def test_build_new_rules_updates(self):
k = {'partition_list': [], 'turn_value': 99, 'stream_id': 'abc123'}
f = ('butter_lowpass', {"partition_list": [], "measure_list": ["where"]})
m = [([('location', 'geo')], [], [], [('turn', k, ['location'])])]
m2 = [([('where', 'geo')], [], [], [('turn', k, ['where'])])]
f2 = ('butter_lowpass', {"partition_list": [], "measure_list": ["location"]})
ff = ('butter_lowpass', {"partition_list": [], "measure_list": ["where"]})
r = update(self.dstream, [{'field': 'user_description', 'type': 'new', 'args': ['new shit'], 'kwargs': {}}], m, [])
self.assertEqual(len(r), 2)
self.assertEqual(r[0], 'ok')
self.assertEqual(len(r[1]['measures']), 1)
self.assertIn('turn_99.000000_location', r[1]['event_rules'])
rawr = deepcopy(r[1])
r2 = update(rawr, [], m2, [ff])
self.assertEqual(len(r2), 2)
self.assertEqual(r[0], 'ok')
self.assertEqual(len(r2[1]['measures']), 2)
self.assertEqual(len(r2[1]['filters']), 3)
self.assertIn('turn_99.000000_location', r2[1]['event_rules'])
self.assertIn('turn_99.000000_where', r2[1]['event_rules'])
d = DStream()
d['measures'] = {'location': {'val': None, 'dtype': 'geo'}}
t = update(d, [], [], [f2])
t2 = update(d, [], [], [f])
self.assertEqual(t[0], 'ok')
self.assertEqual(t2[0], 'invalid update')
def update_template(template_json, updates_list: list):
template = DStream()
template.load_from_json(template_json)
template['stream_token'] = template_json['stream_token']
old_template = deepcopy(template)
for update in updates_list:
update_fn = update_guide[update['field']][update['type']]['function']
args = []
args.append(template)
if update_guide[update['field']][update['type']]['field_key_arg']:
args.append(update['field'])
args.extend(update['args'])
kwargs = update['kwargs']
update_fn(*args, **kwargs)
if valid_update(template) is True:
template.publish_version()
return 'ok', template
else:
return 'invalid update', old_template, valid_update(template)
def new_event(template: DStream, event_key, event_dict):
template.add_event(event_key, event_dict)
def setUp(self):
template_dstream = DStream()
template_dstream.add_measure("length", "float")
template_dstream.add_measure("diameter", "float")
template_dstream.add_measure("whole_weight", "float")
template_dstream.add_measure("viscera_weight", "float")
template_dstream.add_user_id("sex")
template_dstream.add_field("rings")
self.template = template_dstream
self.mapping_list = [(0, ["user_ids", "sex"]),
(1, ["measures", "length", "val"]),
(2, ["measures", "diameter", "val"]),
(4, ["measures", "whole_weight", "val"]),
(6, ["measures", "viscera_weight", "val"]),
(8, ["fields", "rings"]), (3, ["timestamp"])]
self.endpoint = "http://localhost:5000/api/load"
self.data_format = "csv"
self.uid = "abaloneID"
self.test_data = "strom/data_puller/test/abalone0.csv"
config["walltime"] = 10
self.mqtt_context = MQTTContext(self.mapping_list,
self.template,
uid=self.uid,
data_format=self.data_format,
endpoint=self.endpoint,
userdata=config)
self.mqtt_reader = MQTTReader(self.mqtt_context)
def setUp(self):
template_dstream = DStream()
template_dstream.add_measure("length", "float")
template_dstream.add_measure("diameter", "float")
template_dstream.add_measure("whole_weight", "float")
template_dstream.add_measure("viscera_weight", "float")
template_dstream.add_user_id("sex")
template_dstream.add_field("rings")
self.template = template_dstream
self.mapping_list = [(0, ["user_ids", "sex"]),
(1, ["measures", "length", "val"]),
(2, ["measures", "diameter", "val"]),
(4, ["measures", "whole_weight", "val"]),
(6, ["measures", "viscera_weight", "val"]),
(8, ["fields", "rings"]), (3, ["timestamp"])]
self.url = "localhost:9092"
self.topic = b"data_pulling"
self.offset = 28
self.data_format = "csv"
self.timeout = 1000
self.test_data = "strom/data_puller/test/abalone0.csv"
self.kc = KafkaContext(self.mapping_list,
self.template,
url=self.url,
topic=self.topic,
offset=self.offset,
data_format=self.data_format,
timeout=self.timeout)
self.kafka_reader = KafkaReader(self.kc)
def setUp(self):
self.dir = "strom/data_puller/test/"
self.file_type = "csv"
template_dstream = DStream()
template_dstream.add_measure("length", "float")
template_dstream.add_measure("diameter", "float")
template_dstream.add_measure("whole_weight", "float")
template_dstream.add_measure("viscera_weight", "float")
template_dstream.add_user_id("sex")
template_dstream.add_field("rings")
self.template = template_dstream
self.mapping_list = [(0, ["user_ids", "sex"]),
(1, ["measures", "length", "val"]),
(2, ["measures", "diameter", "val"]),
(4, ["measures", "whole_weight", "val"]),
(6, ["measures", "viscera_weight", "val"]),
(8, ["fields", "rings"]), (3, ["timestamp"])]
self.delimiter = ","
self.endpoint = "http://localhost:5000/api/load"
self.dc = DirectoryContext(self.mapping_list,
self.template,
path=self.dir,
file_type=self.file_type,
delimiter=self.delimiter,
endpoint=self.endpoint)
self.source_reader = DirectoryReader(self.dc)
def setUp(self):
self.con1, self.con1b = Pipe()
self.con2, self.con2b = Pipe()
self.con3, self.con3b = Pipe()
self.con4, self.con4b = Pipe()
self.con5, self.con5b = Pipe()
self.con6, self.con6b = Pipe()
self.engine = Engine(self.con1b,
processors=2,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test1')
self.engine2 = Engine(self.con2b,
processors=2,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test2')
self.engine3 = Engine(self.con3b,
processors=2,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test3')
self.engine4 = Engine(self.con4b,
processors=2,
buffer_roll=1,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test4')
self.engine5 = Engine(self.con5b,
processors=2,
buffer_roll=1,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test5')
self.engine6 = Engine(self.con6b,
processors=2,
buffer_roll=1,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test6')
self.test_batch1 = [{
"stream_token": "abc123",
"message": "hi1"
}, {
"stream_token": "abc123",
"message": "hi2"
}, {
"stream_token": "abc123",
"message": "hi3"
}, {
"stream_token": "abc123",
"message": "hi4"
}]
self.test_batch2 = [{
"stream_token": "abc1234",
"message": "hello1"
}, {
"stream_token": "abc1234",
"message": "hello2"
}, {
"stream_token": "abc1234",
"message": "hello3"
}, {
"stream_token": "abc1234",
"message": "hello4"
}]
self.test_batch3 = [{
"stream_token": "abc123",
"message": "hi5"
}, {
"stream_token": "abc123",
"message": "hi6"
}, {
"stream_token": "abc123",
"message": "hi7"
}, {
"stream_token": "abc123",
"message": "hi8"
}]
self.test_batch4 = [{
"stream_token": "abc1234",
"message": "hello5"
}, {
"stream_token": "abc1234",
"message": "hello6"
}, {
"stream_token": "abc1234",
"message": "hello7"
}, {
"stream_token": "abc1234",
"message": "hello8"
