- Patch #284899 by c960657, voxpelli, mattconnolly: fixed Drupal url problem with clean urls.

.htaccess

@@ -88,11 +88,12 @@ DirectoryIndex index.php index.html index.htm
   # uncomment the following line:
   # RewriteBase /
 
-  # Rewrite URLs of the form 'x' to the form 'index.php?q=x'.
+  # Pass all requests not referring directly to files in the filesystem to
+  # index.php. Clean URLs are handled in drupal_environment_initialize().
   RewriteCond %{REQUEST_FILENAME} !-f
   RewriteCond %{REQUEST_FILENAME} !-d
   RewriteCond %{REQUEST_URI} !=/favicon.ico
-  RewriteRule ^(.*)$ index.php?q=$1 [L,QSA]
+  RewriteRule ^ index.php [L]
 </IfModule>
 
 # $Id$

includes/bootstrap.inc

@@ -492,6 +492,13 @@ function drupal_environment_initialize() {
     $_SERVER['HTTP_HOST'] = '';
   }
 
+  // When clean URLs are enabled, emulate ?q=foo/bar using REQUEST_URI. It is
+  // not possible to append the query string using mod_rewrite without the B
+  // flag (this was added in Apache 2.2.8), because mod_rewrite unescapes the
+  // path before passing it on to PHP. This is a problem when the path contains
+  // e.g. "&" or "%" that have special meanings in URLs and must be encoded.
+  $_GET['q'] = request_path();
+
   // Enforce E_ALL, but allow users to set levels not part of E_ALL.
   error_reporting(E_ALL | error_reporting());
 
@@ -559,8 +566,8 @@ function drupal_settings_initialize() {
 
     // $_SERVER['SCRIPT_NAME'] can, in contrast to $_SERVER['PHP_SELF'], not
     // be modified by a visitor.
-    if ($dir = trim(dirname($_SERVER['SCRIPT_NAME']), '\,/')) {
-      $base_path = "/$dir";
+    if ($dir = rtrim(dirname($_SERVER['SCRIPT_NAME']), '\/')) {
+      $base_path = $dir;
       $base_url .= $base_path;
       $base_path .= '/';
     }
@@ -1858,6 +1865,50 @@ function language_default($property = NULL) {
   return $property ? $language->$property : $language;
 }
 
+/**
+ * Returns the requested URL path of the page being viewed.
+ *
+ * Examples:
+ * - http://example.com/node/306 returns "node/306".
+ * - http://example.com/drupalfolder/node/306 returns "node/306" while
+ *   base_path() returns "/drupalfolder/".
+ * - http://example.com/path/alias (which is a path alias for node/306) returns
+ *   "path/alias" as opposed to the internal path.
+ *
+ * @return
+ *   The requested Drupal URL path.
+ *
+ * @see current_path()
+ */
+function request_path() {
+  static $path;
+
+  if (isset($path)) {
+    return $path;
+  }
+
+  if (isset($_GET['q'])) {
+    // This is a request with a ?q=foo/bar query string. $_GET['q'] is
+    // overwritten in drupal_path_initialize(), but request_path() is called
+    // very early in the bootstrap process, so the original value is saved in
+    // $path and returned in later calls.
+    $path = $_GET['q'];
+  }
+  elseif (isset($_SERVER['REQUEST_URI'])) {
+    // This is a request using a clean URL. Extract the path from REQUEST_URI.
+    $request_path = strtok($_SERVER['REQUEST_URI'], '?');
+    $base_path_len = strlen(rtrim(dirname($_SERVER['SCRIPT_NAME']), '\/'));
+    // Unescape and strip $base_path prefix, leaving q without a leading slash.
+    $path = substr(urldecode($request_path), $base_path_len + 1);
+  }
+  else {
+    // This is the front page.
+    $path = '';
+  }
+
+  return $path;
+}
+
 /**
  * If Drupal is behind a reverse proxy, we use the X-Forwarded-For header
  * instead of $_SERVER['REMOTE_ADDR'], which would be the IP address of

includes/common.inc

@@ -501,7 +501,6 @@ function drupal_http_build_query(array $query, $parent = '') {
     }
     else {
       // For better readability of paths in query strings, we decode slashes.
-      // @see drupal_encode_path()
       $params[] = $key . '=' . str_replace('%2F', '/', rawurlencode($value));
     }
   }
@@ -623,38 +622,18 @@ function drupal_parse_url($url) {
 }
 
