mOptions = $options;
- $this->setTitle( $title ); // Page title has to be set for the pre-processor
+ $this->setTitle( $title ); # Page title has to be set for the pre-processor
$oldRevisionId = $this->mRevisionId;
$oldRevisionTimestamp = $this->mRevisionTimestamp;
- if( $revid !== null ) {
+ if ( $revid !== null ) {
$this->mRevisionId = $revid;
$this->mRevisionTimestamp = null;
}
$fixtags = array(
# french spaces, last one Guillemet-left
# only if there is something before the space
- '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1 \\2',
+ '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1 \\2',
# french spaces, Guillemet-right
- '/(\\302\\253) /' => '\\1 ',
- '/ (!\s*important)/' => ' \\1', # Beware of CSS magic word !important, bug #11874.
+ '/(\\302\\253) /' => '\\1 ',
+ '/ (!\s*important)/' => ' \\1', # Beware of CSS magic word !important, bug #11874.
);
- $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
+ $text = preg_replace( array_keys( $fixtags ), array_values( $fixtags ), $text );
$text = $this->doBlockLevels( $text, $linestart );
$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 <nowiki> 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 <nowiki> 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() );
- } elseif ( !( $wgDisableLangConversion
- || isset( $this->mDoubleUnderscores['notitleconvert'] ) ) ) {
- $this->mOutput->setTitleText( $wgContLang->convert( $title ) );
+ /**
+ * 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 );
$elements = array_keys( $this->mTransparentTagHooks );
$text = self::extractTagsAndParams( $elements, $text, $matches, $uniq_prefix );
- foreach( $matches as $marker => $data ) {
+ foreach ( $matches as $marker => $data ) {
list( $element, $content, $params, $tag ) = $data;
$tagName = strtolower( $element );
- if( isset( $this->mTransparentTagHooks[$tagName] ) ) {
- $output = call_user_func_array( $this->mTransparentTagHooks[$tagName],
- array( $content, $params, $this ) );
+ if ( isset( $this->mTransparentTagHooks[$tagName] ) ) {
+ $output = call_user_func_array( $this->mTransparentTagHooks[$tagName], array( $content, $params, $this ) );
} else {
$output = $tag;
}
$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
# 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";
$limitReport =
"NewPP limit report\n" .
- "Preprocessor node count: {$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;
* 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__ );
wfProfileIn( __METHOD__ );
$this->clearState();
$this->setOutputType( self::OT_PREPROCESS );
+ $options->resetUsage();
$this->mOptions = $options;
$this->setTitle( $title );
- if( $revid !== null ) {
+ if ( $revid !== null ) {
$this->mRevisionId = $revid;
}
wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) );
return $text;
}
+ /**
+ * Process the wikitext for the ?preload= feature. (bug 5210)
+ *
+ * <noinclude>, <includeonly> etc. are parsed as for template transclusion,
+ * comments, templates, arguments, tags hooks and parser functions are untouched.
+ */
+ public function getPreloadText( $text, $title, $options ) {
+ # Parser (re)initialisation
+ $this->clearState();
+ $this->setOutputType( self::OT_PLAIN );
+ $options->resetUsage();
+ $this->mOptions = $options;
+ $this->setTitle( $title );
+
+ $flags = PPFrame::NO_ARGS | PPFrame::NO_TEMPLATES;
+ $dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION );
+ return $this->getPreprocessor()->newFrame()->expand( $dom, $flags );
+ }
+
/**
* Get a random string
*
* @private
* @static
*/
- function getRandomString() {
- return dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff));
+ static function getRandomString() {
+ return dechex( mt_rand( 0, 0x7fffffff ) ) . dechex( mt_rand( 0, 0x7fffffff ) );
+ }
+
+ /**
+ * Accessor for mUniqPrefix.
+ *
+ * @return String
+ */
+ public function uniqPrefix() {
+ if ( !isset( $this->mUniqPrefix ) ) {
+ # @todo Fixme: this is probably *horribly wrong*
+ # LanguageConverter seems to want $wgParser's uniqPrefix, however
+ # if this is called for a parser cache hit, the parser may not
+ # have ever been initialized in the first place.
+ # Not really sure what the heck is supposed to be going on here.
+ return '';
+ # throw new MWException( "Accessing uninitialized mUniqPrefix" );
+ }
+ return $this->mUniqPrefix;
+ }
+
+ /**
+ * Set the context title
+ */
+ function setTitle( $t ) {
+ if ( !$t || $t instanceof FakeTitle ) {
+ $t = Title::newFromText( 'NO TITLE' );
+ }
+
+ if ( strval( $t->getFragment() ) !== '' ) {
+ # Strip the fragment to avoid various odd effects
+ $this->mTitle = clone $t;
+ $this->mTitle->setFragment( '' );
+ } else {
+ $this->mTitle = $t;
+ }
+ }
+
+ /**
+ * Accessor for the Title object
+ *
+ * @return Title object
+ */
+ function &getTitle() {
+ return $this->mTitle;
}
- function &getTitle() { return $this->mTitle; }
- function getOptions() { return $this->mOptions; }
- function getRevisionId() { return $this->mRevisionId; }
- function getOutput() { return $this->mOutput; }
- function nextLinkID() { return $this->mLinkID++; }
+ /**
+ * Accessor/mutator for the 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;
/**
* Get a preprocessor object
+ *
+ * @return Preprocessor instance
*/
function getPreprocessor() {
if ( !isset( $this->mPreprocessor ) ) {
*
* @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();
$taglist = implode( '|', $elements );
$start = "/<($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 ) {
+ if ( count( $p ) < 5 ) {
break;
}
- if( count( $p ) > 5 ) {
- // comment
+ if ( count( $p ) > 5 ) {
+ # comment
$element = $p[4];
$attributes = '';
$close = '';
$inside = $p[5];
} else {
- // tag
+ # tag
$element = $p[1];
$attributes = $p[2];
$close = $p[3];
$inside = $p[4];
}
- $marker = "$uniq_prefix-$element-" . sprintf('%08X', $n++) . self::MARKER_SUFFIX;
+ $marker = "$uniq_prefix-$element-" . sprintf( '%08X', $n++ ) . self::MARKER_SUFFIX;
$stripped .= $marker;
if ( $close === '/>' ) {
- // Empty element tag, <tag />
+ # Empty element tag, <tag />
$content = null;
$text = $inside;
$tail = null;
} else {
- if( $element === '!--' ) {
+ if ( $element === '!--' ) {
$end = '/(-->)/';
} else {
$end = "/(<\\/$element\\s*>)/i";
}
$q = preg_split( $end, $inside, 2, PREG_SPLIT_DELIM_CAPTURE );
$content = $q[0];
- if( count( $q ) < 3 ) {
+ if ( count( $q ) < 3 ) {
# No end tag -- let it run out to the end of the text.
$tail = '';
$text = '';
* 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;
}
/**
* @deprecated use replaceVariables
*/
- function strip( $text, $state, $stripcomments = false , $dontstrip = array () ) {
+ function strip( $text, $state, $stripcomments = false , $dontstrip = array() ) {
return $text;
}
*
* @private
*/
- function doTableStuff ( $text ) {
+ 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)
- $tr_history = array (); // Is currently a tr tag open?
- $tr_attributes = array (); // history of tr attributes
- $has_opened_tr = array(); // Did this table open a <tr> element?
- $indent_level = 0; // indent level of the table
+ $td_history = array(); # Is currently a td tag open?
+ $last_tag_history = array(); # Save history of last lag activated (td, th or caption)
+ $tr_history = array(); # Is currently a tr tag open?
+ $tr_attributes = array(); # history of tr attributes
+ $has_opened_tr = array(); # Did this table open a <tr> element?
+ $indent_level = 0; # indent level of the table
+
+ $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];
$matches = array();
if ( preg_match( '/^(:*)\{\|(.*)$/', $line , $matches ) ) {
- // First check if we are starting a new table
+ # First check if we are starting a new table
$indent_level = strlen( $matches[1] );
$attributes = $this->mStripState->unstripBoth( $matches[2] );
- $attributes = Sanitizer::fixTagAttributes ( $attributes , 'table' );
- $outLine = str_repeat( '<dl><dd>' , $indent_level ) . "<table{$attributes}>";
+ $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( '<dl><dd>' , $indent_level ) . "<$table_tag{$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 );
- } elseif ( count ( $td_history ) == 0 ) {
- // Don't do any of the following
+ } elseif ( count( $td_history ) == 0 ) {
+ # Don't do any of the following
$out .= $outLine."\n";
continue;
- } elseif ( substr ( $line , 0 , 2 ) === '|}' ) {
- // We are ending a table
- $line = '</table>' . substr ( $line , 2 );
- $last_tag = array_pop ( $last_tag_history );
+ } elseif ( substr( $line , 0 , 2 ) === '|}' ) {
+ # We are ending a table
+ $line = "</$table_tag>" . substr( $line , 2 );
+ $last_tag = array_pop( $last_tag_history );
- if ( !array_pop ( $has_opened_tr ) ) {
- $line = "<tr><td></td></tr>{$line}";
+ if ( !array_pop( $has_opened_tr ) ) {
+ $line = "<$tr_tag><$td_tag></$td_tag></$tr_tag>{$line}";
}
- if ( array_pop ( $tr_history ) ) {
- $line = "</tr>{$line}";
+ if ( array_pop( $tr_history ) ) {
+ $line = "</$tr_tag>{$line}";
}
- if ( array_pop ( $td_history ) ) {
+ if ( array_pop( $td_history ) ) {
$line = "</{$last_tag}>{$line}";
}
- array_pop ( $tr_attributes );
+ array_pop( $tr_attributes );
$outLine = $line . str_repeat( '</dd></dl>' , $indent_level );
- } elseif ( substr ( $line , 0 , 2 ) === '|-' ) {
- // Now we have a table row
+ } elseif ( substr( $line , 0 , 2 ) === '|-' ) {
+ # Now we have a table row
$line = preg_replace( '#^\|-+#', '', $line );
- // Whats after the tag is now only attributes
+ # Whats after the tag is now only attributes
$attributes = $this->mStripState->unstripBoth( $line );
- $attributes = Sanitizer::fixTagAttributes( $attributes, 'tr' );
+
+ $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 );
$line = '';
- $last_tag = array_pop ( $last_tag_history );
- array_pop ( $has_opened_tr );
- array_push ( $has_opened_tr , true );
+ $last_tag = array_pop( $last_tag_history );
+ array_pop( $has_opened_tr );
+ array_push( $has_opened_tr , true );
- if ( array_pop ( $tr_history ) ) {
- $line = '</tr>';
+ if ( array_pop( $tr_history ) ) {
+ $line = "</$tr_tag>";
}
- if ( array_pop ( $td_history ) ) {
+ if ( array_pop( $td_history ) ) {
$line = "</{$last_tag}>{$line}";
}
$outLine = $line;
- array_push ( $tr_history , false );
- array_push ( $td_history , false );
- array_push ( $last_tag_history , '' );
- }
- elseif ( $first_character === '|' || $first_character === '!' || substr ( $line , 0 , 2 ) === '|+' ) {
- // This might be cell elements, td, th or captions
- if ( substr ( $line , 0 , 2 ) === '|+' ) {
+ array_push( $tr_history , false );
+ array_push( $td_history , false );
+ array_push( $last_tag_history , '' );
+ } elseif ( $first_character === '|' || $first_character === '!' || substr( $line , 0 , 2 ) === '|+' ) {
+ # This might be cell elements, td, th or captions
+ if ( substr( $line , 0 , 2 ) === '|+' ) {
$first_character = '+';
- $line = substr ( $line , 1 );
+ $line = substr( $line , 1 );
}
- $line = substr ( $line , 1 );
+ $line = substr( $line , 1 );
if ( $first_character === '!' ) {
- $line = str_replace ( '!!' , '||' , $line );
+ $line = str_replace( '!!' , '||' , $line );
}
- // Split up multiple cells on the same line.
- // FIXME : This can result in improper nesting of tags processed
- // by earlier parser steps, but should avoid splitting up eg
- // attribute values containing literal "||".
+ # Split up multiple cells on the same line.
+ # FIXME : This can result in improper nesting of tags processed
+ # by earlier parser steps, but should avoid splitting up eg
+ # attribute values containing literal "||".
$cells = StringUtils::explodeMarkup( '||' , $line );
$outLine = '';
- // Loop through each table cell
- foreach ( $cells as $cell )
- {
+ # Loop through each table cell
+ foreach ( $cells as $cell ) {
$previous = '';
- if ( $first_character !== '+' )
- {
- $tr_after = array_pop ( $tr_attributes );
- if ( !array_pop ( $tr_history ) ) {
- $previous = "<tr{$tr_after}>\n";
+ if ( $first_character !== '+' ) {
+ $tr_after = array_pop( $tr_attributes );
+ if ( !array_pop( $tr_history ) ) {
+ $previous = "<$tr_tag{$tr_after}>\n";
}
- array_push ( $tr_history , true );
- array_push ( $tr_attributes , '' );
- array_pop ( $has_opened_tr );
- array_push ( $has_opened_tr , true );
+ array_push( $tr_history , true );
+ array_push( $tr_attributes , '' );
+ array_pop( $has_opened_tr );
+ array_push( $has_opened_tr , true );
}
- $last_tag = array_pop ( $last_tag_history );
+ $last_tag = array_pop( $last_tag_history );
- if ( array_pop ( $td_history ) ) {
- $previous = "</{$last_tag}>{$previous}";
+ if ( array_pop( $td_history ) ) {
+ $previous = "</{$last_tag}>\n{$previous}";
}
if ( $first_character === '|' ) {
- $last_tag = 'td';
+ $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 = '';
}
- array_push ( $last_tag_history , $last_tag );
+ array_push( $last_tag_history , $last_tag );
- // A cell could contain both parameters and data
- $cell_data = explode ( '|' , $cell , 2 );
+ # A cell could contain both parameters and data
+ $cell_data = explode( '|' , $cell , 2 );
- // Bug 553: Note that a '|' inside an invalid link should not
- // be mistaken as delimiting cell parameters
+ $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}";
- } elseif ( count ( $cell_data ) == 1 )
- $cell = "{$previous}<{$last_tag}>{$cell_data[0]}";
- else {
+ if ($extra_td_attribs) $attributes = Sanitizer::collapseTagAttributes( $extra_td_attribs );
+ $cell = "{$previous}<{$last_tag}{$attributes}>{$cell}";
+ } elseif ( count( $cell_data ) == 1 ) {
+ if ($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]}";
}
$outLine .= $cell;
- array_push ( $td_history , true );
+ array_push( $td_history , true );
}
}
$out .= $outLine . "\n";
}
- // Closing open td, tr && table
- while ( count ( $td_history ) > 0 )
- {
- if ( array_pop ( $td_history ) ) {
- $out .= "</td>\n";
+ # Closing open td, tr && table
+ while ( count( $td_history ) > 0 ) {
+ if ( array_pop( $td_history ) ) {
+ $out .= "</$td_tag>\n";
}
- if ( array_pop ( $tr_history ) ) {
- $out .= "</tr>\n";
+ if ( array_pop( $tr_history ) ) {
+ $out .= "</$tr_tag>\n";
}
- if ( !array_pop ( $has_opened_tr ) ) {
- $out .= "<tr><td></td></tr>\n" ;
+ if ( !array_pop( $has_opened_tr ) ) {
+ $out .= "<$tr_tag><$td_tag></$td_tag></$tr_tag>\n" ;
}
- $out .= "</table>\n";
+ $out .= "</$table_tag>\n";
}
- // Remove trailing line-ending (b/c)
+ # Remove trailing line-ending (b/c)
if ( substr( $out, -1 ) === "\n" ) {
$out = substr( $out, 0, -1 );
}
- // special case: don't return empty table
- if( $out === "<table>\n<tr><td></td></tr>\n</table>" ) {
+ # special case: don't return empty table
+ if ( $out === "<$table_tag>\n<$tr_tag><$td_tag></$td_tag></$tr_tag>\n</$table_tag>" ) {
$out = '';
}
return $text ;
}
- // if $frame is provided, then use $frame for replacing any variables
- if ($frame) {
- // use frame depth to infer how include/noinclude tags should be handled
- // depth=0 means this is the top-level document; otherwise it's an included document
- if( !$frame->depth )
+ # if $frame is provided, then use $frame for replacing any variables
+ if ( $frame ) {
+ # use frame depth to infer how include/noinclude tags should be handled
+ # depth=0 means this is the top-level document; otherwise it's an included document
+ if ( !$frame->depth ) {
$flag = 0;
- else
+ } else {
$flag = Parser::PTD_FOR_INCLUSION;
+ }
$dom = $this->preprocessToDom( $text, $flag );
$text = $frame->expand( $dom );
- }
- // if $frame is not provided, then use old-style replaceVariables
- else {
+ } else {
+ # if $frame is not provided, then use old-style replaceVariables
$text = $this->replaceVariables( $text );
}
$text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), false, array_keys( $this->mTransparentTagHooks ) );
wfRunHooks( 'InternalParseBeforeLinks', array( &$this, &$text, &$this->mStripState ) );
- // Tables need to come after variable replacement for things to work
- // properly; putting them before other transformations should keep
- // exciting things like link expansions from showing up in surprising
- // places.
