from datetime import datetime
import time
import pandas as pd
from feast import FeatureStore
from feast.data_source import PushMode
import os
from pyspark.sql import SparkSession
from feast.infra.contrib.stream_processor import ProcessorConfig
from feast.infra.contrib.spark_kafka_processor import SparkProcessorConfig
from feast.infra.contrib.stream_processor import get_stream_processor_object
import sys

def preprocess_fn(rows: pd.DataFrame):
    print(f"df columns: {rows.columns}")
    print(f"df size: {rows.size}")
    print(f"df preview:\n{rows.head()}")
    return rows

def start_spark():
    store = FeatureStore(repo_path=".")

    # See https://spark.apache.org/docs/3.1.2/structured-streaming-kafka-integration.html#deploying for notes on why we need this environment variable.
    # os.environ['PYSPARK_SUBMIT_ARGS'] = "--packages=org.apache.spark:spark-sql-kafka-0-10_2.12:3.0.0 pyspark-shell"
    spark = SparkSession.builder.appName("feast-spark").getOrCreate()
    # spark.conf.set("spark.sql.shuffle.partitions", 5)

    ingestion_config = SparkProcessorConfig(mode="spark", source="kafka", spark_session=spark, processing_time="5 seconds",query_timeout=None)
    sfv = store.get_stream_feature_view("driver_hourly_stats_stream")

    processor = get_stream_processor_object(
        config=ingestion_config,
        fs=store,
        sfv=sfv,
        preprocess_fn=preprocess_fn,
    )
    query = processor.ingest_stream_feature_view(PushMode.OFFLINE)


if __name__ == "__main__":
    start_spark()

spark-submit\
 --master local[1] \
 --name feast-spark\
 --packages=org.apache.spark:spark-sql-kafka-0-10_2.12:3.0.0 \
 --conf spark.sql.shuffle.partitions=5 \
 ./test_spark.py

Traceback (most recent call last):
  File "/home/orestis/Desktop/feast_project/feature_streaming/test_spark.py", line 40, in <module>
    start_spark()
  File "/home/orestis/Desktop/feast_project/feature_streaming/test_spark.py", line 27, in start_spark
    ingestion_config = SparkProcessorConfig(mode="spark", source="kafka", spark_session=spark, processing_time="5 seconds",query_timeout=None)
  File "pydantic/main.py", line 342, in pydantic.main.BaseModel.__init__
pydantic.error_wrappers.ValidationError: 1 validation error for SparkProcessorConfig
query_timeout
  none is not an allowed value (type=type_error.none.not_allowed)