From dcbe8e7dd3c4d54391d30add214e3c31ce381e34 Mon Sep 17 00:00:00 2001
From: Aaron Schulz <aaron@users.mediawiki.org>
Date: Thu, 8 Dec 2011 03:43:07 +0000
Subject: [PATCH] FU r101117: removed cURL thumb handler code and made
 thumb_handler.php a thin wrapper around thumb.php * Moved original URL
 fetching code and parameter extraction code to thumb.php * Made use of local
 repo URL and hash settings to avoid extra config code * This makes it easy to
 add hooks for extensions/config to alter behavoir (ExtractThumbParameters
 hook added) * Added FileRepo::getHashLevels()

---
 docs/hooks.txt                 |   4 +
 includes/filerepo/FileRepo.php |   9 ++
 thumb.config.sample            |  49 -------
 thumb.php                      | 122 +++++++++++++++--
 thumb_handler.php              | 237 ++-------------------------------
 5 files changed, 129 insertions(+), 292 deletions(-)
 delete mode 100644 thumb.config.sample

diff --git a/docs/hooks.txt b/docs/hooks.txt
index 8223ef17c7..dd9a9c1db5 100644
--- a/docs/hooks.txt
+++ b/docs/hooks.txt
@@ -867,6 +867,10 @@ $ip: The ip address of the user
 'ExtensionTypes': called when generating the extensions credits, use this to change the tables headers
 &$extTypes: associative array of extensions types
 
+'ExtractThumbParameters': called when extracting thumbnail parameters from a thumbnail file name
+$thumbname: the base name of the thumbnail file
+&$params: the currently extracted params (has source name, temp or archived zone)
+
 'FetchChangesList': When fetching the ChangesList derivative for
 a particular user
 $user: User the list is being fetched for
diff --git a/includes/filerepo/FileRepo.php b/includes/filerepo/FileRepo.php
index dbda0d1bfc..58b64e4f51 100644
--- a/includes/filerepo/FileRepo.php
+++ b/includes/filerepo/FileRepo.php
@@ -276,6 +276,15 @@ abstract class FileRepo {
 		}
 	}
 
+	/**
+	 * Get the number of hash directory levels
+	 *
+	 * @return integer
+	 */
+	function getHashLevels() {
+		return $this->hashLevels;
+	}
+
 	/**
 	 * Get a relative path including trailing slash, e.g. f/fa/
 	 * If the repo is not hashed, returns an empty string
diff --git a/thumb.config.sample b/thumb.config.sample
deleted file mode 100644
index 84445c1681..0000000000
--- a/thumb.config.sample
+++ /dev/null
@@ -1,49 +0,0 @@
-<?php
-/**
- * @cond file_level_code
- * This is not a valid entry point, perform no further processing unless THUMB_HANDLER is defined
- */
-if ( !defined( 'THUMB_HANDLER' ) ) {
-	echo "This file is part of MediaWiki and is not a valid entry point\n";
-	die( 1 );
-}
-
-/**
- * Sample configuration file for thumb-handler.php.
- * In order to use thumb-handler.php:
- * 1) Copy this file to thumb.config.php and modify the settings.
- * 2) The webserver must be setup to have thumb-handler.php as a 404 handler.
- *    This can be done in apache by editing .htaccess in the /thumb directory by adding:
- *    <IfModule rewrite_module>
- *          RewriteEngine on
- *          RewriteCond %{REQUEST_FILENAME} !-f
- *          RewriteCond %{REQUEST_FILENAME} !-d
- *          RewriteRule ^([^/]+/)?[0-9a-f]/[0-9a-f][0-9a-f]/[^/]+/[^/]+$ /path/to/thumb_handler.php [L]
- *    </IfModule>
- */
-
-$thgThumbUrlMatch = array(
-	# URL name of the server (e.g. "upload.wikipedia.org").
-	'server'        => 'http://localhost',
-	# URL fragment to the thumb/ directory
-	'dirFragment'   => 'MW_trunk/images/thumb',
-	# URL regex fragment correspond to the directory hashing of thumbnails.
-	# This must correspond to $wgLocalFileRepo['hashLevels'].
-	'hashFragment'  => '[0-9a-f]/[0-9a-f][0-9a-f]/' // 2-level directory hashing
-);
-
-$thgThumbCurlConfig = array(
-	# Optionally cURL to thumb.php instead of using it directly
-	'enabled' => false,
-	# The URL to thumb.php, accessible from the web server.
-	'url'     => 'http://localhost/MW_trunk/thumb.php',
-	# Optional proxy server to use to access thumb.php
-	'proxy'   => null,
-	# Timeout to use for cURL request to thumb.php.
-	# Leave it long enough to generate a ulimit timeout in ordinary
-	# cases, but short enough to avoid a local PHP timeout.
-	'timeout' => 53
-);
-
-# Custom functions for overriding aspects of thumb handling
-$thgThumbCallbacks = array();
diff --git a/thumb.php b/thumb.php
index 0bb0b60c34..e16d95b4ce 100644
--- a/thumb.php
+++ b/thumb.php
@@ -13,27 +13,69 @@ if ( isset( $_SERVER['MW_COMPILED'] ) ) {
 	require ( dirname( __FILE__ ) . '/includes/WebStart.php' );
 }
 