+ # Tables need to come after variable replacement for things to work
+ # properly; putting them before other transformations should keep
+ # exciting things like link expansions from showing up in surprising
+ # places.
$text = $this->doTableStuff( $text );
$text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
$text = $this->doDoubleUnderscore( $text );
$text = $this->doHeadings( $text );
- if( $this->mOptions->getUseDynamicDates() ) {
+ if ( $this->mOptions->getUseDynamicDates() ) {
$df = DateFormatter::getInstance();
$text = $df->reformat( $this->mOptions->getDateFormat(), $text );
}
- $text = $this->doAllQuotes( $text );
$text = $this->replaceInternalLinks( $text );
+ $text = $this->doAllQuotes( $text );
$text = $this->replaceExternalLinks( $text );
# replaceInternalLinks may sometimes leave behind
# absolute URLs, which have to be masked to hide them from replaceExternalLinks
- $text = str_replace($this->mUniqPrefix.'NOPARSE', '', $text);
+ $text = str_replace( $this->mUniqPrefix.'NOPARSE', '', $text );
$text = $this->doMagicLinks( $text );
$text = $this->formatHeadings( $text, $origText, $isMain );
$urlChar = self::EXT_LINK_URL_CLASS;
$text = preg_replace_callback(
'!(?: # Start cases
- (<a.*?</a>) | # m[1]: Skip link text
+ (<a[ \t\r\n>].*?</a>) | # m[1]: Skip link text
(<.*?>) | # m[2]: Skip stuff inside HTML elements' . "
(\\b(?:$prots)$urlChar+) | # m[3]: Free external links" . '
(?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number
$id = $m[4];
} else {
throw new MWException( __METHOD__.': unrecognised match type "' .
- substr($m[0], 0, 20 ) . '"' );
+ substr( $m[0], 0, 20 ) . '"' );
}
$url = wfMsg( $urlmsg, $id);
$sk = $this->mOptions->getSkin();
# removeHTMLtags()) should not be included in
# URLs, per RFC 2396.
$m2 = array();
- if (preg_match('/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE)) {
- $trail = substr($url, $m2[0][1]) . $trail;
- $url = substr($url, 0, $m2[0][1]);
+ if ( preg_match( '/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE ) ) {
+ $trail = substr( $url, $m2[0][1] ) . $trail;
+ $url = substr( $url, 0, $m2[0][1] );
}
# Move trailing punctuation to $trail
foreach ( $lines as $line ) {
$outtext .= $this->doQuotes( $line ) . "\n";
}
- $outtext = substr($outtext, 0,-1);
+ $outtext = substr( $outtext, 0,-1 );
wfProfileOut( __METHOD__ );
return $outtext;
}
/**
- * Processes bolds and italics on a single line.
* Helper function for doAllQuotes()
*/
public function doQuotes( $text ) {
- # Counts the number of occurrences of bold and italics mark-ups.
- self::countBoldAndItalic($text, $numbold, $numitalics);
-
- if ( ( $numbold == 0 ) && ( $numitalics == 0 ) )
+ $arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE );
+ if ( count( $arr ) == 1 ) {
return $text;
- else
- {
+ } else {
+ # First, do some preliminary work. This may shift some apostrophes from
+ # being mark-up to being text. It also counts the number of occurrences
+ # of bold and italics mark-ups.
+ $i = 0;
+ $numbold = 0;
+ $numitalics = 0;
+ foreach ( $arr as $r ) {
+ if ( ( $i % 2 ) == 1 ) {
+ # If there are ever four apostrophes, assume the first is supposed to
+ # be text, and the remaining three constitute mark-up for bold text.
+ if ( strlen( $arr[$i] ) == 4 ) {
+ $arr[$i-1] .= "'";
+ $arr[$i] = "'''";
+ } elseif ( strlen( $arr[$i] ) > 5 ) {
+ # If there are more than 5 apostrophes in a row, assume they're all
+ # text except for the last 5.
+ $arr[$i-1] .= str_repeat( "'", strlen( $arr[$i] ) - 5 );
+ $arr[$i] = "'''''";
+ }
+ # Count the number of occurrences of bold and italics mark-ups.
+ # We are not counting sequences of five apostrophes.
+ if ( strlen( $arr[$i] ) == 2 ) {
+ $numitalics++;
+ } elseif ( strlen( $arr[$i] ) == 3 ) {
+ $numbold++;
+ } elseif ( strlen( $arr[$i] ) == 5 ) {
+ $numitalics++;
+ $numbold++;
+ }
+ }
+ $i++;
+ }
+
# If there is an odd number of both bold and italics, it is likely
# that one of the bold ones was meant to be an apostrophe followed
# by italics. Which one we cannot know for certain, but it is more
# likely to be one that has a single-letter word before it.
- if ( ( $numbold % 2 == 1 ) && ( $numitalics % 2 == 1 ) )
- {
-
- # This algorithm moves the literal quote at the
- # right of a single word, at the right of a
- # multiletter word or at the right of a space.
- # Otherwise, it does nothing.
- #
- # The original if-based version can be found at
- # http://svn.wikimedia.org/viewvc/mediawiki/trunk/phase3/includes/parser/Parser.php?revision=61519&view=markup
- #
- # Unlike the original one, here we convert the
- # texty quotes to ' which shouldn't matter.
-
- $quoteBalancerReplacements = array(
- "/(?<= [^ ])'''(?!')/"=>"'''",
- "/(?<=[^ '])'''(?!')/"=>"'''",
- "/(^|(?<=[^'])) '''(?!')/"=>" '''");
+ if ( ( $numbold % 2 == 1 ) && ( $numitalics % 2 == 1 ) ) {
+ $i = 0;
+ $firstsingleletterword = -1;
+ $firstmultiletterword = -1;
+ $firstspace = -1;
+ foreach ( $arr as $r ) {
+ if ( ( $i % 2 == 1 ) and ( strlen( $r ) == 3 ) ) {
+ $x1 = substr( $arr[$i-1], -1 );
+ $x2 = substr( $arr[$i-1], -2, 1 );
+ if ( $x1 === ' ' ) {
+ if ( $firstspace == -1 ) {
+ $firstspace = $i;
+ }
+ } elseif ( $x2 === ' ') {
+ if ( $firstsingleletterword == -1 ) {
+ $firstsingleletterword = $i;
+ }
+ } else {
+ if ( $firstmultiletterword == -1 ) {
+ $firstmultiletterword = $i;
+ }
+ }
+ }
+ $i++;
+ }
- foreach( $quoteBalancerReplacements as $k => $v) {
- $text = preg_replace($k, $v, $text, 1, $count);
- if ($count != 0)
- break;
+ # If there is a single-letter word, use it!
+ if ( $firstsingleletterword > -1 ) {
+ $arr[$firstsingleletterword] = "''";
+ $arr[$firstsingleletterword-1] .= "'";
+ } elseif ( $firstmultiletterword > -1 ) {
+ # If not, but there's a multi-letter word, use that one.
+ $arr[$firstmultiletterword] = "''";
+ $arr[$firstmultiletterword-1] .= "'";
+ } elseif ( $firstspace > -1 ) {
+ # ... otherwise use the first one that has neither.
+ # (notice that it is possible for all three to be -1 if, for example,
+ # there is only one pentuple-apostrophe in the line)
+ $arr[$firstspace] = "''";
+ $arr[$firstspace-1] .= "'";
}
}
- # Split in groups of 2, 3, 5 or 6 apostrophes.
- # If there are ever four apostrophes, assume the first is supposed to
- # be text, and the remaining three constitute mark-up for bold text.
- # If there are more than 6 apostrophes in a row, assume they're all
- # text except for the last 6.
- $arr = Stringutils::preg_split( "/('{2,3}(?:''')?)(?!')/", $text, -1, PREG_SPLIT_DELIM_CAPTURE );
-
-
# Now let's actually convert our apostrophic mush to HTML!
- $output = ''; # Processed text
- $buffer = ''; # Content if $state is 'both'
- $state = ''; # Flags with the order of open tags: '|b|i|bi|ib|both'
+ $output = '';
+ $buffer = '';
+ $state = '';
$i = 0;
- foreach ($arr as $r)
- {
- if (($i % 2) == 0)
- {
- if ($state === 'both')
+ foreach ( $arr as $r ) {
+ if ( ( $i % 2 ) == 0 ) {
+ if ( $state === 'both' ) {
$buffer .= $r;
- else
+ } else {
$output .= $r;
- }
- else
- {
- if (strlen ($r) == 2)
- {
- if ($state === 'i')
- { $output .= '</i>'; $state = ''; }
- elseif ($state === 'bi')
- { $output .= '</i>'; $state = 'b'; }
- elseif ($state === 'ib')
- { $output .= '</b></i><b>'; $state = 'b'; }
- elseif ($state === 'both')
- { $output .= '<b><i>'.$buffer.'</i>'; $state = 'b'; }
- else # $state can be 'b' or ''
- { $output .= '<i>'; $state .= 'i'; }
- }
- elseif (strlen ($r) == 3)
- {
- if ($state === 'b')
- { $output .= '</b>'; $state = ''; }
- elseif ($state === 'bi')
- { $output .= '</i></b><i>'; $state = 'i'; }
- elseif ($state === 'ib')
- { $output .= '</b>'; $state = 'i'; }
- elseif ($state === 'both')
- { $output .= '<i><b>'.$buffer.'</b>'; $state = 'i'; }
- else # $state can be 'i' or ''
- { $output .= '<b>'; $state .= 'b'; }
}
- elseif (strlen ($r) == 5)
- {
- if ($state === 'b')
- { $output .= '</b><i>'; $state = 'i'; }
- elseif ($state === 'i')
- { $output .= '</i><b>'; $state = 'b'; }
- elseif ($state === 'bi')
- { $output .= '</i></b>'; $state = ''; }
- elseif ($state === 'ib')
- { $output .= '</b></i>'; $state = ''; }
- elseif ($state === 'both')
- { $output .= '<i><b>'.$buffer.'</b></i>'; $state = ''; }
- else # ($state == '')
- { $buffer = ''; $state = 'both'; }
- }
- elseif (strlen ($r) == 6)
- {
- if ($state === 'b')
- { $output .= '</b><b>'; $state = 'b'; }
- elseif ($state === 'i')
- { $output .= '\'</i><b>'; $state = 'b'; }
- elseif ($state === 'bi')
- { $output .= '\'</i></b>'; $state = ''; }
- elseif ($state === 'ib')
- { $output .= '\'</b></i>'; $state = ''; }
- elseif ($state === 'both')
- { $output .= '<i><b>'.$buffer.'</b><b>'; $state = 'ib'; }
- else # ($state == '')
- { $buffer = ''; $state = ''; }
+ } else {
+ if ( strlen( $r ) == 2 ) {
+ if ( $state === 'i' ) {
+ $output .= '</i>'; $state = '';
+ } elseif ( $state === 'bi' ) {
+ $output .= '</i>'; $state = 'b';
+ } elseif ( $state === 'ib' ) {
+ $output .= '</b></i><b>'; $state = 'b';
+ } elseif ( $state === 'both' ) {
+ $output .= '<b><i>'.$buffer.'</i>'; $state = 'b';
+ } else { # $state can be 'b' or ''
+ $output .= '<i>'; $state .= 'i';
+ }
+ } elseif ( strlen( $r ) == 3 ) {
+ if ( $state === 'b' ) {
+ $output .= '</b>'; $state = '';
+ } elseif ( $state === 'bi' ) {
+ $output .= '</i></b><i>'; $state = 'i';
+ } elseif ( $state === 'ib' ) {
+ $output .= '</b>'; $state = 'i';
+ } elseif ( $state === 'both' ) {
+ $output .= '<i><b>'.$buffer.'</b>'; $state = 'i';
+ } else { # $state can be 'i' or ''
+ $output .= '<b>'; $state .= 'b';
+ }
+ } elseif ( strlen( $r ) == 5 ) {
+ if ( $state === 'b' ) {
+ $output .= '</b><i>'; $state = 'i';
+ } elseif ( $state === 'i' ) {
+ $output .= '</i><b>'; $state = 'b';
+ } elseif ( $state === 'bi' ) {
+ $output .= '</i></b>'; $state = '';
+ } elseif ( $state === 'ib' ) {
+ $output .= '</b></i>'; $state = '';
+ } elseif ( $state === 'both' ) {
+ $output .= '<i><b>'.$buffer.'</b></i>'; $state = '';
+ } else { # ($state == '')
+ $buffer = ''; $state = 'both';
+ }
}
}
$i++;
}
# Now close all remaining tags. Notice that the order is important.
- if ($state === 'b' || $state === 'ib')
+ if ( $state === 'b' || $state === 'ib' ) {
$output .= '</b>';
- if ($state === 'i' || $state === 'bi' || $state === 'ib')
+ }
+ if ( $state === 'i' || $state === 'bi' || $state === 'ib' ) {
$output .= '</i>';
- if ($state === 'bi')
+ }
+ if ( $state === 'bi' ) {
$output .= '</b>';
+ }
# There might be lonely ''''', so make sure we have a buffer
- if ($state === 'both' && $buffer)
+ if ( $state === 'both' && $buffer ) {
$output .= '<b><i>'.$buffer.'</i></b>';
+ }
return $output;
}
}
- /**
- * Counts the number of bold and italic items from a line of text.
