* <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.
+ * Removes <noinclude> sections, and <includeonly> tags.
*
* Globals used:
* objects: $wgLang, $wgContLang
const EXT_IMAGE_REGEX = '/^(http:\/\/|https:\/\/)([^][<>"\\x00-\\x20\\x7F]+)
\\/([A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]+)\\.((?i)gif|png|jpg|jpeg)$/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; // like parse()
- const OT_WIKI = 2; // like preSaveTransform()
- const OT_PREPROCESS = 3; // like preprocess()
+ # 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.
+ 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;
var $mSubstWords, $mImageParams, $mImageParamsMagicArray, $mStripList, $mMarkerIndex;
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
- 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
-
- /**#@-*/
+ 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->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' ) ) {
wfProfileOut( __METHOD__ );
}
- function setOutputType( $ot ) {
- $this->mOutputType = $ot;
- // Shortcut alias
- $this->ot = array(
- 'html' => $ot == self::OT_HTML,
- 'wiki' => $ot == self::OT_WIKI,
- 'pre' => $ot == self::OT_PREPROCESS,
- 'plain' => $ot == self::OT_PLAIN,
- );
- }
-
- /**
- * Set the context title
- */
- function setTitle( $t ) {
- if ( !$t || $t instanceof FakeTitle ) {
- $t = Title::newFromText( 'NO TITLE' );
- }
-
- if ( strval( $t->getFragment() ) !== '' ) {
- # Strip the fragment to avoid various odd effects
- $this->mTitle = clone $t;
- $this->mTitle->setFragment( '' );
- } else {
- $this->mTitle = $t;
- }
- }
-
- /**
- * Accessor for mUniqPrefix.
- *
- * @public
- */
- function uniqPrefix() {
- if ( !isset( $this->mUniqPrefix ) ) {
- // @todo Fixme: this is probably *horribly wrong*
- // LanguageConverter seems to want $wgParser's uniqPrefix, however
- // if this is called for a parser cache hit, the parser may not
- // have ever been initialized in the first place.
- // Not really sure what the heck is supposed to be going on here.
- return '';
- // throw new MWException( "Accessing uninitialized mUniqPrefix" );
- }
- return $this->mUniqPrefix;
- }
-
/**
* Convert wikitext to HTML
* Do not call this function recursively.
* 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;
$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 );
$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() );
+ /**
+ * 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 );
$uniq_prefix = $this->mUniqPrefix;
$matches = array();
$elements = array_keys( $this->mTransparentTagHooks );
- $text = self::extractTagsAndParams( $elements, $text, $matches, $uniq_prefix );
+ $text = $this->extractTagsAndParams( $elements, $text, $matches, $uniq_prefix );
foreach ( $matches as $marker => $data ) {
list( $element, $content, $params, $tag ) = $data;
$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
$PFreport = "Expensive parser function count: {$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 ) {
/**
* Process the wikitext for the ?preload= feature. (bug 5210)
*
- * <noinclude>, <includeonly> etc. are parsed as for template transclusion,
+ * <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
+ # Parser (re)initialisation
$this->clearState();
$this->setOutputType( self::OT_PLAIN );
+ $options->resetUsage();
$this->mOptions = $options;
- $this->setTitle( $title );
+ $this->setTitle( $title );
$flags = PPFrame::NO_ARGS | PPFrame::NO_TEMPLATES;
$dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION );
* @private
* @static
*/
- function getRandomString() {
+ static private function getRandomString() {
return dechex( mt_rand( 0, 0x7fffffff ) ) . dechex( mt_rand( 0, 0x7fffffff ) );
}
- function &getTitle() { return $this->mTitle; }
- function getOptions() { return $this->mOptions; }
- function getRevisionId() { return $this->mRevisionId; }
- function getOutput() { return $this->mOutput; }
- function nextLinkID() { return $this->mLinkID++; }
+ /**
+ * Accessor for mUniqPrefix.
+ *
+ * @return String
+ */
+ public function uniqPrefix() {
+ if ( !isset( $this->mUniqPrefix ) ) {
+ # @todo Fixme: this is probably *horribly wrong*
+ # LanguageConverter seems to want $wgParser's uniqPrefix, however
+ # if this is called for a parser cache hit, the parser may not
+ # have ever been initialized in the first place.
+ # Not really sure what the heck is supposed to be going on here.
