Teach SiteInstance::GetSiteForURL() about blob and filesystem URLs.

content/browser/site_instance_impl.cc

 // Copyright (c) 2012 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "content/browser/site_instance_impl.h"
 
 #include "content/browser/browsing_instance.h"
 #include "content/browser/child_process_security_policy_impl.h"
 #include "content/browser/frame_host/debug_urls.h"
 #include "content/browser/frame_host/frame_tree_node.h"
@@ skipping 316 matching lines @@
 }
 
 // static
 GURL SiteInstance::GetSiteForURL(BrowserContext* browser_context,
                                  const GURL& real_url) {
   // TODO(fsamuel, creis): For some reason appID is not recognized as a host.
   if (real_url.SchemeIs(kGuestScheme))
     return real_url;
 
   GURL url = SiteInstanceImpl::GetEffectiveURL(browser_context, real_url);
+  url::Origin origin(url);
 
   // If the url has a host, then determine the site.
-  if (url.has_host()) {
-    // Only keep the scheme and registered domain as given by GetOrigin.  This
-    // may also include a port, which we need to drop.
-    GURL site = url.GetOrigin();
-
-    // Remove port, if any.
-    if (site.has_port()) {
-      GURL::Replacements rep;
-      rep.ClearPort();
-      site = site.ReplaceComponents(rep);
-    }
-
-    // If this URL has a registered domain, we only want to remember that part.
-    std::string domain =
-        net::registry_controlled_domains::GetDomainAndRegistry(
-            url,
-            net::registry_controlled_domains::INCLUDE_PRIVATE_REGISTRIES);
-    if (!domain.empty()) {
-      GURL::Replacements rep;
-      rep.SetHostStr(domain);
-      site = site.ReplaceComponents(rep);
-    }
-    return site;
+  if (!origin.host().empty()) {
+    // Only keep the scheme and registered domain of |origin|.
+    std::string domain = net::registry_controlled_domains::GetDomainAndRegistry(
+        origin.host(),
+        net::registry_controlled_domains::INCLUDE_PRIVATE_REGISTRIES);
+    std::string site = origin.scheme();
+    site += url::kStandardSchemeSeparator;
+    site += domain.empty() ? origin.host() : domain;

[Charlie Reis, 2016/04/22 18:11:09]

nit: Maybe add a comment for which case this is for (e.g., IP addresses), since
it's a little tricky to understand at first glance.  (I agree that it's
basically the same as what we did before.)

+    return GURL(site);
   }
 
   // If there is no host but there is a scheme, return the scheme.
   // This is useful for cases like file URLs.
   if (url.has_scheme())
     return GURL(url.scheme() + ":");
 
   // Otherwise the URL should be invalid; return an empty site.
   DCHECK(!url.is_valid());
   return GURL();
@@ skipping 70 matching lines @@
             browsing_instance_->browser_context(), site_))
       return;
 
     ChildProcessSecurityPolicyImpl* policy =
         ChildProcessSecurityPolicyImpl::GetInstance();
     policy->LockToOrigin(process_->GetID(), site_);
   }
 }
 
 }  // namespace content