- * Helper function for doQuotes()
- */
- private static function countBoldAndItalic($text, &$numBold, &$numItalics) {
- $numBold = 0;
- $numItalics = 0;
- $offset = 0;
-
- do {
- $offset = strpos($text, "'", $offset);
- if ($offset === false)
- return;
-
- $quoteLen = strspn($text, "'", $offset);
- $offset += $quoteLen;
-
- switch ($quoteLen) {
- case 0:
- case 1:
- break;
-
- case 2:
- $numItalics++;
- break;
-
- case 3:
- $numBold++;
- break;
-
- case 4:
- # If there are ever four apostrophes, assume the first is supposed to
- # be text, and the remaining three constitute mark-up for bold text.
- $numBold++;
- $numItalics++;
- break;
-
- case 5:
- $numItalics++;
- $numBold++;
- break;
-
- case 6:
- default:
- # If there are more than 6 apostrophes in a row, assume they're all
- # text except for the last 6.
- $numBold+=2;
- }
- } while (true);
- }
-
/**
* Replace external links (REL)
*
# removeHTMLtags()) should not be included in
# URLs, per RFC 2396.
$m2 = array();
- if (preg_match('/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE)) {
- $text = substr($url, $m2[0][1]) . ' ' . $text;
- $url = substr($url, 0, $m2[0][1]);
+ if ( preg_match( '/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE ) ) {
+ $text = substr( $url, $m2[0][1] ) . ' ' . $text;
+ $url = substr( $url, 0, $m2[0][1] );
}
# If the link text is an image URL, replace it with an <img> tag
$dtrail = '';
# Set linktype for CSS - if URL==text, link is essentially free
- $linktype = ($text === $url) ? 'free' : 'text';
+ $linktype = ( $text === $url ) ? 'free' : 'text';
# No link text, e.g. [http://domain.tld/some.link]
if ( $text == '' ) {
# Autonumber if allowed. See bug #5918
- if ( strpos( wfUrlProtocols(), substr($protocol, 0, strpos($protocol, ':')) ) !== false ) {
+ if ( strpos( wfUrlProtocols(), substr( $protocol, 0, strpos( $protocol, ':' ) ) ) !== false ) {
$langObj = $this->getFunctionLang();
$text = '[' . $langObj->formatNum( ++$this->mAutonumber ) . ']';
$linktype = 'autonumber';
list( $dtrail, $trail ) = Linker::splitTrail( $trail );
}
- $text = $wgContLang->markNoConversion($text);
+ $text = $wgContLang->markNoConversion( $text );
$url = Sanitizer::cleanUrl( $url );
# Use the encoded URL
# This means that users can paste URLs directly into the text
- # Funny characters like ö aren't valid in URLs anyway
+ # Funny characters like ö aren't valid in URLs anyway
# This was changed in August 2004
$s .= $sk->makeExternalLink( $url, $text, false, $linktype,
$this->getExternalLinkAttribs( $url ) ) . $dtrail . $trail;
* (depending on configuration, namespace, and the URL's domain) and/or a
* target attribute (depending on configuration).
*
- * @param string $url Optional URL, to extract the domain from for rel =>
+ * @param $url String: optional URL, to extract the domain from for rel =>
* nofollow if appropriate
- * @return array Associative array of HTML attributes
+ * @return Array: associative array of HTML attributes
*/
function getExternalLinkAttribs( $url = false ) {
$attribs = array();
global $wgNoFollowLinks, $wgNoFollowNsExceptions;
$ns = $this->mTitle->getNamespace();
- if( $wgNoFollowLinks && !in_array($ns, $wgNoFollowNsExceptions) ) {
+ if ( $wgNoFollowLinks && !in_array( $ns, $wgNoFollowNsExceptions ) ) {
$attribs['rel'] = 'nofollow';
global $wgNoFollowDomainExceptions;
$bits = wfParseUrl( $url );
if ( is_array( $bits ) && isset( $bits['host'] ) ) {
foreach ( $wgNoFollowDomainExceptions as $domain ) {
- if( substr( $bits['host'], -strlen( $domain ) )
- == $domain ) {
+ if ( substr( $bits['host'], -strlen( $domain ) ) == $domain ) {
unset( $attribs['rel'] );
break;
}
/**
* 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
/**
* Callback function used in replaceUnusualEscapes().
* Replaces unusual URL escape codes with their equivalent character
- * @static
- * @private
*/
private static function replaceUnusualEscapesCallback( $matches ) {
$char = urldecode( $matches[0] );
$ord = ord( $char );
- // Is it an unsafe or HTTP reserved character according to RFC 1738?
+ # Is it an unsafe or HTTP reserved character according to RFC 1738?
if ( $ord > 32 && $ord < 127 && strpos( '<>"#{}|\^~[]`;/?', $char ) === false ) {
- // No, shouldn't be escaped
+ # No, shouldn't be escaped
return $char;
} else {
- // Yes, leave it escaped
+ # Yes, leave it escaped
return $matches[0];
}
}
function maybeMakeExternalImage( $url ) {
$sk = $this->mOptions->getSkin();
$imagesfrom = $this->mOptions->getAllowExternalImagesFrom();
- $imagesexception = !empty($imagesfrom);
+ $imagesexception = !empty( $imagesfrom );
$text = false;
# $imagesfrom could be either a single string or an array of strings, parse out the latter
- if( $imagesexception && is_array( $imagesfrom ) ) {
+ if ( $imagesexception && is_array( $imagesfrom ) ) {
$imagematch = false;
- foreach( $imagesfrom as $match ) {
- if( strpos( $url, $match ) === 0 ) {
+ foreach ( $imagesfrom as $match ) {
+ if ( strpos( $url, $match ) === 0 ) {
$imagematch = true;
break;
}
}
- } elseif( $imagesexception ) {
- $imagematch = (strpos( $url, $imagesfrom ) === 0);
+ } elseif ( $imagesexception ) {
+ $imagematch = ( strpos( $url, $imagesfrom ) === 0 );
} else {
$imagematch = false;
}
$text = $sk->makeExternalImage( $url );
}
}
- if( !$text && $this->mOptions->getEnableImageWhitelist()
+ if ( !$text && $this->mOptions->getEnableImageWhitelist()
&& preg_match( self::EXT_IMAGE_REGEX, $url ) ) {
$whitelist = explode( "\n", wfMsgForContent( 'external_image_whitelist' ) );
- foreach( $whitelist as $entry ) {
+ foreach ( $whitelist as $entry ) {
# Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
- if( strpos( $entry, '#' ) === 0 || $entry === '' )
+ if ( strpos( $entry, '#' ) === 0 || $entry === '' ) {
continue;
- if( preg_match( '/' . str_replace( '/', '\\/', $entry ) . '/i', $url ) ) {
+ }
+ if ( preg_match( '/' . str_replace( '/', '\\/', $entry ) . '/i', $url ) ) {
# Image matches a whitelist entry
$text = $sk->makeExternalImage( $url );
break;
$sk = $this->mOptions->getSkin();
$holders = new LinkHolderArray( $this );
- # split the entire text string on occurences of [[
+ # split the entire text string on occurences of [[
$a = StringUtils::explode( '[[', ' ' . $s );
# get the first element (all text up to first [[), and remove the space we added
$s = $a->current();
$e2 = wfMsgForContent( 'linkprefix' );
}
- if( is_null( $this->mTitle ) ) {
+ if ( is_null( $this->mTitle ) ) {
wfProfileOut( __METHOD__.'-setup' );
wfProfileOut( __METHOD__ );
throw new MWException( __METHOD__.": \$this->mTitle is null\n" );
$prefix = '';
}
- if($wgContLang->hasVariants()) {
- $selflink = $wgContLang->convertLinkToAllVariants($this->mTitle->getPrefixedText());
+ if ( $wgContLang->hasVariants() ) {
+ $selflink = $wgContLang->convertLinkToAllVariants( $this->mTitle->getPrefixedText() );
} else {
- $selflink = array($this->mTitle->getPrefixedText());
+ $selflink = array( $this->mTitle->getPrefixedText() );
}
$useSubpages = $this->areSubpagesAllowed();
wfProfileOut( __METHOD__.'-setup' );
$prefix='';
}
# first link
- if($first_prefix) {
+ if ( $first_prefix ) {
$prefix = $first_prefix;
$first_prefix = false;
}
# Still some problems for cases where the ] is meant to be outside punctuation,
# and no image is in sight. See bug 2095.
#
- if( $text !== '' &&
+ if ( $text !== '' &&
substr( $m[3], 0, 1 ) === ']' &&
- strpos($text, '[') !== false
+ strpos( $text, '[' ) !== false
)
{
$text .= ']'; # so that replaceExternalLinks($text) works later
$m[3] = substr( $m[3], 1 );
}
# fix up urlencoded title texts
- if( strpos( $m[1], '%' ) !== false ) {
+ if ( strpos( $m[1], '%' ) !== false ) {
# Should anchors '#' also be rejected?
- $m[1] = str_replace( array('<', '>'), array('<', '>'), urldecode($m[1]) );
+ $m[1] = str_replace( array('<', '>'), array('<', '>'), urldecode( $m[1] ) );
}
$trail = $m[3];
- } elseif( preg_match($e1_img, $line, $m) ) { # Invalid, but might be an image with a link in its caption
+ } elseif ( preg_match( $e1_img, $line, $m ) ) { # Invalid, but might be an image with a link in its caption
$might_be_img = true;
$text = $m[2];
if ( strpos( $m[1], '%' ) !== false ) {
- $m[1] = urldecode($m[1]);
+ $m[1] = urldecode( $m[1] );
}
$trail = "";
} else { # Invalid form; output directly
$link = $m[1];
}
- $noforce = (substr( $m[1], 0, 1 ) !== ':');
- if (!$noforce) {
+ $noforce = ( substr( $m[1], 0, 1 ) !== ':' );
+ if ( !$noforce ) {
# Strip off leading ':'
$link = substr( $link, 1 );
}
}
$wasblank = ( $text == '' );
- if ( $wasblank ) $text = $link;
+ if ( $wasblank ) {
+ $text = $link;
+ } else {
+ # Bug 4598 madness. Handle the quotes only if they come from the alternate part
+ # [[Lista d''e paise d''o munno]] -> <a href="">Lista d''e paise d''o munno</a>
+ # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']] -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
+ $text = $this->doQuotes($text);
+ }
# Link not escaped by : , create the various objects
if ( $noforce ) {
wfProfileIn( __METHOD__."-interwiki" );
if ( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) {
$this->mOutput->addLanguageLink( $nt->getFullText() );
- $s = rtrim($s . $prefix);
- $s .= trim($trail, "\n") == '' ? '': $prefix . $trail;
+ $s = rtrim( $s . $prefix );
+ $s .= trim( $trail, "\n" ) == '' ? '': $prefix . $trail;
wfProfileOut( __METHOD__."-interwiki" );
continue;
}
# recursively parse links inside the image caption
# actually, this will parse them in any other parameters, too,
# but it might be hard to fix that, and it doesn't matter ATM
- $text = $this->replaceExternalLinks($text);
+ $text = $this->replaceExternalLinks( $text );
$holders->merge( $this->replaceInternalLinks2( $text ) );
}
# cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
$s .= $prefix . $this->armorLinks( $this->makeImage( $nt, $text, $holders ) ) . $trail;
+ } else {
+ $s .= $prefix . $trail;
}
$this->mOutput->addImage( $nt->getDBkey() );
wfProfileOut( __METHOD__."-image" );
if ( $ns == NS_CATEGORY ) {
wfProfileIn( __METHOD__."-category" );
- $s = rtrim($s . "\n"); # bug 87
+ $s = rtrim( $s . "\n" ); # bug 87
if ( $wasblank ) {
$sortkey = $this->getDefaultSort();
* Strip the whitespace Category links produce, see bug 87
* @todo We might want to use trim($tmp, "\n") here.
*/
- $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail;
+ $s .= trim( $prefix . $trail, "\n" ) == '' ? '': $prefix . $trail;
wfProfileOut( __METHOD__."-category" );
continue;
}
# Self-link checking
- if( $nt->getFragment() === '' && $ns !== NS_SPECIAL ) {
- if( in_array( $nt->getPrefixedText(), $selflink, true ) ) {
+ if ( $nt->getFragment() === '' && $ns != NS_SPECIAL ) {
+ if ( in_array( $nt->getPrefixedText(), $selflink, true ) ) {
$s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail );
continue;
}
# NS_MEDIA is a pseudo-namespace for linking directly to a file
# FIXME: Should do batch file existence checks, see comment below
- if( $ns == NS_MEDIA ) {
+ if ( $ns == NS_MEDIA ) {
wfProfileIn( __METHOD__."-media" );
# Give extensions a chance to select the file revision for us
$skip = $time = false;
#
# FIXME: isAlwaysKnown() can be expensive for file links; we should really do
# batch file existence checks for NS_FILE and NS_MEDIA
- if( $iw == '' && $nt->isAlwaysKnown() ) {
+ if ( $iw == '' && $nt->isAlwaysKnown() ) {
$this->mOutput->addLink( $nt );
$s .= $this->makeKnownLinkHolder( $nt, $text, '', $trail, $prefix );
} else {
* breaking URLs in the following text without breaking trails on the
* wiki links, it's been made into a horrible function.
*
- * @param Title $nt
- * @param string $text
- * @param string $query
- * @param string $trail
- * @param string $prefix
- * @return string HTML-wikitext mix oh yuck
+ * @param $nt Title
+ * @param $text String
+ * @param $query String
+ * @param $trail String
+ * @param $prefix String
+ * @return String: HTML-wikitext mix oh yuck
*/
function makeKnownLinkHolder( $nt, $text = '', $query = '', $trail = '', $prefix = '' ) {
list( $inside, $trail ) = Linker::splitTrail( $trail );
$sk = $this->mOptions->getSkin();
- // FIXME: use link() instead of deprecated makeKnownLinkObj()
+ # FIXME: use link() instead of deprecated makeKnownLinkObj()
$link = $sk->makeKnownLinkObj( $nt, $text, $query, $inside, $prefix );
return $this->armorLinks( $link ) . $trail;
}
* 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() . ')/',
/**
* Return true if subpage links should be expanded on this page.
