From 7bb50c630a6b760c0cdc7662c44f8c3607954a19 Mon Sep 17 00:00:00 2001
From: Tim Starling <tstarling@users.mediawiki.org>
Date: Mon, 4 Apr 2011 12:59:55 +0000
Subject: [PATCH] The beginnings of HipHop compiled mode support. It works now
 for parser cache hits.

* Work around HipHop issue 314 (volatile broken) and issue 308 (no compilation detection) by adding some large and ugly compilation detection code to WebStart.php and doMaintenance.php.
* Provide an MW_COMPILED constant which can be used to detect compiled mode throughout the codebase.
* Introduced wfIsHipHop(), which detects either compiled or interpreted mode. Used this to work around unusual eval() return value in eval.php.
* Work around lack of ini_get() in Maintenance.php, by duplicating wfIsHipHop().
* In Maintenance::shouldExecute(), accept "include" as an inclusion function name, since all kinds of inclusion give this string in HipHop.
* Introduced new class MWInit, which provides some static functions in the pre-autoloader environment.
* Introduced MWInit::compiledPath(), which provides a relative path for invoking a compiled file, and MWInit::interpretedPath(), which provides an absolute path for interpreting a PHP file. Used these new functions in the appropriate places.
* When we are running compiled code, don't include files which would generate duplicate class, function or constant definitions. Documented the new requirements on the contents of Defines.php and UtfNormalDefines.php.
* In HipHop compiled mode, it's not possible to have executable code in the same file as a class definition.
  ** Moved MimeMagic initialisation to the constructor.
  ** Moved Namespace.php global variable initialisation to Setup.php.
  ** Moved MemcachedSessions.php initialisation to the caller in GlobalFunctions.php.
  ** Moved Sanitizer.php constants and global variables to static class members. Introduced an accessor function for the attribs regex, as a new place to put code formerly at file level.
  ** Moved Language.php initialisation of $wgLanguageNames to Language::getLanguageNames(). Removed the global variable, marked "private" since forever.

* In two places: don't use error_log() with type=3 to append to a file, HipHop doesn't support it. Use file_put_contents() with FILE_APPEND instead.
* Work around the terrible breakage of class_exists() by using MWInit::classExists() instead in various places. In WebInstaller::getPageByName(), the class_exists() was marked with a fixme comment already, so I replaced it with an autoloader solution.
---
 includes/AutoLoader.php              |  16 +
 includes/DefaultSettings.php         |   4 +-
 includes/Defines.php                 |   7 +-
 includes/ExternalStore.php           |   2 +-
 includes/GlobalFunctions.php         |  27 +-
 includes/MemcachedSessions.php       |   7 -
 includes/MimeMagic.php               |  19 +-
 includes/Namespace.php               |  29 --
 includes/Sanitizer.php               | 656 +++++++++++++--------------
 includes/Setup.php                   |  61 ++-
 includes/Skin.php                    |  20 +-
 includes/Title.php                   |   9 -
 includes/User.php                    |   2 +-
 includes/WebStart.php                |  81 ++--
 includes/installer/WebInstaller.php  |   3 -
 includes/normal/UtfNormalDefines.php |   6 +-
 includes/normal/UtfNormalUtil.php    |   2 +-
 languages/Language.php               |  26 +-
 languages/LanguageConverter.php      |   4 +-
 languages/Names.php                  |   2 +-
 maintenance/Maintenance.php          |   8 +-
 maintenance/addwiki.php              |   2 +-
 maintenance/doMaintenance.php        |  38 +-
 maintenance/eval.php                 |   2 +-
 maintenance/hiphop/file-list.small   |   3 +
 25 files changed, 562 insertions(+), 474 deletions(-)

diff --git a/includes/AutoLoader.php b/includes/AutoLoader.php
index a4c7ae59a6..6bf28544d4 100644
--- a/includes/AutoLoader.php
+++ b/includes/AutoLoader.php
@@ -457,6 +457,22 @@ $wgAutoloadLocalClasses = array(
 	'LBFactory_InstallerFake' => 'includes/installer/Installer.php',
 	'LocalSettingsGenerator' => 'includes/installer/LocalSettingsGenerator.php',
 	'WebInstaller' => 'includes/installer/WebInstaller.php',
+	'WebInstaller_Complete' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Copying' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_DBConnect' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_DBSettings' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Document' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_ExistingWiki' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Install' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Language' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Name' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Options' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Readme' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_ReleaseNotes' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Restart' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Upgrade' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_UpgradeDoc' => 'includes/installer/WebInstallerPage.php',
+	'WebInstaller_Welcome' => 'includes/installer/WebInstallerPage.php',
 	'WebInstallerPage' => 'includes/installer/WebInstallerPage.php',
 	'WebInstallerOutput' => 'includes/installer/WebInstallerOutput.php',
 	'MysqlInstaller' => 'includes/installer/MysqlInstaller.php',
diff --git a/includes/DefaultSettings.php b/includes/DefaultSettings.php
index 36daef4b1d..716162a6bd 100644
--- a/includes/DefaultSettings.php
+++ b/includes/DefaultSettings.php
@@ -28,7 +28,9 @@ if( !defined( 'MEDIAWIKI' ) ) {
 
 # Create a site configuration object. Not used for much in a default install
 if ( !defined( 'MW_PHP4' ) ) {
-	require_once( "$IP/includes/SiteConfiguration.php" );
+	if ( !defined( 'MW_COMPILED' ) ) {
+		require_once( "$IP/includes/SiteConfiguration.php" );
+	}
 	$wgConf = new SiteConfiguration;
 }
 /** @endcond */
diff --git a/includes/Defines.php b/includes/Defines.php
index 94fcf03525..693bd447a3 100644
--- a/includes/Defines.php
+++ b/includes/Defines.php
@@ -1,6 +1,11 @@
 <?php
 /**
- * A few constants that might be needed during LocalSettings.php
+ * A few constants that might be needed during LocalSettings.php.
+ *
+ * Note: these constants must all be resolvable at compile time by HipHop, 
+ * since this file will not be executed during request startup for a compiled
+ * MediaWiki.
+ *
  * @file
  */
 
diff --git a/includes/ExternalStore.php b/includes/ExternalStore.php
index ddb40c32dc..2f6b222cb8 100644
--- a/includes/ExternalStore.php
+++ b/includes/ExternalStore.php
@@ -67,7 +67,7 @@ class ExternalStore {
 
 		$class = 'ExternalStore' . ucfirst( $proto );
 		/* Any custom modules should be added to $wgAutoLoadClasses for on-demand loading */
-		if( !class_exists( $class ) ) {
+		if( !MWInit::classExists( $class ) ) {
 			return false;
 		}
 
diff --git a/includes/GlobalFunctions.php b/includes/GlobalFunctions.php
index 8520ec8f8a..db60f3e1a4 100644
--- a/includes/GlobalFunctions.php
+++ b/includes/GlobalFunctions.php
@@ -8,7 +8,9 @@ if ( !defined( 'MEDIAWIKI' ) ) {
 	die( "This file is part of MediaWiki, it is not a valid entry point" );
 }
 
-require_once dirname( __FILE__ ) . '/normal/UtfNormalUtil.php';
+if ( !defined( 'MW_COMPILED' ) ) {
+	require_once( dirname( __FILE__ ) . '/normal/UtfNormalUtil.php' );
+}
 
 // Hide compatibility functions from Doxygen
 /// @cond
@@ -329,7 +331,7 @@ function wfErrorLog( $text, $file ) {
 		$exists = file_exists( $file );
 		$size = $exists ? filesize( $file ) : false;
 		if ( !$exists || ( $size !== false && $size + strlen( $text ) < 0x7fffffff ) ) {
-			error_log( $text, 3, $file );
+			file_put_contents( $file, $text, FILE_APPEND );
 		}
 		wfRestoreWarnings();
 	}
@@ -495,7 +497,7 @@ function wfMessage( $key /*...*/) {
  */
 function wfMessageFallback( /*...*/ ) {
 	$args = func_get_args();
-	return call_user_func_array( array( 'Message', 'newFallbackSequence' ), $args );
+	return MWFunction::callArray( 'Message::newFallbackSequence', $args );
 }
 
