Make EventFilteringInfo::IsEmpty() take has_instance_id_ into account.

extensions/common/event_filtering_info.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 "extensions/common/event_filtering_info.h"
 
 #include "base/json/json_writer.h"
 #include "base/values.h"
 
 namespace extensions {
 
 EventFilteringInfo::EventFilteringInfo()
     : has_url_(false),

[not at google - send to devlin, 2014/01/16 23:58:59]

having both a url_ and a has_url_ seems redundant. why aren't we just checking
url_.IsValid() (or is that url_.valid()?) I wonder.

[lazyboy, 2014/01/17 01:08:47]

Seems this was checked in from the initial revision.
The only difference I can imagine would be when SetURL(GURL()) is called,
has_url_ would be true. But I don't know if the code calling this function is
counting on this behavior.

       has_instance_id_(false),
       instance_id_(0) {
 }
 
 EventFilteringInfo::~EventFilteringInfo() {
 }
 
 void EventFilteringInfo::SetURL(const GURL& url) {
   url_ = url;
   has_url_ = true;
@@ skipping 15 matching lines @@
   if (has_instance_id_)
     result->SetInteger("instanceId", instance_id_);
 
   if (!service_type_.empty())
     result->SetString("serviceType", service_type_);
 
   return result.PassAs<base::Value>();
 }
 
 bool EventFilteringInfo::IsEmpty() const {
-  return !has_url_ && service_type_.empty();
+  return !has_url_ && service_type_.empty() && !has_instance_id_;
 }
 
 }  // namespace extensions