}]
# roll results
self.test_batch5 = [{
"stream_token": "abc123",
"message": "hi4"
}, {
"stream_token": "abc123",
"message": "hi5"
}, {
"stream_token": "abc123",
"message": "hi6"
}, {
"stream_token": "abc123",
"message": "hi7"
}]
# send batches
self.test_batch6 = [{
"stream_token": "abc123",
"message": "hi9"
}, {
"stream_token": "abc123",
"message": "hi10"
}, {
"stream_token": "abc123",
"message": "hi11"
}, {
"stream_token": "abc123",
"message": "hi12"
}]
self.test_batch7 = [{
"stream_token": "abc123",
"message": "hi13"
}, {
"stream_token": "abc123",
"message": "hi14"
}, {
"stream_token": "abc123",
"message": "hi15"
}, {
"stream_token": "abc123",
"message": "hi16"
}]
# roll results
self.test_batch8 = [{
"stream_token": "abc123",
"message": "hi12"
}, {
"stream_token": "abc123",
"message": "hi13"
}, {
"stream_token": "abc123",
"message": "hi14"
}, {
"stream_token": "abc123",
"message": "hi15"
}]
self.test_batch_mix = [{
"stream_token": "abc123",
"message": "hi1"
}, {
"stream_token": "abc1234",
"message": "hello1"
}, {
"stream_token": "abc123",
"message": "hi2"
}, {
"stream_token": "abc1234",
"message": "hello2"
}, {
"stream_token": "abc123",
"message": "hi3"
}, {
"stream_token": "abc1234",
"message": "hello3"
}, {
"stream_token": "abc123",
"message": "hi4"
}, {
"stream_token": "abc1234",
"message": "hello4"
}]
self.test_batch_1to4 = self.test_batch1 + self.test_batch2 + self.test_batch3 + self.test_batch4
self.outfiles = []
self.abalone_con, self.abalone_conb = Pipe()
self.abalone = json.load(open(demo_data_dir + "demo_template_dir.txt"))
self.abalone_temp = DStream()
self.abalone_temp.load_from_json(self.abalone)
self.abalone_engine = Engine(
self.abalone_conb,
processors=2,
buffer_max_batch=10,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test_abalone')
def update_foreign_key(template: DStream, fk: str, old_fk=None):
if old_fk is not None:
prune_list_key(template, 'foreign_keys', old_fk)
template.add_fk(fk)
def update_field(template: DStream, new_field: str, old_field=None):
if old_field is not None:
prune_key(template, 'fields', old_field)
template.add_field(new_field)
def setUp(self):
demo_data_dir = "demo_data/"
self.dstream_dict = json.load(
open(demo_data_dir + "demo_template_unit_test.txt"))
self.dstream = DStream()
self.dstream.load_from_json(self.dstream_dict)
self.dstream['filters'][0]['transform_id'] = 1
self.dstream['filters'][1]['transform_id'] = 2
counter = 1
for dparam in self.dstream['dparam_rules']:
dparam['transform_id'] = counter
counter += 1
self.test_event_rules = {
"partition_list": [],
"measure_list": ["timestamp", "head1"],
"transform_type": "detect_event",
"transform_name": "DetectThreshold",
"param_dict": {
"event_rules": {
"measure": "head1",
"threshold_value": 69.2,
"comparison_operator": ">=",
"absolute_compare": True
},
"event_name": "nice_event",
"stream_id": "abc123",
},
"logical_comparison": "AND"
}
self.test_dparam_rules_list = [{
"partition_list": [("timestamp", 1510603551106, ">"),
("timestamp", 1510603551391, "<")],
"measure_list": ["timestamp", "timestamp_winning"],
"transform_type":
"derive_param",
"transform_name":
"DeriveSlope",
"param_dict": {
"func_params": {
"window_len": 1
},
"measure_rules": {
"rise_measure": "timestamp_winning",
"run_measure": "timestamp",
"output_name": "time_slope"
}
},
"logical_comparison":
"AND"
}, {
"partition_list": [],
"measure_list": [
"timestamp",
],
"transform_type": "derive_param",
"transform_name": "DeriveChange",
"param_dict": {
"func_params": {
"window_len": 1,
"angle_change": False
},
"measure_rules": {
"target_measure": "timestamp",
"output_name": "time_change"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"timestamp",
],
"transform_type": "derive_param",
"transform_name": "DeriveCumsum",
"param_dict": {
"func_params": {
"offset": 0
},
"measure_rules": {
"target_measure": "timestamp",
"output_name": "time_sum"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"timestamp",
],
"transform_type": "derive_param",
"transform_name": "DeriveWindowSum",
"param_dict": {
"func_params": {
"window_len": 3
},
"measure_rules": {
"target_measure": "timestamp",
"output_name": "time_window_sum"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"timestamp",
],
"transform_type": "derive_param",
"transform_name": "DeriveScaled",
"param_dict": {
"func_params": {
"scalar": -1
},
"measure_rules": {
"target_measure": "timestamp",
"output_name": "negatime"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveDistance",
"param_dict": {
"func_params": {
"window_len": 1,
"distance_func": "euclidean",
"swap_lon_lat": True
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "dist1"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveDistance",
"param_dict": {
"func_params": {
"window_len": 1,
"distance_func": "great_circle",
"swap_lon_lat": True
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "dist2"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveHeading",
"param_dict": {
"func_params": {
"window_len": 1,
"units": "deg",
"heading_type": "bearing",
"swap_lon_lat": True
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "head1"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveHeading",
"param_dict": {
"func_params": {
"window_len": 1,
"units": "deg",
"heading_type": "flat_angle",
"swap_lon_lat": True
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "head2"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveInBox",
"param_dict": {
"func_params": {
"upper_left_corner":
(-122.6835826856399, 45.515814287782455),
"lower_right_corner": (-122.678529, 45.511597)
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "boxy"
}
},
"logical_comparison": "AND"
}]
class TestFunFactory(unittest.TestCase):
def setUp(self):
demo_data_dir = "demo_data/"
self.dstream_dict = json.load(
open(demo_data_dir + "demo_template_unit_test.txt"))
self.dstream = DStream()
self.dstream.load_from_json(self.dstream_dict)
self.dstream['filters'][0]['transform_id'] = 1
self.dstream['filters'][1]['transform_id'] = 2
counter = 1
for dparam in self.dstream['dparam_rules']:
dparam['transform_id'] = counter
counter += 1
self.test_event_rules = {
"partition_list": [],
"measure_list": ["timestamp", "head1"],
"transform_type": "detect_event",
"transform_name": "DetectThreshold",
"param_dict": {
"event_rules": {
"measure": "head1",
"threshold_value": 69.2,
"comparison_operator": ">=",
"absolute_compare": True
},
"event_name": "nice_event",
"stream_id": "abc123",
},
"logical_comparison": "AND"
}
self.test_dparam_rules_list = [{
"partition_list": [("timestamp", 1510603551106, ">"),
("timestamp", 1510603551391, "<")],
"measure_list": ["timestamp", "timestamp_winning"],
"transform_type":
"derive_param",
"transform_name":
"DeriveSlope",