 /**
@@ -1993,6 +1995,13 @@ function wfIsWindows() {
 	return $isWindows;
 }
 
+/**
+ * Check if we are running under HipHop
+ */
+function wfIsHipHop() {
+	return function_exists( 'hphp_thread_set_warmup_enabled' );
+}
+
 /**
  * Swap two variables
  */
@@ -2781,7 +2790,17 @@ function wfSetupSession( $sessionId = false ) {
 	global $wgSessionsInMemcached, $wgCookiePath, $wgCookieDomain,
 			$wgCookieSecure, $wgCookieHttpOnly, $wgSessionHandler;
 	if( $wgSessionsInMemcached ) {
-		require_once( 'MemcachedSessions.php' );
+		if ( !defined( 'MW_COMPILED' ) ) {
+			require_once( 'MemcachedSessions.php' );
+		}
+		session_set_save_handler( 'memsess_open', 'memsess_close', 'memsess_read', 
+			'memsess_write', 'memsess_destroy', 'memsess_gc' );
+
+		// It's necessary to register a shutdown function to call session_write_close(), 
+		// because by the time the request shutdown function for the session module is 
+		// called, $wgMemc has already been destroyed. Shutdown functions registered
+		// this way are called before object destruction.
+		register_shutdown_function( 'memsess_write_close' );
 	} elseif( $wgSessionHandler && $wgSessionHandler != ini_get( 'session.save_handler' ) ) {
 		# Only set this if $wgSessionHandler isn't null and session.save_handler
 		# hasn't already been set to the desired value (that causes errors)
diff --git a/includes/MemcachedSessions.php b/includes/MemcachedSessions.php
index 3825a3ba3d..367335951e 100644
--- a/includes/MemcachedSessions.php
+++ b/includes/MemcachedSessions.php
@@ -96,10 +96,3 @@ function memsess_write_close() {
 	session_write_close();
 }
 
-session_set_save_handler( 'memsess_open', 'memsess_close', 'memsess_read', 'memsess_write', 'memsess_destroy', 'memsess_gc' );
-
-// It's necessary to register a shutdown function to call session_write_close(), 
-// because by the time the request shutdown function for the session module is 
-// called, $wgMemc has already been destroyed. Shutdown functions registered
-// this way are called before object destruction.
-register_shutdown_function( 'memsess_write_close' );
diff --git a/includes/MimeMagic.php b/includes/MimeMagic.php
index ccbc7059fa..248c1e3708 100644
--- a/includes/MimeMagic.php
+++ b/includes/MimeMagic.php
@@ -117,14 +117,6 @@ unknown/unknown application/octet-stream application/x-empty [UNKNOWN]
 END_STRING
 );
 
-// Note: because this file is possibly included by a function,
-// we need to access the global scope explicitely!
-global $wgLoadFileinfoExtension;
-
-if ( $wgLoadFileinfoExtension ) {
-	wfDl( 'fileinfo' );
-}
-
 /**
  * Implements functions related to mime types such as detection and mapping to
  * file extension.
@@ -160,6 +152,10 @@ class MimeMagic {
 	 */
 	private static $instance;
 
+	/** True if the fileinfo extension has been loaded
+	 */
+	private static $extensionLoaded = false;
+
 	/** Initializes the MimeMagic object. This is called by MimeMagic::singleton().
 	*
 	* This constructor parses the mime.types and mime.info files and build internal mappings.
@@ -169,7 +165,7 @@ class MimeMagic {
 		*   --- load mime.types ---
 		*/
 
-		global $wgMimeTypeFile, $IP;
+		global $wgMimeTypeFile, $IP, $wgLoadFileinfoExtension;
 
 		$types = MM_WELL_KNOWN_MIME_TYPES;
 
@@ -177,6 +173,11 @@ class MimeMagic {
 			$wgMimeTypeFile = "$IP/$wgMimeTypeFile";
 		}
 
+		if ( $wgLoadFileinfoExtension && !self::$extensionLoaded ) {
+			self::$extensionLoaded = true;
+			wfDl( 'fileinfo' );
+		}
+
 		if ( $wgMimeTypeFile ) {
 			if ( is_file( $wgMimeTypeFile ) and is_readable( $wgMimeTypeFile ) ) {
 				wfDebug( __METHOD__.": loading mime types from $wgMimeTypeFile\n" );
diff --git a/includes/Namespace.php b/includes/Namespace.php
index 95c58cccdb..34ab6bb6ad 100644
--- a/includes/Namespace.php
+++ b/includes/Namespace.php
@@ -4,35 +4,6 @@
  * @file
  */
 
-/**
- * Definitions of the NS_ constants are in Defines.php
- * @private
- */
-$wgCanonicalNamespaceNames = array(
-	NS_MEDIA            => 'Media',
-	NS_SPECIAL          => 'Special',
-	NS_TALK             => 'Talk',
-	NS_USER             => 'User',
-	NS_USER_TALK        => 'User_talk',
-	NS_PROJECT          => 'Project',
-	NS_PROJECT_TALK     => 'Project_talk',
-	NS_FILE             => 'File',
-	NS_FILE_TALK        => 'File_talk',
-	NS_MEDIAWIKI        => 'MediaWiki',
-	NS_MEDIAWIKI_TALK   => 'MediaWiki_talk',
-	NS_TEMPLATE         => 'Template',
-	NS_TEMPLATE_TALK    => 'Template_talk',
-	NS_HELP             => 'Help',
-	NS_HELP_TALK        => 'Help_talk',
-	NS_CATEGORY         => 'Category',
-	NS_CATEGORY_TALK    => 'Category_talk',
-);
-
-/// @todo UGLY UGLY
-if( is_array( $wgExtraNamespaces ) ) {
-	$wgCanonicalNamespaceNames = $wgCanonicalNamespaceNames + $wgExtraNamespaces;
-}
-
 /**
  * This is a utility class with only static functions
  * for dealing with namespaces that encodes all the
diff --git a/includes/Sanitizer.php b/includes/Sanitizer.php
index 4c99e82d55..27b17ce9f9 100644
--- a/includes/Sanitizer.php
+++ b/includes/Sanitizer.php
@@ -25,322 +25,323 @@
  */
 
 /**
- * Regular expression to match various types of character references in
- * Sanitizer::normalizeCharReferences and Sanitizer::decodeCharReferences
- */
-define( 'MW_CHAR_REFS_REGEX',
-	'/&([A-Za-z0-9\x80-\xff]+);
-	 |&\#([0-9]+);
-	 |&\#[xX]([0-9A-Fa-f]+);
-	 |(&)/x' );
-
-/**
- * Regular expression to match HTML/XML attribute pairs within a tag.
- * Allows some... latitude.
- * Used in Sanitizer::fixTagAttributes and Sanitizer::decodeTagAttributes
+ * XHTML sanitizer for MediaWiki
+ * @ingroup Parser
  */
-$attribFirst = '[:A-Z_a-z0-9]';
-$attrib = '[:A-Z_a-z-.0-9]';
-$space = '[\x09\x0a\x0d\x20]';
-define( 'MW_ATTRIBS_REGEX',
-	"/(?:^|$space)({$attribFirst}{$attrib}*)
-	  ($space*=$space*
-		(?:
-		 # The attribute value: quoted or alone
-		  \"([^<\"]*)\"
-		 | '([^<']*)'
-		 |  ([a-zA-Z0-9!#$%&()*,\\-.\\/:;<>?@[\\]^_`{|}~]+)
-		 |  (\#[0-9a-fA-F]+) # Technically wrong, but lots of
-							 # colors are specified like this.
-							 # We'll be normalizing it.
-		)
-	   )?(?=$space|\$)/sx" );
+class Sanitizer {
+	/**
+	 * Regular expression to match various types of character references in
+	 * Sanitizer::normalizeCharReferences and Sanitizer::decodeCharReferences
+	 */
+	const CHAR_REFS_REGEX = 
+		'/&([A-Za-z0-9\x80-\xff]+);
+		 |&\#([0-9]+);
+		 |&\#[xX]([0-9A-Fa-f]+);
+		 |(&)/x';
 
-/**
- * Regular expression to match URIs that could trigger script execution
- */
-define( 'MW_EVIL_URI_PATTERN', '!(^|\s|\*/\s*)(javascript|vbscript)([^\w]|$)!i' );
+	const EVIL_URI_PATTERN = '!(^|\s|\*/\s*)(javascript|vbscript)([^\w]|$)!i';
+	const XMLNS_ATTRIBUTE_PATTERN = "/^xmlns:[:A-Z_a-z-.0-9]+$/";
 
