Consolidates accessing and setting the UMA pref to be within metrics code

chrome/browser/chromeos/settings/device_settings_provider.cc

Index: chrome/browser/chromeos/settings/device_settings_provider.cc
diff --git a/chrome/browser/chromeos/settings/device_settings_provider.cc b/chrome/browser/chromeos/settings/device_settings_provider.cc
index 3b2b3f318b02f1d92a6a22e9d0e40e4f6ee2dc85..c1d0bff922debf98ae71217bf5654de106384251 100644
--- a/chrome/browser/chromeos/settings/device_settings_provider.cc
+++ b/chrome/browser/chromeos/settings/device_settings_provider.cc
@@ -848,7 +848,7 @@ void DeviceSettingsProvider::ApplyMetricsSetting(bool use_file,
           << "(use file : " << use_file << ")";
   // TODO(pastarmovj): Remove this once we don't need to regenerate the
   // consent file for the GUID anymore.
-  ResolveMetricsReportingEnabled(new_value);
+  InitiateMetricsReportingChange(new_value, base::Callback<void(bool)>());

[Alexei Svitkine (slow), 2014/09/05 15:22:48]

Is this correct?

This is from CrOS code, but I thought the function currently doesn't handle
CrOS?

Also, this code has a bunch of TODOs - can you check with the author to see if
it might make sense to "Remove this" per the comment?

[gayane -on leave until 09-2017, 2014/09/05 18:30:20]

Actually InitiateMetricsReportingChange function would do the same thing for
ChromeOS as before except that it would run on the FILE thread.

For the TODOs I have contacted the author.


On 2014/09/05 15:22:48, Alexei Svitkine wrote:

 }
 
 void DeviceSettingsProvider::ApplySideEffects(