"param_dict": {
"func_params": {
"window_len": 1
},
"measure_rules": {
"rise_measure": "timestamp_winning",
"run_measure": "timestamp",
"output_name": "time_slope"
}
},
"logical_comparison":
"AND"
}, {
"partition_list": [],
"measure_list": [
"timestamp",
],
"transform_type": "derive_param",
"transform_name": "DeriveChange",
"param_dict": {
"func_params": {
"window_len": 1,
"angle_change": False
},
"measure_rules": {
"target_measure": "timestamp",
"output_name": "time_change"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"timestamp",
],
"transform_type": "derive_param",
"transform_name": "DeriveCumsum",
"param_dict": {
"func_params": {
"offset": 0
},
"measure_rules": {
"target_measure": "timestamp",
"output_name": "time_sum"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"timestamp",
],
"transform_type": "derive_param",
"transform_name": "DeriveWindowSum",
"param_dict": {
"func_params": {
"window_len": 3
},
"measure_rules": {
"target_measure": "timestamp",
"output_name": "time_window_sum"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"timestamp",
],
"transform_type": "derive_param",
"transform_name": "DeriveScaled",
"param_dict": {
"func_params": {
"scalar": -1
},
"measure_rules": {
"target_measure": "timestamp",
"output_name": "negatime"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveDistance",
"param_dict": {
"func_params": {
"window_len": 1,
"distance_func": "euclidean",
"swap_lon_lat": True
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "dist1"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveDistance",
"param_dict": {
"func_params": {
"window_len": 1,
"distance_func": "great_circle",
"swap_lon_lat": True
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "dist2"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveHeading",
"param_dict": {
"func_params": {
"window_len": 1,
"units": "deg",
"heading_type": "bearing",
"swap_lon_lat": True
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "head1"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveHeading",
"param_dict": {
"func_params": {
"window_len": 1,
"units": "deg",
"heading_type": "flat_angle",
"swap_lon_lat": True
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "head2"
}
},
"logical_comparison": "AND"
}, {
"partition_list": [],
"measure_list": [
"location",
],
"transform_type": "derive_param",
"transform_name": "DeriveInBox",
"param_dict": {
"func_params": {
"upper_left_corner":
(-122.6835826856399, 45.515814287782455),
"lower_right_corner": (-122.678529, 45.511597)
},
"measure_rules": {
"spatial_measure": "location",
"output_name": "boxy"
}
},
"logical_comparison": "AND"
}]
def test_update_stream_name(self):
update_stream_name(self.dstream, 'chadwick')
self.assertEqual(self.dstream['stream_name'], 'chadwick')
def test_update_source_key(self):
update_source_key(self.dstream, 'idd')
self.assertEqual(self.dstream['source_key'], 'idd')
def test_update_description(self):
update_description(self.dstream, 'lmao')
self.assertEqual(self.dstream['user_description'], 'lmao')
def test_update_user_id(self):
update_user_id(self.dstream, 'galactic_id', old_id='driver-id')
self.assertIn('galactic_id', self.dstream['user_ids'].keys())
self.assertNotIn('driver-id', self.dstream['user_ids'].keys())
def test_update_field(self):
update_field(self.dstream, 'dumb_field')
self.assertIn('dumb_field', self.dstream['fields'])
update_field(self.dstream, 'asteroid_field', old_field='dumb_field')
self.assertNotIn('dumb_field', self.dstream['fields'].keys())
self.assertIn('asteroid_field', self.dstream['fields'].keys())
def test_update_tag(self):
update_tag(self.dstream, 'not_poodles')
self.assertIn('not_poodles', self.dstream['tags'].keys())
update_tag(self.dstream, 'poodles', old_tag='not_poodles')
self.assertIn('poodles', self.dstream['tags'].keys())
self.assertNotIn('not_poodles', self.dstream['tags'].keys())
def test_update_fk(self):
update_foreign_key(self.dstream, 'ugh')
self.assertIn({'ugh': None}, self.dstream['foreign_keys'])
update_foreign_key(self.dstream, 'wah', old_fk='ugh')
self.assertIn({'wah': None}, self.dstream['foreign_keys'])
self.assertNotIn({'ugh': None}, self.dstream['foreign_keys'])
def test_update_rules(self):
update_rules(self.dstream, 'storage_rules',
[('store_raw', False), ('store_filtered', False)])
self.assertDictEqual(
{
"store_raw": False,
"store_filtered": False,
"store_derived": True
}, self.dstream['storage_rules'])
update_rules(self.dstream, 'ingest_rules',
[('is_this_all_fake', True)])
self.assertIn('is_this_all_fake', self.dstream['ingest_rules'].keys())
self.assertTrue(self.dstream['ingest_rules']['is_this_all_fake'])
def test_modify_filter(self):
modify_filter(self.dstream,
1, [('order', 3), ('nyquist', 0.69)],
new_partition_list=['whatever'],
change_comparison=True)
self.assertEqual(self.dstream['filters'][0]['param_dict']['order'], 3)
self.assertEqual(self.dstream['filters'][0]['param_dict']['nyquist'],
0.69)
self.assertEqual(self.dstream['filters'][0]['logical_comparison'],
'OR')
self.assertIn('whatever', self.dstream['filters'][0]['partition_list'])
def test_modify_dparam(self):
modify_dparam(self.dstream, 2, [('window_len', 2)])
self.assertEqual(
self.dstream['dparam_rules'][1]['param_dict']['func_params']
['window_len'], 2)
self.assertEqual(
len(self.dstream['dparam_rules'][1]['partition_list']), 0)
modify_dparam(self.dstream, 2, [], new_partition_list=['new'])
self.assertEqual(
list(self.dstream['dparam_rules'][1]['param_dict']
['func_params'].items()), [('window_len', 2),
('angle_change', False)])
self.assertEqual(self.dstream['dparam_rules'][1]['partition_list'],
['new'])
def test_modify_event(self):
modify_event(self.dstream, 'test_event', [('threshold_value', 70),
('absolute_compare', False)])
self.assertEqual(
self.dstream['event_rules']['test_event']['param_dict']
['event_rules']['threshold_value'], 70)
self.assertFalse(self.dstream['event_rules']['test_event']
['param_dict']['event_rules']['absolute_compare'])
def test_remove_transform(self):
remove_transform(self.dstream, 'filters', 2)
self.assertEqual(len(self.dstream['filters']), 1)
self.assertEqual(self.dstream['filters'][0]['transform_name'],
'ButterLowpass')
class TestFunFactory(unittest.TestCase):
def setUp(self):
demo_data_dir = "demo_data/"
self.dstream_dict = json.load(open(demo_data_dir + "demo_template_unit_test.txt"))
self.dstream = DStream()
self.dstream.load_from_json(self.dstream_dict)
self.dstream['filters'][0]['transform_id'] = 1
self.dstream['filters'][1]['transform_id'] = 2
counter = 1
for dparam in self.dstream['dparam_rules']:
dparam['transform_id'] = counter
counter += 1
self.test_event_rules = {
"partition_list": [],
"measure_list":["timestamp", "head1"],
"transform_type":"detect_event",