+ return '';
+ # throw new MWException( "Accessing uninitialized mUniqPrefix" );
+ }
+ return $this->mUniqPrefix;
+ }
+
+ /**
+ * Set the context title
+ */
+ function setTitle( $t ) {
+ if ( !$t || $t instanceof FakeTitle ) {
+ $t = Title::newFromText( 'NO TITLE' );
+ }
+
+ if ( strval( $t->getFragment() ) !== '' ) {
+ # Strip the fragment to avoid various odd effects
+ $this->mTitle = clone $t;
+ $this->mTitle->setFragment( '' );
+ } else {
+ $this->mTitle = $t;
+ }
+ }
+
+ /**
+ * Accessor for the Title object
+ *
+ * @return Title object
+ */
+ function getTitle() {
+ return $this->mTitle;
+ }
+
+ /**
+ * Accessor/mutator for the Title object
+ *
+ * @param $x New Title object or null to just get the current one
+ * @return Title object
+ */
+ function Title( $x = null ) {
+ return wfSetVar( $this->mTitle, $x );
+ }
+
+ /**
+ * Set the output type
+ *
+ * @param $ot Integer: new value
+ */
+ function setOutputType( $ot ) {
+ $this->mOutputType = $ot;
+ # Shortcut alias
+ $this->ot = array(
+ 'html' => $ot == self::OT_HTML,
+ 'wiki' => $ot == self::OT_WIKI,
+ 'pre' => $ot == self::OT_PREPROCESS,
+ 'plain' => $ot == self::OT_PLAIN,
+ );
+ }
+
+ /**
+ * Accessor/mutator for the output type
+ *
+ * @param $x New value or null to just get the current one
+ * @return Integer
+ */
+ function OutputType( $x = null ) {
+ return wfSetVar( $this->mOutputType, $x );
+ }
+
+ /**
+ * Get the ParserOutput object
+ *
+ * @return ParserOutput object
+ */
+ function getOutput() {
+ return $this->mOutput;
+ }
+
+ /**
+ * Get the ParserOptions object
+ *
+ * @return ParserOptions object
+ */
+ function getOptions() {
+ return $this->mOptions;
+ }
+
+ /**
+ * Accessor/mutator for the ParserOptions object
+ *
+ * @param $x New value or null to just get the current one
+ * @return Current ParserOptions object
+ */
+ function Options( $x = null ) {
+ return wfSetVar( $this->mOptions, $x );
+ }
+
+ function nextLinkID() {
+ return $this->mLinkID++;
+ }
function getFunctionLang() {
global $wgLang, $wgContLang;
/**
* 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: extracted tags
* @param $uniq_prefix
+ * @return String: stripped text
*
- * @public
* @static
*/
- function extractTagsAndParams( $elements, $text, &$matches, $uniq_prefix = '' ) {
+ public function extractTagsAndParams( $elements, $text, &$matches, $uniq_prefix = '' ) {
static $n = 1;
$stripped = '';
$matches = array();
break;
}
if ( count( $p ) > 5 ) {
- // comment
+ # comment
$element = $p[4];
$attributes = '';
$close = '';
$inside = $p[5];
} else {
- // tag
+ # tag
$element = $p[1];
$attributes = $p[2];
$close = $p[3];
$stripped .= $marker;
if ( $close === '/>' ) {
- // Empty element tag, <tag />
+ # Empty element tag, <tag />
$content = null;
$text = $inside;
$tail = null;
*/
function doTableStuff( $text ) {
wfProfileIn( __METHOD__ );
-
+
$lines = StringUtils::explode( "\n", $text );
$out = '';
- $td_history = array(); // Is currently a td tag open?
- $last_tag_history = array(); // Save history of last lag activated (td, th or caption)
- $tr_history = array(); // Is currently a tr tag open?
- $tr_attributes = array(); // history of tr attributes
- $has_opened_tr = array(); // Did this table open a <tr> element?
- $indent_level = 0; // indent level of the table
+ $td_history = array(); # Is currently a td tag open?
+ $last_tag_history = array(); # Save history of last lag activated (td, th or caption)
+ $tr_history = array(); # Is currently a tr tag open?
+ $tr_attributes = array(); # history of tr attributes
+ $has_opened_tr = array(); # Did this table open a <tr> element?