-$wgTrivialMimeDetection = true; //don't use fancy mime detection, just check the file extension for jpg/gif/png.
+// Don't use fancy mime detection, just check the file extension for jpg/gif/png
+$wgTrivialMimeDetection = true;
 
-wfThumbMain();
+if ( defined( 'THUMB_HANDLER' ) ) {
+	// Called from thumb_handler.php via 404; extract params from the URI...
+	wfThumbHandle404();
+} else {
+	// Called directly, use $_REQUEST params
+	wfThumbHandleRequest();
+}
 wfLogProfilingData();
 
 //--------------------------------------------------------------------------
 
-function wfThumbMain() {
-	wfProfileIn( __METHOD__ );
+/**
+ * Handle a thumbnail request via query parameters
+ *
+ * @return void
+ */
+function wfThumbHandleRequest() {
+	$params = get_magic_quotes_gpc()
+		? array_map( 'stripslashes', $_REQUEST )
+		: $_REQUEST;
 
-	$headers = array();
+	wfStreamThumb( $params ); // stream the thumbnail
+}
 
-	// Get input parameters
-	if ( defined( 'THUMB_HANDLER' ) ) {
-		$params = $_REQUEST; // called from thumb_handler.php
+/**
+ * Handle a thumbnail request via thumbnail file URL
+ *
+ * @return void
+ */
+function wfThumbHandle404() {
+	# lighttpd puts the original request in REQUEST_URI, while
+	# sjs sets that to the 404 handler, and puts the original
+	# request in REDIRECT_URL.
+	if ( isset( $_SERVER['REDIRECT_URL'] ) ) {
+		# The URL is un-encoded, so put it back how it was.
+		$uri = str_replace( "%2F", "/", urlencode( $_SERVER['REDIRECT_URL'] ) );
 	} else {
-		$params = get_magic_quotes_gpc()
-			? array_map( 'stripslashes', $_REQUEST )
-			: $_REQUEST;
+		$uri = $_SERVER['REQUEST_URI'];
 	}
 
+	$params = wfExtractThumbParams( $uri ); // basic wiki URL param extracting
+	if ( $params == null ) {
+		wfThumbError( 404, 'The source file for the specified thumbnail does not exist.' );
+		return;
+	}
+
+	wfStreamThumb( $params ); // stream the thumbnail
+}
+
+/**
+ * Stream a thumbnail specified by parameters
+ *
+ * @param $params Array
+ * @return void
+ */
+function wfStreamThumb( array $params ) {
+	wfProfileIn( __METHOD__ );
+
+	$headers = array(); // HTTP headers to send
+
 	$fileName = isset( $params['f'] ) ? $params['f'] : '';
 	unset( $params['f'] );
 
@@ -64,7 +106,7 @@ function wfThumbMain() {
 			return;
 		}
 		$title = Title::makeTitleSafe( NS_FILE, $bits[1] );
-		if ( is_null( $title ) ) {
+		if ( !$title ) {
 			wfThumbError( 404, wfMsg( 'badtitletext' ) );
 			wfProfileOut( __METHOD__ );
 			return;
@@ -169,11 +211,63 @@ function wfThumbMain() {
 }
 
