Chainsaw-OCR is a Python command-line application that retrieves manga chapter data via the MangaDex API, downloads associated page images, performs OCR text extraction, and stores results for search and analysis.
This project was built to explore API interaction, database design, file system automation, and OCR processing in a practical, end-to-end workflow.
- Retrieves structured chapter metadata via the MangaDex API
- Stores chapter and page data in an SQLite database
- Automates bulk image downloading and directory organisation
- Extracts text from images using OCR
- Designed for future expansion into searchable text queries
- Python
- SQLite
- MangaDex API
- Git
- Linux (developed and tested in Arch)
This project began as a curiosity-driven experiment and evolved into a hands-on exercise in building a small data pipeline. The goal was to practice working with external APIs, persistent data storage, file handling, and text extraction within a command-line environment