def on_start(self):
super(VizTransformProcForMatplotlibGraphs, self).on_start()
#assert len(self.streams)==1
self.initDataFlag = True
self.graph_data = {
} # Stores a dictionary of variables : [List of values]
# Need some clients
self.rr_cli = ResourceRegistryServiceProcessClient(
process=self, node=self.container.node)
self.pubsub_cli = PubsubManagementServiceClient(
node=self.container.node)
# extract the various parameters passed to the transform process
self.out_stream_id = self.CFG.get('process').get(
'publish_streams').get('visualization_service_submit_stream_id')
# Create a publisher on the output stream
#stream_route = self.pubsub_cli.register_producer(stream_id=self.out_stream_id)
out_stream_pub_registrar = StreamPublisherRegistrar(
process=self.container, node=self.container.node)
self.out_stream_pub = out_stream_pub_registrar.create_publisher(
stream_id=self.out_stream_id)
self.data_product_id = self.CFG.get('data_product_id')
self.stream_def_id = self.CFG.get("stream_def_id")
self.stream_def = self.rr_cli.read(self.stream_def_id)
# Start the thread responsible for keeping track of time and generating graphs
# Mutex for ensuring proper concurrent communications between threads
self.lock = RLock()
self.rendering_proc = Greenlet(self.rendering_thread)
self.rendering_proc.start()
def on_start(self):
super(VizTransformProcForMatplotlibGraphs,self).on_start()
#assert len(self.streams)==1
self.initDataFlag = True
self.graph_data = {} # Stores a dictionary of variables : [List of values]
# Need some clients
self.rr_cli = ResourceRegistryServiceProcessClient(process = self, node = self.container.node)
self.pubsub_cli = PubsubManagementServiceClient(node=self.container.node)
# extract the various parameters passed to the transform process
self.out_stream_id = self.CFG.get('process').get('publish_streams').get('visualization_service_submit_stream_id')
# Create a publisher on the output stream
#stream_route = self.pubsub_cli.register_producer(stream_id=self.out_stream_id)
out_stream_pub_registrar = StreamPublisherRegistrar(process=self.container, node=self.container.node)
self.out_stream_pub = out_stream_pub_registrar.create_publisher(stream_id=self.out_stream_id)
self.data_product_id = self.CFG.get('data_product_id')
self.stream_def_id = self.CFG.get("stream_def_id")
self.stream_def = self.rr_cli.read(self.stream_def_id)
# Start the thread responsible for keeping track of time and generating graphs
# Mutex for ensuring proper concurrent communications between threads
self.lock = RLock()
self.rendering_proc = Greenlet(self.rendering_thread)
self.rendering_proc.start()
def on_start(self):
super(VizTransformProcForGoogleDT,self).on_start()
self.initDataTableFlag = True
# need some clients
self.rr_cli = ResourceRegistryServiceProcessClient(process = self, node = self.container.node)
self.pubsub_cli = PubsubManagementServiceClient(node=self.container.node)
# extract the various parameters passed
self.out_stream_id = self.CFG.get('process').get('publish_streams').get('visualization_service_submit_stream_id')
# Create a publisher on the output stream
out_stream_pub_registrar = StreamPublisherRegistrar(process=self.container, node=self.container.node)
self.out_stream_pub = out_stream_pub_registrar.create_publisher(stream_id=self.out_stream_id)
self.data_product_id = self.CFG.get('data_product_id')
self.stream_def_id = self.CFG.get("stream_def_id")
stream_def_resource = self.rr_cli.read(self.stream_def_id)
self.stream_def = stream_def_resource.container
self.realtime_flag = False
if self.CFG.get("realtime_flag") == "True":
self.realtime_flag = True
else:
self.data_product_id_token = self.CFG.get('data_product_id_token')
# extract the stream_id associated with the DP. Needed later
stream_ids,_ = self.rr_cli.find_objects(self.data_product_id, PRED.hasStream, None, True)
self.stream_id = stream_ids[0]
self.dataDescription = []
self.dataTableContent = []
self.varTuple = []
self.total_num_of_records_recvd = 0
def setUp(self):
self._start_container()
self.cc = ContainerAgentClient(node=self.container.node,name=self.container.name)
self.cc.start_rel_from_url('res/deploy/r2dm.yml')
self.pubsub_cli = PubsubManagementServiceClient(node=self.cc.node)
self.ctd_stream1_id = self.pubsub_cli.create_stream(name="SampleStream1",
description="Sample Stream 1 Description")
self.ctd_stream2_id = self.pubsub_cli.create_stream(name="SampleStream2",
description="Sample Stream 2 Description")
# Make a subscription to two input streams
exchange_name = "a_queue"
query = StreamQuery([self.ctd_stream1_id, self.ctd_stream2_id])
self.ctd_subscription_id = self.pubsub_cli.create_subscription(query,
exchange_name,
"SampleSubscription",
"Sample Subscription Description")
# Make a subscription to all streams on an exchange point
exchange_name = "another_queue"
query = ExchangeQuery()
self.exchange_subscription_id = self.pubsub_cli.create_subscription(query,
exchange_name,
"SampleExchangeSubscription",
"Sample Exchange Subscription Description")
pid = self.container.spawn_process(name='dummy_process_for_test',
module='pyon.ion.process',
cls='SimpleProcess',
config={})
dummy_process = self.container.proc_manager.procs[pid]