 /**
- * Encode a path for usage in a URL.
+ * Encodes a Drupal path for use in a URL.
  *
- * Wrapper around rawurlencode() which avoids Apache quirks. Should be used when
- * placing arbitrary data into the path component of an URL.
+ * For aesthetic reasons slashes are not escaped.
  *
- * Do not use this function to pass a path to url(). url() properly handles
- * and encodes paths internally.
- * This function should only be used on paths, not on query string arguments.
- * Otherwise, unwanted double encoding will occur.
- *
- * Notes:
- * - For aesthetic reasons, we do not escape slashes. This also avoids a 'feature'
- *   in Apache where it 404s on any path containing '%2F'.
- * - mod_rewrite unescapes %-encoded ampersands, hashes, and slashes when clean
- *   URLs are used, which are interpreted as delimiters by PHP. These
- *   characters are double escaped so PHP will still see the encoded version.
- * - With clean URLs, Apache changes '//' to '/', so every second slash is
- *   double escaped.
+ * Note that url() takes care of calling this function, so a path passed to that
+ * function should not be encoded in advance.
  *
  * @param $path
- *   The URL path component to encode.
+ *   The Drupal path to encode.
  */
 function drupal_encode_path($path) {
-  if (!empty($GLOBALS['conf']['clean_url'])) {
-    return str_replace(array('%2F', '%26', '%23', '//'),
-                       array('/', '%2526', '%2523', '/%252F'),
-                       rawurlencode($path)
-    );
-  }
-  else {
-    return str_replace('%2F', '/', rawurlencode($path));
-  }
+  return str_replace('%2F', '/', rawurlencode($path));
 }
 
 /**

includes/file.inc

@@ -901,6 +901,10 @@ function file_unmunge_filename($filename) {
  *   of $basename.
  */
 function file_create_filename($basename, $directory) {
+  // Strip control characters (ASCII value < 32). Though these are allowed in
+  // some filesystems, not many applications handle them well.
+  $basename = preg_replace('/[\x00-\x1F]/u', '_', $basename);
+
   // A URI or path may already have a trailing slash or look like "public://".
   if (substr($directory, -1) == '/') {
     $separator = '';

includes/path.inc

@@ -384,6 +384,8 @@ function drupal_match_path($path, $patterns) {
  *
  * @return
  *   The current Drupal URL path.
+ *
+ * @see request_path()
  */
 function current_path() {
   return $_GET['q'];

includes/stream_wrappers.inc

@@ -639,7 +639,7 @@ public function getDirectoryPath() {
    */
   function getExternalUrl() {
     $path = str_replace('\\', '/', file_uri_target($this->uri));
-    return $GLOBALS['base_url'] . '/' . self::getDirectoryPath() . '/' . $path;
+    return $GLOBALS['base_url'] . '/' . self::getDirectoryPath() . '/' . drupal_encode_path($path);
   }
 }
 

misc/autocomplete.js

@@ -276,7 +276,7 @@ Drupal.ACDB.prototype.search = function (searchString) {
     // Ajax GET request for autocompletion.
     $.ajax({
       type: 'GET',
-      url: db.uri + '/' + Drupal.encodePath(searchString),
+      url: db.uri + '/' + encodeURIComponent(searchString),
       dataType: 'json',
       success: function (matches) {
         if (typeof matches.status == 'undefined' || matches.status != 0) {

misc/drupal.js

@@ -282,14 +282,13 @@ Drupal.unfreezeHeight = function () {
 };
 
 /**
- * Wrapper around encodeURIComponent() which avoids Apache quirks (equivalent of
- * drupal_encode_path() in PHP). This function should only be used on paths, not
- * on query string arguments.
+ * Encodes a Drupal path for use in a URL.
+ *
+ * For aesthetic reasons slashes are not escaped.
  */
 Drupal.encodePath = function (item, uri) {
   uri = uri || location.href;
-  item = encodeURIComponent(item).replace(/%2F/g, '/');
-  return (uri.indexOf('?q=') != -1) ? item : item.replace(/%26/g, '%2526').replace(/%23/g, '%2523').replace(/\/\//g, '/%252F');
+  return encodeURIComponent(item).replace(/%2F/g, '/');
 };
 
 /**

modules/path/path.test

@@ -66,11 +66,13 @@ class PathTestCase extends DrupalWebTestCase {
     $this->assertText($node1->title, 'Alias works.');
     $this->assertResponse(200);
 