- * @return bool
+ * @return Boolean
*/
function areSubpagesAllowed() {
# Some namespaces don't allow subpages
/**
* Handle link to subpage if necessary
- * @param string $target the source of the link
- * @param string &$text the link text, modified as necessary
+ *
+ * @param $target String: the source of the link
+ * @param &$text String: the link text, modified as necessary
* @return string the full name of the link
* @private
*/
- function maybeDoSubpageLink($target, &$text) {
+ function maybeDoSubpageLink( $target, &$text ) {
return Linker::normalizeSubpageLink( $this->mTitle, $target, $text );
}
* Used by doBlockLevels()
* @private
*/
- /* private */ function closeParagraph() {
+ function closeParagraph() {
$result = '';
- if ( $this->mLastSection !== '' ) {
+ if ( $this->mLastSection != '' ) {
$result = '</' . $this->mLastSection . ">\n";
}
$this->mInPre = false;
$this->mLastSection = '';
return $result;
}
- # getCommon() returns the length of the longest common substring
- # of both arguments, starting at the beginning of both.
- #
- /* private */ function getCommon( $st1, $st2 ) {
+
+ /**
+ * getCommon() returns the length of the longest common substring
+ * of both arguments, starting at the beginning of both.
+ * @private
+ */
+ function getCommon( $st1, $st2 ) {
$fl = strlen( $st1 );
$shorter = strlen( $st2 );
- if ( $fl < $shorter ) { $shorter = $fl; }
+ if ( $fl < $shorter ) {
+ $shorter = $fl;
+ }
for ( $i = 0; $i < $shorter; ++$i ) {
- if ( $st1{$i} !== $st2{$i} ) { break; }
+ if ( $st1{$i} != $st2{$i} ) {
+ break;
+ }
}
return $i;
}
- # These next three functions open, continue, and close the list
- # element appropriate to the prefix character passed into them.
- #
- /* private */ function openList( $char ) {
+
+ /**
+ * These next three functions open, continue, and close the list
+ * element appropriate to the prefix character passed into them.
+ * @private
+ */
+ function openList( $char ) {
$result = $this->closeParagraph();
- if ( '*' === $char ) { $result .= '<ul><li>'; }
- elseif ( '#' === $char ) { $result .= '<ol><li>'; }
- elseif ( ':' === $char ) { $result .= '<dl><dd>'; }
- elseif ( ';' === $char ) {
+ if ( '*' === $char ) {
+ $result .= '<ul><li>';
+ } elseif ( '#' === $char ) {
+ $result .= '<ol><li>';
+ } elseif ( ':' === $char ) {
+ $result .= '<dl><dd>';
+ } elseif ( ';' === $char ) {
$result .= '<dl><dt>';
$this->mDTopen = true;
+ } else {
+ $result = '<!-- ERR 1 -->';
}
- else { $result = '<!-- ERR 1 -->'; }
return $result;
}
- /* private */ function nextItem( $char ) {
- if ( '*' === $char || '#' === $char ) { return '</li><li>'; }
- elseif ( ':' === $char || ';' === $char ) {
+ /**
+ * TODO: document
+ * @param $char String
+ * @private
+ */
+ function nextItem( $char ) {
+ if ( '*' === $char || '#' === $char ) {
+ return '</li><li>';
+ } elseif ( ':' === $char || ';' === $char ) {
$close = '</dd>';
- if ( $this->mDTopen ) { $close = '</dt>'; }
+ if ( $this->mDTopen ) {
+ $close = '</dt>';
+ }
if ( ';' === $char ) {
$this->mDTopen = true;
return $close . '<dt>';
return '<!-- ERR 2 -->';
}
- /* private */ function closeList( $char ) {
- if ( '*' === $char ) { $text = '</li></ul>'; }
- elseif ( '#' === $char ) { $text = '</li></ol>'; }
- elseif ( ':' === $char ) {
+ /**
+ * TODO: document
+ * @param $char String
+ * @private
+ */
+ function closeList( $char ) {
+ if ( '*' === $char ) {
+ $text = '</li></ul>';
+ } elseif ( '#' === $char ) {
+ $text = '</li></ol>';
+ } elseif ( ':' === $char ) {
if ( $this->mDTopen ) {
$this->mDTopen = false;
$text = '</dt></dl>';
} else {
$text = '</dd></dl>';
}
+ } else {
+ return '<!-- ERR 3 -->';
}
- else { return '<!-- ERR 3 -->'; }
return $text."\n";
}
/**#@-*/
/**
* 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
*/
$linestart = true;
continue;
}
- // * = ul
- // # = ol
- // ; = dt
- // : = dd
+ # * = ul
+ # # = ol
+ # ; = dt
+ # : = dd
$lastPrefixLength = strlen( $lastPrefix );
- $preCloseMatch = preg_match('/<\\/pre/i', $oLine );
- $preOpenMatch = preg_match('/<pre/i', $oLine );
- // If not in a <pre> element, scan for and figure out what prefixes are there.
+ $preCloseMatch = preg_match( '/<\\/pre/i', $oLine );
+ $preOpenMatch = preg_match( '/<pre/i', $oLine );
+ # If not in a <pre> element, scan for and figure out what prefixes are there.
if ( !$this->mInPre ) {
# Multiple prefixes may abut each other for nested lists.
$prefixLength = strspn( $oLine, '*#:;' );
$prefix = substr( $oLine, 0, $prefixLength );
# eh?
- // ; and : are both from definition-lists, so they're equivalent
- // for the purposes of determining whether or not we need to open/close
- // elements.
+ # ; and : are both from definition-lists, so they're equivalent
+ # for the purposes of determining whether or not we need to open/close
+ # elements.
$prefix2 = str_replace( ';', ':', $prefix );
$t = substr( $oLine, $prefixLength );
$this->mInPre = (bool)$preOpenMatch;
}
# List generation
- if( $prefixLength && $lastPrefix === $prefix2 ) {
+ if ( $prefixLength && $lastPrefix === $prefix2 ) {
# Same as the last item, so no need to deal with nesting or opening stuff
$output .= $this->nextItem( substr( $prefix, -1 ) );
$paragraphStack = false;
# So we check for : in the remainder text to split up the
# title and definition, without b0rking links.
$term = $t2 = '';
- if ($this->findColonNoLinks($t, $term, $t2) !== false) {
+ if ( $this->findColonNoLinks( $t, $term, $t2 ) !== false ) {
$t = $t2;
$output .= $term . $this->nextItem( ':' );
}
}
- } elseif( $prefixLength || $lastPrefixLength ) {
- // We need to open or close prefixes, or both.
+ } elseif ( $prefixLength || $lastPrefixLength ) {
+ # We need to open or close prefixes, or both.
# Either open or close a level...
$commonPrefixLength = $this->getCommon( $prefix, $lastPrefix );
$paragraphStack = false;
- // Close all the prefixes which aren't shared.
- while( $commonPrefixLength < $lastPrefixLength ) {
+ # Close all the prefixes which aren't shared.
+ while ( $commonPrefixLength < $lastPrefixLength ) {
$output .= $this->closeList( $lastPrefix[$lastPrefixLength-1] );
--$lastPrefixLength;
}
- // Continue the current prefix if appropriate.
+ # Continue the current prefix if appropriate.
if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) {
$output .= $this->nextItem( $prefix[$commonPrefixLength-1] );
}
- // Open prefixes where appropriate.
+ # Open prefixes where appropriate.
while ( $prefixLength > $commonPrefixLength ) {
$char = substr( $prefix, $commonPrefixLength, 1 );
$output .= $this->openList( $char );
if ( ';' === $char ) {
# FIXME: This is dupe of code above
- if ($this->findColonNoLinks($t, $term, $t2) !== false) {
+ if ( $this->findColonNoLinks( $t, $term, $t2 ) !== false ) {
$t = $t2;
$output .= $term . $this->nextItem( ':' );
}
$lastPrefix = $prefix2;
}
- // If we have no prefixes, go to paragraph mode.
- if( 0 == $prefixLength ) {
+ # If we have no prefixes, go to paragraph mode.
+ if ( 0 == $prefixLength ) {
wfProfileIn( __METHOD__."-paragraph" );
# No prefix (not in list)--go to paragraph mode
- // XXX: use a stack for nestable elements like span, table and div
+ # XXX: use a stack for nestable elements like span, table and div
$openmatch = preg_match('/(?:<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
$closematch = preg_match(
'/(?:<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|'.
'<td|<th|<\\/?div|<hr|<\\/pre|<\\/p|'.$this->mUniqPrefix.'-pre|<\\/li|<\\/ul|<\\/ol|<\\/?center)/iS', $t );
if ( $openmatch or $closematch ) {
$paragraphStack = false;
- # TODO bug 5718: paragraph closed
+ #Â TODO bug 5718: paragraph closed
$output .= $this->closeParagraph();
if ( $preOpenMatch and !$preCloseMatch ) {
$this->mInPre = true;
$inBlockElem = true;
}
} elseif ( !$inBlockElem && !$this->mInPre ) {
- if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' or trim($t) !== '' ) ) {
- // pre
- if ($this->mLastSection !== 'pre') {
+ if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' || trim( $t ) != '' ) ) {
+ # pre
+ if ( $this->mLastSection !== 'pre' ) {
$paragraphStack = false;
$output .= $this->closeParagraph().'<pre>';
$this->mLastSection = 'pre';
}
$t = substr( $t, 1 );
} else {
- // paragraph
- if ( trim($t) == '' ) {
+ # paragraph
+ if ( trim( $t ) === '' ) {
if ( $paragraphStack ) {
$output .= $paragraphStack.'<br />';
$paragraphStack = false;
$this->mLastSection = 'p';
} else {
- if ($this->mLastSection !== 'p' ) {
+ if ( $this->mLastSection !== 'p' ) {
$output .= $this->closeParagraph();
$this->mLastSection = '';
$paragraphStack = '<p>';
$output .= $paragraphStack;
$paragraphStack = false;
$this->mLastSection = 'p';
- } elseif ($this->mLastSection !== 'p') {
+ } elseif ( $this->mLastSection !== 'p' ) {
$output .= $this->closeParagraph().'<p>';
$this->mLastSection = 'p';
}
}
wfProfileOut( __METHOD__."-paragraph" );
}
- // somewhere above we forget to get out of pre block (bug 785)
- if($preCloseMatch && $this->mInPre) {
+ # somewhere above we forget to get out of pre block (bug 785)
+ if ( $preCloseMatch && $this->mInPre ) {
$this->mInPre = false;
}
- if ($paragraphStack === false) {
+ if ( $paragraphStack === false ) {
$output .= $t."\n";
}
}
$output .= $this->closeList( $prefix2[$prefixLength-1] );
--$prefixLength;
}
- if ( $this->mLastSection !== '' ) {
+ if ( $this->mLastSection != '' ) {
$output .= '</' . $this->mLastSection . '>';
$this->mLastSection = '';
}
/**
* Split up a string on ':', ignoring any occurences inside tags
* to prevent illegal overlapping.
- * @param string $str the string to split
- * @param string &$before set to everything before the ':'
- * @param string &$after set to everything after the ':'
- * return string the position of the ':', or false if none found
+ *
+ * @param $str String: the string to split
+ * @param &$before String: set to everything before the ':'
+ * @param &$after String: set to everything after the ':'
+ * return String: the position of the ':', or false if none found
*/
- function findColonNoLinks($str, &$before, &$after) {
+ function findColonNoLinks( $str, &$before, &$after ) {
wfProfileIn( __METHOD__ );
$pos = strpos( $str, ':' );
- if( $pos === false ) {
- // Nothing to find!
+ if ( $pos === false ) {
+ # Nothing to find!
wfProfileOut( __METHOD__ );
return false;
}
$lt = strpos( $str, '<' );
- if( $lt === false || $lt > $pos ) {
- // Easy; no tag nesting to worry about
+ if ( $lt === false || $lt > $pos ) {
+ # Easy; no tag nesting to worry about
$before = substr( $str, 0, $pos );
$after = substr( $str, $pos+1 );
wfProfileOut( __METHOD__ );
return $pos;
}
- // Ugly state machine to walk through avoiding tags.
+ # Ugly state machine to walk through avoiding tags.
$state = self::COLON_STATE_TEXT;
$stack = 0;
$len = strlen( $str );
$c = $str{$i};
switch( $state ) {
- // (Using the number is a performance hack for common cases)
- case 0: // self::COLON_STATE_TEXT:
+ # (Using the number is a performance hack for common cases)
+ case 0: # self::COLON_STATE_TEXT:
switch( $c ) {
case "<":
- // Could be either a <start> tag or an </end> tag
+ # Could be either a <start> tag or an </end> tag
$state = self::COLON_STATE_TAGSTART;
break;
case ":":
- if( $stack == 0 ) {
- // We found it!
+ if ( $stack == 0 ) {
+ # We found it!
$before = substr( $str, 0, $i );
$after = substr( $str, $i + 1 );
wfProfileOut( __METHOD__ );
return $i;
}
- // Embedded in a tag; don't break it.
+ # Embedded in a tag; don't break it.
break;
default:
- // Skip ahead looking for something interesting
+ # Skip ahead looking for something interesting
$colon = strpos( $str, ':', $i );
- if( $colon === false ) {
- // Nothing else interesting
+ if ( $colon === false ) {
+ # Nothing else interesting
wfProfileOut( __METHOD__ );
return false;
}
$lt = strpos( $str, '<', $i );
- if( $stack === 0 ) {
- if( $lt === false || $colon < $lt ) {
- // We found it!
+ if ( $stack === 0 ) {
+ if ( $lt === false || $colon < $lt ) {
+ # We found it!
$before = substr( $str, 0, $colon );
$after = substr( $str, $colon + 1 );
wfProfileOut( __METHOD__ );
return $i;
}
}
- if( $lt === false ) {
- // Nothing else interesting to find; abort!
- // We're nested, but there's no close tags left. Abort!
+ if ( $lt === false ) {
+ # Nothing else interesting to find; abort!
+ # We're nested, but there's no close tags left. Abort!
break 2;
}
- // Skip ahead to next tag start
+ # Skip ahead to next tag start
$i = $lt;
$state = self::COLON_STATE_TAGSTART;
}
break;
- case 1: // self::COLON_STATE_TAG:
- // In a <tag>
+ case 1: # self::COLON_STATE_TAG:
+ # In a <tag>
switch( $c ) {
case ">":
$stack++;
$state = self::COLON_STATE_TEXT;
break;
case "/":
- // Slash may be followed by >?
+ # Slash may be followed by >?