-/**
- * Regular expression to match namespace attributes
- */
-define( 'MW_XMLNS_ATTRIBUTE_PATTRN', "/^xmlns:$attrib+$/" );
+	/**
+	 * List of all named character entities defined in HTML 4.01
+	 * http://www.w3.org/TR/html4/sgml/entities.html
+	 * @private
+	 */
+	static $htmlEntities = array(
+		'Aacute'   => 193,
+		'aacute'   => 225,
+		'Acirc'    => 194,
+		'acirc'    => 226,
+		'acute'    => 180,
+		'AElig'    => 198,
+		'aelig'    => 230,
+		'Agrave'   => 192,
+		'agrave'   => 224,
+		'alefsym'  => 8501,
+		'Alpha'    => 913,
+		'alpha'    => 945,
+		'amp'      => 38,
+		'and'      => 8743,
+		'ang'      => 8736,
+		'Aring'    => 197,
+		'aring'    => 229,
+		'asymp'    => 8776,
+		'Atilde'   => 195,
+		'atilde'   => 227,
+		'Auml'     => 196,
+		'auml'     => 228,
+		'bdquo'    => 8222,
+		'Beta'     => 914,
+		'beta'     => 946,
+		'brvbar'   => 166,
+		'bull'     => 8226,
+		'cap'      => 8745,
+		'Ccedil'   => 199,
+		'ccedil'   => 231,
+		'cedil'    => 184,
+		'cent'     => 162,
+		'Chi'      => 935,
+		'chi'      => 967,
+		'circ'     => 710,
+		'clubs'    => 9827,
+		'cong'     => 8773,
+		'copy'     => 169,
+		'crarr'    => 8629,
+		'cup'      => 8746,
+		'curren'   => 164,
+		'dagger'   => 8224,
+		'Dagger'   => 8225,
+		'darr'     => 8595,
+		'dArr'     => 8659,
+		'deg'      => 176,
+		'Delta'    => 916,
+		'delta'    => 948,
+		'diams'    => 9830,
+		'divide'   => 247,
+		'Eacute'   => 201,
+		'eacute'   => 233,
+		'Ecirc'    => 202,
+		'ecirc'    => 234,
+		'Egrave'   => 200,
+		'egrave'   => 232,
+		'empty'    => 8709,
+		'emsp'     => 8195,
+		'ensp'     => 8194,
+		'Epsilon'  => 917,
+		'epsilon'  => 949,
+		'equiv'    => 8801,
+		'Eta'      => 919,
+		'eta'      => 951,
+		'ETH'      => 208,
+		'eth'      => 240,
+		'Euml'     => 203,
+		'euml'     => 235,
+		'euro'     => 8364,
+		'exist'    => 8707,
+		'fnof'     => 402,
+		'forall'   => 8704,
+		'frac12'   => 189,
+		'frac14'   => 188,
+		'frac34'   => 190,
+		'frasl'    => 8260,
+		'Gamma'    => 915,
+		'gamma'    => 947,
+		'ge'       => 8805,
+		'gt'       => 62,
+		'harr'     => 8596,
+		'hArr'     => 8660,
+		'hearts'   => 9829,
+		'hellip'   => 8230,
+		'Iacute'   => 205,
+		'iacute'   => 237,
+		'Icirc'    => 206,
+		'icirc'    => 238,
+		'iexcl'    => 161,
+		'Igrave'   => 204,
+		'igrave'   => 236,
+		'image'    => 8465,
+		'infin'    => 8734,
+		'int'      => 8747,
+		'Iota'     => 921,
+		'iota'     => 953,
+		'iquest'   => 191,
+		'isin'     => 8712,
+		'Iuml'     => 207,
+		'iuml'     => 239,
+		'Kappa'    => 922,
+		'kappa'    => 954,
+		'Lambda'   => 923,
+		'lambda'   => 955,
+		'lang'     => 9001,
+		'laquo'    => 171,
+		'larr'     => 8592,
+		'lArr'     => 8656,
+		'lceil'    => 8968,
+		'ldquo'    => 8220,
+		'le'       => 8804,
+		'lfloor'   => 8970,
+		'lowast'   => 8727,
+		'loz'      => 9674,
+		'lrm'      => 8206,
+		'lsaquo'   => 8249,
+		'lsquo'    => 8216,
+		'lt'       => 60,
+		'macr'     => 175,
+		'mdash'    => 8212,
+		'micro'    => 181,
+		'middot'   => 183,
+		'minus'    => 8722,
+		'Mu'       => 924,
+		'mu'       => 956,
+		'nabla'    => 8711,
+		'nbsp'     => 160,
+		'ndash'    => 8211,
+		'ne'       => 8800,
+		'ni'       => 8715,
+		'not'      => 172,
+		'notin'    => 8713,
+		'nsub'     => 8836,
+		'Ntilde'   => 209,
+		'ntilde'   => 241,
+		'Nu'       => 925,
+		'nu'       => 957,
+		'Oacute'   => 211,
+		'oacute'   => 243,
+		'Ocirc'    => 212,
+		'ocirc'    => 244,
+		'OElig'    => 338,
+		'oelig'    => 339,
+		'Ograve'   => 210,
+		'ograve'   => 242,
+		'oline'    => 8254,
+		'Omega'    => 937,
+		'omega'    => 969,
+		'Omicron'  => 927,
+		'omicron'  => 959,
+		'oplus'    => 8853,
+		'or'       => 8744,
+		'ordf'     => 170,
+		'ordm'     => 186,
+		'Oslash'   => 216,
+		'oslash'   => 248,
+		'Otilde'   => 213,
+		'otilde'   => 245,
+		'otimes'   => 8855,
+		'Ouml'     => 214,
+		'ouml'     => 246,
+		'para'     => 182,
+		'part'     => 8706,
+		'permil'   => 8240,
+		'perp'     => 8869,
+		'Phi'      => 934,
+		'phi'      => 966,
+		'Pi'       => 928,
+		'pi'       => 960,
+		'piv'      => 982,
+		'plusmn'   => 177,
+		'pound'    => 163,
+		'prime'    => 8242,
+		'Prime'    => 8243,
+		'prod'     => 8719,
+		'prop'     => 8733,
+		'Psi'      => 936,
+		'psi'      => 968,
+		'quot'     => 34,
+		'radic'    => 8730,
+		'rang'     => 9002,
+		'raquo'    => 187,
+		'rarr'     => 8594,
+		'rArr'     => 8658,
+		'rceil'    => 8969,
+		'rdquo'    => 8221,
+		'real'     => 8476,
+		'reg'      => 174,
+		'rfloor'   => 8971,
+		'Rho'      => 929,
+		'rho'      => 961,
+		'rlm'      => 8207,
+		'rsaquo'   => 8250,
+		'rsquo'    => 8217,
+		'sbquo'    => 8218,
+		'Scaron'   => 352,
+		'scaron'   => 353,
+		'sdot'     => 8901,
+		'sect'     => 167,
+		'shy'      => 173,
+		'Sigma'    => 931,
+		'sigma'    => 963,
+		'sigmaf'   => 962,
+		'sim'      => 8764,
+		'spades'   => 9824,
+		'sub'      => 8834,
+		'sube'     => 8838,
+		'sum'      => 8721,
+		'sup'      => 8835,
+		'sup1'     => 185,
+		'sup2'     => 178,
+		'sup3'     => 179,
+		'supe'     => 8839,
+		'szlig'    => 223,
+		'Tau'      => 932,
+		'tau'      => 964,
+		'there4'   => 8756,
+		'Theta'    => 920,
+		'theta'    => 952,
+		'thetasym' => 977,
+		'thinsp'   => 8201,
+		'THORN'    => 222,
+		'thorn'    => 254,
+		'tilde'    => 732,
+		'times'    => 215,
+		'trade'    => 8482,
+		'Uacute'   => 218,
+		'uacute'   => 250,
+		'uarr'     => 8593,
+		'uArr'     => 8657,
+		'Ucirc'    => 219,
+		'ucirc'    => 251,
+		'Ugrave'   => 217,
+		'ugrave'   => 249,
+		'uml'      => 168,
+		'upsih'    => 978,
+		'Upsilon'  => 933,
+		'upsilon'  => 965,
+		'Uuml'     => 220,
+		'uuml'     => 252,
+		'weierp'   => 8472,
+		'Xi'       => 926,
+		'xi'       => 958,
+		'Yacute'   => 221,
+		'yacute'   => 253,
+		'yen'      => 165,
+		'Yuml'     => 376,
+		'yuml'     => 255,
+		'Zeta'     => 918,
+		'zeta'     => 950,
+		'zwj'      => 8205,
+		'zwnj'     => 8204
+	);
 