+ $indent_level = 0; # indent level of the table
+
+ $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
+ # 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 );
+ # 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}";
+ $line = "<$tr_tag><$td_tag></$td_tag></$tr_tag>{$line}";
}
if ( array_pop( $tr_history ) ) {
- $line = "</tr>{$line}";
+ $line = "</$tr_tag>{$line}";
}
if ( array_pop( $td_history ) ) {
array_pop( $tr_attributes );
$outLine = $line . str_repeat( '</dd></dl>' , $indent_level );
} elseif ( substr( $line , 0 , 2 ) === '|-' ) {
- // Now we have a table row
+ # 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 );
array_push( $has_opened_tr , true );
if ( array_pop( $tr_history ) ) {
- $line = '</tr>';
+ $line = "</$tr_tag>";
}
if ( array_pop( $td_history ) ) {
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
+ # This might be cell elements, td, th or captions
if ( substr( $line , 0 , 2 ) === '|+' ) {
$first_character = '+';
$line = substr( $line , 1 );
$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
+ # 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";
+ $previous = "<$tr_tag{$tr_after}>\n";
}
array_push( $tr_history , true );
array_push( $tr_attributes , '' );
$last_tag = array_pop( $last_tag_history );
if ( array_pop( $td_history ) ) {
- $previous = "</{$last_tag}>{$previous}";
+ $previous = "</{$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 );
- // A cell could contain both parameters and data
+ # 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}";
+ if ($extra_td_attribs) $attributes = Sanitizer::collapseTagAttributes( $extra_td_attribs );
+ $cell = "{$previous}<{$last_tag}{$attributes}>{$cell}";
} elseif ( count( $cell_data ) == 1 ) {
- $cell = "{$previous}<{$last_tag}>{$cell_data[0]}";
+ if ($extra_td_attribs) $attributes = Sanitizer::collapseTagAttributes( $extra_td_attribs );
+ $cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[0]}";
} else {
$attributes = $this->mStripState->unstripBoth( $cell_data[0] );
- $attributes = Sanitizer::fixTagAttributes( $attributes , $last_tag );
+
+ $attr = Sanitizer::decodeTagAttributes( $attributes );
+ if ($convert_style) $attr['style'] = Sanitizer::styleFromAttributes( $attr );
+ $attr = Sanitizer::validateTagAttributes( $attr, $last_tag );
+ $attributes = Sanitizer::collapseTagAttributes( $attr, $extra_td_attribs );
+
$cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[1]}";
}
$out .= $outLine . "\n";
}
- // Closing open td, tr && table
+ # Closing open td, tr && table
while ( count( $td_history ) > 0 ) {
if ( array_pop( $td_history ) ) {
- $out .= "</td>\n";
+ $out .= "</$td_tag>\n";
}
if ( array_pop( $tr_history ) ) {
- $out .= "</tr>\n";
+ $out .= "</$tr_tag>\n";
}
if ( !array_pop( $has_opened_tr ) ) {
- $out .= "<tr><td></td></tr>\n" ;
+ $out .= "<$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 is provided, then use $frame for replacing any variables
if ( $frame ) {
- // use frame depth to infer how include/noinclude tags should be handled
- // depth=0 means this is the top-level document; otherwise it's an included document
+ # use frame depth to infer how include/noinclude tags should be handled
+ # depth=0 means this is the top-level document; otherwise it's an included document
if ( !$frame->depth ) {
$flag = 0;
} else {
$flag = Parser::PTD_FOR_INCLUSION;
+ }
$dom = $this->preprocessToDom( $text, $flag );
$text = $frame->expand( $dom );
- }
} else {
- // if $frame is not provided, then use old-style replaceVariables
+ # 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 );
$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
return $this->makeFreeExternalLink( $m[0] );
} elseif ( isset( $m[4] ) && $m[4] !== '' ) {
# RFC or PMID
- $CssClass = '';
if ( substr( $m[0], 0, 3 ) === 'RFC' ) {
$keyword = 'RFC';
$urlmsg = 'rfcurl';
throw new MWException( __METHOD__.': unrecognised match type "' .
substr( $m[0], 0, 20 ) . '"' );
}
- $url = wfMsg( $urlmsg, $id);
- $sk = $this->mOptions->getSkin();
+ $url = wfMsgForContent( $urlmsg, $id);
+ $sk = $this->mOptions->getSkin( $this->mTitle );
$la = $sk->getExternalLinkAttributes( "external $CssClass" );
return "<a href=\"{$url}\"{$la}>{$keyword} {$id}</a>";
} elseif ( isset( $m[5] ) && $m[5] !== '' ) {
global $wgContLang;
wfProfileIn( __METHOD__ );
- $sk = $this->mOptions->getSkin();
+ $sk = $this->mOptions->getSkin( $this->mTitle );
$trail = '';
# The characters '<' and '>' (which were escaped by
# First, do some preliminary work. This may shift some apostrophes from
# being mark-up to being text. It also counts the number of occurrences
# of bold and italics mark-ups.
- $i = 0;
$numbold = 0;
$numitalics = 0;
