summaryrefslogtreecommitdiff
path: root/includes/parser
diff options
context:
space:
mode:
Diffstat (limited to 'includes/parser')
-rw-r--r--includes/parser/CacheTime.php4
-rw-r--r--includes/parser/CoreLinkFunctions.php92
-rw-r--r--includes/parser/CoreParserFunctions.php267
-rw-r--r--includes/parser/CoreTagHooks.php3
-rw-r--r--includes/parser/DateFormatter.php76
-rw-r--r--includes/parser/LinkHolderArray.php294
-rw-r--r--includes/parser/Parser.php1544
-rw-r--r--includes/parser/ParserCache.php25
-rw-r--r--includes/parser/ParserOptions.php107
-rw-r--r--includes/parser/ParserOutput.php315
-rw-r--r--includes/parser/Parser_DiffTest.php2
-rw-r--r--includes/parser/Parser_LinkHooks.php324
-rw-r--r--includes/parser/Preprocessor.php7
-rw-r--r--includes/parser/Preprocessor_DOM.php172
-rw-r--r--includes/parser/Preprocessor_Hash.php135
-rw-r--r--includes/parser/Preprocessor_HipHop.hphp2013
-rw-r--r--includes/parser/StripState.php7
-rw-r--r--includes/parser/Tidy.php69
18 files changed, 1878 insertions, 3578 deletions
diff --git a/includes/parser/CacheTime.php b/includes/parser/CacheTime.php
index 881dded7..8190a8a0 100644
--- a/includes/parser/CacheTime.php
+++ b/includes/parser/CacheTime.php
@@ -93,7 +93,7 @@ class CacheTime {
$expire = min( $expire, $wgParserCacheExpireTime );
}
- if( $this->containsOldMagic() ) { //compatibility hack
+ if ( $this->containsOldMagic() ) { //compatibility hack
$expire = min( $expire, 3600 ); # 1 hour
}
@@ -116,7 +116,7 @@ class CacheTime {
* per-article cache invalidation timestamps, or if it comes from
* an incompatible older version.
*
- * @param $touched String: the affected article's last touched timestamp
+ * @param string $touched the affected article's last touched timestamp
* @return Boolean
*/
public function expired( $touched ) {
diff --git a/includes/parser/CoreLinkFunctions.php b/includes/parser/CoreLinkFunctions.php
deleted file mode 100644
index 4bfa9d35..00000000
--- a/includes/parser/CoreLinkFunctions.php
+++ /dev/null
@@ -1,92 +0,0 @@
-<?php
-/**
- * Link functions provided by MediaWiki core; experimental
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
- * http://www.gnu.org/copyleft/gpl.html
- *
- * @file
- * @ingroup Parser
- */
-
-/**
- * Various core link functions, registered in Parser::firstCallInit()
- * @ingroup Parser
- */
-class CoreLinkFunctions {
- /**
- * @param $parser Parser_LinkHooks
- * @return bool
- */
- static function register( $parser ) {
- $parser->setLinkHook( NS_CATEGORY, array( __CLASS__, 'categoryLinkHook' ) );
- return true;
- }
-
- /**
- * @param $parser Parser
- * @param $holders LinkHolderArray
- * @param $markers LinkMarkerReplacer
- * @param Title $title
- * @param $titleText
- * @param null $displayText
- * @param bool $leadingColon
- * @return bool
- */
- static function defaultLinkHook( $parser, $holders, $markers,
- Title $title, $titleText, &$displayText = null, &$leadingColon = false ) {
- if( isset($displayText) && $markers->findMarker( $displayText ) ) {
- # There are links inside of the displayText
- # For backwards compatibility the deepest links are dominant so this
- # link should not be handled
- $displayText = $markers->expand($displayText);
- # Return false so that this link is reverted back to WikiText
- return false;
- }
- return $holders->makeHolder( $title, isset($displayText) ? $displayText : $titleText, array(), '', '' );
- }
-
- /**
- * @param $parser Parser
- * @param $holders LinkHolderArray
- * @param $markers LinkMarkerReplacer
- * @param Title $title
- * @param $titleText
- * @param null $sortText
- * @param bool $leadingColon
- * @return bool|string
- */
- static function categoryLinkHook( $parser, $holders, $markers,
- Title $title, $titleText, &$sortText = null, &$leadingColon = false ) {
- global $wgContLang;
- # When a category link starts with a : treat it as a normal link
- if( $leadingColon ) return true;
- if( isset($sortText) && $markers->findMarker( $sortText ) ) {
- # There are links inside of the sortText
- # For backwards compatibility the deepest links are dominant so this
- # link should not be handled
- $sortText = $markers->expand($sortText);
- # Return false so that this link is reverted back to WikiText
- return false;
- }
- if( !isset($sortText) ) $sortText = $parser->getDefaultSort();
- $sortText = Sanitizer::decodeCharReferences( $sortText );
- $sortText = str_replace( "\n", '', $sortText );
- $sortText = $wgContLang->convertCategoryKey( $sortText );
- $parser->mOutput->addCategory( $title->getDBkey(), $sortText );
- return '';
- }
-
-}
diff --git a/includes/parser/CoreParserFunctions.php b/includes/parser/CoreParserFunctions.php
index 8917b6d0..4b6eeca2 100644
--- a/includes/parser/CoreParserFunctions.php
+++ b/includes/parser/CoreParserFunctions.php
@@ -88,6 +88,8 @@ class CoreParserFunctions {
$parser->setFunctionHook( 'pagenamee', array( __CLASS__, 'pagenamee' ), SFH_NO_HASH );
$parser->setFunctionHook( 'fullpagename', array( __CLASS__, 'fullpagename' ), SFH_NO_HASH );
$parser->setFunctionHook( 'fullpagenamee', array( __CLASS__, 'fullpagenamee' ), SFH_NO_HASH );
+ $parser->setFunctionHook( 'rootpagename', array( __CLASS__, 'rootpagename' ), SFH_NO_HASH );
+ $parser->setFunctionHook( 'rootpagenamee', array( __CLASS__, 'rootpagenamee' ), SFH_NO_HASH );
$parser->setFunctionHook( 'basepagename', array( __CLASS__, 'basepagename' ), SFH_NO_HASH );
$parser->setFunctionHook( 'basepagenamee', array( __CLASS__, 'basepagenamee' ), SFH_NO_HASH );
$parser->setFunctionHook( 'subpagename', array( __CLASS__, 'subpagename' ), SFH_NO_HASH );
@@ -137,9 +139,10 @@ class CoreParserFunctions {
$pref = $parser->getOptions()->getDateFormat();
// Specify a different default date format other than the the normal default
- // iff the user has 'default' for their setting
- if ( $pref == 'default' && $defaultPref )
+ // if the user has 'default' for their setting
+ if ( $pref == 'default' && $defaultPref ) {
$pref = $defaultPref;
+ }
$date = $df->reformat( $pref, $date, array( 'match-whole' ) );
return $date;
@@ -175,8 +178,8 @@ class CoreParserFunctions {
* For links to "wiki"s, or similar software, spaces are encoded as '_',
*
* @param $parser Parser object
- * @param $s String: The text to encode.
- * @param $arg String (optional): The type of encoding.
+ * @param string $s The text to encode.
+ * @param string $arg (optional): The type of encoding.
* @return string
*/
static function urlencode( $parser, $s = '', $arg = null ) {
@@ -184,7 +187,7 @@ class CoreParserFunctions {
if ( is_null( $magicWords ) ) {
$magicWords = new MagicWordArray( array( 'url_path', 'url_query', 'url_wiki' ) );
}
- switch( $magicWords->matchStartToEnd( $arg ) ) {
+ switch ( $magicWords->matchStartToEnd( $arg ) ) {
// Encode as though it's a wiki page, '_' for ' '.
case 'url_wiki':
@@ -248,14 +251,15 @@ class CoreParserFunctions {
# before arriving here; if that's true, then the title can't be created
# and the variable will fail. If we can't get a decent title from the first
# attempt, url-decode and try for a second.
- if( is_null( $title ) )
+ if ( is_null( $title ) ) {
$title = Title::newFromURL( urldecode( $s ) );
- if( !is_null( $title ) ) {
+ }
+ if ( !is_null( $title ) ) {
# Convert NS_MEDIA -> NS_FILE
- if( $title->getNamespace() == NS_MEDIA ) {
+ if ( $title->getNamespace() == NS_MEDIA ) {
$title = Title::makeTitle( NS_FILE, $title->getDBkey() );
}
- if( !is_null( $arg ) ) {
+ if ( !is_null( $arg ) ) {
$text = $title->$func( $arg );
} else {
$text = $title->$func();
@@ -269,12 +273,14 @@ class CoreParserFunctions {
/**
* @param $parser Parser
* @param string $num
- * @param null $raw
- * @return
+ * @param string $arg
+ * @return string
*/
- static function formatnum( $parser, $num = '', $raw = null) {
- if ( self::isRaw( $raw ) ) {
+ static function formatnum( $parser, $num = '', $arg = null ) {
+ if ( self::matchAgainstMagicword( 'rawsuffix', $arg ) ) {
$func = array( $parser->getFunctionLang(), 'parseFormattedNumber' );
+ } elseif ( self::matchAgainstMagicword( 'nocommafysuffix', $arg ) ) {
+ $func = array( $parser->getFunctionLang(), 'formatNumNoSeparators' );
} else {
$func = array( $parser->getFunctionLang(), 'formatNum' );
}
@@ -351,55 +357,77 @@ class CoreParserFunctions {
* title which will normalise to the canonical title
*
* @param $parser Parser: parent parser
- * @param $text String: desired title text
+ * @param string $text desired title text
* @return String
*/
static function displaytitle( $parser, $text = '' ) {
global $wgRestrictDisplayTitle;
- #parse a limited subset of wiki markup (just the single quote items)
+ // parse a limited subset of wiki markup (just the single quote items)
$text = $parser->doQuotes( $text );
- #remove stripped text (e.g. the UNIQ-QINU stuff) that was generated by tag extensions/whatever
+ // remove stripped text (e.g. the UNIQ-QINU stuff) that was generated by tag extensions/whatever
$text = preg_replace( '/' . preg_quote( $parser->uniqPrefix(), '/' ) . '.*?'
. preg_quote( Parser::MARKER_SUFFIX, '/' ) . '/', '', $text );
- #list of disallowed tags for DISPLAYTITLE
- #these will be escaped even though they are allowed in normal wiki text
+ // list of disallowed tags for DISPLAYTITLE
+ // these will be escaped even though they are allowed in normal wiki text
$bad = array( 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'div', 'blockquote', 'ol', 'ul', 'li', 'hr',
'table', 'tr', 'th', 'td', 'dl', 'dd', 'caption', 'p', 'ruby', 'rb', 'rt', 'rp', 'br' );
- #only requested titles that normalize to the actual title are allowed through
- #if $wgRestrictDisplayTitle is true (it is by default)
- #mimic the escaping process that occurs in OutputPage::setPageTitle
- $text = Sanitizer::normalizeCharReferences( Sanitizer::removeHTMLtags( $text, null, array(), array(), $bad ) );
+ // disallow some styles that could be used to bypass $wgRestrictDisplayTitle
+ if ( $wgRestrictDisplayTitle ) {
+ $htmlTagsCallback = function ( &$params ) {
+ $decoded = Sanitizer::decodeTagAttributes( $params );
+
+ if ( isset( $decoded['style'] ) ) {
+ // this is called later anyway, but we need it right now for the regexes below to be safe
+ // calling it twice doesn't hurt
+ $decoded['style'] = Sanitizer::checkCss( $decoded['style'] );
+
+ if ( preg_match( '/(display|user-select|visibility)\s*:/i', $decoded['style'] ) ) {
+ $decoded['style'] = '/* attempt to bypass $wgRestrictDisplayTitle */';
+ }
+ }
+
+ $params = Sanitizer::safeEncodeTagAttributes( $decoded );
+ };
+ } else {
+ $htmlTagsCallback = null;
+ }
+
+ // only requested titles that normalize to the actual title are allowed through
+ // if $wgRestrictDisplayTitle is true (it is by default)
+ // mimic the escaping process that occurs in OutputPage::setPageTitle
+ $text = Sanitizer::normalizeCharReferences( Sanitizer::removeHTMLtags( $text, $htmlTagsCallback, array(), array(), $bad ) );
$title = Title::newFromText( Sanitizer::stripAllTags( $text ) );
- if( !$wgRestrictDisplayTitle ) {
+ if ( !$wgRestrictDisplayTitle ) {
+ $parser->mOutput->setDisplayTitle( $text );
+ } elseif ( $title instanceof Title && $title->getFragment() == '' && $title->equals( $parser->mTitle ) ) {
$parser->mOutput->setDisplayTitle( $text );
- } else {
- if ( $title instanceof Title && $title->getFragment() == '' && $title->equals( $parser->mTitle ) ) {
- $parser->mOutput->setDisplayTitle( $text );
- }
}
return '';
}
- static function isRaw( $param ) {
- static $mwRaw;
- if ( !$mwRaw ) {
- $mwRaw =& MagicWord::get( 'rawsuffix' );
- }
- if ( is_null( $param ) ) {
+ /**
+ * Matches the given value against the value of given magic word
+ *
+ * @param string $magicword magic word key
+ * @param mixed $value value to match
+ * @return boolean true on successful match
+ */
+ static private function matchAgainstMagicword( $magicword, $value ) {
+ if ( strval( $value ) === '' ) {
return false;
- } else {
- return $mwRaw->match( $param );
}
+ $mwObject = MagicWord::get( $magicword );
+ return $mwObject->match( $value );
}
static function formatRaw( $num, $raw ) {
- if( self::isRaw( $raw ) ) {
+ if ( self::matchAgainstMagicword( 'rawsuffix', $raw ) ) {
return $num;
} else {
global $wgContLang;
@@ -422,22 +450,22 @@ class CoreParserFunctions {
return self::formatRaw( SiteStats::images(), $raw );
}
static function numberofadmins( $parser, $raw = null ) {
- return self::formatRaw( SiteStats::numberingroup('sysop'), $raw );
+ return self::formatRaw( SiteStats::numberingroup( 'sysop' ), $raw );
}
static function numberofedits( $parser, $raw = null ) {
return self::formatRaw( SiteStats::edits(), $raw );
}
static function numberofviews( $parser, $raw = null ) {
- return self::formatRaw( SiteStats::views(), $raw );
+ global $wgDisableCounters;
+ return !$wgDisableCounters ? self::formatRaw( SiteStats::views(), $raw ) : '';
}
static function pagesinnamespace( $parser, $namespace = 0, $raw = null ) {
return self::formatRaw( SiteStats::pagesInNs( intval( $namespace ) ), $raw );
}
- static function numberingroup( $parser, $name = '', $raw = null) {
+ static function numberingroup( $parser, $name = '', $raw = null ) {
return self::formatRaw( SiteStats::numberingroup( strtolower( $name ) ), $raw );
}
-
/**
* Given a title, return the namespace name that would be given by the
* corresponding magic word
@@ -447,44 +475,51 @@ class CoreParserFunctions {
*/
static function mwnamespace( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return str_replace( '_', ' ', $t->getNsText() );
}
static function namespacee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfUrlencode( $t->getNsText() );
}
static function namespacenumber( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return $t->getNamespace();
}
static function talkspace( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) || !$t->canTalk() )
+ if ( is_null( $t ) || !$t->canTalk() ) {
return '';
+ }
return str_replace( '_', ' ', $t->getTalkNsText() );
}
static function talkspacee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) || !$t->canTalk() )
+ if ( is_null( $t ) || !$t->canTalk() ) {
return '';
+ }
return wfUrlencode( $t->getTalkNsText() );
}
static function subjectspace( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return str_replace( '_', ' ', $t->getSubjectNsText() );
}
static function subjectspacee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfUrlencode( $t->getSubjectNsText() );
}
@@ -495,75 +530,101 @@ class CoreParserFunctions {
*/
static function pagename( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfEscapeWikiText( $t->getText() );
}
static function pagenamee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfEscapeWikiText( $t->getPartialURL() );
}
static function fullpagename( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) || !$t->canTalk() )
+ if ( is_null( $t ) || !$t->canTalk() ) {
return '';
+ }
return wfEscapeWikiText( $t->getPrefixedText() );
}
static function fullpagenamee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) || !$t->canTalk() )
+ if ( is_null( $t ) || !$t->canTalk() ) {
return '';
+ }
return wfEscapeWikiText( $t->getPrefixedURL() );
}
static function subpagename( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfEscapeWikiText( $t->getSubpageText() );
}
static function subpagenamee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfEscapeWikiText( $t->getSubpageUrlForm() );
}
+ static function rootpagename( $parser, $title = null ) {
+ $t = Title::newFromText( $title );
+ if ( is_null( $t ) ) {
+ return '';
+ }
+ return wfEscapeWikiText( $t->getRootText() );
+ }
+ static function rootpagenamee( $parser, $title = null ) {
+ $t = Title::newFromText( $title );
+ if ( is_null( $t ) ) {
+ return '';
+ }
+ return wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $t->getRootText() ) ) );
+ }
static function basepagename( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfEscapeWikiText( $t->getBaseText() );
}
static function basepagenamee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $t->getBaseText() ) ) );
}
static function talkpagename( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) || !$t->canTalk() )
+ if ( is_null( $t ) || !$t->canTalk() ) {
return '';
+ }
return wfEscapeWikiText( $t->getTalkPage()->getPrefixedText() );
}
static function talkpagenamee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) || !$t->canTalk() )
+ if ( is_null( $t ) || !$t->canTalk() ) {
return '';
- return wfEscapeWikiText( $t->getTalkPage()->getPrefixedUrl() );
+ }
+ return wfEscapeWikiText( $t->getTalkPage()->getPrefixedURL() );
}
static function subjectpagename( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
+ }
return wfEscapeWikiText( $t->getSubjectPage()->getPrefixedText() );
}
static function subjectpagenamee( $parser, $title = null ) {
$t = Title::newFromText( $title );
- if ( is_null( $t ) )
+ if ( is_null( $t ) ) {
return '';
- return wfEscapeWikiText( $t->getSubjectPage()->getPrefixedUrl() );
+ }
+ return wfEscapeWikiText( $t->getSubjectPage()->getPrefixedURL() );
}
/**
@@ -573,6 +634,7 @@ class CoreParserFunctions {
* @return string
*/
static function pagesincategory( $parser, $name = '', $arg1 = null, $arg2 = null ) {
+ global $wgContLang;
static $magicWords = null;
if ( is_null( $magicWords ) ) {
$magicWords = new MagicWordArray( array(
@@ -585,7 +647,7 @@ class CoreParserFunctions {
static $cache = array();
// split the given option to its variable
- if( self::isRaw( $arg1 ) ) {
+ if ( self::matchAgainstMagicword( 'rawsuffix', $arg1 ) ) {
//{{pagesincategory:|raw[|type]}}
$raw = $arg1;
$type = $magicWords->matchStartToEnd( $arg2 );
@@ -594,23 +656,24 @@ class CoreParserFunctions {
$type = $magicWords->matchStartToEnd( $arg1 );
$raw = $arg2;
}
- if( !$type ) { //backward compatibility
+ if ( !$type ) { //backward compatibility
$type = 'pagesincategory_all';
}
$title = Title::makeTitleSafe( NS_CATEGORY, $name );
- if( !$title ) { # invalid title
+ if ( !$title ) { # invalid title
return self::formatRaw( 0, $raw );
}
+ $wgContLang->findVariantLink( $name, $title, true );
// Normalize name for cache
$name = $title->getDBkey();
- if( !isset( $cache[$name] ) ) {
+ if ( !isset( $cache[$name] ) ) {
$category = Category::newFromTitle( $title );
$allCount = $subcatCount = $fileCount = $pagesCount = 0;
- if( $parser->incrementExpensiveFunctionCount() ) {
+ if ( $parser->incrementExpensiveFunctionCount() ) {
// $allCount is the total number of cat members,
// not the count of how many members are normal pages.
$allCount = (int)$category->getPageCount();
@@ -632,8 +695,6 @@ class CoreParserFunctions {
* Return the size of the given page, or 0 if it's nonexistent. This is an
* expensive parser function and can't be called too many times per page.
*
- * @todo FIXME: This doesn't work correctly on preview for getting the size
- * of the current page.
* @todo FIXME: Title::getLength() documentation claims that it adds things
* to the link cache, so the local cache here should be unnecessary, but
* in fact calling getLength() repeatedly for the same $page does seem to
@@ -641,15 +702,15 @@ class CoreParserFunctions {
* @todo Document parameters
*
* @param $parser Parser
- * @param $page String TODO DOCUMENT (Default: empty string)
- * @param $raw TODO DOCUMENT (Default: null)
+ * @param $page String Name of page to check (Default: empty string)
+ * @param $raw String Should number be human readable with commas or just number
* @return string
*/
static function pagesize( $parser, $page = '', $raw = null ) {
static $cache = array();
$title = Title::newFromText( $page );
- if( !is_object( $title ) ) {
+ if ( !is_object( $title ) ) {
$cache[$page] = 0;
return self::formatRaw( 0, $raw );
}
@@ -658,25 +719,35 @@ class CoreParserFunctions {
$page = $title->getPrefixedText();
$length = 0;
- if( isset( $cache[$page] ) ) {
+ if ( isset( $cache[$page] ) ) {
$length = $cache[$page];
- } elseif( $parser->incrementExpensiveFunctionCount() ) {
+ } elseif ( $parser->incrementExpensiveFunctionCount() ) {
$rev = Revision::newFromTitle( $title, false, Revision::READ_NORMAL );
- $id = $rev ? $rev->getPage() : 0;
+ $pageID = $rev ? $rev->getPage() : 0;
+ $revID = $rev ? $rev->getId() : 0;
$length = $cache[$page] = $rev ? $rev->getSize() : 0;
// Register dependency in templatelinks
- $parser->mOutput->addTemplate( $title, $id, $rev ? $rev->getId() : 0 );
+ $parser->mOutput->addTemplate( $title, $pageID, $revID );
}
return self::formatRaw( $length, $raw );
}
/**
- * Returns the requested protection level for the current page
+ * Returns the requested protection level for the current page
+ *
+ * @param Parser $parser
+ * @param string $type
+ * @param string $title
+ *
* @return string
*/
- static function protectionlevel( $parser, $type = '' ) {
- $restrictions = $parser->mTitle->getRestrictions( strtolower( $type ) );
+ static function protectionlevel( $parser, $type = '', $title = '' ) {
+ $titleObject = Title::newFromText( $title );
+ if ( !( $titleObject instanceof Title ) ) {
+ $titleObject = $parser->mTitle;
+ }
+ $restrictions = $titleObject->getRestrictions( strtolower( $type ) );
# Title::getRestrictions returns an array, its possible it may have
# multiple values in the future
return implode( $restrictions, ',' );
@@ -685,8 +756,8 @@ class CoreParserFunctions {
/**
* Gives language names.
* @param $parser Parser
- * @param $code String Language code (of which to get name)
- * @param $inLanguage String Language code (in which to get name)
+ * @param string $code Language code (of which to get name)
+ * @param string $inLanguage Language code (in which to get name)
* @return String
*/
static function language( $parser, $code = '', $inLanguage = '' ) {
@@ -703,7 +774,9 @@ class CoreParserFunctions {
static function pad( $parser, $string, $length, $padding = '0', $direction = STR_PAD_RIGHT ) {
$padding = $parser->killMarkers( $padding );
$lengthOfPadding = mb_strlen( $padding );
- if ( $lengthOfPadding == 0 ) return $string;
+ if ( $lengthOfPadding == 0 ) {
+ return $string;
+ }
# The remaining length to add counts down to 0 as padding is added
$length = min( $length, 500 ) - mb_strlen( $string );
@@ -739,7 +812,7 @@ class CoreParserFunctions {
*/
static function anchorencode( $parser, $text ) {
$text = $parser->killMarkers( $text );
- return substr( $parser->guessSectionNameFromWikiText( $text ), 1);
+ return (string)substr( $parser->guessSectionNameFromWikiText( $text ), 1 );
}
static function special( $parser, $text ) {
@@ -748,7 +821,9 @@ class CoreParserFunctions {
$title = SpecialPage::getTitleFor( $page, $subpage );
return $title->getPrefixedText();
} else {
- return wfMessage( 'nosuchspecialpage' )->inContentLanguage()->text();
+ // unknown special page, just use the given text as its title, if at all possible
+ $title = Title::makeTitleSafe( NS_SPECIAL, $text );
+ return $title ? $title->getPrefixedText() : self::special( $parser, 'Badtitle' );
}
}
@@ -758,8 +833,8 @@ class CoreParserFunctions {
/**
* @param $parser Parser
- * @param $text String The sortkey to use
- * @param $uarg String Either "noreplace" or "noerror" (in en)
+ * @param string $text The sortkey to use
+ * @param string $uarg Either "noreplace" or "noerror" (in en)
* both suppress errors, and noreplace does nothing if
* a default sortkey already exists.
* @return string
@@ -772,35 +847,41 @@ class CoreParserFunctions {
$arg = $magicWords->matchStartToEnd( $uarg );
$text = trim( $text );
- if( strlen( $text ) == 0 )
+ if ( strlen( $text ) == 0 ) {
return '';
+ }
$old = $parser->getCustomDefaultSort();
if ( $old === false || $arg !== 'defaultsort_noreplace' ) {
$parser->setDefaultSort( $text );
}
- if( $old === false || $old == $text || $arg ) {
+ if ( $old === false || $old == $text || $arg ) {
return '';
} else {
- return( '<span class="error">' .
- wfMessage( 'duplicate-defaultsort', $old, $text )->inContentLanguage()->escaped() .
- '</span>' );
+ $converter = $parser->getConverterLanguage()->getConverter();
+ return '<span class="error">' .
+ wfMessage( 'duplicate-defaultsort',
+ // Message should be parsed, but these params should only be escaped.
+ $converter->markNoConversion( wfEscapeWikiText( $old ) ),
+ $converter->markNoConversion( wfEscapeWikiText( $text ) )
+ )->inContentLanguage()->text() .
+ '</span>';
}
}
// Usage {{filepath|300}}, {{filepath|nowiki}}, {{filepath|nowiki|300}} or {{filepath|300|nowiki}}
// or {{filepath|300px}}, {{filepath|200x300px}}, {{filepath|nowiki|200x300px}}, {{filepath|200x300px|nowiki}}
- public static function filepath( $parser, $name='', $argA='', $argB='' ) {
+ public static function filepath( $parser, $name = '', $argA = '', $argB = '' ) {
$file = wfFindFile( $name );
- if( $argA == 'nowiki' ) {
+ if ( $argA == 'nowiki' ) {
// {{filepath: | option [| size] }}
$isNowiki = true;
$parsedWidthParam = $parser->parseWidthParam( $argB );
} else {
// {{filepath: [| size [|option]] }}
$parsedWidthParam = $parser->parseWidthParam( $argA );
- $isNowiki = ($argB == 'nowiki');
+ $isNowiki = ( $argB == 'nowiki' );
}
if ( $file ) {
diff --git a/includes/parser/CoreTagHooks.php b/includes/parser/CoreTagHooks.php
index 296be66f..a2eb6987 100644
--- a/includes/parser/CoreTagHooks.php
+++ b/includes/parser/CoreTagHooks.php
@@ -72,11 +72,12 @@ class CoreTagHooks {
* @param $content string
* @param $attributes array
* @param $parser Parser
+ * @throws MWException
* @return array
*/
static function html( $content, $attributes, $parser ) {
global $wgRawHtml;
- if( $wgRawHtml ) {
+ if ( $wgRawHtml ) {
return array( $content, 'markerType' => 'nowiki' );
} else {
throw new MWException( '<html> extension tag encountered unexpectedly' );
diff --git a/includes/parser/DateFormatter.php b/includes/parser/DateFormatter.php
index 2917b4a7..0a69b045 100644
--- a/includes/parser/DateFormatter.php
+++ b/includes/parser/DateFormatter.php
@@ -22,7 +22,7 @@
*/
/**
- * Date formatter, recognises dates in plain text and formats them accoding to user preferences.
+ * Date formatter, recognises dates in plain text and formats them according to user preferences.
* @todo preferences, OutputPage
* @ingroup Parser
*/
@@ -55,7 +55,7 @@ class DateFormatter {
$this->lang = $lang;
$this->monthNames = $this->getMonthRegex();
- for ( $i=1; $i<=12; $i++ ) {
+ for ( $i = 1; $i <= 12; $i++ ) {
$this->xMonths[$this->lang->lc( $this->lang->getMonthName( $i ) )] = $i;
$this->xMonths[$this->lang->lc( $this->lang->getMonthAbbreviation( $i ) )] = $i;
}
@@ -102,11 +102,11 @@ class DateFormatter {
# Rules
# pref source target
- $this->rules[self::DMY][self::MD] = self::DM;
- $this->rules[self::ALL][self::MD] = self::MD;
- $this->rules[self::MDY][self::DM] = self::MD;
- $this->rules[self::ALL][self::DM] = self::DM;
- $this->rules[self::NONE][self::ISO2] = self::ISO1;
+ $this->rules[self::DMY][self::MD] = self::DM;
+ $this->rules[self::ALL][self::MD] = self::MD;
+ $this->rules[self::MDY][self::DM] = self::MD;
+ $this->rules[self::ALL][self::DM] = self::DM;
+ $this->rules[self::NONE][self::ISO2] = self::ISO1;
$this->preferences = array(
'default' => self::NONE,
@@ -140,12 +140,12 @@ class DateFormatter {
}
/**
- * @param $preference String: User preference
- * @param $text String: Text to reformat
- * @param $options Array: can contain 'linked' and/or 'match-whole'
+ * @param string $preference User preference
+ * @param string $text Text to reformat
+ * @param array $options can contain 'linked' and/or 'match-whole'
* @return mixed|String
*/
- function reformat( $preference, $text, $options = array('linked') ) {
+ function reformat( $preference, $text, $options = array( 'linked' ) ) {
$linked = in_array( 'linked', $options );
$match_whole = in_array( 'match-whole', $options );
@@ -154,12 +154,12 @@ class DateFormatter {
} else {
$preference = self::NONE;
}
- for ( $i=1; $i<=self::LAST; $i++ ) {
+ for ( $i = 1; $i <= self::LAST; $i++ ) {
$this->mSource = $i;
- if ( isset ( $this->rules[$preference][$i] ) ) {
+ if ( isset( $this->rules[$preference][$i] ) ) {
# Specific rules
$this->mTarget = $this->rules[$preference][$i];
- } elseif ( isset ( $this->rules[self::ALL][$i] ) ) {
+ } elseif ( isset( $this->rules[self::ALL][$i] ) ) {
# General rules
$this->mTarget = $this->rules[self::ALL][$i];
} elseif ( $preference ) {
@@ -172,21 +172,21 @@ class DateFormatter {
$regex = $this->regexes[$i];
// Horrible hack
- if (!$linked) {
+ if ( !$linked ) {
$regex = str_replace( array( '\[\[', '\]\]' ), '', $regex );
}
- if ($match_whole) {
+ if ( $match_whole ) {
// Let's hope this works
$regex = preg_replace( '!^/!', '/^', $regex );
$regex = str_replace( $this->regexTrail,
- '$'.$this->regexTrail, $regex );
+ '$' . $this->regexTrail, $regex );
}
// Another horrible hack
$this->mLinked = $linked;
$text = preg_replace_callback( $regex, array( &$this, 'replace' ), $text );
- unset($this->mLinked);
+ unset( $this->mLinked );
}
return $text;
}
@@ -198,14 +198,15 @@ class DateFormatter {
function replace( $matches ) {
# Extract information from $matches
$linked = true;
- if ( isset( $this->mLinked ) )
+ if ( isset( $this->mLinked ) ) {
$linked = $this->mLinked;
-
+ }
+
$bits = array();
$key = $this->keys[$this->mSource];
- for ( $p=0; $p < strlen($key); $p++ ) {
+ for ( $p = 0; $p < strlen( $key ); $p++ ) {
if ( $key[$p] != ' ' ) {
- $bits[$key[$p]] = $matches[$p+1];
+ $bits[$key[$p]] = $matches[$p + 1];
}
}
@@ -219,8 +220,8 @@ class DateFormatter {
*/
function formatDate( $bits, $link = true ) {
$format = $this->targets[$this->mTarget];
-
- if (!$link) {
+
+ if ( !$link ) {
// strip piped links
$format = preg_replace( '/\[\[[^|]+\|([^\]]+)\]\]/', '$1', $format );
// strip remaining links
@@ -232,10 +233,12 @@ class DateFormatter {
$fail = false;
// Pre-generate y/Y stuff because we need the year for the <span> title.
- if ( !isset( $bits['y'] ) && isset( $bits['Y'] ) )
+ if ( !isset( $bits['y'] ) && isset( $bits['Y'] ) ) {
$bits['y'] = $this->makeIsoYear( $bits['Y'] );
- if ( !isset( $bits['Y'] ) && isset( $bits['y'] ) )
+ }
+ if ( !isset( $bits['Y'] ) && isset( $bits['y'] ) ) {
$bits['Y'] = $this->makeNormalYear( $bits['y'] );
+ }
if ( !isset( $bits['m'] ) ) {
$m = $this->makeIsoMonth( $bits['F'] );
@@ -246,11 +249,11 @@ class DateFormatter {
}
}
- if ( !isset($bits['d']) ) {
+ if ( !isset( $bits['d'] ) ) {
$bits['d'] = sprintf( '%02d', $bits['j'] );
}
- for ( $p=0; $p < strlen( $format ); $p++ ) {
+ for ( $p = 0; $p < strlen( $format ); $p++ ) {
$char = $format[$p];
switch ( $char ) {
case 'd': # ISO day of month
@@ -263,7 +266,7 @@ class DateFormatter {
$text .= $bits['y'];
break;
case 'j': # ordinary day of month
- if ( !isset($bits['j']) ) {
+ if ( !isset( $bits['j'] ) ) {
$text .= intval( $bits['d'] );
} else {
$text .= $bits['j'];
@@ -271,7 +274,7 @@ class DateFormatter {
break;
case 'F': # long month
if ( !isset( $bits['F'] ) ) {
- $m = intval($bits['m']);
+ $m = intval( $bits['m'] );
if ( $m > 12 || $m < 1 ) {
$fail = true;
} else {
@@ -293,8 +296,9 @@ class DateFormatter {
}
$isoBits = array();
- if ( isset($bits['y']) )
+ if ( isset( $bits['y'] ) ) {
$isoBits[] = $bits['y'];
+ }
$isoBits[] = $bits['m'];
$isoBits[] = $bits['d'];
$isoDate = implode( '-', $isoBits );
@@ -312,7 +316,7 @@ class DateFormatter {
*/
function getMonthRegex() {
$names = array();
- for( $i = 1; $i <= 12; $i++ ) {
+ for ( $i = 1; $i <= 12; $i++ ) {
$names[] = $this->lang->getMonthName( $i );
$names[] = $this->lang->getMonthAbbreviation( $i );
}
@@ -321,7 +325,7 @@ class DateFormatter {
/**
* Makes an ISO month, e.g. 02, from a month name
- * @param $monthName String: month name
+ * @param string $monthName month name
* @return string ISO month name
*/
function makeIsoMonth( $monthName ) {
@@ -331,13 +335,13 @@ class DateFormatter {
/**
* @todo document
- * @param $year String: Year name
+ * @param string $year Year name
* @return string ISO year name
*/
function makeIsoYear( $year ) {
# Assumes the year is in a nice format, as enforced by the regex
if ( substr( $year, -2 ) == 'BC' ) {
- $num = intval(substr( $year, 0, -3 )) - 1;
+ $num = intval( substr( $year, 0, -3 ) ) - 1;
# PHP bug note: sprintf( "%04d", -1 ) fails poorly
$text = sprintf( '-%04d', $num );
@@ -353,7 +357,7 @@ class DateFormatter {
*/
function makeNormalYear( $iso ) {
if ( $iso[0] == '-' ) {
- $text = (intval( substr( $iso, 1 ) ) + 1) . ' BC';
+ $text = ( intval( substr( $iso, 1 ) ) + 1 ) . ' BC';
} else {
$text = intval( $iso );
}
diff --git a/includes/parser/LinkHolderArray.php b/includes/parser/LinkHolderArray.php
index d9356b48..f1a0b258 100644
--- a/includes/parser/LinkHolderArray.php
+++ b/includes/parser/LinkHolderArray.php
@@ -43,9 +43,9 @@ class LinkHolderArray {
}
}
- /**
+ /**
* Don't serialize the parent object, it is big, and not needed when it is
- * a parameter to mergeForeign(), which is the only application of
+ * a parameter to mergeForeign(), which is the only application of
* serializing at present.
*
* Compact the titles, only serialize the text form.
@@ -103,15 +103,15 @@ class LinkHolderArray {
}
/**
- * Merge a LinkHolderArray from another parser instance into this one. The
- * keys will not be preserved. Any text which went with the old
- * LinkHolderArray and needs to work with the new one should be passed in
+ * Merge a LinkHolderArray from another parser instance into this one. The
+ * keys will not be preserved. Any text which went with the old
+ * LinkHolderArray and needs to work with the new one should be passed in
* the $texts array. The strings in this array will have their link holders
* converted for use in the destination link holder. The resulting array of
* strings will be returned.
*
* @param $other LinkHolderArray
- * @param $texts Array of strings
+ * @param array $texts of strings
* @return Array
*/
function mergeForeign( $other, $texts ) {
@@ -126,7 +126,7 @@ class LinkHolderArray {
$maxId = $newKey > $maxId ? $newKey : $maxId;
}
}
- $texts = preg_replace_callback( '/(<!--LINK \d+:)(\d+)(-->)/',
+ $texts = preg_replace_callback( '/(<!--LINK \d+:)(\d+)(-->)/',
array( $this, 'mergeForeignCallback' ), $texts );
# Renumber interwiki links
@@ -135,7 +135,7 @@ class LinkHolderArray {
$this->interwikis[$newKey] = $entry;
$maxId = $newKey > $maxId ? $newKey : $maxId;
}
- $texts = preg_replace_callback( '/(<!--IWLINK )(\d+)(-->)/',
+ $texts = preg_replace_callback( '/(<!--IWLINK )(\d+)(-->)/',
array( $this, 'mergeForeignCallback' ), $texts );
# Set the parent link ID to be beyond the highest used ID
@@ -159,8 +159,8 @@ class LinkHolderArray {
# Internal links
$pos = 0;
while ( $pos < strlen( $text ) ) {
- if ( !preg_match( '/<!--LINK (\d+):(\d+)-->/',
- $text, $m, PREG_OFFSET_CAPTURE, $pos ) )
+ if ( !preg_match( '/<!--LINK (\d+):(\d+)-->/',
+ $text, $m, PREG_OFFSET_CAPTURE, $pos ) )
{
break;
}
@@ -210,14 +210,14 @@ class LinkHolderArray {
*
* @param $nt Title
* @param $text String
- * @param $query Array [optional]
- * @param $trail String [optional]
- * @param $prefix String [optional]
+ * @param array $query [optional]
+ * @param string $trail [optional]
+ * @param string $prefix [optional]
* @return string
*/
- function makeHolder( $nt, $text = '', $query = array(), $trail = '', $prefix = '' ) {
+ function makeHolder( $nt, $text = '', $query = array(), $trail = '', $prefix = '' ) {
wfProfileIn( __METHOD__ );
- if ( ! is_object($nt) ) {
+ if ( !is_object( $nt ) ) {
# Fail gracefully
$retVal = "<!-- ERROR -->{$prefix}{$text}{$trail}";
} else {
@@ -226,7 +226,7 @@ class LinkHolderArray {
$entry = array(
'title' => $nt,
- 'text' => $prefix.$text.$inside,
+ 'text' => $prefix . $text . $inside,
'pdbk' => $nt->getPrefixedDBkey(),
);
if ( $query !== array() ) {
@@ -254,12 +254,12 @@ class LinkHolderArray {
* @todo FIXME: Update documentation. makeLinkObj() is deprecated.
* Replace <!--LINK--> link placeholders with actual links, in the buffer
* Placeholders created in Skin::makeLinkObj()
- * Returns an array of link CSS classes, indexed by PDBK.
+ * @return array of link CSS classes, indexed by PDBK.