-/**
- * List of all named character entities defined in HTML 4.01
- * http://www.w3.org/TR/html4/sgml/entities.html
- * @private
- */
-global $wgHtmlEntities;
-$wgHtmlEntities = array(
-	'Aacute'   => 193,
-	'aacute'   => 225,
-	'Acirc'    => 194,
-	'acirc'    => 226,
-	'acute'    => 180,
-	'AElig'    => 198,
-	'aelig'    => 230,
-	'Agrave'   => 192,
-	'agrave'   => 224,
-	'alefsym'  => 8501,
-	'Alpha'    => 913,
-	'alpha'    => 945,
-	'amp'      => 38,
-	'and'      => 8743,
-	'ang'      => 8736,
-	'Aring'    => 197,
-	'aring'    => 229,
-	'asymp'    => 8776,
-	'Atilde'   => 195,
-	'atilde'   => 227,
-	'Auml'     => 196,
-	'auml'     => 228,
-	'bdquo'    => 8222,
-	'Beta'     => 914,
-	'beta'     => 946,
-	'brvbar'   => 166,
-	'bull'     => 8226,
-	'cap'      => 8745,
-	'Ccedil'   => 199,
-	'ccedil'   => 231,
-	'cedil'    => 184,
-	'cent'     => 162,
-	'Chi'      => 935,
-	'chi'      => 967,
-	'circ'     => 710,
-	'clubs'    => 9827,
-	'cong'     => 8773,
-	'copy'     => 169,
-	'crarr'    => 8629,
-	'cup'      => 8746,
-	'curren'   => 164,
-	'dagger'   => 8224,
-	'Dagger'   => 8225,
-	'darr'     => 8595,
-	'dArr'     => 8659,
-	'deg'      => 176,
-	'Delta'    => 916,
-	'delta'    => 948,
-	'diams'    => 9830,
-	'divide'   => 247,
-	'Eacute'   => 201,
-	'eacute'   => 233,
-	'Ecirc'    => 202,
-	'ecirc'    => 234,
-	'Egrave'   => 200,
-	'egrave'   => 232,
-	'empty'    => 8709,
-	'emsp'     => 8195,
-	'ensp'     => 8194,
-	'Epsilon'  => 917,
-	'epsilon'  => 949,
-	'equiv'    => 8801,
-	'Eta'      => 919,
-	'eta'      => 951,
-	'ETH'      => 208,
-	'eth'      => 240,
-	'Euml'     => 203,
-	'euml'     => 235,
-	'euro'     => 8364,
-	'exist'    => 8707,
-	'fnof'     => 402,
-	'forall'   => 8704,
-	'frac12'   => 189,
-	'frac14'   => 188,
-	'frac34'   => 190,
-	'frasl'    => 8260,
-	'Gamma'    => 915,
-	'gamma'    => 947,
-	'ge'       => 8805,
-	'gt'       => 62,
-	'harr'     => 8596,
-	'hArr'     => 8660,
-	'hearts'   => 9829,
-	'hellip'   => 8230,
-	'Iacute'   => 205,
-	'iacute'   => 237,
-	'Icirc'    => 206,
-	'icirc'    => 238,
-	'iexcl'    => 161,
-	'Igrave'   => 204,
-	'igrave'   => 236,
-	'image'    => 8465,
-	'infin'    => 8734,
-	'int'      => 8747,
-	'Iota'     => 921,
-	'iota'     => 953,
-	'iquest'   => 191,
-	'isin'     => 8712,
-	'Iuml'     => 207,
-	'iuml'     => 239,
-	'Kappa'    => 922,
-	'kappa'    => 954,
-	'Lambda'   => 923,
-	'lambda'   => 955,
-	'lang'     => 9001,
-	'laquo'    => 171,
-	'larr'     => 8592,
-	'lArr'     => 8656,
-	'lceil'    => 8968,
-	'ldquo'    => 8220,
-	'le'       => 8804,
-	'lfloor'   => 8970,
-	'lowast'   => 8727,
-	'loz'      => 9674,
-	'lrm'      => 8206,
-	'lsaquo'   => 8249,
-	'lsquo'    => 8216,
-	'lt'       => 60,
-	'macr'     => 175,
-	'mdash'    => 8212,
-	'micro'    => 181,
-	'middot'   => 183,
-	'minus'    => 8722,
-	'Mu'       => 924,
-	'mu'       => 956,
-	'nabla'    => 8711,
-	'nbsp'     => 160,
-	'ndash'    => 8211,
-	'ne'       => 8800,
-	'ni'       => 8715,
-	'not'      => 172,
-	'notin'    => 8713,
-	'nsub'     => 8836,
-	'Ntilde'   => 209,
-	'ntilde'   => 241,
-	'Nu'       => 925,
-	'nu'       => 957,
-	'Oacute'   => 211,
-	'oacute'   => 243,
-	'Ocirc'    => 212,
-	'ocirc'    => 244,
-	'OElig'    => 338,
-	'oelig'    => 339,
-	'Ograve'   => 210,
-	'ograve'   => 242,
-	'oline'    => 8254,
-	'Omega'    => 937,
-	'omega'    => 969,
-	'Omicron'  => 927,
-	'omicron'  => 959,
-	'oplus'    => 8853,
-	'or'       => 8744,
-	'ordf'     => 170,
-	'ordm'     => 186,
-	'Oslash'   => 216,
-	'oslash'   => 248,
-	'Otilde'   => 213,
-	'otilde'   => 245,
-	'otimes'   => 8855,
-	'Ouml'     => 214,
-	'ouml'     => 246,
-	'para'     => 182,
-	'part'     => 8706,
-	'permil'   => 8240,
-	'perp'     => 8869,
-	'Phi'      => 934,
-	'phi'      => 966,
-	'Pi'       => 928,
-	'pi'       => 960,
-	'piv'      => 982,
-	'plusmn'   => 177,
-	'pound'    => 163,
-	'prime'    => 8242,
-	'Prime'    => 8243,
-	'prod'     => 8719,
-	'prop'     => 8733,
-	'Psi'      => 936,
-	'psi'      => 968,
-	'quot'     => 34,
-	'radic'    => 8730,
-	'rang'     => 9002,
-	'raquo'    => 187,
-	'rarr'     => 8594,
-	'rArr'     => 8658,
-	'rceil'    => 8969,
-	'rdquo'    => 8221,
-	'real'     => 8476,
-	'reg'      => 174,
-	'rfloor'   => 8971,
-	'Rho'      => 929,
-	'rho'      => 961,
-	'rlm'      => 8207,
-	'rsaquo'   => 8250,
-	'rsquo'    => 8217,
-	'sbquo'    => 8218,
-	'Scaron'   => 352,
-	'scaron'   => 353,
-	'sdot'     => 8901,
-	'sect'     => 167,
-	'shy'      => 173,
-	'Sigma'    => 931,
-	'sigma'    => 963,
-	'sigmaf'   => 962,
-	'sim'      => 8764,
-	'spades'   => 9824,
-	'sub'      => 8834,
-	'sube'     => 8838,
-	'sum'      => 8721,
-	'sup'      => 8835,
-	'sup1'     => 185,
-	'sup2'     => 178,
-	'sup3'     => 179,
-	'supe'     => 8839,
-	'szlig'    => 223,
-	'Tau'      => 932,
-	'tau'      => 964,
-	'there4'   => 8756,
-	'Theta'    => 920,
-	'theta'    => 952,
-	'thetasym' => 977,
-	'thinsp'   => 8201,
-	'THORN'    => 222,
-	'thorn'    => 254,
-	'tilde'    => 732,
-	'times'    => 215,
-	'trade'    => 8482,
-	'Uacute'   => 218,
-	'uacute'   => 250,
-	'uarr'     => 8593,
-	'uArr'     => 8657,
-	'Ucirc'    => 219,
-	'ucirc'    => 251,
-	'Ugrave'   => 217,
-	'ugrave'   => 249,
-	'uml'      => 168,
-	'upsih'    => 978,
-	'Upsilon'  => 933,
-	'upsilon'  => 965,
-	'Uuml'     => 220,
-	'uuml'     => 252,
-	'weierp'   => 8472,
-	'Xi'       => 926,
-	'xi'       => 958,
-	'Yacute'   => 221,
-	'yacute'   => 253,
-	'yen'      => 165,
-	'Yuml'     => 376,
-	'yuml'     => 255,
-	'Zeta'     => 918,
-	'zeta'     => 950,
-	'zwj'      => 8205,
-	'zwnj'     => 8204 );
+	/**
+	 * Character entity aliases accepted by MediaWiki
+	 */
+	static $htmlEntityAliases = array(
+		'רלמ' => 'rlm',
+		'رلم' => 'rlm',
+	);
 