- foreach ( $arr as $r ) {
+ for ( $i = 0; $i < count( $arr ); $i++ ) {
if ( ( $i % 2 ) == 1 ) {
# If there are ever four apostrophes, assume the first is supposed to
# be text, and the remaining three constitute mark-up for bold text.
}
# Count the number of occurrences of bold and italics mark-ups.
# We are not counting sequences of five apostrophes.
- if ( strlen( $arr[$i] ) == 2 ) {
+ if ( strlen( $arr[$i] ) == 2 ) {
$numitalics++;
- } elseif ( strlen( $arr[$i] ) == 3 ) {
+ } elseif ( strlen( $arr[$i] ) == 3 ) {
$numbold++;
- } elseif ( strlen( $arr[$i] ) == 5 ) {
- $numitalics++;
+ } elseif ( strlen( $arr[$i] ) == 5 ) {
+ $numitalics++;
$numbold++;
}
}
- $i++;
}
# If there is an odd number of both bold and italics, it is likely
$firstspace = $i;
}
} elseif ( $x2 === ' ') {
- if ( $firstsingleletterword == -1 ) {
+ if ( $firstsingleletterword == -1 ) {
$firstsingleletterword = $i;
}
} else {
global $wgContLang;
wfProfileIn( __METHOD__ );
- $sk = $this->mOptions->getSkin();
+ $sk = $this->mOptions->getSkin( $this->mTitle );
$bits = preg_split( $this->mExtLinkBracketedRegex, $text, -1, PREG_SPLIT_DELIM_CAPTURE );
$s = array_shift( $bits );
# 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();
/**
* 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];
}
}
* @private
*/
function maybeMakeExternalImage( $url ) {
- $sk = $this->mOptions->getSkin();
+ $sk = $this->mOptions->getSkin( $this->mTitle );
$imagesfrom = $this->mOptions->getAllowExternalImagesFrom();
$imagesexception = !empty( $imagesfrom );
$text = false;
/**
* Process [[ ]] wikilinks
- * @return processed text
+ * @return String: processed text
*
* @private
*/
$e1_img = "/^([{$tc}]+)\\|(.*)\$/sD";
}
- $sk = $this->mOptions->getSkin();
+ $sk = $this->mOptions->getSkin( $this->mTitle );
$holders = new LinkHolderArray( $this );
# split the entire text string on occurences of [[
}
if ( $wgContLang->hasVariants() ) {
- $selflink = $wgContLang->convertLinkToAllVariants( $this->mTitle->getPrefixedText() );
+ $selflink = $wgContLang->autoConvertToAllVariants( $this->mTitle->getPrefixedText() );
} else {
$selflink = array( $this->mTitle->getPrefixedText() );
}
# [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
# the real problem is with the $e1 regex
# See bug 1300.
- #
+ #
# Still some problems for cases where the ] is meant to be outside punctuation,
# and no image is in sight. See bug 2095.
- #
+ #
if ( $text !== '' &&
substr( $m[3], 0, 1 ) === ']' &&
strpos( $text, '[' ) !== false
$wasblank = ( $text == '' );
if ( $wasblank ) {
$text = $link;
+ } else {
+ # Bug 4598 madness. Handle the quotes only if they come from the alternate part
+ # [[Lista d''e paise d''o munno]] -> <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
wfProfileIn( __METHOD__."-always_known" );
# Some titles, such as valid special pages or files in foreign repos, should
# be shown as bluelinks even though they're not included in the page table
- #
+ #
# FIXME: isAlwaysKnown() can be expensive for file links; we should really do
# batch file existence checks for NS_FILE and NS_MEDIA
if ( $iw == '' && $nt->isAlwaysKnown() ) {
* breaking URLs in the following text without breaking trails on the
* wiki links, it's been made into a horrible function.
*
- * @param Title $nt
- * @param string $text
- * @param string $query
- * @param string $trail
- * @param string $prefix
- * @return string HTML-wikitext mix oh yuck
+ * @param $nt Title
+ * @param $text String
+ * @param $query String
+ * @param $trail String
+ * @param $prefix String
+ * @return String: HTML-wikitext mix oh yuck
*/
function makeKnownLinkHolder( $nt, $text = '', $query = '', $trail = '', $prefix = '' ) {
list( $inside, $trail ) = Linker::splitTrail( $trail );
- $sk = $this->mOptions->getSkin();
- // FIXME: use link() instead of deprecated makeKnownLinkObj()
+ $sk = $this->mOptions->getSkin( $this->mTitle );
+ # FIXME: use link() instead of deprecated makeKnownLinkObj()
$link = $sk->makeKnownLinkObj( $nt, $text, $query, $inside, $prefix );
return $this->armorLinks( $link ) . $trail;
}
* 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
*/
* Used by doBlockLevels()
* @private
*/
- /* private */ function closeParagraph() {
+ function closeParagraph() {
$result = '';
if ( $this->mLastSection != '' ) {
$result = '</' . $this->mLastSection . ">\n";
$this->mLastSection = '';
return $result;
}
+
/**
* getCommon() returns the length of the longest common substring
* of both arguments, starting at the beginning of both.
- * @private
- */
+ * @private
+ */
function getCommon( $st1, $st2 ) {
$fl = strlen( $st1 );
$shorter = strlen( $st2 );
- if ( $fl < $shorter ) {
- $shorter = $fl;
+ if ( $fl < $shorter ) {
+ $shorter = $fl;
}
for ( $i = 0; $i < $shorter; ++$i ) {
- if ( $st1{$i} != $st2{$i} ) {
- break;
+ if ( $st1{$i} != $st2{$i} ) {
+ break;
}
}
return $i;
}
+
/**
* These next three functions open, continue, and close the list
* element appropriate to the prefix character passed into them.
