wfRunHooks( 'ParserClearState', array( &$this ) );
}
+ /**
+ * Accessor for mUniqPrefix.
+ *
+ * @access public
+ */
+ function UniqPrefix() {
+ return $this->mUniqPrefix;
+ }
+
/**
* Convert wikitext to HTML
* Do not call this function recursively.
* First pass--just handle <nowiki> sections, pass the rest off
* to internalParse() which does all the real work.
*/
-
- global $wgUseTidy, $wgContLang;
+
+ global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang;
$fname = 'Parser::parse';
wfProfileIn( $fname );
wfRunHooks( 'ParserBeforeTidy', array( &$this, &$text ) );
$text = Sanitizer::normalizeCharReferences( $text );
-
- if ($wgUseTidy) {
+
+ if (($wgUseTidy and $this->mOptions->mTidy) or $wgAlwaysUseTidy) {
$text = Parser::tidy($text);
}
if ( !is_array( $state ) ) {
return $text;
}
-
+
# Must expand in reverse order, otherwise nested tags will be corrupted
foreach( array_reverse( $state, true ) as $tag => $contentDict ) {
if( $tag != 'nowiki' && $tag != 'html' ) {
$fc = substr ( $x , 0 , 1 ) ;
if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
$indent_level = strlen( $matches[1] );
-
+
$attributes = $this->unstripForHTML( $matches[2] );
$t[$k] = str_repeat( '<dl><dd>', $indent_level ) .
# Replace & from obsolete syntax with &.
# All HTML entities will be escaped by makeExternalLink()
- # or maybeMakeExternalImage()
$url = str_replace( '&', '&', $url );
+ # Replace unnecessary URL escape codes with the referenced character
+ # This prevents spammers from hiding links from the filters
+ $url = Parser::replaceUnusualEscapes( $url );
# Process the trail (i.e. everything after this link up until start of the next link),
# replacing any non-bracketed links
$trail = $this->replaceFreeExternalLinks( $trail );
-
# 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
# This was changed in August 2004
$s .= $sk->makeExternalLink( $url, $text, false, $linktype ) . $dtrail . $trail;
+
+ # Register link in the output object
+ $this->mOutput->addExternalLink( $url );
}
wfProfileOut( $fname );
# All HTML entities will be escaped by makeExternalLink()
# or maybeMakeExternalImage()
$url = str_replace( '&', '&', $url );
+ # Replace unnecessary URL escape codes with their equivalent characters
+ $url = Parser::replaceUnusualEscapes( $url );
# Is this an external image?
$text = $this->maybeMakeExternalImage( $url );
if ( $text === false ) {
# Not an image, make a link
$text = $sk->makeExternalLink( $url, $wgContLang->markNoConversion($url), true, 'free' );
+ # Register it in the output object
+ $this->mOutput->addExternalLink( $url );
}
$s .= $text . $trail;
} else {
return $s;
}
+ /**
+ * Replace unusual URL escape codes with their equivalent characters
+ * @param string
+ * @return string
+ * @static
+ */
+ function replaceUnusualEscapes( $url ) {
+ return preg_replace_callback( '/%[0-9A-Fa-f]{2}/',
+ array( 'Parser', 'replaceUnusualEscapesCallback' ), $url );
+ }
+
+ /**
+ * Callback function used in replaceUnusualEscapes().
+ * Replaces unusual URL escape codes with their equivalent character
+ * @static
+ * @access private
+ */
+ function replaceUnusualEscapesCallback( $matches ) {
+ $char = urldecode( $matches[0] );
+ $ord = ord( $char );
+ // 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
+ return $char;
+ } else {
+ // Yes, leave it escaped
+ return $matches[0];
+ }
+ }
+
/**
* make an image if it's allowed, either through the global
* option or through the exception
}
return $retVal;
}
-
+
/**
* Render a forced-blue link inline; protect against double expansion of
* URLs if we're in a mode that prepends full URL prefixes to internal links.
$link = $sk->makeKnownLinkObj( $nt, $text, $query, $inside, $prefix );
return $this->armorLinks( $link ) . $trail;
}
-
+
/**
* Insert a NOPARSE hacky thing into any inline links in a chunk that's
* going to go through further parsing steps before inline URL expansion.