-/**
- * Character entity aliases accepted by MediaWiki
- */
-global $wgHtmlEntityAliases;
-$wgHtmlEntityAliases = array(
-	'רלמ' => 'rlm',
-	'رلم' => 'rlm',
-);
+	/**
+	 * Lazy-initialised attributes regex, see getAttribsRegex()
+	 */
+	static $attribsRegex;
 
+	/**
+	 * Regular expression to match HTML/XML attribute pairs within a tag.
+	 * Allows some... latitude.
+	 * Used in Sanitizer::fixTagAttributes and Sanitizer::decodeTagAttributes
+	 */
+	static function getAttribsRegex() {
+		if ( self::$attribsRegex === null ) {
+			$attribFirst = '[:A-Z_a-z0-9]';
+			$attrib = '[:A-Z_a-z-.0-9]';
+			$space = '[\x09\x0a\x0d\x20]';
+			self::$attribsRegex = 
+				"/(?:^|$space)({$attribFirst}{$attrib}*)
+				  ($space*=$space*
+					(?:
+					 # The attribute value: quoted or alone
+					  \"([^<\"]*)\"
+					 | '([^<']*)'
+					 |  ([a-zA-Z0-9!#$%&()*,\\-.\\/:;<>?@[\\]^_`{|}~]+)
+					 |  (\#[0-9a-fA-F]+) # Technically wrong, but lots of
+										 # colors are specified like this.
+										 # We'll be normalizing it.
+					)
+				)?(?=$space|\$)/sx";
+		}
+		return self::$attribsRegex;
+	}
 
-/**
- * XHTML sanitizer for MediaWiki
- * @ingroup Parser
- */
-class Sanitizer {
 	/**
 	 * Cleans up HTML, removes dangerous tags and attributes, and
 	 * removes HTML comments
@@ -635,8 +636,8 @@ class Sanitizer {
 		$out = array();
 		foreach( $attribs as $attribute => $value ) {
 			#allow XML namespace declaration if RDFa is enabled
-			if ( $wgAllowRdfaAttributes && preg_match( MW_XMLNS_ATTRIBUTE_PATTRN, $attribute ) ) {
-				if ( !preg_match( MW_EVIL_URI_PATTERN, $value ) ) {
+			if ( $wgAllowRdfaAttributes && preg_match( self::XMLNS_ATTRIBUTE_PATTERN, $attribute ) ) {
+				if ( !preg_match( self::EVIL_URI_PATTERN, $value ) ) {
 					$out[$attribute] = $value;
 				}
 
@@ -666,7 +667,7 @@ class Sanitizer {
 				$attribute === 'itemscope' || $attribute === 'itemtype' ) {                         #HTML5 microdata
 
 				//Paranoia. Allow "simple" values but suppress javascript
-				if ( preg_match( MW_EVIL_URI_PATTERN, $value ) ) {
+				if ( preg_match( self::EVIL_URI_PATTERN, $value ) ) {
 					continue; 
 				}
 			}
@@ -1002,7 +1003,7 @@ class Sanitizer {
 		$attribs = array();
 		$pairs = array();
 		if( !preg_match_all(
-			MW_ATTRIBS_REGEX,
+			self::getAttribsRegex(),
 			$text,
 			$pairs,
 			PREG_SET_ORDER ) ) {
@@ -1025,7 +1026,7 @@ class Sanitizer {
 
 	/**
 	 * Pick the appropriate attribute value from a match set from the
-	 * MW_ATTRIBS_REGEX matches.
+	 * attribs regex matches.
 	 *
 	 * @param $set Array
 	 * @return String
@@ -1105,7 +1106,7 @@ class Sanitizer {
 	 */
 	static function normalizeCharReferences( $text ) {
 		return preg_replace_callback(
-			MW_CHAR_REFS_REGEX,
+			self::CHAR_REFS_REGEX,
 			array( 'Sanitizer', 'normalizeCharReferencesCallback' ),
 			$text );
 	}
@@ -1140,14 +1141,13 @@ class Sanitizer {
 	 * @return String
 	 */
 	static function normalizeEntity( $name ) {
-		global $wgHtmlEntities, $wgHtmlEntityAliases;
-		if ( isset( $wgHtmlEntityAliases[$name] ) ) {
-			return "&{$wgHtmlEntityAliases[$name]};";
+		if ( isset( self::$htmlEntityAliases[$name] ) ) {
+			return '&' . self::$htmlEntityAliases[$name] . ';';
 		} elseif ( in_array( $name,
 		array( 'lt', 'gt', 'amp', 'quot' ) ) ) {
 			return "&$name;";
-		} elseif ( isset( $wgHtmlEntities[$name] ) ) {
-			return "&#{$wgHtmlEntities[$name]};";
+		} elseif ( isset( self::$htmlEntities[$name] ) ) {
+			return '&#' . self::$htmlEntities[$name] . ';';
 		} else {
 			return "&amp;$name;";
 		}
@@ -1194,7 +1194,7 @@ class Sanitizer {
 	 */
 	public static function decodeCharReferences( $text ) {
 		return preg_replace_callback(
-			MW_CHAR_REFS_REGEX,
+			self::CHAR_REFS_REGEX,
 			array( 'Sanitizer', 'decodeCharReferencesCallback' ),
 			$text );
 	}
@@ -1212,7 +1212,7 @@ class Sanitizer {
 	public static function decodeCharReferencesAndNormalize( $text ) {
 		global $wgContLang;
 		$text = preg_replace_callback(
-			MW_CHAR_REFS_REGEX,
+			self::CHAR_REFS_REGEX,
 			array( 'Sanitizer', 'decodeCharReferencesCallback' ),
 			$text, /* limit */ -1, $count );
 
@@ -1263,12 +1263,11 @@ class Sanitizer {
 	 * @return String
 	 */
 	static function decodeEntity( $name ) {
-		global $wgHtmlEntities, $wgHtmlEntityAliases;
-		if ( isset( $wgHtmlEntityAliases[$name] ) ) {
-			$name = $wgHtmlEntityAliases[$name];
+		if ( isset( self::$htmlEntityAliases[$name] ) ) {
+			$name = self::$htmlEntityAliases[$name];
 		}
-		if( isset( $wgHtmlEntities[$name] ) ) {
-			return codepointToUtf8( $wgHtmlEntities[$name] );
+		if( isset( self::$htmlEntities[$name] ) ) {
+			return codepointToUtf8( self::$htmlEntities[$name] );
 		} else {
 			return "&$name;";
 		}
@@ -1493,9 +1492,8 @@ class Sanitizer {
 	 * @return String
 	 */
 	static function hackDocType() {
-		global $wgHtmlEntities;
 		$out = "<!DOCTYPE html [\n";
-		foreach( $wgHtmlEntities as $entity => $codepoint ) {
+		foreach( self::$htmlEntities as $entity => $codepoint ) {
 			$out .= "<!ENTITY $entity \"&#$codepoint;\">";
 		}
 		$out .= "]>\n";
diff --git a/includes/Setup.php b/includes/Setup.php
index 9bfe418078..95271df7e3 100644
--- a/includes/Setup.php
+++ b/includes/Setup.php
@@ -218,6 +218,35 @@ if ( $wgMetaNamespace === false ) {
 	$wgMetaNamespace = str_replace( ' ', '_', $wgSitename );
 }
 
