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.

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

simple update to add a lookback_window to the datetime stream slicer #14609

Merged
merged 2 commits into from
Jul 12, 2022
Merged
Show file tree
Hide file tree
Changes from all 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
Original file line number Diff line number Diff line change
Expand Up @@ -35,17 +35,21 @@ def __init__(
cursor_value: Union[InterpolatedString, str],
datetime_format: str,
config: Config,
lookback_window: Union[InterpolatedString, str] = None,
):
self._timezone = datetime.timezone.utc
self._interpolation = JinjaInterpolation()
if isinstance(cursor_value, str):
cursor_value = InterpolatedString(cursor_value)
if isinstance(lookback_window, str):
lookback_window = InterpolatedString(lookback_window)
self._datetime_format = datetime_format
self._start_datetime = start_datetime
self._end_datetime = end_datetime
self._step = self._parse_timedelta(step)
self._config = config
self._cursor_value = cursor_value
self._lookback_window = lookback_window

# If datetime format is not specified then start/end datetime should inherit it from the stream slicer
if not self._start_datetime.datetime_format:
Expand All @@ -58,7 +62,9 @@ def stream_slices(self, sync_mode: SyncMode, stream_state: Mapping[str, Any]) ->
stream_state = stream_state or {}
kwargs = {"stream_state": stream_state}
end_datetime = min(self._end_datetime.get_datetime(self._config, **kwargs), datetime.datetime.now(tz=datetime.timezone.utc))
start_datetime = min(self._start_datetime.get_datetime(self._config, **kwargs), end_datetime)
lookback_delta = self._parse_timedelta(self._lookback_window.eval(self._config, **kwargs) if self._lookback_window else "0d")
start_datetime = self._start_datetime.get_datetime(self._config, **kwargs) - lookback_delta
start_datetime = min(start_datetime, end_datetime)
if self._cursor_value and self._cursor_value.eval(self._config, **kwargs):
cursor_datetime = self.parse_date(self._cursor_value.eval(self._config, **kwargs))
else:
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ def mock_datetime_now(monkeypatch):


