This is a short two week project I completed to find popular movies based on torrent stats. The app is currently deployed at http://filmcloud.appspot.com/. This project was just hacked together, please excuse the quality of the code.
- Scrape torrent search sites for popular movies
- Scrape IMDB for poster + info on movies
- Display movies using a Javascript carousel, grouping movies by genre and ordering by swarm size
- Movie trailer sourced from YouTube and embedded in the page
- YQL + XPath admin interface for adding new source torrent sites.
- Python
- Django
- GAE
- YQL
- ImdbPy
- GData API (YouTube)
- YUI
The project is designed for deployment on AppEngine please review their documentation for deployment instructions. Once deployed hit: http://{YOUR APP}.appspot.com/admin, you should be asked to login with your google username and password. Once inside you will need to setup listing settings, movie sources and categories.
Movie Listing Settings: The settings consist of two paths, used to extract the torrent name and leach count from the YQL result.
E.g. for The Pirate Bay
Name XPath:
//a[starts-with(@href, "/torrent")]
Leaches XPath:
//td[4]/p
Movie Listing Sources: This is a YQL query used to scrape a torrent site, it should be paired with a movie listing setting.
E.g. The Pirate Bay YQL:
select * from html where url="http://thepiratebay.org/browse/201/0/9" and xpath='//table[@id="searchResult"]/tr'
Categories: These are used to group movies on the home page, they are sourced from IMDB genres.
MIT
Sam Cavenagh [email protected]