+/**
+ * Definitions of the NS_ constants are in Defines.php
+ * @private
+ */
+$wgCanonicalNamespaceNames = array(
+	NS_MEDIA            => 'Media',
+	NS_SPECIAL          => 'Special',
+	NS_TALK             => 'Talk',
+	NS_USER             => 'User',
+	NS_USER_TALK        => 'User_talk',
+	NS_PROJECT          => 'Project',
+	NS_PROJECT_TALK     => 'Project_talk',
+	NS_FILE             => 'File',
+	NS_FILE_TALK        => 'File_talk',
+	NS_MEDIAWIKI        => 'MediaWiki',
+	NS_MEDIAWIKI_TALK   => 'MediaWiki_talk',
+	NS_TEMPLATE         => 'Template',
+	NS_TEMPLATE_TALK    => 'Template_talk',
+	NS_HELP             => 'Help',
+	NS_HELP_TALK        => 'Help_talk',
+	NS_CATEGORY         => 'Category',
+	NS_CATEGORY_TALK    => 'Category_talk',
+);
+
+/// @todo UGLY UGLY
+if( is_array( $wgExtraNamespaces ) ) {
+	$wgCanonicalNamespaceNames = $wgCanonicalNamespaceNames + $wgExtraNamespaces;
+}
+
 # These are now the same, always
 # To determine the user language, use $wgLang->getCode()
 $wgContLanguageCode = $wgLanguageCode;
@@ -274,22 +303,24 @@ if ( $wgNewUserLog ) {
 	$wgLogActions['newusers/autocreate'] = 'newuserlog-autocreate-entry';
 }
 
-if ( !class_exists( 'AutoLoader' ) ) {
-	require_once( "$IP/includes/AutoLoader.php" );
-}
+if ( !defined( 'MW_COMPILED' ) ) {
+	if ( !MWInit::classExists( 'AutoLoader' ) ) {
+		require_once( "$IP/includes/AutoLoader.php" );
+	}
 
-wfProfileIn( $fname . '-exception' );
-require_once( "$IP/includes/Exception.php" );
-wfInstallExceptionHandler();
-wfProfileOut( $fname . '-exception' );
-
-wfProfileIn( $fname . '-includes' );
-require_once( "$IP/includes/GlobalFunctions.php" );
-require_once( "$IP/includes/Hooks.php" );
-require_once( "$IP/includes/Namespace.php" );
-require_once( "$IP/includes/ProxyTools.php" );
-require_once( "$IP/includes/ImageFunctions.php" );
-wfProfileOut( $fname . '-includes' );
+	wfProfileIn( $fname . '-exception' );
+	require_once( "$IP/includes/Exception.php" );
+	wfInstallExceptionHandler();
+	wfProfileOut( $fname . '-exception' );
+
+	wfProfileIn( $fname . '-includes' );
+	require_once( "$IP/includes/GlobalFunctions.php" );
+	require_once( "$IP/includes/Hooks.php" );
+	require_once( "$IP/includes/Namespace.php" );
+	require_once( "$IP/includes/ProxyTools.php" );
+	require_once( "$IP/includes/ImageFunctions.php" );
+	wfProfileOut( $fname . '-includes' );
+}
 wfProfileIn( $fname . '-misc1' );
 
 # Raise the memory limit if it's too low
diff --git a/includes/Skin.php b/includes/Skin.php
index 7fbce41fd5..f1f0cde8cb 100644
--- a/includes/Skin.php
+++ b/includes/Skin.php
@@ -138,24 +138,28 @@ abstract class Skin {
 		$className = "Skin{$skinName}";
 
 		# Grab the skin class and initialise it.
-		if ( !class_exists( $className ) ) {
-			// Preload base classes to work around APC/PHP5 bug
-			$deps = "{$wgStyleDirectory}/{$skinName}.deps.php";
+		if ( !MWInit::classExists( $className ) ) {
 
-			if ( file_exists( $deps ) ) {
-				include_once( $deps );
+			if ( !defined( 'MW_COMPILED' ) ) {
+				// Preload base classes to work around APC/PHP5 bug
+				$deps = "{$wgStyleDirectory}/{$skinName}.deps.php";
+				if ( file_exists( $deps ) ) {
+					include_once( $deps );
+				}
+				require_once( "{$wgStyleDirectory}/{$skinName}.php" );
 			}
-			require_once( "{$wgStyleDirectory}/{$skinName}.php" );
 
 			# Check if we got if not failback to default skin
-			if ( !class_exists( $className ) ) {
+			if ( !MWInit::classExists( $className ) ) {
 				# DO NOT die if the class isn't found. This breaks maintenance
 				# scripts and can cause a user account to be unrecoverable
 				# except by SQL manipulation if a previously valid skin name
 				# is no longer valid.
 				wfDebug( "Skin class does not exist: $className\n" );
 				$className = 'SkinVector';
-				require_once( "{$wgStyleDirectory}/Vector.php" );
+				if ( !defined( 'MW_COMPILED' ) ) {
+					require_once( "{$wgStyleDirectory}/Vector.php" );
+				}
 			}
 		}
 		$skin = new $className;
diff --git a/includes/Title.php b/includes/Title.php
index c5fd079c22..051998c0ef 100644
--- a/includes/Title.php
+++ b/includes/Title.php
@@ -4,15 +4,6 @@
  * @file
  */
 
-/**
- * @todo:  determine if it is really necessary to load this.  Appears to be left over from pre-autoloader versions, and
- *   is only really needed to provide access to constant UTF8_REPLACEMENT, which actually resides in UtfNormalDefines.php
- *   and is loaded by UtfNormalUtil.php, which is loaded by UtfNormal.php.
- */
-if ( !class_exists( 'UtfNormal' ) ) {
-	require_once( dirname( __FILE__ ) . '/normal/UtfNormal.php' );
-}
-
 /**
  * @deprecated This used to be a define, but was moved to
  * Title::GAID_FOR_UPDATE in 1.17. This will probably be removed in 1.18
diff --git a/includes/User.php b/includes/User.php
index 968a830056..3ffd4a9276 100644
--- a/includes/User.php
+++ b/includes/User.php
@@ -1333,7 +1333,7 @@ class User {
 				if( $count > $max ) {
 					wfDebug( __METHOD__ . ": tripped! $key at $count $summary\n" );
 					if( $wgRateLimitLog ) {
-						@error_log( wfTimestamp( TS_MW ) . ' ' . wfWikiID() . ': ' . $this->getName() . " tripped $key at $count $summary\n", 3, $wgRateLimitLog );
+						@file_put_contents( $wgRateLimitLog, wfTimestamp( TS_MW ) . ' ' . wfWikiID() . ': ' . $this->getName() . " tripped $key at $count $summary\n", FILE_APPEND );
 					}
 					$triggered = true;
 				} else {
diff --git a/includes/WebStart.php b/includes/WebStart.php
index 2c21cdb2db..4a40343e31 100644
--- a/includes/WebStart.php
+++ b/includes/WebStart.php
@@ -8,6 +8,20 @@
  * @file
  */
 
+/**
+ * Detect compiled mode by looking for a function that only exists if compiled 
+ * in. Note that we can't use function_exists(), because it is terribly broken 
+ * under HipHop due to the "volatile" feature.
+ */
+function wfDetectCompiledMode() {
+	try {
+		$r = new ReflectionFunction( 'wfHipHopCompilerVersion' );
+	} catch ( ReflectionException $e ) {
+		$r = false;
+	}
+	return $r !== false;
+}
+
 # Protect against register_globals
 # This must be done before any globals are set by the code
 if ( ini_get( 'register_globals' ) ) {
@@ -67,40 +81,51 @@ if ( $IP === false ) {
 	$IP = realpath( '.' );
 }
 
-
-# Start profiler
-if( file_exists("$IP/StartProfiler.php") ) {
-	require_once( "$IP/StartProfiler.php" );
-} else {
-	require_once( "$IP/includes/ProfilerStub.php" );
+if ( wfDetectCompiledMode() ) {
+	define( 'MW_COMPILED', 1 );
 }
-wfProfileIn( 'WebStart.php-conf' );
 
