Don't re-run content scripts on fragment navigations.

This regression was inadvertently introduced by a fix to make executeScript 
run after a fragment navigation (bug 29541). That patch was:

http://codereview.chromium.org/566041

The problem is that on frame navigations (didChangeLocationWithinPage), we end
up creating a new UserScriptIdleScheduler, so this patch keeps track of the 
previous has_run state and propagates that to the new UserScriptIdleScheduler.


BUG=35924
TEST=Steps to verify are outlined in bug report


Review URL: http://codereview.chromium.org/646017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@39939 0039d316-1c4b-4281-b951-d872f2087c98

1 files changed, 3 insertions, 4 deletions

diff --git a/chrome/renderer/user_script_idle_scheduler.h b/chrome/renderer/user_script_idle_scheduler.h
index d4825a5..25b75eb 100644
--- a/chrome/renderer/user_script_idle_scheduler.h
+++ b/chrome/renderer/user_script_idle_scheduler.h
@@ -22,22 +22,21 @@ class WebFrame;
 //
 // The intent of this mechanism is to prevent user scripts from slowing down
 // fast pages (run after load), while still allowing them to run relatively
-// timelily for pages with lots of slow subresources.
+// timely for pages with lots of slow subresources.
 class UserScriptIdleScheduler {
  public:
   UserScriptIdleScheduler(RenderView* view, WebKit::WebFrame* frame);
 
   bool has_run() { return has_run_; }
 
+  void set_has_run(bool has_run) { has_run_ = has_run; }
+
   // Called when the DOM has been completely constructed.
   void DidFinishDocumentLoad();
 
   // Called when the document has completed loading.
   void DidFinishLoad();
 
-  // Called when the document has navigated to a fragment.
-  void DidChangeLocationWithinPage();
-
   // Called when the client has gone away and we should no longer run scripts.
   void Cancel();