Separate RenderViewHost from RenderWidgetHost, part 4: delegate

chrome/browser/extensions/api/processes/processes_api.cc

Index: chrome/browser/extensions/api/processes/processes_api.cc
diff --git a/chrome/browser/extensions/api/processes/processes_api.cc b/chrome/browser/extensions/api/processes/processes_api.cc
index 4d33c334ef2fc0c30816d96d310b8ed72f23c3ce..b7d2b81a24fbfbb36737da16134663b03c4d56da 100644
--- a/chrome/browser/extensions/api/processes/processes_api.cc
+++ b/chrome/browser/extensions/api/processes/processes_api.cc
@@ -119,8 +119,6 @@ base::ListValue* GetTabsForProcess(int process_id) {
   while (content::RenderWidgetHost* widget = widgets->GetNextHost()) {
     if (widget->GetProcess()->GetID() != process_id)
       continue;
-    if (!widget->IsRenderView())

[sky, 2015/10/26 21:07:38]

How come this is safe? If widget is not a RenderViewHost would 123 be null, and
127 then crash?

[Avi (use Gerrit), 2015/10/26 23:20:14]

If widget isn't a RVH then the RenderViewHost* |host| on line 123 will be null.
Passing in a null RVH to WebContents::FromRenderViewHost now is safe and _also_
returns null (new in this CL!) and that null result will be tested on line 126.

[sky, 2015/10/27 13:12:33]

Ok, I missed that. I looked at the existing code which doesn't handle null.

-      continue;
 
     content::RenderViewHost* host = content::RenderViewHost::From(widget);
     content::WebContents* contents =