Welcome to Bodhi-Cast, an open-source application designed for surf enthusiasts around the globe. Bodhi-Cast offers comprehensive swell and surf forecasting along with real-time weather conditions, enabling users to explore and analyze surf conditions anywhere in the world.
Bodhi-Cast is more than just a surf forecasting app; it's your go-to platform for discovering and assessing surf spots with ease and precision. With the ability to select any location from a map, users can add surf spots and receive detailed forecasts including primary swell, secondary swell, wind waves, and wind speed/direction.
Bodhi-Cast also aims to bring surf data to life with animated visualizations. Utilizing Three.js, the app generates captivating animations that depict wave height and period based on the primary swell, offering a quick and visually engaging way to evaluate surf conditions at a glance.
- Global Surf Spot Discovery: Choose any location on the map to explore surf conditions.
- Comprehensive Data Forecasts: Get detailed information on primary and secondary swells, wind waves, and wind speed/direction. As well as real-time weather data from nearby available stations(only US based stations currently available)
- Animated Wave Visualizations: Visualize surf data with animated representations of wave height and period, powered by Three.js.
Setting up Bodhi-Cast in your local development environment is straightforward, thanks to Docker and Docker Compose. Follow the steps below to get started.
- A Windows, Mac, or Linux environment.
- Docker and Docker Compose installed on your machine.
Clone the Bodhi-Cast repository from GitHub:
git clone https://github.com/peterbull/bodhi-castNavigate to the project directory and create env file:
cd bodhi-cast
cp .env.example .envInstall npm packages for frontend:
(Doing this locally accounts for host machine differences to prevent volume conflicts)
npm install ./frontend/my-app --prefix ./frontend/my-appBuild and run the docker containers:
(This will run the app with the docker override file that includes debug for backend and airflow-worker)
docker compose up --buildWarning: DAGs will begin to pull data as soon as the container is run, so ensure you have some extra hard drive space and compute. Otherwise change the DAG setting to is_paused_upon_creation=True in the gefs_wave_urls_from_kafka.py and gefs_wave_etl_from_kafka.py files prior to running the above command.
- Frontend: Access the Bodhi-Cast user interface at http://localhost:3001. This is where you can explore surf spots, view forecasts, and interact with the animated visualizations.
- Backend: The backend API can be accessed at http://localhost:8000. This serves as the backbone of Bodhi-Cast, handling data processing, forecasting, and API requests.
- Airflow Webserver: For managing and monitoring your workflows, visit the Airflow webserver at http://localhost:8080. This tool is crucial for orchestrating the data pipeline tasks that power the forecasts in Bodhi-Cast.
Default username and password for the webserver are both airflow
Enabling them will begin processing the latest swell data and writing to your local postgis service.
The architecture of Bodhi-Cast incorporates several advanced components, which might initially appear as an overextension for the application's current scope. However, these choices were strategically made to accommodate future expansion and scalability. The components detailed below are foundational to providing a robust, flexible platform capable of evolving alongside the app's growth and user demands.
- Chosen for:
- Composability
- Scalability
- Existing ecosystem
-
Chosen for:
- High performance
- Ease of API development
- Asynchronous support, enhancing concurrency and scalability
-
Development Approach:
- Initial endpoint setup avoids default async for simplicity and stability
-
ORM for PostGIS: Sqlalchemy 1.4
- Initial use of Sqlalchemy 2.0; reverted to 1.4 for Airflow compatibility
Extension for PostgreSQL:
-
Chosen for:
- Strong database management capabilities
- Superior handling of spatial and geographic data
-
Performance Optimization:
- Outperforms PostgreSQL indexing on latitude and longitude for query optimization
-
Spatial Queries:
- Enables efficient querying of points by bounding box, essential for returning swell data for any lat/lon pairing
- Multi Purpose: Enhances performance and reliability by serving as:
- A messaging system for Airflow's Celery executor, ensuring efficient task management
- A cache for the backend API, reducing database load by storing frequently accessed query results
- A cache for current weather data streaming from Kafka
-
Data Pipeline Orchestration:
- Managing complex workflows with precision
- Offering control, fault tolerance, and transparency
-
Adaptability:
- Ensures Bodhi-Cast's data handling evolves without losing robustness or adaptability
-
Framework for Real-Time Data:
- Kafka's architecture is crucial for handling and processing real-time data streams
-
Future-Proofing Bodhi-Cast:
- Prepares the app for scalable growth and further real-time data integration
Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make to Bodhi-Cast are greatly appreciated.
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/AmazingFeature) - Commit your Changes (
git commit -m 'Add some AmazingFeature') - Push to the Branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Bodhi-Cast is under the Apache License 2.0. See the LICENSE file for more details. This license allows for a permissive usage of the software, provided that the proper credit is given by retaining this license notice. For more details on the terms, please review the license directly.