Case study / Torrent-Api-py
Torrent-Api-py
Torrent-Api-py wraps provider-specific scraping logic behind a versioned FastAPI service. The API exposes normalized metadata, optional API-key auth, pagination, provider routing, and serverless-friendly runtime support.
PythonFastAPIaiohttpBeautifulSoupcloudscraperMangum
Request flow
Provider-normalized API flow
Requests enter a versioned API layer, fan out to provider scrapers, pass through a normalizer, and return in one consistent response contract.
01
Client
Search or metadata request
02
FastAPI Router
Versioned API surface + docs
03
Provider Extractors
15+ site-specific scrapers
04
Normalizer
Consistent schema + pagination
05
JSON Response
Unified torrent metadata
Contributions
What was built
Designed the API surface, provider abstraction, and response normalization layer.
Integrated 15+ torrent providers with a shared contract and deployment-ready docs.
Maintained the project publicly as adoption grew past 400+ stars and 200+ forks.
Technical decisions
How the implementation was shaped
Provider-specific scraping code is isolated so upstream breakage does not leak into the public API contract.
Versioned routes protect existing consumers while adding new providers and endpoints.
Serverless-friendly runtime support keeps deployment flexible beyond a single VPS model.
Challenges
Operational and engineering constraints
Upstream site instability and anti-bot behavior across many providers.
Normalizing inconsistent metadata fields into one reliable response shape.
Maintaining a public interface over sources outside direct control.
Outcomes
Why the project matters
400+ GitHub stars and 200+ forks.
Single interface spanning 15+ supported providers.
Live public docs and production deployment available.