Skip to content

Restore IoTHub metadata in EventHub #42

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Feb 10, 2020
Merged
Show file tree
Hide file tree
Changes from 5 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .flake8
Original file line number Diff line number Diff line change
Expand Up @@ -2,4 +2,4 @@
ignore = W503,E402,E731
exclude =
.git, __pycache__, build, dist, .eggs, .github, .local,
Samples, azure/functions/_thirdparty, docs/
Samples, azure/functions/_thirdparty, docs/, .venv*/, .env*/, .vscode/
85 changes: 71 additions & 14 deletions azure/functions/eventhub.py
Original file line number Diff line number Diff line change
Expand Up @@ -105,12 +105,10 @@ def decode(
typing.List[_eventhub.EventHubEvent]]:
data_type = data.type

if (data_type == 'string' or data_type == 'bytes'
or data_type == 'json'):
if cls._is_cardinary_one(trigger_metadata):
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shouldn't this be "cardinality"? Or is "cardinary" a term I'm not aware of?

return cls.decode_single_event(data, trigger_metadata)

elif (data_type == 'collection_bytes'
or data_type == 'collection_string'):
elif cls._is_cardinary_many(trigger_metadata):
return cls.decode_multiple_events(data, trigger_metadata)

else:
Expand All @@ -129,13 +127,6 @@ def decode_single_event(cls, data,
elif data.type == 'json':
body = data.value.encode('utf-8')

iothub_metadata = {}
for f in trigger_metadata:
if f.startswith('iothub-'):
v = cls._decode_trigger_metadata_field(
trigger_metadata, f, python_type=str)
iothub_metadata[f[len('iothub-'):]] = v

return _eventhub.EventHubEvent(
body=body,
enqueued_time=cls._parse_datetime_metadata(
Expand All @@ -146,7 +137,7 @@ def decode_single_event(cls, data,
trigger_metadata, 'SequenceNumber', python_type=int),
offset=cls._decode_trigger_metadata_field(
trigger_metadata, 'Offset', python_type=str),
iothub_metadata=iothub_metadata
iothub_metadata=cls._decode_iothub_metadata(trigger_metadata)
)

@classmethod
Expand All @@ -159,6 +150,10 @@ def decode_multiple_events(
elif data.type == 'collection_string':
parsed_data = data.value.string

# Input Trigger IotHub Event
elif data.type == 'json':
parsed_data = json.loads(data.value)

sys_props = trigger_metadata.get('SystemPropertiesArray')

parsed_sys_props = json.loads(sys_props.value)
Expand All @@ -180,17 +175,79 @@ def decode_multiple_events(
expected_type=int)

event = _eventhub.EventHubEvent(
body=parsed_data[i],
body=cls._marshall_event_body(parsed_data[i], data.type),
enqueued_time=cls._parse_datetime(enqueued_time),
partition_key=cls._decode_typed_data(
partition_key, python_type=str),
sequence_number=cls._decode_typed_data(
sequence_number, python_type=int),
offset=cls._decode_typed_data(
offset, python_type=int),
iothub_metadata={}
iothub_metadata=cls._extract_iothub_from_dict(
parsed_sys_props[i])
)

events.append(event)

return events

@classmethod
def _marshall_event_body(self, parsed_data, data_type):
# In IoTHub, when setting the eventhub using cardinary = 'many'
# The data is wrapped inside a json (e.g. '[{ "device-id": "1" }]')

# Previously, since the IoTHub events has a 'json' datatype,
# it is handled as single_event by mistake and our users handle the
# data parsing. And we want to keep the same behavior here.
if data_type == 'json':
return json.dumps(parsed_data).encode('utf-8')

return parsed_data

@classmethod
def _decode_iothub_metadata(
cls, trigger_metadata) -> typing.Dict[str, str]:
# Try extracting iothub_metadata from trigger_metadata
iothub_metadata = cls._extract_iothub_from_trigger_metadata(
trigger_metadata)

# Try extracting iothub_metadata from SystemProperties
if not iothub_metadata and trigger_metadata.get('SystemProperties'):
iothub_metadata = cls._extract_iothub_from_system_properties(
trigger_metadata['SystemProperties'].value)

return iothub_metadata

@classmethod
def _extract_iothub_from_trigger_metadata(
cls, metadict: typing.Dict[str, str]) -> typing.Dict[str, str]:
iothub_metadata = {}
for f in metadict:
if f.startswith('iothub-'):
v = cls._decode_trigger_metadata_field(
metadict, f, python_type=str)
iothub_metadata[f[len('iothub-'):]] = v
return iothub_metadata