# Normally the user does not see or create the publisher, this is part of the containers business.
# For the test we need to set it up explicitly
publisher_registrar = StreamPublisherRegistrar(process=dummy_process, node=self.cc.node)
self.ctd_stream1_publisher = publisher_registrar.create_publisher(stream_id=self.ctd_stream1_id)
self.ctd_stream2_publisher = publisher_registrar.create_publisher(stream_id=self.ctd_stream2_id)
# Cheat and use the cc as the process - I don't think it is used for anything...
self.stream_subscriber = StreamSubscriberRegistrar(process=dummy_process, node=self.cc.node)
def setUp(self):
self._start_container()
self.container.start_rel_from_url('res/deploy/r2dm.yml')
self.pubsub_cli = PubsubManagementServiceClient(node=self.container.node)
self.ctd_stream1_id = self.pubsub_cli.create_stream(name="SampleStream1",
description="Sample Stream 1 Description")
self.ctd_stream2_id = self.pubsub_cli.create_stream(name="SampleStream2",
description="Sample Stream 2 Description")
# Make a subscription to two input streams
exchange_name = "a_queue"
query = StreamQuery([self.ctd_stream1_id, self.ctd_stream2_id])
self.ctd_subscription_id = self.pubsub_cli.create_subscription(query,
exchange_name,
"SampleSubscription",
"Sample Subscription Description")
# Make a subscription to all streams on an exchange point
exchange_name = "another_queue"
query = ExchangeQuery()
self.exchange_subscription_id = self.pubsub_cli.create_subscription(query,
exchange_name,
"SampleExchangeSubscription",
"Sample Exchange Subscription Description")
pid = self.container.spawn_process(name='dummy_process_for_test',
module='pyon.ion.process',
cls='SimpleProcess',
config={})
dummy_process = self.container.proc_manager.procs[pid]
# Normally the user does not see or create the publisher, this is part of the containers business.
# For the test we need to set it up explicitly
publisher_registrar = StreamPublisherRegistrar(process=dummy_process, node=self.container.node)
self.ctd_stream1_publisher = publisher_registrar.create_publisher(stream_id=self.ctd_stream1_id)
self.ctd_stream2_publisher = publisher_registrar.create_publisher(stream_id=self.ctd_stream2_id)