*/
function replace( &$text ) {
wfProfileIn( __METHOD__ );
- $colours = $this->replaceInternal( $text );
+ $colours = $this->replaceInternal( $text ); // FIXME: replaceInternal doesn't return a value
$this->replaceInterwiki( $text );
wfProfileOut( __METHOD__ );
@@ -281,93 +281,95 @@ class LinkHolderArray {
$linkCache = LinkCache::singleton();
$output = $this->parent->getOutput();
- wfProfileIn( __METHOD__.'-check' );
- $dbr = wfGetDB( DB_SLAVE );
- $threshold = $this->parent->getOptions()->getStubThreshold();
+ if ( $linkCache->useDatabase() ) {
+ wfProfileIn( __METHOD__ . '-check' );
+ $dbr = wfGetDB( DB_SLAVE );
+ $threshold = $this->parent->getOptions()->getStubThreshold();
- # Sort by namespace
- ksort( $this->internals );
+ # Sort by namespace
+ ksort( $this->internals );
- $linkcolour_ids = array();
+ $linkcolour_ids = array();
- # Generate query
- $queries = array();
- foreach ( $this->internals as $ns => $entries ) {
- foreach ( $entries as $entry ) {
- $title = $entry['title'];
- $pdbk = $entry['pdbk'];
+ # Generate query
+ $queries = array();
+ foreach ( $this->internals as $ns => $entries ) {
+ foreach ( $entries as $entry ) {
+ $title = $entry['title'];
+ $pdbk = $entry['pdbk'];
- # Skip invalid entries.
- # Result will be ugly, but prevents crash.
- if ( is_null( $title ) ) {
- continue;
- }
+ # Skip invalid entries.
+ # Result will be ugly, but prevents crash.
+ if ( is_null( $title ) ) {
+ continue;
+ }
- # Check if it's a static known link, e.g. interwiki
- if ( $title->isAlwaysKnown() ) {
- $colours[$pdbk] = '';
- } elseif ( $ns == NS_SPECIAL ) {
- $colours[$pdbk] = 'new';
- } elseif ( ( $id = $linkCache->getGoodLinkID( $pdbk ) ) != 0 ) {
- $colours[$pdbk] = Linker::getLinkColour( $title, $threshold );
- $output->addLink( $title, $id );
- $linkcolour_ids[$id] = $pdbk;
- } elseif ( $linkCache->isBadLink( $pdbk ) ) {
- $colours[$pdbk] = 'new';
- } else {
- # Not in the link cache, add it to the query
- $queries[$ns][] = $title->getDBkey();
+ # Check if it's a static known link, e.g. interwiki
+ if ( $title->isAlwaysKnown() ) {
+ $colours[$pdbk] = '';
+ } elseif ( $ns == NS_SPECIAL ) {
+ $colours[$pdbk] = 'new';
+ } elseif ( ( $id = $linkCache->getGoodLinkID( $pdbk ) ) != 0 ) {
+ $colours[$pdbk] = Linker::getLinkColour( $title, $threshold );
+ $output->addLink( $title, $id );
+ $linkcolour_ids[$id] = $pdbk;
+ } elseif ( $linkCache->isBadLink( $pdbk ) ) {
+ $colours[$pdbk] = 'new';
+ } else {
+ # Not in the link cache, add it to the query
+ $queries[$ns][] = $title->getDBkey();
+ }
}
}
- }
- if ( $queries ) {
- $where = array();
- foreach( $queries as $ns => $pages ){
- $where[] = $dbr->makeList(
- array(
- 'page_namespace' => $ns,
- 'page_title' => $pages,
- ),
- LIST_AND
- );
- }
+ if ( $queries ) {
+ $where = array();
+ foreach ( $queries as $ns => $pages ) {
+ $where[] = $dbr->makeList(
+ array(
+ 'page_namespace' => $ns,
+ 'page_title' => $pages,
+ ),
+ LIST_AND
+ );
+ }
- $res = $dbr->select(
- 'page',
- array( 'page_id', 'page_namespace', 'page_title', 'page_is_redirect', 'page_len', 'page_latest' ),
- $dbr->makeList( $where, LIST_OR ),
- __METHOD__
- );
+ $res = $dbr->select(
+ 'page',
+ array( 'page_id', 'page_namespace', 'page_title', 'page_is_redirect', 'page_len', 'page_latest' ),
+ $dbr->makeList( $where, LIST_OR ),
+ __METHOD__
+ );
- # Fetch data and form into an associative array
- # non-existent = broken
- foreach ( $res as $s ) {
- $title = Title::makeTitle( $s->page_namespace, $s->page_title );
- $pdbk = $title->getPrefixedDBkey();
- $linkCache->addGoodLinkObjFromRow( $title, $s );
- $output->addLink( $title, $s->page_id );
- # @todo FIXME: Convoluted data flow
- # The redirect status and length is passed to getLinkColour via the LinkCache
- # Use formal parameters instead
- $colours[$pdbk] = Linker::getLinkColour( $title, $threshold );
- //add id to the extension todolist
- $linkcolour_ids[$s->page_id] = $pdbk;
+ # Fetch data and form into an associative array
+ # non-existent = broken
+ foreach ( $res as $s ) {
+ $title = Title::makeTitle( $s->page_namespace, $s->page_title );
+ $pdbk = $title->getPrefixedDBkey();
+ $linkCache->addGoodLinkObjFromRow( $title, $s );
+ $output->addLink( $title, $s->page_id );
+ # @todo FIXME: Convoluted data flow
+ # The redirect status and length is passed to getLinkColour via the LinkCache
+ # Use formal parameters instead
+ $colours[$pdbk] = Linker::getLinkColour( $title, $threshold );
+ //add id to the extension todolist
+ $linkcolour_ids[$s->page_id] = $pdbk;
+ }
+ unset( $res );
}
- unset( $res );
- }
- if ( count($linkcolour_ids) ) {
- //pass an array of page_ids to an extension
- wfRunHooks( 'GetLinkColours', array( $linkcolour_ids, &$colours ) );
+ if ( count( $linkcolour_ids ) ) {
+ //pass an array of page_ids to an extension
+ wfRunHooks( 'GetLinkColours', array( $linkcolour_ids, &$colours ) );
+ }
+ wfProfileOut( __METHOD__ . '-check' );
}
- wfProfileOut( __METHOD__.'-check' );
# Do a second query for different language variants of links and categories
- if($wgContLang->hasVariants()) {
+ if ( $wgContLang->hasVariants() ) {
$this->doVariants( $colours );
}
# Construct search and replace arrays
- wfProfileIn( __METHOD__.'-construct' );
+ wfProfileIn( __METHOD__ . '-construct' );
$replacePairs = array();
foreach ( $this->internals as $ns => $entries ) {
foreach ( $entries as $index => $entry ) {
@@ -377,6 +379,10 @@ class LinkHolderArray {
$key = "$ns:$index";
$searchkey = "<!--LINK $key-->";
$displayText = $entry['text'];
+ if ( isset( $entry['selflink'] ) ) {
+ $replacePairs[$searchkey] = Linker::makeSelfLinkObj( $title, $displayText, $query );
+ continue;
+ }
if ( $displayText === '' ) {
$displayText = null;
}
@@ -399,16 +405,17 @@ class LinkHolderArray {
}
}
$replacer = new HashtableReplacer( $replacePairs, 1 );
- wfProfileOut( __METHOD__.'-construct' );
+ wfProfileOut( __METHOD__ . '-construct' );
# Do the thing
- wfProfileIn( __METHOD__.'-replace' );
+ wfProfileIn( __METHOD__ . '-replace' );
$text = preg_replace_callback(
'/(<!--LINK .*?-->)/',
$replacer->cb(),
- $text);
+ $text
+ );
- wfProfileOut( __METHOD__.'-replace' );
+ wfProfileOut( __METHOD__ . '-replace' );
wfProfileOut( __METHOD__ );
}
@@ -424,7 +431,7 @@ class LinkHolderArray {
# Make interwiki link HTML
$output = $this->parent->getOutput();
$replacePairs = array();
- foreach( $this->interwikis as $key => $link ) {
+ foreach ( $this->interwikis as $key => $link ) {
$replacePairs[$key] = Linker::link( $link['title'], $link['text'] );
$output->addInterwikiLink( $link['title'] );
}
@@ -454,20 +461,17 @@ class LinkHolderArray {
// single string to all variants. This would improve parser's performance
// significantly.
foreach ( $this->internals as $ns => $entries ) {
+ if ( $ns == NS_SPECIAL ) {
+ continue;
+ }
foreach ( $entries as $index => $entry ) {
$pdbk = $entry['pdbk'];
// we only deal with new links (in its first query)
if ( !isset( $colours[$pdbk] ) || $colours[$pdbk] === 'new' ) {
- $title = $entry['title'];
- $titleText = $title->getText();
- $titlesAttrs[] = array(
- 'ns' => $ns,
- 'key' => "$ns:$index",
- 'titleText' => $titleText,
- );
+ $titlesAttrs[] = array( $index, $entry['title'] );
// separate titles with \0 because it would never appears
// in a valid title
- $titlesToBeConverted .= $titleText . "\0";
+ $titlesToBeConverted .= $entry['title']->getText() . "\0";
}
}
}
@@ -478,39 +482,58 @@ class LinkHolderArray {
foreach ( $titlesAllVariants as &$titlesVariant ) {
$titlesVariant = explode( "\0", $titlesVariant );
}
- $l = count( $titlesAttrs );
+
// Then add variants of links to link batch
- for ( $i = 0; $i < $l; $i ++ ) {
+ $parentTitle = $this->parent->getTitle();
+ foreach ( $titlesAttrs as $i => $attrs ) {
+ list( $index, $title ) = $attrs;
+ $ns = $title->getNamespace();
+ $text = $title->getText();
+
foreach ( $allVariantsName as $variantName ) {
$textVariant = $titlesAllVariants[$variantName][$i];
- if ( $textVariant != $titlesAttrs[$i]['titleText'] ) {
- $variantTitle = Title::makeTitle( $titlesAttrs[$i]['ns'], $textVariant );
- if( is_null( $variantTitle ) ) {
- continue;
- }
- $linkBatch->addObj( $variantTitle );
- $variantMap[$variantTitle->getPrefixedDBkey()][] = $titlesAttrs[$i]['key'];
+ if ( $textVariant === $text ) {
+ continue;
+ }
+
+ $variantTitle = Title::makeTitle( $ns, $textVariant );
+ if ( is_null( $variantTitle ) ) {
+ continue;
+ }
+
+ // Self-link checking for mixed/different variant titles. At this point, we
+ // already know the exact title does not exist, so the link cannot be to a
+ // variant of the current title that exists as a separate page.
+ if ( $variantTitle->equals( $parentTitle ) && $title->getFragment() === '' ) {
+ $this->internals[$ns][$index]['selflink'] = true;
+ continue 2;
}
+
+ $linkBatch->addObj( $variantTitle );
+ $variantMap[$variantTitle->getPrefixedDBkey()][] = "$ns:$index";
}
}
// process categories, check if a category exists in some variant
$categoryMap = array(); // maps $category_variant => $category (dbkeys)
$varCategories = array(); // category replacements oldDBkey => newDBkey
- foreach( $output->getCategoryLinks() as $category ){
+ foreach ( $output->getCategoryLinks() as $category ) {
+ $categoryTitle = Title::makeTitleSafe( NS_CATEGORY, $category );
+ $linkBatch->addObj( $categoryTitle );
$variants = $wgContLang->autoConvertToAllVariants( $category );
- foreach($variants as $variant){
- if($variant != $category){
- $variantTitle = Title::newFromDBkey( Title::makeName(NS_CATEGORY,$variant) );
- if(is_null($variantTitle)) continue;
+ foreach ( $variants as $variant ) {
+ if ( $variant !== $category ) {
+ $variantTitle = Title::makeTitleSafe( NS_CATEGORY, $variant );
+ if ( is_null( $variantTitle ) ) {
+ continue;
+ }
$linkBatch->addObj( $variantTitle );
- $categoryMap[$variant] = $category;
+ $categoryMap[$variant] = array( $category, $categoryTitle );
}
}
}
-
- if(!$linkBatch->isEmpty()){
+ if ( !$linkBatch->isEmpty() ) {
// construct query
$dbr = wfGetDB( DB_SLAVE );
$varRes = $dbr->select( 'page',
@@ -529,14 +552,14 @@ class LinkHolderArray {
$vardbk = $variantTitle->getDBkey();
$holderKeys = array();
- if( isset( $variantMap[$varPdbk] ) ) {
+ if ( isset( $variantMap[$varPdbk] ) ) {
$holderKeys = $variantMap[$varPdbk];
$linkCache->addGoodLinkObjFromRow( $variantTitle, $s );
$output->addLink( $variantTitle, $s->page_id );
}
// loop over link holders
- foreach( $holderKeys as $key ) {
+ foreach ( $holderKeys as $key ) {
list( $ns, $index ) = explode( ':', $key, 2 );
$entry =& $this->internals[$ns][$index];
$pdbk = $entry['pdbk'];
@@ -556,25 +579,28 @@ class LinkHolderArray {
}
// check if the object is a variant of a category
- if(isset($categoryMap[$vardbk])){
- $oldkey = $categoryMap[$vardbk];
- if($oldkey != $vardbk)
- $varCategories[$oldkey]=$vardbk;
+ if ( isset( $categoryMap[$vardbk] ) ) {
+ list( $oldkey, $oldtitle ) = $categoryMap[$vardbk];
+ if ( !isset( $varCategories[$oldkey] ) && !$oldtitle->exists() ) {
+ $varCategories[$oldkey] = $vardbk;
+ }
}
}
wfRunHooks( 'GetLinkColours', array( $linkcolour_ids, &$colours ) );
// rebuild the categories in original order (if there are replacements)
- if(count($varCategories)>0){
+ if ( count( $varCategories ) > 0 ) {
$newCats = array();
$originalCats = $output->getCategories();
- foreach($originalCats as $cat => $sortkey){
+ foreach ( $originalCats as $cat => $sortkey ) {
// make the replacement
- if( array_key_exists($cat,$varCategories) )
+ if ( array_key_exists( $cat, $varCategories ) ) {
$newCats[$varCategories[$cat]] = $sortkey;
- else $newCats[$cat] = $sortkey;
+ } else {
+ $newCats[$cat] = $sortkey;
+ }
}
- $output->setCategoryLinks($newCats);
+ $output->setCategoryLinks( $newCats );
}
}
}
@@ -607,14 +633,14 @@ class LinkHolderArray {
*/
function replaceTextCallback( $matches ) {
$type = $matches[1];
- $key = $matches[2];
- if( $type == 'LINK' ) {
+ $key = $matches[2];
+ if ( $type == 'LINK' ) {
list( $ns, $index ) = explode( ':', $key, 2 );
- if( isset( $this->internals[$ns][$index]['text'] ) ) {
+ if ( isset( $this->internals[$ns][$index]['text'] ) ) {
return $this->internals[$ns][$index]['text'];
}
- } elseif( $type == 'IWLINK' ) {
- if( isset( $this->interwikis[$key]['text'] ) ) {
+ } elseif ( $type == 'IWLINK' ) {
+ if ( isset( $this->interwikis[$key]['text'] ) ) {
return $this->interwikis[$key]['text'];
}
}
diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php
index 10765de2..1f14223d 100644
--- a/includes/parser/Parser.php
+++ b/includes/parser/Parser.php
@@ -28,7 +28,7 @@
/**
* PHP Parser - Processes wiki markup (which uses a more user-friendly
* syntax, such as "[[link]]" for making links), and provides a one-way
- * transformation of that wiki markup it into XHTML output / markup
+ * transformation of that wiki markup it into (X)HTML output / markup
* (which in turn the browser understands, and can display).
*
* There are seven main entry points into the Parser class:
@@ -54,7 +54,6 @@
* @warning $wgUser or $wgTitle or $wgRequest or $wgLang. Keep them away!
*
* @par Settings:
- * $wgLocaltimezone
* $wgNamespacesWithSubpages
*
* @par Settings only within ParserOptions:
@@ -62,7 +61,6 @@
* $wgAllowSpecialInclusion
* $wgInterwikiMagic
* $wgMaxArticleSize
- * $wgUseDynamicDates
*
* @ingroup Parser
*/
@@ -117,14 +115,18 @@ class Parser {
# Marker Suffix needs to be accessible staticly.
const MARKER_SUFFIX = "-QINU\x7f";
+ # Markers used for wrapping the table of contents
+ const TOC_START = '<mw:toc>';
+ const TOC_END = '</mw:toc>';
+
# Persistent:
var $mTagHooks = array();
var $mTransparentTagHooks = array();
var $mFunctionHooks = array();
var $mFunctionSynonyms = array( 0 => array(), 1 => array() );
var $mFunctionTagHooks = array();
- var $mStripList = array();
- var $mDefaultStripList = array();
+ var $mStripList = array();
+ var $mDefaultStripList = array();
var $mVarCache = array();
var $mImageParams = array();
var $mImageParamsMagicArray = array();
@@ -193,7 +195,9 @@ class Parser {
var $mRevisionId; # ID to display in {{REVISIONID}} tags
var $mRevisionTimestamp; # The timestamp of the specified revision ID
var $mRevisionUser; # User to display in {{REVISIONUSER}} tag
+ var $mRevisionSize; # Size to display in {{REVISIONSIZE}} variable
var $mRevIdForTs; # The revision ID which was used to fetch the timestamp
+ var $mInputSize = false; # For {{PAGESIZE}} on current page.
/**
* @var string
@@ -201,6 +205,13 @@ class Parser {
var $mUniqPrefix;
/**
+ * @var Array with the language name of each language link (i.e. the
+ * interwiki prefix) in the key, value arbitrary. Used to avoid sending
+ * duplicate language links to the ParserOutput.
+ */
+ var $mLangLinkLanguages;
+
+ /**
* Constructor
*
* @param $conf array
@@ -208,12 +219,12 @@ class Parser {
public function __construct( $conf = array() ) {
$this->mConf = $conf;
$this->mUrlProtocols = wfUrlProtocols();
- $this->mExtLinkBracketedRegex = '/\[(((?i)' . $this->mUrlProtocols . ')'.
- self::EXT_LINK_URL_CLASS.'+)\p{Zs}*([^\]\\x00-\\x08\\x0a-\\x1F]*?)\]/Su';
+ $this->mExtLinkBracketedRegex = '/\[(((?i)' . $this->mUrlProtocols . ')' .
+ self::EXT_LINK_URL_CLASS . '+)\p{Zs}*([^\]\\x00-\\x08\\x0a-\\x1F]*?)\]/Su';
if ( isset( $conf['preprocessorClass'] ) ) {
$this->mPreprocessorClass = $conf['preprocessorClass'];
- } elseif ( defined( 'MW_COMPILED' ) ) {
- # Preprocessor_Hash is much faster than Preprocessor_DOM in compiled mode
+ } elseif ( defined( 'HPHP_VERSION' ) ) {
+ # Preprocessor_Hash is much faster than Preprocessor_DOM under HipHop
$this->mPreprocessorClass = 'Preprocessor_Hash';
} elseif ( extension_loaded( 'domxml' ) ) {
# PECL extension that conflicts with the core DOM extension (bug 13770)
@@ -240,6 +251,13 @@ class Parser {
}
/**
+ * Allow extensions to clean up when the parser is cloned
+ */
+ function __clone() {
+ wfRunHooks( 'ParserCloned', array( $this ) );
+ }
+
+ /**
* Do various kinds of initialisation on the first call of the parser
*/
function firstCallInit() {
@@ -279,9 +297,10 @@ class Parser {
$this->mLinkHolders = new LinkHolderArray( $this );
$this->mLinkID = 0;
$this->mRevisionObject = $this->mRevisionTimestamp =
- $this->mRevisionId = $this->mRevisionUser = null;
+ $this->mRevisionId = $this->mRevisionUser = $this->mRevisionSize = null;
$this->mVarCache = array();
$this->mUser = null;
+ $this->mLangLinkLanguages = array();
/**
* Prefix for temporary replacement strings for the multipass parser.
@@ -291,12 +310,11 @@ class Parser {
* string constructs.
*
* Must not consist of all title characters, or else it will change
- * the behaviour of <nowiki> in a link.
+ * the behavior of <nowiki> in a link.
*/
$this->mUniqPrefix = "\x7fUNIQ" . self::getRandomString();
$this->mStripState = new StripState( $this->mUniqPrefix );
-
# Clear these on every parse, bug 4549
$this->mTplExpandCache = $this->mTplRedirCache = $this->mTplDomCache = array();
@@ -327,12 +345,12 @@ class Parser {
* Convert wikitext to HTML
* Do not call this function recursively.
*
- * @param $text String: text we want to parse
+ * @param string $text text we want to parse
* @param $title Title object
* @param $options ParserOptions
* @param $linestart boolean
* @param $clearState boolean
- * @param $revid Int: number to pass in {{REVISIONID}}
+ * @param int $revid number to pass in {{REVISIONID}}
* @return ParserOutput a ParserOutput
*/
public function parse( $text, Title $title, ParserOptions $options, $linestart = true, $clearState = true, $revid = null ) {
@@ -341,13 +359,18 @@ class Parser {
* to internalParse() which does all the real work.
*/
- global $wgUseTidy, $wgAlwaysUseTidy;
- $fname = __METHOD__.'-' . wfGetCaller();
+ global $wgUseTidy, $wgAlwaysUseTidy, $wgShowHostnames;
+ $fname = __METHOD__ . '-' . wfGetCaller();
wfProfileIn( __METHOD__ );
wfProfileIn( $fname );
$this->startParse( $title, $options, self::OT_HTML, $clearState );
+ $this->mInputSize = strlen( $text );
+ if ( $this->mOptions->getEnableLimitReport() ) {
+ $this->mOutput->resetParseStartTime();
+ }
+
# Remove the strip marker tag prefix from the input, if present.
if ( $clearState ) {
$text = str_replace( $this->mUniqPrefix, '', $text );
@@ -357,11 +380,13 @@ class Parser {
$oldRevisionObject = $this->mRevisionObject;
$oldRevisionTimestamp = $this->mRevisionTimestamp;
$oldRevisionUser = $this->mRevisionUser;
+ $oldRevisionSize = $this->mRevisionSize;
if ( $revid !== null ) {
$this->mRevisionId = $revid;
$this->mRevisionObject = null;
$this->mRevisionTimestamp = null;
$this->mRevisionUser = null;
+ $this->mRevisionSize = null;
}
wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) );
@@ -397,9 +422,7 @@ class Parser {
if ( !( $options->getDisableContentConversion()
|| isset( $this->mDoubleUnderscores['nocontentconvert'] ) ) )
{
- # Run convert unconditionally in 1.18-compatible mode
- global $wgBug34832TransitionalRollback;
- if ( $wgBug34832TransitionalRollback || !$this->mOptions->getInterfaceMessage() ) {
+ if ( !$this->mOptions->getInterfaceMessage() ) {
# 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.
@@ -479,23 +502,73 @@ class Parser {
# Information on include size limits, for the benefit of users who try to skirt them
if ( $this->mOptions->getEnableLimitReport() ) {
$max = $this->mOptions->getMaxIncludeSize();
- $PFreport = "Expensive parser function count: {$this->mExpensiveFunctionCount}/{$this->mOptions->getExpensiveParserFunctionLimit()}\n";
- $limitReport =
- "NewPP limit report\n" .
- "Preprocessor visited node count: {$this->mPPNodeCount}/{$this->mOptions->getMaxPPNodeCount()}\n" .
- "Preprocessor generated node count: " .
- "{$this->mGeneratedPPNodeCount}/{$this->mOptions->getMaxGeneratedPPNodeCount()}\n" .
- "Post-expand include size: {$this->mIncludeSizes['post-expand']}/$max bytes\n" .
- "Template argument size: {$this->mIncludeSizes['arg']}/$max bytes\n".
- "Highest expansion depth: {$this->mHighestExpansionDepth}/{$this->mOptions->getMaxPPExpandDepth()}\n".
- $PFreport;
+
+ $cpuTime = $this->mOutput->getTimeSinceStart( 'cpu' );
+ if ( $cpuTime !== null ) {
+ $this->mOutput->setLimitReportData( 'limitreport-cputime',
+ sprintf( "%.3f", $cpuTime )
+ );
+ }
+
+ $wallTime = $this->mOutput->getTimeSinceStart( 'wall' );
+ $this->mOutput->setLimitReportData( 'limitreport-walltime',
+ sprintf( "%.3f", $wallTime )
+ );
+
+ $this->mOutput->setLimitReportData( 'limitreport-ppvisitednodes',
+ array( $this->mPPNodeCount, $this->mOptions->getMaxPPNodeCount() )
+ );
+ $this->mOutput->setLimitReportData( 'limitreport-ppgeneratednodes',
+ array( $this->mGeneratedPPNodeCount, $this->mOptions->getMaxGeneratedPPNodeCount() )
+ );
+ $this->mOutput->setLimitReportData( 'limitreport-postexpandincludesize',
+ array( $this->mIncludeSizes['post-expand'], $max )
+ );
+ $this->mOutput->setLimitReportData( 'limitreport-templateargumentsize',
+ array( $this->mIncludeSizes['arg'], $max )
+ );
+ $this->mOutput->setLimitReportData( 'limitreport-expansiondepth',
+ array( $this->mHighestExpansionDepth, $this->mOptions->getMaxPPExpandDepth() )
+ );
+ $this->mOutput->setLimitReportData( 'limitreport-expensivefunctioncount',
+ array( $this->mExpensiveFunctionCount, $this->mOptions->getExpensiveParserFunctionLimit() )
+ );
+ wfRunHooks( 'ParserLimitReportPrepare', array( $this, $this->mOutput ) );
+
+ $limitReport = "NewPP limit report\n";
+ if ( $wgShowHostnames ) {
+ $limitReport .= 'Parsed by ' . wfHostname() . "\n";
+ }
+ foreach ( $this->mOutput->getLimitReportData() as $key => $value ) {
+ if ( wfRunHooks( 'ParserLimitReportFormat',
+ array( $key, $value, &$limitReport, false, false )
+ ) ) {
+ $keyMsg = wfMessage( $key )->inLanguage( 'en' )->useDatabase( false );
+ $valueMsg = wfMessage( array( "$key-value-text", "$key-value" ) )
+ ->inLanguage( 'en' )->useDatabase( false );
+ if ( !$valueMsg->exists() ) {
+ $valueMsg = new RawMessage( '$1' );
+ }
+ if ( !$keyMsg->isDisabled() && !$valueMsg->isDisabled() ) {
+ $valueMsg->params( $value );
+ $limitReport .= "{$keyMsg->text()}: {$valueMsg->text()}\n";
+ }
+ }
+ }
+ // Since we're not really outputting HTML, decode the entities and
+ // then re-encode the things that need hiding inside HTML comments.
+ $limitReport = htmlspecialchars_decode( $limitReport );
wfRunHooks( 'ParserLimitReport', array( $this, &$limitReport ) );
// Sanitize for comment. Note '‐' in the replacement is U+2010,
// which looks much like the problematic '-'.
$limitReport = str_replace( array( '-', '&' ), array( '‐', '&amp;' ), $limitReport );
-
$text .= "\n<!-- \n$limitReport-->\n";
+
+ if ( $this->mGeneratedPPNodeCount > $this->mOptions->getMaxGeneratedPPNodeCount() / 10 ) {
+ wfDebugLog( 'generated-pp-node-count', $this->mGeneratedPPNodeCount . ' ' .
+ $this->mTitle->getPrefixedDBkey() );
+ }
}
$this->mOutput->setText( $text );
@@ -503,6 +576,8 @@ class Parser {
$this->mRevisionObject = $oldRevisionObject;
$this->mRevisionTimestamp = $oldRevisionTimestamp;
$this->mRevisionUser = $oldRevisionUser;
+ $this->mRevisionSize = $oldRevisionSize;
+ $this->mInputSize = false;
wfProfileOut( $fname );
wfProfileOut( __METHOD__ );
@@ -515,12 +590,12 @@ class Parser {
*
* If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded
*
- * @param $text String: text extension wants to have parsed
+ * @param string $text text extension wants to have parsed
* @param $frame PPFrame: The frame to use for expanding any template variables
*
* @return string
*/
- function recursiveTagParse( $text, $frame=false ) {
+ function recursiveTagParse( $text, $frame = false ) {
wfProfileIn( __METHOD__ );
wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) );
wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) );
@@ -534,7 +609,7 @@ class Parser {
* Also removes comments.
* @return mixed|string
*/
- function preprocess( $text, Title $title, ParserOptions $options, $revid = null ) {
+ function preprocess( $text, Title $title = null, ParserOptions $options, $revid = null ) {
wfProfileIn( __METHOD__ );
$this->startParse( $title, $options, self::OT_PREPROCESS, true );
if ( $revid !== null ) {
@@ -552,7 +627,7 @@ class Parser {
* Recursive parser entry point that can be called from an extension tag
* hook.
*
- * @param $text String: text to be expanded
+ * @param string $text text to be expanded
* @param $frame PPFrame: The frame to use for expanding any template variables
* @return String
* @since 1.19
@@ -593,7 +668,7 @@ class Parser {
*
* @return string
*/
- static public function getRandomString() {
+ public static function getRandomString() {
return wfRandomString( 16 );
}
@@ -682,7 +757,7 @@ class Parser {
/**
* Accessor/mutator for the output type
*
- * @param $x int|null New value or null to just get the current one
+ * @param int|null $x New value or null to just get the current one
* @return Integer
*/
function OutputType( $x = null ) {
@@ -745,6 +820,7 @@ class Parser {
*
* @since 1.19
*
+ * @throws MWException
* @return Language|null
*/
public function getTargetLanguage() {
@@ -752,9 +828,9 @@ class Parser {
if ( $target !== null ) {
return $target;
- } elseif( $this->mOptions->getInterfaceMessage() ) {
+ } elseif ( $this->mOptions->getInterfaceMessage() ) {
return $this->mOptions->getUserLangObj();
- } elseif( is_null( $this->mTitle ) ) {
+ } elseif ( is_null( $this->mTitle ) ) {
throw new MWException( __METHOD__ . ': $this->mTitle is null' );
}
@@ -765,12 +841,7 @@ class Parser {
* Get the language object for language conversion
*/
function getConverterLanguage() {
- global $wgBug34832TransitionalRollback, $wgContLang;
- if ( $wgBug34832TransitionalRollback ) {
- return $wgContLang;
- } else {
- return $this->getTargetLanguage();
- }
+ return $this->getTargetLanguage();
}
/**
@@ -813,9 +884,9 @@ class Parser {
* '<element param="x">tag content</element>' ) )
* @endcode
*
- * @param $elements array list of element names. Comments are always extracted.
- * @param $text string Source text string.
- * @param $matches array Out parameter, Array: extracted tags
+ * @param array $elements list of element names. Comments are always extracted.
+ * @param string $text Source text string.
+ * @param array $matches Out parameter, Array: extracted tags
* @param $uniq_prefix string
* @return String: stripped text
*/
@@ -835,16 +906,16 @@ class Parser {
}
if ( count( $p ) > 5 ) {
# comment
- $element = $p[4];
+ $element = $p[4];
$attributes = '';
- $close = '';
- $inside = $p[5];
+ $close = '';
+ $inside = $p[5];
} else {
# tag
- $element = $p[1];
+ $element = $p[1];
$attributes = $p[2];
- $close = $p[3];
- $inside = $p[4];
+ $close = $p[3];
+ $inside = $p[4];
}
$marker = "$uniq_prefix-$element-" . sprintf( '%08X', $n++ ) . self::MARKER_SUFFIX;
@@ -928,33 +999,33 @@ class Parser {
$line = trim( $outLine );
if ( $line === '' ) { # empty line, go to next line
- $out .= $outLine."\n";
+ $out .= $outLine . "\n";
continue;
}
$first_character = $line[0];
$matches = array();
- if ( preg_match( '/^(:*)\{\|(.*)$/', $line , $matches ) ) {
+ if ( preg_match( '/^(:*)\{\|(.*)$/', $line, $matches ) ) {
# 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}>";
- 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 );
+ $attributes = Sanitizer::fixTagAttributes( $attributes, 'table' );
+
+ $outLine = str_repeat( '<dl><dd>', $indent_level ) . "<table{$attributes}>";
+ array_push( $td_history, false );
+ array_push( $last_tag_history, '' );
+ array_push( $tr_history, false );
+ array_push( $tr_attributes, '' );
+ array_push( $has_opened_tr, false );
} elseif ( count( $td_history ) == 0 ) {
# Don't do any of the following
- $out .= $outLine."\n";
+ $out .= $outLine . "\n";
continue;
- } elseif ( substr( $line , 0 , 2 ) === '|}' ) {
+ } elseif ( substr( $line, 0, 2 ) === '|}' ) {
# We are ending a table
- $line = '</table>' . substr( $line , 2 );
+ $line = '</table>' . substr( $line, 2 );
$last_tag = array_pop( $last_tag_history );
if ( !array_pop( $has_opened_tr ) ) {
@@ -969,8 +1040,8 @@ class Parser {
$line = "</{$last_tag}>{$line}";
}
array_pop( $tr_attributes );
- $outLine = $line . str_repeat( '</dd></dl>' , $indent_level );
- } elseif ( substr( $line , 0 , 2 ) === '|-' ) {
+ $outLine = $line . str_repeat( '</dd></dl>', $indent_level );
+ } elseif ( substr( $line, 0, 2 ) === '|-' ) {
# Now we have a table row
$line = preg_replace( '#^\|-+#', '', $line );
@@ -983,7 +1054,7 @@ class Parser {
$line = '';
$last_tag = array_pop( $last_tag_history );
array_pop( $has_opened_tr );
- array_push( $has_opened_tr , true );
+ array_push( $has_opened_tr, true );
if ( array_pop( $tr_history ) ) {
$line = '</tr>';
@@ -994,27 +1065,27 @@ class Parser {
}
$outLine = $line;
- array_push( $tr_history , false );
- array_push( $td_history , false );
- array_push( $last_tag_history , '' );
- } elseif ( $first_character === '|' || $first_character === '!' || substr( $line , 0 , 2 ) === '|+' ) {
+ array_push( $tr_history, false );
+ array_push( $td_history, false );
+ array_push( $last_tag_history, '' );
+ } elseif ( $first_character === '|' || $first_character === '!' || substr( $line, 0, 2 ) === '|+' ) {
# This might be cell elements, td, th or captions
- if ( substr( $line , 0 , 2 ) === '|+' ) {
+ if ( substr( $line, 0, 2 ) === '|+' ) {
$first_character = '+';
- $line = substr( $line , 1 );
+ $line = substr( $line, 1 );
}
- $line = substr( $line , 1 );
+ $line = substr( $line, 1 );
if ( $first_character === '!' ) {
- $line = str_replace( '!!' , '||' , $line );
+ $line = str_replace( '!!', '||', $line );
}
# Split up multiple cells on the same line.
# FIXME : This can result in improper nesting of tags processed
# by earlier parser steps, but should avoid splitting up eg
# attribute values containing literal "||".
- $cells = StringUtils::explodeMarkup( '||' , $line );
+ $cells = StringUtils::explodeMarkup( '||', $line );
$outLine = '';
@@ -1026,10 +1097,10 @@ class Parser {
if ( !array_pop( $tr_history ) ) {
$previous = "<tr{$tr_after}>\n";
}
- array_push( $tr_history , true );
- array_push( $tr_attributes , '' );
+ array_push( $tr_history, true );
+ array_push( $tr_attributes, '' );
array_pop( $has_opened_tr );
- array_push( $has_opened_tr , true );
+ array_push( $has_opened_tr, true );
}
$last_tag = array_pop( $last_tag_history );
@@ -1048,10 +1119,10 @@ class Parser {
$last_tag = '';
}
- array_push( $last_tag_history , $last_tag );
+ array_push( $last_tag_history, $last_tag );
# A cell could contain both parameters and data
- $cell_data = explode( '|' , $cell , 2 );
+ $cell_data = explode( '|', $cell, 2 );
# Bug 553: Note that a '|' inside an invalid link should not
# be mistaken as delimiting cell parameters
@@ -1061,12 +1132,12 @@ class Parser {
$cell = "{$previous}<{$last_tag}>{$cell_data[0]}";
} else {
$attributes = $this->mStripState->unstripBoth( $cell_data[0] );
- $attributes = Sanitizer::fixTagAttributes( $attributes , $last_tag );
+ $attributes = Sanitizer::fixTagAttributes( $attributes, $last_tag );
$cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[1]}";
}
$outLine .= $cell;
- array_push( $td_history , true );
+ array_push( $td_history, true );
}
}
$out .= $outLine . "\n";
@@ -1081,7 +1152,7 @@ class Parser {
$out .= "</tr>\n";
}
if ( !array_pop( $has_opened_tr ) ) {
- $out .= "<tr><td></td></tr>\n" ;
+ $out .= "<tr><td></td></tr>\n";
}
$out .= "</table>\n";
@@ -1122,7 +1193,7 @@ class Parser {
# Hook to suspend the parser in this state
if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
wfProfileOut( __METHOD__ );
- return $text ;
+ return $text;
}
# if $frame is provided, then use $frame for replacing any variables
@@ -1156,17 +1227,13 @@ class Parser {
$text = $this->doDoubleUnderscore( $text );
$text = $this->doHeadings( $text );
- if ( $this->mOptions->getUseDynamicDates() ) {
- $df = DateFormatter::getInstance();
- $text = $df->reformat( $this->mOptions->getDateFormat(), $text );
- }
$text = $this->replaceInternalLinks( $text );
$text = $this->doAllQuotes( $text );
$text = $this->replaceExternalLinks( $text );
# replaceInternalLinks may sometimes leave behind
# absolute URLs, which have to be masked to hide them from replaceExternalLinks
- $text = str_replace( $this->mUniqPrefix.'NOPARSE', '', $text );
+ $text = str_replace( $this->mUniqPrefix . 'NOPARSE', '', $text );
$text = $this->doMagicLinks( $text );
$text = $this->formatHeadings( $text, $origText, $isMain );
@@ -1234,7 +1301,7 @@ class Parser {
$CssClass = 'mw-magiclink-pmid';
$id = $m[4];
} else {
- throw new MWException( __METHOD__.': unrecognised match type "' .
+ throw new MWException( __METHOD__ . ': unrecognised match type "' .
substr( $m[0], 0, 20 ) . '"' );
}
$url = wfMessage( $urlmsg, $id )->inContentLanguage()->text();
@@ -1248,8 +1315,8 @@ class Parser {
'x' => 'X',
));
$titleObj = SpecialPage::getTitleFor( 'Booksources', $num );
- return'<a href="' .
- htmlspecialchars( $titleObj->getLocalUrl() ) .
+ return '<a href="' .
+ htmlspecialchars( $titleObj->getLocalURL() ) .
"\" class=\"internal mw-magiclink-isbn\">ISBN $isbn</a>";
} else {
return $m[0];
@@ -1298,7 +1365,8 @@ class Parser {
if ( $text === false ) {
# Not an image, make a link
$text = Linker::makeExternalLink( $url,
- $this->getConverterLanguage()->markNoConversion($url), true, 'free',
+ $this->getConverterLanguage()->markNoConversion( $url, true ),
+ true, 'free',
$this->getExternalLinkAttribs( $url ) );
# Register it in the output object...
# Replace unnecessary URL escape codes with their equivalent characters
@@ -1309,7 +1377,6 @@ class Parser {
return $text . $trail;
}
-
/**
* Parse headers and return html
*
@@ -1323,8 +1390,7 @@ class Parser {
wfProfileIn( __METHOD__ );
for ( $i = 6; $i >= 1; --$i ) {
$h = str_repeat( '=', $i );
- $text = preg_replace( "/^$h(.+)$h\\s*$/m",
- "<h$i>\\1</h$i>", $text );
+ $text = preg_replace( "/^$h(.+)$h\\s*$/m", "<h$i>\\1</h$i>", $text );
}
wfProfileOut( __METHOD__ );
return $text;
@@ -1345,7 +1411,7 @@ class Parser {
foreach ( $lines as $line ) {
$outtext .= $this->doQuotes( $line ) . "\n";
}
- $outtext = substr( $outtext, 0,-1 );
+ $outtext = substr( $outtext, 0, -1 );
wfProfileOut( __METHOD__ );
return $outtext;
}
@@ -1359,170 +1425,192 @@ class Parser {
*/
public function doQuotes( $text ) {
$arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE );
- if ( count( $arr ) == 1 ) {
+ $countarr = count( $arr );
+ if ( $countarr == 1 ) {
return $text;
- } else {
- # First, do some preliminary work. This may shift some apostrophes from
- # being mark-up to being text. It also counts the number of occurrences
- # of bold and italics mark-ups.