return $varCache[$index] = $wgContLang->getMonthAbbreviation( date( 'n', $ts ) );
case MAG_CURRENTDAY:
return $varCache[$index] = $wgContLang->formatNum( date( 'j', $ts ) );
+ case MAG_CURRENTDAY2:
+ return $varCache[$index] = $wgContLang->formatNum( date( 'd', $ts ) );
case MAG_PAGENAME:
return $this->mTitle->getText();
case MAG_PAGENAMEE:
case MAG_CURRENTTIME:
return $varCache[$index] = $wgContLang->time( wfTimestamp( TS_MW, $ts ), false, false );
case MAG_CURRENTWEEK:
- return $varCache[$index] = $wgContLang->formatNum( date( 'W', $ts ) );
+ // @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
+ // int to remove the padding
+ return $varCache[$index] = $wgContLang->formatNum( (int)date( 'W', $ts ) );
case MAG_CURRENTDOW:
return $varCache[$index] = $wgContLang->formatNum( date( 'w', $ts ) );
case MAG_NUMBEROFARTICLES:
if ($lastOpeningBrace >= 0) {
$pos = strpos ($text, $openingBraceStack[$lastOpeningBrace]['braceEnd'], $i);
-
+
if (false !== $pos && (-1 == $nextPos || $pos < $nextPos)){
$rule = null;
$nextPos = $pos;
}
$pos = strpos ($text, '|', $i);
-
+
if (false !== $pos && (-1 == $nextPos || $pos < $nextPos)){
$rule = null;
$nextPos = $pos;
$matchingCallback = $fn;
}
}
-
+
if ($matchingCount == 0) {
$i += $count - 1;
continue;
$pieceStart = $openingBraceStack[$lastOpeningBrace]['startAt'] - $matchingCount;
$pieceEnd = $i + $matchingCount;
-
+
if( is_callable( $matchingCallback ) ) {
$cbArgs = array (
'text' => substr($text, $pieceStart, $pieceEnd - $pieceStart),
}
else
$openingBraceStack[$lastOpeningBrace]['parts'][] = substr($text, $openingBraceStack[$lastOpeningBrace]['partStart'], $i - $openingBraceStack[$lastOpeningBrace]['partStart']);
-
+
$openingBraceStack[$lastOpeningBrace]['partStart'] = $i + 1;
}
}
$found = false;
$nowiki = false;
$noparse = false;
+ $replaceHeadings = false;
+ $isHTML = false;
$title = NULL;
# Check for NS: (namespace expansion)
$mwNs = MagicWord::get( MAG_NS );
if ( $mwNs->matchStartAndRemove( $part1 ) ) {
- if ( intval( $part1 ) ) {
+ if ( intval( $part1 ) || $part1 == "0" ) {
$text = $linestart . $wgContLang->getNsText( intval( $part1 ) );
$found = true;
} else {
$mwLocalE =& MagicWord::get( MAG_LOCALURLE );
$mwFull =& MagicWord::get( MAG_FULLURL );
$mwFullE =& MagicWord::get( MAG_FULLURLE );
-
+
if ( $mwLocal->matchStartAndRemove( $part1 ) ) {
$func = 'getLocalURL';
}
# Load from database
- $replaceHeadings = false;
- $isHTML = false;
$lastPathLevel = $this->mTemplatePath;
if ( !$found ) {
$ns = NS_TEMPLATE;
}
$title = Title::newFromText( $part1, $ns );
- if ($title) {
- $interwiki = Title::getInterwikiLink($title->getInterwiki());
- if ($interwiki != '' && $title->isTrans()) {
- return $this->scarytransclude($title, $interwiki);
- }
- }
-
- if ( !is_null( $title ) && !$title->isExternal() ) {
- # Check for excessive inclusion
- $dbk = $title->getPrefixedDBkey();
- if ( $this->incrementIncludeCount( $dbk ) ) {
- if ( $title->getNamespace() == NS_SPECIAL && $this->mOptions->getAllowSpecialInclusion() ) {
- # Capture special page output
- $text = SpecialPage::capturePath( $title );
- if ( is_string( $text ) ) {
- $found = true;
- $noparse = true;
- $isHTML = true;
- $this->disableCache();
- }
- } else {
- $article = new Article( $title );
- $articleContent = $article->fetchContent(0, false);
- if ( $articleContent !== false ) {
- $found = true;
- $text = $articleContent;
- $replaceHeadings = true;
+ if ( !is_null( $title ) ) {
+ if ( !$title->isExternal() ) {
+ # Check for excessive inclusion
+ $dbk = $title->getPrefixedDBkey();
+ if ( $this->incrementIncludeCount( $dbk ) ) {
+ if ( $title->getNamespace() == NS_SPECIAL && $this->mOptions->getAllowSpecialInclusion() ) {
