Bug 1328453 - Only use completed jobs for calculating average job duration

Completed jobs = those with an end time > start_time

tests/model/test_calculate_durations.py

@@ -42,3 +42,15 @@ def test_calculate_durations(test_repository, failure_classifications,
     assert len(durations) == 1
     expected_duration = int(round((first_job_duration + second_job_duration) / 2))
     assert durations[0].average_duration == expected_duration
+
+    # Add a fake job with an end time > start time, verify that it is
+    # ignored and average duration remains the same
+    third_job = job_data(revision=result_set_stored[0]['revision'],
+                         start_timestamp=now,
+                         end_timestamp=now - second_job_duration,
+                         job_guid='another-unique-guid')
+    store_job_data(test_repository, [third_job])
+    call_command('calculate_durations')
+    durations = JobDuration.objects.all()
+    assert len(durations) == 1
+    assert durations[0].average_duration == expected_duration

treeherder/model/models.py

@@ -14,7 +14,8 @@ from django.core.cache import cache
 from django.core.validators import MinLengthValidator
 from django.db import (models,
                        transaction)
-from django.db.models import (Q,
+from django.db.models import (F,
+                              Q,
                               Case,
                               Count,
                               When)
@@ -572,7 +573,8 @@ class JobManager(models.Manager):
             num_jobs = 0
             total_time = 0.0
             for (start_time, end_time) in jobs.filter(
-                    signature__signature=signature_hash).values_list(
+                    signature__signature=signature_hash,
+                    end_time__gt=F('start_time')).values_list(
                         'start_time', 'end_time'):
                 total_time += (end_time - start_time).total_seconds()
                 num_jobs += 1