- $numbold = 0;
- $numitalics = 0;
- 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.
- if ( strlen( $arr[$i] ) == 4 ) {
- $arr[$i-1] .= "'";
- $arr[$i] = "'''";
- } elseif ( strlen( $arr[$i] ) > 5 ) {
- # If there are more than 5 apostrophes in a row, assume they're all
- # text except for the last 5.
- $arr[$i-1] .= str_repeat( "'", strlen( $arr[$i] ) - 5 );
- $arr[$i] = "'''''";
- }
- # Count the number of occurrences of bold and italics mark-ups.
- # We are not counting sequences of five apostrophes.
- if ( strlen( $arr[$i] ) == 2 ) {
- $numitalics++;
- } elseif ( strlen( $arr[$i] ) == 3 ) {
- $numbold++;
- } elseif ( strlen( $arr[$i] ) == 5 ) {
- $numitalics++;
- $numbold++;
- }
- }
+ }
+
+ // 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.
+ $numbold = 0;
+ $numitalics = 0;
+ for ( $i = 1; $i < $countarr; $i += 2 ) {
+ $thislen = strlen( $arr[$i] );
+ // If there are ever four apostrophes, assume the first is supposed to
+ // be text, and the remaining three constitute mark-up for bold text.
+ // (bug 13227: ''''foo'''' turns into ' ''' foo ' ''')
+ if ( $thislen == 4 ) {
+ $arr[$i - 1] .= "'";
+ $arr[$i] = "'''";
+ $thislen = 3;
+ } elseif ( $thislen > 5 ) {
+ // If there are more than 5 apostrophes in a row, assume they're all
+ // text except for the last 5.
+ // (bug 13227: ''''''foo'''''' turns into ' ''''' foo ' ''''')
+ $arr[$i - 1] .= str_repeat( "'", $thislen - 5 );
+ $arr[$i] = "'''''";
+ $thislen = 5;
+ }
+ // Count the number of occurrences of bold and italics mark-ups.
+ if ( $thislen == 2 ) {
+ $numitalics++;
+ } elseif ( $thislen == 3 ) {
+ $numbold++;
+ } elseif ( $thislen == 5 ) {
+ $numitalics++;
+ $numbold++;
}
+ }
- # If there is an odd number of both bold and italics, it is likely
- # that one of the bold ones was meant to be an apostrophe followed
- # by italics. Which one we cannot know for certain, but it is more
- # likely to be one that has a single-letter word before it.
- if ( ( $numbold % 2 == 1 ) && ( $numitalics % 2 == 1 ) ) {
- $i = 0;
- $firstsingleletterword = -1;
- $firstmultiletterword = -1;
- $firstspace = -1;
- foreach ( $arr as $r ) {
- if ( ( $i % 2 == 1 ) and ( strlen( $r ) == 3 ) ) {
- $x1 = substr( $arr[$i-1], -1 );
- $x2 = substr( $arr[$i-1], -2, 1 );
- if ( $x1 === ' ' ) {
- if ( $firstspace == -1 ) {
- $firstspace = $i;
- }
- } elseif ( $x2 === ' ') {
- if ( $firstsingleletterword == -1 ) {
- $firstsingleletterword = $i;
- }
- } else {
- if ( $firstmultiletterword == -1 ) {
- $firstmultiletterword = $i;
- }
+ // If there is an odd number of both bold and italics, it is likely
+ // that one of the bold ones was meant to be an apostrophe followed
+ // by italics. Which one we cannot know for certain, but it is more
+ // likely to be one that has a single-letter word before it.
+ if ( ( $numbold % 2 == 1 ) && ( $numitalics % 2 == 1 ) ) {
+ $firstsingleletterword = -1;
+ $firstmultiletterword = -1;
+ $firstspace = -1;
+ for ( $i = 1; $i < $countarr; $i += 2 ) {
+ if ( strlen( $arr[$i] ) == 3 ) {
+ $x1 = substr( $arr[$i - 1], -1 );
+ $x2 = substr( $arr[$i - 1], -2, 1 );
+ if ( $x1 === ' ' ) {
+ if ( $firstspace == -1 ) {
+ $firstspace = $i;
+ }
+ } elseif ( $x2 === ' ' ) {
+ if ( $firstsingleletterword == -1 ) {
+ $firstsingleletterword = $i;
+ // if $firstsingleletterword is set, we don't
+ // look at the other options, so we can bail early.
+ break;
+ }
+ } else {
+ if ( $firstmultiletterword == -1 ) {
+ $firstmultiletterword = $i;
}
}
- $i++;
}
+ }
- # If there is a single-letter word, use it!
- if ( $firstsingleletterword > -1 ) {
- $arr[$firstsingleletterword] = "''";
- $arr[$firstsingleletterword-1] .= "'";
- } elseif ( $firstmultiletterword > -1 ) {
- # If not, but there's a multi-letter word, use that one.
- $arr[$firstmultiletterword] = "''";
- $arr[$firstmultiletterword-1] .= "'";
- } elseif ( $firstspace > -1 ) {
- # ... otherwise use the first one that has neither.
- # (notice that it is possible for all three to be -1 if, for example,
- # there is only one pentuple-apostrophe in the line)
- $arr[$firstspace] = "''";
- $arr[$firstspace-1] .= "'";
- }
+ // If there is a single-letter word, use it!
+ if ( $firstsingleletterword > -1 ) {
+ $arr[$firstsingleletterword] = "''";
+ $arr[$firstsingleletterword - 1] .= "'";
+ } elseif ( $firstmultiletterword > -1 ) {
+ // If not, but there's a multi-letter word, use that one.
+ $arr[$firstmultiletterword] = "''";
+ $arr[$firstmultiletterword - 1] .= "'";
+ } elseif ( $firstspace > -1 ) {
+ // ... otherwise use the first one that has neither.
+ // (notice that it is possible for all three to be -1 if, for example,
+ // there is only one pentuple-apostrophe in the line)
+ $arr[$firstspace] = "''";
+ $arr[$firstspace - 1] .= "'";
}
+ }
- # Now let's actually convert our apostrophic mush to HTML!
- $output = '';
- $buffer = '';
- $state = '';
- $i = 0;
- foreach ( $arr as $r ) {
- if ( ( $i % 2 ) == 0 ) {
- if ( $state === 'both' ) {
- $buffer .= $r;
- } else {
- $output .= $r;
- }
+ // Now let's actually convert our apostrophic mush to HTML!
+ $output = '';
+ $buffer = '';
+ $state = '';
+ $i = 0;
+ foreach ( $arr as $r ) {
+ if ( ( $i % 2 ) == 0 ) {
+ if ( $state === 'both' ) {
+ $buffer .= $r;
} else {
- if ( strlen( $r ) == 2 ) {
- if ( $state === 'i' ) {
- $output .= '</i>'; $state = '';
- } elseif ( $state === 'bi' ) {
- $output .= '</i>'; $state = 'b';
- } elseif ( $state === 'ib' ) {
- $output .= '</b></i><b>'; $state = 'b';
- } elseif ( $state === 'both' ) {
- $output .= '<b><i>'.$buffer.'</i>'; $state = 'b';
- } else { # $state can be 'b' or ''
- $output .= '<i>'; $state .= 'i';
- }
- } elseif ( strlen( $r ) == 3 ) {
- if ( $state === 'b' ) {
- $output .= '</b>'; $state = '';
- } elseif ( $state === 'bi' ) {
- $output .= '</i></b><i>'; $state = 'i';
- } elseif ( $state === 'ib' ) {
- $output .= '</b>'; $state = 'i';
- } elseif ( $state === 'both' ) {
- $output .= '<i><b>'.$buffer.'</b>'; $state = 'i';
- } else { # $state can be 'i' or ''
- $output .= '<b>'; $state .= 'b';
- }
- } elseif ( strlen( $r ) == 5 ) {
- if ( $state === 'b' ) {
- $output .= '</b><i>'; $state = 'i';
- } elseif ( $state === 'i' ) {
- $output .= '</i><b>'; $state = 'b';
- } elseif ( $state === 'bi' ) {
- $output .= '</i></b>'; $state = '';
- } elseif ( $state === 'ib' ) {
- $output .= '</b></i>'; $state = '';
- } elseif ( $state === 'both' ) {
- $output .= '<i><b>'.$buffer.'</b></i>'; $state = '';
- } else { # ($state == '')
- $buffer = ''; $state = 'both';
- }
+ $output .= $r;
+ }
+ } else {
+ $thislen = strlen( $r );
+ if ( $thislen == 2 ) {
+ if ( $state === 'i' ) {
+ $output .= '</i>';
+ $state = '';
+ } elseif ( $state === 'bi' ) {
+ $output .= '</i>';
+ $state = 'b';
+ } elseif ( $state === 'ib' ) {
+ $output .= '</b></i><b>';
+ $state = 'b';
+ } elseif ( $state === 'both' ) {
+ $output .= '<b><i>' . $buffer . '</i>';
+ $state = 'b';
+ } else { // $state can be 'b' or ''
+ $output .= '<i>';
+ $state .= 'i';
+ }
+ } elseif ( $thislen == 3 ) {
+ if ( $state === 'b' ) {
+ $output .= '</b>';
+ $state = '';
+ } elseif ( $state === 'bi' ) {
+ $output .= '</i></b><i>';
+ $state = 'i';
+ } elseif ( $state === 'ib' ) {
+ $output .= '</b>';
+ $state = 'i';
+ } elseif ( $state === 'both' ) {
+ $output .= '<i><b>' . $buffer . '</b>';
+ $state = 'i';
+ } else { // $state can be 'i' or ''
+ $output .= '<b>';
+ $state .= 'b';
+ }
+ } elseif ( $thislen == 5 ) {
+ if ( $state === 'b' ) {
+ $output .= '</b><i>';
+ $state = 'i';
+ } elseif ( $state === 'i' ) {
+ $output .= '</i><b>';
+ $state = 'b';
+ } elseif ( $state === 'bi' ) {
+ $output .= '</i></b>';
+ $state = '';
+ } elseif ( $state === 'ib' ) {
+ $output .= '</b></i>';
+ $state = '';
+ } elseif ( $state === 'both' ) {
+ $output .= '<i><b>' . $buffer . '</b></i>';
+ $state = '';
+ } else { // ($state == '')
+ $buffer = '';
+ $state = 'both';
}
}
- $i++;
- }
- # Now close all remaining tags. Notice that the order is important.
- if ( $state === 'b' || $state === 'ib' ) {
- $output .= '</b>';
}
- if ( $state === 'i' || $state === 'bi' || $state === 'ib' ) {
- $output .= '</i>';
- }
- if ( $state === 'bi' ) {
- $output .= '</b>';
- }
- # There might be lonely ''''', so make sure we have a buffer
- if ( $state === 'both' && $buffer ) {
- $output .= '<b><i>'.$buffer.'</i></b>';
- }
- return $output;
+ $i++;
+ }
+ // Now close all remaining tags. Notice that the order is important.
+ if ( $state === 'b' || $state === 'ib' ) {
+ $output .= '</b>';
+ }
+ if ( $state === 'i' || $state === 'bi' || $state === 'ib' ) {
+ $output .= '</i>';
+ }
+ if ( $state === 'bi' ) {
+ $output .= '</b>';
+ }
+ // There might be lonely ''''', so make sure we have a buffer
+ if ( $state === 'both' && $buffer ) {
+ $output .= '<b><i>' . $buffer . '</i></b>';
}
+ return $output;
}
/**
* Replace external links (REL)
*
* Note: this is all very hackish and the order of execution matters a lot.
- * Make sure to run maintenance/parserTests.php if you change this code.
+ * Make sure to run tests/parserTests.php if you change this code.
*
* @private
*
* @param $text string
*
+ * @throws MWException
* @return string
*/
function replaceExternalLinks( $text ) {
@@ -1530,14 +1618,15 @@ class Parser {
$bits = preg_split( $this->mExtLinkBracketedRegex, $text, -1, PREG_SPLIT_DELIM_CAPTURE );
if ( $bits === false ) {
+ wfProfileOut( __METHOD__ );
throw new MWException( "PCRE needs to be compiled with --enable-unicode-properties in order for MediaWiki to function" );
}
$s = array_shift( $bits );
$i = 0;
- while ( $i<count( $bits ) ) {
+ while ( $i < count( $bits ) ) {
$url = $bits[$i++];
- $protocol = $bits[$i++];
+ $i++; // protocol
$text = $bits[$i++];
$trail = $bits[$i++];
@@ -1595,26 +1684,39 @@ class Parser {
wfProfileOut( __METHOD__ );
return $s;
}
-
+ /**
+ * Get the rel attribute for a particular external link.
+ *
+ * @since 1.21
+ * @param string|bool $url optional URL, to extract the domain from for rel =>
+ * nofollow if appropriate
+ * @param $title Title optional Title, for wgNoFollowNsExceptions lookups
+ * @return string|null rel attribute for $url
+ */
+ public static function getExternalLinkRel( $url = false, $title = null ) {
+ global $wgNoFollowLinks, $wgNoFollowNsExceptions, $wgNoFollowDomainExceptions;
+ $ns = $title ? $title->getNamespace() : false;
+ if ( $wgNoFollowLinks && !in_array( $ns, $wgNoFollowNsExceptions ) &&
+ !wfMatchesDomainList( $url, $wgNoFollowDomainExceptions ) )
+ {
+ return 'nofollow';
+ }
+ return null;
+ }
/**
* Get an associative array of additional HTML attributes appropriate for a
* particular external link. This currently may include rel => nofollow
* (depending on configuration, namespace, and the URL's domain) and/or a
* target attribute (depending on configuration).
*
- * @param $url String|bool optional URL, to extract the domain from for rel =>
+ * @param string|bool $url optional URL, to extract the domain from for rel =>
* nofollow if appropriate
* @return Array associative array of HTML attributes
*/
function getExternalLinkAttribs( $url = false ) {
$attribs = array();
- global $wgNoFollowLinks, $wgNoFollowNsExceptions, $wgNoFollowDomainExceptions;
- $ns = $this->mTitle->getNamespace();
- if ( $wgNoFollowLinks && !in_array( $ns, $wgNoFollowNsExceptions ) &&
- !wfMatchesDomainList( $url, $wgNoFollowDomainExceptions ) )
- {
- $attribs['rel'] = 'nofollow';
- }
+ $attribs['rel'] = self::getExternalLinkRel( $url, $this->mTitle );
+
if ( $this->mOptions->getExternalLinkTarget() ) {
$attribs['target'] = $this->mOptions->getExternalLinkTarget();
}
@@ -1726,6 +1828,8 @@ class Parser {
/**
* Process [[ ]] wikilinks (RIL)
+ * @param $s
+ * @throws MWException
* @return LinkHolderArray
*
* @private
@@ -1733,8 +1837,8 @@ class Parser {
function replaceInternalLinks2( &$s ) {
wfProfileIn( __METHOD__ );
- wfProfileIn( __METHOD__.'-setup' );
- static $tc = FALSE, $e1, $e1_img;
+ wfProfileIn( __METHOD__ . '-setup' );
+ static $tc = false, $e1, $e1_img;
# the % is needed to support urlencoded titles as well
if ( !$tc ) {
$tc = Title::legalChars() . '#%';
@@ -1763,9 +1867,9 @@ class Parser {
}
if ( is_null( $this->mTitle ) ) {
- wfProfileOut( __METHOD__.'-setup' );
+ wfProfileOut( __METHOD__ . '-setup' );
wfProfileOut( __METHOD__ );
- throw new MWException( __METHOD__.": \$this->mTitle is null\n" );
+ throw new MWException( __METHOD__ . ": \$this->mTitle is null\n" );
}
$nottalk = !$this->mTitle->isTalkPage();
@@ -1780,17 +1884,11 @@ class Parser {
$prefix = '';
}
- if ( $this->getConverterLanguage()->hasVariants() ) {
- $selflink = $this->getConverterLanguage()->autoConvertToAllVariants(
- $this->mTitle->getPrefixedText() );
- } else {
- $selflink = array( $this->mTitle->getPrefixedText() );
- }
$useSubpages = $this->areSubpagesAllowed();
- wfProfileOut( __METHOD__.'-setup' );
+ wfProfileOut( __METHOD__ . '-setup' );
# Loop for each link
- for ( ; $line !== false && $line !== null ; $a->next(), $line = $a->current() ) {
+ for ( ; $line !== false && $line !== null; $a->next(), $line = $a->current() ) {
# Check for excessive memory usage
if ( $holders->isBig() ) {
# Too big
@@ -1800,24 +1898,24 @@ class Parser {
}
if ( $useLinkPrefixExtension ) {
- wfProfileIn( __METHOD__.'-prefixhandling' );
+ wfProfileIn( __METHOD__ . '-prefixhandling' );
if ( preg_match( $e2, $s, $m ) ) {
$prefix = $m[2];
$s = $m[1];
} else {
- $prefix='';
+ $prefix = '';
}
# first link
if ( $first_prefix ) {
$prefix = $first_prefix;
$first_prefix = false;
}
- wfProfileOut( __METHOD__.'-prefixhandling' );
+ wfProfileOut( __METHOD__ . '-prefixhandling' );
}
$might_be_img = false;
- wfProfileIn( __METHOD__."-e1" );
+ wfProfileIn( __METHOD__ . "-e1" );
if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
$text = $m[2];
# If we get a ] at the beginning of $m[3] that means we have a link that's something like:
@@ -1839,7 +1937,7 @@ class Parser {
# fix up urlencoded title texts
if ( strpos( $m[1], '%' ) !== false ) {
# Should anchors '#' also be rejected?
- $m[1] = str_replace( array('<', '>'), array('&lt;', '&gt;'), rawurldecode( $m[1] ) );
+ $m[1] = str_replace( array( '<', '>' ), array( '&lt;', '&gt;' ), rawurldecode( $m[1] ) );
}
$trail = $m[3];
} elseif ( preg_match( $e1_img, $line, $m ) ) { # Invalid, but might be an image with a link in its caption
@@ -1850,19 +1948,19 @@ class Parser {
}
$trail = "";
} else { # Invalid form; output directly
- $s .= $prefix . '[[' . $line ;
- wfProfileOut( __METHOD__."-e1" );
+ $s .= $prefix . '[[' . $line;
+ wfProfileOut( __METHOD__ . "-e1" );
continue;
}
- wfProfileOut( __METHOD__."-e1" );
- wfProfileIn( __METHOD__."-misc" );
+ wfProfileOut( __METHOD__ . "-e1" );
+ wfProfileIn( __METHOD__ . "-misc" );
# Don't allow internal links to pages containing
# PROTO: where PROTO is a valid URL protocol; these
# should be external links.
if ( preg_match( '/^(?i:' . $this->mUrlProtocols . ')/', $m[1] ) ) {
- $s .= $prefix . '[[' . $line ;
- wfProfileOut( __METHOD__."-misc" );
+ $s .= $prefix . '[[' . $line;
+ wfProfileOut( __METHOD__ . "-misc" );
continue;
}
@@ -1879,21 +1977,21 @@ class Parser {
$link = substr( $link, 1 );
}
- wfProfileOut( __METHOD__."-misc" );
- wfProfileIn( __METHOD__."-title" );
+ wfProfileOut( __METHOD__ . "-misc" );
+ wfProfileIn( __METHOD__ . "-title" );
$nt = Title::newFromText( $this->mStripState->unstripNoWiki( $link ) );
if ( $nt === null ) {
$s .= $prefix . '[[' . $line;
- wfProfileOut( __METHOD__."-title" );
+ wfProfileOut( __METHOD__ . "-title" );
continue;
}
$ns = $nt->getNamespace();
$iw = $nt->getInterWiki();
- wfProfileOut( __METHOD__."-title" );
+ wfProfileOut( __METHOD__ . "-title" );
if ( $might_be_img ) { # if this is actually an invalid link
- wfProfileIn( __METHOD__."-might_be_img" );
+ wfProfileIn( __METHOD__ . "-might_be_img" );
if ( $ns == NS_FILE && $noforce ) { # but might be an image
$found = false;
while ( true ) {
@@ -1925,19 +2023,19 @@ class Parser {
$holders->merge( $this->replaceInternalLinks2( $text ) );
$s .= "{$prefix}[[$link|$text";
# note: no $trail, because without an end, there *is* no trail
- wfProfileOut( __METHOD__."-might_be_img" );
+ wfProfileOut( __METHOD__ . "-might_be_img" );
continue;
}
} else { # it's not an image, so output it raw
$s .= "{$prefix}[[$link|$text";
# note: no $trail, because without an end, there *is* no trail
- wfProfileOut( __METHOD__."-might_be_img" );
+ wfProfileOut( __METHOD__ . "-might_be_img" );
continue;
}
- wfProfileOut( __METHOD__."-might_be_img" );
+ wfProfileOut( __METHOD__ . "-might_be_img" );
}
- $wasblank = ( $text == '' );
+ $wasblank = ( $text == '' );
if ( $wasblank ) {
$text = $link;
} else {
@@ -1951,18 +2049,25 @@ class Parser {
# Link not escaped by : , create the various objects
if ( $noforce ) {
# Interwikis
- wfProfileIn( __METHOD__."-interwiki" );
+ wfProfileIn( __METHOD__ . "-interwiki" );
if ( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && Language::fetchLanguageName( $iw, null, 'mw' ) ) {
- $this->mOutput->addLanguageLink( $nt->getFullText() );
+ // XXX: the above check prevents links to sites with identifiers that are not language codes
+
+ # Bug 24502: filter duplicates
+ if ( !isset( $this->mLangLinkLanguages[$iw] ) ) {
+ $this->mLangLinkLanguages[$iw] = true;
+ $this->mOutput->addLanguageLink( $nt->getFullText() );
+ }
+
$s = rtrim( $s . $prefix );
$s .= trim( $trail, "\n" ) == '' ? '': $prefix . $trail;
- wfProfileOut( __METHOD__."-interwiki" );
+ wfProfileOut( __METHOD__ . "-interwiki" );
continue;
}
- wfProfileOut( __METHOD__."-interwiki" );
+ wfProfileOut( __METHOD__ . "-interwiki" );
if ( $ns == NS_FILE ) {
- wfProfileIn( __METHOD__."-image" );
+ wfProfileIn( __METHOD__ . "-image" );
if ( !wfIsBadImage( $nt->getDBkey(), $this->mTitle ) ) {
if ( $wasblank ) {
# if no parameters were passed, $text
@@ -1983,12 +2088,12 @@ class Parser {
} else {
$s .= $prefix . $trail;
}
- wfProfileOut( __METHOD__."-image" );
+ wfProfileOut( __METHOD__ . "-image" );
continue;
}
if ( $ns == NS_CATEGORY ) {
- wfProfileIn( __METHOD__."-category" );
+ wfProfileIn( __METHOD__ . "-category" );
$s = rtrim( $s . "\n" ); # bug 87
if ( $wasblank ) {
@@ -2007,23 +2112,23 @@ class Parser {
*/
$s .= trim( $prefix . $trail, "\n" ) == '' ? '' : $prefix . $trail;
- wfProfileOut( __METHOD__."-category" );
+ wfProfileOut( __METHOD__ . "-category" );
continue;
}
}
- # Self-link checking
- if ( $nt->getFragment() === '' && $ns != NS_SPECIAL ) {
- if ( in_array( $nt->getPrefixedText(), $selflink, true ) ) {
- $s .= $prefix . Linker::makeSelfLinkObj( $nt, $text, '', $trail );
- continue;
- }
+ # Self-link checking. For some languages, variants of the title are checked in
+ # LinkHolderArray::doVariants() to allow batching the existence checks necessary
+ # for linking to a different variant.
+ if ( $ns != NS_SPECIAL && $nt->equals( $this->mTitle ) && $nt->getFragment() === '' ) {
+ $s .= $prefix . Linker::makeSelfLinkObj( $nt, $text, '', $trail );
+ continue;
}
# NS_MEDIA is a pseudo-namespace for linking directly to a file
# @todo FIXME: Should do batch file existence checks, see comment below
if ( $ns == NS_MEDIA ) {
- wfProfileIn( __METHOD__."-media" );
+ wfProfileIn( __METHOD__ . "-media" );
# Give extensions a chance to select the file revision for us
$options = array();
$descQuery = false;
@@ -2034,11 +2139,11 @@ class Parser {
# Cloak with NOPARSE to avoid replacement in replaceExternalLinks
$s .= $prefix . $this->armorLinks(
Linker::makeMediaLinkFile( $nt, $file, $text ) ) . $trail;
- wfProfileOut( __METHOD__."-media" );
+ wfProfileOut( __METHOD__ . "-media" );
continue;
}
- wfProfileIn( __METHOD__."-always_known" );
+ 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
#
@@ -2051,7 +2156,7 @@ class Parser {
# Links will be added to the output link list after checking
$s .= $holders->makeHolder( $nt, $text, array(), $trail, $prefix );
}
- wfProfileOut( __METHOD__."-always_known" );
+ wfProfileOut( __METHOD__ . "-always_known" );
}
wfProfileOut( __METHOD__ );
return $holders;
@@ -2066,7 +2171,7 @@ class Parser {
*
* @param $nt Title
* @param $text String
- * @param $query Array or String
+ * @param array $query or String
* @param $trail String
* @param $prefix String
* @return String: HTML-wikitext mix oh yuck
@@ -2093,7 +2198,7 @@ class Parser {
* Not needed quite as much as it used to be since free links are a bit
* more sensible these days. But bracketed links are still an issue.
*
- * @param $text String: more-or-less HTML
+ * @param string $text more-or-less HTML
* @return String: less-or-more HTML with NOPARSE bits
*/
function armorLinks( $text ) {
@@ -2113,7 +2218,7 @@ class Parser {
/**
* Handle link to subpage if necessary
*
- * @param $target String: the source of the link
+ * @param string $target the source of the link
* @param &$text String: the link text, modified as necessary
* @return string the full name of the link
* @private
@@ -2131,7 +2236,7 @@ class Parser {
function closeParagraph() {
$result = '';
if ( $this->mLastSection != '' ) {
- $result = '</' . $this->mLastSection . ">\n";
+ $result = '</' . $this->mLastSection . ">\n";
}
$this->mInPre = false;
$this->mLastSection = '';
@@ -2176,13 +2281,13 @@ class Parser {
$result = $this->closeParagraph();
if ( '*' === $char ) {
- $result .= '<ul><li>';
+ $result .= "<ul>\n<li>";
} elseif ( '#' === $char ) {
- $result .= '<ol><li>';
+ $result .= "<ol>\n<li>";
} elseif ( ':' === $char ) {
- $result .= '<dl><dd>';
+ $result .= "<dl>\n<dd>";
} elseif ( ';' === $char ) {
- $result .= '<dl><dt>';
+ $result .= "<dl>\n<dt>";
$this->mDTopen = true;
} else {
$result = '<!-- ERR 1 -->';
@@ -2200,11 +2305,11 @@ class Parser {
*/
function nextItem( $char ) {
if ( '*' === $char || '#' === $char ) {
- return '</li><li>';
+ return "</li>\n<li>";
} elseif ( ':' === $char || ';' === $char ) {
- $close = '</dd>';
+ $close = "</dd>\n";
if ( $this->mDTopen ) {
- $close = '</dt>';
+ $close = "</dt>\n";
}
if ( ';' === $char ) {
$this->mDTopen = true;
@@ -2226,20 +2331,20 @@ class Parser {
*/
function closeList( $char ) {
if ( '*' === $char ) {
- $text = '</li></ul>';
+ $text = "</li>\n</ul>";
} elseif ( '#' === $char ) {
- $text = '</li></ol>';
+ $text = "</li>\n</ol>";
} elseif ( ':' === $char ) {
if ( $this->mDTopen ) {
$this->mDTopen = false;
- $text = '</dt></dl>';
+ $text = "</dt>\n</dl>";
} else {
- $text = '</dd></dl>';
+ $text = "</dd>\n</dl>";
}
} else {
return '<!-- ERR 3 -->';
}
- return $text."\n";
+ return $text . "\n";
}
/**#@-*/
@@ -2264,6 +2369,7 @@ class Parser {
$this->mDTopen = $inBlockElem = false;
$prefixLength = 0;
$paragraphStack = false;
+ $inBlockquote = false;
foreach ( $textLines as $oLine ) {
# Fix up $linestart
@@ -2306,7 +2412,7 @@ class Parser {
$output .= $this->nextItem( substr( $prefix, -1 ) );
$paragraphStack = false;
- if ( substr( $prefix, -1 ) === ';') {
+ if ( substr( $prefix, -1 ) === ';' ) {
# The one nasty exception: definition lists work like this:
# ; title : definition text
# So we check for : in the remainder text to split up the
@@ -2326,13 +2432,13 @@ class Parser {
# Close all the prefixes which aren't shared.
while ( $commonPrefixLength < $lastPrefixLength ) {
- $output .= $this->closeList( $lastPrefix[$lastPrefixLength-1] );
+ $output .= $this->closeList( $lastPrefix[$lastPrefixLength - 1] );
--$lastPrefixLength;
}
# Continue the current prefix if appropriate.
if ( $prefixLength <= $commonPrefixLength && $commonPrefixLength > 0 ) {
- $output .= $this->nextItem( $prefix[$commonPrefixLength-1] );
+ $output .= $this->nextItem( $prefix[$commonPrefixLength - 1] );
}
# Open prefixes where appropriate.
@@ -2354,13 +2460,13 @@ class Parser {
# If we have no prefixes, go to paragraph mode.
if ( 0 == $prefixLength ) {
- wfProfileIn( __METHOD__."-paragraph" );
+ wfProfileIn( __METHOD__ . "-paragraph" );
# No prefix (not in list)--go to paragraph mode
# 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 );
+ $openmatch = preg_match( '/(?:<table|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<dl|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
$closematch = preg_match(
- '/(?:<\\/table|<\\/blockquote|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|'.
- '<td|<th|<\\/?div|<hr|<\\/pre|<\\/p|'.$this->mUniqPrefix.'-pre|<\\/li|<\\/ul|<\\/ol|<\\/?center)/iS', $t );
+ '/(?:<\\/table|<\\/h1|<\\/h2|<\\/h3|<\\/h4|<\\/h5|<\\/h6|' .
+ '<td|<th|<\\/?blockquote|<\\/?div|<hr|<\\/pre|<\\/p|<\\/mw:|' . $this->mUniqPrefix . '-pre|<\\/li|<\\/ul|<\\/ol|<\\/dl|<\\/?center)/iS', $t );
if ( $openmatch or $closematch ) {
$paragraphStack = false;
# TODO bug 5718: paragraph closed
@@ -2368,13 +2474,18 @@ class Parser {
if ( $preOpenMatch and !$preCloseMatch ) {
$this->mInPre = true;
}
+ $bqOffset = 0;
+ while ( preg_match( '/<(\\/?)blockquote[\s>]/i', $t, $bqMatch, PREG_OFFSET_CAPTURE, $bqOffset ) ) {
+ $inBlockquote = !$bqMatch[1][0]; // is this a close tag?
+ $bqOffset = $bqMatch[0][1] + strlen( $bqMatch[0][0] );
+ }
$inBlockElem = !$closematch;
} elseif ( !$inBlockElem && !$this->mInPre ) {
- if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' || trim( $t ) != '' ) ) {
+ if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' || trim( $t ) != '' ) and !$inBlockquote ) {
# pre
if ( $this->mLastSection !== 'pre' ) {
$paragraphStack = false;
- $output .= $this->closeParagraph().'<pre>';
+ $output .= $this->closeParagraph() . '<pre>';
$this->mLastSection = 'pre';
}
$t = substr( $t, 1 );
@@ -2382,7 +2493,7 @@ class Parser {
# paragraph
if ( trim( $t ) === '' ) {
if ( $paragraphStack ) {
- $output .= $paragraphStack.'<br />';
+ $output .= $paragraphStack . '<br />';
$paragraphStack = false;
$this->mLastSection = 'p';
} else {
@@ -2400,24 +2511,24 @@ class Parser {
$paragraphStack = false;
$this->mLastSection = 'p';
} elseif ( $this->mLastSection !== 'p' ) {
- $output .= $this->closeParagraph().'<p>';
+ $output .= $this->closeParagraph() . '<p>';
$this->mLastSection = 'p';
}
}
}
}
- wfProfileOut( __METHOD__."-paragraph" );
+ wfProfileOut( __METHOD__ . "-paragraph" );
}
# somewhere above we forget to get out of pre block (bug 785)
if ( $preCloseMatch && $this->mInPre ) {
$this->mInPre = false;
}
if ( $paragraphStack === false ) {
- $output .= $t."\n";
+ $output .= $t . "\n";
}
}
while ( $prefixLength ) {
- $output .= $this->closeList( $prefix2[$prefixLength-1] );
+ $output .= $this->closeList( $prefix2[$prefixLength - 1] );
--$prefixLength;
}
if ( $this->mLastSection != '' ) {
@@ -2433,9 +2544,10 @@ class Parser {
* Split up a string on ':', ignoring any occurrences inside tags
* to prevent illegal overlapping.
*
- * @param $str String the string to split
+ * @param string $str the string to split
* @param &$before String set to everything before the ':'
* @param &$after String set to everything after the ':'
+ * @throws MWException
* @return String the position of the ':', or false if none found
*/
function findColonNoLinks( $str, &$before, &$after ) {
@@ -2452,7 +2564,7 @@ class Parser {
if ( $lt === false || $lt > $pos ) {
# Easy; no tag nesting to worry about
$before = substr( $str, 0, $pos );
- $after = substr( $str, $pos+1 );
+ $after = substr( $str, $pos + 1 );
wfProfileOut( __METHOD__ );
return $pos;
}
@@ -2461,13 +2573,13 @@ class Parser {
$state = self::COLON_STATE_TEXT;
$stack = 0;
$len = strlen( $str );
- for( $i = 0; $i < $len; $i++ ) {
+ for ( $i = 0; $i < $len; $i++ ) {
$c = $str[$i];
- switch( $state ) {
+ switch ( $state ) {
# (Using the number is a performance hack for common cases)
case 0: # self::COLON_STATE_TEXT:
- switch( $c ) {
+ switch ( $c ) {
case "<":
# Could be either a <start> tag or an </end> tag
$state = self::COLON_STATE_TAGSTART;
@@ -2512,7 +2624,7 @@ class Parser {
break;
case 1: # self::COLON_STATE_TAG:
# In a <tag>
- switch( $c ) {
+ switch ( $c ) {
case ">":
$stack++;
$state = self::COLON_STATE_TEXT;
@@ -2526,7 +2638,7 @@ class Parser {
}
break;
case 2: # self::COLON_STATE_TAGSTART:
- switch( $c ) {
+ switch ( $c ) {
case "/":
$state = self::COLON_STATE_CLOSETAG;
break;
@@ -2546,7 +2658,7 @@ class Parser {
if ( $c === ">" ) {
$stack--;
if ( $stack < 0 ) {
- wfDebug( __METHOD__.": Invalid input; too many close tags\n" );
+ wfDebug( __METHOD__ . ": Invalid input; too many close tags\n" );
wfProfileOut( __METHOD__ );
return false;
}
@@ -2582,11 +2694,12 @@ class Parser {
}
break;
default:
+ wfProfileOut( __METHOD__ );
throw new MWException( "State machine error in " . __METHOD__ );
}
}
if ( $stack > 0 ) {
- wfDebug( __METHOD__.": Invalid input; not enough close tags (stack $stack, state $state)\n" );
+ wfDebug( __METHOD__ . ": Invalid input; not enough close tags (stack $stack, state $state)\n" );
wfProfileOut( __METHOD__ );
return false;
}
@@ -2600,8 +2713,9 @@ class Parser {
* @private
*
* @param $index integer
- * @param $frame PPFrame
+ * @param bool|\PPFrame $frame
*
+ * @throws MWException
* @return string
*/
function getVariableValue( $index, $frame = false ) {
@@ -2630,71 +2744,50 @@ class Parser {
$ts = wfTimestamp( TS_UNIX, $this->mOptions->getTimestamp() );
wfRunHooks( 'ParserGetVariableValueTs', array( &$this, &$ts ) );
- # Use the time zone
- global $wgLocaltimezone;
- if ( isset( $wgLocaltimezone ) ) {
- $oldtz = date_default_timezone_get();
- date_default_timezone_set( $wgLocaltimezone );
- }
-
- $localTimestamp = date( 'YmdHis', $ts );
- $localMonth = date( 'm', $ts );
- $localMonth1 = date( 'n', $ts );
- $localMonthName = date( 'n', $ts );
- $localDay = date( 'j', $ts );
- $localDay2 = date( 'd', $ts );
- $localDayOfWeek = date( 'w', $ts );
- $localWeek = date( 'W', $ts );
- $localYear = date( 'Y', $ts );
- $localHour = date( 'H', $ts );
- if ( isset( $wgLocaltimezone ) ) {
- date_default_timezone_set( $oldtz );
- }
-
$pageLang = $this->getFunctionLang();
switch ( $index ) {
case 'currentmonth':
- $value = $pageLang->formatNum( gmdate( 'm', $ts ) );
+ $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'm' ) );
break;
case 'currentmonth1':
- $value = $pageLang->formatNum( gmdate( 'n', $ts ) );
+ $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'n' ) );
break;
case 'currentmonthname':
- $value = $pageLang->getMonthName( gmdate( 'n', $ts ) );
+ $value = $pageLang->getMonthName( MWTimestamp::getInstance( $ts )->format( 'n' ) );
break;
case 'currentmonthnamegen':
- $value = $pageLang->getMonthNameGen( gmdate( 'n', $ts ) );
+ $value = $pageLang->getMonthNameGen( MWTimestamp::getInstance( $ts )->format( 'n' ) );
break;
case 'currentmonthabbrev':
- $value = $pageLang->getMonthAbbreviation( gmdate( 'n', $ts ) );
+ $value = $pageLang->getMonthAbbreviation( MWTimestamp::getInstance( $ts )->format( 'n' ) );
break;
case 'currentday':
- $value = $pageLang->formatNum( gmdate( 'j', $ts ) );
+ $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'j' ) );
break;
case 'currentday2':
- $value = $pageLang->formatNum( gmdate( 'd', $ts ) );
+ $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'd' ) );
break;
case 'localmonth':
- $value = $pageLang->formatNum( $localMonth );
+ $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'm' ) );
break;
case 'localmonth1':
- $value = $pageLang->formatNum( $localMonth1 );
+ $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
break;
case 'localmonthname':
- $value = $pageLang->getMonthName( $localMonthName );
+ $value = $pageLang->getMonthName( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
break;
case 'localmonthnamegen':
- $value = $pageLang->getMonthNameGen( $localMonthName );
+ $value = $pageLang->getMonthNameGen( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
break;
case 'localmonthabbrev':
- $value = $pageLang->getMonthAbbreviation( $localMonthName );
+ $value = $pageLang->getMonthAbbreviation( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
break;
case 'localday':
- $value = $pageLang->formatNum( $localDay );
+ $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'j' ) );
break;
case 'localday2':
- $value = $pageLang->formatNum( $localDay2 );
+ $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'd' ) );
break;
case 'pagename':
$value = wfEscapeWikiText( $this->mTitle->getText() );
@@ -2714,6 +2807,12 @@ class Parser {
case 'subpagenamee':
$value = wfEscapeWikiText( $this->mTitle->getSubpageUrlForm() );
break;
+ case 'rootpagename':
+ $value = wfEscapeWikiText( $this->mTitle->getRootText() );
+ break;
+ case 'rootpagenamee':
+ $value = wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getRootText() ) ) );
+ break;
case 'basepagename':
$value = wfEscapeWikiText( $this->mTitle->getBaseText() );
break;
@@ -2731,7 +2830,7 @@ class Parser {
case 'talkpagenamee':
if ( $this->mTitle->canTalk() ) {
$talkPage = $this->mTitle->getTalkPage();
- $value = wfEscapeWikiText( $talkPage->getPrefixedUrl() );
+ $value = wfEscapeWikiText( $talkPage->getPrefixedURL() );
} else {
$value = '';
}
@@ -2742,11 +2841,11 @@ class Parser {
break;
case 'subjectpagenamee':
$subjPage = $this->mTitle->getSubjectPage();
- $value = wfEscapeWikiText( $subjPage->getPrefixedUrl() );
+ $value = wfEscapeWikiText( $subjPage->getPrefixedURL() );
break;
case 'pageid': // requested in bug 23427
- $pageid = $this->getTitle()->getArticleId();
- if( $pageid == 0 ) {
+ $pageid = $this->getTitle()->getArticleID();
+ if ( $pageid == 0 ) {
# 0 means the page doesn't exist in the database,
# which means the user is previewing a new page.
# The vary-revision flag must be set, because the magic word
@@ -2812,8 +2911,15 @@ class Parser {
wfDebug( __METHOD__ . ": {{REVISIONUSER}} used, setting vary-revision...\n" );
$value = $this->getRevisionUser();
break;
+ case 'revisionsize':
+ # 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__ . ": {{REVISIONSIZE}} used, setting vary-revision...\n" );
+ $value = $this->getRevisionSize();
+ break;
case 'namespace':
- $value = str_replace( '_',' ',$wgContLang->getNsText( $this->mTitle->getNamespace() ) );
+ $value = str_replace( '_', ' ', $wgContLang->getNsText( $this->mTitle->getNamespace() ) );
break;
case 'namespacee':
$value = wfUrlencode( $wgContLang->getNsText( $this->mTitle->getNamespace() ) );
@@ -2822,56 +2928,56 @@ class Parser {
$value = $this->mTitle->getNamespace();
break;
case 'talkspace':
- $value = $this->mTitle->canTalk() ? str_replace( '_',' ',$this->mTitle->getTalkNsText() ) : '';
+ $value = $this->mTitle->canTalk() ? str_replace( '_', ' ', $this->mTitle->getTalkNsText() ) : '';
break;
case 'talkspacee':
$value = $this->mTitle->canTalk() ? wfUrlencode( $this->mTitle->getTalkNsText() ) : '';
break;
case 'subjectspace':
- $value = $this->mTitle->getSubjectNsText();
+ $value = str_replace( '_', ' ', $this->mTitle->getSubjectNsText() );
break;
case 'subjectspacee':
$value = ( wfUrlencode( $this->mTitle->getSubjectNsText() ) );
break;
case 'currentdayname':
- $value = $pageLang->getWeekdayName( gmdate( 'w', $ts ) + 1 );
+ $value = $pageLang->getWeekdayName( MWTimestamp::getInstance( $ts )->format( 'w' ) + 1 );
break;
case 'currentyear':
- $value = $pageLang->formatNum( gmdate( 'Y', $ts ), true );
+ $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'Y' ), true );
break;
case 'currenttime':
$value = $pageLang->time( wfTimestamp( TS_MW, $ts ), false, false );
break;
case 'currenthour':
- $value = $pageLang->formatNum( gmdate( 'H', $ts ), true );
+ $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'H' ), true );
break;
case 'currentweek':
# @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
# int to remove the padding
- $value = $pageLang->formatNum( (int)gmdate( 'W', $ts ) );
+ $value = $pageLang->formatNum( (int)MWTimestamp::getInstance( $ts )->format( 'W' ) );
break;
case 'currentdow':
- $value = $pageLang->formatNum( gmdate( 'w', $ts ) );
+ $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'w' ) );
break;
case 'localdayname':
- $value = $pageLang->getWeekdayName( $localDayOfWeek + 1 );
+ $value = $pageLang->getWeekdayName( MWTimestamp::getLocalInstance( $ts )->format( 'w' ) + 1 );
break;
case 'localyear':
- $value = $pageLang->formatNum( $localYear, true );
+ $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'Y' ), true );
break;
case 'localtime':
- $value = $pageLang->time( $localTimestamp, false, false );
+ $value = $pageLang->time( MWTimestamp::getLocalInstance( $ts )->format( 'YmdHis' ), false, false );
break;
case 'localhour':
- $value = $pageLang->formatNum( $localHour, true );
+ $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'H' ), true );
break;
case 'localweek':
# @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
# int to remove the padding
- $value = $pageLang->formatNum( (int)$localWeek );
+ $value = $pageLang->formatNum( (int)MWTimestamp::getLocalInstance( $ts )->format( 'W' ) );
break;
case 'localdow':
- $value = $pageLang->formatNum( $localDayOfWeek );
+ $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'w' ) );
break;
case 'numberofarticles':
$value = $pageLang->formatNum( SiteStats::articles() );
@@ -2902,7 +3008,7 @@ class Parser {
$value = wfTimestamp( TS_MW, $ts );
break;
case 'localtimestamp':
- $value = $localTimestamp;
+ $value = MWTimestamp::getLocalInstance( $ts )->format( 'YmdHis' );
break;
case 'currentversion':
$value = SpecialVersion::getVersion();
@@ -2960,7 +3066,7 @@ class Parser {
* Preprocess some wikitext and return the document tree.
