Skip to content

Latest commit

 

History

History

etl

ETL

ETL pipelines with Apache Airflow.

⚠️Currently, airflow pypi package is only compatible with Python 3.12

Setup (for MacOS only)

Running airflow for the first tme

  1. First update the postgres volume's absolute path manually
  2. Run airflow with docker compose:
make airflow-init
make airflow

From then on, you just need to run the following:

make airflow
  1. Make sure Rust & Cargo are installed on your machine.
  2. Make sure you have Python version 3.12 installed.
  3. Run pipenv install. We only install apache-airflow for the typings when developing airflow DAGS.

Load

After extraction, the data is loaded into the postgres location_etl_db database running on port 5433