Scraping the High Rollers Tennis bets
This project consists of two Python scripts to scrape data from the Stake website using GraphQL API and Flask to serve the scraped data via a RESTful API. The first script (Stake_Scraper.py
) continuously fetches betting data, converts currency, and stores it in a CSV file. The second script (APIStake.py
) creates a Fast-API web server to serve the scraped data.
- Python 3.x
- Libraries: requests, bs4 (BeautifulSoup), Flask
-
Clone the repository: git clone https://github.com/mdSHash/Stake-Scraper cd Stake-Scraper
-
Install dependencies: pip install -r requirements.txt
-
Run the
Stake_Scraper.py
script to continuously fetch and store data: python stake_scraper.py -
Run the
APIStake.py
script to start the Flask server: python app.py -
Access the API endpoint to get the scraped data: http://localhost:8000/api/data
- Ensure that you have proper internet connectivity to scrape data from the Stake website.
- The CSV file (
Tennis_Bets.csv
) will be continuously updated with new data. - Adjust the scraping interval and other parameters as needed.