[Findit] don't included skipped or unknown tests in swarming tasks into failed tests.

appengine/findit/waterfall/process_swarming_task_result_pipeline.py

Index: appengine/findit/waterfall/process_swarming_task_result_pipeline.py
diff --git a/appengine/findit/waterfall/process_swarming_task_result_pipeline.py b/appengine/findit/waterfall/process_swarming_task_result_pipeline.py
index 06d55d6de4d23ccba2e8dceb475278617b6aa2b4..f8dae19ca11bad13e08d32584e46e274a7e2bfb1 100644
--- a/appengine/findit/waterfall/process_swarming_task_result_pipeline.py
+++ b/appengine/findit/waterfall/process_swarming_task_result_pipeline.py
@@ -15,11 +15,12 @@ from waterfall import swarming_util
 from waterfall import waterfall_config
 
 
-def _CheckTestsRunStatuses(output_json):
+def _CheckTestsRunStatuses(output_json, targeted_tests):
   """Checks result status for each test run and saves the numbers accordingly.
 
   Args:
     output_json (dict): A dict of all test results in the swarming task.
+    targeted_tests (list): A list of targeted tests.
 
   Returns:
     tests_statuses (dict): A dict of different statuses for each test.
@@ -31,6 +32,10 @@ def _CheckTestsRunStatuses(output_json):
   if output_json:
     for iteration in output_json.get('per_iteration_data'):
       for test_name, tests in iteration.iteritems():
+        if test_name not in targeted_tests:

[stgao, 2016/06/02 01:31:20]

If PRE_A failed on Waterfall and PRE_PRE_A failed on Swarming rerun (PRE_A is
skipped), how do we handle this case in the end?

[chanli, 2016/06/02 22:40:09]

In the end we should still say PRE_A failed.

[stgao, 2016/06/03 06:59:39]

OK. The code seems fixed for this now.

+          # Some special pseudo tests like PRE_ tests may be added to task,
+          # should filter them out when check results.
+          continue
         tests_statuses[test_name]['total_run'] += len(tests)
         for test in tests:
           tests_statuses[test_name][test['status']] += 1
@@ -62,7 +67,8 @@ class ProcessSwarmingTaskResultPipeline(BasePipeline):
 
   HTTP_CLIENT = HttpClient()
   # Arguments number differs from overridden method - pylint: disable=W0221
-  def run(self, master_name, builder_name, build_number, step_name, task_id):
+  def run(
+      self, master_name, builder_name, build_number, step_name, task_id, tests):
     """
     Args:
       master_name (str): The master name.
@@ -70,6 +76,7 @@ class ProcessSwarmingTaskResultPipeline(BasePipeline):
       build_number (str): The build number.
       step_name (str): The failed test step name.
       task_id (str): Id for the swarming task which is triggered by Findit.
+      tests (str): A list of test cases, eg: ['suite1.test1', 'suite2.test2'].
 
     Returns:
       A dict of lists for reliable/flaky tests.
@@ -102,7 +109,7 @@ class ProcessSwarmingTaskResultPipeline(BasePipeline):
           outputs_ref = data.get('outputs_ref')
           output_json = swarming_util.GetSwarmingTaskFailureLog(
               outputs_ref, self.HTTP_CLIENT)
-          tests_statuses = _CheckTestsRunStatuses(output_json)
+          tests_statuses = _CheckTestsRunStatuses(output_json, tests)
 
           task.status = analysis_status.COMPLETED
           task.tests_statuses = tests_statuses