-    // Change alias.
+    // Change alias to one containing "exotic" characters.
     $pid = $this->getPID($edit['alias']);
 
     $previous = $edit['alias'];
-    $edit['alias'] = $this->randomName(8);
+    $edit['alias'] = "- ._~!$'\"()*@[]?&+%#,;=:" . // "Special" ASCII characters.
+      "%23%25%26%2B%2F%3F" . // Characters that look like a percent-escaped string.
+      "éøïвβ中國書۞"; // Characters from various non-ASCII alphabets.
     $this->drupalPost('admin/config/search/path/edit/' . $pid, $edit, t('Save'));
 
     // Confirm that the alias works.
@@ -122,9 +124,11 @@ class PathTestCase extends DrupalWebTestCase {
     $this->assertText($node1->title, 'Alias works.');
     $this->assertResponse(200);
 
-    // Change alias.
+    // Change alias to one containing "exotic" characters.
     $previous = $edit['path[alias]'];
-    $edit['path[alias]'] = $this->randomName(8);
+    $edit['path[alias]'] = "- ._~!$'\"()*@[]?&+%#,;=:" . // "Special" ASCII characters.
+      "%23%25%26%2B%2F%3F" . // Characters that look like a percent-escaped string.
+      "éøïвβ中國書۞"; // Characters from various non-ASCII alphabets.
     $this->drupalPost('node/' . $node1->nid . '/edit', $edit, t('Save'));
 
     // Confirm that the alias works.

modules/simpletest/tests/file.test

@@ -1888,6 +1888,8 @@ class FileDownloadTest extends FileTestCase {
 
   function setUp() {
     parent::setUp('file_test');
+    // Clear out any hook calls.
+    file_test_reset();
   }
 
   /**
@@ -1937,6 +1939,70 @@ class FileDownloadTest extends FileTestCase {
     $this->drupalHead($url);
     $this->assertResponse(404, t('Correctly returned 404 response for a non-existent file.'));
   }
+
+  /**
+   * Test file_create_url().
+   */
+  function testFileCreateUrl() {
+    global $base_url;
+
+    $basename = " -._~!$'\"()*@[]?&+%#,;=:\n\x00" . // "Special" ASCII characters.
+      "%23%25%26%2B%2F%3F" . // Characters that look like a percent-escaped string.
+      "éøïвβ中國書۞"; // Characters from various non-ASCII alphabets.
+    $basename_encoded = '%20-._%7E%21%24%27%22%28%29%2A%40%5B%5D%3F%26%2B%25%23%2C%3B%3D%3A__' .
+      '%2523%2525%2526%252B%252F%253F' .
+      '%C3%A9%C3%B8%C3%AF%D0%B2%CE%B2%E4%B8%AD%E5%9C%8B%E6%9B%B8%DB%9E';
+
+    $this->checkUrl('public', '', $basename, $base_url . '/' . file_directory_path() . '/' . $basename_encoded);
+    $this->checkUrl('private', '', $basename, $base_url . '/system/files/' . $basename_encoded);
+    $this->checkUrl('private', '', $basename, $base_url . '/?q=system/files/' . $basename_encoded, '0');
+  }
+
+  /**
+   * Download a file from the URL generated by file_create_url().
+   *
+   * Create a file with the specified scheme, directory and filename; check that
+   * the URL generated by file_create_url() for the specified file equals the
+   * specified URL; fetch the URL and then compare the contents to the file.
+   *
+   * @param $scheme
+   *   A scheme, e.g. "public"
+   * @param $directory
+   *   A directory, possibly ""
+   * @param $filename
+   *   A filename
+   * @param $expected_url
+   *   The expected URL
+   * @param $clean_url
+   *   The value of the clean_url setting
+   */
+  private function checkUrl($scheme, $directory, $filename, $expected_url, $clean_url = '1') {
+    variable_set('clean_url', $clean_url);
+
+    // Convert $filename to a valid filename, i.e. strip characters not
+    // supported by the filesystem, and create the file in the specified
+    // directory.
+    $filepath = file_create_filename($filename, $directory);
+    $directory_uri = $scheme . '://' . dirname($filepath);
+    file_prepare_directory($directory_uri, FILE_CREATE_DIRECTORY);
+    $file = $this->createFile($filepath, NULL, $scheme);
+
+    $url = file_create_url($file->uri);
+    $this->assertEqual($url, $expected_url, t('Generated URL matches expected URL.'));
+
+    if ($scheme == 'private') {
+      // Tell the implementation of hook_file_download() in file_test.module
+      // that this file may be downloaded.
+      file_test_set_return('download', array('x-foo' => 'Bar'));
+    }
+
+    $this->drupalGet($url);
+    if ($this->assertResponse(200) == 'pass') {
+      $this->assertRaw(file_get_contents($file->uri), t('Contents of the file are correct.'));
+    }
+
+    file_delete($file);
+  }
 }
 