- * @private
+ * @private
*/
function openList( $char ) {
$result = $this->closeParagraph();
- if ( '*' === $char ) {
- $result .= '<ul><li>';
- } elseif ( '#' === $char ) {
- $result .= '<ol><li>';
- } elseif ( ':' === $char ) {
- $result .= '<dl><dd>';
+ 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>';
+ /**
+ * 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 '<!-- ERR 2 -->';
}
- /* private */ function closeList( $char ) {
- if ( '*' === $char ) {
- $text = '</li></ul>';
- } elseif ( '#' === $char ) {
- $text = '</li></ol>';
+ /**
+ * 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;
} 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
*/
# Parsing through the text line by line. The main thing
# happening here is handling of block-level elements p, pre,
# and making lists from lines starting with * # : etc.
- #
+ #
$textLines = StringUtils::explode( "\n", $text );
$lastPrefix = $output = '';
$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', $oLine );
- // If not in a <pre> element, scan for and figure out what prefixes are there.
+ # If not in a <pre> element, scan for and figure out what prefixes are there.
if ( !$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;
}
}
} elseif ( $prefixLength || $lastPrefixLength ) {
- // We need to open or close prefixes, or both.
+ # 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.
+ # 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 );
$lastPrefix = $prefix2;
}
- // If we have no prefixes, go to paragraph mode.
+ # 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|'.
if ( $preOpenMatch and !$preCloseMatch ) {
$this->mInPre = true;
}
- if ( $closematch ) {
- $inBlockElem = false;
- } else {
- $inBlockElem = true;
- }
+ $inBlockElem = !$closematch;
} elseif ( !$inBlockElem && !$this->mInPre ) {
if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' || trim( $t ) != '' ) ) {
- // pre
+ # pre
if ( $this->mLastSection !== 'pre' ) {
$paragraphStack = false;
$output .= $this->closeParagraph().'<pre>';
}
$t = substr( $t, 1 );
} else {
- // paragraph
- if ( trim( $t ) == '' ) {
+ # paragraph
+ if ( trim( $t ) === '' ) {
if ( $paragraphStack ) {
$output .= $paragraphStack.'<br />';
$paragraphStack = false;
}
wfProfileOut( __METHOD__."-paragraph" );
}
- // somewhere above we forget to get out of pre block (bug 785)
+ # somewhere above we forget to get out of pre block (bug 785)
if ( $preCloseMatch && $this->mInPre ) {
$this->mInPre = false;
}
/**
* Split up a string on ':', ignoring any occurences inside tags
* to prevent illegal overlapping.
- * @param string $str the string to split
- * @param string &$before set to everything before the ':'
- * @param string &$after set to everything after the ':'
- * return string the position of the ':', or false if none found
+ *
+ * @param $str String: the string to split
+ * @param &$before String: set to everything before the ':'
+ * @param &$after String: set to everything after the ':'
+ * return String: the position of the ':', or false if none found
*/
function findColonNoLinks( $str, &$before, &$after ) {
wfProfileIn( __METHOD__ );
$pos = strpos( $str, ':' );
if ( $pos === false ) {
- // Nothing to find!
+ # Nothing to find!
wfProfileOut( __METHOD__ );
return false;
}
$lt = strpos( $str, '<' );
if ( $lt === false || $lt > $pos ) {
- // Easy; no tag nesting to worry about
+ # 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!
+ # 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
+ # Nothing else interesting
wfProfileOut( __METHOD__ );
return false;
}
$lt = strpos( $str, '<', $i );
if ( $stack === 0 ) {
if ( $lt === false || $colon < $lt ) {
- // We found it!
+ # We found it!
$before = substr( $str, 0, $colon );
$after = substr( $str, $colon + 1 );
wfProfileOut( __METHOD__ );
}
}
if ( $lt === false ) {
- // Nothing else interesting to find; abort!
- // We're nested, but there's no close tags left. Abort!
+ # 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>
+ case 3: # self::COLON_STATE_CLOSETAG:
+ # In a </tag>
if ( $c === ">" ) {
$stack--;
if ( $stack < 0 ) {
break;
case self::COLON_STATE_TAGSLASH:
if ( $c === ">" ) {
- // Yes, a self-closed tag <blah/>
+ # 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:
+ case 5: # self::COLON_STATE_COMMENT:
if ( $c === "-" ) {
$state = self::COLON_STATE_COMMENTDASH;
}
* @private
*/
function getVariableValue( $index, $frame=false ) {
- global $wgContLang, $wgSitename, $wgServer, $wgServerName;
- global $wgScriptPath, $wgStylePath;
+ global $wgContLang, $wgSitename, $wgServer;
+ global $wgArticlePath, $wgScriptPath, $wgStylePath;
/**
* Some of these require message or data lookups and can be
$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();
$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':
case 'currentversion':
$value = SpecialVersion::getVersion();
break;
+ case 'articlepath':
+ return $wgArticlePath;
case 'sitename':
return $wgSitename;
case 'server':
return $wgServer;
case 'servername':
- return $wgServerName;
+ wfSuppressWarnings(); # May give an E_WARNING in PHP < 5.3.3
+ $serverName = parse_url( $wgServer, PHP_URL_HOST );
+ wfRestoreWarnings();
+ return $serverName ? $serverName : $wgServer;
case 'scriptpath':
return $wgScriptPath;
case 'stylepath':
case 'directionmark':
return $wgContLang->getDirMark();
case 'contentlanguage':
- global $wgContLanguageCode;
- return $wgContLanguageCode;
+ global $wgLanguageCode;
+ return $wgLanguageCode;
default:
$ret = null;
if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) {
}
/**
- * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
+ * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
*
* @private
*/
* 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.