 /**
- * @param $status
- * @param $msg
+ * Extract the required params for thumb.php from the thumbnail request URI.
+ * At least 'width' and 'f' should be set if the result is an array.
+ *
+ * @param $uri String Thumbnail request URI
+ * @return Array|null associative params array or null
+ */
+function wfExtractThumbParams( $uri ) {
+	$repo = RepoGroup::singleton()->getLocalRepo();
+
+	$hashDirRegex = $subdirRegex = '';
+	for ( $i = 0; $i < $repo->getHashLevels(); $i++ ) {
+		$subdirRegex .= '[0-9a-f]';
+		$hashDirRegex .= "$subdirRegex/";
+	}
+	$zoneUrlRegex = preg_quote( $repo->getZoneUrl( 'thumb' ) );
+
+	$thumbUrlRegex = "!^$zoneUrlRegex(/archive|/temp|)/$hashDirRegex([^/]*)/([^/]*)$!";
+
+	// Check if this is a valid looking thumbnail request...
+	if ( preg_match( $thumbUrlRegex, $uri, $matches ) ) {
+		list( /* all */, $archOrTemp, $filename, $thumbname ) = $matches;
+
+		$params = array( 'f' => $filename );
+		if ( $archOrTemp == '/archive' ) {
+			$params['archived'] = 1;
+		} elseif ( $archOrTemp == '/temp' ) {
+			$params['temp'] = 1;
+		}
+
+		// Check if the parameters can be extracted from the thumbnail name...
+		// @TODO: remove 'page' stuff and make ProofreadPage handle it via hook.
+		if ( preg_match( '!^(page(\d*)-)*(\d*)px-[^/]*$!', $thumbname, $matches ) ) {
+			list( /* all */, $pagefull, $pagenum, $size ) = $matches;
+			$params['width'] = $size;
+			if ( $pagenum ) {
+				$params['page'] = $pagenum;
+			}
+			return $params; // valid thumbnail URL
+		// Hooks return false if they manage to *resolve* the parameters
+		} elseif ( !wfRunHooks( 'ExtractThumbParameters', array( $thumbname, &$params ) ) ) {
+			return $params; // valid thumbnail URL (via extension or config)
+		}
+	}
+
+	return null; // not a valid thumbnail URL
+}
+
+/**
+ * Output a thumbnail generation error message
+ *
+ * @param $status integer
+ * @param $msg string
+ * @return void
  */
 function wfThumbError( $status, $msg ) {
 	global $wgShowHostnames;
+
 	header( 'Cache-Control: no-cache' );
 	header( 'Content-Type: text/html; charset=utf-8' );
 	if ( $status == 404 ) {
diff --git a/thumb_handler.php b/thumb_handler.php
index 073a87929e..39a3e7c057 100644
--- a/thumb_handler.php
+++ b/thumb_handler.php
@@ -3,239 +3,18 @@
 # Valid web server entry point
 define( 'THUMB_HANDLER', true );
 
-# Load thumb-handler configuration. Avoids WebStart.php for performance.
-if ( !file_exists( dirname( __FILE__ ) . "/thumb.config.php" ) ) {
-	die( "thumb_handler.php is not enabled for this wiki.\n" );
-}
-require( dirname( __FILE__ ) . "/thumb.config.php" );
-
-# Execute thumb.php if not handled via cURL
-if ( wfHandleThumb404Main() === 'wfThumbMain' ) {
+if ( $_SERVER['REQUEST_URI'] === $_SERVER['SCRIPT_NAME'] ) {
+	# Directly requesting this script is not a use case.
+	# Instead of giving a thumbnail error, give a generic 404.
+	wfDisplay404Error(); // go away, nothing to see here
+} else {
+	# Execute thumb.php, having set THUMB_HANDLER so that
+	# it knows to extract params from a thumbnail file URL.
 	require( dirname( __FILE__ ) . '/thumb.php' );
 }
 