* This is the ghost of replace_variables().
*
- * @param $text String: The text to parse
+ * @param string $text The text to parse
* @param $flags Integer: bitwise combination of:
* self::PTD_FOR_INCLUSION Handle "<noinclude>" and "<includeonly>" as if the text is being
* included. Default is to assume a direct page view.
@@ -3015,7 +3121,7 @@ class Parser {
* self::OT_PREPROCESS: templates but not extension tags
* self::OT_HTML: all templates and extension tags
*
- * @param $text String the text to transform
+ * @param string $text 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.
@@ -3034,7 +3140,7 @@ class Parser {
if ( $frame === false ) {
$frame = $this->getPreprocessor()->newFrame();
} elseif ( !( $frame instanceof PPFrame ) ) {
- wfDebug( __METHOD__." called using plain parameters instead of a PPFrame instance. Creating custom frame.\n" );
+ wfDebug( __METHOD__ . " called using plain parameters instead of a PPFrame instance. Creating custom frame.\n" );
$frame = $this->getPreprocessor()->newCustomFrame( $frame );
}
@@ -3062,7 +3168,7 @@ class Parser {
$assocArgs[$index++] = $arg;
} else {
$name = trim( substr( $arg, 0, $eqpos ) );
- $value = trim( substr( $arg, $eqpos+1 ) );
+ $value = trim( substr( $arg, $eqpos + 1 ) );
if ( $value === false ) {
$value = '';
}
@@ -3079,7 +3185,7 @@ class Parser {
* Warn the user when a parser limitation is reached
* Will warn at most once the user per limitation type
*
- * @param $limitationType String: should be one of:
+ * @param string $limitationType should be one of:
* 'expensive-parserfunction' (corresponding messages:
* 'expensive-parserfunction-warning',
* 'expensive-parserfunction-category')
@@ -3089,14 +3195,20 @@ class Parser {
* 'post-expand-template-inclusion' (corresponding messages:
* 'post-expand-template-inclusion-warning',
* 'post-expand-template-inclusion-category')
- * @param $current int|null Current value
- * @param $max int|null Maximum allowed, when an explicit limit has been
+ * 'node-count-exceeded' (corresponding messages:
+ * 'node-count-exceeded-warning',
+ * 'node-count-exceeded-category')
+ * 'expansion-depth-exceeded' (corresponding messages:
+ * 'expansion-depth-exceeded-warning',
+ * 'expansion-depth-exceeded-category')
+ * @param int|null $current Current value
+ * @param int|null $max Maximum allowed, when an explicit limit has been
* exceeded, provide the values (optional)
*/
function limitationWarn( $limitationType, $current = '', $max = '' ) {
# does no harm if $current and $max are present but are unnecessary for the message
$warning = wfMessage( "$limitationType-warning" )->numParams( $current, $max )
- ->inContentLanguage()->escaped();
+ ->inLanguage( $this->mOptions->getUserLangObj() )->text();
$this->mOutput->addWarning( $warning );
$this->addTrackingCategory( "$limitationType-category" );
}
@@ -3105,18 +3217,18 @@ class Parser {
* Return the text of a template, after recursively
* replacing any variables or templates within the template.
*
- * @param $piece Array: the parts of the template
+ * @param array $piece 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 $frame PPFrame The current frame, contains template arguments
+ * @throws MWException
* @return String: the text of the template
* @private
*/
function braceSubstitution( $piece, $frame ) {
- global $wgContLang;
wfProfileIn( __METHOD__ );
- wfProfileIn( __METHOD__.'-setup' );
+ wfProfileIn( __METHOD__ . '-setup' );
# Flags
$found = false; # $text has been filled
@@ -3141,12 +3253,12 @@ class Parser {
# $args is a list of argument nodes, starting from index 0, not including $part1
# @todo FIXME: If piece['parts'] is null then the call to getLength() below won't work b/c this $args isn't an object
$args = ( null == $piece['parts'] ) ? array() : $piece['parts'];
- wfProfileOut( __METHOD__.'-setup' );
+ wfProfileOut( __METHOD__ . '-setup' );
$titleProfileIn = null; // profile templates
# SUBST
- wfProfileIn( __METHOD__.'-modifiers' );
+ wfProfileIn( __METHOD__ . '-modifiers' );
if ( !$found ) {
$substMatch = $this->mSubstWords->matchStartAndRemove( $part1 );
@@ -3203,7 +3315,7 @@ class Parser {
$forceRawInterwiki = true;
}
}
- wfProfileOut( __METHOD__.'-modifiers' );
+ wfProfileOut( __METHOD__ . '-modifiers' );
# Parser functions
if ( !$found ) {
@@ -3211,70 +3323,23 @@ class Parser {
$colonPos = strpos( $part1, ':' );
if ( $colonPos !== false ) {
- # Case sensitive functions
- $function = substr( $part1, 0, $colonPos );
- if ( isset( $this->mFunctionSynonyms[1][$function] ) ) {
- $function = $this->mFunctionSynonyms[1][$function];
- } else {
- # Case insensitive functions
- $function = $wgContLang->lc( $function );
- if ( isset( $this->mFunctionSynonyms[0][$function] ) ) {
- $function = $this->mFunctionSynonyms[0][$function];
- } else {
- $function = false;
- }
+ $func = substr( $part1, 0, $colonPos );
+ $funcArgs = array( trim( substr( $part1, $colonPos + 1 ) ) );
+ for ( $i = 0; $i < $args->getLength(); $i++ ) {
+ $funcArgs[] = $args->item( $i );
}
- if ( $function ) {
- wfProfileIn( __METHOD__ . '-pfunc-' . $function );
- list( $callback, $flags ) = $this->mFunctionHooks[$function];
- $initialArgs = array( &$this );
- $funcArgs = array( trim( substr( $part1, $colonPos + 1 ) ) );
- if ( $flags & SFH_OBJECT_ARGS ) {
- # Add a frame parameter, and pass the arguments as an array
- $allArgs = $initialArgs;
- $allArgs[] = $frame;
- for ( $i = 0; $i < $args->getLength(); $i++ ) {
- $funcArgs[] = $args->item( $i );
- }
- $allArgs[] = $funcArgs;
- } else {
- # Convert arguments to plain text
- for ( $i = 0; $i < $args->getLength(); $i++ ) {
- $funcArgs[] = trim( $frame->expand( $args->item( $i ) ) );
- }
- $allArgs = array_merge( $initialArgs, $funcArgs );
- }
-
- # Workaround for PHP bug 35229 and similar
- if ( !is_callable( $callback ) ) {
- wfProfileOut( __METHOD__ . '-pfunc-' . $function );
- wfProfileOut( __METHOD__ . '-pfunc' );
- wfProfileOut( __METHOD__ );
- throw new MWException( "Tag hook for $function is not callable\n" );
- }
- $result = call_user_func_array( $callback, $allArgs );
- $found = true;
- $noparse = true;
- $preprocessFlags = 0;
-
- if ( is_array( $result ) ) {
- if ( isset( $result[0] ) ) {
- $text = $result[0];
- unset( $result[0] );
- }
-
- # Extract flags into the local scope
- # This allows callers to set flags such as nowiki, found, etc.
- extract( $result );
- } else {
- $text = $result;
- }
- if ( !$noparse ) {
- $text = $this->preprocessToDom( $text, $preprocessFlags );
- $isChildObj = true;
- }
- wfProfileOut( __METHOD__ . '-pfunc-' . $function );
+ try {
+ $result = $this->callParserFunction( $frame, $func, $funcArgs );
+ } catch ( Exception $ex ) {
+ wfProfileOut( __METHOD__ . '-pfunc' );
+ wfProfileOut( __METHOD__ );
+ throw $ex;
}
+
+ # The interface for parser functions allows for extracting
+ # flags into the local scope. Extract any forwarded flags
+ # here.
+ extract( $result );
}
wfProfileOut( __METHOD__ . '-pfunc' );
}
@@ -3285,8 +3350,9 @@ class Parser {
$ns = NS_TEMPLATE;
# Split the title into page and subpage
$subpage = '';
- $part1 = $this->maybeDoSubpageLink( $part1, $subpage );
- if ( $subpage !== '' ) {
+ $relative = $this->maybeDoSubpageLink( $part1, $subpage );
+ if ( $part1 !== $relative ) {
+ $part1 = $relative;
$ns = $this->mTitle->getNamespace();
}
$title = Title::newFromText( $part1, $ns );
@@ -3312,7 +3378,7 @@ class Parser {
if ( !$found && $title ) {
if ( !Profiler::instance()->isPersistent() ) {
# Too many unique items can kill profiling DBs/collectors
- $titleProfileIn = __METHOD__ . "-title-" . $title->getDBKey();
+ $titleProfileIn = __METHOD__ . "-title-" . $title->getPrefixedDBkey();
wfProfileIn( $titleProfileIn ); // template in
}
wfProfileIn( __METHOD__ . '-loadtpl' );
@@ -3350,7 +3416,7 @@ class Parser {
}
} elseif ( MWNamespace::isNonincludable( $title->getNamespace() ) ) {
$found = false; # access denied
- wfDebug( __METHOD__.": template inclusion denied for " . $title->getPrefixedDBkey() );
+ wfDebug( __METHOD__ . ": template inclusion denied for " . $title->getPrefixedDBkey() );
} else {
list( $text, $title ) = $this->getTemplateDom( $title );
if ( $text !== false ) {
@@ -3385,7 +3451,7 @@ class Parser {
$text = '<span class="error">'
. wfMessage( 'parser-template-loop-warning', $titleText )->inContentLanguage()->text()
. '</span>';
- wfDebug( __METHOD__.": template loop broken at '$titleText'\n" );
+ wfDebug( __METHOD__ . ": template loop broken at '$titleText'\n" );
}
wfProfileOut( __METHOD__ . '-loadtpl' );
}
@@ -3442,7 +3508,7 @@ class Parser {
{
# Bug 529: if the template begins with a table or block-level
# element, it should be treated as beginning a new line.
- # This behaviour is somewhat controversial.
+ # This behavior is somewhat controversial.
$text = "\n" . $text;
}
@@ -3472,6 +3538,120 @@ class Parser {
}
/**
+ * Call a parser function and return an array with text and flags.
+ *
+ * The returned array will always contain a boolean 'found', indicating
+ * whether the parser function was found or not. It may also contain the
+ * following:
+ * text: string|object, resulting wikitext or PP DOM object
+ * isHTML: bool, $text is HTML, armour it against wikitext transformation
+ * isChildObj: bool, $text is a DOM node needing expansion in a child frame
+ * isLocalObj: bool, $text is a DOM node needing expansion in the current frame
+ * nowiki: bool, wiki markup in $text should be escaped
+ *
+ * @since 1.21
+ * @param $frame PPFrame The current frame, contains template arguments
+ * @param $function string Function name
+ * @param $args array Arguments to the function
+ * @return array
+ */
+ public function callParserFunction( $frame, $function, array $args = array() ) {
+ global $wgContLang;
+
+ wfProfileIn( __METHOD__ );
+
+ # Case sensitive functions
+ if ( isset( $this->mFunctionSynonyms[1][$function] ) ) {
+ $function = $this->mFunctionSynonyms[1][$function];
+ } else {
+ # Case insensitive functions
+ $function = $wgContLang->lc( $function );
+ if ( isset( $this->mFunctionSynonyms[0][$function] ) ) {
+ $function = $this->mFunctionSynonyms[0][$function];
+ } else {
+ wfProfileOut( __METHOD__ );
+ return array( 'found' => false );
+ }
+ }
+
+ wfProfileIn( __METHOD__ . '-pfunc-' . $function );
+ list( $callback, $flags ) = $this->mFunctionHooks[$function];
+
+ # Workaround for PHP bug 35229 and similar
+ if ( !is_callable( $callback ) ) {
+ wfProfileOut( __METHOD__ . '-pfunc-' . $function );
+ wfProfileOut( __METHOD__ );
+ throw new MWException( "Tag hook for $function is not callable\n" );
+ }
+
+ $allArgs = array( &$this );
+ if ( $flags & SFH_OBJECT_ARGS ) {
+ # Convert arguments to PPNodes and collect for appending to $allArgs
+ $funcArgs = array();
+ foreach ( $args as $k => $v ) {
+ if ( $v instanceof PPNode || $k === 0 ) {
+ $funcArgs[] = $v;
+ } else {
+ $funcArgs[] = $this->mPreprocessor->newPartNodeArray( array( $k => $v ) )->item( 0 );
+ }
+ }
+
+ # Add a frame parameter, and pass the arguments as an array
+ $allArgs[] = $frame;
+ $allArgs[] = $funcArgs;
+ } else {
+ # Convert arguments to plain text and append to $allArgs
+ foreach ( $args as $k => $v ) {
+ if ( $v instanceof PPNode ) {
+ $allArgs[] = trim( $frame->expand( $v ) );
+ } elseif ( is_int( $k ) && $k >= 0 ) {
+ $allArgs[] = trim( $v );
+ } else {
+ $allArgs[] = trim( "$k=$v" );
+ }
+ }
+ }
+
+ $result = call_user_func_array( $callback, $allArgs );
+
+ # The interface for function hooks allows them to return a wikitext
+ # string or an array containing the string and any flags. This mungs
+ # things around to match what this method should return.
+ if ( !is_array( $result ) ) {
+ $result = array(
+ 'found' => true,
+ 'text' => $result,
+ );
+ } else {
+ if ( isset( $result[0] ) && !isset( $result['text'] ) ) {
+ $result['text'] = $result[0];
+ }
+ unset( $result[0] );
+ $result += array(
+ 'found' => true,
+ );
+ }
+
+ $noparse = true;
+ $preprocessFlags = 0;
+ if ( isset( $result['noparse'] ) ) {
+ $noparse = $result['noparse'];
+ }
+ if ( isset( $result['preprocessFlags'] ) ) {
+ $preprocessFlags = $result['preprocessFlags'];
+ }
+
+ if ( !$noparse ) {
+ $result['text'] = $this->preprocessToDom( $result['text'], $preprocessFlags );
+ $result['isChildObj'] = true;
+ }
+ wfProfileOut( __METHOD__ . '-pfunc-' . $function );
+ wfProfileOut( __METHOD__ );
+
+ return $result;
+ }
+
+ /**
* Get the semi-parsed DOM representation of a template with a given title,
* and its redirect destination title. Cached.
*
@@ -3501,7 +3681,7 @@ class Parser {
}
$dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION );
- $this->mTplDomCache[ $titleText ] = $dom;
+ $this->mTplDomCache[$titleText] = $dom;
if ( !$title->equals( $cacheTitle ) ) {
$this->mTplRedirCache[$cacheTitle->getPrefixedDBkey()] =
@@ -3524,6 +3704,11 @@ class Parser {
if ( isset( $stuff['deps'] ) ) {
foreach ( $stuff['deps'] as $dep ) {
$this->mOutput->addTemplate( $dep['title'], $dep['page_id'], $dep['rev_id'] );
+ if ( $dep['title']->equals( $this->getTitle() ) ) {
+ // If we transclude ourselves, the final result
+ // will change based on the new version of the page
+ $this->mOutput->setFlag( 'vary-revision' );
+ }
}
}
return array( $text, $finalTitle );
@@ -3563,9 +3748,9 @@ class Parser {
if ( $skip ) {
$text = false;
$deps[] = array(
- 'title' => $title,
- 'page_id' => $title->getArticleID(),
- 'rev_id' => null
+ 'title' => $title,
+ 'page_id' => $title->getArticleID(),
+ 'rev_id' => null
);
break;
}
@@ -3581,19 +3766,25 @@ class Parser {
}
$deps[] = array(
- 'title' => $title,
- 'page_id' => $title->getArticleID(),
- 'rev_id' => $rev_id );
+ 'title' => $title,
+ 'page_id' => $title->getArticleID(),
+ 'rev_id' => $rev_id );
if ( $rev && !$title->equals( $rev->getTitle() ) ) {
# We fetched a rev from a different title; register it too...
$deps[] = array(
- 'title' => $rev->getTitle(),
- 'page_id' => $rev->getPage(),
- 'rev_id' => $rev_id );
+ 'title' => $rev->getTitle(),
+ 'page_id' => $rev->getPage(),
+ 'rev_id' => $rev_id );
}
if ( $rev ) {
- $text = $rev->getText();
+ $content = $rev->getContent();
+ $text = $content ? $content->getWikitextForTransclusion() : null;
+
+ if ( $text === false || $text === null ) {
+ $text = false;
+ break;
+ }
} elseif ( $title->getNamespace() == NS_MEDIAWIKI ) {
global $wgContLang;
$message = wfMessage( $wgContLang->lcfirst( $title->getText() ) )->inContentLanguage();
@@ -3601,16 +3792,17 @@ class Parser {
$text = false;
break;
}
+ $content = $message->content();
$text = $message->plain();
} else {
break;
}
- if ( $text === false ) {
+ if ( !$content ) {
break;
}
# Redirect?
$finalTitle = $title;
- $title = Title::newFromRedirect( $text );
+ $title = $content->getRedirectTarget();
}
return array(
'text' => $text,
@@ -3622,7 +3814,7 @@ class Parser {
* Fetch a file and its title and register a reference to it.
* If 'broken' is a key in $options then the file will appear as a broken thumbnail.
* @param Title $title
- * @param Array $options Array of options to RepoGroup::findFile
+ * @param array $options Array of options to RepoGroup::findFile
* @return File|bool
*/
function fetchFile( $title, $options = array() ) {
@@ -3634,17 +3826,12 @@ class Parser {
* Fetch a file and its title and register a reference to it.
* If 'broken' is a key in $options then the file will appear as a broken thumbnail.
* @param Title $title
- * @param Array $options Array of options to RepoGroup::findFile
+ * @param array $options Array of options to RepoGroup::findFile
* @return Array ( File or false, Title of file )
*/
function fetchFileAndTitle( $title, $options = array() ) {
- if ( isset( $options['broken'] ) ) {
- $file = false; // broken thumbnail forced by hook
- } elseif ( isset( $options['sha1'] ) ) { // get by (sha1,timestamp)
- $file = RepoGroup::singleton()->findFileFromKey( $options['sha1'], $options );
- } else { // get by (name,timestamp)
- $file = wfFindFile( $title, $options );
- }
+ $file = $this->fetchFileNoRegister( $title, $options );
+
$time = $file ? $file->getTimestamp() : false;
$sha1 = $file ? $file->getSha1() : false;
# Register the file as a dependency...
@@ -3663,6 +3850,27 @@ class Parser {
}
/**
+ * Helper function for fetchFileAndTitle.
+ *
+ * Also useful if you need to fetch a file but not use it yet,
+ * for example to get the file's handler.
+ *
+ * @param Title $title
+ * @param array $options Array of options to RepoGroup::findFile
+ * @return File or false
+ */
+ protected function fetchFileNoRegister( $title, $options = array() ) {
+ if ( isset( $options['broken'] ) ) {
+ $file = false; // broken thumbnail forced by hook
+ } elseif ( isset( $options['sha1'] ) ) { // get by (sha1,timestamp)
+ $file = RepoGroup::singleton()->findFileFromKey( $options['sha1'], $options );
+ } else { // get by (name,timestamp)
+ $file = wfFindFile( $title, $options );
+ }
+ return $file;
+ }
+
+ /**
* Transclude an interwiki link.
*
* @param $title Title
@@ -3674,10 +3882,10 @@ class Parser {
global $wgEnableScaryTranscluding;
if ( !$wgEnableScaryTranscluding ) {
- return wfMessage('scarytranscludedisabled')->inContentLanguage()->text();
+ return wfMessage( 'scarytranscludedisabled' )->inContentLanguage()->text();
}
- $url = $title->getFullUrl( "action=$action" );
+ $url = $title->getFullURL( array( 'action' => $action ) );
if ( strlen( $url ) > 255 ) {
return wfMessage( 'scarytranscludetoolong' )->inContentLanguage()->text();
@@ -3693,23 +3901,28 @@ class Parser {
global $wgTranscludeCacheExpiry;
$dbr = wfGetDB( DB_SLAVE );
$tsCond = $dbr->timestamp( time() - $wgTranscludeCacheExpiry );
- $obj = $dbr->selectRow( 'transcache', array('tc_time', 'tc_contents' ),
+ $obj = $dbr->selectRow( 'transcache', array( 'tc_time', 'tc_contents' ),
array( 'tc_url' => $url, "tc_time >= " . $dbr->addQuotes( $tsCond ) ) );
if ( $obj ) {
return $obj->tc_contents;
}
- $text = Http::get( $url );
- if ( !$text ) {
+ $req = MWHttpRequest::factory( $url );
+ $status = $req->execute(); // Status object
+ if ( $status->isOK() ) {
+ $text = $req->getContent();
+ } elseif ( $req->getStatus() != 200 ) { // Though we failed to fetch the content, this status is useless.
+ return wfMessage( 'scarytranscludefailed-httpstatus', $url, $req->getStatus() /* HTTP status */ )->inContentLanguage()->text();
+ } else {
return wfMessage( 'scarytranscludefailed', $url )->inContentLanguage()->text();
}
$dbw = wfGetDB( DB_MASTER );
- $dbw->replace( 'transcache', array('tc_url'), array(
+ $dbw->replace( 'transcache', array( 'tc_url' ), array(
'tc_url' => $url,
'tc_time' => $dbw->timestamp( time() ),
- 'tc_contents' => $text)
- );
+ 'tc_contents' => $text
+ ) );
return $text;
}
@@ -3731,7 +3944,7 @@ class Parser {
$argName = trim( $nameWithSpaces );
$object = false;
$text = $frame->getArgument( $argName );
- if ( $text === false && $parts->getLength() > 0
+ if ( $text === false && $parts->getLength() > 0
&& (
$this->ot['html']
|| $this->ot['pre']
@@ -3767,7 +3980,7 @@ class Parser {
* Return the text to be used for a given extension tag.
* This is the ghost of strip().
*
- * @param $params array Associative array of parameters:
+ * @param array $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
@@ -3775,6 +3988,7 @@ class Parser {
* noClose Original text did not have a close tag
* @param $frame PPFrame
*
+ * @throws MWException
* @return string
*/
function extensionSubstitution( $params, $frame ) {
@@ -3783,7 +3997,7 @@ class Parser {
$content = !isset( $params['inner'] ) ? null : $frame->expand( $params['inner'] );
$marker = "{$this->mUniqPrefix}-$name-" . sprintf( '%08X', $this->mMarkerIndex++ ) . self::MARKER_SUFFIX;
- $isFunctionTag = isset( $this->mFunctionTagHooks[strtolower($name)] ) &&
+ $isFunctionTag = isset( $this->mFunctionTagHooks[strtolower( $name )] ) &&
( $this->ot['html'] || $this->ot['pre'] );
if ( $isFunctionTag ) {
$markerType = 'none';
@@ -3805,7 +4019,7 @@ class Parser {
$output = call_user_func_array( $this->mTagHooks[$name],
array( $content, $attributes, $this, $frame ) );
} elseif ( isset( $this->mFunctionTagHooks[$name] ) ) {
- list( $callback, $flags ) = $this->mFunctionTagHooks[$name];
+ list( $callback, ) = $this->mFunctionTagHooks[$name];
if ( !is_callable( $callback ) ) {
throw new MWException( "Tag hook for $name is not callable\n" );
}
@@ -3848,7 +4062,7 @@ class Parser {
} elseif ( $markerType === 'general' ) {
$this->mStripState->addGeneral( $marker, $output );
} else {
- throw new MWException( __METHOD__.': invalid marker type' );
+ throw new MWException( __METHOD__ . ': invalid marker type' );
}
return $marker;
}
@@ -3856,7 +4070,7 @@ class Parser {
/**
* Increment an include size counter
*
- * @param $type String: the type of expansion
+ * @param string $type 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
*/
@@ -3942,12 +4156,12 @@ class Parser {
* Add a tracking category, getting the title from a system message,
* or print a debug message if the title is invalid.
*
- * @param $msg String: message key
+ * @param string $msg message key
* @return Boolean: whether the addition was successful
*/
public function addTrackingCategory( $msg ) {
if ( $this->mTitle->getNamespace() === NS_SPECIAL ) {
- wfDebug( __METHOD__.": Not adding tracking category $msg to special page!\n" );
+ wfDebug( __METHOD__ . ": Not adding tracking category $msg to special page!\n" );
return false;
}
// Important to parse with correct title (bug 31469)
@@ -3966,7 +4180,7 @@ class Parser {
$this->mOutput->addCategory( $containerCategory->getDBkey(), $this->getDefaultSort() );
return true;
} else {
- wfDebug( __METHOD__.": [[MediaWiki:$msg]] is not a valid title!\n" );
+ wfDebug( __METHOD__ . ": [[MediaWiki:$msg]] is not a valid title!\n" );
return false;
}
}
@@ -3982,13 +4196,13 @@ class Parser {
* string and re-inserts the newly formatted headlines.
*
* @param $text String
- * @param $origText String: original, untouched wikitext
+ * @param string $origText original, untouched wikitext
* @param $isMain Boolean
* @return mixed|string
* @private
*/
- function formatHeadings( $text, $origText, $isMain=true ) {
- global $wgMaxTocLevel, $wgHtml5, $wgExperimentalHtmlIds;
+ function formatHeadings( $text, $origText, $isMain = true ) {
+ global $wgMaxTocLevel, $wgExperimentalHtmlIds;
# Inhibit editsection links if requested in the page
if ( isset( $this->mDoubleUnderscores['noeditsection'] ) ) {
@@ -4004,7 +4218,7 @@ class Parser {
# Get all headlines for numbering them and adding funky stuff like [edit]
# links - this is for later, but we need the number of headlines right now
$matches = array();
- $numMatches = preg_match_all( '/<H(?P<level>[1-6])(?P<attrib>.*?'.'>)(?P<header>.*?)<\/H[1-6] *>/i', $text, $matches );
+ $numMatches = preg_match_all( '/<H(?P<level>[1-6])(?P<attrib>.*?' . '>)\s*(?P<header>[\s\S]*?)\s*<\/H[1-6] *>/i', $text, $matches );
# if there are fewer than 4 headlines in the article, do not show TOC
# unless it's been explicitly enabled.
@@ -4062,11 +4276,11 @@ class Parser {
$sectionIndex = false;
$numbering = '';
$markerMatches = array();
- if ( preg_match("/^$markerRegex/", $headline, $markerMatches ) ) {
+ if ( preg_match( "/^$markerRegex/", $headline, $markerMatches ) ) {
$serial = $markerMatches[1];
list( $titleText, $sectionIndex ) = $this->mHeadings[$serial];
$isTemplate = ( $titleText != $baseTitleText );
- $headline = preg_replace( "/^$markerRegex/", "", $headline );
+ $headline = preg_replace( "/^$markerRegex\\s*/", "", $headline );
}
if ( $toclevel ) {
@@ -4078,7 +4292,7 @@ class Parser {
# Increase TOC level
$toclevel++;
$sublevelCount[$toclevel] = 0;
- if ( $toclevel<$wgMaxTocLevel ) {
+ if ( $toclevel < $wgMaxTocLevel ) {
$prevtoclevel = $toclevel;
$toc .= Linker::tocIndent();
$numVisible++;
@@ -4100,7 +4314,7 @@ class Parser {
if ( $i == 0 ) {
$toclevel = 1;
}
- if ( $toclevel<$wgMaxTocLevel ) {
+ if ( $toclevel < $wgMaxTocLevel ) {
if ( $prevtoclevel < $wgMaxTocLevel ) {
# Unindent only if the previous toc level was shown :p
$toc .= Linker::tocUnindent( $prevtoclevel - $toclevel );
@@ -4111,7 +4325,7 @@ class Parser {
}
} else {
# No change in level, end TOC line
- if ( $toclevel<$wgMaxTocLevel ) {
+ if ( $toclevel < $wgMaxTocLevel ) {
$toc .= Linker::tocLineEnd();
}
}
@@ -4119,9 +4333,9 @@ class Parser {
$levelCount[$toclevel] = $level;
# count number of headlines for each level
- @$sublevelCount[$toclevel]++;
+ $sublevelCount[$toclevel]++;
$dot = 0;
- for( $i = 1; $i <= $toclevel; $i++ ) {
+ for ( $i = 1; $i <= $toclevel; $i++ ) {
if ( !empty( $sublevelCount[$i] ) ) {
if ( $dot ) {
$numbering .= '.';
@@ -4144,23 +4358,29 @@ class Parser {
$safeHeadline = $this->mStripState->unstripBoth( $safeHeadline );
# Strip out HTML (first regex removes any tag not allowed)
- # Allowed tags are <sup> and <sub> (bug 8393), <i> (bug 26375) and <b> (r105284)
- # We strip any parameter from accepted tags (second regex)
+ # Allowed tags are:
+ # * <sup> and <sub> (bug 8393)
+ # * <i> (bug 26375)
+ # * <b> (r105284)
+ # * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
+ #
+ # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
+ # to allow setting directionality in toc items.
$tocline = preg_replace(
- array( '#<(?!/?(sup|sub|i|b)(?: [^>]*)?>).*?'.'>#', '#<(/?(sup|sub|i|b))(?: .*?)?'.'>#' ),
- array( '', '<$1>' ),
+ array( '#<(?!/?(span|sup|sub|i|b)(?: [^>]*)?>).*?' . '>#', '#<(/?(?:span(?: dir="(?:rtl|ltr)")?|sup|sub|i|b))(?: .*?)?' . '>#' ),
+ array( '', '<$1>' ),
$safeHeadline
);
$tocline = trim( $tocline );
# For the anchor, strip out HTML-y stuff period
- $safeHeadline = preg_replace( '/<.*?'.'>/', '', $safeHeadline );
+ $safeHeadline = preg_replace( '/<.*?' . '>/', '', $safeHeadline );
$safeHeadline = Sanitizer::normalizeSectionNameWhitespace( $safeHeadline );
# Save headline for section edit hint before it's escaped
$headlineHint = $safeHeadline;
- if ( $wgHtml5 && $wgExperimentalHtmlIds ) {
+ if ( $wgExperimentalHtmlIds ) {
# For reverse compatibility, provide an id that's
# HTML4-compatible, like we used to.
#
@@ -4230,7 +4450,8 @@ class Parser {
# Add the section to the section tree
# Find the DOM node for this header
- while ( $node && !$isTemplate ) {
+ $noOffset = ( $isTemplate || $sectionIndex === false );
+ while ( $node && !$noOffset ) {
if ( $node->getName() === 'h' ) {
$bits = $node->splitHeading();
if ( $bits['i'] == $sectionIndex ) {
@@ -4248,7 +4469,7 @@ class Parser {
'number' => $numbering,
'index' => ( $isTemplate ? 'T-' : '' ) . $sectionIndex,
'fromtitle' => $titleText,
- 'byteoffset' => ( $isTemplate ? null : $byteOffset ),
+ 'byteoffset' => ( $noOffset ? null : $byteOffset ),
'anchor' => $anchor,
);
@@ -4269,9 +4490,9 @@ class Parser {
// We use a page and section attribute to stop the language converter from converting these important bits
// of data, but put the headline hint inside a content block because the language converter is supposed to
// be able to convert that piece of data.
- $editlink = '<mw:editsection page="' . htmlspecialchars($editlinkArgs[0]);
- $editlink .= '" section="' . htmlspecialchars($editlinkArgs[1]) .'"';
- if ( isset($editlinkArgs[2]) ) {
+ $editlink = '<mw:editsection page="' . htmlspecialchars( $editlinkArgs[0] );
+ $editlink .= '" section="' . htmlspecialchars( $editlinkArgs[1] ) . '"';
+ if ( isset( $editlinkArgs[2] ) ) {
$editlink .= '>' . $editlinkArgs[2] . '</mw:editsection>';
} else {
$editlink .= '/>';
@@ -4299,6 +4520,7 @@ class Parser {
}
$toc = Linker::tocList( $toc, $this->mOptions->getUserLangObj() );
$this->mOutput->setTOCHTML( $toc );
+ $toc = self::TOC_START . $toc . self::TOC_END;
}
if ( $isMain ) {
@@ -4306,7 +4528,7 @@ class Parser {
}
# split up and insert constructed headlines
- $blocks = preg_split( '/<H[1-6].*?' . '>.*?<\/H[1-6]>/i', $text );
+ $blocks = preg_split( '/<H[1-6].*?' . '>[\s\S]*?<\/H[1-6]>/i', $text );
$i = 0;
// build an array of document sections
@@ -4353,7 +4575,7 @@ class Parser {
* Transform wiki markup when saving a page by doing "\r\n" -> "\n"
* conversion, substitting signatures, {{subst:}} templates, etc.
*
- * @param $text String: the text to transform
+ * @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 $options ParserOptions: parsing options
@@ -4368,7 +4590,7 @@ class Parser {
"\r\n" => "\n",
);
$text = str_replace( array_keys( $pairs ), array_values( $pairs ), $text );
- if( $options->getPreSaveTransform() ) {
+ if ( $options->getPreSaveTransform() ) {
$text = $this->pstPass2( $text, $user );
}
$text = $this->mStripState->unstripBoth( $text );
@@ -4388,7 +4610,7 @@ class Parser {
* @return string
*/
function pstPass2( $text, $user ) {
- global $wgContLang, $wgLocaltimezone;
+ global $wgContLang;
# Note: This is the timestamp saved as hardcoded wikitext to
# the database, we use $wgContLang here in order to give
@@ -4396,19 +4618,11 @@ class Parser {
# than the one selected in each user's preferences.
# (see also bug 12815)
$ts = $this->mOptions->getTimestamp();
- if ( isset( $wgLocaltimezone ) ) {
- $tz = $wgLocaltimezone;
- } else {
- $tz = date_default_timezone_get();
- }
+ $timestamp = MWTimestamp::getLocalInstance( $ts );
+ $ts = $timestamp->format( 'YmdHis' );
+ $tzMsg = $timestamp->format( 'T' ); # might vary on DST changeover!
- $unixts = wfTimestamp( TS_UNIX, $ts );
- $oldtz = date_default_timezone_get();
- date_default_timezone_set( $tz );
- $ts = date( 'YmdHis', $unixts );
- $tzMsg = date( 'T', $unixts ); # might vary on DST changeover!
-
- # Allow translation of timezones through wiki. date() can return
+ # Allow translation of timezones through wiki. format() can return
# whatever crap the system uses, localised or not, so we cannot
# ship premade translations.
$key = 'timezone-' . strtolower( trim( $tzMsg ) );
@@ -4417,8 +4631,6 @@ class Parser {
$tzMsg = $msg->text();
}
- date_default_timezone_set( $oldtz );
-
$d = $wgContLang->timeanddate( $ts, false, false ) . " ($tzMsg)";
# Variable replacement
@@ -4436,14 +4648,14 @@ class Parser {
'~~~' => $sigText
) );
- # Context links: [[|name]] and [[name (context)|]]
+ # Context links ("pipe tricks"): [[|name]] and [[name (context)|]]
$tc = '[' . Title::legalChars() . ']';
$nc = '[ _0-9A-Za-z\x80-\xff-]'; # Namespaces can use non-ascii!
- $p1 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\))\\|]]/"; # [[ns:page (context)|]]
- $p4 = "/\[\[(:?$nc+:|:|)($tc+?)( ?（$tc+）)\\|]]/"; # [[ns:page（context）|]]
- $p3 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\)|)((?:, |，)$tc+|)\\|]]/"; # [[ns:page (context), context|]]
- $p2 = "/\[\[\\|($tc+)]]/"; # [[|page]]
+ $p1 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\))\\|]]/"; # [[ns:page (context)|]]
+ $p4 = "/\[\[(:?$nc+:|:|)($tc+?)( ?（$tc+）)\\|]]/"; # [[ns:page（context）|]] (double-width brackets, added in r40257)
+ $p3 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\)|)((?:, |，)$tc+|)\\|]]/"; # [[ns:page (context), context|]] (using either single or double-width comma)
+ $p2 = "/\[\[\\|($tc+)]]/"; # [[|page]] (reverse pipe trick: add context from page title)
# try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
$text = preg_replace( $p1, '[[\\1\\2\\3|\\2]]', $text );
@@ -4476,7 +4688,7 @@ class Parser {
* as it may have changed if it's the $wgParser.