"transform_name":"DetectThreshold",
"param_dict":{
"event_rules":{
"measure":"head1",
"threshold_value":69.2,
"comparison_operator":">=",
"absolute_compare":True
},
"event_name":"nice_event",
"stream_id":"abc123",
},
"logical_comparison": "AND"
}
self.test_dparam_rules_list = [
{
"partition_list": [("timestamp", 1510603551106, ">"), ("timestamp", 1510603551391, "<")],
"measure_list":["timestamp", "timestamp_winning"],
"transform_type": "derive_param",
"transform_name": "DeriveSlope",
"param_dict":{
"func_params":{"window_len":1},
"measure_rules":{"rise_measure":"timestamp_winning", "run_measure":"timestamp","output_name":"time_slope"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["timestamp",],
"transform_type": "derive_param",
"transform_name": "DeriveChange",
"param_dict":{
"func_params":{"window_len":1, "angle_change":False},
"measure_rules":{"target_measure":"timestamp","output_name":"time_change"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["timestamp",],
"transform_type": "derive_param",
"transform_name": "DeriveCumsum",
"param_dict":{
"func_params":{"offset":0},
"measure_rules":{"target_measure":"timestamp","output_name":"time_sum"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["timestamp",],
"transform_type": "derive_param",
"transform_name": "DeriveWindowSum",
"param_dict":{
"func_params":{"window_len":3},
"measure_rules":{"target_measure":"timestamp","output_name":"time_window_sum"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["timestamp",],
"transform_type": "derive_param",
"transform_name": "DeriveScaled",
"param_dict":{
"func_params":{"scalar":-1},
"measure_rules":{"target_measure":"timestamp","output_name":"negatime"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["location",],
"transform_type": "derive_param",
"transform_name": "DeriveDistance",
"param_dict":{
"func_params":{"window_len":1, "distance_func":"euclidean", "swap_lon_lat":True},
"measure_rules":{"spatial_measure":"location","output_name":"dist1"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["location",],
"transform_type": "derive_param",
"transform_name": "DeriveDistance",
"param_dict":{
"func_params":{"window_len":1, "distance_func":"great_circle", "swap_lon_lat":True},
"measure_rules":{"spatial_measure":"location","output_name":"dist2"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["location",],
"transform_type": "derive_param",
"transform_name": "DeriveHeading",
"param_dict":{
"func_params":{"window_len":1, "units":"deg","heading_type":"bearing", "swap_lon_lat":True},
"measure_rules":{"spatial_measure":"location","output_name":"head1"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["location",],
"transform_type": "derive_param",
"transform_name": "DeriveHeading",
"param_dict":{
"func_params":{"window_len":1, "units":"deg","heading_type":"flat_angle", "swap_lon_lat":True},
"measure_rules":{"spatial_measure":"location","output_name":"head2"}
},
"logical_comparison":"AND"
},
{
"partition_list":[],
"measure_list":["location",],
"transform_type": "derive_param",
"transform_name": "DeriveInBox",
"param_dict":{
"func_params":{"upper_left_corner":(-122.6835826856399, 45.515814287782455), "lower_right_corner":(-122.678529, 45.511597)},
"measure_rules":{"spatial_measure":"location","output_name":"boxy"}
},
"logical_comparison":"AND"
}
]
def test_create_template(self):
t1 = create_template_dstream('tester', 'driver_id',[('location', 'geo')], ['driver-id', 'idd'], [('test_event', self.test_event_rules)], self.test_dparam_rules_list, [], {})
self.assertEqual(t1['stream_name'], 'tester')
self.assertEqual(t1['source_key'], 'driver_id')
self.assertIn('location', t1['measures'].keys())
self.assertEqual(t1['measures']['location']['dtype'], 'geo')
self.assertIn('driver-id', t1['user_ids'])
self.assertIn('idd', t1['user_ids'])
self.assertDictEqual(t1['storage_rules'], {"store_raw":True, "store_filtered":True, "store_derived":True})
self.assertIn('test_event', t1['event_rules'].keys())
def test_build_rules_from_event(self):
k1 = {'partition_list': [], 'turn_value': 45, 'stream_id': 'abc123'}
r1 = build_rules_from_event('turn', [('location', 'geo')], **k1)
for i in ['event_rules', 'dparam_rules', 'filter_rules']:
self.assertIn(i, r1)
self.assertEqual(r1['event_rules'][0], 'turn_45.000000_location')
self.assertEqual(len(r1['dparam_rules']), 2)
# with self.assertRaises(ValueError):
# build_rules_from_event('turn', [('smokeation', 'smokey')], **k1)
k2 = {'partition_list': [], 'urn_value': 45, 'stream_id': 'abc123'}
with self.assertRaises(ValueError):
build_rules_from_event('turn', [('location', 'geo')], **k2)
k3 = {'partition_list': [], 'turn_value': 45}
with self.assertRaises(ValueError):
build_rules_from_event('turn', [('location', 'geo')], **k3)
k4 = {'turn_value': 45, 'stream_id': 'abc123'}
with self.assertRaises(ValueError):
build_rules_from_event('turn', [('location', 'geo')], **k4)
def test_build_temp_event_filters(self):
skey = 'driver_id'
uids = ['driver-id', 'idd']
k1 = {'partition_list': [], 'turn_value': 45, 'stream_id': 'abc123'}
f1 = ('butter_lowpass', {"partition_list": [], "measure_list": ["location"]})
f2 = ('butter_lowpass', {"partition_list": [], "measure_list": ["location"]})
m = [([('location', 'geo')], [], [], [('turn', k1, ['location'])])]
m2 = [([('location', 'geo')], [], [], [('turn', k1, ['location_buttered']),])]
t = build_template('test', skey, m, uids, [f1])
self.assertIn('turn_45.000000_location', t['event_rules'])
t2 = build_template('test', skey, m2, uids, [f2])
self.assertIn('turn_45.000000_location_buttered', t2['event_rules'])
def test_build_temp_dparam(self):
f = ('butter_lowpass', {"partition_list": [], "measure_list": ["location"]})
m = [([('location', 'geo')], [], [('heading', {'partition_list': [], 'measure_list': ['location'], }, {'spatial_measure': 'location'})], [])]
t = build_template('test','driver_id', m, ['driver-id', 'idd'], [])
self.assertEqual(len(t['dparam_rules']), 1)
self.assertEqual(t['dparam_rules'][0]['transform_name'], 'DeriveHeading')
m2 = [([('location', 'geo')], [], [('heading', {'partition_list': [], 'measure_list': ['location_buttered'], }, {'spatial_measure': 'location_buttered'})], [])]
t2 = build_template('test','driver_id', m2, ['driver-id', 'idd'], [f])
def test_build_template_event(self):
skey = 'driver_id'
uids = ['driver-id', 'idd']
k1 = {'partition_list': [], 'turn_value': 45, 'stream_id': 'abc123'}
k2 = {'partition_list': [], 'turn_value': 66, 'stream_id': 'abc123'}
k3 = {'partition_list': [], 'turn_value': 30, 'stream_id': 'abc123'}
k4 = {'partition_list': [], 'turn_value': 57, 'stream_id': 'abc123'}
m = [([('location', 'geo')], [], [], [('turn', k1, ['location']), ('turn', k2, ['location'])])]
