Adds a feature to MHTML serialization that omits subframes and subresources marked no-store.

content/renderer/render_frame_impl.cc

 // Copyright 2013 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/renderer/render_frame_impl.h"
 
 #include <map>
 #include <string>
 #include <utility>
 #include <vector>
@@ skipping 689 matching lines @@
     int routing_id = GetRoutingIdForFrameOrProxy(frame);
 
     auto it = params_.frame_routing_id_to_content_id.find(routing_id);
     if (it == params_.frame_routing_id_to_content_id.end())
       return WebString();
 
     const std::string& content_id = it->second;
     return WebString::fromUTF8(content_id);
   }
 
+  blink::WebFrameSerializerCacheControlPolicy cacheControlPolicy() override {
+    return static_cast<blink::WebFrameSerializerCacheControlPolicy>(
+        params_.mhtml_cache_control_policy);
+  }
+
+  bool useBinaryEncoding() override { return params_.mhtml_binary_encoding; }
+
  private:
   const FrameMsg_SerializeAsMHTML_Params& params_;
   std::set<std::string>* digests_of_uris_of_serialized_resources_;
 
   DISALLOW_COPY_AND_ASSIGN(MHTMLPartsGenerationDelegate);
 };
 
 // Returns true if a subresource certificate error (described by |url|
 // and |security_info|) is "interesting" to the browser process. The
 // browser process is interested in certificate errors that differ from
@@ skipping 4221 matching lines @@
 
 void RenderFrameImpl::OnSerializeAsMHTML(
     const FrameMsg_SerializeAsMHTML_Params& params) {
   // Unpack IPC payload.
   base::File file = IPC::PlatformFileForTransitToFile(params.destination_file);
   const WebString mhtml_boundary =
       WebString::fromUTF8(params.mhtml_boundary_marker);
   DCHECK(!mhtml_boundary.isEmpty());
 
   WebData data;
-  bool success = true;
   std::set<std::string> digests_of_uris_of_serialized_resources;
   MHTMLPartsGenerationDelegate delegate(
       params, &digests_of_uris_of_serialized_resources);
 
+  bool should_serialize_frame =
+      WebFrameSerializer::frameShouldBeSerializedAsMHTML(GetWebFrame(),
+                                                         &delegate);

[Łukasz Anforowicz, 2016/05/19 17:00:13]

Do we still need this (this = |should_serialize_frame| variable + *public*
WebFrameSerializer::frameShouldBeSerializedAsMHTML method)?  It seems that you
are now handling no-store check within the serializer, in
WebFrameSerializer::generateMHTMLHeader and
WebFrameSerializer::generateMHTMLParts (which I think is great) so, checking
return value of generateMHTMLHeader and generateMHTMLParts should be sufficient?

[dewittj, 2016/05/19 18:18:36]

Done.

+  bool success = true;
+
   // Generate MHTML header if needed.
   if (IsMainFrame()) {
-    blink::WebFrameSerializerCacheControlPolicy policy =
-        static_cast<blink::WebFrameSerializerCacheControlPolicy>(
-            params.mhtml_cache_control_policy);
-    success = WebFrameSerializer::generateMHTMLHeader(mhtml_boundary, policy,
-                                                      GetWebFrame(), &data);
-    if (success && file.WriteAtCurrentPos(data.data(), data.size()) < 0) {
+    if (should_serialize_frame) {
+      WebData data = WebFrameSerializer::generateMHTMLHeader(
+          mhtml_boundary, GetWebFrame(), &delegate);
+      if (file.WriteAtCurrentPos(data.data(), data.size()) < 0) {

[Łukasz Anforowicz, 2016/05/19 17:00:13]

Just double-checking: do we need to check here is generateMHTMLHeader returned
an empty, default WebData() - it seems to be returned this when frame should not
be serialized.

[dewittj, 2016/05/19 18:18:35]

Done.

+        success = false;
+      }
+    } else {
+      // If we can't serialize the main frame, the whole archive is bad.
       success = false;
     }
   }
 
-  // Generate MHTML parts.
-  if (success) {
-    data = WebFrameSerializer::generateMHTMLParts(
-        mhtml_boundary, GetWebFrame(), params.mhtml_binary_encoding, &delegate);
+  // Generate MHTML parts.  Note that if this is not the main frame, then even
+  // skipping the whole parts generation step is not an error - it simply
+  // results in an omitted resource in the final file.
+  if (success && should_serialize_frame) {
+    data = WebFrameSerializer::generateMHTMLParts(mhtml_boundary, GetWebFrame(),
+                                                  &delegate);

[Łukasz Anforowicz, 2016/05/19 17:00:13]

Same question as above - do we need to check if generateMHTMLParts returned an
empty WebData (and in this case we would need to set success=false)?

[dewittj, 2016/05/19 18:18:36]

Done.

     // TODO(jcivelli): write the chunks in deferred tasks to give a chance to
     //                 the message loop to process other events.
     if (file.WriteAtCurrentPos(data.data(), data.size()) < 0) {
       success = false;
     }
   }
 
   // Generate MHTML footer if needed.
   if (success && params.is_last_frame) {
     data = WebFrameSerializer::generateMHTMLFooter(mhtml_boundary);
@@ skipping 1161 matching lines @@
   // event target. Potentially a Pepper plugin will receive the event.
   // In order to tell whether a plugin gets the last mouse event and which it
   // is, we set |pepper_last_mouse_event_target_| to null here. If a plugin gets
   // the event, it will notify us via DidReceiveMouseEvent() and set itself as
   // |pepper_last_mouse_event_target_|.
   pepper_last_mouse_event_target_ = nullptr;
 #endif
 }
 
 }  // namespace content