*
* @param $user User
- * @param $nickname String|bool nickname to use or false to use user's default nickname
+ * @param string|bool $nickname nickname to use or false to use user's default nickname
* @param $fancySig Boolean|null whether the nicknname is the complete signature
* or null to use default value
* @return string
@@ -4487,8 +4699,9 @@ class Parser {
$username = $user->getName();
# If not given, retrieve from the user object.
- if ( $nickname === false )
+ if ( $nickname === false ) {
$nickname = $user->getOption( 'nickname' );
+ }
if ( is_null( $fancySig ) ) {
$fancySig = $user->getBoolOption( 'fancysig' );
@@ -4507,7 +4720,7 @@ class Parser {
} else {
# Failed to validate; fall back to the default
$nickname = $username;
- wfDebug( __METHOD__.": $username has bad XML tags in signature.\n" );
+ wfDebug( __METHOD__ . ": $username has bad XML tags in signature.\n" );
}
}
@@ -4529,7 +4742,7 @@ class Parser {
* @return mixed An expanded string, or false if invalid.
*/
function validateSig( $text ) {
- return( Xml::isWellFormedXmlFragment( $text ) ? $text : false );
+ return Xml::isWellFormedXmlFragment( $text ) ? $text : false;
}
/**
@@ -4539,7 +4752,7 @@ class Parser {
* 2) Substitute all transclusions
*
* @param $text String
- * @param $parsing bool Whether we're cleaning (preferences save) or parsing
+ * @param bool $parsing Whether we're cleaning (preferences save) or parsing
* @return String: signature text
*/
public function cleanSig( $text, $parsing = false ) {
@@ -4614,7 +4827,7 @@ class Parser {
/**
* Wrapper for preprocess()
*
- * @param $text String: the text to preprocess
+ * @param string $text the text to preprocess
* @param $options ParserOptions: options
* @param $title Title object or null to use $wgTitle
* @return String
@@ -4633,11 +4846,7 @@ class Parser {
global $wgTitle;
$title = $wgTitle;
}
- if ( !$title ) {
- # It's not uncommon having a null $wgTitle in scripts. See r80898
- # Create a ghost title in such case
- $title = Title::newFromText( 'Dwimmerlaik' );
- }
+
$text = $this->preprocess( $text, $title, $options );
$executing = false;
@@ -4666,6 +4875,7 @@ class Parser {
*
* @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
+ * @throws MWException
* @return Mixed|null The old value of the mTagHooks array associated with the hook
*/
public function setHook( $tag, $callback ) {
@@ -4696,6 +4906,7 @@ class Parser {
*
* @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
+ * @throws MWException
* @return Mixed|null The old value of the mTagHooks array associated with the hook
*/
function setTransparentTagHook( $tag, $callback ) {
@@ -4734,7 +4945,7 @@ class Parser {
* nowiki Wiki markup in the return value should be escaped
* isHTML The returned text is HTML, armour it against wikitext transformation
*
- * @param $id String: The magic word ID
+ * @param string $id 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:...}}
@@ -4758,6 +4969,7 @@ class Parser {
* Please read the documentation in includes/parser/Preprocessor.php for more information
* about the methods available in PPFrame and PPNode.
*
+ * @throws MWException
* @return string|callback The old callback function for this name, if any
*/
public function setFunctionHook( $id, $callback, $flags = 0 ) {
@@ -4768,8 +4980,9 @@ class Parser {
# Add to function cache
$mw = MagicWord::get( $id );
- if ( !$mw )
- throw new MWException( __METHOD__.'() expecting a magic word identifier.' );
+ if ( !$mw ) {
+ throw new MWException( __METHOD__ . '() expecting a magic word identifier.' );
+ }
$synonyms = $mw->getSynonyms();
$sensitive = intval( $mw->isCaseSensitive() );
@@ -4805,11 +5018,17 @@ class Parser {
* Create a tag function, e.g. "<test>some stuff</test>".
* Unlike tag hooks, tag functions are parsed at preprocessor level.
* Unlike parser functions, their content is not preprocessed.
+ * @param $tag
+ * @param $callback
+ * @param $flags
+ * @throws MWException
* @return null
*/
function setFunctionTagHook( $tag, $callback, $flags ) {
$tag = strtolower( $tag );
- if ( preg_match( '/[<>\r\n]/', $tag, $m ) ) throw new MWException( "Invalid character {$m[0]} in setFunctionTagHook('$tag', ...) call" );
+ if ( preg_match( '/[<>\r\n]/', $tag, $m ) ) {
+ throw new MWException( "Invalid character {$m[0]} in setFunctionTagHook('$tag', ...) call" );
+ }
$old = isset( $this->mFunctionTagHooks[$tag] ) ?
$this->mFunctionTagHooks[$tag] : null;
$this->mFunctionTagHooks[$tag] = array( $callback, $flags );
@@ -4860,7 +5079,20 @@ class Parser {
* @return string HTML
*/
function renderImageGallery( $text, $params ) {
- $ig = new ImageGallery();
+ wfProfileIn( __METHOD__ );
+
+ $mode = false;
+ if ( isset( $params['mode'] ) ) {
+ $mode = $params['mode'];
+ }
+
+ try {
+ $ig = ImageGalleryBase::factory( $mode );
+ } catch ( MWException $e ) {
+ // If invalid type set, fallback to default.
+ $ig = ImageGalleryBase::factory( false );
+ }
+
$ig->setContextTitle( $this->mTitle );
$ig->setShowBytes( false );
$ig->setShowFilename( false );
@@ -4888,6 +5120,7 @@ class Parser {
if ( isset( $params['heights'] ) ) {
$ig->setHeights( $params['heights'] );
}
+ $ig->setAdditionalOptions( $params );
wfRunHooks( 'BeforeParserrenderImageGallery', array( &$this, &$ig ) );
@@ -4911,34 +5144,81 @@ class Parser {
continue;
}
+ # We need to get what handler the file uses, to figure out parameters.
+ # Note, a hook can overide the file name, and chose an entirely different
+ # file (which potentially could be of a different type and have different handler).
+ $options = array();
+ $descQuery = false;
+ wfRunHooks( 'BeforeParserFetchFileAndTitle',
+ array( $this, $title, &$options, &$descQuery ) );
+ # Don't register it now, as ImageGallery does that later.
+ $file = $this->fetchFileNoRegister( $title, $options );
+ $handler = $file ? $file->getHandler() : false;
+
+ wfProfileIn( __METHOD__ . '-getMagicWord' );
+ $paramMap = array(
+ 'img_alt' => 'gallery-internal-alt',
+ 'img_link' => 'gallery-internal-link',
+ );
+ if ( $handler ) {
+ $paramMap = $paramMap + $handler->getParamMap();
+ // We don't want people to specify per-image widths.
+ // Additionally the width parameter would need special casing anyhow.
+ unset( $paramMap['img_width'] );
+ }
+
+ $mwArray = new MagicWordArray( array_keys( $paramMap ) );
+ wfProfileOut( __METHOD__ . '-getMagicWord' );
+
$label = '';
$alt = '';
$link = '';
+ $handlerOptions = array();
if ( isset( $matches[3] ) ) {
// look for an |alt= definition while trying not to break existing
// captions with multiple pipes (|) in it, until a more sensible grammar
// is defined for images in galleries
+ // FIXME: Doing recursiveTagParse at this stage, and the trim before
+ // splitting on '|' is a bit odd, and different from makeImage.
$matches[3] = $this->recursiveTagParse( trim( $matches[3] ) );
- $parameterMatches = StringUtils::explode('|', $matches[3]);
- $magicWordAlt = MagicWord::get( 'img_alt' );
- $magicWordLink = MagicWord::get( 'img_link' );
+ $parameterMatches = StringUtils::explode( '|', $matches[3] );
foreach ( $parameterMatches as $parameterMatch ) {
- if ( $match = $magicWordAlt->matchVariableStartToEnd( $parameterMatch ) ) {
- $alt = $this->stripAltText( $match, false );
- }
- elseif( $match = $magicWordLink->matchVariableStartToEnd( $parameterMatch ) ){
- $link = strip_tags($this->replaceLinkHoldersText($match));
- $chars = self::EXT_LINK_URL_CLASS;
- $prots = $this->mUrlProtocols;
- //check to see if link matches an absolute url, if not then it must be a wiki link.
- if(!preg_match( "/^($prots)$chars+$/u", $link)){
- $localLinkTitle = Title::newFromText($link);
- $link = $localLinkTitle->getLocalURL();
+ list( $magicName, $match ) = $mwArray->matchVariableStartToEnd( $parameterMatch );
+ if ( $magicName ) {
+ $paramName = $paramMap[$magicName];
+
+ switch ( $paramName ) {
+ case 'gallery-internal-alt':
+ $alt = $this->stripAltText( $match, false );
+ break;
+ case 'gallery-internal-link':
+ $linkValue = strip_tags( $this->replaceLinkHoldersText( $match ) );
+ $chars = self::EXT_LINK_URL_CLASS;
+ $prots = $this->mUrlProtocols;
+ //check to see if link matches an absolute url, if not then it must be a wiki link.
+ if ( preg_match( "/^($prots)$chars+$/u", $linkValue ) ) {
+ $link = $linkValue;
+ } else {
+ $localLinkTitle = Title::newFromText( $linkValue );
+ if ( $localLinkTitle !== null ) {
+ $link = $localLinkTitle->getLocalURL();
+ }
+ }
+ break;
+ default:
+ // Must be a handler specific parameter.
+ if ( $handler->validateParam( $paramName, $match ) ) {
+ $handlerOptions[$paramName] = $match;
+ } else {
+ // Guess not. Append it to the caption.
+ wfDebug( "$parameterMatch failed parameter validation" );
+ $label .= '|' . $parameterMatch;
+ }
}
- }
- else {
+
+ } else {
// concatenate all other pipes
$label .= '|' . $parameterMatch;
}
@@ -4947,9 +5227,11 @@ class Parser {
$label = substr( $label, 1 );
}
- $ig->add( $title, $label, $alt ,$link);
+ $ig->add( $title, $label, $alt, $link, $handlerOptions );
}
- return $ig->toHTML();
+ $html = $ig->toHTML();
+ wfProfileOut( __METHOD__ );
+ return $html;
}
/**
@@ -4962,7 +5244,7 @@ class Parser {
} else {
$handlerClass = '';
}
- if ( !isset( $this->mImageParams[$handlerClass] ) ) {
+ if ( !isset( $this->mImageParams[$handlerClass] ) ) {
# Initialise static lists
static $internalParamNames = array(
'horizAlign' => array( 'left', 'right', 'center', 'none' ),
@@ -5064,14 +5346,14 @@ class Parser {
# Special case; width and height come in one variable together
if ( $type === 'handler' && $paramName === 'width' ) {
$parsedWidthParam = $this->parseWidthParam( $value );
- if( isset( $parsedWidthParam['width'] ) ) {
+ if ( isset( $parsedWidthParam['width'] ) ) {
$width = $parsedWidthParam['width'];
if ( $handler->validateParam( 'width', $width ) ) {
$params[$type]['width'] = $width;
$validated = true;
}
}
- if( isset( $parsedWidthParam['height'] ) ) {
+ if ( isset( $parsedWidthParam['height'] ) ) {
$height = $parsedWidthParam['height'];
if ( $handler->validateParam( 'height', $height ) ) {
$params[$type]['height'] = $height;
@@ -5085,7 +5367,7 @@ class Parser {
$validated = $handler->validateParam( $paramName, $value );
} else {
# Validate internal parameters
- switch( $paramName ) {
+ switch ( $paramName ) {
case 'manualthumb':
case 'alt':
case 'class':
@@ -5180,7 +5462,7 @@ class Parser {
} else { # Inline image
if ( !isset( $params['frame']['alt'] ) ) {
# No alt text, use the "caption" for the alt text
- if ( $caption !== '') {
+ if ( $caption !== '' ) {
$params['frame']['alt'] = $this->stripAltText( $caption, $holders );
} else {
# No caption, fall back to using the filename for the
@@ -5303,8 +5585,8 @@ class Parser {
*
* External callers should use the getSection and replaceSection methods.
*
- * @param $text String: Page wikitext
- * @param $section String: a section identifier string of the form:
+ * @param string $text Page wikitext
+ * @param string $section a section identifier string of the form:
* "<flag1> - <flag2> - ... - <section number>"
*
* Currently the only recognised flag is "T", which means the target section number
@@ -5321,12 +5603,12 @@ class Parser {
* string. If $text is the empty string and section 0 is replaced, $newText is
* returned.
*
- * @param $mode String: one of "get" or "replace"
- * @param $newText String: replacement text for section data.
+ * @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.
*/
- private function extractSections( $text, $section, $mode, $newText='' ) {
+ private function extractSections( $text, $section, $mode, $newText = '' ) {
global $wgTitle; # not generally used but removes an ugly failure mode
$this->startParse( $wgTitle, new ParserOptions, self::OT_PLAIN, true );
$outText = '';
@@ -5442,12 +5724,12 @@ class Parser {
*
* If a section contains subsections, these are also returned.
*
- * @param $text String: text to look in
- * @param $section String: section identifier
- * @param $deftext String: default to return if section is not found
+ * @param string $text text to look in
+ * @param string $section section identifier
+ * @param string $deftext default to return if section is not found
* @return string text of the requested section
*/
- public function getSection( $text, $section, $deftext='' ) {
+ public function getSection( $text, $section, $deftext = '' ) {
return $this->extractSections( $text, $section, "get", $deftext );
}
@@ -5456,9 +5738,9 @@ class Parser {
* specified by $section has been replaced with $text. If the target
* section does not exist, $oldtext is returned unchanged.
*
- * @param $oldtext String: former text of the article
- * @param $section int section identifier
- * @param $text String: replacing text
+ * @param string $oldtext former text of the article
+ * @param int $section section identifier
+ * @param string $text replacing text
* @return String: modified text
*/
public function replaceSection( $oldtext, $section, $text ) {
@@ -5523,14 +5805,14 @@ class Parser {
* @return String: user name
*/
function getRevisionUser() {
- if( is_null( $this->mRevisionUser ) ) {
+ if ( is_null( $this->mRevisionUser ) ) {
$revObject = $this->getRevisionObject();
# if this template is subst: the revision id will be blank,
# so just use the current user's name
- if( $revObject ) {
+ if ( $revObject ) {
$this->mRevisionUser = $revObject->getUserText();
- } elseif( $this->ot['wiki'] || $this->mOptions->getIsPreview() ) {
+ } elseif ( $this->ot['wiki'] || $this->mOptions->getIsPreview() ) {
$this->mRevisionUser = $this->getUser()->getName();
}
}
@@ -5538,9 +5820,30 @@ class Parser {
}
/**
+ * Get the size of the revision
+ *
+ * @return int|null revision size
+ */
+ function getRevisionSize() {
+ if ( is_null( $this->mRevisionSize ) ) {
+ $revObject = $this->getRevisionObject();
+
+ # if this variable is subst: the revision id will be blank,
+ # so just use the parser input size, because the own substituation
+ # will change the size.
+ if ( $revObject ) {
+ $this->mRevisionSize = $revObject->getSize();
+ } elseif ( $this->ot['wiki'] || $this->mOptions->getIsPreview() ) {
+ $this->mRevisionSize = $this->mInputSize;
+ }
+ }
+ return $this->mRevisionSize;
+ }
+
+ /**
* Mutator for $mDefaultSort
*
- * @param $sort string New value
+ * @param string $sort New value
*/
public function setDefaultSort( $sort ) {
$this->mDefaultSort = $sort;
@@ -5596,7 +5899,7 @@ class Parser {
* 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
+ * @param string $text The section name
* @return string An anchor
*/
public function guessLegacySectionNameFromWikiText( $text ) {
@@ -5616,7 +5919,7 @@ class Parser {
* to create valid section anchors by mimicing the output of the
* parser when headings are parsed.
*
- * @param $text String: text string to be stripped of wikitext
+ * @param string $text text string to be stripped of wikitext
* for use in a Section anchor
* @return string Filtered text string
*/
@@ -5767,12 +6070,13 @@ class Parser {
* If the $data array has been stored persistently, the caller should first
* check whether it is still valid, by calling isValidHalfParsedText().
*
- * @param $data array Serialized data
+ * @param array $data Serialized data
+ * @throws MWException
* @return String
*/
function unserializeHalfParsedText( $data ) {
if ( !isset( $data['version'] ) || $data['version'] != self::HALF_PARSED_VERSION ) {
- throw new MWException( __METHOD__.': invalid version' );
+ throw new MWException( __METHOD__ . ': invalid version' );
}
# First, extract the strip state.
@@ -5809,7 +6113,7 @@ class Parser {
*/
public function parseWidthParam( $value ) {
$parsedWidthParam = array();
- if( $value === '' ) {
+ if ( $value === '' ) {
return $parsedWidthParam;
}
$m = array();
diff --git a/includes/parser/ParserCache.php b/includes/parser/ParserCache.php
index 6a4ef0c5..7053f134 100644
--- a/includes/parser/ParserCache.php
+++ b/includes/parser/ParserCache.php
@@ -48,6 +48,7 @@ class ParserCache {
* May be a memcached client or a BagOStuff derivative.
*
* @param $memCached Object
+ * @throws MWException
*/
protected function __construct( $memCached ) {
if ( !$memCached ) {
@@ -66,7 +67,7 @@ class ParserCache {
// idhash seem to mean 'page id' + 'rendering hash' (r3710)
$pageid = $article->getID();
- $renderkey = (int)($wgRequest->getVal('action') == 'render');
+ $renderkey = (int)( $wgRequest->getVal( 'action' ) == 'render' );
$key = wfMemcKey( 'pcache', 'idhash', "{$pageid}-{$renderkey}!{$hash}" );
return $key;
@@ -127,7 +128,7 @@ class ParserCache {
public function getKey( $article, $popts, $useOutdated = true ) {
global $wgCacheEpoch;
- if( $popts instanceof User ) {
+ if ( $popts instanceof User ) {
wfWarn( "Use of outdated prototype ParserCache::getKey( &\$article, &\$user )\n" );
$popts = ParserOptions::newFromUser( $popts );
}
@@ -200,8 +201,8 @@ class ParserCache {
wfDebug( "ParserOutput cache found.\n" );
- // The edit section preference may not be the appropiate one in
- // the ParserOutput, as we are not storing it in the parsercache
+ // The edit section preference may not be the appropiate one in
+ // the ParserOutput, as we are not storing it in the parsercache
// key. Force it here. See bug 31445.
$value->setEditSectionTokens( $popts->getEditSection() );
@@ -222,19 +223,19 @@ class ParserCache {
* @param $parserOutput ParserOutput
* @param $article Article
* @param $popts ParserOptions
+ * @param $cacheTime Time when the cache was generated
*/
- public function save( $parserOutput, $article, $popts ) {
+ public function save( $parserOutput, $article, $popts, $cacheTime = null ) {
$expire = $parserOutput->getCacheExpiry();
-
- if( $expire > 0 ) {
- $now = wfTimestampNow();
+ if ( $expire > 0 ) {
+ $cacheTime = $cacheTime ?: wfTimestampNow();
$optionsKey = new CacheTime;
$optionsKey->mUsedOptions = $parserOutput->getUsedOptions();
$optionsKey->updateCacheExpiry( $expire );
- $optionsKey->setCacheTime( $now );
- $parserOutput->setCacheTime( $now );
+ $optionsKey->setCacheTime( $cacheTime );
+ $parserOutput->setCacheTime( $cacheTime );
$optionsKey->setContainsOldMagic( $parserOutput->containsOldMagic() );
@@ -244,8 +245,8 @@ class ParserCache {
// Save the timestamp so that we don't have to load the revision row on view
$parserOutput->setTimestamp( $article->getTimestamp() );
- $parserOutput->mText .= "\n<!-- Saved in parser cache with key $parserOutputKey and timestamp $now -->\n";
- wfDebug( "Saved in parser cache with key $parserOutputKey and timestamp $now\n" );
+ $parserOutput->mText .= "\n<!-- Saved in parser cache with key $parserOutputKey and timestamp $cacheTime\n -->\n";
+ wfDebug( "Saved in parser cache with key $parserOutputKey and timestamp $cacheTime\n" );
// Save the parser output
$this->mMemc->set( $parserOutputKey, $parserOutput, $expire );
diff --git a/includes/parser/ParserOptions.php b/includes/parser/ParserOptions.php
index 009b18a1..e12f32d8 100644
--- a/includes/parser/ParserOptions.php
+++ b/includes/parser/ParserOptions.php
@@ -29,67 +29,62 @@
* @ingroup Parser
*/
class ParserOptions {
-
- /**
- * Use DateFormatter to format dates
- */
- var $mUseDynamicDates;
-
+
/**
* Interlanguage links are removed and returned in an array
*/
var $mInterwikiMagic;
-
+
/**
* Allow external images inline?
*/
var $mAllowExternalImages;
-
+
/**
* If not, any exception?
*/
var $mAllowExternalImagesFrom;
-
+
/**
* If not or it doesn't match, should we check an on-wiki whitelist?
*/
var $mEnableImageWhitelist;
-
+
/**
* Date format index
*/
var $mDateFormat = null;
-
+
/**
* Create "edit section" links?
*/
var $mEditSection = true;
-
+
/**
* Allow inclusion of special pages?
*/
var $mAllowSpecialInclusion;
-
+
/**
* Use tidy to cleanup output HTML?
*/
var $mTidy = false;
-
+
/**
* Which lang to call for PLURAL and GRAMMAR
*/
var $mInterfaceMessage = false;
-
+
/**
* Overrides $mInterfaceMessage with arbitrary language
*/
var $mTargetLanguage = null;
-
+
/**
* Maximum size of template expansions, in bytes
*/
var $mMaxIncludeSize;
-
+
/**
* Maximum number of nodes touched by PPFrame::expand()
*/
@@ -99,56 +94,56 @@ class ParserOptions {
* Maximum number of nodes generated by Preprocessor::preprocessToObj()
*/
var $mMaxGeneratedPPNodeCount;
-
+
/**
* Maximum recursion depth in PPFrame::expand()
*/
var $mMaxPPExpandDepth;
-
+
/**
* Maximum recursion depth for templates within templates
*/
var $mMaxTemplateDepth;
-
+
/**
* Maximum number of calls per parse to expensive parser functions
*/
var $mExpensiveParserFunctionLimit;
-
+
/**
* Remove HTML comments. ONLY APPLIES TO PREPROCESS OPERATIONS
*/
var $mRemoveComments = true;
-
+
/**
* Callback for template fetching. Used as first argument to call_user_func().
*/
var $mTemplateCallback =
array( 'Parser', 'statelessFetchTemplate' );
-
+
/**
* Enable limit report in an HTML comment on output
*/
var $mEnableLimitReport = false;
-
+
/**
* Timestamp used for {{CURRENTDAY}} etc.
*/
var $mTimestamp;
-
+
/**
* Target attribute for external links
*/
var $mExternalLinkTarget;
-
+
/**
- * Clean up signature texts?
+ * Clean up signature texts?
*
* 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures
* 2) Substitute all transclusions
*/
var $mCleanSignatures;
-
+
/**
* Transform wiki markup when saving the page?
*/
@@ -168,43 +163,43 @@ class ParserOptions {
* Automatically number headings?
*/
var $mNumberHeadings;
-
+
/**
* User math preference (as integer). Not used (1.19)
*/
var $mMath;
-
+
/**
* Thumb size preferred by the user.
*/
var $mThumbSize;
-
+
/**
* Maximum article size of an article to be marked as "stub"
*/
private $mStubThreshold;
-
+
/**
* Language object of the User language.
*/
var $mUserLang;
/**
- * @var User
+ * @var User
* Stored user object
*/
var $mUser;
-
+
/**
* Parsing the page for a "preview" operation?
*/
var $mIsPreview = false;
-
+
/**
* Parsing the page for a "preview" operation on a single section?
*/
var $mIsSectionPreview = false;
-
+
/**
* Parsing the printable version of the page?
*/
@@ -220,7 +215,6 @@ class ParserOptions {
*/
protected $onAccessCallback = null;
- function getUseDynamicDates() { return $this->mUseDynamicDates; }
function getInterwikiMagic() { return $this->mInterwikiMagic; }
function getAllowExternalImages() { return $this->mAllowExternalImages; }
function getAllowExternalImagesFrom() { return $this->mAllowExternalImagesFrom; }
@@ -246,6 +240,7 @@ class ParserOptions {
function getExternalLinkTarget() { return $this->mExternalLinkTarget; }
function getDisableContentConversion() { return $this->mDisableContentConversion; }
function getDisableTitleConversion() { return $this->mDisableTitleConversion; }
+ /** @deprecated since 1.22 use User::getOption('math') instead */
function getMath() { $this->optionUsed( 'math' );
return $this->mMath; }
function getThumbSize() { $this->optionUsed( 'thumbsize' );
@@ -286,9 +281,17 @@ class ParserOptions {
}
/**
+ * Get the user language used by the parser for this page.
+ *
* You shouldn't use this. Really. $parser->getFunctionLang() is all you need.
- * Using this fragments the cache and is discouraged. Yes, {{int: }} uses this,
- * producing inconsistent tables (Bug 14404).
+ *
+ * To avoid side-effects where the page will be rendered based on the language
+ * of the user who last saved, this function will triger a cache fragmentation.
+ * Usage of this method is discouraged for that reason.
+ *
+ * When saving, this will return the default language instead of the user's.
+ *
+ * {{int: }} uses this which used to produce inconsistent link tables (bug 14404).
*
* @return Language object
* @since 1.19
@@ -308,7 +311,6 @@ class ParserOptions {
return $this->getUserLangObj()->getCode();
}
- function setUseDynamicDates( $x ) { return wfSetVar( $this->mUseDynamicDates, $x ); }
function setInterwikiMagic( $x ) { return wfSetVar( $this->mInterwikiMagic, $x ); }
function setAllowExternalImages( $x ) { return wfSetVar( $this->mAllowExternalImages, $x ); }
function setAllowExternalImagesFrom( $x ) { return wfSetVar( $this->mAllowExternalImagesFrom, $x ); }
@@ -319,7 +321,7 @@ class ParserOptions {
function setAllowSpecialInclusion( $x ) { return wfSetVar( $this->mAllowSpecialInclusion, $x ); }
function setTidy( $x ) { return wfSetVar( $this->mTidy, $x ); }
- /** @deprecated in 1.19; will be removed in 1.20 */
+ /** @deprecated in 1.19 */
function setSkin( $x ) { wfDeprecated( __METHOD__, '1.19' ); }
function setInterfaceMessage( $x ) { return wfSetVar( $this->mInterfaceMessage, $x ); }
function setTargetLanguage( $x ) { return wfSetVar( $this->mTargetLanguage, $x, true ); }
@@ -337,6 +339,7 @@ class ParserOptions {
function setExternalLinkTarget( $x ) { return wfSetVar( $this->mExternalLinkTarget, $x ); }
function disableContentConversion( $x = true ) { return wfSetVar( $this->mDisableContentConversion, $x ); }
function disableTitleConversion( $x = true ) { return wfSetVar( $this->mDisableTitleConversion, $x ); }
+ /** @deprecated since 1.22 */
function setMath( $x ) { return wfSetVar( $this->mMath, $x ); }
function setUserLang( $x ) {
if ( is_string( $x ) ) {
@@ -415,14 +418,14 @@ class ParserOptions {
return new ParserOptions( $context->getUser(), $context->getLanguage() );
}
- /**
- * Get user options
+ /**
+ * Get user options
*
* @param $user User object
* @param $lang Language object
*/
private function initialiseFromUser( $user, $lang ) {
- global $wgUseDynamicDates, $wgInterwikiMagic, $wgAllowExternalImages,
+ global $wgInterwikiMagic, $wgAllowExternalImages,
$wgAllowExternalImagesFrom, $wgEnableImageWhitelist, $wgAllowSpecialInclusion,
$wgMaxArticleSize, $wgMaxPPNodeCount, $wgMaxTemplateDepth, $wgMaxPPExpandDepth,
$wgCleanSignatures, $wgExternalLinkTarget, $wgExpensiveParserFunctionLimit,
@@ -430,7 +433,6 @@ class ParserOptions {
wfProfileIn( __METHOD__ );
- $this->mUseDynamicDates = $wgUseDynamicDates;
$this->mInterwikiMagic = $wgInterwikiMagic;
$this->mAllowExternalImages = $wgAllowExternalImages;
$this->mAllowExternalImagesFrom = $wgAllowExternalImagesFrom;
@@ -481,12 +483,7 @@ class ParserOptions {
* @return array
*/
public static function legacyOptions() {
- global $wgUseDynamicDates;
- $legacyOpts = array( 'math', 'stubthreshold', 'numberheadings', 'userlang', 'thumbsize', 'editsection', 'printable' );
- if ( $wgUseDynamicDates ) {
- $legacyOpts[] = 'dateformat';
- }
- return $legacyOpts;
+ return array( 'math', 'stubthreshold', 'numberheadings', 'userlang', 'thumbsize', 'editsection', 'printable' );
}
/**
@@ -518,14 +515,13 @@ class ParserOptions {
$confstr .= '*';
}
-
// Space assigned for the stubthreshold but unused
// since it disables the parser cache, its value will always
// be 0 when this function is called by parsercache.
if ( in_array( 'stubthreshold', $forOptions ) ) {
$confstr .= '!' . $this->mStubThreshold;
} else {
- $confstr .= '!*' ;
+ $confstr .= '!*';
}
if ( in_array( 'dateformat', $forOptions ) ) {
@@ -552,7 +548,7 @@ class ParserOptions {
// add in language specific options, if any
// @todo FIXME: This is just a way of retrieving the url/user preferred variant
- if( !is_null( $title ) ) {
+ if ( !is_null( $title ) ) {
$confstr .= $title->getPageLanguage()->getExtraHashOptions();
} else {
global $wgContLang;
@@ -571,8 +567,9 @@ class ParserOptions {
$confstr .= '!printable=1';
}
- if ( $this->mExtraKey != '' )
+ if ( $this->mExtraKey != '' ) {
$confstr .= $this->mExtraKey;
+ }
// Give a chance for extensions to modify the hash, if they have
// extra options or other effects on the parser cache.
diff --git a/includes/parser/ParserOutput.php b/includes/parser/ParserOutput.php
index 41b4a385..502f0fd1 100644
--- a/includes/parser/ParserOutput.php
+++ b/includes/parser/ParserOutput.php
@@ -47,10 +47,14 @@ class ParserOutput extends CacheTime {
$mEditSectionTokens = false, # prefix/suffix markers if edit sections were output as tokens
$mProperties = array(), # Name/value pairs to be cached in the DB
$mTOCHTML = '', # HTML of the TOC
- $mTimestamp; # Timestamp of the revision
+ $mTimestamp, # Timestamp of the revision
+ $mTOCEnabled = true; # Whether TOC should be shown, can't override __NOTOC__
private $mIndexPolicy = ''; # 'index' or 'noindex'? Any other value will result in no change.
private $mAccessedOptions = array(); # List of ParserOptions (stored in the keys)
- private $mSecondaryDataUpdates = array(); # List of instances of SecondaryDataObject(), used to cause some information extracted from the page in a custom place.
+ private $mSecondaryDataUpdates = array(); # List of DataUpdate, used to save info from the page somewhere else.
+ private $mExtensionData = array(); # extra data used by extensions
+ private $mLimitReportData = array(); # Parser limit report data
+ private $mParseStartTime = array(); # Timestamps for getTimeSinceStart()
const EDITSECTION_REGEX = '#<(?:mw:)?editsection page="(.*?)" section="(.*?)"(?:/>|>(.*?)(</(?:mw:)?editsection>))#';
@@ -65,28 +69,46 @@ class ParserOutput extends CacheTime {
}
function getText() {
+ wfProfileIn( __METHOD__ );
+ $text = $this->mText;
if ( $this->mEditSectionTokens ) {
- return preg_replace_callback( ParserOutput::EDITSECTION_REGEX,
- array( &$this, 'replaceEditSectionLinksCallback' ), $this->mText );
+ $text = preg_replace_callback( ParserOutput::EDITSECTION_REGEX,
+ array( &$this, 'replaceEditSectionLinksCallback' ), $text );
+ } else {
+ $text = preg_replace( ParserOutput::EDITSECTION_REGEX, '', $text );
}
- return preg_replace( ParserOutput::EDITSECTION_REGEX, '', $this->mText );
+
+ // If you have an old cached version of this class - sorry, you can't disable the TOC
+ if ( isset( $this->mTOCEnabled ) && $this->mTOCEnabled ) {
+ $text = str_replace( array( Parser::TOC_START, Parser::TOC_END ), '', $text );
+ } else {
+ $text = preg_replace(
+ '#'. preg_quote( Parser::TOC_START ) . '.*?' . preg_quote( Parser::TOC_END ) . '#s',
+ '',
+ $text
+ );
+ }
+ wfProfileOut( __METHOD__ );
+ return $text;
}
/**
* callback used by getText to replace editsection tokens
* @private
+ * @param $m
+ * @throws MWException
* @return mixed
*/
function replaceEditSectionLinksCallback( $m ) {
global $wgOut, $wgLang;
$args = array(
- htmlspecialchars_decode($m[1]),
- htmlspecialchars_decode($m[2]),
- isset($m[4]) ? $m[3] : null,
+ htmlspecialchars_decode( $m[1] ),
+ htmlspecialchars_decode( $m[2] ),
+ isset( $m[4] ) ? $m[3] : null,
);
$args[0] = Title::newFromText( $args[0] );
- if ( !is_object($args[0]) ) {
- throw new MWException("Bad parser output text.");
+ if ( !is_object( $args[0] ) ) {
+ throw new MWException( "Bad parser output text." );
}
$args[] = $wgLang->getCode();
$skin = $wgOut->getSkin();
@@ -117,6 +139,8 @@ class ParserOutput extends CacheTime {
function getIndexPolicy() { return $this->mIndexPolicy; }
function getTOCHTML() { return $this->mTOCHTML; }
function getTimestamp() { return $this->mTimestamp; }
+ function getLimitReportData() { return $this->mLimitReportData; }
+ function getTOCEnabled() { return $this->mTOCEnabled; }
function setText( $text ) { return wfSetVar( $this->mText, $text ); }
function setLanguageLinks( $ll ) { return wfSetVar( $this->mLanguageLinks, $ll ); }
@@ -128,6 +152,7 @@ class ParserOutput extends CacheTime {
function setIndexPolicy( $policy ) { return wfSetVar( $this->mIndexPolicy, $policy ); }
function setTOCHTML( $tochtml ) { return wfSetVar( $this->mTOCHTML, $tochtml ); }
function setTimestamp( $timestamp ) { return wfSetVar( $this->mTimestamp, $timestamp ); }
+ function setTOCEnabled( $flag ) { return wfSetVar( $this->mTOCEnabled, $flag ); }
function addCategory( $c, $sort ) { $this->mCategories[$c] = $sort; }
function addLanguageLink( $t ) { $this->mLanguageLinks[] = $t; }
@@ -140,21 +165,45 @@ class ParserOutput extends CacheTime {
function setNewSection( $value ) {
$this->mNewSection = (bool)$value;
}
- function hideNewSection ( $value ) {
+ function hideNewSection( $value ) {
$this->mHideNewSection = (bool)$value;
}
- function getHideNewSection () {
+ function getHideNewSection() {
return (bool)$this->mHideNewSection;
}
function getNewSection() {
return (bool)$this->mNewSection;
}
+ /**
+ * Checks, if a url is pointing to the own server
+ *
+ * @param string $internal the server to check against
+ * @param string $url the url to check
+ * @return bool
+ */
+ static function isLinkInternal( $internal, $url ) {
+ return (bool)preg_match( '/^' .
+ # If server is proto relative, check also for http/https links
+ ( substr( $internal, 0, 2 ) === '//' ? '(?:https?:)?' : '' ) .
+ preg_quote( $internal, '/' ) .
+ # check for query/path/anchor or end of link in each case
+ '(?:[\?\/\#]|$)/i',
+ $url
+ );
+ }
+
function addExternalLink( $url ) {
# We don't register links pointing to our own server, unless... :-)
global $wgServer, $wgRegisterInternalExternals;
- if( $wgRegisterInternalExternals or stripos($url,$wgServer.'/')!==0)
+
+ $registerExternalLink = true;
+ if ( !$wgRegisterInternalExternals ) {
+ $registerExternalLink = !self::isLinkInternal( $wgServer, $url );
+ }
+ if ( $registerExternalLink ) {
$this->mExternalLinks[$url] = 1;
+ }
}
/**
@@ -163,7 +212,7 @@ class ParserOutput extends CacheTime {
* @param $title Title object
* @param $id Mixed: optional known page_id so we can skip the lookup
*/
- function addLink( $title, $id = null ) {
+ function addLink( Title $title, $id = null ) {
if ( $title->isExternal() ) {
// Don't record interwikis in pagelinks
$this->addInterwikiLink( $title );
@@ -174,11 +223,11 @@ class ParserOutput extends CacheTime {
if ( $ns == NS_MEDIA ) {
// Normalize this pseudo-alias if it makes it down here...
$ns = NS_FILE;
- } elseif( $ns == NS_SPECIAL ) {
+ } elseif ( $ns == NS_SPECIAL ) {
// We don't record Special: links currently
// It might actually be wise to, but we'd need to do some normalization.
return;
- } elseif( $dbk === '' ) {
+ } elseif ( $dbk === '' ) {
// Don't record self links - [[#Foo]]
return;
}
@@ -193,9 +242,9 @@ class ParserOutput extends CacheTime {
/**
* Register a file dependency for this output
- * @param $name string Title dbKey
- * @param $timestamp string MW timestamp of file creation (or false if non-existing)
- * @param $sha1 string base 36 SHA-1 of file (or false if non-existing)
+ * @param string $name Title dbKey
+ * @param string $timestamp MW timestamp of file creation (or false if non-existing)
+ * @param string $sha1 base 36 SHA-1 of file (or false if non-existing)
* @return void
*/
function addImage( $name, $timestamp = null, $sha1 = null ) {
@@ -231,10 +280,10 @@ class ParserOutput extends CacheTime {
*/
function addInterwikiLink( $title ) {
$prefix = $title->getInterwiki();
- if( $prefix == '' ) {
+ if ( $prefix == '' ) {
throw new MWException( 'Non-interwiki link passed, internal parser error.' );
}
- if (!isset($this->mInterwikiLinks[$prefix])) {
+ if ( !isset( $this->mInterwikiLinks[$prefix] ) ) {
$this->mInterwikiLinks[$prefix] = array();
}
$this->mInterwikiLinks[$prefix][$title->getDBkey()] = 1;
@@ -254,7 +303,7 @@ class ParserOutput extends CacheTime {
}
public function addModules( $modules ) {
- $this->mModules = array_merge( $this->mModules, (array) $modules );
+ $this->mModules = array_merge( $this->mModules, (array)$modules );
}
public function addModuleScripts( $modules ) {
@@ -288,7 +337,7 @@ class ParserOutput extends CacheTime {
* -- this is assumed to have been validated
* (check equal normalisation, etc.)
*
- * @param $text String: desired title text
+ * @param string $text desired title text
*/
public function setDisplayTitle( $text ) {
$this->setTitleText( $text );
@@ -302,7 +351,7 @@ class ParserOutput extends CacheTime {
*/
public function getDisplayTitle() {
$t = $this->getTitleText();
- if( $t === '' ) {
+ if ( $t === '' ) {
return false;
}
return $t;
@@ -320,13 +369,67 @@ class ParserOutput extends CacheTime {
}
/**
- * Set a property to be cached in the DB
+ * Set a property to be stored in the page_props database table.
+ *
+ * page_props is a key value store indexed by the page ID. This allows
+ * the parser to set a property on a page which can then be quickly
+ * retrieved given the page ID or via a DB join when given the page
+ * title.
+ *
+ * setProperty() is thus used to propagate properties from the parsed
+ * page to request contexts other than a page view of the currently parsed
+ * article.
+ *
+ * Some applications examples:
+ *
+ * * To implement hidden categories, hiding pages from category listings
+ * by storing a property.
+ *
+ * * Overriding the displayed article title.
+ * @see ParserOutput::setDisplayTitle()
+ *
+ * * To implement image tagging, for example displaying an icon on an
+ * image thumbnail to indicate that it is listed for deletion on
+ * Wikimedia Commons.
+ * This is not actually implemented, yet but would be pretty cool.