# Cheat and use the cc as the process - I don't think it is used for anything...
self.stream_subscriber = StreamSubscriberRegistrar(process=dummy_process, node=self.container.node)
def on_start(self):
super(VizTransformProcForGoogleDT, self).on_start()
self.initDataTableFlag = True
# need some clients
self.rr_cli = ResourceRegistryServiceProcessClient(
process=self, node=self.container.node)
self.pubsub_cli = PubsubManagementServiceClient(
node=self.container.node)
# extract the various parameters passed
self.out_stream_id = self.CFG.get('process').get(
'publish_streams').get('visualization_service_submit_stream_id')
# Create a publisher on the output stream
out_stream_pub_registrar = StreamPublisherRegistrar(
process=self.container, node=self.container.node)
self.out_stream_pub = out_stream_pub_registrar.create_publisher(
stream_id=self.out_stream_id)
self.data_product_id = self.CFG.get('data_product_id')
self.stream_def_id = self.CFG.get("stream_def_id")
stream_def_resource = self.rr_cli.read(self.stream_def_id)
self.stream_def = stream_def_resource.container
self.realtime_flag = False
if self.CFG.get("realtime_flag") == "True":
self.realtime_flag = True
else:
self.data_product_id_token = self.CFG.get('data_product_id_token')
# extract the stream_id associated with the DP. Needed later
stream_ids, _ = self.rr_cli.find_objects(self.data_product_id,
PRED.hasStream, None, True)
self.stream_id = stream_ids[0]
self.dataDescription = []
self.dataTableContent = []
self.varTuple = []
self.total_num_of_records_recvd = 0
def test_createDataProcess(self):
#-------------------------------
# Data Process Definition
#-------------------------------
log.debug("TestIntDataProcessManagementService: create data process definition")
dpd_obj = IonObject(RT.DataProcessDefinition,
name='data_process_definition',
description='some new dpd',
module='ion.processes.data.transforms.transform_example',
class_name='TransformExample',
process_source='some_source_reference')
try:
dprocdef_id = self.Processclient.create_data_process_definition(dpd_obj)
except BadRequest as ex:
self.fail("failed to create new data process definition: %s" %ex)
# test Data Process Definition creation in rr
dprocdef_obj = self.Processclient.read_data_process_definition(dprocdef_id)
self.assertEquals(dprocdef_obj.name,'data_process_definition')
# Create an input instrument
instrument_obj = IonObject(RT.InstrumentDevice, name='Inst1',description='an instrument that is creating the data product')
instrument_id, rev = self.RRclient.create(instrument_obj)
# Register the instrument so that the data producer and stream object are created
data_producer_id = self.DAMSclient.register_instrument(instrument_id)
log.debug("TestIntDataProcessManagementService data_producer_id %s" % data_producer_id)
#-------------------------------
# Input Data Product
#-------------------------------
log.debug("TestIntDataProcessManagementService: create input data product")
input_dp_obj = IonObject(RT.DataProduct, name='InputDataProduct', description='some new dp')
try:
input_dp_id = self.DPMSclient.create_data_product(input_dp_obj, instrument_id)
except BadRequest as ex:
self.fail("failed to create new input data product: %s" %ex)
# Retrieve the stream via the DataProduct->Stream associations
stream_ids, _ = self.RRclient.find_objects(input_dp_id, PRED.hasStream, None, True)
log.debug("TestIntDataProcessManagementService: in stream_ids " + str(stream_ids))
self.in_stream_id = stream_ids[0]
log.debug("TestIntDataProcessManagementService: Input Stream: " + str( self.in_stream_id))
#-------------------------------
# Output Data Product
#-------------------------------
log.debug("TestIntDataProcessManagementService: create output data product")
output_dp_obj = IonObject(RT.DataProduct, name='OutDataProduct',description='transform output')
output_dp_id = self.DPMSclient.create_data_product(output_dp_obj)
# this will NOT create a stream for the product becuase the data process (source) resource has not been created yet.
#-------------------------------
# Create the data process
#-------------------------------
log.debug("TestIntDataProcessManagementService: create_data_process start")
try:
dproc_id = self.Processclient.create_data_process(dprocdef_id, input_dp_id, output_dp_id)
except BadRequest as ex:
self.fail("failed to create new data process: %s" %ex)
self.DAMSclient.assign_data_product(dproc_id, output_dp_id, False)
log.debug("TestIntDataProcessManagementService: create_data_process return")
#-------------------------------
# Producer (Sample Input)
#-------------------------------
# Create a producing example process
# cheat to make a publisher object to send messages in the test.
# it is really hokey to pass process=self.cc but it works
#stream_route = self.PubSubClient.register_producer(exchange_name='producer_doesnt_have_a_name1', stream_id=self.in_stream_id)
#self.ctd_stream1_publisher = StreamPublisher(node=self.container.node, name=('science_data',stream_route.routing_key), process=self.container)
pid = self.container.spawn_process(name='dummy_process_for_test',
module='pyon.ion.process',
cls='SimpleProcess',
config={})
dummy_process = self.container.proc_manager.procs[pid]
publisher_registrar = StreamPublisherRegistrar(process=dummy_process, node=self.container.node)
self.ctd_stream1_publisher = publisher_registrar.create_publisher(stream_id=self.in_stream_id)
msg = {'num':'3'}
self.ctd_stream1_publisher.publish(msg)
time.sleep(1)
msg = {'num':'5'}
self.ctd_stream1_publisher.publish(msg)
time.sleep(1)
msg = {'num':'9'}
self.ctd_stream1_publisher.publish(msg)
