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

fix: Fix materialization with ttl=0 bug #2666

Merged
merged 2 commits into from
May 11, 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
13 changes: 11 additions & 2 deletions sdk/python/feast/feature_store.py
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@
import os
import warnings
from collections import Counter, defaultdict
from datetime import datetime
from datetime import datetime, timedelta
from pathlib import Path
from typing import (
TYPE_CHECKING,
Expand Down Expand Up @@ -1080,7 +1080,16 @@ def materialize_incremental(
f"No start time found for feature view {feature_view.name}. materialize_incremental() requires"
f" either a ttl to be set or for materialize() to have been run at least once."
)
start_date = datetime.utcnow() - feature_view.ttl
elif feature_view.ttl.total_seconds() > 0:
start_date = datetime.utcnow() - feature_view.ttl
else:
# TODO(felixwang9817): Find the earliest timestamp for this specific feature
# view from the offline store, and set the start date to that timestamp.
print(
f"Since the ttl is 0 for feature view {Style.BRIGHT + Fore.GREEN}{feature_view.name}{Style.RESET_ALL}, "
"the start date will be set to 1 year before the current time."
)
start_date = datetime.utcnow() - timedelta(weeks=52)
provider = self._get_provider()
print(
f"{Style.BRIGHT + Fore.GREEN}{feature_view.name}{Style.RESET_ALL}"
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,48 @@
from datetime import timedelta

from feast import Entity, FeatureView, Field, FileSource, ValueType
from feast.types import Float32, Int32, Int64

driver_hourly_stats = FileSource(
path="%PARQUET_PATH%", # placeholder to be replaced by the test
timestamp_field="event_timestamp",
created_timestamp_column="created",
)

driver = Entity(name="driver_id", value_type=ValueType.INT64, description="driver id")


driver_hourly_stats_view = FeatureView(
name="driver_hourly_stats",
entities=[driver],
ttl=timedelta(days=0),
schema=[
Field(name="conv_rate", dtype=Float32),
Field(name="acc_rate", dtype=Float32),
Field(name="avg_daily_trips", dtype=Int64),
],
online=True,
source=driver_hourly_stats,
tags={},
)


global_daily_stats = FileSource(
path="%PARQUET_PATH_GLOBAL%", # placeholder to be replaced by the test
timestamp_field="event_timestamp",
created_timestamp_column="created",
)


global_stats_feature_view = FeatureView(
name="global_daily_stats",
entities=[],
ttl=timedelta(days=0),
schema=[
Field(name="num_rides", dtype=Int32),
Field(name="avg_ride_length", dtype=Float32),
],
online=True,
source=global_daily_stats,
tags={},
)
20 changes: 15 additions & 5 deletions sdk/python/tests/integration/online_store/test_e2e_local.py
Original file line number Diff line number Diff line change
Expand Up @@ -101,12 +101,12 @@ def _test_materialize_and_online_retrieval(

def test_e2e_local() -> None:
"""
A more comprehensive than "basic" test, using local provider.
Tests the end-to-end workflow of apply, materialize, and online retrieval.

1. Create a repo.
2. Apply
3. Ingest some data to online store from parquet
4. Read from the online store to make sure it made it there.
This test runs against several different types of repos:
1. A repo with a normal FV and an entity-less FV.
2. A repo using the SDK from version 0.19.0.
3. A repo with a FV with a ttl of 0.
"""
runner = CliRunner()
with tempfile.TemporaryDirectory() as data_dir:
Expand Down Expand Up @@ -143,6 +143,16 @@ def test_e2e_local() -> None:
runner, store, start_date, end_date, driver_df
)

with runner.local_repo(
get_example_repo("example_feature_repo_with_ttl_0.py")
.replace("%PARQUET_PATH%", driver_stats_path)
.replace("%PARQUET_PATH_GLOBAL%", global_stats_path),
"file",
) as store:
_test_materialize_and_online_retrieval(
runner, store, start_date, end_date, driver_df
)

# Test a failure case when the parquet file doesn't include a join key
with runner.local_repo(
get_example_repo("example_feature_repo_with_entity_join_key.py").replace(
Expand Down