+ # Capture special page output
+ $text = SpecialPage::capturePath( $title );
+ if ( is_string( $text ) ) {
+ $found = true;
+ $noparse = true;
+ $isHTML = true;
+ $this->disableCache();
+ }
+ } else {
+ $articleContent = $this->fetchTemplate( $title );
+ if ( $articleContent !== false ) {
+ $found = true;
+ $text = $articleContent;
+ $replaceHeadings = true;
+ }
}
- # Register a template reference whether or not the template exists
- $this->mOutput->addTemplate( $title, $article->getID() );
}
- }
- # If the title is valid but undisplayable, make a link to it
- if ( $this->mOutputType == OT_HTML && !$found ) {
- $text = '[['.$title->getPrefixedText().']]';
- $found = true;
- }
+ # If the title is valid but undisplayable, make a link to it
+ if ( $this->mOutputType == OT_HTML && !$found ) {
+ $text = '[['.$title->getPrefixedText().']]';
+ $found = true;
+ }
- # Template cache array insertion
- if( $found ) {
- $this->mTemplates[$part1] = $text;
- $text = $linestart . $text;
+ # Template cache array insertion
+ if( $found ) {
+ $this->mTemplates[$part1] = $text;
+ $text = $linestart . $text;
+ }
+ } elseif ( $title->isTrans() ) {
+ // Interwiki transclusion
+ if ( $this->mOutputType == OT_HTML ) {
+ $text = $this->interwikiTransclude( $title, 'render' );
+ $isHTML = true;
+ $noparse = true;
+ } else {
+ $text = $this->interwikiTransclude( $title, 'raw' );
+ }
+ $found = true;
}
}
}
# Remove <noinclude> sections and <includeonly> tags
$text = preg_replace( '/<noinclude>.*?<\/noinclude>/s', '', $text );
$text = strtr( $text, array( '<includeonly>' => '' , '</includeonly>' => '' ) );
-
+
if( $this->mOutputType == OT_HTML ) {
# Strip <nowiki>, <pre>, etc.
$text = $this->strip( $text, $this->mStripState );
return $text;
}
}
-
+
/**
- * Translude an interwiki link.
+ * Fetch the unparsed text of a template and register a reference to it.
*/
- function scarytransclude($title, $interwiki) {
- global $wgEnableScaryTranscluding;
+ function fetchTemplate( $title ) {
+ $text = false;
+ // Loop to fetch the article, with up to 1 redirect
+ for ( $i = 0; $i < 2 && is_object( $title ); $i++ ) {
+ $rev = Revision::newFromTitle( $title );
+ $this->mOutput->addTemplate( $title, $title->getArticleID() );
+ if ( !$rev ) {
+ break;
+ }
+ $text = $rev->getText();
+ if ( $text === false ) {
+ break;
+ }
+ // Redirect?
+ $title = Title::newFromRedirect( $text );
+ }
+ return $text;
+ }
+
+ /**
+ * Transclude an interwiki link.
+ */
+ function interwikiTransclude( $title, $action ) {
+ global $wgEnableScaryTranscluding, $wgCanonicalNamespaceNames;
if (!$wgEnableScaryTranscluding)
return wfMsg('scarytranscludedisabled');
- $articlename = "Template:" . $title->getDBkey();
- $url = str_replace('$1', urlencode($articlename), $interwiki);
+ // The namespace will actually only be 0 or 10, depending on whether there was a leading :
+ // But we'll handle it generally anyway
+ if ( $title->getNamespace() ) {
+ // Use the canonical namespace, which should work anywhere
+ $articleName = $wgCanonicalNamespaceNames[$title->getNamespace()] . ':' . $title->getDBkey();
+ } else {
+ $articleName = $title->getDBkey();
+ }
+
+ $url = str_replace('$1', urlencode($articleName), Title::getInterwikiLink($title->getInterwiki()));
+ $url .= "?action=$action";
if (strlen($url) > 255)
return wfMsg('scarytranscludetoolong');
- $text = $this->fetchScaryTemplateMaybeFromCache($url);
- $this->mIWTransData[] = $text;
- return "<!--IW_TRANSCLUDE ".(count($this->mIWTransData) - 1)."-->";
+ return $this->fetchScaryTemplateMaybeFromCache($url);
}
function fetchScaryTemplateMaybeFromCache($url) {
+ global $wgTranscludeCacheExpiry;
$dbr =& wfGetDB(DB_SLAVE);