+ *
+ * @note: Do not use setProperty() to set a property which is only used
+ * in a context where the ParserOutput object itself is already available,
+ * for example a normal page view. There is no need to save such a property
+ * in the database since it the text is already parsed. You can just hook
+ * OutputPageParserOutput and get your data out of the ParserOutput object.
+ *
+ * If you are writing an extension where you want to set a property in the
+ * parser which is used by an OutputPageParserOutput hook, you have to
+ * associate the extension data directly with the ParserOutput object.
+ * Since MediaWiki 1.21, you can use setExtensionData() to do this:
+ *
+ * @par Example:
+ * @code
+ * $parser->getOutput()->setExtensionData( 'my_ext_foo', '...' );
+ * @endcode
+ *
+ * And then later, in OutputPageParserOutput or similar:
+ *
+ * @par Example:
+ * @code
+ * $output->getExtensionData( 'my_ext_foo' );
+ * @endcode
+ *
+ * In MediaWiki 1.20 and older, you have to use a custom member variable
+ * within the ParserOutput object:
+ *
+ * @par Example:
+ * @code
+ * $parser->getOutput()->my_ext_foo = '...';
+ * @endcode
+ *
*/
public function setProperty( $name, $value ) {
$this->mProperties[$name] = $value;
}
- public function getProperty( $name ){
+ public function getProperty( $name ) {
return isset( $this->mProperties[$name] ) ? $this->mProperties[$name] : false;
}
@@ -337,26 +440,25 @@ class ParserOutput extends CacheTime {
return $this->mProperties;
}
-
/**
* Returns the options from its ParserOptions which have been taken
* into account to produce this output or false if not available.
* @return mixed Array
*/
- public function getUsedOptions() {
+ public function getUsedOptions() {
if ( !isset( $this->mAccessedOptions ) ) {
return array();
}
return array_keys( $this->mAccessedOptions );
- }
+ }
- /**
- * Callback passed by the Parser to the ParserOptions to keep track of which options are used.
- * @access private
- */
- function recordOption( $option ) {
- $this->mAccessedOptions[$option] = true;
- }
+ /**
+ * Callback passed by the Parser to the ParserOptions to keep track of which options are used.
+ * @access private
+ */
+ function recordOption( $option ) {
+ $this->mAccessedOptions[$option] = true;
+ }
/**
* Adds an update job to the output. Any update jobs added to the output will eventually bexecuted in order to
@@ -375,9 +477,13 @@ class ParserOutput extends CacheTime {
* extracted from the page's content, including a LinksUpdate object for all links stored in
* this ParserOutput object.
*
+ * @note: Avoid using this method directly, use ContentHandler::getSecondaryDataUpdates() instead! The content
+ * handler may provide additional update objects.
+ *
* @since 1.20
*
- * @param $title Title of the page we're updating. If not given, a title object will be created based on $this->getTitleText()
+ * @param $title Title The title of the page we're updating. If not given, a title object will be created
+ * based on $this->getTitleText()
* @param $recursive Boolean: queue jobs for recursive updates?
*
* @return Array. An array of instances of DataUpdate
@@ -389,13 +495,138 @@ class ParserOutput extends CacheTime {
$linksUpdate = new LinksUpdate( $title, $this, $recursive );
- if ( $this->mSecondaryDataUpdates === array() ) {
- return array( $linksUpdate );
+ return array_merge( $this->mSecondaryDataUpdates, array( $linksUpdate ) );
+ }
+
+ /**
+ * Attaches arbitrary data to this ParserObject. This can be used to store some information in
+ * the ParserOutput object for later use during page output. The data will be cached along with
+ * the ParserOutput object, but unlike data set using setProperty(), it is not recorded in the
+ * database.
+ *
+ * This method is provided to overcome the unsafe practice of attaching extra information to a
+ * ParserObject by directly assigning member variables.
+ *
+ * To use setExtensionData() to pass extension information from a hook inside the parser to a
+ * hook in the page output, use this in the parser hook:
+ *
+ * @par Example:
+ * @code
+ * $parser->getOutput()->setExtensionData( 'my_ext_foo', '...' );
+ * @endcode
+ *
+ * And then later, in OutputPageParserOutput or similar:
+ *
+ * @par Example:
+ * @code
+ * $output->getExtensionData( 'my_ext_foo' );
+ * @endcode
+ *
+ * In MediaWiki 1.20 and older, you have to use a custom member variable
+ * within the ParserOutput object:
+ *
+ * @par Example:
+ * @code
+ * $parser->getOutput()->my_ext_foo = '...';
+ * @endcode
+ *
+ * @since 1.21
+ *
+ * @param string $key The key for accessing the data. Extensions should take care to avoid
+ * conflicts in naming keys. It is suggested to use the extension's name as a
+ * prefix.
+ *
+ * @param mixed $value The value to set. Setting a value to null is equivalent to removing
+ * the value.
+ */
+ public function setExtensionData( $key, $value ) {
+ if ( $value === null ) {
+ unset( $this->mExtensionData[$key] );
} else {
- $updates = array_merge( $this->mSecondaryDataUpdates, array( $linksUpdate ) );
+ $this->mExtensionData[$key] = $value;
+ }
+ }
+
+ /**
+ * Gets extensions data previously attached to this ParserOutput using setExtensionData().
+ * Typically, such data would be set while parsing the page, e.g. by a parser function.
+ *
+ * @since 1.21
+ *
+ * @param string $key The key to look up.
+ *
+ * @return mixed The value previously set for the given key using setExtensionData( $key ),
+ * or null if no value was set for this key.
+ */
+ public function getExtensionData( $key ) {
+ if ( isset( $this->mExtensionData[$key] ) ) {
+ return $this->mExtensionData[$key];
}
- return $updates;
- }
+ return null;
+ }
+ private static function getTimes( $clock = null ) {
+ $ret = array();
+ if ( !$clock || $clock === 'wall' ) {
+ $ret['wall'] = microtime( true );
+ }
+ if ( ( !$clock || $clock === 'cpu' ) && function_exists( 'getrusage' ) ) {
+ $ru = getrusage();
+ $ret['cpu'] = $ru['ru_utime.tv_sec'] + $ru['ru_utime.tv_usec'] / 1e6;
+ $ret['cpu'] += $ru['ru_stime.tv_sec'] + $ru['ru_stime.tv_usec'] / 1e6;
+ }
+ return $ret;
+ }
+
+ /**
+ * Resets the parse start timestamps for future calls to getTimeSinceStart()
+ * @since 1.22
+ */
+ function resetParseStartTime() {
+ $this->mParseStartTime = self::getTimes();
+ }
+
+ /**
+ * Returns the time since resetParseStartTime() was last called
+ *
+ * Clocks available are:
+ * - wall: Wall clock time
+ * - cpu: CPU time (requires getrusage)
+ *
+ * @since 1.22
+ * @param string $clock
+ * @return float|null
+ */
+ function getTimeSinceStart( $clock ) {
+ if ( !isset( $this->mParseStartTime[$clock] ) ) {
+ return null;
+ }
+
+ $end = self::getTimes( $clock );
+ return $end[$clock] - $this->mParseStartTime[$clock];
+ }
+
+ /**
+ * Sets parser limit report data for a key
+ *
+ * The key is used as the prefix for various messages used for formatting:
+ * - $key: The label for the field in the limit report
+ * - $key-value-text: Message used to format the value in the "NewPP limit
+ * report" HTML comment. If missing, uses $key-format.
+ * - $key-value-html: Message used to format the value in the preview
+ * limit report table. If missing, uses $key-format.
+ * - $key-value: Message used to format the value. If missing, uses "$1".
+ *
+ * Note that all values are interpreted as wikitext, and so should be
+ * encoded with htmlspecialchars() as necessary, but should avoid complex
+ * HTML for sanity of display in the "NewPP limit report" comment.
+ *
+ * @since 1.22
+ * @param string $key Message key
+ * @param mixed $value Appropriate for Message::params()
+ */
+ function setLimitReportData( $key, $value ) {
+ $this->mLimitReportData[$key] = $value;
+ }
}
diff --git a/includes/parser/Parser_DiffTest.php b/includes/parser/Parser_DiffTest.php
index f25340fa..aeae234a 100644
--- a/includes/parser/Parser_DiffTest.php
+++ b/includes/parser/Parser_DiffTest.php
@@ -122,7 +122,7 @@ class Parser_DiffTest
function setFunctionHook( $id, $callback, $flags = 0 ) {
$this->init();
- foreach ( $this->parsers as $parser ) {
+ foreach ( $this->parsers as $parser ) {
$parser->setFunctionHook( $id, $callback, $flags );
}
}
diff --git a/includes/parser/Parser_LinkHooks.php b/includes/parser/Parser_LinkHooks.php
deleted file mode 100644
index 6bcc324d..00000000
--- a/includes/parser/Parser_LinkHooks.php
+++ /dev/null
@@ -1,324 +0,0 @@
-<?php
-/**
- * Modified version of the PHP parser with hooks for wiki links; experimental
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
- * http://www.gnu.org/copyleft/gpl.html
- *
- * @file
- * @ingroup Parser
- */
-
-/**
- * Parser with LinkHooks experiment
- * @ingroup Parser
- */
-class Parser_LinkHooks extends Parser {
- /**
- * Update this version number when the ParserOutput format
- * changes in an incompatible way, so the parser cache
- * can automatically discard old data.
- */
- const VERSION = '1.6.4';
-
- # Flags for Parser::setLinkHook
- # Also available as global constants from Defines.php
- const SLH_PATTERN = 1;
-
- # Constants needed for external link processing
- # Everything except bracket, space, or control characters
- const EXT_LINK_URL_CLASS = '[^][<>"\\x00-\\x20\\x7F]';
- const EXT_IMAGE_REGEX = '/^(http:\/\/|https:\/\/)([^][<>"\\x00-\\x20\\x7F]+)
- \\/([A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]+)\\.((?i)gif|png|jpg|jpeg)$/Sx';
-
- /**#@+
- * @private
- */
- # Persistent:
- var $mLinkHooks;
-
- /**#@-*/
-
- /**
- * Constructor
- */
- public function __construct( $conf = array() ) {
- parent::__construct( $conf );
- $this->mLinkHooks = array();
- }
-
- /**
- * Do various kinds of initialisation on the first call of the parser
- */
- function firstCallInit() {
- parent::__construct();
- if ( !$this->mFirstCall ) {
- return;
- }
- $this->mFirstCall = false;
-
- wfProfileIn( __METHOD__ );
-
- $this->setHook( 'pre', array( $this, 'renderPreTag' ) );
- CoreParserFunctions::register( $this );
- CoreLinkFunctions::register( $this );
- $this->initialiseVariables();
-
- wfRunHooks( 'ParserFirstCallInit', array( &$this ) );
- wfProfileOut( __METHOD__ );
- }
-
- /**
- * Create a link hook, e.g. [[Namepsace:...|display}}
- * The callback function should have the form:
- * function myLinkCallback( $parser, $holders, $markers,
- * Title $title, $titleText, &$sortText = null, &$leadingColon = false ) { ... }
- *
- * Or with SLH_PATTERN:
- * function myLinkCallback( $parser, $holders, $markers, )
- * &$titleText, &$sortText = null, &$leadingColon = false ) { ... }
- *
- * The callback may either return a number of different possible values:
- * String) Text result of the link
- * True) (Treat as link) Parse the link according to normal link rules
- * False) (Bad link) Just output the raw wikitext (You may modify the text first)
- *
- * @param $ns Integer or String: the Namespace ID or regex pattern if SLH_PATTERN is set
- * @param $callback Mixed: the callback function (and object) to use
- * @param $flags Integer: a combination of the following flags:
- * SLH_PATTERN Use a regex link pattern rather than a namespace
- *
- * @return callback|null The old callback function for this name, if any
- */
- public function setLinkHook( $ns, $callback, $flags = 0 ) {
- if( $flags & SLH_PATTERN && !is_string($ns) )
- throw new MWException( __METHOD__.'() expecting a regex string pattern.' );
- elseif( $flags | ~SLH_PATTERN && !is_int($ns) )
- throw new MWException( __METHOD__.'() expecting a namespace index.' );
- $oldVal = isset( $this->mLinkHooks[$ns] ) ? $this->mLinkHooks[$ns][0] : null;
- $this->mLinkHooks[$ns] = array( $callback, $flags );
- return $oldVal;
- }
-
- /**
- * Get all registered link hook identifiers
- *
- * @return array
- */
- function getLinkHooks() {
- return array_keys( $this->mLinkHooks );
- }
-
- /**
- * Process [[ ]] wikilinks
- * @return LinkHolderArray
- *
- * @private
- */
- function replaceInternalLinks2( &$s ) {
- wfProfileIn( __METHOD__ );
-
- wfProfileIn( __METHOD__.'-setup' );
- static $tc = FALSE, $titleRegex;//$e1, $e1_img;
- if( !$tc ) {
- # the % is needed to support urlencoded titles as well
- $tc = Title::legalChars() . '#%';
- # Match a link having the form [[namespace:link|alternate]]trail
- //$e1 = "/^([{$tc}]+)(?:\\|(.+?))?]](.*)\$/sD";
- # Match cases where there is no "]]", which might still be images
- //$e1_img = "/^([{$tc}]+)\\|(.*)\$/sD";
- # Match a valid plain title
- $titleRegex = "/^([{$tc}]+)$/sD";
- }
-
- $holders = new LinkHolderArray( $this );
-
- if( is_null( $this->mTitle ) ) {
- wfProfileOut( __METHOD__ );
- wfProfileOut( __METHOD__.'-setup' );
- throw new MWException( __METHOD__.": \$this->mTitle is null\n" );
- }
-
- wfProfileOut( __METHOD__.'-setup' );
-
- $offset = 0;
- $offsetStack = array();
- $markers = new LinkMarkerReplacer( $this, $holders, array( &$this, 'replaceInternalLinksCallback' ) );
- while( true ) {
- $startBracketOffset = strpos( $s, '[[', $offset );
- $endBracketOffset = strpos( $s, ']]', $offset );
- # Finish when there are no more brackets
- if( $startBracketOffset === false && $endBracketOffset === false ) break;
- # Determine if the bracket is a starting or ending bracket
- # When we find both, use the first one
- elseif( $startBracketOffset !== false && $endBracketOffset !== false )
- $isStart = $startBracketOffset <= $endBracketOffset;
- # When we only found one, check which it is
- else $isStart = $startBracketOffset !== false;
- $bracketOffset = $isStart ? $startBracketOffset : $endBracketOffset;
- if( $isStart ) {
- /** Opening bracket **/
- # Just push our current offset in the string onto the stack
- $offsetStack[] = $startBracketOffset;
- } else {
- /** Closing bracket **/
- # Pop the start pos for our current link zone off the stack
- $startBracketOffset = array_pop($offsetStack);
- # Just to clean up the code, lets place offsets on the outer ends
- $endBracketOffset += 2;
-
- # Only do logic if we actually have a opening bracket for this
- if( isset($startBracketOffset) ) {
- # Extract text inside the link
- @list( $titleText, $paramText ) = explode('|',
- substr($s, $startBracketOffset+2, $endBracketOffset-$startBracketOffset-4), 2);
- # Create markers only for valid links
- if( preg_match( $titleRegex, $titleText ) ) {
- # Store the text for the marker
- $marker = $markers->addMarker($titleText, $paramText);
- # Replace the current link with the marker
- $s = substr($s,0,$startBracketOffset).
- $marker.
- substr($s, $endBracketOffset);
- # We have modified $s, because of this we need to set the
- # offset manually since the end position is different now
- $offset = $startBracketOffset+strlen($marker);
- continue;
- }
- # ToDo: Some LinkHooks may allow recursive links inside of
- # the link text, create a regex that also matches our
- # <!-- LINKMARKER ### --> sequence in titles
- # ToDO: Some LinkHooks use patterns rather than namespaces
- # these need to be tested at this point here
- }
-
- }
- # Bump our offset to after our current bracket
- $offset = $bracketOffset+2;
- }
-
-
- # Now expand our tree
- wfProfileIn( __METHOD__.'-expand' );
- $s = $markers->expand( $s );
- wfProfileOut( __METHOD__.'-expand' );
-
- wfProfileOut( __METHOD__ );
- return $holders;
- }
-
- function replaceInternalLinksCallback( $parser, $holders, $markers, $titleText, $paramText ) {
- wfProfileIn( __METHOD__ );
- $wt = isset($paramText) ? "[[$titleText|$paramText]]" : "[[$titleText]]";
- wfProfileIn( __METHOD__."-misc" );
- # Don't allow internal links to pages containing
- # PROTO: where PROTO is a valid URL protocol; these
- # should be external links.
- if( preg_match('/^\b(?i:' . wfUrlProtocols() . ')/', $titleText) ) {
- wfProfileOut( __METHOD__ );
- return $wt;
- }
-
- # Make subpage if necessary
- if( $this->areSubpagesAllowed() ) {
- $titleText = $this->maybeDoSubpageLink( $titleText, $paramText );
- }
-
- # Check for a leading colon and strip it if it is there
- $leadingColon = $titleText[0] == ':';
- if( $leadingColon ) $titleText = substr( $titleText, 1 );
-
- wfProfileOut( __METHOD__."-misc" );
- # Make title object
- wfProfileIn( __METHOD__."-title" );
- $title = Title::newFromText( $this->mStripState->unstripNoWiki( $titleText ) );
- if( !$title ) {
- wfProfileOut( __METHOD__."-title" );
- wfProfileOut( __METHOD__ );
- return $wt;
- }
- $ns = $title->getNamespace();
- wfProfileOut( __METHOD__."-title" );
-
- # Default for Namespaces is a default link
- # ToDo: Default for patterns is plain wikitext
- $return = true;
- if( isset( $this->mLinkHooks[$ns] ) ) {
- list( $callback, $flags ) = $this->mLinkHooks[$ns];
- if( $flags & SLH_PATTERN ) {
- $args = array( $parser, $holders, $markers, $titleText, &$paramText, &$leadingColon );
- } else {
- $args = array( $parser, $holders, $markers, $title, $titleText, &$paramText, &$leadingColon );
- }
- # Workaround for PHP bug 35229 and similar
- if ( !is_callable( $callback ) ) {
- throw new MWException( "Tag hook for namespace $ns is not callable\n" );
- }
- $return = call_user_func_array( $callback, $args );
- }
- if( $return === true ) {
- # True (treat as plain link) was returned, call the defaultLinkHook
- $return = CoreLinkFunctions::defaultLinkHook( $parser, $holders, $markers, $title,
- $titleText, $paramText, $leadingColon );
- }
- if( $return === false ) {
- # False (no link) was returned, output plain wikitext
- # Build it again as the hook is allowed to modify $paramText
- $return = isset($paramText) ? "[[$titleText|$paramText]]" : "[[$titleText]]";
- }
- # Content was returned, return it
- wfProfileOut( __METHOD__ );
- return $return;
- }
-
-}
-
-class LinkMarkerReplacer {
-
- protected $markers, $nextId, $parser, $holders, $callback;
-
- function __construct( $parser, $holders, $callback ) {
- $this->nextId = 0;
- $this->markers = array();
- $this->parser = $parser;
- $this->holders = $holders;
- $this->callback = $callback;
- }
-
- function addMarker($titleText, $paramText) {
- $id = $this->nextId++;
- $this->markers[$id] = array( $titleText, $paramText );
- return "<!-- LINKMARKER $id -->";
- }
-
- function findMarker( $string ) {
- return (bool) preg_match('/<!-- LINKMARKER [0-9]+ -->/', $string );
- }
-
- function expand( $string ) {
- return StringUtils::delimiterReplaceCallback( "<!-- LINKMARKER ", " -->", array( &$this, 'callback' ), $string );
- }
-
- function callback( $m ) {
- $id = intval($m[1]);
- if( !array_key_exists($id, $this->markers) ) return $m[0];
- $args = $this->markers[$id];
- array_unshift( $args, $this );
- array_unshift( $args, $this->holders );
- array_unshift( $args, $this->parser );
- return call_user_func_array( $this->callback, $args );
- }
-
-}
diff --git a/includes/parser/Preprocessor.php b/includes/parser/Preprocessor.php
index bd13f9ae..aeacd2e1 100644
--- a/includes/parser/Preprocessor.php
+++ b/includes/parser/Preprocessor.php
@@ -84,9 +84,9 @@ interface PPFrame {
/**
* Create a child frame
*
- * @param $args array
- * @param $title Title
- * @param $indexOffset A number subtracted from the index attributes of the arguments
+ * @param array $args
+ * @param Title $title
+ * @param int $indexOffset A number subtracted from the index attributes of the arguments
*
* @return PPFrame
*/
@@ -205,7 +205,6 @@ interface PPNode {
*/
function getChildrenOfType( $type );
-
/**
* Returns the length of the array, or false if this is not an array-type node
*/
diff --git a/includes/parser/Preprocessor_DOM.php b/includes/parser/Preprocessor_DOM.php
index 34de0ba5..3138f483 100644
--- a/includes/parser/Preprocessor_DOM.php
+++ b/includes/parser/Preprocessor_DOM.php
@@ -72,9 +72,8 @@ class Preprocessor_DOM implements Preprocessor {
$xml = "<list>";
foreach ( $values as $k => $val ) {
-
if ( is_int( $k ) ) {
- $xml .= "<part><name index=\"$k\"/><value>" . htmlspecialchars( $val ) ."</value></part>";
+ $xml .= "<part><name index=\"$k\"/><value>" . htmlspecialchars( $val ) . "</value></part>";
} else {
$xml .= "<part><name>" . htmlspecialchars( $k ) . "</name>=<value>" . htmlspecialchars( $val ) . "</value></part>";
}
@@ -110,7 +109,7 @@ class Preprocessor_DOM implements Preprocessor {
* Preprocess some wikitext and return the document tree.
* This is the ghost of Parser::replace_variables().
*
- * @param $text String: the text to parse
+ * @param string $text the text to parse
* @param $flags Integer: bitwise combination of:
* Parser::PTD_FOR_INCLUSION Handle "<noinclude>" and "<includeonly>" as if the text is being
* included. Default is to assume a direct page view.
@@ -126,6 +125,7 @@ class Preprocessor_DOM implements Preprocessor {
* cache may be implemented at a later date which takes further advantage of these strict
* dependency requirements.
*
+ * @throws MWException
* @return PPNode_DOM
*/
function preprocessToObj( $text, $flags = 0 ) {
@@ -136,9 +136,9 @@ class Preprocessor_DOM implements Preprocessor {
$cacheable = ( $wgPreprocessorCacheThreshold !== false
&& strlen( $text ) > $wgPreprocessorCacheThreshold );
if ( $cacheable ) {
- wfProfileIn( __METHOD__.'-cacheable' );
+ wfProfileIn( __METHOD__ . '-cacheable' );
- $cacheKey = wfMemcKey( 'preprocess-xml', md5($text), $flags );
+ $cacheKey = wfMemcKey( 'preprocess-xml', md5( $text ), $flags );
$cacheValue = $wgMemc->get( $cacheKey );
if ( $cacheValue ) {
$version = substr( $cacheValue, 0, 8 );
@@ -148,30 +148,32 @@ class Preprocessor_DOM implements Preprocessor {
wfDebugLog( "Preprocessor", "Loaded preprocessor XML from memcached (key $cacheKey)" );
}
}
- }
- if ( $xml === false ) {
- if ( $cacheable ) {
- wfProfileIn( __METHOD__.'-cache-miss' );
+ if ( $xml === false ) {
+ wfProfileIn( __METHOD__ . '-cache-miss' );
$xml = $this->preprocessToXml( $text, $flags );
$cacheValue = sprintf( "%08d", self::CACHE_VERSION ) . $xml;
$wgMemc->set( $cacheKey, $cacheValue, 86400 );
- wfProfileOut( __METHOD__.'-cache-miss' );
+ wfProfileOut( __METHOD__ . '-cache-miss' );
wfDebugLog( "Preprocessor", "Saved preprocessor XML to memcached (key $cacheKey)" );
- } else {
- $xml = $this->preprocessToXml( $text, $flags );
}
-
+ } else {
+ $xml = $this->preprocessToXml( $text, $flags );
}
+
// Fail if the number of elements exceeds acceptable limits
- // Do not attempt to generate the DOM
+ // Do not attempt to generate the DOM
$this->parser->mGeneratedPPNodeCount += substr_count( $xml, '<' );
$max = $this->parser->mOptions->getMaxGeneratedPPNodeCount();
if ( $this->parser->mGeneratedPPNodeCount > $max ) {
- throw new MWException( __METHOD__.': generated node count limit exceeded' );
+ if ( $cacheable ) {
+ wfProfileOut( __METHOD__ . '-cacheable' );
+ }
+ wfProfileOut( __METHOD__ );
+ throw new MWException( __METHOD__ . ': generated node count limit exceeded' );
}
- wfProfileIn( __METHOD__.'-loadXML' );
+ wfProfileIn( __METHOD__ . '-loadXML' );
$dom = new DOMDocument;
wfSuppressWarnings();
$result = $dom->loadXML( $xml );
@@ -181,16 +183,21 @@ class Preprocessor_DOM implements Preprocessor {
$xml = UtfNormal::cleanUp( $xml );
// 1 << 19 == XML_PARSE_HUGE, needed so newer versions of libxml2 don't barf when the XML is >256 levels deep
$result = $dom->loadXML( $xml, 1 << 19 );
- if ( !$result ) {
- throw new MWException( __METHOD__.' generated invalid XML' );
- }
}
- $obj = new PPNode_DOM( $dom->documentElement );
- wfProfileOut( __METHOD__.'-loadXML' );
+ if ( $result ) {
+ $obj = new PPNode_DOM( $dom->documentElement );
+ }
+ wfProfileOut( __METHOD__ . '-loadXML' );
+
if ( $cacheable ) {
- wfProfileOut( __METHOD__.'-cacheable' );
+ wfProfileOut( __METHOD__ . '-cacheable' );
}
+
wfProfileOut( __METHOD__ );
+
+ if ( !$result ) {
+ throw new MWException( __METHOD__ . ' generated invalid XML' );
+ }
return $obj;
}
@@ -354,9 +361,11 @@ class Preprocessor_DOM implements Preprocessor {
}
// Handle comments
if ( isset( $matches[2] ) && $matches[2] == '!--' ) {
- // To avoid leaving blank lines, when a comment is both preceded
- // and followed by a newline (ignoring spaces), trim leading and
- // trailing spaces and one of the newlines.
+
+ // To avoid leaving blank lines, when a sequence of
+ // space-separated comments is both preceded and followed by
+ // a newline (ignoring spaces), then
+ // trim leading and trailing spaces and the trailing newline.
// Find the end
$endPos = strpos( $text, '-->', $i + 4 );
@@ -367,10 +376,25 @@ class Preprocessor_DOM implements Preprocessor {
$i = $lengthText;
} else {
// Search backwards for leading whitespace
- $wsStart = $i ? ( $i - strspn( $revText, ' ', $lengthText - $i ) ) : 0;
+ $wsStart = $i ? ( $i - strspn( $revText, " \t", $lengthText - $i ) ) : 0;
+
// Search forwards for trailing whitespace
// $wsEnd will be the position of the last space (or the '>' if there's none)
- $wsEnd = $endPos + 2 + strspn( $text, ' ', $endPos + 3 );
+ $wsEnd = $endPos + 2 + strspn( $text, " \t", $endPos + 3 );
+
+ // Keep looking forward as long as we're finding more
+ // comments.
+ $comments = array( array( $wsStart, $wsEnd ) );
+ while ( substr( $text, $wsEnd + 1, 4 ) == '<!--' ) {
+ $c = strpos( $text, '-->', $wsEnd + 4 );
+ if ( $c === false ) {
+ break;
+ }
+ $c = $c + 2 + strspn( $text, " \t", $c + 3 );
+ $comments[] = array( $wsEnd + 1, $c );
+ $wsEnd = $c;
+ }
+
// Eat the line if possible
// TODO: This could theoretically be done if $wsStart == 0, i.e. for comments at
// the overall start. That's not how Sanitizer::removeHTMLcomments() did it, but
@@ -378,14 +402,26 @@ class Preprocessor_DOM implements Preprocessor {
if ( $wsStart > 0 && substr( $text, $wsStart - 1, 1 ) == "\n"
&& substr( $text, $wsEnd + 1, 1 ) == "\n" )
{
- $startPos = $wsStart;
- $endPos = $wsEnd + 1;
// Remove leading whitespace from the end of the accumulator
// Sanity check first though
$wsLength = $i - $wsStart;
- if ( $wsLength > 0 && substr( $accum, -$wsLength ) === str_repeat( ' ', $wsLength ) ) {
+ if ( $wsLength > 0
+ && strspn( $accum, " \t", -$wsLength ) === $wsLength )
+ {
$accum = substr( $accum, 0, -$wsLength );
}
+
+ // Dump all but the last comment to the accumulator
+ foreach ( $comments as $j => $com ) {
+ $startPos = $com[0];
+ $endPos = $com[1] + 1;
+ if ( $j == ( count( $comments ) - 1 ) ) {
+ break;
+ }
+ $inner = substr( $text, $startPos, $endPos - $startPos );
+ $accum .= '<comment>' . htmlspecialchars( $inner ) . '</comment>';
+ }
+
// Do a line-start run next time to look for headings after the comment
$fakeLineStart = true;
} else {
@@ -396,7 +432,7 @@ class Preprocessor_DOM implements Preprocessor {
if ( $stack->top ) {
$part = $stack->top->getCurrentPart();
- if ( ! (isset( $part->commentEnd ) && $part->commentEnd == $wsStart - 1 )) {
+ if ( !( isset( $part->commentEnd ) && $part->commentEnd == $wsStart - 1 ) ) {
$part->visualEnd = $wsStart;
}
// Else comments abutting, no change in visual end
@@ -431,7 +467,7 @@ class Preprocessor_DOM implements Preprocessor {
}
$tagStartPos = $i;
- if ( $text[$tagEndPos-1] == '/' ) {
+ if ( $text[$tagEndPos - 1] == '/' ) {
$attrEnd = $tagEndPos - 1;
$inner = null;
$i = $tagEndPos + 1;
@@ -521,7 +557,7 @@ class Preprocessor_DOM implements Preprocessor {
if ( $equalsLength > 0 ) {
if ( $searchStart - $equalsLength == $piece->startPos ) {
// This is just a single string of equals signs on its own line
- // Replicate the doHeadings behaviour /={count}(.+)={count}/
+ // Replicate the doHeadings behavior /={count}(.+)={count}/
// First find out how many equals signs there really are (don't stop at 6)
$count = $equalsLength;
if ( $count < 3 ) {
@@ -568,7 +604,7 @@ class Preprocessor_DOM implements Preprocessor {
'open' => $curChar,
'close' => $rule['end'],
'count' => $count,
- 'lineStart' => ($i > 0 && $text[$i-1] == "\n"),
+ 'lineStart' => ( $i > 0 && $text[$i - 1] == "\n" ),
);
$stack->push( $piece );
@@ -657,19 +693,13 @@ class Preprocessor_DOM implements Preprocessor {
$piece->parts = array( new PPDPart );
$piece->count -= $matchingCount;
# do we still qualify for any callback with remaining count?
- $names = $rules[$piece->open]['names'];
- $skippedBraces = 0;
- $enclosingAccum =& $accum;
- while ( $piece->count ) {
- if ( array_key_exists( $piece->count, $names ) ) {
- $stack->push( $piece );
- $accum =& $stack->getAccum();
- break;
- }
- --$piece->count;
- $skippedBraces ++;
+ $min = $rules[$piece->open]['min'];
+ if ( $piece->count >= $min ) {
+ $stack->push( $piece );
+ $accum =& $stack->getAccum();
+ } else {
+ $accum .= str_repeat( $piece->open, $piece->count );
}
- $enclosingAccum .= str_repeat( $piece->open, $skippedBraces );
}
$flags = $stack->getFlags();
extract( $flags );
@@ -751,18 +781,18 @@ class PPDStack {
$class = $this->elementClass;
$this->stack[] = new $class( $data );
}
- $this->top = $this->stack[ count( $this->stack ) - 1 ];
+ $this->top = $this->stack[count( $this->stack ) - 1];
$this->accum =& $this->top->getAccum();
}
function pop() {
if ( !count( $this->stack ) ) {
- throw new MWException( __METHOD__.': no elements remaining' );
+ throw new MWException( __METHOD__ . ': no elements remaining' );
}
$temp = array_pop( $this->stack );
if ( count( $this->stack ) ) {
- $this->top = $this->stack[ count( $this->stack ) - 1 ];
+ $this->top = $this->stack[count( $this->stack ) - 1];
$this->accum =& $this->top->getAccum();
} else {
$this->top = self::$false;
@@ -796,8 +826,8 @@ class PPDStack {
* @ingroup Parser
*/
class PPDStackElement {
- var $open, // Opening character (\n for heading)
- $close, // Matching closing character
+ var $open, // Opening character (\n for heading)
+ $close, // Matching closing character
$count, // Number of opening characters found (number of "=" for heading)
$parts, // Array of PPDPart objects describing pipe-separated parts.
$lineStart; // True if the open char appeared at the start of the input line. Not set for headings.
@@ -814,7 +844,7 @@ class PPDStackElement {
}
function &getAccum() {
- return $this->parts[count($this->parts) - 1]->out;
+ return $this->parts[count( $this->parts ) - 1]->out;
}
function addPart( $s = '' ) {
@@ -823,7 +853,7 @@ class PPDStackElement {
}
function getCurrentPart() {
- return $this->parts[count($this->parts) - 1];
+ return $this->parts[count( $this->parts ) - 1];
}
/**
@@ -916,7 +946,6 @@ class PPFrame_DOM implements PPFrame {
*/
var $depth;
-
/**
* Construct a new preprocessor frame.
* @param $preprocessor Preprocessor The parent preprocessor
@@ -1020,11 +1049,13 @@ class PPFrame_DOM implements PPFrame {
while ( count( $iteratorStack ) > 1 ) {
$level = count( $outStack ) - 1;
- $iteratorNode =& $iteratorStack[ $level ];
+ $iteratorNode =& $iteratorStack[$level];
$out =& $outStack[$level];
$index =& $indexStack[$level];
- if ( $iteratorNode instanceof PPNode_DOM ) $iteratorNode = $iteratorNode->node;
+ if ( $iteratorNode instanceof PPNode_DOM ) {
+ $iteratorNode = $iteratorNode->node;
+ }
if ( is_array( $iteratorNode ) ) {
if ( $index >= count( $iteratorNode ) ) {
@@ -1117,7 +1148,7 @@ class PPFrame_DOM implements PPFrame {
}
# Add a strip marker in PST mode so that pstPass2() can run some old-fashioned regexes on the result
# Not in RECOVER_COMMENTS mode (extractSections) though
- elseif ( $this->parser->ot['wiki'] && ! ( $flags & PPFrame::RECOVER_COMMENTS ) ) {
+ elseif ( $this->parser->ot['wiki'] && !( $flags & PPFrame::RECOVER_COMMENTS ) ) {
$out .= $this->parser->insertStripItem( $contextNode->textContent );
}
# Recover the literal comment in RECOVER_COMMENTS and pre+no-remove
@@ -1154,9 +1185,7 @@ class PPFrame_DOM implements PPFrame {
# Insert a heading marker only for <h> children of <root>
# This is to stop extractSections from going over multiple tree levels
- if ( $contextNode->parentNode->nodeName == 'root'
- && $this->parser->ot['html'] )
- {
+ if ( $contextNode->parentNode->nodeName == 'root' && $this->parser->ot['html'] ) {
# Insert heading index marker
$headingIndex = $contextNode->getAttribute( 'i' );
$titleText = $this->title->getPrefixedDBkey();
@@ -1174,7 +1203,7 @@ class PPFrame_DOM implements PPFrame {
}
} else {
wfProfileOut( __METHOD__ );
- throw new MWException( __METHOD__.': Invalid parameter type' );
+ throw new MWException( __METHOD__ . ': Invalid parameter type' );
}
if ( $newIterator !== false ) {
@@ -1212,7 +1241,9 @@ class PPFrame_DOM implements PPFrame {
$first = true;
$s = '';
foreach ( $args as $root ) {
- if ( $root instanceof PPNode_DOM ) $root = $root->node;
+ if ( $root instanceof PPNode_DOM ) {
+ $root = $root->node;
+ }
if ( !is_array( $root ) && !( $root instanceof DOMNodeList ) ) {
$root = array( $root );
}
@@ -1458,25 +1489,25 @@ class PPTemplateFrame_DOM extends PPFrame_DOM {
function getArguments() {
$arguments = array();
foreach ( array_merge(
- array_keys($this->numberedArgs),
- array_keys($this->namedArgs)) as $key ) {
- $arguments[$key] = $this->getArgument($key);
+ array_keys( $this->numberedArgs ),
+ array_keys( $this->namedArgs ) ) as $key ) {
+ $arguments[$key] = $this->getArgument( $key );
}
return $arguments;
}
function getNumberedArguments() {
$arguments = array();
- foreach ( array_keys($this->numberedArgs) as $key ) {
- $arguments[$key] = $this->getArgument($key);
+ foreach ( array_keys( $this->numberedArgs ) as $key ) {
+ $arguments[$key] = $this->getArgument( $key );
}
return $arguments;
}
function getNamedArguments() {
$arguments = array();
- foreach ( array_keys($this->namedArgs) as $key ) {
- $arguments[$key] = $this->getArgument($key);
+ foreach ( array_keys( $this->namedArgs ) as $key ) {
+ $arguments[$key] = $this->getArgument( $key );
}
return $arguments;
}
@@ -1673,6 +1704,7 @@ class PPNode_DOM implements PPNode {
* - index String index
* - value PPNode value
*
+ * @throws MWException
* @return array
*/
function splitArg() {
@@ -1694,6 +1726,7 @@ class PPNode_DOM implements PPNode {
* Split an "<ext>" node into an associative array containing name, attr, inner and close
* All values in the resulting array are PPNodes. Inner and close are optional.
*
+ * @throws MWException
* @return array
*/
function splitExt() {
@@ -1719,6 +1752,7 @@ class PPNode_DOM implements PPNode {
/**
* Split a "<h>" node
+ * @throws MWException
* @return array
*/
function splitHeading() {
diff --git a/includes/parser/Preprocessor_Hash.php b/includes/parser/Preprocessor_Hash.php
index 4f04c865..2fc5e118 100644
--- a/includes/parser/Preprocessor_Hash.php
+++ b/includes/parser/Preprocessor_Hash.php
@@ -89,7 +89,7 @@ class Preprocessor_Hash implements Preprocessor {
* Preprocess some wikitext and return the document tree.
* This is the ghost of Parser::replace_variables().
*
- * @param $text String: the text to parse
+ * @param string $text the text to parse
* @param $flags Integer: bitwise combination of:
* Parser::PTD_FOR_INCLUSION Handle "<noinclude>" and "<includeonly>" as if the text is being
* included. Default is to assume a direct page view.
@@ -105,6 +105,7 @@ class Preprocessor_Hash implements Preprocessor {
* cache may be implemented at a later date which takes further advantage of these strict
* dependency requirements.
*
+ * @throws MWException
* @return PPNode_Hash_Tree
*/
function preprocessToObj( $text, $flags = 0 ) {
@@ -115,9 +116,9 @@ class Preprocessor_Hash implements Preprocessor {
$cacheable = $wgPreprocessorCacheThreshold !== false && strlen( $text ) > $wgPreprocessorCacheThreshold;
if ( $cacheable ) {
- wfProfileIn( __METHOD__.'-cacheable' );
+ wfProfileIn( __METHOD__ . '-cacheable' );
- $cacheKey = wfMemcKey( 'preprocess-hash', md5($text), $flags );
+ $cacheKey = wfMemcKey( 'preprocess-hash', md5( $text ), $flags );
$cacheValue = $wgMemc->get( $cacheKey );
if ( $cacheValue ) {
$version = substr( $cacheValue, 0, 8 );
@@ -126,12 +127,12 @@ class Preprocessor_Hash implements Preprocessor {
// From the cache
wfDebugLog( "Preprocessor",
"Loaded preprocessor hash from memcached (key $cacheKey)" );
- wfProfileOut( __METHOD__.'-cacheable' );
+ wfProfileOut( __METHOD__ . '-cacheable' );
wfProfileOut( __METHOD__ );
return $hash;
}
}
- wfProfileIn( __METHOD__.'-cache-miss' );
+ wfProfileIn( __METHOD__ . '-cache-miss' );
}
$rules = array(
@@ -286,9 +287,11 @@ class Preprocessor_Hash implements Preprocessor {
}
// Handle comments
if ( isset( $matches[2] ) && $matches[2] == '!--' ) {
- // To avoid leaving blank lines, when a comment is both preceded
- // and followed by a newline (ignoring spaces), trim leading and
- // trailing spaces and one of the newlines.