$state = self::COLON_STATE_TAGSLASH;
break;
default:
- // ignore
+ # ignore
}
break;
- case 2: // self::COLON_STATE_TAGSTART:
+ case 2: # self::COLON_STATE_TAGSTART:
switch( $c ) {
case "/":
$state = self::COLON_STATE_CLOSETAG;
$state = self::COLON_STATE_COMMENT;
break;
case ">":
- // Illegal early close? This shouldn't happen D:
+ # Illegal early close? This shouldn't happen D:
$state = self::COLON_STATE_TEXT;
break;
default:
$state = self::COLON_STATE_TAG;
}
break;
- case 3: // self::COLON_STATE_CLOSETAG:
- // In a </tag>
- if( $c === ">" ) {
+ case 3: # self::COLON_STATE_CLOSETAG:
+ # In a </tag>
+ if ( $c === ">" ) {
$stack--;
- if( $stack < 0 ) {
+ if ( $stack < 0 ) {
wfDebug( __METHOD__.": Invalid input; too many close tags\n" );
wfProfileOut( __METHOD__ );
return false;
}
break;
case self::COLON_STATE_TAGSLASH:
- if( $c === ">" ) {
- // Yes, a self-closed tag <blah/>
+ if ( $c === ">" ) {
+ # Yes, a self-closed tag <blah/>
$state = self::COLON_STATE_TEXT;
} else {
- // Probably we're jumping the gun, and this is an attribute
+ # Probably we're jumping the gun, and this is an attribute
$state = self::COLON_STATE_TAG;
}
break;
- case 5: // self::COLON_STATE_COMMENT:
- if( $c === "-" ) {
+ case 5: # self::COLON_STATE_COMMENT:
+ if ( $c === "-" ) {
$state = self::COLON_STATE_COMMENTDASH;
}
break;
case self::COLON_STATE_COMMENTDASH:
- if( $c === "-" ) {
+ if ( $c === "-" ) {
$state = self::COLON_STATE_COMMENTDASHDASH;
} else {
$state = self::COLON_STATE_COMMENT;
}
break;
case self::COLON_STATE_COMMENTDASHDASH:
- if( $c === ">" ) {
+ if ( $c === ">" ) {
$state = self::COLON_STATE_TEXT;
} else {
$state = self::COLON_STATE_COMMENT;
throw new MWException( "State machine error in " . __METHOD__ );
}
}
- if( $stack > 0 ) {
+ if ( $stack > 0 ) {
wfDebug( __METHOD__.": Invalid input; not enough close tags (stack $stack, state $state)\n" );
return false;
}
*/
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
$value = wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) );
break;
case 'talkpagename':
- if( $this->mTitle->canTalk() ) {
+ if ( $this->mTitle->canTalk() ) {
$talkPage = $this->mTitle->getTalkPage();
$value = wfEscapeWikiText( $talkPage->getPrefixedText() );
} else {
}
break;
case 'talkpagenamee':
- if( $this->mTitle->canTalk() ) {
+ if ( $this->mTitle->canTalk() ) {
$talkPage = $this->mTitle->getTalkPage();
$value = $talkPage->getPrefixedUrl();
} else {
$value = $subjPage->getPrefixedUrl();
break;
case 'revisionid':
- // Let the edit saving system know we should parse the page
- // *after* a revision ID has been assigned.
+ # Let the edit saving system know we should parse the page
+ # *after* a revision ID has been assigned.
$this->mOutput->setFlag( 'vary-revision' );
wfDebug( __METHOD__ . ": {{REVISIONID}} used, setting vary-revision...\n" );
$value = $this->mRevisionId;
break;
case 'revisionday':
- // Let the edit saving system know we should parse the page
- // *after* a revision ID has been assigned. This is for null edits.
+ # Let the edit saving system know we should parse the page
+ # *after* a revision ID has been assigned. This is for null edits.
$this->mOutput->setFlag( 'vary-revision' );
wfDebug( __METHOD__ . ": {{REVISIONDAY}} used, setting vary-revision...\n" );
$value = intval( substr( $this->getRevisionTimestamp(), 6, 2 ) );
break;
case 'revisionday2':
- // Let the edit saving system know we should parse the page
- // *after* a revision ID has been assigned. This is for null edits.
+ # Let the edit saving system know we should parse the page
+ # *after* a revision ID has been assigned. This is for null edits.
$this->mOutput->setFlag( 'vary-revision' );
wfDebug( __METHOD__ . ": {{REVISIONDAY2}} used, setting vary-revision...\n" );
$value = substr( $this->getRevisionTimestamp(), 6, 2 );
break;
case 'revisionmonth':
- // Let the edit saving system know we should parse the page
- // *after* a revision ID has been assigned. This is for null edits.
+ # Let the edit saving system know we should parse the page
+ # *after* a revision ID has been assigned. This is for null edits.
$this->mOutput->setFlag( 'vary-revision' );
wfDebug( __METHOD__ . ": {{REVISIONMONTH}} used, setting vary-revision...\n" );
+ $value = substr( $this->getRevisionTimestamp(), 4, 2 );
+ break;
+ case 'revisionmonth1':
+ # Let the edit saving system know we should parse the page
+ # *after* a revision ID has been assigned. This is for null edits.
+ $this->mOutput->setFlag( 'vary-revision' );
+ wfDebug( __METHOD__ . ": {{REVISIONMONTH1}} used, setting vary-revision...\n" );
$value = intval( substr( $this->getRevisionTimestamp(), 4, 2 ) );
break;
case 'revisionyear':
- // Let the edit saving system know we should parse the page
- // *after* a revision ID has been assigned. This is for null edits.
+ # Let the edit saving system know we should parse the page
+ # *after* a revision ID has been assigned. This is for null edits.
$this->mOutput->setFlag( 'vary-revision' );
wfDebug( __METHOD__ . ": {{REVISIONYEAR}} used, setting vary-revision...\n" );
$value = substr( $this->getRevisionTimestamp(), 0, 4 );
break;
case 'revisiontimestamp':
- // Let the edit saving system know we should parse the page
- // *after* a revision ID has been assigned. This is for null edits.
+ # Let the edit saving system know we should parse the page
+ # *after* a revision ID has been assigned. This is for null edits.
$this->mOutput->setFlag( 'vary-revision' );
wfDebug( __METHOD__ . ": {{REVISIONTIMESTAMP}} used, setting vary-revision...\n" );
$value = $this->getRevisionTimestamp();
break;
case 'revisionuser':
- // Let the edit saving system know we should parse the page
- // *after* a revision ID has been assigned. This is for null edits.
+ # Let the edit saving system know we should parse the page
+ # *after* a revision ID has been assigned. This is for null edits.
$this->mOutput->setFlag( 'vary-revision' );
wfDebug( __METHOD__ . ": {{REVISIONUSER}} used, setting vary-revision...\n" );
$value = $this->getRevisionUser();
break;
case 'namespace':
- $value = str_replace('_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) );
+ $value = str_replace( '_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) );
break;
case 'namespacee':
$value = wfUrlencode( $wgContLang->getNsText( $this->mTitle->getNamespace() ) );
break;
case 'talkspace':
- $value = $this->mTitle->canTalk() ? str_replace('_',' ',$this->mTitle->getTalkNsText()) : '';
+ $value = $this->mTitle->canTalk() ? str_replace( '_',' ',$this->mTitle->getTalkNsText() ) : '';
break;
case 'talkspacee':
$value = $this->mTitle->canTalk() ? wfUrlencode( $this->mTitle->getTalkNsText() ) : '';
$value = $wgContLang->formatNum( gmdate( 'H', $ts ), true );
break;
case 'currentweek':
- // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
- // int to remove the padding
+ # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
+ # int to remove the padding
$value = $wgContLang->formatNum( (int)gmdate( 'W', $ts ) );
break;
case 'currentdow':
$value = $wgContLang->formatNum( $localHour, true );
break;
case 'localweek':
- // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
- // int to remove the padding
+ # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
+ # int to remove the padding
$value = $wgContLang->formatNum( (int)$localWeek );
break;
case 'localdow':
$value = $wgContLang->formatNum( SiteStats::pages() );
break;
case 'numberofadmins':
- $value = $wgContLang->formatNum( SiteStats::numberingroup('sysop') );
+ $value = $wgContLang->formatNum( SiteStats::numberingroup( 'sysop' ) );
break;
case 'numberofedits':
$value = $wgContLang->formatNum( SiteStats::edits() );
case 'currentversion':
$value = SpecialVersion::getVersion();
break;
+ case 'articlepath':
+ return $wgArticlePath;
case 'sitename':
return $wgSitename;
case 'server':
return $wgContLanguageCode;
default:
$ret = null;
- if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) )
+ if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) {
return $ret;
- else
+ } else {
return null;
+ }
}
if ( $index )
}
/**
- * 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__ );
}
* Preprocess some wikitext and return the document tree.
* This is the ghost of replace_variables().
*
- * @param string $text The text to parse
- * @param integer flags Bitwise combination of:
+ * @param $text String: The text to parse
+ * @param $flags Integer: bitwise combination of:
* self::PTD_FOR_INCLUSION Handle <noinclude>/<includeonly> as if the text is being
* included. Default is to assume a direct page view.
*
*
* @private
*/
- function preprocessToDom ( $text, $flags = 0 ) {
+ function preprocessToDom( $text, $flags = 0 ) {
$dom = $this->getPreprocessor()->preprocessToObj( $text, $flags );
return $dom;
}
- /*
+ /**
* Return a three-element array: leading whitespace, string contents, trailing whitespace
*/
public static function splitWhitespace( $s ) {
* 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 ) {
$frame = $this->getPreprocessor()->newFrame();
} elseif ( !( $frame instanceof PPFrame ) ) {
wfDebug( __METHOD__." called using plain parameters instead of a PPFrame instance. Creating custom frame.\n" );
- $frame = $this->getPreprocessor()->newCustomFrame($frame);
+ $frame = $this->getPreprocessor()->newCustomFrame( $frame );
}
$dom = $this->preprocessToDom( $text );
return $text;
}
- /// Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
+ # Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
static function createAssocArgs( $args ) {
$assocArgs = array();
$index = 1;
- foreach( $args as $arg ) {
+ foreach ( $args as $arg ) {
$eqpos = strpos( $arg, '=' );
if ( $eqpos === false ) {
$assocArgs[$index++] = $arg;
* Warn the user when a parser limitation is reached
* Will warn at most once the user per limitation type
*
- * @param string $limitationType, should be one of:
- * 'expensive-parserfunction' (corresponding messages: 'expensive-parserfunction-warning', 'expensive-parserfunction-category')
- * 'post-expand-template-argument' (corresponding messages: 'post-expand-template-argument-warning', 'post-expand-template-argument-category')
- * 'post-expand-template-inclusion' (corresponding messages: 'post-expand-template-inclusion-warning', 'post-expand-template-inclusion-category')
- * @params int $current, $max When an explicit limit has been
+ * @param $limitationType String: should be one of:
+ * 'expensive-parserfunction' (corresponding messages:
+ * 'expensive-parserfunction-warning',
+ * 'expensive-parserfunction-category')
+ * 'post-expand-template-argument' (corresponding messages:
+ * 'post-expand-template-argument-warning',
+ * 'post-expand-template-argument-category')
+ * 'post-expand-template-inclusion' (corresponding messages:
+ * 'post-expand-template-inclusion-warning',
+ * 'post-expand-template-inclusion-category')
+ * @param $current Current value
+ * @param $max Maximum allowed, when an explicit limit has been
* exceeded, provide the values (optional)
*/
function limitationWarn( $limitationType, $current=null, $max=null) {
- //does no harm if $current and $max are present but are unnecessary for the message
+ # does no harm if $current and $max are present but are unnecessary for the message
$warning = wfMsgExt( "$limitationType-warning", array( 'parsemag', 'escape' ), $current, $max );
$this->mOutput->addWarning( $warning );
$this->addTrackingCategory( "$limitationType-category" );
* 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 ) {
$originalTitle = $part1;
# $args is a list of argument nodes, starting from index 0, not including $part1
- $args = (null == $piece['parts']) ? array() : $piece['parts'];
+ $args = ( null == $piece['parts'] ) ? array() : $piece['parts'];
wfProfileOut( __METHOD__.'-setup' );
# SUBST
wfProfileIn( __METHOD__.'-modifiers' );
if ( !$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;
$id = $this->mVariables->matchStartToEnd( $part1 );
if ( $id !== false ) {
$text = $this->getVariableValue( $id, $frame );
- if (MagicWord::getCacheTTL($id)>-1)
- $this->mOutput->mContainsOldMagic = true;
+ if ( MagicWord::getCacheTTL( $id ) > -1 ) {
+ $this->mOutput->updateCacheExpiry( MagicWord::getCacheTTL( $id ) );
+ }
$found = true;
}
}
unset( $result[0] );
}
- // Extract flags into the local scope
- // This allows callers to set flags such as nowiki, found, etc.
+ # Extract flags into the local scope
+ # This allows callers to set flags such as nowiki, found, etc.
extract( $result );
} else {
$text = $result;
# Split the title into page and subpage
$subpage = '';
$part1 = $this->maybeDoSubpageLink( $part1, $subpage );
- if ($subpage !== '') {
+ if ( $subpage !== '' ) {
$ns = $this->mTitle->getNamespace();
}
$title = Title::newFromText( $part1, $ns );
if ( $title ) {
$titleText = $title->getPrefixedText();
# Check for language variants if the template is not found
- if($wgContLang->hasVariants() && $title->getArticleID() == 0){
+ if ( $wgContLang->hasVariants() && $title->getArticleID() == 0 ) {
$wgContLang->findVariantLink( $part1, $title, true );
}
# Do recursion depth check
$limit = $this->mOptions->getMaxTemplateDepth();
if ( $frame->depth >= $limit ) {
$found = true;
- $text = '<span class="error">' . wfMsgForContent( 'parser-template-recursion-depth-warning', $limit ) . '</span>';
+ $text = '<span class="error">'
+ . wfMsgForContent( 'parser-template-recursion-depth-warning', $limit )
+ . '</span>';
}
}
}
if ( !$found && $title ) {
wfProfileIn( __METHOD__ . '-loadtpl' );
if ( !$title->isExternal() ) {
- if ( $title->getNamespace() == NS_SPECIAL && $this->mOptions->getAllowSpecialInclusion() && $this->ot['html'] ) {
+ if ( $title->getNamespace() == NS_SPECIAL
+ && $this->mOptions->getAllowSpecialInclusion()
+ && $this->ot['html'] )
+ {
$text = SpecialPage::capturePath( $title );
if ( is_string( $text ) ) {
$found = true;
$this->disableCache();
}
} elseif ( $wgNonincludableNamespaces && in_array( $title->getNamespace(), $wgNonincludableNamespaces ) ) {
- $found = false; //access denied
+ $found = false; # access denied
wfDebug( __METHOD__.": template inclusion denied for " . $title->getPrefixedDBkey() );
} else {
list( $text, $title ) = $this->getTemplateDom( $title );
$found = true;
}
} elseif ( $title->isTrans() ) {
- // Interwiki transclusion
+ # Interwiki transclusion
if ( $this->ot['html'] && !$forceRawInterwiki ) {
$text = $this->interwikiTransclude( $title, 'render' );
$isHTML = true;
} else {
$text = $this->interwikiTransclude( $title, 'raw' );
- // Preprocess it like a template
+ # Preprocess it like a template
$text = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION );
$isChildObj = true;
}
# immediately preceding headings
if ( $isHTML ) {
$text = "\n\n" . $this->insertStripItem( $text );
- }
- # Escape nowiki-style return values
- elseif ( $nowiki && ( $this->ot['html'] || $this->ot['pre'] ) ) {
+ } elseif ( $nowiki && ( $this->ot['html'] || $this->ot['pre'] ) ) {
+ # Escape nowiki-style return values
$text = wfEscapeWikiText( $text );
- }
- # Bug 529: if the template begins with a table or block-level
- # element, it should be treated as beginning a new line.