$obj = $dbr->selectRow('transcache', array('tc_time', 'tc_contents'),
array('tc_url' => $url));
if ($obj) {
$time = $obj->tc_time;
$text = $obj->tc_contents;
- if ($time && $time < (time() + (60*60))) {
+ if ($time && time() < $time + $wgTranscludeCacheExpiry ) {
return $text;
}
}
- $text = wfGetHTTP($url . '?action=render');
+ $text = wfGetHTTP($url);
if (!$text)
return wfMsg('scarytranscludefailed', $url);
$dbw =& wfGetDB(DB_MASTER);
- $dbw->replace('transcache', array(), array(
+ $dbw->replace('transcache', array('tc_url'), array(
'tc_url' => $url,
'tc_time' => time(),
'tc_contents' => $text));
function pstPass2( $text, &$user ) {
global $wgContLang, $wgLocaltimezone;
- # Variable replacement
- # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
- $text = $this->replaceVariables( $text );
-
- # Signatures
- #
- $sigText = $this->getUserSig( $user );
-
/* Note: This is the timestamp saved as hardcoded wikitext to
* the database, we use $wgContLang here in order to give
* everyone the same signiture and use the default one rather
putenv( 'TZ='.$oldtz );
}
- $text = preg_replace( '/~~~~~/', $d, $text );
- $text = preg_replace( '/~~~~/', "$sigText $d", $text );
- $text = preg_replace( '/~~~/', $sigText, $text );
+ # Variable replacement
+ # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
+ $text = $this->replaceVariables( $text );
+
+ # Signatures
+ $sigText = $this->getUserSig( $user );
+ $text = strtr( $text, array(
+ '~~~~~' => $d,
+ '~~~~' => "$sigText $d",
+ '~~~' => $sigText
+ ) );
# Context links: [[|name]] and [[name (context)|]]
#
global $wgLegalTitleChars;
$tc = "[$wgLegalTitleChars]";
$np = str_replace( array( '(', ')' ), array( '', '' ), $tc ); # No parens
-
+
$namespacechar = '[ _0-9A-Za-z\x80-\xff]'; # Namespaces can use non-ascii!
$conpat = "/^({$np}+) \\(({$tc}+)\\)$/";
return $text;
}
-
+
/**
* Fetch the user's signature text, if any, and normalize to
* validated, ready-to-insert wikitext.
* @access private
*/
function getUserSig( &$user ) {
- $name = $user->getName();
- $nick = trim( $user->getOption( 'nickname' ) );
- if ( '' == $nick ) {
- $nick = $name;
- }
-
- if( $user->getOption( 'fancysig' ) ) {
- // A wikitext signature.
- $valid = $this->validateSig( $nick );
- if( $valid === false ) {
- // Fall back to default sig
- $nick = $name;
- wfDebug( "Parser::getUserSig: $name has bad XML tags in signature.\n" );
+ global $wgContLang;
+
+ $username = $user->getName();
+ $nickname = $user->getOption( 'nickname' );
+ $nickname = $nickname === '' ? $username : $nickname;
+
+ if( $user->getBoolOption( 'fancysig' ) !== false ) {
+ # Sig. might contain markup; validate this
+ if( $this->validateSig( $nickname ) !== false ) {
+ # Validated; clean up (if needed) and return it
+ return( $this->cleanSig( $nickname ) );
} else {
- return $nick;
+ # Failed to validate; fall back to the default
+ $nickname = $username;
+ wfDebug( "Parser::getUserSig: $username has bad XML tags in signature.\n" );
}
}
-
- // Plain text linking to the user's homepage
- global $wgContLang;
- $page = $user->getUserPage();
- return '[[' .
- $page->getPrefixedText() .
- "|" .
- wfEscapeWikIText( $nick ) .
- "]]";
+
+ # If we're still here, make it a link to the user page
+ $userpage = $user->getUserPage();
+ return( '[[' . $userpage->getPrefixedText() . '|' . wfEscapeWikiText( $nickname ) . ']]' );
}
-
+
/**
- * We want to enforce two rules on wikitext sigs here:
- * 1) Expand any templates at save time (forced subst:)
- * 2) Check for unbalanced XML tags, and reject if so.
+ * Check that the user's signature contains no bad XML
*
* @param string $text
* @return mixed An expanded string, or false if invalid.