@classmethod
def _extract_iothub_from_system_properties(
cls, system_properties_string: str) -> typing.Dict[str, str]:
system_properties = json.loads(system_properties_string)
return cls._extract_iothub_from_dict(system_properties)

@classmethod
def _extract_iothub_from_dict(
cls, metadict: typing.Dict[str, str]) -> typing.Dict[str, str]:
iothub_metadata = {}
for f in metadict:
if f.startswith('iothub-'):
iothub_metadata[f[len('iothub-'):]] = metadict[f]
return iothub_metadata

@classmethod
def _is_cardinary_many(cls, trigger_metadata) -> bool:
return 'SystemPropertiesArray' in trigger_metadata

@classmethod
def _is_cardinary_one(cls, trigger_metadata) -> bool:
return 'SystemProperties' in trigger_metadata
200 changes: 200 additions & 0 deletions tests/test_eventhub.py
Original file line number Diff line number Diff line change
@@ -1,11 +1,27 @@
from typing import List
import unittest
import json
from unittest.mock import patch
from datetime import datetime

import azure.functions as func
import azure.functions.eventhub as azf_eh
import azure.functions.meta as meta


class CollectionBytes:
def __init__(self, data: List[bytes]):
self.bytes = data


class CollectionString:
def __init__(self, data: List[str]):
self.string = list(map(lambda x: x.encode('utf-8'), data))


class TestEventHub(unittest.TestCase):
MOCKED_ENQUEUE_TIME = datetime.utcnow()

def test_eventhub_input_type(self):
check_input_type = (
azf_eh.EventHubConverter.check_input_type_annotation
Expand All @@ -26,3 +42,187 @@ def test_eventhub_output_type(self):
self.assertFalse(check_output_type(func.EventHubEvent))
self.assertFalse(check_output_type(List[bytes]))
self.assertFalse(check_output_type(List[func.EventHubEvent]))

@patch('azure.functions.eventhub.EventHubTriggerConverter'
'.decode_single_event')
@patch('azure.functions.eventhub.EventHubTriggerConverter'
'.decode_multiple_events')
def test_eventhub_decode_single_event(self, dme_mock, dse_mock):
azf_eh.EventHubTriggerConverter.decode(
data=self._generate_single_iothub_datum(),
trigger_metadata=self._generate_single_trigger_metadatum()
)
dse_mock.assert_called_once()
dme_mock.assert_not_called()

@patch('azure.functions.eventhub.EventHubTriggerConverter'
'.decode_single_event')
@patch('azure.functions.eventhub.EventHubTriggerConverter'
'.decode_multiple_events')
def test_eventhub_decode_multiple_events(self, dme_mock, dse_mock):
azf_eh.EventHubTriggerConverter.decode(
data=self._generate_multiple_iothub_data(),
trigger_metadata=self._generate_multiple_trigger_metadata()
)
dse_mock.assert_not_called()
dme_mock.assert_called_once()

def test_eventhub_trigger_single_event_json(self):
result = azf_eh.EventHubTriggerConverter.decode(
data=self._generate_single_iothub_datum('json'),
trigger_metadata=self._generate_single_trigger_metadatum()
)
# Result body always has the datatype of bytes
self.assertEqual(
result.get_body().decode('utf-8'), '{"device-status": "good"}'
)
self.assertEqual(result.enqueued_time, self.MOCKED_ENQUEUE_TIME)

def test_eventhub_trigger_single_event_bytes(self):
result = azf_eh.EventHubTriggerConverter.decode(
data=self._generate_single_iothub_datum('bytes'),
trigger_metadata=self._generate_single_trigger_metadatum()
)
self.assertEqual(
result.get_body().decode('utf-8'), '{"device-status": "good"}'
)
self.assertEqual(result.enqueued_time, self.MOCKED_ENQUEUE_TIME)

def test_iothub_metadata_single_event(self):
result = azf_eh.EventHubTriggerConverter.decode(
data=self._generate_single_iothub_datum('json'),
trigger_metadata=self._generate_single_trigger_metadatum()
)
self.assertIsNotNone(result.iothub_metadata)
self.assertEqual(
result.iothub_metadata['connection-device-id'], 'MyTestDevice'
)

def test_eventhub_trigger_single_event_string(self):
result = azf_eh.EventHubTriggerConverter.decode(
data=self._generate_single_iothub_datum('string'),
trigger_metadata=self._generate_single_trigger_metadatum()
)
self.assertEqual(
result.get_body().decode('utf-8'), '{"device-status": "good"}'
)
self.assertEqual(result.enqueued_time, self.MOCKED_ENQUEUE_TIME)

