feat: Airflow ELT 파이프라인 (#182)

01-batch-serving(airflow)/dags/05-python-operator-with-slack-noti.py

@@ -11,7 +11,7 @@
     'owner': 'kyle',
     'depends_on_past': False,
     'start_date': datetime(2024, 1, 1),
-    'start_date': datetime(2024, 1, 4),
+    'end_date': datetime(2024, 1, 4),
     'retires': 1,
     'retry_delay': timedelta(minutes=5),
 }

01-batch-serving(airflow)/dags/06-simple_elt.py

@@ -0,0 +1,143 @@
+# ELT 파이프라인을 개발합니다(Extract - Load - Transfer)
+# 1) Extract : 데이터를 추출하는 과정. 여기선 Cloud Storage에 업로드(현업에선 Database에서 추출)
+# 2) Load : Cloud Storage의 Bucket에 Data 저장된 것을 데이터 웨어하우스인 BigQuery로 저장
+# 3) Transform : Load로 저장된 데이터를 Transform. BigQuery 쿼리로 데이터를 처리
+# pip install apache-airflow-providers-google==10.14.0
+
+from airflow import DAG
+from datetime import datetime, timedelta
+from pathlib import Path
+from airflow.providers.google.cloud.transfers.gcs_to_bigquery import GCSToBigQueryOperator
+from airflow.providers.google.cloud.operators.bigquery import BigQueryExecuteQueryOperator
+from airflow.providers.google.cloud.transfers.local_to_gcs import LocalFilesystemToGCSOperator
+from utils.slack_notifier import task_fail_slack_alert, task_succ_slack_alert
+
+execution_date = '{{ ds_nodash }}' #20240101
+execution_date_with_slash = "{{ macros.ds_format(ds, '%Y-%m-%d', '%Y/%m/%d') }}" # 2024/01/01
+
+# 아래 2개는 여러분들의 Google Cloud Project, Bucket 입력
+PROJECT_ID = "boostcamp-ai-tech-serving"
+BUCKET_NAME = "boostcamp-ai-tech-gcs"
+
+FILE_NAME = f"bike_data_{execution_date}.csv"
+LOCAL_FILE_PATH = str(Path(__file__).parent.parent / "data" / FILE_NAME) # 파일의 할아버지(dags -> 01-batch-serving(airflow))의 data 폴더
+
+GCS_PATH = f"{execution_date_with_slash}/bike_data.csv"
+
+default_args = {
+    'owner': 'kyle',
+    'depends_on_past': False,
+    'start_date': datetime(2024, 1, 1),
+    'end_date': datetime(2024, 1, 4),
+    'retires': 1,
+    'retry_delay': timedelta(minutes=5),
+}
+
+schema_fields = [
+  {
+    "mode": "NULLABLE",
+    "name": "trip_id",
+    "type": "INTEGER"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "subscriber_type",
+    "type": "STRING"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "bikeid",
+    "type": "STRING"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "start_time",
+    "type": "TIMESTAMP"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "start_station_id",
+    "type": "INTEGER"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "start_station_name",
+    "type": "STRING"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "end_station_id",
+    "type": "STRING"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "end_station_name",
+    "type": "STRING"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "duration_minutes",
+    "type": "INTEGER"
+  },
+  {
+    "mode": "NULLABLE",
+    "name": "dummy_date",
+    "type": "DATE"
+  }
+]
+
+
+with DAG(
+    dag_id='simple_elt_pipeline',
+    default_args=default_args,
+    schedule_interval='30 0 * * *',
+    tags=['my_dags'],
+    catchup=False,
+    on_failure_callback=task_fail_slack_alert,
+    # on_success_callback=task_succ_slack_alert  # 성공 알림 필요 시 추가
+) as dag:
+
+    extract_data = LocalFilesystemToGCSOperator(
+        task_id="extract_data",
+        src=LOCAL_FILE_PATH,
+        bucket=BUCKET_NAME,
+        dst=GCS_PATH
+    )
+
+
+    load_csv = GCSToBigQueryOperator(
+        task_id="gcs_to_bigquery",
+        gcp_conn_id="google_cloud_default",
+        bucket=f'{BUCKET_NAME}', 
+        source_objects=[GCS_PATH], 
+        schema_fields=schema_fields,
+        source_format='CSV',
+        skip_leading_rows=1,
+        create_disposition='CREATE_IF_NEEDED',
+        destination_project_dataset_table=f'{PROJECT_ID}.temp.bike_{execution_date}', # temp dataset 생성 필요
+        write_disposition='WRITE_TRUNCATE'
+    )
+
+    # dummy_date 별 COUNT
+    sql_query = f"""
+    SELECT 
+      dummy_date, 
+      start_station_id, 
+      end_station_id, 
+      COUNT(bikeid) as cnt
+    FROM `{PROJECT_ID}.temp.bike_{execution_date}`
+    GROUP BY 
+      dummy_date, 
+      start_station_id, 
+      end_station_id
+    """
+
+    transform = BigQueryExecuteQueryOperator(
+        task_id='run_query',
+        sql=sql_query,
+        use_legacy_sql=False,
+        write_disposition='WRITE_TRUNCATE',
+        destination_dataset_table=f"temp.bike_agg_{execution_date}"
+    )
+
+    extract_data >> load_csv >> transform