- *
- * @todo Run brace substitutions
- * @todo ?? Check for unbalanced '' and ''' quotes, etc
*/
function validateSig( $text ) {
- if( wfIsWellFormedXmlFragment( $text ) ) {
- return $text;
- } else {
- return false;
- }
+ return( wfIsWellFormedXmlFragment( $text ) ? $text : false );
}
+
+ /**
+ * Clean up signature text
+ *
+ * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures
+ * 2) Substitute all transclusions
+ *
+ * @param string $text
+ * @return string Signature text
+ */
+ function cleanSig( $text ) {
+ $substWord = MagicWord::get( MAG_SUBST );
+ $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
+ $substText = '{{' . $substWord->getSynonym( 0 );
+ $text = preg_replace( $substRegex, $substText, $text );
+ $text = preg_replace( '/~{3,5}/', '', $text );
+ $text = $this->replaceVariables( $text );
+
+ return $text;
+ }
+
/**
* Set up some variables which are usually set up in parse()
* so that an external function can call some class members with confidence
global $wgTitle;
static $executing = false;
+ $fname = "Parser::transformMsg";
+
# Guard against infinite recursion
if ( $executing ) {
return $text;
}
$executing = true;
+ wfProfileIn($fname);
+
$this->mTitle = $wgTitle;
$this->mOptions = $options;
$this->mOutputType = OT_MSG;
$text = $this->replaceVariables( $text );
$executing = false;
+ wfProfileOut($fname);
return $text;
}
function setHook( $tag, $callback ) {
$oldVal = @$this->mTagHooks[$tag];
$this->mTagHooks[$tag] = $callback;
-
+
return $oldVal;
}
function disableCache() {
$this->mOutput->mCacheTime = -1;
}
-
- /**
+
+ /**#@+
* Callback from the Sanitizer for expanding items found in HTML attribute
* values, so they can be safely tested and escaped.
* @param string $text
$text = $this->unstripForHTML( $text );
return $text;
}
-
+
function unstripForHTML( $text ) {
$text = $this->unstrip( $text, $this->mStripState );
$text = $this->unstripNoWiki( $text, $this->mStripState );
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_keys( $this->mTagHooks ); }
+ /**#@-*/
}
/**
$mTitleText, # title text of the chosen language variant
$mLinks, # 2-D map of NS/DBK to ID for the links in the document. ID=zero for broken.
$mTemplates, # 2-D map of NS/DBK to ID for the template references. ID=zero for broken.
- $mImages; # DB keys of the images used, in the array key only
+ $mImages, # DB keys of the images used, in the array key only
+ $mExternalLinks; # External link URLs, in the key only
function ParserOutput( $text = '', $languageLinks = array(), $categoryLinks = array(),
$containsOldMagic = false, $titletext = '' )
$this->mLinks = array();
$this->mTemplates = array();
$this->mImages = array();
+ $this->mExternalLinks = array();
}
function getText() { return $this->mText; }
function &getLinks() { return $this->mLinks; }
function &getTemplates() { return $this->mTemplates; }
function &getImages() { return $this->mImages; }
+ function &getExternalLinks() { return $this->mExternalLinks; }
function containsOldMagic() { return $this->mContainsOldMagic; }
function setText( $text ) { return wfSetVar( $this->mText, $text ); }
function addCategory( $c, $sort ) { $this->mCategories[$c] = $sort; }
function addImage( $name ) { $this->mImages[$name] = 1; }
function addLanguageLink( $t ) { $this->mLanguageLinks[] = $t; }
+ function addExternalLink( $url ) { $this->mExternalLinks[$url] = 1; }
function addLink( $title, $id ) {
$ns = $title->getNamespace();
function addTemplate( $title, $id ) {
$ns = $title->getNamespace();
- $dbk = $title->getDBkey();
+ $dbk = $title->getDBkey();
if ( !isset( $this->mTemplates[$ns] ) ) {
$this->mTemplates[$ns] = array();
}
var $mEditSection; # Create "edit section" links
var $mNumberHeadings; # Automatically number headings
var $mAllowSpecialInclusion; # Allow inclusion of special pages
+ var $mTidy; # Ask for tidy cleanup
function getUseTeX() { return $this->mUseTeX; }
function getUseDynamicDates() { return $this->mUseDynamicDates; }
function getEditSection() { return $this->mEditSection; }
function getNumberHeadings() { return $this->mNumberHeadings; }
function getAllowSpecialInclusion() { return $this->mAllowSpecialInclusion; }
-
+ function getTidy() { return $this->mTidy; }
function setUseTeX( $x ) { return wfSetVar( $this->mUseTeX, $x ); }
function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates, $x ); }
function setEditSection( $x ) { return wfSetVar( $this->mEditSection, $x ); }
function setNumberHeadings( $x ) { return wfSetVar( $this->mNumberHeadings, $x ); }
function setAllowSpecialInclusion( $x ) { return wfSetVar( $this->mAllowSpecialInclusion, $x ); }
-
+ function setTidy( $x ) { return wfSetVar( $this->mTidy, $x); }
function setSkin( &$x ) { $this->mSkin =& $x; }
function ParserOptions() {
$this->mEditSection = true;
$this->mNumberHeadings = $user->getOption( 'numberheadings' );
$this->mAllowSpecialInclusion = $wgAllowSpecialInclusion;
+ $this->mTidy = false;
wfProfileOut( $fname );
}
}