Skip to content
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.

Sign up for GitHub

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

airbyte-lib: Fix telemetry for streaming #34955

Merged
merged 1 commit into from
Feb 9, 2024
Merged

airbyte-lib: Fix telemetry for streaming #34955

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
22 changes: 13 additions & 9 deletions airbyte-lib/airbyte_lib/source.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -407,6 +407,8 @@ def _read_with_catalog(
"""
source_tracking_information = self.executor.get_telemetry_info()
send_telemetry(source_tracking_information, cache_info, SyncState.STARTED)
sync_failed = False
self._processed_records = 0 # Reset the counter before we start
try:
with as_temp_files(
[self._config, catalog.json(), json.dumps(state) if state else "[]"]
Expand All @@ -430,14 +432,16 @@ def _read_with_catalog(
send_telemetry(
source_tracking_information, cache_info, SyncState.FAILED, self._processed_records
)
sync_failed = True
raise
finally:
send_telemetry(
source_tracking_information,
cache_info,
SyncState.SUCCEEDED,
self._processed_records,
)
if not sync_failed:
send_telemetry(
source_tracking_information,
cache_info,
SyncState.SUCCEEDED,
self._processed_records,
)

def _add_to_logs(self, message: str) -> None:
self._last_log_messages.append(message)
Expand All @@ -460,11 +464,13 @@ def _execute(self, args: list[str]) -> Iterator[AirbyteMessage]:
for line in self.executor.execute(args):
try:
message = AirbyteMessage.parse_raw(line)
yield message
if message.type is Type.RECORD:
self._processed_records += 1
if message.type == Type.LOG:
self._add_to_logs(message.log.message)
if message.type == Type.TRACE and message.trace.type == TraceType.ERROR:
self._add_to_logs(message.trace.error.message)
yield message
except Exception:
self._add_to_logs(line)
except Exception as e:
Expand All @@ -481,8 +487,6 @@ def _tally_records(
progress.reset(len(self._selected_stream_names or []))

for message in messages:
if message.type is Type.RECORD:
self._processed_records += 1
yield message
progress.log_records_read(self._processed_records)

Expand Down
44 changes: 29 additions & 15 deletions airbyte-lib/tests/integration_tests/test_source_test_fixture.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,8 @@
from collections.abc import Mapping
import os
import shutil
import subprocess
import itertools
from contextlib import nullcontext as does_not_raise
from typing import Any
from unittest.mock import Mock, call, patch
import tempfile
Expand Down Expand Up @@ -559,16 +560,30 @@ def test_airbyte_lib_version() -> None:
@patch('airbyte_lib.telemetry.requests')
@patch('airbyte_lib.telemetry.datetime')
@pytest.mark.parametrize(
"raises, api_key, expected_state, expected_number_of_records, request_call_fails, extra_env, expected_flags",
"raises, api_key, expected_state, expected_number_of_records, request_call_fails, extra_env, expected_flags, cache_type, number_of_records_read",
[
pytest.param(True, "test_fail_during_sync", "failed", 1, False, {"CI": ""}, {"CI": False}, id="fail_during_sync"),
pytest.param(False, "test", "succeeded", 3, False, {"CI": ""}, {"CI": False}, id="succeed_during_sync"),
pytest.param(False, "test", "succeeded", 3, True, {"CI": ""}, {"CI": False}, id="fail_request_without_propagating"),
pytest.param(False, "test", "succeeded", 3, False, {"CI": ""}, {"CI": False}, id="falsy_ci_flag"),
pytest.param(False, "test", "succeeded", 3, False, {"CI": "true"}, {"CI": True}, id="truthy_ci_flag"),
pytest.param(pytest.raises(Exception), "test_fail_during_sync", "failed", 1, False, {"CI": ""}, {"CI": False}, "duckdb", None, id="fail_during_sync"),
pytest.param(does_not_raise(), "test", "succeeded", 3, False, {"CI": ""}, {"CI": False}, "duckdb", None, id="succeed_during_sync"),
pytest.param(does_not_raise(), "test", "succeeded", 3, True, {"CI": ""}, {"CI": False}, "duckdb", None,id="fail_request_without_propagating"),
pytest.param(does_not_raise(), "test", "succeeded", 3, False, {"CI": ""}, {"CI": False}, "duckdb", None,id="falsy_ci_flag"),
pytest.param(does_not_raise(), "test", "succeeded", 3, False, {"CI": "true"}, {"CI": True}, "duckdb", None,id="truthy_ci_flag"),
pytest.param(pytest.raises(Exception), "test_fail_during_sync", "failed", 1, False, {"CI": ""}, {"CI": False}, "streaming", 3, id="streaming_fail_during_sync"),
pytest.param(does_not_raise(), "test", "succeeded", 2, False, {"CI": ""}, {"CI": False}, "streaming", 2, id="streaming_succeed"),
pytest.param(does_not_raise(), "test", "succeeded", 1, False, {"CI": ""}, {"CI": False}, "streaming", 1, id="streaming_partial_read"),
],
)
def test_tracking(mock_datetime: Mock, mock_requests: Mock, raises: bool, api_key: str, expected_state: str, expected_number_of_records: int, request_call_fails: bool, extra_env: dict[str, str], expected_flags: dict[str, bool]):
def test_tracking(
mock_datetime: Mock,
mock_requests: Mock,
raises, api_key: str,
expected_state: str,
expected_number_of_records: int,
request_call_fails: bool,
extra_env: dict[str, str],
expected_flags: dict[str, bool],
cache_type: str,
number_of_records_read: int
):
"""
Test that the telemetry is sent when the sync is successful.
This is done by mocking the requests.post method and checking that it is called with the right arguments.
Expand All @@ -588,12 +603,11 @@ def test_tracking(mock_datetime: Mock, mock_requests: Mock, raises: bool, api_ke
mock_post.side_effect = Exception("test exception")

with patch.dict('os.environ', extra_env):
if raises:
with pytest.raises(Exception):
with raises:
if cache_type == "streaming":
list(itertools.islice(source.get_records("stream1"), number_of_records_read))
else:
source.read(cache)
else:
source.read(cache)


mock_post.assert_has_calls([
call("https://api.segment.io/v1/track",
Expand All @@ -605,7 +619,7 @@ def test_tracking(mock_datetime: Mock, mock_requests: Mock, raises: bool, api_ke
"version": get_version(),
"source": {'name': 'source-test', 'version': '0.0.1', 'type': 'venv'},
"state": "started",
"cache": {"type": "duckdb"},
"cache": {"type": cache_type},
"ip": "0.0.0.0",
"flags": expected_flags
},
Expand All @@ -623,7 +637,7 @@ def test_tracking(mock_datetime: Mock, mock_requests: Mock, raises: bool, api_ke
"source": {'name': 'source-test', 'version': '0.0.1', 'type': 'venv'},
"state": expected_state,
"number_of_records": expected_number_of_records,
"cache": {"type": "duckdb"},
"cache": {"type": cache_type},
"ip": "0.0.0.0",
"flags": expected_flags
},
Expand Down
Loading