@pytest.mark.parametrize(
"test_name, stream_state, start, end, step, cursor, expected_slices",
"test_name, stream_state, start, end, step, cursor, lookback_window, expected_slices",
[
(
"test_1_day",
Expand All @@ -39,6 +39,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("2021-01-10", datetime_format=datetime_format),
"1d",
cursor_value,
None,
[
{"start_date": "2021-01-01", "end_date": "2021-01-01"},
{"start_date": "2021-01-02", "end_date": "2021-01-02"},
Expand All @@ -59,6 +60,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("2021-01-10", datetime_format=datetime_format),
"2d",
cursor_value,
None,
[
{"start_date": "2021-01-01", "end_date": "2021-01-02"},
{"start_date": "2021-01-03", "end_date": "2021-01-04"},
Expand All @@ -74,6 +76,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("2021-01-10", datetime_format=datetime_format),
"1d",
cursor_value,
None,
[
# FIXME: should this include 2021-01-05?
{"start_date": "2021-01-05", "end_date": "2021-01-05"},
Expand All @@ -91,6 +94,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("2021-01-10", datetime_format=datetime_format),
"12d",
cursor_value,
None,
[
{"start_date": "2021-01-01", "end_date": "2021-01-10"},
],
Expand All @@ -102,6 +106,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime(f"{(FAKE_NOW + datetime.timedelta(days=1)).strftime(datetime_format)}", datetime_format=datetime_format),
"1d",
cursor_value,
None,
[
{"start_date": "2021-12-28", "end_date": "2021-12-28"},
{"start_date": "2021-12-29", "end_date": "2021-12-29"},
Expand All @@ -117,6 +122,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("{{ stream_state['date'] }}", datetime_format=datetime_format),
"1d",
cursor_value,
None,
[
{"start_date": "2021-01-05", "end_date": "2021-01-05"},
],
Expand All @@ -128,6 +134,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("2021-01-10", datetime_format=datetime_format),
"1d",
InterpolatedString("{{ stream_state['date'] }}"),
None,
[
{"start_date": "2021-01-05", "end_date": "2021-01-05"},
{"start_date": "2021-01-06", "end_date": "2021-01-06"},
Expand All @@ -144,6 +151,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("2021-01-10", datetime_format=datetime_format),
"1d",
InterpolatedString("{{ stream_state['date'] }}"),
None,
[
{"start_date": "2021-01-05", "end_date": "2021-01-05"},
{"start_date": "2021-01-06", "end_date": "2021-01-06"},
Expand All @@ -160,6 +168,7 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("2021-01-10", max_datetime="{{ stream_state['date'] }}", datetime_format=datetime_format),
"1d",
None,
None,
[
{"start_date": "2021-01-01", "end_date": "2021-01-01"},
{"start_date": "2021-01-02", "end_date": "2021-01-02"},
Expand All @@ -175,6 +184,42 @@ def mock_datetime_now(monkeypatch):
MinMaxDatetime("2021-01-10", max_datetime="{{ stream_state['date'] }}"),
"1d",
None,
None,
[
{"start_date": "2021-01-01", "end_date": "2021-01-01"},
{"start_date": "2021-01-02", "end_date": "2021-01-02"},
{"start_date": "2021-01-03", "end_date": "2021-01-03"},
{"start_date": "2021-01-04", "end_date": "2021-01-04"},
{"start_date": "2021-01-05", "end_date": "2021-01-05"},
],
),
(
"test_with_lookback_window_from_start_date",
{"date": "2021-01-05"},
MinMaxDatetime("{{ config['start_date'] }}"),
MinMaxDatetime("2021-01-10", max_datetime="{{ stream_state['date'] }}"),
"1d",
None,
"3d",
[
{"start_date": "2020-12-29", "end_date": "2020-12-29"},
{"start_date": "2020-12-30", "end_date": "2020-12-30"},
{"start_date": "2020-12-31", "end_date": "2020-12-31"},
{"start_date": "2021-01-01", "end_date": "2021-01-01"},
{"start_date": "2021-01-02", "end_date": "2021-01-02"},
{"start_date": "2021-01-03", "end_date": "2021-01-03"},
{"start_date": "2021-01-04", "end_date": "2021-01-04"},
{"start_date": "2021-01-05", "end_date": "2021-01-05"},
],
),
(
"test_with_lookback_window_defaults_to_0d",
{"date": "2021-01-05"},
MinMaxDatetime("{{ config['start_date'] }}"),
MinMaxDatetime("2021-01-10", max_datetime="{{ stream_state['date'] }}"),
"1d",
None,
"{{ config['does_not_exist'] }}",
[
{"start_date": "2021-01-01", "end_date": "2021-01-01"},
{"start_date": "2021-01-02", "end_date": "2021-01-02"},
Expand All @@ -185,8 +230,16 @@ def mock_datetime_now(monkeypatch):
),
],
)
def test_stream_slices(mock_datetime_now, test_name, stream_state, start, end, cursor, step, expected_slices):
slicer = DatetimeStreamSlicer(start, end, step, cursor, datetime_format, config)
def test_stream_slices(mock_datetime_now, test_name, stream_state, start, end, cursor, step, lookback_window, expected_slices):
slicer = DatetimeStreamSlicer(
start_datetime=start,
end_datetime=end,
step=step,
cursor_value=cursor,
datetime_format=datetime_format,
lookback_window=lookback_window,
config=config,
)
stream_slices = slicer.stream_slices(SyncMode.incremental, stream_state)

assert expected_slices == stream_slices
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -115,6 +115,7 @@ def test_datetime_stream_slicer():
datetime: "{{ config['end_time'] }}"
step: "10d"
cursor_value: "created"
lookback_window: "5d"
"""

config = parser.parse(content)
Expand All @@ -129,6 +130,7 @@ def test_datetime_stream_slicer():
assert stream_slicer._end_datetime._datetime_interpolator._string == "{{ config['end_time'] }}"
assert stream_slicer._step == datetime.timedelta(days=10)
assert stream_slicer._cursor_value._string == "created"
assert stream_slicer._lookback_window._string == "5d"


def test_full_config():
Expand Down