Added of new UMA signals in order to be able to discover early if the "save password" feature gets …

chrome/browser/password_manager/password_manager.h

Index: chrome/browser/password_manager/password_manager.h
diff --git a/chrome/browser/password_manager/password_manager.h b/chrome/browser/password_manager/password_manager.h
index d65d1c7d449d5c2d7ca9280649eca73ad3e28a67..7f7ccfa72b980c2283087c323e55fc3275849307 100644
--- a/chrome/browser/password_manager/password_manager.h
+++ b/chrome/browser/password_manager/password_manager.h
@@ -35,6 +35,23 @@ class PasswordManager : public LoginModel,
                         public content::WebContentsObserver,
                         public content::WebContentsUserData<PasswordManager> {
  public:
+  enum PotentialInfoBarErrorReasons {
+    SPDY_PROXY,
+    SAVE_PASSWORD_DISABLE,

[vabr (Chromium), 2013/08/14 14:39:58]

nit: DISABLE->DISABLED ?

+    EMPTY_PASSWORD,
+    FIRST_PAGE_NOT_LOADED,
+    MATCHING_ONGOING,
+    NEVER_REMEMBER,
+    INVALID_FORM,
+    PASSWORD_GENERATED_OR_AUTOCOMPLETED,
+    CANNOT_GET_THE_PROVIOSIONAL_PASSWORD,

[vabr (Chromium), 2013/08/14 14:39:58]

typo: PROVIOSIONAL

+    FORM_REAPPEARED,
+    LOGIN_ALREADY_KNEW,

[vabr (Chromium), 2013/08/14 14:39:58]

nit: KNEW->KNOWN ?

+    PASSWORD_GENERATED,

[vabr (Chromium), 2013/08/14 14:39:58]

How does this differ from PASSWORD_GENERATED_OR_AUTOCOMPLETED?

+    ORIGIN_MATCHING_OF_PUBLIC_SUFFIX,

[vabr (Chromium), 2013/08/14 14:39:58]

nit: leave out "OF" ?

+    NUM_ERROR_TYPES

[vabr (Chromium), 2013/08/14 14:39:58]

Here I am not sure about the C++ standard -- are we guaranteed that the enum
values are numbered from 0 up by +1 increments?
If yes, then this is fine.
Otherwise we need to assign values explicitly.

[Garrett Casto, 2013/08/14 23:02:47]

Yep.

+  };
+
   static void RegisterProfilePrefs(user_prefs::PrefRegistrySyncable* registry);
 
   static void CreateForWebContentsAndDelegate(