# See /tmp/transform_output for results.....
# clean up the data process
try:
self.Processclient.delete_data_process(dproc_id)
except BadRequest as ex:
self.fail("failed to create new data process definition: %s" %ex)
with self.assertRaises(NotFound) as e:
self.Processclient.read_data_process(dproc_id)
try:
self.Processclient.delete_data_process_definition(dprocdef_id)
except BadRequest as ex:
self.fail("failed to create new data process definition: %s" %ex)
with self.assertRaises(NotFound) as e:
self.Processclient.read_data_process_definition(dprocdef_id)
def test_integrated_transform(self):
'''
This example script runs a chained three way transform:
B
A <
C
Where A is the even_odd transform (generates a stream of even and odd numbers from input)
and B and C are the basic transforms that receive even and odd input
'''
cc = self.container
assertions = self.assertTrue
pubsub_cli = PubsubManagementServiceClient(node=cc.node)
rr_cli = ResourceRegistryServiceClient(node=cc.node)
tms_cli = TransformManagementServiceClient(node=cc.node)
#-------------------------------
# Process Definition
#-------------------------------
# Create the process definition for the basic transform
process_definition = IonObject(RT.ProcessDefinition, name='basic_transform_definition')
process_definition.executable = {
'module': 'ion.processes.data.transforms.transform_example',
'class':'TransformExample'
}
basic_transform_definition_id, _ = rr_cli.create(process_definition)
# Create The process definition for the TransformEvenOdd
process_definition = IonObject(RT.ProcessDefinition, name='evenodd_transform_definition')
process_definition.executable = {
'module': 'ion.processes.data.transforms.transform_example',
'class':'TransformEvenOdd'
}
evenodd_transform_definition_id, _ = rr_cli.create(process_definition)
#-------------------------------
# Streams
#-------------------------------
streams = [pubsub_cli.create_stream() for i in xrange(5)]
#-------------------------------
# Subscriptions
#-------------------------------
query = StreamQuery(stream_ids=[streams[0]])
input_subscription_id = pubsub_cli.create_subscription(query=query, exchange_name='input_queue')
query = StreamQuery(stream_ids = [streams[1]]) # even output
even_subscription_id = pubsub_cli.create_subscription(query=query, exchange_name='even_queue')
query = StreamQuery(stream_ids = [streams[2]]) # odd output
odd_subscription_id = pubsub_cli.create_subscription(query=query, exchange_name='odd_queue')
#-------------------------------
# Launch the EvenOdd Transform
#-------------------------------
evenodd_id = tms_cli.create_transform(name='even_odd',
in_subscription_id=input_subscription_id,
out_streams={'even':streams[1], 'odd':streams[2]},
process_definition_id=evenodd_transform_definition_id,
configuration={})
tms_cli.activate_transform(evenodd_id)
#-------------------------------
# Launch the Even Processing Transform
#-------------------------------
even_transform_id = tms_cli.create_transform(name='even_transform',
in_subscription_id = even_subscription_id,
out_streams={'even_plus1':streams[3]},
process_definition_id=basic_transform_definition_id,
configuration={})
tms_cli.activate_transform(even_transform_id)
#-------------------------------
# Launch the Odd Processing Transform
#-------------------------------
odd_transform_id = tms_cli.create_transform(name='odd_transform',
in_subscription_id = odd_subscription_id,
out_streams={'odd_plus1':streams[4]},
process_definition_id=basic_transform_definition_id,
configuration={})
tms_cli.activate_transform(odd_transform_id)
#-------------------------------
# Set up final subscribers
#-------------------------------
evenplus1_subscription_id = pubsub_cli.create_subscription(
query=StreamQuery([streams[3]]),
exchange_name='evenplus1_queue',
name='EvenPlus1Subscription',
description='EvenPlus1 SubscriptionDescription'
)
oddplus1_subscription_id = pubsub_cli.create_subscription(
query=StreamQuery([streams[4]]),
exchange_name='oddplus1_queue',
name='OddPlus1Subscription',
description='OddPlus1 SubscriptionDescription'
)
total_msg_count = 2
msgs = gevent.queue.Queue()
def even1_message_received(message, headers):