- # This behaviour is somewhat controversial.
- elseif ( is_string( $text ) && !$piece['lineStart'] && preg_match('/^(?:{\\||:|;|#|\*)/', $text)) /*}*/{
+ } elseif ( is_string( $text )
+ && !$piece['lineStart']
+ && preg_match( '/^(?:{\\||:|;|#|\*)/', $text ) )
+ {
+ # Bug 529: if the template begins with a table or block-level
+ # element, it should be treated as beginning a new line.
+ # This behaviour is somewhat controversial.
$text = "\n" . $text;
}
if ( is_string( $text ) && !$this->incrementIncludeSize( 'post-expand', strlen( $text ) ) ) {
# Error, oversize inclusion
- $text = "[[$originalTitle]]" .
- $this->insertStripItem( '<!-- WARNING: template omitted, post-expand include size too large -->' );
+ if ( $titleText !== false ) {
+ # Make a working, properly escaped link if possible (bug 23588)
+ $text = "[[:$titleText]]";
+ } else {
+ # This will probably not be a working link, but at least it may
+ # provide some hint of where the problem is
+ preg_replace( '/^:/', '', $originalTitle );
+ $text = "[[:$originalTitle]]";
+ }
+ $text .= $this->insertStripItem( '<!-- WARNING: template omitted, post-expand include size too large -->' );
$this->limitationWarn( 'post-expand-template-inclusion' );
}
return array( $this->mTplDomCache[$titleText], $title );
}
- // Cache miss, go to the database
+ # Cache miss, go to the database
list( $text, $title ) = $this->fetchTemplateAndTitle( $title );
if ( $text === false ) {
$dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION );
$this->mTplDomCache[ $titleText ] = $dom;
- if (! $title->equals($cacheTitle)) {
+ if ( !$title->equals( $cacheTitle ) ) {
$this->mTplRedirCache[$cacheTitle->getPrefixedDBkey()] =
array( $title->getNamespace(),$cdb = $title->getDBkey() );
}
* 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;
$this->mOutput->addTemplate( $dep['title'], $dep['page_id'], $dep['rev_id'] );
}
}
- return array($text,$finalTitle);
+ return array( $text, $finalTitle );
}
function fetchTemplate( $title ) {
- $rv = $this->fetchTemplateAndTitle($title);
+ $rv = $this->fetchTemplateAndTitle( $title );
return $rv[0];
}
$finalTitle = $title;
$deps = array();
- // Loop to fetch the article, with up to 1 redirect
+ # Loop to fetch the article, with up to 1 redirect
for ( $i = 0; $i < 2 && is_object( $title ); $i++ ) {
# Give extensions a chance to select the revision instead
- $id = false; // Assume current
+ $id = false; # Assume current
wfRunHooks( 'BeforeParserFetchTemplateAndtitle', array( $parser, &$title, &$skip, &$id ) );
- if( $skip ) {
+ if ( $skip ) {
$text = false;
$deps[] = array(
'title' => $title,
}
$rev = $id ? Revision::newFromId( $id ) : Revision::newFromTitle( $title );
$rev_id = $rev ? $rev->getId() : 0;
- // If there is no current revision, there is no page
- if( $id === false && !$rev ) {
+ # If there is no current revision, there is no page
+ if ( $id === false && !$rev ) {
$linkCache = LinkCache::singleton();
$linkCache->addBadLinkObj( $title );
}
'page_id' => $title->getArticleID(),
'rev_id' => $rev_id );
- if( $rev ) {
+ if ( $rev ) {
$text = $rev->getText();
- } elseif( $title->getNamespace() == NS_MEDIAWIKI ) {
+ } elseif ( $title->getNamespace() == NS_MEDIAWIKI ) {
global $wgContLang;
$message = $wgContLang->lcfirst( $title->getText() );
$text = wfMsgForContentNoTrans( $message );
- if( wfEmptyMsg( $message, $text ) ) {
+ if ( wfEmptyMsg( $message, $text ) ) {
$text = false;
break;
}
if ( $text === false ) {
break;
}
- // Redirect?
+ # Redirect?
$finalTitle = $title;
$title = Title::newFromRedirect( $text );
}
function interwikiTransclude( $title, $action ) {
global $wgEnableScaryTranscluding;
- if (!$wgEnableScaryTranscluding)
+ if ( !$wgEnableScaryTranscluding ) {
return wfMsg('scarytranscludedisabled');
+ }
$url = $title->getFullUrl( "action=$action" );
- if (strlen($url) > 255)
- return wfMsg('scarytranscludetoolong');
- return $this->fetchScaryTemplateMaybeFromCache($url);
+ if ( strlen( $url ) > 255 ) {
+ return wfMsg( 'scarytranscludetoolong' );
+ }
+ return $this->fetchScaryTemplateMaybeFromCache( $url );
}
- function fetchScaryTemplateMaybeFromCache($url) {
+ function fetchScaryTemplateMaybeFromCache( $url ) {
global $wgTranscludeCacheExpiry;
- $dbr = wfGetDB(DB_SLAVE);
+ $dbr = wfGetDB( DB_SLAVE );
$tsCond = $dbr->timestamp( time() - $wgTranscludeCacheExpiry );
- $obj = $dbr->selectRow('transcache', array('tc_time', 'tc_contents'),
- array('tc_url' => $url, "tc_time >= " . $dbr->addQuotes( $tsCond ) ) );
- if ($obj) {
+ $obj = $dbr->selectRow( 'transcache', array('tc_time', 'tc_contents' ),
+ array( 'tc_url' => $url, "tc_time >= " . $dbr->addQuotes( $tsCond ) ) );
+ if ( $obj ) {
return $obj->tc_contents;
}
- $text = Http::get($url);
- if (!$text)
- return wfMsg('scarytranscludefailed', $url);
+ $text = Http::get( $url );
+ if ( !$text ) {
+ return wfMsg( 'scarytranscludefailed', $url );
+ }
- $dbw = wfGetDB(DB_MASTER);
- $dbw->replace('transcache', array('tc_url'), array(
+ $dbw = wfGetDB( DB_MASTER );
+ $dbw->replace( 'transcache', array('tc_url'), array(
'tc_url' => $url,
'tc_time' => $dbw->timestamp( time() ),
- 'tc_contents' => $text));
+ 'tc_contents' => $text)
+ );
return $text;
}
* Return the text to be used for a given extension tag.
* This is the ghost of strip().
*
- * @param array $params Associative array of parameters:
+ * @param $params Associative array of parameters:
* name PPNode for the tag name
* attr PPNode for unparsed text where tag attributes are thought to be
* attributes Optional associative array of parsed attributes
* inner Contents of extension element
* noClose Original text did not have a close tag
- * @param PPFrame $frame
+ * @param $frame PPFrame
*/
function extensionSubstitution( $params, $frame ) {
- global $wgRawHtml, $wgContLang;
-
$name = $frame->expand( $params['name'] );
$attrText = !isset( $params['attr'] ) ? null : $frame->expand( $params['attr'] );
$content = !isset( $params['inner'] ) ? null : $frame->expand( $params['inner'] );
-
- $marker = "{$this->mUniqPrefix}-$name-" . sprintf('%08X', $this->mMarkerIndex++) . self::MARKER_SUFFIX;
+ $marker = "{$this->mUniqPrefix}-$name-" . sprintf( '%08X', $this->mMarkerIndex++ ) . self::MARKER_SUFFIX;
$isFunctionTag = isset( $this->mFunctionTagHooks[strtolower($name)] ) &&
( $this->ot['html'] || $this->ot['pre'] );
+ 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 'gallery':
- $output = $this->renderImageGallery( $content, $attributes );
- break;
- case 'a':
- $output = $this->renderHyperlink( $content, $attributes, $frame );
- break;
- case 'math':
- if ( $this->mOptions->getUseTeX() ) {
- $output = $wgContLang->armourMath(
- MathRenderer::renderMath( $content, $attributes ) );
- break;
- }
- /* else let a tag hook handle it (bug 21222) */
- 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, $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 ( 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 ) ) {
}
}
- if( $isFunctionTag ) {
+ if ( $markerType === 'none' ) {
return $output;
- } elseif ( $name === 'html' || $name === 'nowiki' ) {
+ } 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;
}
/**
* 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 ) ) {
/**
* Increment the expensive function count
*
- * @return boolean False if the limit has been exceeded
+ * @return Boolean: false if the limit has been exceeded
*/
function incrementExpensiveFunctionCount() {
global $wgExpensiveParserFunctionLimit;
$this->mExpensiveFunctionCount++;
- if($this->mExpensiveFunctionCount <= $wgExpensiveParserFunctionLimit) {
+ if ( $this->mExpensiveFunctionCount <= $wgExpensiveParserFunctionLimit ) {
return true;
}
return false;
function doDoubleUnderscore( $text ) {
wfProfileIn( __METHOD__ );
- // The position of __TOC__ needs to be recorded
+ # The position of __TOC__ needs to be recorded
$mw = MagicWord::get( 'toc' );
- if( $mw->match( $text ) ) {
+ if ( $mw->match( $text ) ) {
$this->mShowToc = true;
$this->mForceTocPosition = true;
- // Set a placeholder. At the end we'll fill it in with the TOC.
+ # Set a placeholder. At the end we'll fill it in with the TOC.
$text = $mw->replace( '<!--MWTOC-->', $text, 1 );
- // Only keep the first one.
+ # Only keep the first one.
$text = $mw->replace( '', $text );
}
- // Now match and remove the rest of them
+ # Now match and remove the rest of them
$mwa = MagicWord::getDoubleUnderscoreArray();
$this->mDoubleUnderscores = $mwa->matchAndRemove( $text );
$this->mShowToc = false;
}
if ( isset( $this->mDoubleUnderscores['hiddencat'] ) && $this->mTitle->getNamespace() == NS_CATEGORY ) {
- $this->mOutput->setProperty( 'hiddencat', 'y' );
$this->addTrackingCategory( 'hidden-category-category' );
}
# (bug 8068) Allow control over whether robots index a page.
#
# FIXME (bug 14899): __INDEX__ always overrides __NOINDEX__ here! This
# is not desirable, the last one on the page should win.
- if( isset( $this->mDoubleUnderscores['noindex'] ) && $this->mTitle->canUseNoindex() ) {
+ if ( isset( $this->mDoubleUnderscores['noindex'] ) && $this->mTitle->canUseNoindex() ) {
$this->mOutput->setIndexPolicy( 'noindex' );
$this->addTrackingCategory( 'noindex-category' );
}
- if( isset( $this->mDoubleUnderscores['index'] ) && $this->mTitle->canUseNoindex() ){
+ if ( isset( $this->mDoubleUnderscores['index'] ) && $this->mTitle->canUseNoindex() ) {
$this->mOutput->setIndexPolicy( 'index' );
$this->addTrackingCategory( 'index-category' );
}
+
+ # Cache all double underscores in the database
+ foreach ( $this->mDoubleUnderscores as $key => $val ) {
+ $this->mOutput->setProperty( $key, '' );
+ }
wfProfileOut( __METHOD__ );
return $text;
/**
* Add a tracking category, getting the title from a system message,
* or print a debug message if the title is invalid.
- * @param $msg String message key
- * @return Bool whether the addition was successful
+ *
+ * @param $msg String: message key
+ * @return Boolean: whether the addition was successful
*/
- protected function addTrackingCategory( $msg ){
+ protected function addTrackingCategory( $msg ) {
$cat = wfMsgForContent( $msg );
# Allow tracking categories to be disabled by setting them to "-"
- if( $cat === '-' ) return false;
+ if ( $cat === '-' ) {
+ return false;
+ }
$containerCategory = Title::makeTitleSafe( NS_CATEGORY, $cat );
if ( $containerCategory ) {
* 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, $wgEnforceHtmlIds;
+ 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]
# if there are fewer than 4 headlines in the article, do not show TOC
# unless it's been explicitly enabled.
$enoughToc = $this->mShowToc &&
- (($numMatches >= 4) || $this->mForceTocPosition);
+ ( ( $numMatches >= 4 ) || $this->mForceTocPosition );
# Allow user to stipulate that a page should have a "new section"
# link added via __NEWSECTIONLINK__
$byteOffset = 0;
$tocraw = array();
- foreach( $matches[3] as $headline ) {
+ foreach ( $matches[3] as $headline ) {
$isTemplate = false;
$titleText = false;
$sectionIndex = false;
$numbering = '';
$markerMatches = array();
- if (preg_match("/^$markerRegex/", $headline, $markerMatches)) {
+ if ( preg_match("/^$markerRegex/", $headline, $markerMatches ) ) {
$serial = $markerMatches[1];
list( $titleText, $sectionIndex ) = $this->mHeadings[$serial];
- $isTemplate = ($titleText !== $baseTitleText);
- $headline = preg_replace("/^$markerRegex/", "", $headline);
+ $isTemplate = ( $titleText != $baseTitleText );
+ $headline = preg_replace( "/^$markerRegex/", "", $headline );
}
- if( $toclevel ) {
+ if ( $toclevel ) {
$prevlevel = $level;
$prevtoclevel = $toclevel;
}
# Increase TOC level
$toclevel++;
$sublevelCount[$toclevel] = 0;
- if( $toclevel<$wgMaxTocLevel ) {
+ if ( $toclevel<$wgMaxTocLevel ) {
$prevtoclevel = $toclevel;
$toc .= $sk->tocIndent();
$numVisible++;
}
- }
- elseif ( $level < $prevlevel && $toclevel > 1 ) {
+ } elseif ( $level < $prevlevel && $toclevel > 1 ) {
# Decrease TOC level, find level to jump to
- for ($i = $toclevel; $i > 0; $i--) {
+ for ( $i = $toclevel; $i > 0; $i-- ) {
if ( $levelCount[$i] == $level ) {
# Found last matching level
$toclevel = $i;
break;
- }
- elseif ( $levelCount[$i] < $level ) {
+ } elseif ( $levelCount[$i] < $level ) {
# Found first matching level below current level
$toclevel = $i + 1;
break;
}
}
- if( $i == 0 ) $toclevel = 1;
- if( $toclevel<$wgMaxTocLevel ) {
- if($prevtoclevel < $wgMaxTocLevel) {
+ if ( $i == 0 ) {
+ $toclevel = 1;
+ }
+ if ( $toclevel<$wgMaxTocLevel ) {
+ if ( $prevtoclevel < $wgMaxTocLevel ) {
# Unindent only if the previous toc level was shown :p
$toc .= $sk->tocUnindent( $prevtoclevel - $toclevel );
$prevtoclevel = $toclevel;
$toc .= $sk->tocLineEnd();
}
}
- }
- else {
+ } else {
# No change in level, end TOC line
- if( $toclevel<$wgMaxTocLevel ) {
+ if ( $toclevel<$wgMaxTocLevel ) {
$toc .= $sk->tocLineEnd();
}
}
@$sublevelCount[$toclevel]++;
$dot = 0;
for( $i = 1; $i <= $toclevel; $i++ ) {
- if( !empty( $sublevelCount[$i] ) ) {
- if( $dot ) {
+ if ( !empty( $sublevelCount[$i] ) ) {
+ if ( $dot ) {
$numbering .= '.';
}
$numbering .= $wgContLang->formatNum( $sublevelCount[$i] );
# Strip out HTML (other than plain <sup> and <sub>: bug 8393)
$tocline = preg_replace(
array( '#<(?!/?(sup|sub)).*?'.'>#', '#<(/?(sup|sub)).*?'.'>#' ),
- array( '', '<$1>'),
+ array( '', '<$1>' ),
$safeHeadline
);
$tocline = trim( $tocline );
# For the anchor, strip out HTML-y stuff period
$safeHeadline = preg_replace( '/<.*?'.'>/', '', $safeHeadline );
- $safeHeadline = preg_replace( '/[ _]+/', ' ', $safeHeadline );
- $safeHeadline = trim( $safeHeadline );
+ $safeHeadline = Sanitizer::normalizeSectionNameWhitespace( $safeHeadline );
# Save headline for section edit hint before it's escaped
$headlineHint = $safeHeadline;
- 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.