-# Load up some global defines.
-require_once( "$IP/includes/Defines.php" );
-
-# Check for PHP 5
-if ( !function_exists( 'version_compare' ) 
-	|| version_compare( phpversion(), '5.0.0' ) < 0
-) {
-	define( 'MW_PHP4', '1' );
-	require( "$IP/includes/DefaultSettings.php" );
-	require( "$IP/includes/templates/PHP4.php" );
-	exit;
+if ( !defined( 'MW_COMPILED' ) ) {
+	# Get MWInit class
+	require_once( "$IP/includes/Init.php" );
+
+	# Start profiler
+	# FIXME: rewrite wfProfileIn/wfProfileOut so that they can work in compiled mode
+	if ( file_exists( "$IP/StartProfiler.php" ) ) {
+		require_once( "$IP/StartProfiler.php" );
+	} else {
+		require_once( "$IP/includes/ProfilerStub.php" );
+	}
+
+	# Load up some global defines.
+	require_once( "$IP/includes/Defines.php" );
+
+	# Check for PHP 5
+	if ( !function_exists( 'version_compare' ) 
+		|| version_compare( phpversion(), '5.0.0' ) < 0
+	) {
+		define( 'MW_PHP4', '1' );
+		require( "$IP/includes/DefaultSettings.php" );
+		require( "$IP/includes/templates/PHP4.php" );
+		exit;
+	}
+
+	# Start the autoloader, so that extensions can derive classes from core files
+	require_once( "$IP/includes/AutoLoader.php" );
 }
 
-# Start the autoloader, so that extensions can derive classes from core files
-require_once( "$IP/includes/AutoLoader.php" );
+wfProfileIn( 'WebStart.php-conf' );
+
 # Load default settings
-require_once( "$IP/includes/DefaultSettings.php" );
+require_once( MWInit::compiledPath( "includes/DefaultSettings.php" ) );
 
 if ( defined( 'MW_CONFIG_CALLBACK' ) ) {
 	# Use a callback function to configure MediaWiki
 	MWFunction::call( MW_CONFIG_CALLBACK );
-	
 } else {
-	if ( !defined('MW_CONFIG_FILE') )
-		define('MW_CONFIG_FILE', "$IP/LocalSettings.php");
+	if ( !defined( 'MW_CONFIG_FILE' ) ) {
+		define('MW_CONFIG_FILE', MWInit::interpretedPath( 'LocalSettings.php' ) );
+	}
 	
 	# LocalSettings.php is the per site customization file. If it does not exist
 	# the wiki installer needs to be launched or the generated file uploaded to
@@ -115,7 +140,7 @@ if ( defined( 'MW_CONFIG_CALLBACK' ) ) {
 }
 
 if ( $wgEnableSelenium ) {
-	require_once( "$IP/includes/SeleniumWebSettings.php" );
+	require_once( MWInit::compiledPath( "includes/SeleniumWebSettings.php" ) );
 }
 
 wfProfileOut( 'WebStart.php-conf' );
@@ -126,12 +151,14 @@ wfProfileIn( 'WebStart.php-ob_start' );
 # that would cause us to potentially mix gzip and non-gzip output, creating a
 # big mess.
 if ( !defined( 'MW_NO_OUTPUT_BUFFER' ) && ob_get_level() == 0 ) {
-	require_once( "$IP/includes/OutputHandler.php" );
+	if ( !defined( 'MW_COMPILED' ) ) {
+		require_once( "$IP/includes/OutputHandler.php" );
+	}
 	ob_start( 'wfOutputHandler' );
 }
 wfProfileOut( 'WebStart.php-ob_start' );
 
 if ( !defined( 'MW_NO_SETUP' ) ) {
-	require_once( "$IP/includes/Setup.php" );
+	require_once( MWInit::compiledPath( "includes/Setup.php" ) );
 }
 
diff --git a/includes/installer/WebInstaller.php b/includes/installer/WebInstaller.php
index 493e81f193..0d007260c3 100644
--- a/includes/installer/WebInstaller.php
+++ b/includes/installer/WebInstaller.php
@@ -408,9 +408,6 @@ class WebInstaller extends Installer {
 	 * @return WebInstallerPage
 	 */
 	public function getPageByName( $pageName ) {
-		// Totally lame way to force autoload of WebInstallerPage.php
-		class_exists( 'WebInstallerPage' );
-
 		$pageClass = 'WebInstaller_' . $pageName;
 
 		return new $pageClass( $this );
diff --git a/includes/normal/UtfNormalDefines.php b/includes/normal/UtfNormalDefines.php
index d759c64c62..6c4d8b76d8 100644
--- a/includes/normal/UtfNormalDefines.php
+++ b/includes/normal/UtfNormalDefines.php
@@ -1,6 +1,10 @@
 <?php
 /**
- * Some constant definitions for the unicode normalization module
+ * Some constant definitions for the unicode normalization module.
+ *
+ * Note: these constants must all be resolvable at compile time by HipHop, 
+ * since this file will not be executed during request startup for a compiled
+ * MediaWiki.
  *
  * @file
  * @ingroup UtfNormal
diff --git a/includes/normal/UtfNormalUtil.php b/includes/normal/UtfNormalUtil.php
index 0c78e5ec04..2df55775e9 100644
--- a/includes/normal/UtfNormalUtil.php
+++ b/includes/normal/UtfNormalUtil.php
@@ -25,7 +25,7 @@
  * @ingroup UtfNormal
  */
 
-require_once dirname(__FILE__).'/UtfNormalDefines.php';
+require_once( MWInit::compiledPath( 'includes/normal/UtfNormalDefines.php' ) );
 
 /**
  * Return UTF-8 sequence for a given Unicode code point.
diff --git a/languages/Language.php b/languages/Language.php
index 1acc15012c..e92fdc0fb7 100644
--- a/languages/Language.php
+++ b/languages/Language.php
@@ -174,12 +174,14 @@ class Language {
 			$class = 'Language';
 		} else {
 			$class = 'Language' . str_replace( '-', '_', ucfirst( $code ) );
-			// Preload base classes to work around APC/PHP5 bug
-			if ( file_exists( "$IP/languages/classes/$class.deps.php" ) ) {
-				include_once( "$IP/languages/classes/$class.deps.php" );
-			}
-			if ( file_exists( "$IP/languages/classes/$class.php" ) ) {
-				include_once( "$IP/languages/classes/$class.php" );
+			if ( !defined( 'MW_COMPILED' ) ) {
+				// Preload base classes to work around APC/PHP5 bug
+				if ( file_exists( "$IP/languages/classes/$class.deps.php" ) ) {
+					include_once( "$IP/languages/classes/$class.deps.php" );
+				}
+				if ( file_exists( "$IP/languages/classes/$class.php" ) ) {
+					include_once( "$IP/languages/classes/$class.php" );
+				}
 			}
 		}
 
@@ -187,7 +189,7 @@ class Language {
 			throw new MWException( "Language fallback loop detected when creating class $class\n" );
 		}
 