*
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 ) {
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;
* Warn the user when a parser limitation is reached
* Will warn at most once the user per limitation type
*
- * @param string $limitationType, should be one of:
- * 'expensive-parserfunction' (corresponding messages:
- * 'expensive-parserfunction-warning',
+ * @param $limitationType String: should be one of:
+ * 'expensive-parserfunction' (corresponding messages:
+ * 'expensive-parserfunction-warning',
* 'expensive-parserfunction-category')
- * 'post-expand-template-argument' (corresponding messages:
- * 'post-expand-template-argument-warning',
+ * 'post-expand-template-argument' (corresponding messages:
+ * 'post-expand-template-argument-warning',
* 'post-expand-template-argument-category')
- * 'post-expand-template-inclusion' (corresponding messages:
- * 'post-expand-template-inclusion-warning',
+ * 'post-expand-template-inclusion' (corresponding messages:
+ * 'post-expand-template-inclusion-warning',
* 'post-expand-template-inclusion-category')
- * @params int $current, $max When an explicit limit has been
+ * @param $current Current value
+ * @param $max Maximum allowed, when an explicit limit has been
* exceeded, provide the values (optional)
*/
function limitationWarn( $limitationType, $current=null, $max=null) {
- // 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 ) {
if ( $id !== false ) {
$text = $this->getVariableValue( $id, $frame );
if ( MagicWord::getCacheTTL( $id ) > -1 ) {
- $this->mOutput->mContainsOldMagic = true;
+ $this->mOutput->updateCacheExpiry( MagicWord::getCacheTTL( $id ) );
}
$found = true;
}
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;
$limit = $this->mOptions->getMaxTemplateDepth();
if ( $frame->depth >= $limit ) {
$found = true;
- $text = '<span class="error">'
- . wfMsgForContent( 'parser-template-recursion-depth-warning', $limit )
+ $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 ) ) {
$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;
}
# Escape nowiki-style return values
$text = wfEscapeWikiText( $text );
} elseif ( is_string( $text )
- && !$piece['lineStart']
+ && !$piece['lineStart']
&& preg_match( '/^(?:{\\||:|;|#|\*)/', $text ) )
{
# Bug 529: if the template begins with a table or block-level
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 ( $text === false ) {
if ( !$title->equals( $cacheTitle ) ) {
$this->mTplRedirCache[$cacheTitle->getPrefixedDBkey()] =
- array( $title->getNamespace(),$cdb = $title->getDBkey() );
+ array( $title->getNamespace(), $cdb = $title->getDBkey() );
}
return array( $dom, $title );
* 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;
$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 ) {
}
$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 there is no current revision, there is no page
if ( $id === false && !$rev ) {
$linkCache = LinkCache::singleton();
$linkCache->addBadLinkObj( $title );
if ( $text === false ) {
break;
}
- // Redirect?
+ # Redirect?
$finalTitle = $title;
$title = Title::newFromRedirect( $text );
}
global $wgEnableScaryTranscluding;
if ( !$wgEnableScaryTranscluding ) {
- return wfMsg('scarytranscludedisabled');
+ return wfMsgForContent('scarytranscludedisabled');
}
$url = $title->getFullUrl( "action=$action" );
if ( strlen( $url ) > 255 ) {
- return wfMsg( 'scarytranscludetoolong' );
+ return wfMsgForContent( 'scarytranscludetoolong' );
}
return $this->fetchScaryTemplateMaybeFromCache( $url );
}
$text = Http::get( $url );
if ( !$text ) {
- return wfMsg( 'scarytranscludefailed', $url );
+ return wfMsgForContent( 'scarytranscludefailed', $url );
}
$dbw = wfGetDB( DB_MASTER );
* 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'] );
}
if ( is_array( $output ) ) {
- // Extract flags to local scope (to override $markerType)
+ # Extract flags to local scope (to override $markerType)
$flags = $output;
$output = $flags[0];
unset( $flags[0] );
/**
* 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;
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 ) ) {
$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() ) {
$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 ) {
$cat = wfMsgForContent( $msg );
* It loops through all headlines, collects the necessary data, then splits up the
* string and re-inserts the newly formatted headlines.