input = int(message.get('num'))
assertions( (input % 2) ) # Assert it is odd (transform adds 1)
msgs.put(True)
def odd1_message_received(message, headers):
input = int(message.get('num'))
assertions(not (input % 2)) # Assert it is even
msgs.put(True)
subscriber_registrar = StreamSubscriberRegistrar(process=cc, container=cc)
even_subscriber = subscriber_registrar.create_subscriber(exchange_name='evenplus1_queue', callback=even1_message_received)
odd_subscriber = subscriber_registrar.create_subscriber(exchange_name='oddplus1_queue', callback=odd1_message_received)
# Start subscribers
even_subscriber.start()
odd_subscriber.start()
# Activate subscriptions
pubsub_cli.activate_subscription(evenplus1_subscription_id)
pubsub_cli.activate_subscription(oddplus1_subscription_id)
#-------------------------------
# Set up fake stream producer
#-------------------------------
pid = cc.spawn_process(name='dummy_process_for_test',
module='pyon.ion.process',
cls='SimpleProcess',
config={})
dummy_process = cc.proc_manager.procs[pid]
# Normally the user does not see or create the publisher, this is part of the containers business.
# For the test we need to set it up explicitly
publisher_registrar = StreamPublisherRegistrar(process=dummy_process, container=cc)
stream_publisher = publisher_registrar.create_publisher(stream_id=streams[0])
#-------------------------------
# Start test
#-------------------------------
# Publish a stream
for i in xrange(total_msg_count):
stream_publisher.publish({'num':str(i)})
time.sleep(0.5)
for i in xrange(total_msg_count * 2):
try:
msgs.get()
except Empty:
assertions(False, "Failed to process all messages correctly.")
def test_integrated_transform(self):
'''
This example script runs a chained three way transform:
B
A <
C
Where A is the even_odd transform (generates a stream of even and odd numbers from input)
and B and C are the basic transforms that receive even and odd input
'''
cc = self.container
assertions = self.assertTrue
pubsub_cli = PubsubManagementServiceClient(node=cc.node)
rr_cli = ResourceRegistryServiceClient(node=cc.node)
tms_cli = TransformManagementServiceClient(node=cc.node)
#-------------------------------
# Process Definition
#-------------------------------
# Create the process definition for the basic transform
process_definition = IonObject(RT.ProcessDefinition, name='basic_transform_definition')
process_definition.executable = {
'module': 'ion.processes.data.transforms.transform_example',
'class':'TransformExample'
}
basic_transform_definition_id, _ = rr_cli.create(process_definition)
# Create The process definition for the TransformEvenOdd
process_definition = IonObject(RT.ProcessDefinition, name='evenodd_transform_definition')
process_definition.executable = {
'module': 'ion.processes.data.transforms.transform_example',
'class':'TransformEvenOdd'
}
evenodd_transform_definition_id, _ = rr_cli.create(process_definition)
#-------------------------------
# Streams
#-------------------------------
streams = [pubsub_cli.create_stream() for i in xrange(5)]
#-------------------------------
# Subscriptions
#-------------------------------
query = StreamQuery(stream_ids=[streams[0]])
input_subscription_id = pubsub_cli.create_subscription(query=query, exchange_name='input_queue')
query = StreamQuery(stream_ids = [streams[1]]) # even output
even_subscription_id = pubsub_cli.create_subscription(query=query, exchange_name='even_queue')
query = StreamQuery(stream_ids = [streams[2]]) # odd output
odd_subscription_id = pubsub_cli.create_subscription(query=query, exchange_name='odd_queue')
#-------------------------------
# Launch the EvenOdd Transform
#-------------------------------
evenodd_id = tms_cli.create_transform(name='even_odd',
in_subscription_id=input_subscription_id,
out_streams={'even':streams[1], 'odd':streams[2]},
process_definition_id=evenodd_transform_definition_id,
configuration={})
tms_cli.activate_transform(evenodd_id)
#-------------------------------
# Launch the Even Processing Transform
#-------------------------------
even_transform_id = tms_cli.create_transform(name='even_transform',
in_subscription_id = even_subscription_id,
