Added handling of $sitekey exceptions

lib/basedomain.js

Index: lib/basedomain.js
===================================================================
--- a/lib/basedomain.js
+++ b/lib/basedomain.js
@@ -134,32 +134,109 @@ function isThirdParty(/**String*/ reques
     return (requestHost != documentDomain);
 }
 
 /**
  * Extracts host name from a URL.
  */
 function extractHostFromURL(/**String*/ url)
 {
-  if (!url)
-    return "";
-
-  if (extractHostFromURL._lastURL == url)
+  if (url && extractHostFromURL._lastURL == url)
     return extractHostFromURL._lastDomain;
 
-  var x = url.substr(url.indexOf("://") + 3);
-  x = x.substr(0, x.indexOf("/"));
-  x = x.substr(x.indexOf("@") + 1);
-  if (x.indexOf("[") == 0 && x.indexOf("]") > 0)
+  var host = "";
+  try
   {
-    x = x.substring(1,x.indexOf("]"));
+    host = new URI(url).host;
+  }
+  catch (e)
+  {
+    // Keep the empty string for invalid URIs.
+  }
+
+  extractHostFromURL._lastURL = url;
+  extractHostFromURL._lastDomain = host;
+  return host;
+}
+
+/**
+ * Parses URLs and provides an interface similar to nsIURI in Gecko, see
+ * https://developer.mozilla.org/en-US/docs/XPCOM_Interface_Reference/nsIURI.
+ * TODO: Make sure the parsing actually works the same as nsStandardURL.
+ * @constructor
+ */
+function URI(/**String*/ spec)
+{
+  this.spec = spec;

[Felix Dahlke, 2012/09/25 12:40:53]

Shouldn't this be this._spec? Seems to be the convention here.

[Wladimir Palant, 2012/09/25 14:09:41]

It's actually a public member like in nsIURI - that's how you get the string
representation of the URI object.

+  this._schemeEnd = spec.indexOf(":");

[Felix Dahlke, 2012/09/25 12:40:53]

You could probably get rid of a lot of code here if you parse the spec using a
regex with capture groups.

[Wladimir Palant, 2012/09/25 14:09:41]

Maybe. Then again, that regexp will probably be absolutely unintelligible -
there are just too many optional parts. And there is a slight hope that this
version is actually faster (it does matter here).

+  if (this._schemeEnd < 0)
+    throw new Error("Invalid URI scheme");
+
+  if (spec[this._schemeEnd + 1] != "/" || spec[this._schemeEnd + 2] != "/")

[Felix Dahlke, 2012/09/25 12:40:53]

How about this:
if (spec.substr(this._schemeEnd + 1, 2) !== "//")

[Wladimir Palant, 2012/09/25 14:09:41]

Given that we are testing for string equality here and not whether we actually
deal with the same string instances I'll rather use != operator :)

+    throw new Error("Unexpected URI structure");
+
+  this._hostPortStart = this._schemeEnd + 3;
+  this._hostPortEnd = spec.indexOf("/", this._hostPortStart);
+  if (this._hostPortEnd < 0)
+    throw new Error("Invalid URI host");
+
+  var authEnd = spec.indexOf("@", this._hostPortStart);
+  if (authEnd >= 0 && authEnd < this._hostPortEnd)
+    this._hostPortStart = authEnd + 1;
+
+  this._portStart = -1;
+  this._hostEnd = spec.indexOf("]", this._hostPortStart + 1);
+  if (spec[this._hostPortStart] == "[" && this._hostEnd >= 0 && this._hostEnd < this._hostPortEnd)
+  {
+    // The host is an IPv6 literal
+    this._hostStart = this._hostPortStart + 1;
+    if (spec[this._hostEnd + 1] == ":")
+      this._portStart = this._hostEnd + 2;
   }
   else
   {
-    colPos = x.indexOf(":");
-    if (colPos >= 0)
-      x = x.substr(0, colPos);
+    this._hostStart = this._hostPortStart;
+    this._hostEnd = spec.indexOf(":", this._hostStart);
+    if (this._hostEnd >= 0 && this._hostEnd < this._hostPortEnd)
+      this._portStart = this._hostEnd + 1;
+    else
+      this._hostEnd = this._hostPortEnd;
   }
-
-  extractHostFromURL._lastURL = url;
-  extractHostFromURL._lastDomain = x;
-  return x;
 }
+URI.prototype =
+{
+  spec: null,
+  get scheme()
+  {
+    return this.spec.substring(0, this._schemeEnd).toLowerCase();
+  },
+  get host()
+  {
+    return this.spec.substring(this._hostStart, this._hostEnd);
+  },
+  get asciiHost()
+  {
+    var host = this.host;
+    if (/^[\x00-\x7F]+$/.test(host))
+      return host;
+    else
+      return punycode.toASCII(host);
+  },
+  get hostPort()
+  {
+    return this.spec.substring(this._hostPortStart, this._hostPortEnd);
+  },
+  get port()
+  {
+    if (this._portStart < 0)
+      return -1;
+    else
+      return parseInt(this.spec.substring(this._portStart, this._hostPortEnd), 10);
+  },
+  get path()
+  {
+    return this.spec.substring(this._hostPortEnd);
+  },
+  get prePath()
+  {
+    return this.spec.substring(0, this._hostPortEnd);
+  }
+};