-function wfHandleThumb404Main() {
-	global $thgThumbCallbacks, $thgThumbCurlConfig;
-
-	# lighttpd puts the original request in REQUEST_URI, while
-	# sjs sets that to the 404 handler, and puts the original
-	# request in REDIRECT_URL.
-	if ( isset( $_SERVER['REDIRECT_URL'] ) ) {
-		# The URL is un-encoded, so put it back how it was.
-		$uri = str_replace( "%2F", "/", urlencode( $_SERVER['REDIRECT_URL'] ) );
-	} else {
-		$uri = $_SERVER['REQUEST_URI'];
-	}
-
-	# Extract thumb.php params from the URI...
-	if ( isset( $thgThumbCallbacks['extractParams'] )
-		&& is_callable( $thgThumbCallbacks['extractParams'] ) ) // overridden by configuration?
-	{
-		$params = call_user_func_array( $thgThumbCallbacks['extractParams'], array( $uri ) );
-	} else {
-		$params = wfExtractThumbParams( $uri ); // basic wiki URL param extracting
-	}
-
-	# Show 404 error if this is not a valid thumb request...
-	if ( !is_array( $params ) ) {
-		header( 'X-Debug: no regex match' ); // useful for debugging
-		if ( isset( $thgThumbCallbacks['error404'] )
-			&& is_callable( $thgThumbCallbacks['error404'] ) ) // overridden by configuration?
-		{
-			call_user_func( $thgThumbCallbacks['error404'] );
-		} else {
-			wfDisplay404Error(); // standard 404 message
-		}
-		return;
-	}
-
-	# Obtain and stream the thumbnail or setup for wfThumbMain() call...
-	if ( $thgThumbCurlConfig['enabled'] ) {
-		wfStreamThumbViaCurl( $params, $uri );
-		return true; // done
-	} else {
-		$_REQUEST = $params; // pass params to thumb.php
-		return 'wfThumbMain';
-	}
-}
-
-/**
- * Extract the required params for thumb.php from the thumbnail request URI.
- * At least 'width' and 'f' should be set if the result is an array.
- *
- * @param $uri String Thumbnail request URI
- * @return Array|null associative params array or null
- */
-function wfExtractThumbParams( $uri ) {
-	global $thgThumbUrlMatch;
-
-	$thumbRegex = '!^(?:' . preg_quote( $thgThumbUrlMatch['server'] ) . ')?/' .
-		preg_quote( $thgThumbUrlMatch['dirFragment'] ) . '(/archive|/temp|)/' .
-		$thgThumbUrlMatch['hashFragment'] . '([^/]*)/(page(\d*)-)*(\d*)px-[^/]*$!';
-
-	if ( preg_match( $thumbRegex, $uri, $matches ) ) {
-		list( $all, $archOrTemp, $filename, $pagefull, $pagenum, $size ) = $matches;
-		$params = array( 'f' => $filename, 'width' => $size );
-		if ( $pagenum ) {
-			$params['page'] = $pagenum;
-		}
-		if ( $archOrTemp == '/archive' ) {
-			$params['archived'] = 1;
-		} elseif ( $archOrTemp == '/temp' ) {
-			$params['temp'] = 1;
-		}
-	} else {
-		$params = null; // not a valid thumbnail URL
-	}
-
-	return $params;
-}
-
-/**
- * cURL to thumb.php and stream back the resulting file or give an error message.
- *
- * @param $params Array Parameters to thumb.php
- * @param $uri String Thumbnail request URI
- * @return void
- */
-function wfStreamThumbViaCurl( array $params, $uri ) {
-	global $thgThumbCallbacks, $thgThumbCurlConfig;
-
-	# Check any backend caches for the thumbnail...
-	if ( isset( $thgThumbCallbacks['checkCache'] )
-		&& is_callable( $thgThumbCallbacks['checkCache'] ) )
-	{
-		if ( call_user_func_array( $thgThumbCallbacks['checkCache'], array( $uri, $params ) ) ) {
-			return; // file streamed from backend thumb cache
-		}
-	}
-
-	if ( !extension_loaded( 'curl' ) ) {
-		die( "cURL is not enabled for PHP on this wiki.\n" ); // sanity
-	}
-
-	# Build up the request URL to use with CURL...
-	$reqURL = $thgThumbCurlConfig['url'] . '?';
-	$first = true;
-	foreach ( $params as $name => $value ) {
-		if ( $first ) {
-			$first = false;