out_streams={'even_plus1':streams[3]},
process_definition_id=basic_transform_definition_id,
configuration={})
tms_cli.activate_transform(even_transform_id)
#-------------------------------
# Launch the Odd Processing Transform
#-------------------------------
odd_transform_id = tms_cli.create_transform(name='odd_transform',
in_subscription_id = odd_subscription_id,
out_streams={'odd_plus1':streams[4]},
process_definition_id=basic_transform_definition_id,
configuration={})
tms_cli.activate_transform(odd_transform_id)
#-------------------------------
# Set up final subscribers
#-------------------------------
evenplus1_subscription_id = pubsub_cli.create_subscription(
query=StreamQuery([streams[3]]),
exchange_name='evenplus1_queue',
name='EvenPlus1Subscription',
description='EvenPlus1 SubscriptionDescription'
)
oddplus1_subscription_id = pubsub_cli.create_subscription(
query=StreamQuery([streams[4]]),
exchange_name='oddplus1_queue',
name='OddPlus1Subscription',
description='OddPlus1 SubscriptionDescription'
)
total_msg_count = 2
msgs = gevent.queue.Queue()
def even1_message_received(message, headers):
input = int(message.get('num'))
assertions( (input % 2) ) # Assert it is odd (transform adds 1)
msgs.put(True)
def odd1_message_received(message, headers):
input = int(message.get('num'))
assertions(not (input % 2)) # Assert it is even
msgs.put(True)
subscriber_registrar = StreamSubscriberRegistrar(process=cc, node=cc.node)
even_subscriber = subscriber_registrar.create_subscriber(exchange_name='evenplus1_queue', callback=even1_message_received)
odd_subscriber = subscriber_registrar.create_subscriber(exchange_name='oddplus1_queue', callback=odd1_message_received)
# Start subscribers
even_subscriber.start()
odd_subscriber.start()
# Activate subscriptions
pubsub_cli.activate_subscription(evenplus1_subscription_id)
pubsub_cli.activate_subscription(oddplus1_subscription_id)
#-------------------------------
# Set up fake stream producer
#-------------------------------
pid = cc.spawn_process(name='dummy_process_for_test',
module='pyon.ion.process',
cls='SimpleProcess',
config={})
dummy_process = cc.proc_manager.procs[pid]
# Normally the user does not see or create the publisher, this is part of the containers business.
# For the test we need to set it up explicitly
publisher_registrar = StreamPublisherRegistrar(process=dummy_process, node=cc.node)
stream_publisher = publisher_registrar.create_publisher(stream_id=streams[0])
#-------------------------------
# Start test
#-------------------------------
# Publish a stream
for i in xrange(total_msg_count):
stream_publisher.publish({'num':str(i)})
time.sleep(0.5)
for i in xrange(total_msg_count * 2):
try:
msgs.get()
except Empty:
assertions(False, "Failed to process all messages correctly.")
def test_replay_integration(self):
'''
Test full DM Services Integration
'''
cc = self.container
### Every thing below here can be run as a script:
pubsub_management_service = PubsubManagementServiceClient(node=cc.node)
ingestion_management_service = IngestionManagementServiceClient(node=cc.node)
dataset_management_service = DatasetManagementServiceClient(node=cc.node)
data_retriever_service = DataRetrieverServiceClient(node=cc.node)
resource_registry_service = ResourceRegistryServiceClient(node=cc.node)
#------------------------------------------------------------------------------------------------------
# Datastore name
#------------------------------------------------------------------------------------------------------
datastore_name = 'test_replay_integration'
#------------------------------------------------------------------------------------------------------
# Spawn process
#------------------------------------------------------------------------------------------------------
pid = cc.spawn_process(name='dummy_process_for_test',
module='pyon.ion.process',
cls='SimpleProcess',
config={})
dummy_process = cc.proc_manager.procs[pid]
#------------------------------------------------------------------------------------------------------
# Set up subscriber
#------------------------------------------------------------------------------------------------------
