Set the target type when creating the request for main resource

Source/platform/network/ResourceRequest.cpp

 /*
  * Copyright (C) 2003, 2006 Apple Computer, Inc.  All rights reserved.
  * Copyright (C) 2009, 2012 Google Inc. All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
  * are met:
  * 1. Redistributions of source code must retain the above copyright
  *    notice, this list of conditions and the following disclaimer.
  * 2. Redistributions in binary form must reproduce the above copyright
@@ skipping 64 matching lines @@
     data->m_hasUserGesture = m_hasUserGesture;
     data->m_downloadToFile = m_downloadToFile;
     data->m_requestorID = m_requestorID;
     data->m_requestorProcessID = m_requestorProcessID;
     data->m_appCacheHostID = m_appCacheHostID;
     data->m_targetType = m_targetType;
     data->m_referrerPolicy = m_referrerPolicy;
     return data.release();
 }
 
+void ResourceRequest::initializeForMainResource(bool isMainFrame)

[ppi, 2014/06/19 18:25:42]

This can be called at any point, while ::initialize() has the nice property of
being private and called only from constructors.

There seems to be existing pattern here of setting additional parameters by
adding a specialized constructor that calls ::initialize() to set up defaults
first, and then overrides some of them, see "ResourceRequest(const KURL& url,
const Referrer& referrer ..." .

Would this pattern work for this case?

For overrides from content we could use simple setters (setTargetType /
setPriority).

+{
+    m_targetType = isMainFrame ? TargetIsMainFrame : TargetIsSubframe;
+    m_priority = ResourceLoadPriorityVeryHigh;
+}
+
 bool ResourceRequest::isEmpty() const
 {
     return m_url.isEmpty();
 }
 
 bool ResourceRequest::isNull() const
 {
     return m_url.isNull();
 }
 
@@ skipping 279 matching lines @@
 // This is used by the loader to control the number of issued parallel load requests.
 unsigned initializeMaximumHTTPConnectionCountPerHost()
 {
     // The chromium network stack already handles limiting the number of
     // parallel requests per host, so there's no need to do it here.  Therefore,
     // this is set to a high value that should never be hit in practice.
     return 10000;
 }
 
 }