-		} else {
-			$reqURL .= '&';
-		}
-		$reqURL .= "$name=$value"; // Note: value is already urlencoded
-	}
-
-	# Set relevant HTTP headers...
-	$headers = array();
-	$headers[] = "X-Original-URI: " . str_replace( "\n", '', $uri );
-	if ( isset( $thgThumbCallbacks['curlHeaders'] )
-		&& is_callable( $thgThumbCallbacks['curlHeaders'] ) )
-	{
-		# Add on any custom headers (like XFF)
-		call_user_func_array( $thgThumbCallbacks['curlHeaders'], array( &$headers ) );
-	}
-
-	# Pass through some other headers...
-	$passThrough = array( 'If-Modified-Since', 'Referer', 'User-Agent' );
-	foreach ( $passThrough as $headerName ) {
-		$serverVarName = 'HTTP_' . str_replace( '-', '_', strtoupper( $headerName ) );
-		if ( !empty( $_SERVER[$serverVarName] ) ) {
-			$headers[] = $headerName . ': ' . 
-				str_replace( "\n", '', $_SERVER[$serverVarName] );
-		}
-	}
-
-	$ch = curl_init( $reqURL );
-	if ( $thgThumbCurlConfig['proxy'] ) {
-		curl_setopt( $ch, CURLOPT_PROXY, $thgThumbCurlConfig['proxy'] );
-	}
-
-	curl_setopt( $ch, CURLOPT_HTTPHEADER, $headers );
-	curl_setopt( $ch, CURLOPT_RETURNTRANSFER, true );
-	curl_setopt( $ch, CURLOPT_TIMEOUT, $thgThumbCurlConfig['timeout'] );
-
-	# Actually make the request
-	$text = curl_exec( $ch );
-
-	# Send it on to the client...
-	$errno = curl_errno( $ch );
-	$contentType = curl_getinfo( $ch, CURLINFO_CONTENT_TYPE );
-	$httpCode = curl_getinfo( $ch, CURLINFO_HTTP_CODE );
-	if ( $errno ) {
-		header( 'HTTP/1.1 500 Internal server error' );
-		header( 'Cache-Control: no-cache' );
-		$contentType = 'text/html';
-		$text = wfCurlErrorText( $ch );
-	} elseif ( $httpCode == 304 ) { // OK
-		header( 'HTTP/1.1 304 Not modified' );
-		$contentType = '';
-		$text = '';
-	} elseif ( strval( $text ) == '' ) {
-		header( 'HTTP/1.1 500 Internal server error' );
-		header( 'Cache-Control: no-cache' );
-		$contentType = 'text/html';
-		$text = wfCurlEmptyText( $ch );
-	} elseif ( $httpCode == 404 ) {
-		header( 'HTTP/1.1 404 Not found' );
-		header( 'Cache-Control: s-maxage=300, must-revalidate, max-age=0' );
-	} elseif ( $httpCode != 200 || substr( $contentType, 0, 9 ) == 'text/html' ) {
-		# Error message, suppress cache
-		header( 'HTTP/1.1 500 Internal server error' );
-		header( 'Cache-Control: no-cache' );
-	} else {
-		# OK thumbnail; save to any backend caches...
-		if ( isset( $thgThumbCallbacks['fillCache'] )
-			&& is_callable( $thgThumbCallbacks['fillCache'] ) )
-		{
-			call_user_func_array( $thgThumbCallbacks['fillCache'], array( $uri, $text ) );
-		}
-	}
-
-	if ( !$contentType ) {
-		header( 'Content-Type:' );
-	} else {
-		header( "Content-Type: $contentType" );
-	}
-
-	print $text; // thumb data or error text
-
-	curl_close( $ch );
-}
-
-/**
- * Get error message HTML for when the cURL response is an error.
- *
- * @param $ch cURL handle
- * @return string
- */
-function wfCurlErrorText( $ch ) {
-	$error = htmlspecialchars( curl_error( $ch ) );
-	return <<<EOT
-<html>
-<head><title>Thumbnail error</title></head>
-<body>Error retrieving thumbnail from scaling server: $error</body>
-</html>
-EOT;
-}
-
-/**
- * Get error message HTML for when the cURL response is empty.
- *
- * @param $ch cURL handle
- * @return string
- */
-function wfCurlEmptyText( $ch ) {
-	return <<<EOT
-<html>
-<head><title>Thumbnail error</title></head>
-<body>Error retrieving thumbnail from scaling server: empty response</body>
-</html>
-EOT;
-}
-
 /**
- * Print out a generic 404 error message.
+ * Print out a generic 404 error message
  *
  * @return void
  */
-- 
2.20.1