Add an iterator to prevent OOM when cleaning large databases (#604)

* Fixes OOM issue when cleaning over 8M records

* Adding name to Authors

* Added changes to Changelog

* Removing redundant .all()

* Removing unnecessary typo.

* Adding black formatting to unrelated files

Author: craigmaloney

AUTHORS.rst

@@ -22,6 +22,7 @@ Authors
 - Brian Dixon
 - Christopher Broderick (`uhurusurfa <https://github.com/uhurusurfa>`_)
 - Corey Bertram
+- Craig Maloney (`craigmaloney <https://github.com/craigmaloney>`_)
 - Damien Nozay
 - Daniel Gilge
 - Daniel Levy

CHANGES.rst

@@ -9,6 +9,7 @@ Unreleased
 - Fix `model_to_dict` to detect changes in a parent model when using
   `inherit=True` (backwards-incompatible for users who were directly
   using previous version) (gh-576)
+- Use an iterator for `clean_duplicate_history`
 
 2.7.3 (2019-07-15)
 ------------------

simple_history/management/commands/clean_duplicate_history.py

@@ -69,7 +69,7 @@ def _process(self, to_process, date_back=None, dry_run=True):
             # have changes (in the given period) but
             # `m_qs.values(model._meta.pk.name).distinct()`
             # is actually slower than looping all and filtering in the code...
-            for o in model.objects.all():
+            for o in model.objects.iterator():
                 self._process_instance(o, model, stop_date=stop_date, dry_run=dry_run)
 
     def _process_instance(self, instance, model, stop_date=None, dry_run=True):

simple_history/tests/tests/test_manager.py

@@ -22,9 +22,11 @@ def setUp(self):
         self.obj.changed_by = user
         self.obj.save()
         self.model.objects.all().delete()  # allows us to leave PK on instance
-        self.delete_history, self.change_history, self.create_history = (
-            self.model.history.all()
-        )
+        (
+            self.delete_history,
+            self.change_history,
+            self.create_history,
+        ) = self.model.history.all()
         self.create_history.history_date = self.now - timedelta(days=2)
         self.create_history.save()
         self.change_history.history_date = self.now - timedelta(days=1)

simple_history/tests/tests/test_models.py

@@ -123,7 +123,7 @@ def assertRecordValues(self, record, klass, values_dict):
     def test_create(self):
         p = Poll(question="what's up?", pub_date=today)
         p.save()
-        record, = p.history.all()
+        (record,) = p.history.all()
         self.assertRecordValues(
             record,
             Poll,