m2 = [([('location', 'geo'), ('smokeation', 'geo')], [], [], [('turn', k1, ['location']), ('turn', k2, ['smokeation'])])]
m3 = [([('location', 'geo')], [], [], [('turn', k1, ['location']), ('turn', k2, ['location'])]), ([('smokeation', 'geo')], [], [], [('turn', k3, ['smokeation']), ('turn', k4, ['smokeation'])])]
m4 = [([('location', 'geo')], [], [], [('turn', k1, ['smokeation'])])]
t = build_template('test', skey, m, uids, [])
self.assertEqual(len(t['event_rules']), 2)
self.assertEqual(len(t['measures']), 1)
self.assertEqual(len(t['dparam_rules']), 4)
for i in ['turn_45.000000_location', 'turn_66.000000_location']:
self.assertIn(i, t['event_rules'])
t2 = build_template('test', skey, m2, uids, [])
self.assertEqual(len(t2['event_rules']), 2)
self.assertEqual(len(t2['measures']), 2)
self.assertEqual(len(t2['dparam_rules']), 4)
for i in ['turn_45.000000_location', 'turn_66.000000_smokeation']:
self.assertIn(i, t2['event_rules'])
t3 = build_template('test', skey, m3, uids, [])
self.assertEqual(len(t3['event_rules']), 4)
self.assertEqual(len(t3['measures']), 2)
self.assertEqual(len(t3['dparam_rules']), 8)
for i in ['turn_45.000000_location', 'turn_66.000000_location', 'turn_30.000000_smokeation', 'turn_57.000000_smokeation']:
self.assertIn(i, t3['event_rules'])
with self.assertRaises(ValueError):
build_template('test', skey, m4, uids, [])
#
# with self.assertRaises(TypeError):
# build_template('test', skey, [1,2,3,4], uids, [])
#
# with self.assertRaises(TypeError):
# build_template('test', skey,(1,2,3), uids, [])
def test_update_template(self):
name_update = {'field': 'stream_name', 'type': 'new', 'args': ['shit'], 'kwargs': {}}
desc_update = {'field': 'user_description', 'type': 'new', 'args': ['i hate this shit'], 'kwargs': {}}
source_key_update = {'field': 'source_key', 'type': 'new', 'args': ['vom_id'], 'kwargs': {}}
user_id_update1 = {'field': 'user_ids', 'type': 'new', 'args': ['bananas'], 'kwargs': {}}
user_id_update2 = {'field': 'user_ids', 'type': 'new', 'args': ['shit_kiwis'], 'kwargs': {'old_id': 'id'}}
user_id_update3 = {'field': 'user_ids', 'type': 'remove', 'args': ['driver-id'], 'kwargs': {}}
field_update1 = {'field': 'fields', 'type': 'new', 'args': ['field_of_garbage'], 'kwargs': {}}
field_update2 = {'field': 'fields', 'type': 'new', 'args': ['field_of_trash'], 'kwargs': {'old_field': 'region-code'}}
field_update3 = {'field': 'fields', 'type': 'remove', 'args': ['field_of_garbage'], 'kwargs': {}}
tag_update1 = {'field': 'tags', 'type': 'new', 'args': ['hash'], 'kwargs': {}}
tag_update2 = {'field': 'tags', 'type': 'new', 'args': ['price'], 'kwargs': {}}
tag_update3 = {'field': 'tags', 'type': 'new', 'args': ['toe',], 'kwargs': {'old_tag': 'hash'}}
tag_update4 = {'field': 'tags', 'type': 'remove', 'args': ['price'], 'kwargs': {}}
fk_update1 = {'field': 'foreign_keys', 'type': 'new', 'args': ['romania'], 'kwargs': {}}
fk_update2 = {'field': 'foreign_keys', 'type': 'new', 'args': ['slovakia'], 'kwargs': {}}
fk_update3 = {'field': 'foreign_keys', 'type': 'new', 'args': ['lithuania',], 'kwargs': {'old_fk': 'romania'}}
fk_update4 = {'field': 'foreign_keys', 'type': 'remove', 'args': ['slovakia'], 'kwargs': {}}
storage_update = {'field': 'storage_rules', 'type': 'modify', 'args': [[('store_raw', False), ('store_derived', False)]], 'kwargs': {}}
ingest_update = {'field': 'ingest_rules', 'type': 'modify', 'args': [[('im_real', False),]], 'kwargs': {}}
engine_update = {'field': 'engine_rules', 'type': 'modify',
'args': [[('fuck_buffer', True), ('notfuck_buffer', False)]], 'kwargs': {}}
measure_update_add = {'field': 'measures', 'type': 'new', 'args': [('poodles', 'poodle')], 'kwargs': {}}
measure_update_remove_ok = {'field': 'measures', 'type': 'remove', 'args': ['poodles'], 'kwargs': {}}
measure_update_remove_bad = {'field': 'measures', 'type': 'remove', 'args': ['location'], 'kwargs': {}}
filter_update_add = {'field': 'filters', 'type': 'new', 'args': [{ 'transform_id': 3,'fake_filter': 'yes', 'param_dict': {'filter_name': '_fake'}, 'measure_list': []}], 'kwargs': {}}
filter_update_modify = {'field': 'filters', 'type': 'modify', 'args': [1, [('order', 1)]], 'kwargs': {}}
filter_update_remove_ok = {'field': 'filters', 'type': 'remove', 'args': [3], 'kwargs': {}}
filter_update_remove_bad = {'field': 'filters', 'type': 'remove', 'args': [2], 'kwargs': {}}
dparam_update_add = {'field': 'dparam_rules', 'type': 'new', 'args': [{'transform_id': 15, 'fake_param': 'yes', 'measure_list': [], 'param_dict': {'measure_rules': {'output_name': 'new'}}}], 'kwargs': {}}
dparam_update_modify = {'field': 'dparam_rules', 'type': 'modify', 'args': [9, [('window_len', 2)]], 'kwargs': {'new_partition_list': ['dumb']}}
dparam_update_remove_ok = {'field': 'dparam_rules', 'type': 'remove', 'args': [15], 'kwargs': {}}
dparam_update_remove_bad = {'field': 'dparam_rules', 'type': 'remove', 'args': [8], 'kwargs': {}}
event_update_add = {'field': 'event_rules', 'type': 'new', 'args': ['fuck_this_event', {
"partition_list": [],
"measure_list":["timestamp", "head1"],
"transform_type":"detect_event",
"transform_name":"DetectThreshold",
"param_dict":{
"event_rules":{
"measure":"head1",
"threshold_value":69.2,
"comparison_operator":">=",
"absolute_compare":True
},
"event_name":"nice_event",
"stream_id":"abc123",
},
"logical_comparison": "AND"
}], 'kwargs': {}}
event_update_modify = {'field': 'event_rules', 'type': 'modify', 'args': ['fuck_this_event', [('threshold_value', 70)]], 'kwargs': {}}
update1 = [name_update, desc_update, source_key_update, user_id_update1, user_id_update1, field_update1, fk_update1, tag_update1, storage_update, engine_update, ingest_update, measure_update_add, filter_update_add, dparam_update_add, event_update_add]
update2 = [user_id_update2, field_update2, tag_update2, tag_update3, fk_update2, fk_update3, filter_update_modify, dparam_update_modify, event_update_modify]
update3 = [user_id_update3, field_update3, fk_update4, tag_update4, measure_update_remove_ok, filter_update_remove_ok, dparam_update_remove_ok]
update4 = [measure_update_remove_bad, filter_update_remove_bad, dparam_update_remove_bad]
# update 1
result1 = update_template(self.dstream, update1)
updated_template = result1[1]
self.assertEqual(result1[0], 'ok')
self.assertEqual(updated_template['stream_name'], 'shit')
self.assertEqual(updated_template['user_description'], 'i hate this shit')
self.assertEqual(updated_template['source_key'], 'vom_id')
self.assertIn('bananas', updated_template['user_ids'])
self.assertIn('field_of_garbage', updated_template['fields'])
self.assertIn({'romania': None}, updated_template['foreign_keys'])
self.assertIn('hash', updated_template['tags'])