#
# 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:
}
# Don't number the heading if it is the only one (looks silly)
- if( $doNumberHeadings && count( $matches[3] ) > 1) {
+ if ( $doNumberHeadings && count( $matches[3] ) > 1) {
# the two are different if the line contains a link
- $headline=$numbering . ' ' . $headline;
+ $headline = $numbering . ' ' . $headline;
}
# Create the anchor for linking from the TOC to the section
if ( $legacyHeadline !== false && $refers[$legacyArrayKey] > 1 ) {
$legacyAnchor .= '_' . $refers[$legacyArrayKey];
}
- if( $enoughToc && ( !isset($wgMaxTocLevel) || $toclevel<$wgMaxTocLevel ) ) {
- $toc .= $sk->tocLine($anchor, $tocline,
- $numbering, $toclevel, ($isTemplate ? false : $sectionIndex));
+ if ( $enoughToc && ( !isset( $wgMaxTocLevel ) || $toclevel < $wgMaxTocLevel ) ) {
+ $toc .= $sk->tocLine( $anchor, $tocline,
+ $numbering, $toclevel, ( $isTemplate ? false : $sectionIndex ) );
}
# Add the section to the section tree
'level' => $level,
'line' => $tocline,
'number' => $numbering,
- 'index' => ($isTemplate ? 'T-' : '' ) . $sectionIndex,
+ 'index' => ( $isTemplate ? 'T-' : '' ) . $sectionIndex,
'fromtitle' => $titleText,
'byteoffset' => ( $isTemplate ? null : $byteOffset ),
'anchor' => $anchor,
);
# give headline the correct <h#> tag
- if( $showEditLink && $sectionIndex !== false ) {
- if( $isTemplate ) {
+ if ( $showEditLink && $sectionIndex !== false ) {
+ if ( $isTemplate ) {
# Put a T flag in the section identifier, to indicate to extractSections()
# that sections inside <includeonly> should be counted.
- $editlink = $sk->doEditSectionLink(Title::newFromText( $titleText ), "T-$sectionIndex");
+ $editlink = $sk->doEditSectionLink( Title::newFromText( $titleText ), "T-$sectionIndex" );
} else {
- $editlink = $sk->doEditSectionLink($this->mTitle, $sectionIndex, $headlineHint);
+ $editlink = $sk->doEditSectionLink( $this->mTitle, $sectionIndex, $headlineHint );
}
} else {
$editlink = '';
$this->setOutputType( $oldType );
# Never ever show TOC if no headers
- if( $numVisible < 1 ) {
+ if ( $numVisible < 1 ) {
$enoughToc = false;
}
- if( $enoughToc ) {
- if( $prevtoclevel > 0 && $prevtoclevel < $wgMaxTocLevel ) {
+ if ( $enoughToc ) {
+ if ( $prevtoclevel > 0 && $prevtoclevel < $wgMaxTocLevel ) {
$toc .= $sk->tocUnindent( $prevtoclevel - 1 );
}
$toc = $sk->tocList( $toc );
$blocks = preg_split( '/<H[1-6].*?' . '>.*?<\/H[1-6]>/i', $text );
$i = 0;
- foreach( $blocks as $block ) {
- if( $showEditLink && $headlineCount > 0 && $i == 0 && $block !== "\n" ) {
+ foreach ( $blocks as $block ) {
+ if ( $showEditLink && $headlineCount > 0 && $i == 0 && $block !== "\n" ) {
# This is the [edit] link that appears for the top block of text when
# section editing is enabled
# $full .= $sk->editSectionLink(0);
}
$full .= $block;
- if( $enoughToc && !$i && $isMain && !$this->mForceTocPosition ) {
+ if ( $enoughToc && !$i && $isMain && !$this->mForceTocPosition ) {
# Top anchor now in skin
$full = $full.$toc;
}
- if( !empty( $head[$i] ) ) {
+ if ( !empty( $head[$i] ) ) {
$full .= $head[$i];
}
$i++;
}
- if( $this->mForceTocPosition ) {
+ if ( $this->mForceTocPosition ) {
return str_replace( '<!--MWTOC-->', $toc, $full );
} else {
return $full;
}
}
- /**
- * 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 ) {
+ $options->resetUsage();
$this->mOptions = $options;
$this->setTitle( $title );
$this->setOutputType( self::OT_WIKI );
function pstPass2( $text, $user ) {
global $wgContLang, $wgLocaltimezone;
- /* Note: This is the timestamp saved as hardcoded wikitext to
- * the database, we use $wgContLang here in order to give
- * everyone the same signature and use the default one rather
- * than the one selected in each user's preferences.
- *
- * (see also bug 12815)
- */
+ # Note: This is the timestamp saved as hardcoded wikitext to
+ # the database, we use $wgContLang here in order to give
+ # everyone the same signature and use the default one rather
+ # than the one selected in each user's preferences.
+ # (see also bug 12815)
$ts = $this->mOptions->getTimestamp();
if ( isset( $wgLocaltimezone ) ) {
$tz = $wgLocaltimezone;
$ts = date( 'YmdHis', $unixts );
$tzMsg = date( 'T', $unixts ); # might vary on DST changeover!
- /* Allow translation of timezones trough wiki. date() can return
- * whatever crap the system uses, localised or not, so we cannot
- * ship premade translations.
- */
+ # Allow translation of timezones through wiki. date() can return
+ # whatever crap the system uses, localised or not, so we cannot
+ # ship premade translations.
$key = 'timezone-' . strtolower( trim( $tzMsg ) );
$value = wfMsgForContent( $key );
- if ( !wfEmptyMsg( $key, $value ) ) $tzMsg = $value;
+ if ( !wfEmptyMsg( $key, $value ) ) {
+ $tzMsg = $value;
+ }
date_default_timezone_set( $oldtz );
) );
# Context links: [[|name]] and [[name (context)|]]
- #
global $wgLegalTitleChars;
$tc = "[$wgLegalTitleChars]";
$nc = '[ _0-9A-Za-z\x80-\xff-]'; # Namespaces can use non-ascii!
$m = array();
if ( preg_match( "/^($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
* 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 ) {
$username = $user->getName();
- // If not given, retrieve from the user object.
+ # If not given, retrieve from the user object.
if ( $nickname === false )
$nickname = $user->getOption( 'nickname' );
- if ( is_null( $fancySig ) )
+ if ( is_null( $fancySig ) ) {
$fancySig = $user->getBoolOption( 'fancysig' );
+ }
$nickname = $nickname == null ? $username : $nickname;
- if( mb_strlen( $nickname ) > $wgMaxSigChars ) {
+ if ( mb_strlen( $nickname ) > $wgMaxSigChars ) {
$nickname = $username;
wfDebug( __METHOD__ . ": $username has overlong signature.\n" );
- } elseif( $fancySig !== false ) {
+ } elseif ( $fancySig !== false ) {
# Sig. might contain markup; validate this
- if( $this->validateSig( $nickname ) !== false ) {
+ if ( $this->validateSig( $nickname ) !== false ) {
# Validated; clean up (if needed) and return it
return $this->cleanSig( $nickname, true );
} else {
}
}
- // Make sure nickname doesnt get a sig in a sig
+ # Make sure nickname doesnt get a sig in a sig
$nickname = $this->cleanSigInSig( $nickname );
# If we're still here, make it a link to the user page
/**
* 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 ) {
* 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 ) {
/**
* 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}/', '', $text );
/**
* Set up some variables which are usually set up in parse()
* so that an external function can call some class members with confidence
- * @public
*/
- function startExternalParse( &$title, $options, $outputType, $clearState = true ) {
+ public function startExternalParse( &$title, $options, $outputType, $clearState = true ) {
$this->setTitle( $title );
+ $options->resetUsage();
$this->mOptions = $options;
$this->setOutputType( $outputType );
if ( $clearState ) {
/**
* 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;
}
$executing = true;
- wfProfileIn(__METHOD__);
+ wfProfileIn( __METHOD__ );
$text = $this->preprocess( $text, $wgTitle, $options );
$executing = false;
- wfProfileOut(__METHOD__);
+ wfProfileOut( __METHOD__ );
return $text;
}
/**
* Create an HTML-style tag, e.g. <yourtag>special text</yourtag>
* The callback should have the following form:
- * function myParserHook( $text, $params, &$parser ) { ... }
+ * function myParserHook( $text, $params, $parser ) { ... }
*
* Transform and return $text. Use $parser for any required context, e.g. use
* $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions
*
- * @public
- *
- * @param mixed $tag The tag to use, e.g. 'hook' for <hook>
- * @param mixed $callback The callback function (and object) to use for the tag
- *
+ * @param $tag Mixed: the tag to use, e.g. 'hook' for <hook>
+ * @param $callback Mixed: the callback function (and object) to use for the tag
* @return The old value of the mTagHooks array associated with the hook
*/
- function setHook( $tag, $callback ) {
+ public function setHook( $tag, $callback ) {
$tag = strtolower( $tag );
$oldVal = isset( $this->mTagHooks[$tag] ) ? $this->mTagHooks[$tag] : null;
$this->mTagHooks[$tag] = $callback;
- if( !in_array( $tag, $this->mStripList ) ) {
+ if ( !in_array( $tag, $this->mStripList ) ) {
$this->mStripList[] = $tag;
}
* 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
*
* @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;
# Add to function cache
$mw = MagicWord::get( $id );
- if( !$mw )
+ if ( !$mw )
throw new MWException( __METHOD__.'() expecting a magic word identifier.' );
$synonyms = $mw->getSynonyms();
/**
* Get all registered function hook identifiers
*
- * @return array
+ * @return Array
*/
function getFunctionHooks() {
return array_keys( $this->mFunctionHooks );
$this->mFunctionTagHooks[$tag] : null;
$this->mFunctionTagHooks[$tag] = array( $callback, $flags );
- if( !in_array( $tag, $this->mStripList ) ) {
+ if ( !in_array( $tag, $this->mStripList ) ) {
$this->mStripList[] = $tag;
}
/**
* Replace <!--LINK--> link placeholders with plain text of links
* (not HTML-formatted).
- * @param string $text
- * @return string
+ *
+ * @param $text String
+ * @return String
*/
function replaceLinkHoldersText( $text ) {
return $this->mLinkHolders->replaceText( $text );
}
- /**
- * 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>';
- }
-
- /**
- * Tag hook handler for 'a'. Renders a HTML <a> tag, allowing most attributes, filtering href against
- * allowed protocols and spam blacklist.
- **/
- function renderHyperlink( $text, $params, $frame = false ) {
- foreach ( $params as $name => $value ) {
- $params[ $name ] = $this->replaceVariables( $value, $frame );
- }
-
- $whitelist = Sanitizer::attributeWhitelist( 'a' );
- $params = Sanitizer::validateAttributes( $params, $whitelist );
-
- $content = $this->recursiveTagParse( trim( $text ), $frame );
-
- if ( isset( $params[ 'href' ] ) ) {
- $href = $params[ 'href' ];
- $this->mOutput->addExternalLink( $href );
- unset( $params[ 'href' ] );
- } else {
- # Non-link <a> tag
- return Xml::openElement( 'a', $params ) . $content . Xml::closeElement( 'a' );
- }
-
- $sk = $this->mOptions->getSkin();
- $html = $sk->makeExternalLink( $href, $content, false, '', $params );
-
- return $html;
- }
-
/**
* Renders an image gallery from a text with one line per image.
* text labels may be given by using |-style alternative text. E.g.
