Airflow ETL Pipeline

Project Summary

Create an ETL pipeline using Airflow as the scheduler grab book list from https://www.amazon.com/s?k=machine+learning+interview and store the unique book list to postgres database including price, author, price, rating

Tech Stack

• Python3
• Postgres
• Docker
• Airflow

What is ETL?

ETL stands for Extract, Transform and Load.

What is Airflow?

Apache Airflow is an open-source platform for orchestrating and scheduling complex workflows, particularly ETL pipelines and data processes.

What is DAG?

DAG (Directed Acyclic Graph) in Airflow is core structure defining a workflow, a DAG organized the tasks in a sequnce and in one direction so that each task won't depend on itself directly or indirectly. Each DAG contains Tasks, Operators and Dependencies. Tasks as units of work, Operators as Task Types like PythonOperator, BashOperator whereas Dependencies defining the execution order.

ETL Flow

• Step 1: Create Books Table (PostgresOperator)
• Step 1: Extract data from Amazon via Web Scraping (PostgresOperator)
• Step 2: Transform data to book table for storing (PostgresOperator)
• Step 3: Load the data to Postgres SQL

Step by Step Guide on creating Airflow pipeline

1. Open Visual Studio and Create a Folder
2. Run below command for grabbing docker compose file

   curl -LfO 'https://airflow.apache.org/docs/apache-airflow/2.10.2/docker-compose.yaml'

3. Create Environment variable files .env

You can Run below command for grabbing env variable file

copy below content

# Common environment variables
AIRFLOW_IMAGE_NAME=apache/airflow:2.10.2
AIRFLOW_UID=501
AIRFLOW_PROJ_DIR=.
POSTGRES_USER=airflow
POSTGRES_PASSWORD=airflow
POSTGRES_DB=airflow

# Airflow settings
AIRFLOW__CORE__EXECUTOR=CeleryExecutor
AIRFLOW__DATABASE__SQL_ALCHEMY_CONN=postgresql+psycopg2://airflow:airflow@postgres/airflow
AIRFLOW__CELERY__RESULT_BACKEND=db+postgresql://airflow:airflow@postgres/airflow
AIRFLOW__CELERY__BROKER_URL=redis://:@redis:6379/0
AIRFLOW__CORE__FERNET_KEY=
AIRFLOW__CORE__DAGS_ARE_PAUSED_AT_CREATION=true
AIRFLOW__CORE__LOAD_EXAMPLES=true
AIRFLOW__API__AUTH_BACKENDS=airflow.api.auth.backend.basic_auth,airflow.api.auth.backend.session
AIRFLOW__SCHEDULER__ENABLE_HEALTH_CHECK=true
_AIRFLOW_WWW_USER_CREATE=true
_AIRFLOW_WWW_USER_USERNAME=airflow
_AIRFLOW_WWW_USER_PASSWORD=airflow
_PIP_ADDITIONAL_REQUIREMENTS=

# PgAdmin settings
PGADMIN_DEFAULT_EMAIL=admin@admin.com
PGADMIN_DEFAULT_PASSWORD=root

# Celery worker settings
DUMB_INIT_SETSID=0
CONNECTION_CHECK_MAX_COUNT=0

4. Run below commands to set the right airflow user.

     mkdir -p ./dags ./logs ./plugins ./config
     echo -e "AIRFLOW_UID=$(id -u)" > .env

5. Create first user account via below command

docker compose up airflow-init

6. Run the docker compose

docker compose up

1. Set data in pgAdmin Web UI

• Open localhost:5050
• Username: admin@admin.com, password root
• Create server Go to General Tab: Name: ps_db
• Set connection parameters Go to Connection Tab: Host name/address = postgres Port: 5432 Username: airflow Password: airflow
• create database called amazon-books

1. Set DB Connection string in airflow Go to Admin Tab, add a new connection. Connection Id: books_connection Connection Type: Postgres Description: Postgres Connection Host: postgres Database: amazon-books

2. Set Variable in airflow Go to Admin Tab, add a new variable Variable: key = amazon_book_search_term, val: <Your preferred term>

3. Run the job Go to DAGs Tab, in Search DAGS, search fetch_and_store_amazon_books Hit Icon for running the job

4. (Optional) Clean up Environment

docker compose down --volumes --remove-orphansd