self.assertDictEqual(updated_template['storage_rules'], {"store_raw":False, "store_filtered":True, "store_derived":False})
self.assertDictEqual(updated_template['engine_rules'], {'kafka': 'test', 'fuck_buffer': True, 'notfuck_buffer': False})
self.assertDictEqual(updated_template['ingest_rules'], {'im_real': False})
self.assertEqual(len(updated_template['measures'].keys()), 2)
self.assertEqual(len(updated_template['filters']), 3)
self.assertEqual(len(updated_template['dparam_rules']), 11)
self.assertIn('fuck_this_event', updated_template['event_rules'])
# update 2
result2 = update_template(updated_template, update2)
updated_template2 = result2[1]
self.assertEqual(result2[0], 'ok')
self.assertEqual(len(updated_template2['user_ids']), 3)
self.assertIn('shit_kiwis', updated_template2['user_ids'])
self.assertNotIn('id', updated_template2['user_ids'])
self.assertEqual(len(updated_template2['fields']), 2)
self.assertIn('field_of_trash', updated_template2['fields'])
self.assertNotIn('region-code', updated_template2['fields'])
self.assertEqual(len(updated_template2['tags']), 2)
self.assertIn('price', updated_template2['tags'])
self.assertIn('toe', updated_template2['tags'])
self.assertNotIn('hash', updated_template2['tags'])
self.assertEqual(len(updated_template2['foreign_keys']), 2)
self.assertIn({'slovakia': None}, updated_template2['foreign_keys'])
self.assertIn({'lithuania': None}, updated_template2['foreign_keys'])
self.assertNotIn({'romania': None}, updated_template2['foreign_keys'])
self.assertEqual(updated_template2['filters'][0]['param_dict']['order'], 1)
self.assertEqual(updated_template2['dparam_rules'][8]['param_dict']['func_params']['window_len'], 2)
self.assertEqual(updated_template2['dparam_rules'][8]['partition_list'], ['dumb'])
self.assertEqual(updated_template2['event_rules']['fuck_this_event']['param_dict']['event_rules']['threshold_value'], 70)
# update 3
result3 = update_template(updated_template2, update3)
updated_template3 = result3[1]
self.assertEqual(result3[0], 'ok')
self.assertEqual(len(updated_template3['user_ids']), 2)
self.assertNotIn('driver-id', updated_template3['user_ids'])
self.assertEqual(len(updated_template3['fields']), 1)
self.assertNotIn('field_of_garbage', updated_template3['fields'])
self.assertEqual(len(updated_template3['tags']), 1)
self.assertNotIn('price', updated_template3['tags'])
self.assertEqual(len(updated_template3['foreign_keys']), 1)
self.assertNotIn({'slovakia': None}, updated_template3['foreign_keys'])
self.assertEqual(len(updated_template3['measures'].keys()), 1)
self.assertEqual(len(updated_template3['filters']), 2)
self.assertEqual(len(updated_template3['dparam_rules']), 10)
# update 4
result4 = update_template(updated_template3, update4)
self.assertEqual(result4[0], 'invalid update')
bad_guys = result4[2]
self.assertEqual(len(bad_guys), 7)
self.assertIn(('derived param', 'DeriveSlope', 'measure', 'timestamp_winning'), bad_guys)
self.assertIn(('derived param', 'DeriveDistance', 'measure', 'location'), bad_guys)
self.assertIn(('derived param', 'DeriveDistance', 'measure', 'location'), bad_guys)
self.assertIn(('derived param', 'DeriveHeading', 'measure', 'location'), bad_guys)
self.assertIn(('derived param', 'DeriveInBox', 'measure', 'location'), bad_guys)
self.assertIn(('event', 'test_event', 'derived param', 'head1'), bad_guys)
self.assertIn(('event', 'fuck_this_event', 'derived param', 'head1'), bad_guys)
#
def test_build_data_rules(self):
source_inds = [0,1,2,4,6,8,3]
t_keys = [["user_ids","sex"],["measures","length", "val"],["measures","diameter", "val"],["measures","whole_weight", "val"],["measures","viscera_weight", "val"],["fields","rings"],["timestamp"]]
d = build_data_rules(source_inds, t_keys)
self.assertDictEqual(d, {'mapping_list': [(0, ['user_ids', 'sex']), (1, ['measures', 'length', 'val']), (2, ['measures', 'diameter', 'val']), (4, ['measures', 'whole_weight', 'val']), (6, ['measures', 'viscera_weight', 'val']), (8, ['fields', 'rings']), (3, ['timestamp'])], 'date_format': None, 'puller': {}, 'pull': False})
#
d2 = build_data_rules(source_inds, t_keys, puller=['dir', [['path', 'strom/data_puller/test/'], ['file_type', 'csv']]])
self.assertDictEqual(d2, {'mapping_list': [(0, ['user_ids', 'sex']), (1, ['measures', 'length', 'val']), (2, ['measures', 'diameter', 'val']), (4, ['measures', 'whole_weight', 'val']), (6, ['measures', 'viscera_weight', 'val']), (8, ['fields', 'rings']), (3, ['timestamp'])], 'date_format': None, 'puller': {'type': 'dir', 'inputs': {'path': 'strom/data_puller/test/', 'file_type': 'csv'}}, 'pull': True})
d3 = build_data_rules(source_inds, t_keys, puller=['dir', [['path', 'strom/data_puller/test/'], ['file_type', 'csv'], ['delimiter', ',']]])
self.assertDictEqual(d3, {'mapping_list': [(0, ['user_ids', 'sex']), (1, ['measures', 'length', 'val']), (2, ['measures', 'diameter', 'val']), (4, ['measures', 'whole_weight', 'val']), (6, ['measures', 'viscera_weight', 'val']), (8, ['fields', 'rings']), (3, ['timestamp'])], 'date_format': None, 'puller': {'type': 'dir', 'inputs': {'path': 'strom/data_puller/test/', 'file_type': 'csv', 'delimiter': ','}}, 'pull': True} )
def test_build_new_rules_updates(self):
k = {'partition_list': [], 'turn_value': 99, 'stream_id': 'abc123'}
f = ('butter_lowpass', {"partition_list": [], "measure_list": ["where"]})
m = [([('location', 'geo')], [], [], [('turn', k, ['location'])])]
m2 = [([('where', 'geo')], [], [], [('turn', k, ['where'])])]
f2 = ('butter_lowpass', {"partition_list": [], "measure_list": ["location"]})
ff = ('butter_lowpass', {"partition_list": [], "measure_list": ["where"]})
r = update(self.dstream, [{'field': 'user_description', 'type': 'new', 'args': ['new shit'], 'kwargs': {}}], m, [])
self.assertEqual(len(r), 2)
self.assertEqual(r[0], 'ok')
self.assertEqual(len(r[1]['measures']), 1)
self.assertIn('turn_99.000000_location', r[1]['event_rules'])
rawr = deepcopy(r[1])
r2 = update(rawr, [], m2, [ff])
self.assertEqual(len(r2), 2)
self.assertEqual(r[0], 'ok')
self.assertEqual(len(r2[1]['measures']), 2)
self.assertEqual(len(r2[1]['filters']), 3)
self.assertIn('turn_99.000000_location', r2[1]['event_rules'])
self.assertIn('turn_99.000000_where', r2[1]['event_rules'])
d = DStream()
d['measures'] = {'location': {'val': None, 'dtype': 'geo'}}
t = update(d, [], [], [f2])
t2 = update(d, [], [], [f])
self.assertEqual(t[0], 'ok')
self.assertEqual(t2[0], 'invalid update')
def setUp(self):
template_dstream = DStream()
template_dstream['data_rules'] = {
"pull": True,
"puller": {
"type": "dir",
"inputs": {
"path": "strom/data_puller/test/",
"file_type": "csv",
"delimiter": ","
}
},
"mapping_list": [(0,["user_ids","sex"]), (1,["measures","length", "val"]), (2,["measures","diameter", "val"]), (4,["measures","whole_weight", "val"]), (6,["measures","viscera_weight", "val"]), (8,["fields","rings"]), (3,["timestamp"])]