 /**

modules/simpletest/tests/menu.test

@@ -51,6 +51,17 @@ class MenuRouterTestCase extends DrupalWebTestCase {
     $this->assertRaw('seven/style.css', t("The administrative theme's CSS appears on the page."));
   }
 
+  /**
+   * Test path containing "exotic" characters.
+   */
+  function testExoticPath() {
+    $path = "menu-test/ -._~!$'\"()*@[]?&+%#,;=:" . // "Special" ASCII characters.
+      "%23%25%26%2B%2F%3F" . // Characters that look like a percent-escaped string.
+      "éøïвβ中國書۞"; // Characters from various non-ASCII alphabets.
+    $this->drupalGet($path);
+    $this->assertRaw('This is menu_test_callback().');
+  }
+
   /**
    * Test the theme callback when the site is in maintenance mode.
    */

modules/simpletest/tests/menu_test.module

@@ -58,6 +58,15 @@ function menu_test_menu() {
     'page arguments' => array(TRUE),
     'access arguments' => array('access content'),
   );
+  // Path containing "exotic" characters.
+  $path = "menu-test/ -._~!$'\"()*@[]?&+%#,;=:" . // "Special" ASCII characters.
+    "%23%25%26%2B%2F%3F" . // Characters that look like a percent-escaped string.
+    "éøïвβ中國書۞"; // Characters from various non-ASCII alphabets.
+  $items[$path] = array(
+    'title' => '"Exotic" path',
+    'page callback' => 'menu_test_callback',
+    'access arguments' => array('access content'),
+  );
 
   // Hidden tests; base parents.
   // Same structure as in Menu and Block modules. Since those structures can
@@ -174,7 +183,7 @@ function menu_test_menu() {
  *  A random string.
  */
 function menu_test_callback() {
-  return $this->randomName();
+  return 'This is menu_test_callback().';
 }
 
 /**

modules/simpletest/tests/path.test

@@ -200,6 +200,15 @@ class UrlAlterFunctionalTest extends DrupalWebTestCase {
     $this->assertUrlOutboundAlter("taxonomy/term/$tid", "taxonomy/term/$tid");
   }
 
+  /**
+   * Test current_path() and request_path().
+   */
+  function testCurrentUrlRequestedPath() {
+    $this->drupalGet('url-alter-test/bar');
+    $this->assertRaw('request_path=url-alter-test/bar', t('request_path() returns the requested path.'));
+    $this->assertRaw('current_path=url-alter-test/foo', t('current_path() returns the internal path.'));
+  }
+
   /**
    * Assert that an outbound path is altered to an expected value.
    *

modules/simpletest/tests/url_alter_test.module

@@ -6,6 +6,27 @@
  * Module to help test hook_url_inbound_alter() and hook_url_outbound_alter().
  */
 
+/**
+ * Implements hook_menu().
+ */
+function url_alter_test_menu() {
+  $items['url-alter-test/foo'] = array(
+    'title' => 'Foo',
+    'page callback' => 'url_alter_test_foo',
+    'access arguments' => array('access content'),
+    'type' => MENU_CALLBACK,
+  );
+  return $items;
+}
+
+/**
+ * Menu callback.
+ */
+function url_alter_test_foo() {
+  print 'current_path=' . current_path() . ' request_path=' . request_path();
+  exit;
+}
+
 /**
  * Implements hook_url_inbound_alter().
  */
@@ -22,6 +43,10 @@ function url_alter_test_url_inbound_alter(&$path, $original_path, $path_language
   if ($path == 'community' || strpos($path, 'community/') === 0) {
     $path = 'forum' . substr($path, 9);
   }
+
+  if ($path == 'url-alter-test/bar') {
+    $path = 'url-alter-test/foo';
+  }
 }
 
 /**