+
+ // To avoid leaving blank lines, when a sequence of
+ // space-separated comments is both preceded and followed by
+ // a newline (ignoring spaces), then
+ // trim leading and trailing spaces and the trailing newline.
// Find the end
$endPos = strpos( $text, '-->', $i + 4 );
@@ -299,10 +302,25 @@ class Preprocessor_Hash implements Preprocessor {
$i = $lengthText;
} else {
// Search backwards for leading whitespace
- $wsStart = $i ? ( $i - strspn( $revText, ' ', $lengthText - $i ) ) : 0;
+ $wsStart = $i ? ( $i - strspn( $revText, " \t", $lengthText - $i ) ) : 0;
+
// Search forwards for trailing whitespace
// $wsEnd will be the position of the last space (or the '>' if there's none)
- $wsEnd = $endPos + 2 + strspn( $text, ' ', $endPos + 3 );
+ $wsEnd = $endPos + 2 + strspn( $text, " \t", $endPos + 3 );
+
+ // Keep looking forward as long as we're finding more
+ // comments.
+ $comments = array( array( $wsStart, $wsEnd ) );
+ while ( substr( $text, $wsEnd + 1, 4 ) == '<!--' ) {
+ $c = strpos( $text, '-->', $wsEnd + 4 );
+ if ( $c === false ) {
+ break;
+ }
+ $c = $c + 2 + strspn( $text, " \t", $c + 3 );
+ $comments[] = array( $wsEnd + 1, $c );
+ $wsEnd = $c;
+ }
+
// Eat the line if possible
// TODO: This could theoretically be done if $wsStart == 0, i.e. for comments at
// the overall start. That's not how Sanitizer::removeHTMLcomments() did it, but
@@ -310,17 +328,27 @@ class Preprocessor_Hash implements Preprocessor {
if ( $wsStart > 0 && substr( $text, $wsStart - 1, 1 ) == "\n"
&& substr( $text, $wsEnd + 1, 1 ) == "\n" )
{
- $startPos = $wsStart;
- $endPos = $wsEnd + 1;
// Remove leading whitespace from the end of the accumulator
// Sanity check first though
$wsLength = $i - $wsStart;
if ( $wsLength > 0
&& $accum->lastNode instanceof PPNode_Hash_Text
- && substr( $accum->lastNode->value, -$wsLength ) === str_repeat( ' ', $wsLength ) )
+ && strspn( $accum->lastNode->value, " \t", -$wsLength ) === $wsLength )
{
$accum->lastNode->value = substr( $accum->lastNode->value, 0, -$wsLength );
}
+
+ // Dump all but the last comment to the accumulator
+ foreach ( $comments as $j => $com ) {
+ $startPos = $com[0];
+ $endPos = $com[1] + 1;
+ if ( $j == ( count( $comments ) - 1 ) ) {
+ break;
+ }
+ $inner = substr( $text, $startPos, $endPos - $startPos );
+ $accum->addNodeWithText( 'comment', $inner );
+ }
+
// Do a line-start run next time to look for headings after the comment
$fakeLineStart = true;
} else {
@@ -331,7 +359,7 @@ class Preprocessor_Hash implements Preprocessor {
if ( $stack->top ) {
$part = $stack->top->getCurrentPart();
- if ( ! (isset( $part->commentEnd ) && $part->commentEnd == $wsStart - 1 )) {
+ if ( !( isset( $part->commentEnd ) && $part->commentEnd == $wsStart - 1 ) ) {
$part->visualEnd = $wsStart;
}
// Else comments abutting, no change in visual end
@@ -366,7 +394,7 @@ class Preprocessor_Hash implements Preprocessor {
}
$tagStartPos = $i;
- if ( $text[$tagEndPos-1] == '/' ) {
+ if ( $text[$tagEndPos - 1] == '/' ) {
// Short end tag
$attrEnd = $tagEndPos - 1;
$inner = null;
@@ -390,7 +418,7 @@ class Preprocessor_Hash implements Preprocessor {
}
// <includeonly> and <noinclude> just become <ignore> tags
if ( in_array( $lowerName, $ignoredElements ) ) {
- $accum->addNodeWithText( 'ignore', substr( $text, $tagStartPos, $i - $tagStartPos ) );
+ $accum->addNodeWithText( 'ignore', substr( $text, $tagStartPos, $i - $tagStartPos ) );
continue;
}
@@ -461,7 +489,7 @@ class Preprocessor_Hash implements Preprocessor {
if ( $equalsLength > 0 ) {
if ( $searchStart - $equalsLength == $piece->startPos ) {
// This is just a single string of equals signs on its own line
- // Replicate the doHeadings behaviour /={count}(.+)={count}/
+ // Replicate the doHeadings behavior /={count}(.+)={count}/
// First find out how many equals signs there really are (don't stop at 6)
$count = $equalsLength;
if ( $count < 3 ) {
@@ -514,7 +542,7 @@ class Preprocessor_Hash implements Preprocessor {
'open' => $curChar,
'close' => $rule['end'],
'count' => $count,
- 'lineStart' => ($i > 0 && $text[$i-1] == "\n"),
+ 'lineStart' => ( $i > 0 && $text[$i - 1] == "\n" ),
);
$stack->push( $piece );
@@ -548,7 +576,7 @@ class Preprocessor_Hash implements Preprocessor {
}
}
- if ($matchingCount <= 0) {
+ if ( $matchingCount <= 0 ) {
# No matching element found in callback array
# Output a literal closing brace and continue
$accum->addLiteral( str_repeat( $curChar, $count ) );
@@ -590,10 +618,20 @@ class Preprocessor_Hash implements Preprocessor {
$lastNode = $node;
}
if ( !$node ) {
- throw new MWException( __METHOD__. ': eqpos not found' );
+ if ( $cacheable ) {
+ wfProfileOut( __METHOD__ . '-cache-miss' );
+ wfProfileOut( __METHOD__ . '-cacheable' );
+ }
+ wfProfileOut( __METHOD__ );
+ throw new MWException( __METHOD__ . ': eqpos not found' );
}
if ( $node->name !== 'equals' ) {
- throw new MWException( __METHOD__ .': eqpos is not equals' );
+ if ( $cacheable ) {
+ wfProfileOut( __METHOD__ . '-cache-miss' );
+ wfProfileOut( __METHOD__ . '-cacheable' );
+ }
+ wfProfileOut( __METHOD__ );
+ throw new MWException( __METHOD__ . ': eqpos is not equals' );
}
$equalsNode = $node;
@@ -638,23 +676,17 @@ class Preprocessor_Hash implements Preprocessor {
$accum =& $stack->getAccum();
# Re-add the old stack element if it still has unmatched opening characters remaining
- if ($matchingCount < $piece->count) {
+ if ( $matchingCount < $piece->count ) {
$piece->parts = array( new PPDPart_Hash );
$piece->count -= $matchingCount;
# do we still qualify for any callback with remaining count?
- $names = $rules[$piece->open]['names'];
- $skippedBraces = 0;
- $enclosingAccum =& $accum;
- while ( $piece->count ) {
- if ( array_key_exists( $piece->count, $names ) ) {
- $stack->push( $piece );
- $accum =& $stack->getAccum();
- break;
- }
- --$piece->count;
- $skippedBraces ++;
+ $min = $rules[$piece->open]['min'];
+ if ( $piece->count >= $min ) {
+ $stack->push( $piece );
+ $accum =& $stack->getAccum();
+ } else {
+ $accum->addLiteral( str_repeat( $piece->open, $piece->count ) );
}
- $enclosingAccum->addLiteral( str_repeat( $piece->open, $skippedBraces ) );
}
extract( $stack->getFlags() );
@@ -695,11 +727,11 @@ class Preprocessor_Hash implements Preprocessor {
$rootNode->lastChild = $stack->rootAccum->lastNode;
// Cache
- if ($cacheable) {
+ if ( $cacheable ) {
$cacheValue = sprintf( "%08d", self::CACHE_VERSION ) . serialize( $rootNode );
$wgMemc->set( $cacheKey, $cacheValue, 86400 );
- wfProfileOut( __METHOD__.'-cache-miss' );
- wfProfileOut( __METHOD__.'-cacheable' );
+ wfProfileOut( __METHOD__ . '-cache-miss' );
+ wfProfileOut( __METHOD__ . '-cacheable' );
wfDebugLog( "Preprocessor", "Saved preprocessor Hash to memcached (key $cacheKey)" );
}
@@ -866,7 +898,6 @@ class PPFrame_Hash implements PPFrame {
*/
var $depth;
-
/**
* Construct a new preprocessor frame.
* @param $preprocessor Preprocessor: the parent preprocessor
@@ -884,9 +915,11 @@ class PPFrame_Hash implements PPFrame {
* Create a new child frame
* $args is optionally a multi-root PPNode or array containing the template arguments
*
- * @param $args PPNode_Hash_Array|array
+ * @param array|bool|\PPNode_Hash_Array $args PPNode_Hash_Array|array
* @param $title Title|bool
*
+ * @param int $indexOffset
+ * @throws MWException
* @return PPTemplateFrame_Hash
*/
function newChild( $args = false, $title = false, $indexOffset = 0 ) {
@@ -956,7 +989,7 @@ class PPFrame_Hash implements PPFrame {
while ( count( $iteratorStack ) > 1 ) {
$level = count( $outStack ) - 1;
- $iteratorNode =& $iteratorStack[ $level ];
+ $iteratorNode =& $iteratorStack[$level];
$out =& $outStack[$level];
$index =& $indexStack[$level];
@@ -1035,7 +1068,7 @@ class PPFrame_Hash implements PPFrame {
}
# Add a strip marker in PST mode so that pstPass2() can run some old-fashioned regexes on the result
# Not in RECOVER_COMMENTS mode (extractSections) though
- elseif ( $this->parser->ot['wiki'] && ! ( $flags & PPFrame::RECOVER_COMMENTS ) ) {
+ elseif ( $this->parser->ot['wiki'] && !( $flags & PPFrame::RECOVER_COMMENTS ) ) {
$out .= $this->parser->insertStripItem( $contextNode->firstChild->value );
}
# Recover the literal comment in RECOVER_COMMENTS and pre+no-remove
@@ -1082,7 +1115,7 @@ class PPFrame_Hash implements PPFrame {
$newIterator = $contextNode->getChildren();
}
} else {
- throw new MWException( __METHOD__.': Invalid parameter type' );
+ throw new MWException( __METHOD__ . ': Invalid parameter type' );
}
if ( $newIterator !== false ) {
@@ -1371,9 +1404,9 @@ class PPTemplateFrame_Hash extends PPFrame_Hash {
function getArguments() {
$arguments = array();
foreach ( array_merge(
- array_keys($this->numberedArgs),
- array_keys($this->namedArgs)) as $key ) {
- $arguments[$key] = $this->getArgument($key);
+ array_keys( $this->numberedArgs ),
+ array_keys( $this->namedArgs ) ) as $key ) {
+ $arguments[$key] = $this->getArgument( $key );
}
return $arguments;
}
@@ -1383,8 +1416,8 @@ class PPTemplateFrame_Hash extends PPFrame_Hash {
*/
function getNumberedArguments() {
$arguments = array();
- foreach ( array_keys($this->numberedArgs) as $key ) {
- $arguments[$key] = $this->getArgument($key);
+ foreach ( array_keys( $this->numberedArgs ) as $key ) {
+ $arguments[$key] = $this->getArgument( $key );
}
return $arguments;
}
@@ -1394,8 +1427,8 @@ class PPTemplateFrame_Hash extends PPFrame_Hash {
*/
function getNamedArguments() {
$arguments = array();
- foreach ( array_keys($this->namedArgs) as $key ) {
- $arguments[$key] = $this->getArgument($key);
+ foreach ( array_keys( $this->namedArgs ) as $key ) {
+ $arguments[$key] = $this->getArgument( $key );
}
return $arguments;
}
@@ -1609,6 +1642,7 @@ class PPNode_Hash_Tree implements PPNode {
* - index String index
* - value PPNode value
*
+ * @throws MWException
* @return array
*/
function splitArg() {
@@ -1642,6 +1676,7 @@ class PPNode_Hash_Tree implements PPNode {
* Split an "<ext>" node into an associative array containing name, attr, inner and close
* All values in the resulting array are PPNodes. Inner and close are optional.
*
+ * @throws MWException
* @return array
*/
function splitExt() {
@@ -1669,6 +1704,7 @@ class PPNode_Hash_Tree implements PPNode {
/**
* Split an "<h>" node
*
+ * @throws MWException
* @return array
*/
function splitHeading() {
@@ -1695,6 +1731,7 @@ class PPNode_Hash_Tree implements PPNode {
/**
* Split a "<template>" or "<tplarg>" node
*
+ * @throws MWException
* @return array
*/
function splitTemplate() {
diff --git a/includes/parser/Preprocessor_HipHop.hphp b/includes/parser/Preprocessor_HipHop.hphp
deleted file mode 100644
index 8b71a1b5..00000000
--- a/includes/parser/Preprocessor_HipHop.hphp
+++ /dev/null
@@ -1,2013 +0,0 @@
-<?php
-/**
- * A preprocessor optimised for HipHop, using HipHop-specific syntax.
- * vim: ft=php
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
- * http://www.gnu.org/copyleft/gpl.html
- *
- * @file
- * @ingroup Parser
- */
-
-/**
- * @ingroup Parser
- */
-class Preprocessor_HipHop implements Preprocessor {
- /**
- * @var Parser
- */
- var $parser;
-
- const CACHE_VERSION = 1;
-
- /**
- * @param $parser Parser
- */
- function __construct( $parser ) {
- $this->parser = $parser;
- }
-
- /**
- * @return PPFrame_HipHop
- */
- function newFrame() {
- return new PPFrame_HipHop( $this );
- }
-
- /**
- * @param $args array
- * @return PPCustomFrame_HipHop
- */
- function newCustomFrame( $args ) {
- return new PPCustomFrame_HipHop( $this, $args );
- }
-
- /**
- * @param $values array
- * @return PPNode_HipHop_Array
- */
- function newPartNodeArray( $values ) {
- $list = array();
-
- foreach ( $values as $k => $val ) {
- $partNode = new PPNode_HipHop_Tree( 'part' );
- $nameNode = new PPNode_HipHop_Tree( 'name' );
-
- if ( is_int( $k ) ) {
- $nameNode->addChild( new PPNode_HipHop_Attr( 'index', $k ) );
- $partNode->addChild( $nameNode );
- } else {
- $nameNode->addChild( new PPNode_HipHop_Text( $k ) );
- $partNode->addChild( $nameNode );
- $partNode->addChild( new PPNode_HipHop_Text( '=' ) );
- }
-
- $valueNode = new PPNode_HipHop_Tree( 'value' );
- $valueNode->addChild( new PPNode_HipHop_Text( $val ) );
- $partNode->addChild( $valueNode );
-
- $list[] = $partNode;
- }
-
- $node = new PPNode_HipHop_Array( $list );
- return $node;
- }
-
- /**
- * Preprocess some wikitext and return the document tree.
- * This is the ghost of Parser::replace_variables().
- *
- * @param $text String: the text to parse
- * @param $flags Integer: bitwise combination of:
- * Parser::PTD_FOR_INCLUSION Handle <noinclude>/<includeonly> as if the text is being
- * included. Default is to assume a direct page view.
- *
- * The generated DOM tree must depend only on the input text and the flags.
- * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899.
- *
- * Any flag added to the $flags parameter here, or any other parameter liable to cause a
- * change in the DOM tree for a given text, must be passed through the section identifier
- * in the section edit link and thus back to extractSections().
- *
- * The output of this function is currently only cached in process memory, but a persistent
- * cache may be implemented at a later date which takes further advantage of these strict
- * dependency requirements.
- *
- * @throws MWException
- * @return PPNode_HipHop_Tree
- */
- function preprocessToObj( $text, $flags = 0 ) {
- wfProfileIn( __METHOD__ );
-
- // Check cache.
- global $wgMemc, $wgPreprocessorCacheThreshold;
-
- $lengthText = strlen( $text );
-
- $cacheable = ($wgPreprocessorCacheThreshold !== false && $lengthText > $wgPreprocessorCacheThreshold);
- if ( $cacheable ) {
- wfProfileIn( __METHOD__.'-cacheable' );
-
- $cacheKey = strval( wfMemcKey( 'preprocess-hash', md5($text), $flags ) );
- $cacheValue = strval( $wgMemc->get( $cacheKey ) );
- if ( $cacheValue !== '' ) {
- $version = substr( $cacheValue, 0, 8 );
- if ( intval( $version ) == self::CACHE_VERSION ) {
- $hash = unserialize( substr( $cacheValue, 8 ) );
- // From the cache
- wfDebugLog( "Preprocessor",
- "Loaded preprocessor hash from memcached (key $cacheKey)" );
- wfProfileOut( __METHOD__.'-cacheable' );
- wfProfileOut( __METHOD__ );
- return $hash;
- }
- }
- wfProfileIn( __METHOD__.'-cache-miss' );
- }
-
- $rules = array(
- '{' => array(
- 'end' => '}',
- 'names' => array(
- 2 => 'template',
- 3 => 'tplarg',
- ),
- 'min' => 2,
- 'max' => 3,
- ),
- '[' => array(
- 'end' => ']',
- 'names' => array( 2 => 'LITERAL' ),
- 'min' => 2,
- 'max' => 2,
- )
- );
-
- $forInclusion = (bool)( $flags & Parser::PTD_FOR_INCLUSION );
-
- $xmlishElements = (array)$this->parser->getStripList();
- $enableOnlyinclude = false;
- if ( $forInclusion ) {
- $ignoredTags = array( 'includeonly', '/includeonly' );
- $ignoredElements = array( 'noinclude' );
- $xmlishElements[] = 'noinclude';
- if ( strpos( $text, '<onlyinclude>' ) !== false && strpos( $text, '</onlyinclude>' ) !== false ) {
- $enableOnlyinclude = true;
- }
- } else if ( $this->parser->ot['wiki'] ) {
- $ignoredTags = array( 'noinclude', '/noinclude', 'onlyinclude', '/onlyinclude', 'includeonly', '/includeonly' );
- $ignoredElements = array();
- } else {
- $ignoredTags = array( 'noinclude', '/noinclude', 'onlyinclude', '/onlyinclude' );
- $ignoredElements = array( 'includeonly' );
- $xmlishElements[] = 'includeonly';
- }
- $xmlishRegex = implode( '|', array_merge( $xmlishElements, $ignoredTags ) );
-
- // Use "A" modifier (anchored) instead of "^", because ^ doesn't work with an offset
- $elementsRegex = "~($xmlishRegex)(?:\s|\/>|>)|(!--)~iA";
-
- $stack = new PPDStack_HipHop;
-
- $searchBase = "[{<\n";
- $revText = strrev( $text ); // For fast reverse searches
-
- $i = 0; # Input pointer, starts out pointing to a pseudo-newline before the start
- $accum = $stack->getAccum(); # Current accumulator
- $headingIndex = 1;
- $stackFlags = array(
- 'findPipe' => false, # True to take notice of pipe characters
- 'findEquals' => false, # True to find equals signs in arguments
- 'inHeading' => false, # True if $i is inside a possible heading
- );
- $noMoreGT = false; # True if there are no more greater-than (>) signs right of $i
- $findOnlyinclude = $enableOnlyinclude; # True to ignore all input up to the next <onlyinclude>
- $fakeLineStart = true; # Do a line-start run without outputting an LF character
-
- while ( true ) {
- //$this->memCheck();
-
- if ( $findOnlyinclude ) {
- // Ignore all input up to the next <onlyinclude>
- $variantStartPos = strpos( $text, '<onlyinclude>', $i );
- if ( $variantStartPos === false ) {
- // Ignored section runs to the end
- $accum->addNodeWithText( 'ignore', strval( substr( $text, $i ) ) );
- break;
- }
- $startPos1 = intval( $variantStartPos );
- $tagEndPos = $startPos1 + strlen( '<onlyinclude>' ); // past-the-end
- $accum->addNodeWithText( 'ignore', strval( substr( $text, $i, $tagEndPos - $i ) ) );
- $i = $tagEndPos;
- $findOnlyinclude = false;
- }
-
- if ( $fakeLineStart ) {
- $found = 'line-start';
- $curChar = '';
- } else {
- # Find next opening brace, closing brace or pipe
- $search = $searchBase;
- if ( $stack->top === false ) {
- $currentClosing = '';
- } else {
- $currentClosing = strval( $stack->getTop()->close );
- $search .= $currentClosing;
- }
- if ( $stackFlags['findPipe'] ) {
- $search .= '|';
- }
- if ( $stackFlags['findEquals'] ) {
- // First equals will be for the template
- $search .= '=';
- }
- $rule = null;
- # Output literal section, advance input counter
- $literalLength = intval( strcspn( $text, $search, $i ) );
- if ( $literalLength > 0 ) {
- $accum->addLiteral( strval( substr( $text, $i, $literalLength ) ) );
- $i += $literalLength;
- }
- if ( $i >= $lengthText ) {
- if ( $currentClosing === "\n" ) {
- // Do a past-the-end run to finish off the heading
- $curChar = '';
- $found = 'line-end';
- } else {
- # All done
- break;
- }
- } else {
- $curChar = $text[$i];
- if ( $curChar === '|' ) {
- $found = 'pipe';
- } elseif ( $curChar === '=' ) {
- $found = 'equals';
- } elseif ( $curChar === '<' ) {
- $found = 'angle';
- } elseif ( $curChar === "\n" ) {
- if ( $stackFlags['inHeading'] ) {
- $found = 'line-end';
- } else {
- $found = 'line-start';
- }
- } elseif ( $curChar === $currentClosing ) {
- $found = 'close';
- } elseif ( isset( $rules[$curChar] ) ) {
- $found = 'open';
- $rule = $rules[$curChar];
- } else {
- # Some versions of PHP have a strcspn which stops on null characters
- # Ignore and continue
- ++$i;
- continue;
- }
- }
- }
-
- if ( $found === 'angle' ) {
- $matches = false;
- // Handle </onlyinclude>
- if ( $enableOnlyinclude
- && substr( $text, $i, strlen( '</onlyinclude>' ) ) === '</onlyinclude>' )
- {
- $findOnlyinclude = true;
- continue;
- }
-
- // Determine element name
- if ( !preg_match( $elementsRegex, $text, $matches, 0, $i + 1 ) ) {
- // Element name missing or not listed
- $accum->addLiteral( '<' );
- ++$i;
- continue;
- }
- // Handle comments
- if ( isset( $matches[2] ) && $matches[2] === '!--' ) {
- // To avoid leaving blank lines, when a comment is both preceded
- // and followed by a newline (ignoring spaces), trim leading and
- // trailing spaces and one of the newlines.
-
- // Find the end
- $variantEndPos = strpos( $text, '-->', $i + 4 );
- if ( $variantEndPos === false ) {
- // Unclosed comment in input, runs to end
- $inner = strval( substr( $text, $i ) );
- $accum->addNodeWithText( 'comment', $inner );
- $i = $lengthText;
- } else {
- $endPos = intval( $variantEndPos );
- // Search backwards for leading whitespace
- if ( $i ) {
- $wsStart = $i - intval( strspn( $revText, ' ', $lengthText - $i ) );
- } else {
- $wsStart = 0;
- }
- // Search forwards for trailing whitespace
- // $wsEnd will be the position of the last space (or the '>' if there's none)
- $wsEnd = $endPos + 2 + intval( strspn( $text, ' ', $endPos + 3 ) );
- // Eat the line if possible
- // TODO: This could theoretically be done if $wsStart == 0, i.e. for comments at
- // the overall start. That's not how Sanitizer::removeHTMLcomments() did it, but
- // it's a possible beneficial b/c break.
- if ( $wsStart > 0 && substr( $text, $wsStart - 1, 1 ) === "\n"
- && substr( $text, $wsEnd + 1, 1 ) === "\n" )
- {
- $startPos2 = $wsStart;
- $endPos = $wsEnd + 1;
- // Remove leading whitespace from the end of the accumulator
- // Sanity check first though
- $wsLength = $i - $wsStart;
- if ( $wsLength > 0
- && $accum->lastNode instanceof PPNode_HipHop_Text
- && substr( $accum->lastNode->value, -$wsLength ) === str_repeat( ' ', $wsLength ) )
- {
- $accum->lastNode->value = strval( substr( $accum->lastNode->value, 0, -$wsLength ) );
- }
- // Do a line-start run next time to look for headings after the comment
- $fakeLineStart = true;
- } else {
- // No line to eat, just take the comment itself
- $startPos2 = $i;
- $endPos += 2;
- }
-
- if ( $stack->top ) {
- $part = $stack->getTop()->getCurrentPart();
- if ( ! (isset( $part->commentEnd ) && $part->commentEnd == $wsStart - 1 )) {
- $part->visualEnd = $wsStart;
- }
- // Else comments abutting, no change in visual end
- $part->commentEnd = $endPos;
- }
- $i = $endPos + 1;
- $inner = strval( substr( $text, $startPos2, $endPos - $startPos2 + 1 ) );
- $accum->addNodeWithText( 'comment', $inner );
- }
- continue;
- }
- $name = strval( $matches[1] );
- $lowerName = strtolower( $name );
- $attrStart = $i + strlen( $name ) + 1;
-
- // Find end of tag
- $variantTagEndPos = $noMoreGT ? false : strpos( $text, '>', $attrStart );
- if ( $variantTagEndPos === false ) {
- // Infinite backtrack
- // Disable tag search to prevent worst-case O(N^2) performance
- $noMoreGT = true;
- $accum->addLiteral( '<' );
- ++$i;
- continue;
- }
- $tagEndPos = intval( $variantTagEndPos );
-
- // Handle ignored tags
- if ( in_array( $lowerName, $ignoredTags ) ) {
- $accum->addNodeWithText( 'ignore', strval( substr( $text, $i, $tagEndPos - $i + 1 ) ) );
- $i = $tagEndPos + 1;
- continue;
- }
-
- $tagStartPos = $i;
- $close = '';
- if ( $text[$tagEndPos-1] === '/' ) {
- // Short end tag
- $attrEnd = $tagEndPos - 1;
- $shortEnd = true;
- $inner = '';
- $i = $tagEndPos + 1;
- $haveClose = false;
- } else {
- $attrEnd = $tagEndPos;
- $shortEnd = false;
- // Find closing tag
- if ( preg_match( "/<\/" . preg_quote( $name, '/' ) . "\s*>/i",
- $text, $matches, PREG_OFFSET_CAPTURE, $tagEndPos + 1 ) )
- {
- $inner = strval( substr( $text, $tagEndPos + 1, $matches[0][1] - $tagEndPos - 1 ) );
- $i = intval( $matches[0][1] ) + strlen( $matches[0][0] );
- $close = strval( $matches[0][0] );
- $haveClose = true;
- } else {
- // No end tag -- let it run out to the end of the text.
- $inner = strval( substr( $text, $tagEndPos + 1 ) );
- $i = $lengthText;
- $haveClose = false;
- }
- }
- // <includeonly> and <noinclude> just become <ignore> tags
- if ( in_array( $lowerName, $ignoredElements ) ) {
- $accum->addNodeWithText( 'ignore', strval( substr( $text, $tagStartPos, $i - $tagStartPos ) ) );
- continue;
- }
-
- if ( $attrEnd <= $attrStart ) {
- $attr = '';
- } else {
- // Note that the attr element contains the whitespace between name and attribute,
- // this is necessary for precise reconstruction during pre-save transform.
- $attr = strval( substr( $text, $attrStart, $attrEnd - $attrStart ) );
- }
-
- $extNode = new PPNode_HipHop_Tree( 'ext' );
- $extNode->addChild( PPNode_HipHop_Tree::newWithText( 'name', $name ) );
- $extNode->addChild( PPNode_HipHop_Tree::newWithText( 'attr', $attr ) );
- if ( !$shortEnd ) {
- $extNode->addChild( PPNode_HipHop_Tree::newWithText( 'inner', $inner ) );
- }
- if ( $haveClose ) {
- $extNode->addChild( PPNode_HipHop_Tree::newWithText( 'close', $close ) );
- }
- $accum->addNode( $extNode );
- }
-
- elseif ( $found === 'line-start' ) {
- // Is this the start of a heading?
- // Line break belongs before the heading element in any case
- if ( $fakeLineStart ) {
- $fakeLineStart = false;
- } else {
- $accum->addLiteral( $curChar );
- $i++;
- }
-
- $count = intval( strspn( $text, '=', $i, 6 ) );
- if ( $count == 1 && $stackFlags['findEquals'] ) {
- // DWIM: This looks kind of like a name/value separator
- // Let's let the equals handler have it and break the potential heading
- // This is heuristic, but AFAICT the methods for completely correct disambiguation are very complex.
- } elseif ( $count > 0 ) {
- $partData = array(
- 'open' => "\n",
- 'close' => "\n",
- 'parts' => array( new PPDPart_HipHop( str_repeat( '=', $count ) ) ),
- 'startPos' => $i,
- 'count' => $count );
- $stack->push( $partData );
- $accum = $stack->getAccum();
- $stackFlags = $stack->getFlags();
- $i += $count;
- }
- } elseif ( $found === 'line-end' ) {
- $piece = $stack->getTop();
- // A heading must be open, otherwise \n wouldn't have been in the search list
- assert( $piece->open === "\n" ); // Passing the assert condition directly instead of string, as
- // HPHP /compiler/ chokes on strings when ASSERT_ACTIVE != 0.
- $part = $piece->getCurrentPart();
- // Search back through the input to see if it has a proper close
- // Do this using the reversed string since the other solutions (end anchor, etc.) are inefficient
- $wsLength = intval( strspn( $revText, " \t", $lengthText - $i ) );
- $searchStart = $i - $wsLength;
- if ( isset( $part->commentEnd ) && $searchStart - 1 == $part->commentEnd ) {
- // Comment found at line end
- // Search for equals signs before the comment
- $searchStart = intval( $part->visualEnd );
- $searchStart -= intval( strspn( $revText, " \t", $lengthText - $searchStart ) );
- }
- $count = intval( $piece->count );
- $equalsLength = intval( strspn( $revText, '=', $lengthText - $searchStart ) );
- $isTreeNode = false;
- $resultAccum = $accum;
- if ( $equalsLength > 0 ) {
- if ( $searchStart - $equalsLength == $piece->startPos ) {
- // This is just a single string of equals signs on its own line
- // Replicate the doHeadings behaviour /={count}(.+)={count}/
- // First find out how many equals signs there really are (don't stop at 6)
- $count = $equalsLength;
- if ( $count < 3 ) {
- $count = 0;
- } else {
- $count = intval( ( $count - 1 ) / 2 );
- if ( $count > 6 ) {
- $count = 6;
- }
- }
- } else {
- if ( $count > $equalsLength ) {
- $count = $equalsLength;
- }
- }
- if ( $count > 0 ) {
- // Normal match, output <h>
- $tree = new PPNode_HipHop_Tree( 'possible-h' );
- $tree->addChild( new PPNode_HipHop_Attr( 'level', $count ) );
- $tree->addChild( new PPNode_HipHop_Attr( 'i', $headingIndex++ ) );
- $tree->lastChild->nextSibling = $accum->firstNode;
- $tree->lastChild = $accum->lastNode;
- $isTreeNode = true;
- } else {
- // Single equals sign on its own line, count=0
- // Output $resultAccum
- }
- } else {
- // No match, no <h>, just pass down the inner text
- // Output $resultAccum
- }
- // Unwind the stack
- $stack->pop();
- $accum = $stack->getAccum();
- $stackFlags = $stack->getFlags();
-
- // Append the result to the enclosing accumulator
- if ( $isTreeNode ) {
- $accum->addNode( $tree );
- } else {
- $accum->addAccum( $resultAccum );
- }
- // Note that we do NOT increment the input pointer.
- // This is because the closing linebreak could be the opening linebreak of
- // another heading. Infinite loops are avoided because the next iteration MUST
- // hit the heading open case above, which unconditionally increments the
- // input pointer.
- } elseif ( $found === 'open' ) {
- # count opening brace characters
- $count = intval( strspn( $text, $curChar, $i ) );
-
- # we need to add to stack only if opening brace count is enough for one of the rules
- if ( $count >= $rule['min'] ) {
- # Add it to the stack
- $partData = array(
- 'open' => $curChar,
- 'close' => $rule['end'],
- 'count' => $count,
- 'lineStart' => ($i == 0 || $text[$i-1] === "\n"),
- );
-
- $stack->push( $partData );
- $accum = $stack->getAccum();
- $stackFlags = $stack->getFlags();
- } else {
- # Add literal brace(s)
- $accum->addLiteral( str_repeat( $curChar, $count ) );
- }
- $i += $count;
- } elseif ( $found === 'close' ) {
- $piece = $stack->getTop();
- # lets check if there are enough characters for closing brace
- $maxCount = intval( $piece->count );
- $count = intval( strspn( $text, $curChar, $i, $maxCount ) );
-
- # check for maximum matching characters (if there are 5 closing
- # characters, we will probably need only 3 - depending on the rules)
- $rule = $rules[$piece->open];
- if ( $count > $rule['max'] ) {
- # The specified maximum exists in the callback array, unless the caller
- # has made an error
- $matchingCount = intval( $rule['max'] );
- } else {
- # Count is less than the maximum
- # Skip any gaps in the callback array to find the true largest match
- # Need to use array_key_exists not isset because the callback can be null
- $matchingCount = $count;
- while ( $matchingCount > 0 && !array_key_exists( $matchingCount, $rule['names'] ) ) {
- --$matchingCount;
- }
- }
-
- if ($matchingCount <= 0) {
- # No matching element found in callback array
- # Output a literal closing brace and continue
- $accum->addLiteral( str_repeat( $curChar, $count ) );
- $i += $count;
- continue;
- }
- $name = strval( $rule['names'][$matchingCount] );
- $isTreeNode = false;
- if ( $name === 'LITERAL' ) {
- // No element, just literal text
- $resultAccum = $piece->breakSyntax( $matchingCount );
- $resultAccum->addLiteral( str_repeat( $rule['end'], $matchingCount ) );
- } else {
- # Create XML element
- # Note: $parts is already XML, does not need to be encoded further
- $isTreeNode = true;
- $parts = $piece->parts;
- $titleAccum = PPDAccum_HipHop::cast( $parts[0]->out );
- unset( $parts[0] );
-
- $tree = new PPNode_HipHop_Tree( $name );
-
- # The invocation is at the start of the line if lineStart is set in
- # the stack, and all opening brackets are used up.
- if ( $maxCount == $matchingCount && !empty( $piece->lineStart ) ) {
- $tree->addChild( new PPNode_HipHop_Attr( 'lineStart', 1 ) );
- }
- $titleNode = new PPNode_HipHop_Tree( 'title' );
- $titleNode->firstChild = $titleAccum->firstNode;
- $titleNode->lastChild = $titleAccum->lastNode;
- $tree->addChild( $titleNode );
- $argIndex = 1;
- foreach ( $parts as $variantPart ) {
- $part = PPDPart_HipHop::cast( $variantPart );
- if ( isset( $part->eqpos ) ) {
- // Find equals
- $lastNode = false;
- for ( $node = $part->out->firstNode; $node; $node = $node->nextSibling ) {
- if ( $node === $part->eqpos ) {
- break;
- }
- $lastNode = $node;
- }
- if ( !$node ) {
- throw new MWException( __METHOD__. ': eqpos not found' );
- }
- if ( $node->name !== 'equals' ) {
- throw new MWException( __METHOD__ .': eqpos is not equals' );
- }
- $equalsNode = $node;
-
- // Construct name node
- $nameNode = new PPNode_HipHop_Tree( 'name' );
- if ( $lastNode !== false ) {
- $lastNode->nextSibling = false;
- $nameNode->firstChild = $part->out->firstNode;
- $nameNode->lastChild = $lastNode;
- }
-
- // Construct value node
- $valueNode = new PPNode_HipHop_Tree( 'value' );
- if ( $equalsNode->nextSibling !== false ) {
- $valueNode->firstChild = $equalsNode->nextSibling;
- $valueNode->lastChild = $part->out->lastNode;
- }
- $partNode = new PPNode_HipHop_Tree( 'part' );
- $partNode->addChild( $nameNode );
- $partNode->addChild( $equalsNode->firstChild );
- $partNode->addChild( $valueNode );
- $tree->addChild( $partNode );
- } else {
- $partNode = new PPNode_HipHop_Tree( 'part' );
- $nameNode = new PPNode_HipHop_Tree( 'name' );
- $nameNode->addChild( new PPNode_HipHop_Attr( 'index', $argIndex++ ) );
- $valueNode = new PPNode_HipHop_Tree( 'value' );
- $valueNode->firstChild = $part->out->firstNode;
- $valueNode->lastChild = $part->out->lastNode;
- $partNode->addChild( $nameNode );
- $partNode->addChild( $valueNode );
- $tree->addChild( $partNode );
- }
- }
- }
-
- # Advance input pointer
- $i += $matchingCount;
-
- # Unwind the stack
- $stack->pop();
- $accum = $stack->getAccum();
-
- # Re-add the old stack element if it still has unmatched opening characters remaining
- if ($matchingCount < $piece->count) {
- $piece->parts = array( new PPDPart_HipHop );
- $piece->count -= $matchingCount;
- # do we still qualify for any callback with remaining count?
- $names = $rules[$piece->open]['names'];
- $skippedBraces = 0;
- $enclosingAccum = $accum;
- while ( $piece->count ) {
- if ( array_key_exists( $piece->count, $names ) ) {
- $stack->push( $piece );
- $accum = $stack->getAccum();
- break;
- }
- --$piece->count;
- $skippedBraces ++;
- }
- $enclosingAccum->addLiteral( str_repeat( $piece->open, $skippedBraces ) );
- }
-
- $stackFlags = $stack->getFlags();
-
- # Add XML element to the enclosing accumulator
- if ( $isTreeNode ) {
- $accum->addNode( $tree );
- } else {
- $accum->addAccum( $resultAccum );
- }
- } elseif ( $found === 'pipe' ) {
- $stackFlags['findEquals'] = true; // shortcut for getFlags()
- $stack->addPart();
- $accum = $stack->getAccum();
- ++$i;
- } elseif ( $found === 'equals' ) {
- $stackFlags['findEquals'] = false; // shortcut for getFlags()
- $accum->addNodeWithText( 'equals', '=' );
- $stack->getCurrentPart()->eqpos = $accum->lastNode;
- ++$i;
- }
- }
-
- # Output any remaining unclosed brackets
- foreach ( $stack->stack as $variantPiece ) {
- $piece = PPDStackElement_HipHop::cast( $variantPiece );
- $stack->rootAccum->addAccum( $piece->breakSyntax() );
- }
-
- # Enable top-level headings
- for ( $node = $stack->rootAccum->firstNode; $node; $node = $node->nextSibling ) {
- if ( isset( $node->name ) && $node->name === 'possible-h' ) {
- $node->name = 'h';
- }
- }
-
- $rootNode = new PPNode_HipHop_Tree( 'root' );
- $rootNode->firstChild = $stack->rootAccum->firstNode;
- $rootNode->lastChild = $stack->rootAccum->lastNode;
-
- // Cache
- if ($cacheable) {
- $cacheValue = sprintf( "%08d", self::CACHE_VERSION ) . serialize( $rootNode );
- $wgMemc->set( $cacheKey, $cacheValue, 86400 );
- wfProfileOut( __METHOD__.'-cache-miss' );
- wfProfileOut( __METHOD__.'-cacheable' );
- wfDebugLog( "Preprocessor", "Saved preprocessor Hash to memcached (key $cacheKey)" );
- }
-
- wfProfileOut( __METHOD__ );
- return $rootNode;
- }
-}
-
-
-
-/**
- * Stack class to help Preprocessor::preprocessToObj()
- * @ingroup Parser
- */
-class PPDStack_HipHop {
- var $stack, $rootAccum;
-
- /**
- * @var PPDStack
- */
- var $top;
- var $out;
-
- static $false = false;
-
- function __construct() {
- $this->stack = array();
- $this->top = false;
- $this->rootAccum = new PPDAccum_HipHop;
- $this->accum = $this->rootAccum;
- }
-
- /**
- * @return int
- */
- function count() {
- return count( $this->stack );
- }
-
- function getAccum() {
- return PPDAccum_HipHop::cast( $this->accum );
- }
-
- function getCurrentPart() {
- return $this->getTop()->getCurrentPart();
- }
-
- function getTop() {
- return PPDStackElement_HipHop::cast( $this->top );
- }
-
- function push( $data ) {
- if ( $data instanceof PPDStackElement_HipHop ) {
- $this->stack[] = $data;
- } else {
- $this->stack[] = new PPDStackElement_HipHop( $data );
- }
- $this->top = $this->stack[ count( $this->stack ) - 1 ];
- $this->accum = $this->top->getAccum();
- }
-
- function pop() {
- if ( !count( $this->stack ) ) {
- throw new MWException( __METHOD__.': no elements remaining' );
- }
- $temp = array_pop( $this->stack );
-
- if ( count( $this->stack ) ) {
- $this->top = $this->stack[ count( $this->stack ) - 1 ];
- $this->accum = $this->top->getAccum();
- } else {
- $this->top = self::$false;
- $this->accum = $this->rootAccum;
- }
- return $temp;
- }
-
- function addPart( $s = '' ) {
- $this->top->addPart( $s );
- $this->accum = $this->top->getAccum();
- }
-
- /**
- * @return array
- */
- function getFlags() {
- if ( !count( $this->stack ) ) {
- return array(
- 'findEquals' => false,
- 'findPipe' => false,
- 'inHeading' => false,
- );
- } else {
- return $this->top->getFlags();
- }
- }
-}
-
-/**
- * @ingroup Parser
- */
-class PPDStackElement_HipHop {
- var $open, // Opening character (\n for heading)
- $close, // Matching closing character
- $count, // Number of opening characters found (number of "=" for heading)
- $parts, // Array of PPDPart objects describing pipe-separated parts.