}
template_dstream.add_measure("length", "float")
template_dstream.add_measure("diameter", "float")
template_dstream.add_measure("whole_weight", "float")
template_dstream.add_measure("viscera_weight", "float")
template_dstream.add_user_id("sex")
template_dstream.add_field("rings")
self.template = template_dstream
self.q = Queue()
self.puller = DataPuller(self.template, self.q)
def setUp(self):
self.dstream = DStream()
class TestDStream(unittest.TestCase):
def setUp(self):
self.dstream = DStream()
def test_init(self):
init_keys = [
'stream_name', 'user_description', 'version', 'stream_token',
'source_key', 'template_id', 'storage_rules', 'ingest_rules',
'engine_rules', 'timestamp', 'measures', 'fields', 'user_ids',
'tags', 'foreign_keys', 'filters', 'dparam_rules', 'event_rules',
'data_rules'
]
for key in init_keys:
print(self.dstream.keys())
self.assertIn(key, self.dstream.keys())
def testadd_methods(self):
self.assertIsInstance(self.dstream["stream_token"], str)
m_name = "viscosity"
m_dtype = "float"
self.dstream.add_measure(m_name, m_dtype)
self.assertTrue(m_name in self.dstream["measures"].keys())
self.assertEqual(self.dstream["measures"][m_name]["dtype"], m_dtype)
f_name = "strawberry"
self.dstream.add_field(f_name)
self.assertTrue(f_name in self.dstream["fields"].keys())
uid_name = "my_id"
self.dstream.add_user_id(uid_name)
self.assertTrue(uid_name in self.dstream["user_ids"].keys())
tag_name = "Really good sensor"
self.dstream.add_tag(tag_name)
self.assertIn(tag_name, self.dstream["tags"])
fk = "key to the city"
self.dstream.add_fk(fk)
self.assertTrue({fk: None} in self.dstream["foreign_keys"])
fake_filter = {"func_name": "Make all values 0"}
self.dstream.add_filter(fake_filter)
self.assertEqual(fake_filter, self.dstream["filters"][0])
fake_dparam = {"measure": "viscosity", "drule": "max of mins"}
self.dstream.add_derived_param(fake_dparam)
self.assertEqual(fake_dparam, self.dstream["dparam_rules"][0])
fake_event_name = "My birthday"
fake_event = {"param": "viscosity", "threshold": "too viscous"}
self.dstream.add_event(fake_event_name, fake_event)
self.assertEqual(fake_event,
self.dstream["event_rules"][fake_event_name])
old_version = self.dstream["version"]
self.dstream.publish_version()
self.assertEqual(old_version + 1, self.dstream["version"])
fake_mapping = ["fake", "mapping", "list"]
self.dstream.add_data_rules(fake_mapping)
self.assertEqual(fake_mapping, self.dstream["data_rules"])
def test_load_from_json(self):
test_dict = {"stream_token": "foo", "version": 900}
self.dstream.load_from_json((test_dict))
self.assertEqual(test_dict["version"], self.dstream["version"])
self.assertIsInstance(self.dstream["stream_token"], str)
def test_filter(self):
self.dstream.add_filter({"test": "filterdict"})
def new_measure(template: DStream, measure_tup):
template.add_measure(measure_tup[0], measure_tup[1])
def update_user_id(template: DStream, new_id: str, old_id=None):
if old_id is not None:
prune_key(template, 'user_ids', old_id)
template.add_user_id(new_id)
def setUp(self):
template_dstream = DStream()
template_dstream.add_measure("length", "float")
template_dstream.add_measure("diameter", "float")
template_dstream.add_measure("whole_weight", "float")
template_dstream.add_measure("viscera_weight", "float")
template_dstream.add_user_id("sex")
template_dstream.add_field("rings")
self.template = template_dstream
self.mapping_list = [(0, ["user_ids", "sex"]),
(1, ["measures", "length", "val"]),
(2, ["measures", "diameter", "val"]),
(4, ["measures", "whole_weight", "val"]),
(6, ["measures", "viscera_weight", "val"]),
(8, ["fields", "rings"]), (3, ["timestamp"])]
self.csvf = CSVFormatter(self.mapping_list, self.template)
def update_tag(template: DStream, tag_name: str, old_tag=None):
if old_tag is not None:
prune_key(template, 'tags', old_tag)
template.add_tag(tag_name)
def create_template_dstream(strm_nm, src_key, measures: list, uids: list, events: list, dparam_rules: list, filters: list, data_rules: dict, usr_dsc="", storage_rules=None, ingest_rules=None, engine_rules=None, foreign_keys=None, tags=None, fields=None):
# measures: list of tuples (measure_name, dtype)
template = DStream()
template['stream_name'] = strm_nm
template['source_key'] = src_key
template.add_measures(measures) # expects list of tuples (measure, dtype)
template.add_user_ids(uids)
template['user_description'] = usr_dsc
template.add_dparams(dparam_rules)
template.add_events(events)
template.add_filters(filters)
template.add_data_rules(data_rules)
if storage_rules is not None: # add else branch w default
template['storage_rules'] = storage_rules
else:
template['storage_rules'] = {"store_raw":True, "store_filtered":True, "store_derived":True}
if ingest_rules is not None: # add else branch w default
template['ingest_rules'] = ingest_rules
if engine_rules is not None: # add else branch w default
template['engine_rules'] = engine_rules
# the totally optional section
if tags is not None:
template.add_tags(tags)
if foreign_keys is not None:
template.add_foreign_keys(foreign_keys)
if fields is not None:
template.add_fields(fields)
return template
def _dstream_new(self):
tk['Server._dstream_new'].start()
dstream = DStream()
tk['Server._dstream_new'].stop()
return dstream
class TestEngineThread(unittest.TestCase):
def setUp(self):
self.con1, self.con1b = Pipe()
self.con2, self.con2b = Pipe()
self.con3, self.con3b = Pipe()
self.con4, self.con4b = Pipe()
self.con5, self.con5b = Pipe()
self.con6, self.con6b = Pipe()
self.engine = Engine(self.con1b,
processors=2,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test1')
self.engine2 = Engine(self.con2b,
processors=2,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test2')
self.engine3 = Engine(self.con3b,
processors=2,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test3')
self.engine4 = Engine(self.con4b,
processors=2,
buffer_roll=1,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test4')
self.engine5 = Engine(self.con5b,
processors=2,
buffer_roll=1,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test5')
self.engine6 = Engine(self.con6b,
processors=2,
buffer_roll=1,
buffer_max_batch=4,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test6')
self.test_batch1 = [{
"stream_token": "abc123",
"message": "hi1"
}, {
"stream_token": "abc123",
"message": "hi2"
}, {
"stream_token": "abc123",
"message": "hi3"
}, {
"stream_token": "abc123",
"message": "hi4"
}]
self.test_batch2 = [{
"stream_token": "abc1234",
"message": "hello1"
}, {
"stream_token": "abc1234",
"message": "hello2"
}, {
"stream_token": "abc1234",
"message": "hello3"
}, {
"stream_token": "abc1234",
"message": "hello4"
