[Findit] Adding data model for capturing check flake analysis metadata

appengine/findit/model/base_swarming_task.py

 # 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.
 
 from google.appengine.ext import ndb
 
 from model import analysis_status
 
+
 class BaseSwarmingTask(ndb.Model):
   """Represents the progress of a general swarming task."""
   # The id of the Swarming task scheduled or running on Swarming Server.
   task_id = ndb.StringProperty(indexed=True)
 
   # A dict to keep track of running information for each test:
   # number of total runs, number of each status (such as 'SUCCESS' or 'FAILED')
   tests_statuses = ndb.JsonProperty(indexed=False, compressed=True)
 
   # The status of the swarming task.
   status = ndb.IntegerProperty(
       default=analysis_status.PENDING, indexed=False)
 
+  # An error dict containing an error code and message should this task fail
+  # unexpectedly. For example:
+  # {
+  #     'code': 1,
+  #     'message': 'Timed out'
+  # }
+  error = ndb.JsonProperty(indexed=False)
+
   # The revision of the failed build.
   build_revision = ndb.StringProperty(indexed=False)
 
   # Time when the task is created.
   created_time = ndb.DateTimeProperty(indexed=True)
   # Time when the task is started.
   started_time = ndb.DateTimeProperty(indexed=False)
   # Time when the task is completed.
   completed_time = ndb.DateTimeProperty(indexed=False)
 
   # parameters need to be stored and analyzed later.
   parameters = ndb.JsonProperty(default={}, indexed=False, compressed=True)