- $lineStart; // True if the open char appeared at the start of the input line. Not set for headings.
-
- /**
- * @param $obj PPDStackElement_HipHop
- * @return PPDStackElement_HipHop
- */
- static function cast( PPDStackElement_HipHop $obj ) {
- return $obj;
- }
-
- /**
- * @param $data array
- */
- function __construct( $data = array() ) {
- $this->parts = array( new PPDPart_HipHop );
-
- foreach ( $data as $name => $value ) {
- $this->$name = $value;
- }
- }
-
- /**
- * @return PPDAccum_HipHop
- */
- function getAccum() {
- return PPDAccum_HipHop::cast( $this->parts[count($this->parts) - 1]->out );
- }
-
- /**
- * @param $s string
- */
- function addPart( $s = '' ) {
- $this->parts[] = new PPDPart_HipHop( $s );
- }
-
- /**
- * @return PPDPart_HipHop
- */
- function getCurrentPart() {
- return PPDPart_HipHop::cast( $this->parts[count($this->parts) - 1] );
- }
-
- /**
- * @return array
- */
- function getFlags() {
- $partCount = count( $this->parts );
- $findPipe = $this->open !== "\n" && $this->open !== '[';
- return array(
- 'findPipe' => $findPipe,
- 'findEquals' => $findPipe && $partCount > 1 && !isset( $this->parts[$partCount - 1]->eqpos ),
- 'inHeading' => $this->open === "\n",
- );
- }
-
- /**
- * Get the accumulator that would result if the close is not found.
- *
- * @param $openingCount bool
- * @return PPDAccum_HipHop
- */
- function breakSyntax( $openingCount = false ) {
- if ( $this->open === "\n" ) {
- $accum = PPDAccum_HipHop::cast( $this->parts[0]->out );
- } else {
- if ( $openingCount === false ) {
- $openingCount = $this->count;
- }
- $accum = new PPDAccum_HipHop;
- $accum->addLiteral( str_repeat( $this->open, $openingCount ) );
- $first = true;
- foreach ( $this->parts as $part ) {
- if ( $first ) {
- $first = false;
- } else {
- $accum->addLiteral( '|' );
- }
- $accum->addAccum( $part->out );
- }
- }
- return $accum;
- }
-}
-
-/**
- * @ingroup Parser
- */
-class PPDPart_HipHop {
- var $out; // Output accumulator object
-
- // Optional member variables:
- // eqpos Position of equals sign in output accumulator
- // commentEnd Past-the-end input pointer for the last comment encountered
- // visualEnd Past-the-end input pointer for the end of the accumulator minus comments
-
- function __construct( $out = '' ) {
- $this->out = new PPDAccum_HipHop;
- if ( $out !== '' ) {
- $this->out->addLiteral( $out );
- }
- }
-
- static function cast( PPDPart_HipHop $obj ) {
- return $obj;
- }
-}
-
-/**
- * @ingroup Parser
- */
-class PPDAccum_HipHop {
- var $firstNode, $lastNode;
-
- function __construct() {
- $this->firstNode = $this->lastNode = false;
- }
-
- static function cast( PPDAccum_HipHop $obj ) {
- return $obj;
- }
-
- /**
- * Append a string literal
- */
- function addLiteral( string $s ) {
- if ( $this->lastNode === false ) {
- $this->firstNode = $this->lastNode = new PPNode_HipHop_Text( $s );
- } elseif ( $this->lastNode instanceof PPNode_HipHop_Text ) {
- $this->lastNode->value .= $s;
- } else {
- $this->lastNode->nextSibling = new PPNode_HipHop_Text( $s );
- $this->lastNode = $this->lastNode->nextSibling;
- }
- }
-
- /**
- * Append a PPNode
- */
- function addNode( PPNode $node ) {
- if ( $this->lastNode === false ) {
- $this->firstNode = $this->lastNode = $node;
- } else {
- $this->lastNode->nextSibling = $node;
- $this->lastNode = $node;
- }
- }
-
- /**
- * Append a tree node with text contents
- */
- function addNodeWithText( string $name, string $value ) {
- $node = PPNode_HipHop_Tree::newWithText( $name, $value );
- $this->addNode( $node );
- }
-
- /**
- * Append a PPDAccum_HipHop
- * Takes over ownership of the nodes in the source argument. These nodes may
- * subsequently be modified, especially nextSibling.
- */
- function addAccum( PPDAccum_HipHop $accum ) {
- if ( $accum->lastNode === false ) {
- // nothing to add
- } elseif ( $this->lastNode === false ) {
- $this->firstNode = $accum->firstNode;
- $this->lastNode = $accum->lastNode;
- } else {
- $this->lastNode->nextSibling = $accum->firstNode;
- $this->lastNode = $accum->lastNode;
- }
- }
-}
-
-/**
- * An expansion frame, used as a context to expand the result of preprocessToObj()
- * @ingroup Parser
- */
-class PPFrame_HipHop implements PPFrame {
-
- /**
- * @var Parser
- */
- var $parser;
-
- /**
- * @var Preprocessor
- */
- var $preprocessor;
-
- /**
- * @var Title
- */
- var $title;
- var $titleCache;
-
- /**
- * Hashtable listing templates which are disallowed for expansion in this frame,
- * having been encountered previously in parent frames.
- */
- var $loopCheckHash;
-
- /**
- * Recursion depth of this frame, top = 0
- * Note that this is NOT the same as expansion depth in expand()
- */
- var $depth;
-
- /**
- * Construct a new preprocessor frame.
- * @param $preprocessor Preprocessor: the parent preprocessor
- */
- function __construct( $preprocessor ) {
- $this->preprocessor = $preprocessor;
- $this->parser = $preprocessor->parser;
- $this->title = $this->parser->mTitle;
- $this->titleCache = array( $this->title ? $this->title->getPrefixedDBkey() : false );
- $this->loopCheckHash = array();
- $this->depth = 0;
- }
-
- /**
- * Create a new child frame
- * $args is optionally a multi-root PPNode or array containing the template arguments
- *
- * @param $args PPNode_HipHop_Array|array|bool
- * @param $title Title|bool
- * @param $indexOffset A number subtracted from the index attributes of the arguments
- *
- * @throws MWException
- * @return PPTemplateFrame_HipHop
- */
- function newChild( $args = false, $title = false, $indexOffset = 0 ) {
- $namedArgs = array();
- $numberedArgs = array();
- if ( $title === false ) {
- $title = $this->title;
- }
- if ( $args !== false ) {
- if ( $args instanceof PPNode_HipHop_Array ) {
- $args = $args->value;
- } elseif ( !is_array( $args ) ) {
- throw new MWException( __METHOD__ . ': $args must be array or PPNode_HipHop_Array' );
- }
- foreach ( $args as $arg ) {
- $bits = $arg->splitArg();
- if ( $bits['index'] !== '' ) {
- // Numbered parameter
- $numberedArgs[$bits['index']] = $bits['value'];
- unset( $namedArgs[$bits['index']] );
- } else {
- // Named parameter
- $name = trim( $this->expand( $bits['name'], PPFrame::STRIP_COMMENTS ) );
- $namedArgs[$name] = $bits['value'];
- unset( $numberedArgs[$name] );
- }
- }
- }
- return new PPTemplateFrame_HipHop( $this->preprocessor, $this, $numberedArgs, $namedArgs, $title );
- }
-
- /**
- * @throws MWException
- * @param $root
- * @param $flags int
- * @return string
- */
- function expand( $root, $flags = 0 ) {
- static $expansionDepth = 0;
- if ( is_string( $root ) ) {
- return $root;
- }
-
- if ( ++$this->parser->mPPNodeCount > $this->parser->mOptions->getMaxPPNodeCount() ) {
- $this->parser->limitationWarn( 'node-count-exceeded',
- $this->parser->mPPNodeCount,
- $this->parser->mOptions->getMaxPPNodeCount()
- );
- return '<span class="error">Node-count limit exceeded</span>';
- }
- if ( $expansionDepth > $this->parser->mOptions->getMaxPPExpandDepth() ) {
- $this->parser->limitationWarn( 'expansion-depth-exceeded',
- $expansionDepth,
- $this->parser->mOptions->getMaxPPExpandDepth()
- );
- return '<span class="error">Expansion depth limit exceeded</span>';
- }
- ++$expansionDepth;
- if ( $expansionDepth > $this->parser->mHighestExpansionDepth ) {
- $this->parser->mHighestExpansionDepth = $expansionDepth;
- }
-
- $outStack = array( '', '' );
- $iteratorStack = array( false, $root );
- $indexStack = array( 0, 0 );
-
- while ( count( $iteratorStack ) > 1 ) {
- $level = count( $outStack ) - 1;
- $iteratorNode =& $iteratorStack[ $level ];
- $out =& $outStack[$level];
- $index =& $indexStack[$level];
-
- if ( is_array( $iteratorNode ) ) {
- if ( $index >= count( $iteratorNode ) ) {
- // All done with this iterator
- $iteratorStack[$level] = false;
- $contextNode = false;
- } else {
- $contextNode = $iteratorNode[$index];
- $index++;
- }
- } elseif ( $iteratorNode instanceof PPNode_HipHop_Array ) {
- if ( $index >= $iteratorNode->getLength() ) {
- // All done with this iterator
- $iteratorStack[$level] = false;
- $contextNode = false;
- } else {
- $contextNode = $iteratorNode->item( $index );
- $index++;
- }
- } else {
- // Copy to $contextNode and then delete from iterator stack,
- // because this is not an iterator but we do have to execute it once
- $contextNode = $iteratorStack[$level];
- $iteratorStack[$level] = false;
- }
-
- $newIterator = false;
-
- if ( $contextNode === false ) {
- // nothing to do
- } elseif ( is_string( $contextNode ) ) {
- $out .= $contextNode;
- } elseif ( is_array( $contextNode ) || $contextNode instanceof PPNode_HipHop_Array ) {
- $newIterator = $contextNode;
- } elseif ( $contextNode instanceof PPNode_HipHop_Attr ) {
- // No output
- } elseif ( $contextNode instanceof PPNode_HipHop_Text ) {
- $out .= $contextNode->value;
- } elseif ( $contextNode instanceof PPNode_HipHop_Tree ) {
- if ( $contextNode->name === 'template' ) {
- # Double-brace expansion
- $bits = $contextNode->splitTemplate();
- if ( $flags & PPFrame::NO_TEMPLATES ) {
- $newIterator = $this->virtualBracketedImplode( '{{', '|', '}}', $bits['title'], $bits['parts'] );
- } else {
- $ret = $this->parser->braceSubstitution( $bits, $this );
- if ( isset( $ret['object'] ) ) {
- $newIterator = $ret['object'];
- } else {
- $out .= $ret['text'];
- }
- }
- } elseif ( $contextNode->name === 'tplarg' ) {
- # Triple-brace expansion
- $bits = $contextNode->splitTemplate();
- if ( $flags & PPFrame::NO_ARGS ) {
- $newIterator = $this->virtualBracketedImplode( '{{{', '|', '}}}', $bits['title'], $bits['parts'] );
- } else {
- $ret = $this->parser->argSubstitution( $bits, $this );
- if ( isset( $ret['object'] ) ) {
- $newIterator = $ret['object'];
- } else {
- $out .= $ret['text'];
- }
- }
- } elseif ( $contextNode->name === 'comment' ) {
- # HTML-style comment
- # Remove it in HTML, pre+remove and STRIP_COMMENTS modes
- if ( $this->parser->ot['html']
- || ( $this->parser->ot['pre'] && $this->parser->mOptions->getRemoveComments() )
- || ( $flags & PPFrame::STRIP_COMMENTS ) )
- {
- $out .= '';
- }
- # Add a strip marker in PST mode so that pstPass2() can run some old-fashioned regexes on the result
- # Not in RECOVER_COMMENTS mode (extractSections) though
- elseif ( $this->parser->ot['wiki'] && ! ( $flags & PPFrame::RECOVER_COMMENTS ) ) {
- $out .= $this->parser->insertStripItem( $contextNode->firstChild->value );
- }
- # Recover the literal comment in RECOVER_COMMENTS and pre+no-remove
- else {
- $out .= $contextNode->firstChild->value;
- }
- } elseif ( $contextNode->name === 'ignore' ) {
- # Output suppression used by <includeonly> etc.
- # OT_WIKI will only respect <ignore> in substed templates.
- # The other output types respect it unless NO_IGNORE is set.
- # extractSections() sets NO_IGNORE and so never respects it.
- if ( ( !isset( $this->parent ) && $this->parser->ot['wiki'] ) || ( $flags & PPFrame::NO_IGNORE ) ) {
- $out .= $contextNode->firstChild->value;
- } else {
- //$out .= '';
- }
- } elseif ( $contextNode->name === 'ext' ) {
- # Extension tag
- $bits = $contextNode->splitExt() + array( 'attr' => null, 'inner' => null, 'close' => null );
- $out .= $this->parser->extensionSubstitution( $bits, $this );
- } elseif ( $contextNode->name === 'h' ) {
- # Heading
- if ( $this->parser->ot['html'] ) {
- # Expand immediately and insert heading index marker
- $s = '';
- for ( $node = $contextNode->firstChild; $node; $node = $node->nextSibling ) {
- $s .= $this->expand( $node, $flags );
- }
-
- $bits = $contextNode->splitHeading();
- $titleText = $this->title->getPrefixedDBkey();
- $this->parser->mHeadings[] = array( $titleText, $bits['i'] );
- $serial = count( $this->parser->mHeadings ) - 1;
- $marker = "{$this->parser->mUniqPrefix}-h-$serial-" . Parser::MARKER_SUFFIX;
- $s = substr( $s, 0, $bits['level'] ) . $marker . substr( $s, $bits['level'] );
- $this->parser->mStripState->addGeneral( $marker, '' );
- $out .= $s;
- } else {
- # Expand in virtual stack
- $newIterator = $contextNode->getChildren();
- }
- } else {
- # Generic recursive expansion
- $newIterator = $contextNode->getChildren();
- }
- } else {
- throw new MWException( __METHOD__.': Invalid parameter type' );
- }
-
- if ( $newIterator !== false ) {
- $outStack[] = '';
- $iteratorStack[] = $newIterator;
- $indexStack[] = 0;
- } elseif ( $iteratorStack[$level] === false ) {
- // Return accumulated value to parent
- // With tail recursion
- while ( $iteratorStack[$level] === false && $level > 0 ) {
- $outStack[$level - 1] .= $out;
- array_pop( $outStack );
- array_pop( $iteratorStack );
- array_pop( $indexStack );
- $level--;
- }
- }
- }
- --$expansionDepth;
- return $outStack[0];
- }
-
- /**
- * @param $sep
- * @param $flags
- * @return string
- */
- function implodeWithFlags( $sep, $flags /*, ... */ ) {
- $args = array_slice( func_get_args(), 2 );
-
- $first = true;
- $s = '';
- foreach ( $args as $root ) {
- if ( $root instanceof PPNode_HipHop_Array ) {
- $root = $root->value;
- }
- if ( !is_array( $root ) ) {
- $root = array( $root );
- }
- foreach ( $root as $node ) {
- if ( $first ) {
- $first = false;
- } else {
- $s .= $sep;
- }
- $s .= $this->expand( $node, $flags );
- }
- }
- return $s;
- }
-
- /**
- * Implode with no flags specified
- * This previously called implodeWithFlags but has now been inlined to reduce stack depth
- * @param $sep
- * @return string
- */
- function implode( $sep /*, ... */ ) {
- $args = array_slice( func_get_args(), 1 );
-
- $first = true;
- $s = '';
- foreach ( $args as $root ) {
- if ( $root instanceof PPNode_HipHop_Array ) {
- $root = $root->value;
- }
- if ( !is_array( $root ) ) {
- $root = array( $root );
- }
- foreach ( $root as $node ) {
- if ( $first ) {
- $first = false;
- } else {
- $s .= $sep;
- }
- $s .= $this->expand( $node );
- }
- }
- return $s;
- }
-
- /**
- * Makes an object that, when expand()ed, will be the same as one obtained
- * with implode()
- *
- * @param $sep
- * @return PPNode_HipHop_Array
- */
- function virtualImplode( $sep /*, ... */ ) {
- $args = array_slice( func_get_args(), 1 );
- $out = array();
- $first = true;
-
- foreach ( $args as $root ) {
- if ( $root instanceof PPNode_HipHop_Array ) {
- $root = $root->value;
- }
- if ( !is_array( $root ) ) {
- $root = array( $root );
- }
- foreach ( $root as $node ) {
- if ( $first ) {
- $first = false;
- } else {
- $out[] = $sep;
- }
- $out[] = $node;
- }
- }
- return new PPNode_HipHop_Array( $out );
- }
-
- /**
- * Virtual implode with brackets
- *
- * @param $start
- * @param $sep
- * @param $end
- * @return PPNode_HipHop_Array
- */
- function virtualBracketedImplode( $start, $sep, $end /*, ... */ ) {
- $args = array_slice( func_get_args(), 3 );
- $out = array( $start );
- $first = true;
-
- foreach ( $args as $root ) {
- if ( $root instanceof PPNode_HipHop_Array ) {
- $root = $root->value;
- }
- if ( !is_array( $root ) ) {
- $root = array( $root );
- }
- foreach ( $root as $node ) {
- if ( $first ) {
- $first = false;
- } else {
- $out[] = $sep;
- }
- $out[] = $node;
- }
- }
- $out[] = $end;
- return new PPNode_HipHop_Array( $out );
- }
-
- function __toString() {
- return 'frame{}';
- }
-
- /**
- * @param $level bool
- * @return array|bool|String
- */
- function getPDBK( $level = false ) {
- if ( $level === false ) {
- return $this->title->getPrefixedDBkey();
- } else {
- return isset( $this->titleCache[$level] ) ? $this->titleCache[$level] : false;
- }
- }
-
- /**
- * @return array
- */
- function getArguments() {
- return array();
- }
-
- /**
- * @return array
- */
- function getNumberedArguments() {
- return array();
- }
-
- /**
- * @return array
- */
- function getNamedArguments() {
- return array();
- }
-
- /**
- * Returns true if there are no arguments in this frame
- *
- * @return bool
- */
- function isEmpty() {
- return true;
- }
-
- /**
- * @param $name
- * @return bool
- */
- function getArgument( $name ) {
- return false;
- }
-
- /**
- * Returns true if the infinite loop check is OK, false if a loop is detected
- *
- * @param $title Title
- *
- * @return bool
- */
- function loopCheck( $title ) {
- return !isset( $this->loopCheckHash[$title->getPrefixedDBkey()] );
- }
-
- /**
- * Return true if the frame is a template frame
- *
- * @return bool
- */
- function isTemplate() {
- return false;
- }
-
- /**
- * Get a title of frame
- *
- * @return Title
- */
- function getTitle() {
- return $this->title;
- }
-}
-
-/**
- * Expansion frame with template arguments
- * @ingroup Parser
- */
-class PPTemplateFrame_HipHop extends PPFrame_HipHop {
- var $numberedArgs, $namedArgs, $parent;
- var $numberedExpansionCache, $namedExpansionCache;
-
- /**
- * @param $preprocessor Preprocessor_HipHop
- * @param $parent bool
- * @param $numberedArgs array
- * @param $namedArgs array
- * @param $title Title|bool
- */
- function __construct( $preprocessor, $parent = false, $numberedArgs = array(), $namedArgs = array(), $title = false ) {
- parent::__construct( $preprocessor );
-
- $this->parent = $parent;
- $this->numberedArgs = $numberedArgs;
- $this->namedArgs = $namedArgs;
- $this->title = $title;
- $pdbk = $title ? $title->getPrefixedDBkey() : false;
- $this->titleCache = $parent->titleCache;
- $this->titleCache[] = $pdbk;
- $this->loopCheckHash = /*clone*/ $parent->loopCheckHash;
- if ( $pdbk !== false ) {
- $this->loopCheckHash[$pdbk] = true;
- }
- $this->depth = $parent->depth + 1;
- $this->numberedExpansionCache = $this->namedExpansionCache = array();
- }
-
- function __toString() {
- $s = 'tplframe{';
- $first = true;
- $args = $this->numberedArgs + $this->namedArgs;
- foreach ( $args as $name => $value ) {
- if ( $first ) {
- $first = false;
- } else {
- $s .= ', ';
- }
- $s .= "\"$name\":\"" .
- str_replace( '"', '\\"', $value->__toString() ) . '"';
- }
- $s .= '}';
- return $s;
- }
- /**
- * Returns true if there are no arguments in this frame
- *
- * @return bool
- */
- function isEmpty() {
- return !count( $this->numberedArgs ) && !count( $this->namedArgs );
- }
-
- /**
- * @return array
- */
- function getArguments() {
- $arguments = array();
- foreach ( array_merge(
- array_keys($this->numberedArgs),
- array_keys($this->namedArgs)) as $key ) {
- $arguments[$key] = $this->getArgument($key);
- }
- return $arguments;
- }
-
- /**
- * @return array
- */
- function getNumberedArguments() {
- $arguments = array();
- foreach ( array_keys($this->numberedArgs) as $key ) {
- $arguments[$key] = $this->getArgument($key);
- }
- return $arguments;
- }
-
- /**
- * @return array
- */
- function getNamedArguments() {
- $arguments = array();
- foreach ( array_keys($this->namedArgs) as $key ) {
- $arguments[$key] = $this->getArgument($key);
- }
- return $arguments;
- }
-
- /**
- * @param $index
- * @return array|bool
- */
- function getNumberedArgument( $index ) {
- if ( !isset( $this->numberedArgs[$index] ) ) {
- return false;
- }
- if ( !isset( $this->numberedExpansionCache[$index] ) ) {
- # No trimming for unnamed arguments
- $this->numberedExpansionCache[$index] = $this->parent->expand( $this->numberedArgs[$index], PPFrame::STRIP_COMMENTS );
- }
- return $this->numberedExpansionCache[$index];
- }
-
- /**
- * @param $name
- * @return bool
- */
- function getNamedArgument( $name ) {
- if ( !isset( $this->namedArgs[$name] ) ) {
- return false;
- }
- if ( !isset( $this->namedExpansionCache[$name] ) ) {
- # Trim named arguments post-expand, for backwards compatibility
- $this->namedExpansionCache[$name] = trim(
- $this->parent->expand( $this->namedArgs[$name], PPFrame::STRIP_COMMENTS ) );
- }
- return $this->namedExpansionCache[$name];
- }
-
- /**
- * @param $name
- * @return array|bool
- */
- function getArgument( $name ) {
- $text = $this->getNumberedArgument( $name );
- if ( $text === false ) {
- $text = $this->getNamedArgument( $name );
- }
- return $text;
- }
-
- /**
- * Return true if the frame is a template frame
- *
- * @return bool
- */
- function isTemplate() {
- return true;
- }
-}
-
-/**
- * Expansion frame with custom arguments
- * @ingroup Parser
- */
-class PPCustomFrame_HipHop extends PPFrame_HipHop {
- var $args;
-
- function __construct( $preprocessor, $args ) {
- parent::__construct( $preprocessor );
- $this->args = $args;
- }
-
- function __toString() {
- $s = 'cstmframe{';
- $first = true;
- foreach ( $this->args as $name => $value ) {
- if ( $first ) {
- $first = false;
- } else {
- $s .= ', ';
- }
- $s .= "\"$name\":\"" .
- str_replace( '"', '\\"', $value->__toString() ) . '"';
- }
- $s .= '}';
- return $s;
- }
-
- /**
- * @return bool
- */
- function isEmpty() {
- return !count( $this->args );
- }
-
- /**
- * @param $index
- * @return bool
- */
- function getArgument( $index ) {
- if ( !isset( $this->args[$index] ) ) {
- return false;
- }
- return $this->args[$index];
- }
-}
-
-/**
- * @ingroup Parser
- */
-class PPNode_HipHop_Tree implements PPNode {
- var $name, $firstChild, $lastChild, $nextSibling;
-
- function __construct( $name ) {
- $this->name = $name;
- $this->firstChild = $this->lastChild = $this->nextSibling = false;
- }
-
- function __toString() {
- $inner = '';
- $attribs = '';
- for ( $node = $this->firstChild; $node; $node = $node->nextSibling ) {
- if ( $node instanceof PPNode_HipHop_Attr ) {
- $attribs .= ' ' . $node->name . '="' . htmlspecialchars( $node->value ) . '"';
- } else {
- $inner .= $node->__toString();
- }
- }
- if ( $inner === '' ) {
- return "<{$this->name}$attribs/>";
- } else {
- return "<{$this->name}$attribs>$inner</{$this->name}>";
- }
- }
-
- /**
- * @param $name
- * @param $text
- * @return PPNode_HipHop_Tree
- */
- static function newWithText( $name, $text ) {
- $obj = new self( $name );
- $obj->addChild( new PPNode_HipHop_Text( $text ) );
- return $obj;
- }
-
- function addChild( $node ) {
- if ( $this->lastChild === false ) {
- $this->firstChild = $this->lastChild = $node;
- } else {
- $this->lastChild->nextSibling = $node;
- $this->lastChild = $node;
- }
- }
-
- /**
- * @return PPNode_HipHop_Array
- */
- function getChildren() {
- $children = array();
- for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) {
- $children[] = $child;
- }
- return new PPNode_HipHop_Array( $children );
- }
-
- function getFirstChild() {
- return $this->firstChild;
- }
-
- function getNextSibling() {
- return $this->nextSibling;
- }
-
- /**
- * @param $name string
- * @return array
- */
- function getChildrenOfType( $name ) {
- $children = array();
- for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) {
- if ( isset( $child->name ) && $child->name === $name ) {
- $children[] = $child;
- }
- }
- return $children;
- }
-
- /**
- * @return bool
- */
- function getLength() {
- return false;
- }
-
- /**
- * @param $i
- * @return bool
- */
- function item( $i ) {
- return false;
- }
-
- /**
- * @return string
- */
- function getName() {
- return $this->name;
- }
-
- /**
- * Split a <part> node into an associative array containing:
- * name PPNode name
- * index String index
- * value PPNode value
- *
- * @throws MWException
- * @return array
- */
- function splitArg() {
- $bits = array();
- for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) {
- if ( !isset( $child->name ) ) {
- continue;
- }
- if ( $child->name === 'name' ) {
- $bits['name'] = $child;
- if ( $child->firstChild instanceof PPNode_HipHop_Attr
- && $child->firstChild->name === 'index' )
- {
- $bits['index'] = $child->firstChild->value;
- }
- } elseif ( $child->name === 'value' ) {
- $bits['value'] = $child;
- }
- }
-
- if ( !isset( $bits['name'] ) ) {
- throw new MWException( 'Invalid brace node passed to ' . __METHOD__ );
- }
- if ( !isset( $bits['index'] ) ) {
- $bits['index'] = '';
- }
- return $bits;
- }
-
- /**
- * Split an <ext> node into an associative array containing name, attr, inner and close
- * All values in the resulting array are PPNodes. Inner and close are optional.
- *
- * @throws MWException
- * @return array
- */
- function splitExt() {
- $bits = array();
- for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) {
- if ( !isset( $child->name ) ) {
- continue;
- }
- if ( $child->name === 'name' ) {
- $bits['name'] = $child;
- } elseif ( $child->name === 'attr' ) {
- $bits['attr'] = $child;
- } elseif ( $child->name === 'inner' ) {
- $bits['inner'] = $child;
- } elseif ( $child->name === 'close' ) {
- $bits['close'] = $child;
- }
- }
- if ( !isset( $bits['name'] ) ) {
- throw new MWException( 'Invalid ext node passed to ' . __METHOD__ );
- }
- return $bits;
- }
-
- /**
- * Split an <h> node
- *
- * @throws MWException
- * @return array
- */
- function splitHeading() {
- if ( $this->name !== 'h' ) {
- throw new MWException( 'Invalid h node passed to ' . __METHOD__ );
- }
- $bits = array();
- for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) {
- if ( !isset( $child->name ) ) {
- continue;
- }
- if ( $child->name === 'i' ) {
- $bits['i'] = $child->value;
- } elseif ( $child->name === 'level' ) {
- $bits['level'] = $child->value;
- }
- }
- if ( !isset( $bits['i'] ) ) {
- throw new MWException( 'Invalid h node passed to ' . __METHOD__ );
- }
- return $bits;
- }
-
- /**
- * Split a <template> or <tplarg> node
- *
- * @return array
- */
- function splitTemplate() {
- $parts = array();
- $bits = array( 'lineStart' => '' );
- for ( $child = $this->firstChild; $child; $child = $child->nextSibling ) {
- if ( !isset( $child->name ) ) {
- continue;
- }
- if ( $child->name === 'title' ) {
- $bits['title'] = $child;
- }
- if ( $child->name === 'part' ) {
- $parts[] = $child;
- }
- if ( $child->name === 'lineStart' ) {
- $bits['lineStart'] = '1';
- }
- }
- if ( !isset( $bits['title'] ) ) {
- throw new MWException( 'Invalid node passed to ' . __METHOD__ );
- }
- $bits['parts'] = new PPNode_HipHop_Array( $parts );
- return $bits;
- }
-}
-
-/**
- * @ingroup Parser
- */
-class PPNode_HipHop_Text implements PPNode {
- var $value, $nextSibling;
-
- function __construct( $value ) {
- if ( is_object( $value ) ) {
- throw new MWException( __CLASS__ . ' given object instead of string' );
- }
- $this->value = $value;
- }
-
- function __toString() {
- return htmlspecialchars( $this->value );
- }
-
- function getNextSibling() {
- return $this->nextSibling;
- }
-
- function getChildren() { return false; }
- function getFirstChild() { return false; }
- function getChildrenOfType( $name ) { return false; }
- function getLength() { return false; }
- function item( $i ) { return false; }
- function getName() { return '#text'; }
- function splitArg() { throw new MWException( __METHOD__ . ': not supported' ); }
- function splitExt() { throw new MWException( __METHOD__ . ': not supported' ); }
- function splitHeading() { throw new MWException( __METHOD__ . ': not supported' ); }
-}
-
-/**
- * @ingroup Parser
- */
-class PPNode_HipHop_Array implements PPNode {
- var $value, $nextSibling;
-
- function __construct( $value ) {
- $this->value = $value;
- }
-
- function __toString() {
- return var_export( $this, true );
- }
-
- function getLength() {
- return count( $this->value );
- }
-
- function item( $i ) {
- return $this->value[$i];
- }
-
- function getName() { return '#nodelist'; }
-
- function getNextSibling() {
- return $this->nextSibling;
- }
-
- function getChildren() { return false; }
- function getFirstChild() { return false; }
- function getChildrenOfType( $name ) { return false; }
- function splitArg() { throw new MWException( __METHOD__ . ': not supported' ); }
- function splitExt() { throw new MWException( __METHOD__ . ': not supported' ); }
- function splitHeading() { throw new MWException( __METHOD__ . ': not supported' ); }
-}
-
-/**
- * @ingroup Parser
- */
-class PPNode_HipHop_Attr implements PPNode {
- var $name, $value, $nextSibling;
-
- function __construct( $name, $value ) {
- $this->name = $name;
- $this->value = $value;
- }
-
- function __toString() {
- return "<@{$this->name}>" . htmlspecialchars( $this->value ) . "</@{$this->name}>";
- }
-
- function getName() {
- return $this->name;
- }
-
- function getNextSibling() {
- return $this->nextSibling;
- }
-
- function getChildren() { return false; }
- function getFirstChild() { return false; }
- function getChildrenOfType( $name ) { return false; }
- function getLength() { return false; }
- function item( $i ) { return false; }
- function splitArg() { throw new MWException( __METHOD__ . ': not supported' ); }
- function splitExt() { throw new MWException( __METHOD__ . ': not supported' ); }
- function splitHeading() { throw new MWException( __METHOD__ . ': not supported' ); }
-}
diff --git a/includes/parser/StripState.php b/includes/parser/StripState.php
index ad95d5f7..5f3f18ea 100644
--- a/includes/parser/StripState.php
+++ b/includes/parser/StripState.php
@@ -112,7 +112,7 @@ class StripState {
* @return mixed
*/
protected function unstripType( $type, $text ) {
- // Shortcut
+ // Shortcut
if ( !count( $this->data[$type] ) ) {
return $text;
}
@@ -139,7 +139,7 @@ class StripState {
. '</span>';
}
if ( $this->recursionLevel >= self::UNSTRIP_RECURSION_LIMIT ) {
- return '<span class="error">' .
+ return '<span class="error">' .
wfMessage( 'parser-unstrip-recursion-limit' )
->numParams( self::UNSTRIP_RECURSION_LIMIT )->inContentLanguage()->text() .
'</span>';
@@ -156,7 +156,7 @@ class StripState {
}
/**
- * Get a StripState object which is sufficient to unstrip the given text.
+ * Get a StripState object which is sufficient to unstrip the given text.
* It will contain the minimum subset of strip items necessary.
*
* @param $text string
@@ -233,4 +233,3 @@ class StripState {
return preg_replace( $this->regex, '', $text );
}
}
-
diff --git a/includes/parser/Tidy.php b/includes/parser/Tidy.php
index ed2d436d..32b16aaf 100644
--- a/includes/parser/Tidy.php
+++ b/includes/parser/Tidy.php
@@ -59,12 +59,21 @@ class MWTidyWrapper {
dechex( mt_rand( 0, 0x7fffffff ) ) . dechex( mt_rand( 0, 0x7fffffff ) );
$this->mMarkerIndex = 0;
+ // Replace <mw:editsection> elements with placeholders
$wrappedtext = preg_replace_callback( ParserOutput::EDITSECTION_REGEX,
- array( &$this, 'replaceEditSectionLinksCallback' ), $text );
+ array( &$this, 'replaceCallback' ), $text );
+ // ...and <mw:toc> markers
+ $wrappedtext = preg_replace_callback( '/\<\\/?mw:toc\>/',
+ array( &$this, 'replaceCallback' ), $wrappedtext );
- $wrappedtext = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"'.
- ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>'.
- '<head><title>test</title></head><body>'.$wrappedtext.'</body></html>';
+ // Modify inline Microdata <link> and <meta> elements so they say <html-link> and <html-meta> so
+ // we can trick Tidy into not stripping them out by including them in tidy's new-empty-tags config
+ $wrappedtext = preg_replace( '!<(link|meta)([^>]*?)(/{0,1}>)!', '<html-$1$2$3', $wrappedtext );
+
+ // Wrap the whole thing in a doctype and body for Tidy.
+ $wrappedtext = '<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"' .
+ ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html>' .
+ '<head><title>test</title></head><body>' . $wrappedtext . '</body></html>';
return $wrappedtext;
}
@@ -74,7 +83,7 @@ class MWTidyWrapper {
*
* @return string
*/
- function replaceEditSectionLinksCallback( $m ) {
+ function replaceCallback( $m ) {
$marker = "{$this->mUniqPrefix}-item-{$this->mMarkerIndex}" . Parser::MARKER_SUFFIX;
$this->mMarkerIndex++;
$this->mTokens->setPair( $marker, $m[0] );
@@ -86,7 +95,13 @@ class MWTidyWrapper {
* @return string
*/
public function postprocess( $text ) {
- return $this->mTokens->replace( $text );
+ // Revert <html-{link,meta}> back to <{link,meta}>
+ $text = preg_replace( '!<html-(link|meta)([^>]*?)(/{0,1}>)!', '<$1$2$3', $text );
+
+ // Restore the contents of placeholder tokens
+ $text = $this->mTokens->replace( $text );
+
+ return $text;
}
}
@@ -106,7 +121,7 @@ class MWTidy {
* If tidy isn't able to correct the markup, the original will be
* returned in all its glory with a warning comment appended.
*
- * @param $text String: hideous HTML input
+ * @param string $text hideous HTML input
* @return String: corrected HTML output
*/
public static function tidy( $text ) {
@@ -146,7 +161,7 @@ class MWTidy {
global $wgTidyInternal;
$retval = 0;
- if( $wgTidyInternal ) {
+ if ( $wgTidyInternal ) {
$errorStr = self::execInternalTidy( $text, true, $retval );
} else {
$errorStr = self::execExternalTidy( $text, true, $retval );
@@ -159,7 +174,7 @@ class MWTidy {
* Spawn an external HTML tidy process and get corrected markup back from it.
* Also called in OutputHandler.php for full page validation
*
- * @param $text String: HTML to check
+ * @param string $text HTML to check
* @param $stderr Boolean: Whether to read result from STDERR rather than STDOUT
* @param &$retval int Exit code (-1 on internal error)
* @return mixed String or null
@@ -223,7 +238,7 @@ class MWTidy {
* Use the HTML tidy extension to use the tidy library in-process,
* saving the overhead of spawning a new process.
*
- * @param $text String: HTML to check
+ * @param string $text HTML to check
* @param $stderr Boolean: Whether to read result from error status instead of output
* @param &$retval int Exit code (-1 on internal error)
* @return mixed String or null
@@ -232,7 +247,7 @@ class MWTidy {
global $wgTidyConf, $wgDebugTidy;
wfProfileIn( __METHOD__ );
- if ( !MWInit::classExists( 'tidy' ) ) {
+ if ( !class_exists( 'tidy' ) ) {
wfWarn( "Unable to load internal tidy class." );
$retval = -1;
@@ -248,24 +263,24 @@ class MWTidy {
wfProfileOut( __METHOD__ );
return $tidy->errorBuffer;
+ }
+
+ $tidy->cleanRepair();
+ $retval = $tidy->getStatus();
+ if ( $retval == 2 ) {
+ // 2 is magic number for fatal error
+ // http://www.php.net/manual/en/function.tidy-get-status.php
+ $cleansource = null;
} else {
- $tidy->cleanRepair();
- $retval = $tidy->getStatus();
- if ( $retval == 2 ) {
- // 2 is magic number for fatal error
- // http://www.php.net/manual/en/function.tidy-get-status.php
- $cleansource = null;
- } else {
- $cleansource = tidy_get_output( $tidy );
- if ( $wgDebugTidy && $retval > 0 ) {
- $cleansource .= "<!--\nTidy reports:\n" .
- str_replace( '-->', '--&gt;', $tidy->errorBuffer ) .
- "\n-->";
- }
+ $cleansource = tidy_get_output( $tidy );
+ if ( $wgDebugTidy && $retval > 0 ) {
+ $cleansource .= "<!--\nTidy reports:\n" .
+ str_replace( '-->', '--&gt;', $tidy->errorBuffer ) .
+ "\n-->";
}
-
- wfProfileOut( __METHOD__ );
- return $cleansource;
}
+
+ wfProfileOut( __METHOD__ );
+ return $cleansource;
}
}
generated by cgit v1.2.2 (git 2.25.0) at 2024-05-01 08:16:05 +0000