Remove WebURLResponse::initialize() [revised]

third_party/WebKit/Source/platform/exported/WebURLResponsePrivate.h

Index: third_party/WebKit/Source/platform/exported/WebURLResponsePrivate.h
diff --git a/third_party/WebKit/Source/platform/exported/WebURLResponsePrivate.h b/third_party/WebKit/Source/platform/exported/WebURLResponsePrivate.h
index 81cb5c0649e665bdd83a4d7fab1230c82061cd3c..fa9382168299dbee9c171eff81f843f147ba7fc5 100644
--- a/third_party/WebKit/Source/platform/exported/WebURLResponsePrivate.h
+++ b/third_party/WebKit/Source/platform/exported/WebURLResponsePrivate.h
@@ -41,12 +41,12 @@ class ResourceResponse;
 class WebURLResponsePrivate {
     WTF_MAKE_NONCOPYABLE(WebURLResponsePrivate);
 public:
-    WebURLResponsePrivate() : m_resourceResponse(0) { }
+    WebURLResponsePrivate() {}
+    virtual ~WebURLResponsePrivate() {}
 
-    // Called by WebURLResponse when it no longer needs this object.
-    virtual void dispose() = 0;
-
-    ResourceResponse* m_resourceResponse;
+    virtual const ResourceResponse* resourceResponse() const = 0;
+    virtual ResourceResponse* resourceResponse() = 0;

[Adam Rice, 2016/07/06 01:39:00]

I have been assuming that whoever originally wrote this was trying to avoid
virtual method dispatch on the regular setters and getters. Otherwise why make
it so complicated?

They might have been wrong about the need for this optimisation, but they might
have been right.

[kinuko, 2016/07/06 09:45:57]

That might be the case (given that WebKit code base generally tried to avoid
doing that), but it's making the code harder to read.  I could revert this part
though and try to land it separately (so that we can easily revert this specific
change if we see any regression)

[Adam Rice, 2016/07/06 10:06:50]

I think makes sense to split out the removal of initialize() from the
refactoring so we have less to revert if we see a regression.

+    virtual void setResourceResponse(const ResourceResponse& response) = 0;
 };
 
 } // namespace blink