def test_eventhub_trigger_multiple_events_json(self):
result = azf_eh.EventHubTriggerConverter.decode(
data=self._generate_multiple_iothub_data('json'),
trigger_metadata=self._generate_multiple_trigger_metadata()
)
self.assertIsInstance(result, list)
self.assertEqual(len(result), 2)

self.assertEqual(result[0].enqueued_time, self.MOCKED_ENQUEUE_TIME)
self.assertEqual(
result[0].get_body().decode('utf-8'), '{"device-status": "good1"}'
)

self.assertEqual(result[1].enqueued_time, self.MOCKED_ENQUEUE_TIME)
self.assertEqual(
result[1].get_body().decode('utf-8'), '{"device-status": "good2"}'
)

def test_eventhub_trigger_multiple_events_collection_string(self):
result = azf_eh.EventHubTriggerConverter.decode(
data=self._generate_multiple_iothub_data('collection_string'),
trigger_metadata=self._generate_multiple_trigger_metadata()
)
self.assertIsInstance(result, list)
self.assertEqual(len(result), 2)

self.assertEqual(result[0].enqueued_time, self.MOCKED_ENQUEUE_TIME)
self.assertEqual(
result[0].get_body().decode('utf-8'), '{"device-status": "good1"}'
)

self.assertEqual(result[1].enqueued_time, self.MOCKED_ENQUEUE_TIME)
self.assertEqual(
result[1].get_body().decode('utf-8'), '{"device-status": "good2"}'
)

def test_eventhub_trigger_multiple_events_collection_bytes(self):
result = azf_eh.EventHubTriggerConverter.decode(
data=self._generate_multiple_iothub_data('collection_bytes'),
trigger_metadata=self._generate_multiple_trigger_metadata()
)
self.assertIsInstance(result, list)
self.assertEqual(len(result), 2)

self.assertEqual(result[0].enqueued_time, self.MOCKED_ENQUEUE_TIME)
self.assertEqual(
result[0].get_body().decode('utf-8'), '{"device-status": "good1"}'
)

self.assertEqual(result[1].enqueued_time, self.MOCKED_ENQUEUE_TIME)
self.assertEqual(
result[1].get_body().decode('utf-8'), '{"device-status": "good2"}'
)

def test_iothub_metadata_events(self):
result = azf_eh.EventHubTriggerConverter.decode(
data=self._generate_multiple_iothub_data('json'),
trigger_metadata=self._generate_multiple_trigger_metadata()
)
self.assertIsNotNone(result[0].iothub_metadata)
self.assertEqual(
result[0].iothub_metadata['connection-device-id'], 'MyTestDevice1'
)

self.assertIsNotNone(result[1].iothub_metadata)
self.assertEqual(
result[1].iothub_metadata['connection-device-id'], 'MyTestDevice2'
)

def _generate_single_iothub_datum(self, datum_type='json'):
datum = '{"device-status": "good"}'
if datum_type == 'bytes':
datum = datum.encode('utf-8')

return meta.Datum(datum, datum_type)

def _generate_multiple_iothub_data(self, data_type='json'):
data = '[{"device-status": "good1"}, {"device-status": "good2"}]'
if data_type == 'collection_bytes':
data = list(
map(lambda x: json.dumps(x).encode('utf-8'), json.loads(data))
)
data = CollectionBytes(data)
elif data_type == 'collection_string':
data = list(
map(lambda x: json.dumps(x), json.loads(data))
)
data = CollectionString(data)

return meta.Datum(data, data_type)

def _generate_single_trigger_metadatum(self):
return {
'EnqueuedTime': meta.Datum(
f'"{self.MOCKED_ENQUEUE_TIME.isoformat()}"', 'json'
),
'SystemProperties': meta.Datum(
'{"iothub-connection-device-id": "MyTestDevice"}', 'json'
)
}

def _generate_multiple_trigger_metadata(self):
system_props_array = [
{
'EnqueuedTimeUtc': self.MOCKED_ENQUEUE_TIME.isoformat(),
'iothub-connection-device-id': 'MyTestDevice1',
},
{
'EnqueuedTimeUtc': self.MOCKED_ENQUEUE_TIME.isoformat(),
'iothub-connection-device-id': 'MyTestDevice2',
}
]

return {
'SystemPropertiesArray': meta.Datum(
json.dumps(system_props_array), 'json'
)
}