# Normally the user does not see or create the publisher, this is part of the containers business.
# For the test we need to set it up explicitly
publisher_registrar = StreamPublisherRegistrar(process=dummy_process, node=cc.node)
subscriber_registrar = StreamSubscriberRegistrar(process=cc, node=cc.node)
#------------------------------------------------------------------------------------------------------
# Set up ingestion
#------------------------------------------------------------------------------------------------------
# Configure ingestion using eight workers, ingesting to test_dm_integration datastore with the SCIDATA profile
ingestion_configuration_id = ingestion_management_service.create_ingestion_configuration(
exchange_point_id='science_data',
couch_storage=CouchStorage(datastore_name=datastore_name, datastore_profile='SCIDATA'),
hdf_storage=HdfStorage(),
number_of_workers=1,
)
ingestion_management_service.activate_ingestion_configuration(
ingestion_configuration_id=ingestion_configuration_id)
#------------------------------------------------------------------------------------------------------
# Grab the transforms acting as ingestion workers
#------------------------------------------------------------------------------------------------------
transforms = [resource_registry_service.read(assoc.o)
for assoc in resource_registry_service.find_associations(ingestion_configuration_id, PRED.hasTransform)]
proc_1 = cc.proc_manager.procs[transforms[0].process_id]
log.info("PROCESS 1: %s" % str(proc_1))
#------------------------------------------------------------------------------------------------------
# Set up the test hooks for the gevent event AsyncResult object
#------------------------------------------------------------------------------------------------------
def ingestion_worker_received(message, headers):
ar.set(message)
proc_1.ingest_process_test_hook = ingestion_worker_received
#------------------------------------------------------------------------------------------------------
# Set up the producers (CTD Simulators)
#------------------------------------------------------------------------------------------------------
ctd_stream_def = ctd_stream_definition()
stream_def_id = pubsub_management_service.create_stream_definition(container=ctd_stream_def, name='Junk definition')
stream_id = pubsub_management_service.create_stream(stream_definition_id=stream_def_id)
#------------------------------------------------------------------------------------------------------
# Set up the dataset config
#------------------------------------------------------------------------------------------------------
dataset_id = dataset_management_service.create_dataset(
stream_id=stream_id,
datastore_name=datastore_name,
view_name='datasets/stream_join_granule'
)
dataset_config_id = ingestion_management_service.create_dataset_configuration(
dataset_id = dataset_id,
archive_data = True,
archive_metadata = True,
ingestion_configuration_id = ingestion_configuration_id
)
#------------------------------------------------------------------------------------------------------
# Launch a ctd_publisher
#------------------------------------------------------------------------------------------------------
publisher = publisher_registrar.create_publisher(stream_id=stream_id)
#------------------------------------------------------------------------
# Create a packet and publish it
#------------------------------------------------------------------------
ctd_packet = _create_packet(stream_id)
published_hdfstring = ctd_packet.identifiables['ctd_data'].values
publisher.publish(ctd_packet)
#------------------------------------------------------------------------------------------------------
# Catch what the ingestion worker gets! Assert it is the same packet that was published!
#------------------------------------------------------------------------------------------------------
packet = ar.get(timeout=2)
#------------------------------------------------------------------------------------------------------
# Create subscriber to listen to the replays
#------------------------------------------------------------------------------------------------------
replay_id, replay_stream_id = data_retriever_service.define_replay(dataset_id)
query = StreamQuery(stream_ids=[replay_stream_id])
subscription_id = pubsub_management_service.create_subscription(query = query, exchange_name='replay_capture_point' ,name = 'replay_capture_point')
# It is not required or even generally a good idea to use the subscription resource name as the queue name, but it makes things simple here
# Normally the container creates and starts subscribers for you when a transform process is spawned
subscriber = subscriber_registrar.create_subscriber(exchange_name='replay_capture_point', callback=_subscriber_call_back)
subscriber.start()
pubsub_management_service.activate_subscription(subscription_id)
#------------------------------------------------------------------------------------------------------
# Start the replay
#------------------------------------------------------------------------------------------------------
data_retriever_service.start_replay(replay_id)
#------------------------------------------------------------------------------------------------------
# Get the hdf string from the captured stream in the replay
#------------------------------------------------------------------------------------------------------
retrieved_hdf_string = ar2.get(timeout=2)
### Non scriptable portion of the test
#------------------------------------------------------------------------------------------------------
# Assert that it matches the message we sent
#------------------------------------------------------------------------------------------------------
self.assertEquals(packet.identifiables['stream_encoding'].sha1, ctd_packet.identifiables['stream_encoding'].sha1)
self.assertEquals(retrieved_hdf_string, published_hdfstring)