[Findit] Refactor Findit pipeline.

appengine/findit/waterfall/test/schedule_test_try_job_pipeline_test.py

Index: appengine/findit/waterfall/test/schedule_test_try_job_pipeline_test.py
diff --git a/appengine/findit/waterfall/test/schedule_test_try_job_pipeline_test.py b/appengine/findit/waterfall/test/schedule_test_try_job_pipeline_test.py
new file mode 100644
index 0000000000000000000000000000000000000000..940a8875369d5374598f9946ec5059a2cfb82bcc
--- /dev/null
+++ b/appengine/findit/waterfall/test/schedule_test_try_job_pipeline_test.py
@@ -0,0 +1,93 @@
+# Copyright 2016 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+import mock
+
+from common.waterfall import buildbucket_client
+from common.waterfall import failure_type
+from model.wf_try_job import WfTryJob
+from model.wf_try_job_data import WfTryJobData
+from waterfall import schedule_test_try_job_pipeline
+from waterfall import schedule_try_job_pipeline
+from waterfall.schedule_test_try_job_pipeline import (
+    ScheduleTestTryJobPipeline)
+from waterfall.test import wf_testcase
+
+
+class ScheduleTestTryjobPipelineTest(wf_testcase.WaterfallTestCase):
+
+  def testGetTargetedTests(self):
+    reliable_tests = {
+        'a on platform': ('a', ['t1', 't2']),
+        'b on platform': ('b', [])
+    }
+    targeted_tests = schedule_test_try_job_pipeline._GetTargetedTests(
+        reliable_tests)
+    expected_targeted_tests = {
+        'a': ['t1', 't2']
+    }
+    self.assertEqual(targeted_tests, expected_targeted_tests)
+
+  def testGetBuildPropertiesForTestFailure(self):
+    master_name = 'm'
+    builder_name = 'b'
+    build_number = 1
+
+    expected_properties = {
+        'recipe': 'findit/chromium/test',
+        'good_revision': 1,
+        'bad_revision': 2,
+        'target_mastername': master_name,
+        'target_testername': 'b',
+        'referenced_build_url': ('https://build.chromium.org/p/%s/builders'
+                                 '/%s/builds/%s') % (
+                                     master_name, builder_name, build_number)
+    }
+    try_job_pipeline = ScheduleTestTryJobPipeline()
+    properties = try_job_pipeline._GetBuildProperties(
+        master_name, builder_name, build_number, 1, 2, failure_type.TEST, None)
+
+    self.assertEqual(properties, expected_properties)
+
+  @mock.patch.object(schedule_try_job_pipeline, 'buildbucket_client')
+  def testSuccessfullyScheduleNewTryJobForTest(self, mock_other):
+    master_name = 'm'
+    builder_name = 'b'
+    build_number = 223
+    good_revision = 'rev1'
+    bad_revision = 'rev2'
+    targeted_tests = ['a on platform', ['a', ['test1', 'test2']]]

[Sharu Jiang, 2016/08/09 23:22:15]

If there are multiply platforms:

targeted_tests = [('a on platform1', ['a', ['test1', 'test2']]),
                  ('a on platform2', ['a', ['test3', 'test4']])]

try_job_id = try_job_pipeline.run(
    master_name, builder_name, build_number, good_revision, bad_revision,
    failure_type.TEST, None, *targeted_tests)

Is this correct?

[chanli, 2016/08/11 01:27:54]

No. In test we are only testing this pipeline so we just pass the targeted tests
directly. But when the pipeline is actually running, reliable_tests are a list
of futures from last pipeline.

+    build_id = '1'
+
+    response = {
+        'build': {
+            'id': build_id,
+            'url': 'url',
+            'status': 'SCHEDULED',
+        }
+    }
+    results = [(None, buildbucket_client.BuildbucketBuild(response['build']))]
+    mock_other.TriggerTryJobs.return_value = results
+
+    WfTryJob.Create(master_name, builder_name, build_number).put()
+
+    try_job_pipeline = ScheduleTestTryJobPipeline()
+    try_job_id = try_job_pipeline.run(
+        master_name, builder_name, build_number, good_revision, bad_revision,
+        failure_type.TEST, None, targeted_tests)
+
+    try_job = WfTryJob.Get(master_name, builder_name, build_number)
+    self.assertEqual(try_job_id, build_id)
+    self.assertEqual(try_job.test_results[-1]['try_job_id'], build_id)
+
+    try_job_data = WfTryJobData.Get(try_job_id)
+    self.assertIsNotNone(try_job_data)
+    self.assertEqual(try_job_data.master_name, master_name)
+    self.assertEqual(try_job_data.builder_name, builder_name)
+    self.assertEqual(try_job_data.build_number, build_number)
+    self.assertEqual(
+        try_job_data.try_job_type,
+        failure_type.GetDescriptionForFailureType(failure_type.TEST))
+    self.assertFalse(try_job_data.has_compile_targets)
+    self.assertFalse(try_job_data.has_heuristic_results)