*
- * @param string $text
- * @param string $origText Original, untouched wikitext
- * @param boolean $isMain
+ * @param $text String
+ * @param $origText String: original, untouched wikitext
+ * @param $isMain Boolean
* @private
*/
function formatHeadings( $text, $origText, $isMain=true ) {
global $wgMaxTocLevel, $wgContLang, $wgHtml5, $wgExperimentalHtmlIds;
$doNumberHeadings = $this->mOptions->getNumberHeadings();
- $showEditLink = $this->mOptions->getEditSection();
-
- // Do not call quickUserCan unless necessary
- if ( $showEditLink && !$this->mTitle->quickUserCan( 'edit' ) ) {
- $showEditLink = 0;
- }
-
+
# Inhibit editsection links if requested in the page
- if ( isset( $this->mDoubleUnderscores['noeditsection'] ) || $this->mOptions->getIsPrintable() ) {
+ if ( isset( $this->mDoubleUnderscores['noeditsection'] ) ) {
$showEditLink = 0;
+ } else {
+ $showEditLink = $this->mOptions->getEditSection();
}
# Get all headlines for numbering them and adding funky stuff like [edit]
}
# We need this to perform operations on the HTML
- $sk = $this->mOptions->getSkin();
+ $sk = $this->mOptions->getSkin( $this->mTitle );
# headline counter
$headlineCount = 0;
$head = array();
$sublevelCount = array();
$levelCount = array();
- $toclevel = 0;
$level = 0;
$prevlevel = 0;
$toclevel = 0;
$node = $root->getFirstChild();
$byteOffset = 0;
$tocraw = array();
+ $refers = array();
foreach ( $matches[3] as $headline ) {
$isTemplate = false;
if ( $toclevel ) {
$prevlevel = $level;
- $prevtoclevel = $toclevel;
}
$level = $matches[1][$headlineCount];
# 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 ( $wgHtml5 && $wgExperimentalHtmlIds ) {
# For reverse compatibility, provide an id that's
# HTML4-compatible, like we used to.
- #
+ #
# It may be worth noting, academically, that it's possible for
# the legacy anchor to conflict with a non-legacy headline
# anchor on the page. In this case likely the "correct" thing
'noninitial' );
}
- # HTML names must be case-insensitively unique (bug 10721). FIXME:
- # Does this apply to Unicode characters? Because we aren't
- # handling those here.
+ # HTML names must be case-insensitively unique (bug 10721).
+ # This does not apply to Unicode characters per
+ # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison
+ # FIXME: We may be changing them depending on the current locale.
$arrayKey = strtolower( $safeHeadline );
if ( $legacyHeadline === false ) {
$legacyArrayKey = false;
while ( $node && !$isTemplate ) {
if ( $node->getName() === 'h' ) {
$bits = $node->splitHeading();
- if ( $bits['i'] == $sectionIndex )
+ if ( $bits['i'] == $sectionIndex ) {
break;
+ }
}
$byteOffset += mb_strlen( $this->mStripState->unstripBoth(
$frame->expand( $node, PPFrame::RECOVER_ORIG ) ) );
if ( $isTemplate ) {
# Put a T flag in the section identifier, to indicate to extractSections()
# that sections inside <includeonly> should be counted.
- $editlink = $sk->doEditSectionLink( Title::newFromText( $titleText ), "T-$sectionIndex" );
+ $editlink = $sk->doEditSectionLink( Title::newFromText( $titleText ), "T-$sectionIndex", null, $this->mOptions->getUserLang() );
} else {
- $editlink = $sk->doEditSectionLink( $this->mTitle, $sectionIndex, $headlineHint );
+ $editlink = $sk->doEditSectionLink( $this->mTitle, $sectionIndex, $headlineHint, $this->mOptions->getUserLang() );
}
} else {
$editlink = '';
}
}
- /**
- * 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 );
$ts = date( 'YmdHis', $unixts );
$tzMsg = date( 'T', $unixts ); # might vary on DST changeover!