}]
self.test_batch3 = [{
"stream_token": "abc123",
"message": "hi5"
}, {
"stream_token": "abc123",
"message": "hi6"
}, {
"stream_token": "abc123",
"message": "hi7"
}, {
"stream_token": "abc123",
"message": "hi8"
}]
self.test_batch4 = [{
"stream_token": "abc1234",
"message": "hello5"
}, {
"stream_token": "abc1234",
"message": "hello6"
}, {
"stream_token": "abc1234",
"message": "hello7"
}, {
"stream_token": "abc1234",
"message": "hello8"
}]
# roll results
self.test_batch5 = [{
"stream_token": "abc123",
"message": "hi4"
}, {
"stream_token": "abc123",
"message": "hi5"
}, {
"stream_token": "abc123",
"message": "hi6"
}, {
"stream_token": "abc123",
"message": "hi7"
}]
# send batches
self.test_batch6 = [{
"stream_token": "abc123",
"message": "hi9"
}, {
"stream_token": "abc123",
"message": "hi10"
}, {
"stream_token": "abc123",
"message": "hi11"
}, {
"stream_token": "abc123",
"message": "hi12"
}]
self.test_batch7 = [{
"stream_token": "abc123",
"message": "hi13"
}, {
"stream_token": "abc123",
"message": "hi14"
}, {
"stream_token": "abc123",
"message": "hi15"
}, {
"stream_token": "abc123",
"message": "hi16"
}]
# roll results
self.test_batch8 = [{
"stream_token": "abc123",
"message": "hi12"
}, {
"stream_token": "abc123",
"message": "hi13"
}, {
"stream_token": "abc123",
"message": "hi14"
}, {
"stream_token": "abc123",
"message": "hi15"
}]
self.test_batch_mix = [{
"stream_token": "abc123",
"message": "hi1"
}, {
"stream_token": "abc1234",
"message": "hello1"
}, {
"stream_token": "abc123",
"message": "hi2"
}, {
"stream_token": "abc1234",
"message": "hello2"
}, {
"stream_token": "abc123",
"message": "hi3"
}, {
"stream_token": "abc1234",
"message": "hello3"
}, {
"stream_token": "abc123",
"message": "hi4"
}, {
"stream_token": "abc1234",
"message": "hello4"
}]
self.test_batch_1to4 = self.test_batch1 + self.test_batch2 + self.test_batch3 + self.test_batch4
self.outfiles = []
self.abalone_con, self.abalone_conb = Pipe()
self.abalone = json.load(open(demo_data_dir + "demo_template_dir.txt"))
self.abalone_temp = DStream()
self.abalone_temp.load_from_json(self.abalone)
self.abalone_engine = Engine(
self.abalone_conb,
processors=2,
buffer_max_batch=10,
buffer_max_seconds=5,
test_mode=True,
test_outfile='engine_test_output/engine_test_abalone')
def tearDown(self):
print("Tear it all down")
sleep(1)
for o in self.outfiles:
print(o)
try:
remove_outfile(o)
except OSError as oserr:
print(oserr)
self.outfiles = []
def test_buffer1(self):
outfiles = [
'engine_test_output/engine_test1_abc123_1.txt',
'engine_test_output/engine_test1_abc123_2.txt',
'engine_test_output/engine_test1_abc1234_1.txt',
'engine_test_output/engine_test1_abc1234_2.txt',
]
self.engine.start()
# sleep(5)
# test all sent to processor in batches of 4
for i in self.test_batch_1to4:
self.con1.send((i, 'load'))
sleep(5)
result = []
for o in outfiles:
result.extend(read_outfile(o))
self.assertEqual(len(result), 4)
self.assertIn(self.test_batch1, result)
self.assertIn(self.test_batch2, result)
self.assertIn(self.test_batch3, result)
self.assertIn(self.test_batch4, result)
self.outfiles.extend(outfiles)
self.con1.send("stop_poison_pill")
def test_buffer2(self):
# test all sent to processor in batches of 4 - GROUPED BY TOKEN (2 buffs)
outfiles = [
'engine_test_output/engine_test2_abc123_1.txt',
'engine_test_output/engine_test2_abc1234_1.txt'
]
self.engine2.start()
sleep(5)
for i in self.test_batch_mix:
self.con2.send((i, 'load'))
sleep(5)
result2 = []
for o in outfiles:
result2.extend(read_outfile(o))
self.assertEqual(len(result2), 2)
self.assertIn(self.test_batch1, result2)
self.assertIn(self.test_batch2, result2)
self.outfiles.extend(outfiles)
self.con2.send("stop_poison_pill")
def test_buffer3(self):
# leftovers
outfiles = [
'engine_test_output/engine_test3_abc123_1.txt',
'engine_test_output/engine_test3_abc1234_1.txt'
]
self.engine3.start()
sleep(5)
for i in self.test_batch1[:2]:
self.con3.send((i, 'load'))
for i in self.test_batch2[:2]:
self.con3.send((i, 'load'))
sleep(7)
result3 = []
for o in outfiles:
result3.extend(read_outfile(o))
self.assertEqual(len(result3), 2)
for x in result3:
self.assertEqual(len(x), 2)
self.outfiles.extend(outfiles)
self.con3.send("stop_poison_pill")
def test_buffer4(self):
# w rolling window
outfiles = [
'engine_test_output/engine_test4_abc123_1.txt',
'engine_test_output/engine_test4_abc123_2.txt',
'engine_test_output/engine_test4_abc123_3.txt',
]
self.engine4.start()
sleep(5)
for i in self.test_batch1 + self.test_batch3:
self.con4.send((i, 'load'))
sleep(7)
result4 = []
for o in outfiles:
result4.extend(read_outfile(o))
# for r in result4:
# print(r)
self.assertEqual(len(result4), 3)
self.assertIn(self.test_batch1, result4)
self.assertIn(self.test_batch5, result4)
self.assertIn(self.test_batch3[-2:], result4)
self.outfiles.extend(outfiles)
self.con4.send("stop_poison_pill")
def test_buffer5(self):
# no leftovers that are just buffer roll
outfiles = [
'engine_test_output/engine_test5_abc123_1.txt',
]
self.engine5.start()
sleep(5)
for i in self.test_batch1:
self.con5.send((i, 'load'))
sleep(7)
result5 = []
for o in outfiles:
result5.extend(read_outfile(o))
self.assertEqual(len(result5), 1)
self.assertIn(self.test_batch1, result5)
self.outfiles.extend(outfiles)
self.con5.send("stop_poison_pill")
def test_buffer6(self):
# row resets correctly after leftovers
outfiles = [
'engine_test_output/engine_test6_abc123_1.txt',
'engine_test_output/engine_test6_abc123_2.txt',
'engine_test_output/engine_test6_abc123_3.txt',
'engine_test_output/engine_test6_abc123_4.txt',
'engine_test_output/engine_test6_abc123_5.txt',
'engine_test_output/engine_test6_abc123_6.txt',
]
self.engine6.start()
sleep(5)
for i in self.test_batch1:
self.con6.send((i, 'load'))
for i in self.test_batch3:
self.con6.send((i, 'load'))
sleep(7)
for i in self.test_batch6:
self.con6.send((i, 'load'))
for i in self.test_batch7:
self.con6.send((i, 'load'))
sleep(7)
result6 = []
for o in outfiles:
result6.extend(read_outfile(o))
# for i in result6:
# print(i)
self.assertEqual(len(result6), 6)
self.assertIn(self.test_batch1, result6)
self.assertIn(self.test_batch5, result6)
self.assertIn(self.test_batch3[-2:], result6)
self.assertIn(self.test_batch6, result6)
self.assertIn(self.test_batch8, result6)
self.assertIn(self.test_batch7[-2:], result6)
self.outfiles.extend(outfiles)
self.con6.send("stop_poison_pill")
def test_new_with_puller(self):
self.abalone_engine.start()
sleep(3)
self.abalone_con.send((self.abalone_temp, 'new'))
sleep(5)
outfiles = glob.glob('engine_test_output/engine_test_abalone*')
result = []
for o in outfiles:
result.extend(read_outfile(o))
self.assertEqual(len(result), 2)
self.outfiles.extend(outfiles)
self.abalone_con.send("stop_poison_pill")
