Reload: FrameMsg_Navigate_Type cleanup to rename RELOAD_MAIN_RESOURCE

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 637 matching lines @@
 media::Context3D GetSharedMainThreadContext3D(
     scoped_refptr<ui::ContextProviderCommandBuffer> provider) {
   if (!provider)
     return media::Context3D();
   return media::Context3D(provider->ContextGL(), provider->GrContext());
 }
 
 bool IsReload(FrameMsg_Navigate_Type::Value navigation_type) {
   switch (navigation_type) {
     case FrameMsg_Navigate_Type::RELOAD:
-    case FrameMsg_Navigate_Type::RELOAD_MAIN_RESOURCE:
     case FrameMsg_Navigate_Type::RELOAD_BYPASSING_CACHE:
     case FrameMsg_Navigate_Type::RELOAD_ORIGINAL_REQUEST_URL:
       return true;
     case FrameMsg_Navigate_Type::RESTORE:
     case FrameMsg_Navigate_Type::RESTORE_WITH_POST:
     case FrameMsg_Navigate_Type::NORMAL:
       return false;
   }
   NOTREACHED();
   return false;
 }
 
 WebFrameLoadType ReloadFrameLoadTypeFor(
     FrameMsg_Navigate_Type::Value navigation_type) {
   switch (navigation_type) {
     case FrameMsg_Navigate_Type::RELOAD:
     case FrameMsg_Navigate_Type::RELOAD_ORIGINAL_REQUEST_URL:
-      return WebFrameLoadType::Reload;

[Charlie Reis, 2017/01/11 18:21:54]

This looks like a possible behavior change, since WebFrameLoadType::Reload is
still used differently than WebFrameLoadType::ReloadMainResource.  For example,
see RenderFrameImpl::LoadDataURL and the comment in WebFrameLoadType.h.

Do we need to update those as well?

[Takashi Toyoshima, 2017/01/12 11:11:00]

Since FrameMsg_Navigate_Type::RELOAD isn't used any more, this just change a
behavior for RELOAD_ORIGINAL_REQUEST_URL, and I think this should be changed
together to return ReloadMainResource when I introduced it.

I also feel I should have added "load_type ==
WebFrameLoadType::ReloadMainResource" in replace logic of LoadDataURL() though I
still do not understand completely how and when this code is called.

I already switched to use ReloadMainResource in almost all places in Blink side
though location.reload() still use Reload unless the experimental flag is
enabled. Once the study is fully launched, I'd also change Reload's behavior
completely as I'm doing in content side. This effort is tracked by
crbug.com/674404.

[Charlie Reis, 2017/01/13 00:19:47]

Ok.  I'm alright with changing that behavior here as long as it's intentional,
and so that we're aware (in case there's a regression).  Can you mention the
change in the CL description?
I think that's primarily for Android Webview's loadDataWithBaseURL (sigh).  I
think you're right-- can you go ahead and update it to include
ReloadMainResource in this CL, as long as we're changing behavior?
Ok, so the WebFrameLoadType cleanup is blocked on the experiment finishing. 
Sounds good.

[Takashi Toyoshima, 2017/01/13 06:45:23]

Done.

-    case FrameMsg_Navigate_Type::RELOAD_MAIN_RESOURCE:
       return WebFrameLoadType::ReloadMainResource;
     case FrameMsg_Navigate_Type::RELOAD_BYPASSING_CACHE:
       return WebFrameLoadType::ReloadBypassingCache;
     case FrameMsg_Navigate_Type::RESTORE:
     case FrameMsg_Navigate_Type::RESTORE_WITH_POST:
     case FrameMsg_Navigate_Type::NORMAL:
       NOTREACHED();
       return WebFrameLoadType::Standard;
   }
   NOTREACHED();
@@ skipping 6107 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