- # Allow translation of timezones trough wiki. date() can return
+ # Allow translation of timezones through wiki. date() can return
# whatever crap the system uses, localised or not, so we cannot
# ship premade translations.
$key = 'timezone-' . strtolower( trim( $tzMsg ) );
* 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' );
}
}
- // 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;
/**
* 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;
* 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;
/**
* Get all registered function hook identifiers
*
- * @return array
+ * @return Array
*/
function getFunctionHooks() {
return array_keys( $this->mFunctionHooks );
/**
* 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 );
$ig->setParser( $this );
$ig->setHideBadImages();
$ig->setAttributes( Sanitizer::validateTagAttributes( $params, 'table' ) );
- $ig->useSkin( $this->mOptions->getSkin() );
+ $ig->useSkin( $this->mOptions->getSkin( $this->mTitle ) );
$ig->mRevisionId = $this->mRevisionId;
if ( isset( $params['showfilename'] ) ) {
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 ) ) {
# Bogus title. Ignore these so we don't bomb out later.
$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"
# * text-bottom
$parts = StringUtils::explode( "|", $options );
- $sk = $this->mOptions->getSkin();
+ $sk = $this->mOptions->getSkin( $this->mTitle );
# Give extensions a chance to select the file revision for us
$skip = $time = $descQuery = false;
}
# Get the file
- $imagename = $title->getDBkey();
$file = wfFindFile( $title, array( 'time' => $time ) );
# Get parameter map
$handler = $file ? $file->getHandler() : false;
if ( isset( $paramMap[$magicName] ) ) {
list( $type, $paramName ) = $paramMap[$magicName];
- // Special case; width and height come in one variable together
+ # 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),
$params[$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, $holders );
break;
if ( preg_match( "/^($prots)$chars+$/", $value, $m ) ) {
$paramName = 'link-url';
$this->mOutput->addExternalLink( $value );
+ if ( $this->mOptions->getExternalLinkTarget() ) {
+ $params[$type]['link-target'] = $this->mOptions->getExternalLinkTarget();
+ }
$validated = true;
}
} else {
}
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 ) ) );
}
}
# came to also set the caption, ordinary text after the image -- which
# makes no sense, because that just repeats the text multiple times in
# screen readers. It *also* came to set the title attribute.
- #
+ #
# Now that we have an alt attribute, we should not set the alt text to
# equal the caption: that's worse than useless, it just repeats the
# text. This is the framed/thumbnail case. If there's no caption, we
# use the unnamed parameter for alt text as well, just for the time be-
# ing, if the unnamed param is set and the alt param is not.
- #
+ #
# For the future, we need to figure out if we want to tweak this more,
# e.g., introducing a title= parameter for the title; ignoring the un-
# named parameter entirely for images without a caption; adding an ex-
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_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();
$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
+ # 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 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();
*/
public function setDefaultSort( $sort ) {
$this->mDefaultSort = $sort;
+ $this->mOutput->setProperty( 'defaultsort', $sort );
}
/**
* @return string
*/
public function getDefaultSort() {
- global $wgCategoryPrefixedDefaultSortkey;
if ( $this->mDefaultSort !== false ) {
return $this->mDefaultSort;
- } elseif ( $this->mTitle->getNamespace() == NS_CATEGORY ||
- !$wgCategoryPrefixedDefaultSortkey )
- {
- return $this->mTitle->getText();
} else {
- return $this->mTitle->getPrefixedText();
+ return $this->mTitle->getCategorySortkey();
}
}
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
*/
return $text;
}
- function srvus( $text ) {
- return $this->testSrvus( $text, $this->mOutputType );
- }
-
/**
* strip/replaceVariables/unstrip for preprocessor regression testing
*/
$title = Title::newFromText( $title );
}
$this->mTitle = $title;
+ $options->resetUsage();
$this->mOptions = $options;
$this->setOutputType( $outputType );
$text = $this->replaceVariables( $text );
$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
+ # Internal links
while ( ( $start_pos = strpos( $text, '<!--LINK ', $pos ) ) ) {
list( $ns, $trail ) = explode( ':', substr( $text, $start_pos + strlen( '<!--LINK ' ) ), 2 );
$pos = 0;
- // Interwiki links
+ # Interwiki links
while ( ( $start_pos = strpos( $text, '<!--IWLINK ', $pos ) ) ) {
$data = substr( $text, $start_pos );
$key = trim( substr( $data, 0, strpos( $data, '-->' ) ) );
return $data;
}
- function unserialiseHalfParsedText( $data, $intPrefix = null /* Unique identifying prefix */ ) {
+ /**
+ * TODO: document
+ * @param $data Array
+ * @param $intPrefix String unique identifying prefix
+ * @return String
+ */
+ function unserialiseHalfParsedText( $data, $intPrefix = null ) {
if ( !$intPrefix ) {
- $intPrefix = $this->getRandomString();
+ $intPrefix = self::getRandomString();
}
- // First, extract the strip state.
+ # First, extract the strip state.
$stripState = $data['stripstate'];
$this->mStripState->general->merge( $stripState->general );
$this->mStripState->nowiki->merge( $stripState->nowiki );
- // Now, extract the text, and renumber links
+ # Now, extract the text, and renumber links
$text = $data['text'];
$links = $data['linkholder'];
- // Internal...
+ # Internal...
foreach ( $links['internal'] as $ns => $nsLinks ) {
foreach ( $nsLinks as $key => $entry ) {
$newKey = $intPrefix . '-' . $key;
}
}
- // Interwiki...
+ # 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;
}
}