-		if ( !class_exists( $class ) ) {
+		if ( !MWInit::classExists( $class ) ) {
 			$fallback = Language::getFallbackFor( $code );
 			++$recursionLevel;
 			$lang = Language::newFromCode( $fallback );
@@ -540,8 +542,14 @@ class Language {
 	 * If $customisedOnly is true, only returns codes with a messages file
 	 */
 	public static function getLanguageNames( $customisedOnly = false ) {
-		global $wgLanguageNames, $wgExtraLanguageNames;
-		$allNames = $wgExtraLanguageNames + $wgLanguageNames;
+		global $wgExtraLanguageNames;
+		static $coreLanguageNames;
+
+		if ( $coreLanguageNames === null ) {
+			include( MWInit::compiledPath( 'languages/Names.php' ) );
+		}
+
+		$allNames = $wgExtraLanguageNames + $coreLanguageNames;
 		if ( !$customisedOnly ) {
 			return $allNames;
 		}
diff --git a/languages/LanguageConverter.php b/languages/LanguageConverter.php
index 730191e881..42b9660771 100644
--- a/languages/LanguageConverter.php
+++ b/languages/LanguageConverter.php
@@ -67,12 +67,12 @@ class LanguageConverter {
 	public function __construct( $langobj, $maincode, $variants = array(),
 								$variantfallbacks = array(), $flags = array(),
 								$manualLevel = array() ) {
-		global $wgDisabledVariants, $wgLanguageNames;
+		global $wgDisabledVariants;
 		$this->mLangObj = $langobj;
 		$this->mMainLanguageCode = $maincode;
 		$this->mVariants = array_diff( $variants, $wgDisabledVariants );
 		$this->mVariantFallbacks = $variantfallbacks;
-		$this->mVariantNames = $wgLanguageNames;
+		$this->mVariantNames = Language::getLanguageNames();
 		$this->mCacheKey = wfMemcKey( 'conversiontables', $maincode );
 		$defaultflags = array(
 			// 'S' show converted text
diff --git a/languages/Names.php b/languages/Names.php
index e9153b5e0e..89027e7d02 100644
--- a/languages/Names.php
+++ b/languages/Names.php
@@ -5,7 +5,7 @@
   *
   * @ingroup Language
   */
-/* private */ $wgLanguageNames = array(
+/* private */ $coreLanguageNames = array(
 	'aa' => 'Qafár af',	# Afar
 	'ab' => 'Аҧсуа',	# Abkhaz, should possibly add ' бысжѡа'
 	'ace' => 'Acèh',	# Aceh
diff --git a/maintenance/Maintenance.php b/maintenance/Maintenance.php
index c238b9c12e..67c5faf035 100644
--- a/maintenance/Maintenance.php
+++ b/maintenance/Maintenance.php
@@ -139,7 +139,7 @@ abstract class Maintenance {
 		if( count( $bt ) !== 2 ) {
 			return false;
 		}
-		return ( $bt[1]['function'] == 'require_once' || $bt[1]['function'] == 'require' ) &&
+		return in_array( $bt[1]['function'], array( 'require_once', 'require', 'include' ) ) &&
 			$bt[0]['class'] == 'Maintenance' &&
 			$bt[0]['function'] == 'shouldExecute';
 	}
@@ -430,11 +430,11 @@ abstract class Maintenance {
 	 */
 	public function runChild( $maintClass, $classFile = null ) {
 		// Make sure the class is loaded first
-		if ( !class_exists( $maintClass ) ) {
+		if ( !MWInit::classExists( $maintClass ) ) {
 			if ( $classFile ) {
 				require_once( $classFile );
 			}
-			if ( !class_exists( $maintClass ) ) {
+			if ( !MWInit::classExists( $maintClass ) ) {
 				$this->error( "Cannot spawn child: $maintClass" );
 			}
 		}
@@ -456,7 +456,7 @@ abstract class Maintenance {
 		}
 
 		# Make sure we can handle script parameters
-		if ( !ini_get( 'register_argc_argv' ) ) {
+		if ( !function_exists( 'hphp_thread_set_warmup_enabled' ) && !ini_get( 'register_argc_argv' ) ) {
 			$this->error( 'Cannot get command line arguments, register_argc_argv is set to false', true );
 		}
 
diff --git a/maintenance/addwiki.php b/maintenance/addwiki.php
index bed2ad6940..5d0a54a68f 100644
--- a/maintenance/addwiki.php
+++ b/maintenance/addwiki.php
@@ -57,7 +57,7 @@ class AddWiki extends Maintenance {
 		$languageNames = Language::getLanguageNames();
 
 		if ( !isset( $languageNames[$lang] ) ) {
-			$this->error( "Language $lang not found in \$wgLanguageNames", true );
+			$this->error( "Language $lang not found in Names.php", true );
 		}
 		$name = $languageNames[$lang];
 
diff --git a/maintenance/doMaintenance.php b/maintenance/doMaintenance.php
index 0ed50bad75..437ca83c32 100644
--- a/maintenance/doMaintenance.php
+++ b/maintenance/doMaintenance.php
@@ -53,17 +53,35 @@ $maintenance->setup();
 // to $maintenance->mSelf. Keep that here for b/c
 $self = $maintenance->getName();
 
+// Detect compiled mode
+try {
+	$r = new ReflectionFunction( 'wfHipHopCompilerVersion' );
+} catch ( ReflectionException $e ) {
+	$r = false;
+}
+
+if ( $r ) {
+	define( 'MW_COMPILED', 1 );
+}
+
+# Get the MWInit class
+if ( !defined( 'MW_COMPILED' ) ) {
+	require_once( "$IP/includes/Init.php" );
+}
+
 # Setup the profiler
 global $IP;
-if ( file_exists( "$IP/StartProfiler.php" ) ) {
+if ( !defined( 'MW_COMPILED' ) && file_exists( "$IP/StartProfiler.php" ) ) {
 	require_once( "$IP/StartProfiler.php" );
 } else {
-	require_once( "$IP/includes/ProfilerStub.php" );
+	require_once( MWInit::compiledPath( 'includes/ProfilerStub.php' ) );
 }
 
 // Some other requires
-require_once( "$IP/includes/AutoLoader.php" );
-require_once( "$IP/includes/Defines.php" );
+if ( !defined( 'MW_COMPILED' ) ) {
+	require_once( "$IP/includes/AutoLoader.php" );
+	require_once( "$IP/includes/Defines.php" );
+}
 require_once( "$IP/includes/DefaultSettings.php" );
 
 if ( defined( 'MW_CONFIG_CALLBACK' ) ) {
@@ -77,10 +95,10 @@ if ( defined( 'MW_CONFIG_CALLBACK' ) ) {
 	global $cluster;
 	$wgWikiFarm = true;
 	$cluster = 'pmtpa';
-	require_once( "$IP/includes/SiteConfiguration.php" );
-	require( "$IP/wmf-config/wgConf.php" );
+	require_once( MWInit::compiledPath( 'includes/SiteConfiguration.php' ) );
+	require( MWInit::interpretedPath( 'wmf-config/wgConf.php' ) );
 	$maintenance->loadWikimediaSettings();
-	require( $IP . '/wmf-config/CommonSettings.php' );
+	require( MWInit::interpretedPath( '/wmf-config/CommonSettings.php' ) );
 } else {
 	require_once( $maintenance->loadSettings() );
 }
@@ -88,12 +106,12 @@ if ( defined( 'MW_CONFIG_CALLBACK' ) ) {
 if ( $maintenance->getDbType() === Maintenance::DB_ADMIN &&
 		is_readable( "$IP/AdminSettings.php" ) )
 {
-	require( "$IP/AdminSettings.php" );
+	require( MWInit::interpretedPath( 'AdminSettings.php' ) );
 }
 $maintenance->finalSetup();
 // Some last includes
-require_once( "$IP/includes/Setup.php" );
-require_once( "$IP/maintenance/install-utils.inc" );
+require_once( MWInit::compiledPath( 'includes/Setup.php' ) );
+require_once( MWInit::compiledPath( 'maintenance/install-utils.inc' ) );
 
 // Much much faster startup than creating a title object
 $wgTitle = null;
diff --git a/maintenance/eval.php b/maintenance/eval.php
index 9229ae2421..74143b5dd4 100644
--- a/maintenance/eval.php
+++ b/maintenance/eval.php
@@ -74,7 +74,7 @@ while ( ( $line = Maintenance::readconsole() ) !== false ) {
 		readline_write_history( $historyFile );
 	}
 	$val = eval( $line . ";" );
-	if ( is_null( $val ) ) {
+	if ( wfIsHipHop() || is_null( $val ) ) {
 		echo "\n";
 	} elseif ( is_string( $val ) || is_numeric( $val ) ) {
 		echo "$val\n";
diff --git a/maintenance/hiphop/file-list.small b/maintenance/hiphop/file-list.small
index 84acdda147..0f7a8ba0fe 100644
--- a/maintenance/hiphop/file-list.small
+++ b/maintenance/hiphop/file-list.small
@@ -158,6 +158,7 @@ includes/filerepo/NullRepo.php
 includes/filerepo/OldLocalFile.php
 includes/filerepo/RepoGroup.php
 includes/filerepo/UnregisteredLocalFile.php
+includes/installer/Installer.php
 includes/job/DoubleRedirectJob.php
 includes/job/EmaillingJob.php
 includes/job/EnotifNotifyJob.php
@@ -280,6 +281,8 @@ profileinfo.php
 redirect.php
 resources/Resources.php
 serialized/serialize.php
+skins/MonoBook.deps.php
+skins/MonoBook.php
 skins/Vector.deps.php
 skins/Vector.php
 thumb.php
-- 
2.20.1