1 files changed, 54 insertions, 7 deletions

diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php
index 039710fd1..25eb8410b 100755..100644
--- a/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php
+++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php
@@ -3,11 +3,13 @@
 /**
  * Validator for the components of a URI for a specific scheme
  */
-class HTMLPurifier_URIScheme
+abstract class HTMLPurifier_URIScheme
 {
 
     /**
-     * Scheme's default port (integer)
+     * Scheme's default port (integer).  If an explicit port number is
+     * specified that coincides with the default port, it will be
+     * elided.
      */
     public $default_port = null;
 
@@ -24,17 +26,62 @@ class HTMLPurifier_URIScheme
     public $hierarchical = false;
 
     /**
-     * Validates the components of a URI
-     * @note This implementation should be called by children if they define
-     *       a default port, as it does port processing.
-     * @param $uri Instance of HTMLPurifier_URI
+     * Whether or not the URI may omit a hostname when the scheme is
+     * explicitly specified, ala file:///path/to/file. As of writing,
+     * 'file' is the only scheme that browsers support his properly.
+     */
+    public $may_omit_host = false;
+
+    /**
+     * Validates the components of a URI for a specific scheme.
+     * @param $uri Reference to a HTMLPurifier_URI object
+     * @param $config HTMLPurifier_Config object
+     * @param $context HTMLPurifier_Context object
+     * @return Bool success or failure
+     */
+    public abstract function doValidate(&$uri, $config, $context);
+
+    /**
+     * Public interface for validating components of a URI.  Performs a
+     * bunch of default actions. Don't overload this method.
+     * @param $uri Reference to a HTMLPurifier_URI object
      * @param $config HTMLPurifier_Config object
      * @param $context HTMLPurifier_Context object
      * @return Bool success or failure
      */
     public function validate(&$uri, $config, $context) {
         if ($this->default_port == $uri->port) $uri->port = null;
-        return true;
+        // kludge: browsers do funny things when the scheme but not the
+        // authority is set
+        if (!$this->may_omit_host &&
+            // if the scheme is present, a missing host is always in error
+            (!is_null($uri->scheme) && ($uri->host === '' || is_null($uri->host))) ||
+            // if the scheme is not present, a *blank* host is in error,
+            // since this translates into '///path' which most browsers
+            // interpret as being 'http://path'.
+             (is_null($uri->scheme) && $uri->host === '')
+        ) {
+            do {
+                if (is_null($uri->scheme)) {
+                    if (substr($uri->path, 0, 2) != '//') {
+                        $uri->host = null;
+                        break;
+                    }
+                    // URI is '////path', so we cannot nullify the
+                    // host to preserve semantics.  Try expanding the
+                    // hostname instead (fall through)
+                }
+                // first see if we can manually insert a hostname
+                $host = $config->get('URI.Host');
+                if (!is_null($host)) {
+                    $uri->host = $host;
+                } else {
+                    // we can't do anything sensible, reject the URL.
+                    return false;
+                }
+            } while (false);
+        }
+        return $this->doValidate($uri, $config, $context);
     }
 
 }