$ig->useSkin( $this->mOptions->getSkin() );
$ig->mRevisionId = $this->mRevisionId;
- if( isset( $params['caption'] ) ) {
+ if ( isset( $params['showfilename'] ) ) {
+ $ig->setShowFilename( true );
+ } else {
+ $ig->setShowFilename( false );
+ }
+ if ( isset( $params['caption'] ) ) {
$caption = $params['caption'];
$caption = htmlspecialchars( $caption );
$caption = $this->replaceInternalLinks( $caption );
$ig->setCaptionHtml( $caption );
}
- if( isset( $params['perrow'] ) ) {
+ if ( isset( $params['perrow'] ) ) {
$ig->setPerRow( $params['perrow'] );
}
- if( isset( $params['widths'] ) ) {
+ if ( isset( $params['widths'] ) ) {
$ig->setWidths( $params['widths'] );
}
- if( isset( $params['heights'] ) ) {
+ if ( isset( $params['heights'] ) ) {
$ig->setHeights( $params['heights'] );
}
continue;
}
- if ( strpos( $matches[0], '%' ) !== false )
+ if ( strpos( $matches[0], '%' ) !== false ) {
$matches[1] = urldecode( $matches[1] );
- $tp = Title::newFromText( $matches[1]/*, NS_FILE*/ );
+ }
+ $tp = Title::newFromText( $matches[1] );
$nt =& $tp;
- if( is_null( $nt ) ) {
+ if ( is_null( $nt ) ) {
# Bogus title. Ignore these so we don't bomb out later.
continue;
}
$handlerClass = '';
}
if ( !isset( $this->mImageParams[$handlerClass] ) ) {
- // Initialise static lists
+ # Initialise static lists
static $internalParamNames = array(
'horizAlign' => array( 'left', 'right', 'center', 'none' ),
'vertAlign' => array( 'baseline', 'sub', 'super', 'top', 'text-top', 'middle',
}
}
- // Add handler params
+ # Add handler params
$paramMap = $internalParamMap;
if ( $handler ) {
$handlerParamMap = $handler->getParamMap();
/**
* 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"
$caption = '';
$params = array( 'frame' => array(), 'handler' => array(),
'horizAlign' => array(), 'vertAlign' => array() );
- foreach( $parts as $part ) {
+ foreach ( $parts as $part ) {
$part = trim( $part );
list( $magicName, $value ) = $mwArray->matchVariableStartToEnd( $part );
$validated = false;
- if( isset( $paramMap[$magicName] ) ) {
+ if ( isset( $paramMap[$magicName] ) ) {
list( $type, $paramName ) = $paramMap[$magicName];
- // Special case; width and height come in one variable together
- if( $type === 'handler' && $paramName === 'width' ) {
+ # Special case; width and height come in one variable together
+ if ( $type === 'handler' && $paramName === 'width' ) {
$m = array();
# (bug 13500) In both cases (width/height and width only),
# permit trailing "px" for backward compatibility.
$params[$type]['width'] = $width;
$validated = true;
}
- } // else no validation -- bug 13436
+ } # else no validation -- bug 13436
} else {
if ( $type === 'handler' ) {
# Validate handler parameter
switch( $paramName ) {
case 'manualthumb':
case 'alt':
- // @todo Fixme: possibly check validity here for
- // manualthumb? downstream behavior seems odd with
- // missing manual thumbs.
+ # @todo Fixme: possibly check validity here for
+ # manualthumb? downstream behavior seems odd with
+ # missing manual thumbs.
$validated = true;
$value = $this->stripAltText( $value, $holders );
break;
}
break;
default:
- // Most other things appear to be empty or numeric...
+ # Most other things appear to be empty or numeric...
$validated = ( $value === false || is_numeric( trim( $value ) ) );
}
}
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 ) {
*/
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 ) {
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
*
* External callers should use the getSection and replaceSection methods.
*
- * @param string $text Page wikitext
- * @param string $section A section identifier string of the form:
+ * @param $text String: Page wikitext
+ * @param $section String: a section identifier string of the form:
* <flag1> - <flag2> - ... - <section number>
*
* Currently the only recognised flag is "T", which means the target section number
* pull the given section along with its lower-level subsections. If the section is
* not found, $mode=get will return $newtext, and $mode=replace will return $text.
*
- * @param string $mode One of "get" or "replace"
- * @param string $newText Replacement text for section data.
- * @return string for "get", the extracted section text.
- * for "replace", the whole page with the section replaced.
+ * @param $mode String: one of "get" or "replace"
+ * @param $newText String: replacement text for section data.
+ * @return String: for "get", the extracted section text.
+ * for "replace", the whole page with the section replaced.
*/
private function extractSections( $text, $section, $mode, $newText='' ) {
global $wgTitle;
$this->clearState();
- $this->setTitle( $wgTitle ); // not generally used but removes an ugly failure mode
+ $this->setTitle( $wgTitle ); # not generally used but removes an ugly failure mode
$this->mOptions = new ParserOptions;
- $this->setOutputType( self::OT_WIKI );
+ $this->setOutputType( self::OT_PLAIN );
$outText = '';
$frame = $this->getPreprocessor()->newFrame();
- // Process section extraction flags
+ # Process section extraction flags
$flags = 0;
$sectionParts = explode( '-', $section );
$sectionIndex = array_pop( $sectionParts );
$flags |= self::PTD_FOR_INCLUSION;
}
}
- // Preprocess the text
+ # Preprocess the text
$root = $this->preprocessToDom( $text, $flags );
- // <h> nodes indicate section breaks
- // They can only occur at the top level, so we can find them by iterating the root's children
+ # <h> nodes indicate section breaks
+ # They can only occur at the top level, so we can find them by iterating the root's children
$node = $root->getFirstChild();
- // Find the target section
+ # Find the target section
if ( $sectionIndex == 0 ) {
- // Section zero doesn't nest, level=big
+ # Section zero doesn't nest, level=big
$targetLevel = 1000;
} else {
- while ( $node ) {
- if ( $node->getName() === 'h' ) {
- $bits = $node->splitHeading();
+ while ( $node ) {
+ if ( $node->getName() === 'h' ) {
+ $bits = $node->splitHeading();
if ( $bits['i'] == $sectionIndex ) {
- $targetLevel = $bits['level'];
+ $targetLevel = $bits['level'];
break;
}
}
}
if ( !$node ) {
- // Not found
+ # Not found
if ( $mode === 'get' ) {
return $newText;
} else {
}
}
- // Find the end of the section, including nested sections
+ # Find the end of the section, including nested sections
do {
if ( $node->getName() === 'h' ) {
$bits = $node->splitHeading();
$curLevel = $bits['level'];
- if ( $bits['i'] !== $sectionIndex && $curLevel <= $targetLevel ) {
+ if ( $bits['i'] != $sectionIndex && $curLevel <= $targetLevel ) {
break;
}
}
$node = $node->getNextSibling();
} while ( $node );
- // Write out the remainder (in replace mode only)
+ # Write out the remainder (in replace mode only)
if ( $mode === 'replace' ) {
- // Output the replacement text
- // Add two newlines on -- trailing whitespace in $newText is conventionally
- // stripped by the editor, so we need both newlines to restore the paragraph gap
- // Only add trailing whitespace if there is newText
- if($newText !== "") {
+ # Output the replacement text
+ # Add two newlines on -- trailing whitespace in $newText is conventionally
+ # stripped by the editor, so we need both newlines to restore the paragraph gap
+ # Only add trailing whitespace if there is newText
+ if ( $newText != "" ) {
$outText .= $newText . "\n\n";
}
}
if ( is_string( $outText ) ) {
- // Re-insert stripped tags
+ # Re-insert stripped tags
$outText = rtrim( $this->mStripState->unstripBoth( $outText ) );
}
*
* If a section contains subsections, these are also returned.
*
- * @param string $text text to look in
- * @param string $section section identifier
- * @param string $deftext default to return if section is not found
+ * @param $text String: text to look in
+ * @param $section String: section identifier
+ * @param $deftext String: default to return if section is not found
* @return string text of the requested section
*/
public function getSection( $text, $section, $deftext='' ) {
return $this->extractSections( $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
$timestamp = $dbr->selectField( 'revision', 'rev_timestamp',
array( 'rev_id' => $this->mRevisionId ), __METHOD__ );
- // Normalize timestamp to internal MW format for timezone processing.
- // This has the added side-effect of replacing a null value with
- // the current time, which gives us more sensible behavior for
- // previews.
+ # Normalize timestamp to internal MW format for timezone processing.
+ # This has the added side-effect of replacing a null value with
+ # the current time, which gives us more sensible behavior for
+ # previews.
$timestamp = wfTimestamp( TS_MW, $timestamp );
- // The cryptic '' timezone parameter tells to use the site-default
- // timezone offset instead of the user settings.
- //
- // Since this value will be saved into the parser cache, served
- // to other users, and potentially even used inside links and such,
- // it needs to be consistent for all visitors.
+ # The cryptic '' timezone parameter tells to use the site-default
+ # timezone offset instead of the user settings.
+ #
+ # Since this value will be saved into the parser cache, served
+ # to other users, and potentially even used inside links and such,
+ # it needs to be consistent for all visitors.
$this->mRevisionTimestamp = $wgContLang->userAdjust( $timestamp, '' );
wfProfileOut( __METHOD__ );
/**
* Get the name of the user that edited the last revision
+ *
+ * @return String: user name
*/
function getRevisionUser() {
- // if this template is subst: the revision id will be blank,
- // so just use the current user's name
- if( $this->mRevisionId ) {
+ # if this template is subst: the revision id will be blank,
+ # so just use the current user's name
+ if ( $this->mRevisionId ) {
$revision = Revision::newFromId( $this->mRevisionId );
$revuser = $revision->getUserText();
} else {
*/
public function setDefaultSort( $sort ) {
$this->mDefaultSort = $sort;
+ $this->mOutput->setProperty( 'defaultsort', $sort );
}
/**
* @return string
*/
public function getDefaultSort() {
- global $wgCategoryPrefixedDefaultSortkey;
- if( $this->mDefaultSort !== false ) {
+ if ( $this->mDefaultSort !== false ) {
return $this->mDefaultSort;
- } elseif ($this->mTitle->getNamespace() == NS_CATEGORY ||
- !$wgCategoryPrefixedDefaultSortkey) {
- return $this->mTitle->getText();
} else {
- return $this->mTitle->getPrefixedText();
+ return $this->mTitle->getCategorySortkey();
}
}
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' ) );
}
/**
* to create valid section anchors by mimicing the output of the
* parser when headings are parsed.
*
- * @param $text string Text string to be stripped of wikitext
+ * @param $text String: text string to be stripped of wikitext
* for use in a Section anchor
* @return Filtered text string
*/
public function stripSectionName( $text ) {
# Strip internal link markup
- $text = preg_replace('/\[\[:?([^[|]+)\|([^[]+)\]\]/','$2',$text);
- $text = preg_replace('/\[\[:?([^[]+)\|?\]\]/','$1',$text);
+ $text = preg_replace( '/\[\[:?([^[|]+)\|([^[]+)\]\]/', '$2', $text );
+ $text = preg_replace( '/\[\[:?([^[]+)\|?\]\]/', '$1', $text );
# Strip external link markup (FIXME: Not Tolerant to blank link text
# I.E. [http://www.mediawiki.org] will render as [1] or something depending
# on how many empty links there are on the page - need to figure that out.
- $text = preg_replace('/\[(?:' . wfUrlProtocols() . ')([^ ]+?) ([^[]+)\]/','$2',$text);
+ $text = preg_replace( '/\[(?:' . wfUrlProtocols() . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
# Parse wikitext quotes (italics & bold)
- $text = $this->doQuotes($text);
+ $text = $this->doQuotes( $text );
# Strip HTML tags
$text = StringUtils::delimiterReplace( '<', '>', '', $text );
*/
function testSrvus( $text, $title, $options, $outputType = self::OT_HTML ) {
$this->clearState();
- if ( ! ( $title instanceof Title ) ) {
+ if ( !$title instanceof Title ) {
$title = Title::newFromText( $title );
}
$this->mTitle = $title;
+ $options->resetUsage();
$this->mOptions = $options;
$this->setOutputType( $outputType );
$text = $this->replaceVariables( $text );
function testPst( $text, $title, $options ) {
global $wgUser;
- if ( ! ( $title instanceof Title ) ) {
+ if ( !$title instanceof Title ) {
$title = Title::newFromText( $title );
}
return $this->preSaveTransform( $text, $title, $wgUser, $options );
}
function testPreprocess( $text, $title, $options ) {
- if ( ! ( $title instanceof Title ) ) {
+ if ( !$title instanceof Title ) {
$title = Title::newFromText( $title );
}
return $this->testSrvus( $text, $title, $options, self::OT_PREPROCESS );
$data = array();
$data['text'] = $text;
- // First, find all strip markers, and store their
- // data in an array.
+ # First, find all strip markers, and store their
+ # data in an array.
$stripState = new StripState;
$pos = 0;
- while( ( $start_pos = strpos( $text, $this->mUniqPrefix, $pos ) ) && ( $end_pos = strpos( $text, self::MARKER_SUFFIX, $pos ) ) ) {
+ while ( ( $start_pos = strpos( $text, $this->mUniqPrefix, $pos ) )
+ && ( $end_pos = strpos( $text, self::MARKER_SUFFIX, $pos ) ) )
+ {
$end_pos += strlen( self::MARKER_SUFFIX );
$marker = substr( $text, $start_pos, $end_pos-$start_pos );
}
$data['stripstate'] = $stripState;
- // Now, find all of our links, and store THEIR
- // data in an array! :)
+ # Now, find all of our links, and store THEIR
+ # data in an array! :)
$links = array( 'internal' => array(), 'interwiki' => array() );
$pos = 0;
- // Internal links
- while( ( $start_pos = strpos( $text, '<!--LINK ', $pos ) ) ) {
+ # Internal links
+ while ( ( $start_pos = strpos( $text, '<!--LINK ', $pos ) ) ) {
list( $ns, $trail ) = explode( ':', substr( $text, $start_pos + strlen( '<!--LINK ' ) ), 2 );
- $ns = trim($ns);
- if (empty( $links['internal'][$ns] )) {
+ $ns = trim( $ns );
+ if ( empty( $links['internal'][$ns] ) ) {
$links['internal'][$ns] = array();
}
$pos = 0;
- // Interwiki links
- while( ( $start_pos = strpos( $text, '<!--IWLINK ', $pos ) ) ) {
+ # Interwiki links
+ while ( ( $start_pos = strpos( $text, '<!--IWLINK ', $pos ) ) ) {
$data = substr( $text, $start_pos );
$key = trim( substr( $data, 0, strpos( $data, '-->' ) ) );
$links['interwiki'][] = $this->mLinkHolders->interwiki[$key];
return $data;
}
- function unserialiseHalfParsedText( $data, $intPrefix = null /* Unique identifying prefix */ ) {
- if (!$intPrefix)
+ /**
+ * TODO: document
+ * @param $data Array
+ * @param $intPrefix String unique identifying prefix
+ * @return String
+ */
+ function unserialiseHalfParsedText( $data, $intPrefix = null ) {
+ if ( !$intPrefix ) {
$intPrefix = $this->getRandomString();
+ }
- // First, extract the strip state.
+ # First, extract the strip state.
$stripState = $data['stripstate'];
$this->mStripState->general->merge( $stripState->general );
$this->mStripState->nowiki->merge( $stripState->nowiki );
- // Now, extract the text, and renumber links
+ # Now, extract the text, and renumber links
$text = $data['text'];
$links = $data['linkholder'];
- // Internal...
- foreach( $links['internal'] as $ns => $nsLinks ) {
- foreach( $nsLinks as $key => $entry ) {
+ # Internal...
+ foreach ( $links['internal'] as $ns => $nsLinks ) {
+ foreach ( $nsLinks as $key => $entry ) {
$newKey = $intPrefix . '-' . $key;
$this->mLinkHolders->internals[$ns][$newKey] = $entry;
}
}
- // Interwiki...
- foreach( $links['interwiki'] as $key => $entry ) {
+ # Interwiki...
+ foreach ( $links['interwiki'] as $key => $entry ) {
$newKey = "$intPrefix-$key";
$this->mLinkHolders->interwikis[$newKey] = $entry;
$text = str_replace( "<!--IWLINK $key-->", "<!--IWLINK $newKey-->", $text );
}
- // Should be good to go.
+ # Should be good to go.
return $text;
}
}