dayrize-usecase/dags/sustainability_score/__init__.py

"""
DAG IDs: sustainability_score
"""

import os
from datetime import datetime

import utils

from airflow import DAG
from airflow.operators.dummy_operator import DummyOperator

from airflow.providers.apache.beam.operators.beam import BeamRunPythonPipelineOperator
from airflow.providers.postgres.operators.postgres import PostgresOperator


HOME = os.environ["HOME"]
CSV_FNAME = (
    "large_target_store_products_dataset_sample - "
    "large_target_store_products_dataset_sample.csv"
)
CONFIG = {
    "input": f"{ HOME }/gcs/data/{ CSV_FNAME }",
    "beam_etl_path": "/etl/main.py",
    "products_table": "sustainability_score.products",
    "scored_table": "sustainability_score.scored_products",
}

with DAG(
    "sustainability_score",
    schedule_interval="0 * * * 1-5",
    catchup=False,
    max_active_runs=10,
    start_date=datetime(2023, 6, 21),
    doc_md=utils.load_docs(__file__),
    params=CONFIG,
    template_searchpath=["/sql"],
) as dag:
    create_products_table = PostgresOperator(
        task_id="create_products_table",
        sql="products_schema.sql",
        postgres_conn_id="pg_db",
    )
    create_scores_table = PostgresOperator(
        task_id="create_scored_products_table",
        sql="scored_products_schema.sql",
        postgres_conn_id="pg_db",
    )

    etl_pipeline = BeamRunPythonPipelineOperator(
        task_id="beam_etl",
        py_file="{{ params.beam_etl_path }}",
        pipeline_options={
            "input": "{{ params.input }}",
            "pg_hostname": "{{ conn.get('pg_db').host }}",
            "pg_port": "{{ conn.get('pg_db').port }}",
            "pg_username": "{{ conn.get('pg_db').login }}",
            "pg_password": "{{ conn.get('pg_db').password }}",
            "pg_database": "{{ conn.get('pg_db').schema }}",
            "pg_table": "{{ params.products_table }}",
        },
    )

    calculate_score = PostgresOperator(
        task_id="calculate_score",
        sql="calculate_score.sql",
        postgres_conn_id="pg_db",
    )

    create_products_table >> etl_pipeline
    [etl_pipeline, create_scores_table] >> calculate_score