How cdhigh/kindleear works
Project type and purpose
• From the file tree and README, this is a full-stack Python web application (“KindleEar”) that aggregates arbitrary web content (RSS/ATOM/Calibre recipes/custom feeds), converts it into EPUB/MOBI/MP3 via built-in ebook generation, summarization, translation and TTS engines, and then delivers it to users’ Kindles or serves it through an online reader.
File structure
• Top-level directories:
• application/ – core code, split into:
· back_end/ – adapters for DB, mail, task queues
· view/ – URL handlers (admin, library, deliver, reader, translator, etc.)
· lib/ – domain libraries (Calibre engine, summarizer, translator, TTS, dictionary, scraping, ebook building)
· templates/ & static/ – HTML templates and front-end assets
· images/, recipes/, translations/ – data assets
· routes.py, main.py – WSGI entrypoints and URL routing
• docker/, tools/ – Dockerfiles, CI/CD scripts, deployment helpers
• config.py, app.yaml, queue.yaml, cron.yaml, dispatch.yaml – configuration for GAE, cron jobs, task queues
• docs/, tests/ – documentation and test suite
• Architectural patterns:
• Adapter pattern in back_end (send_mail_adpt.py, task_queue_adpt.py)
• Plugin structure in lib (calibre/customize, ebook_translator/engines, ebook_tts/engines)
• Monolithic codebase with modular subpackages
• Support for multiple task queue engines (Celery, RQ, APScheduler, GAE push queue)
Extract additional insights from the README
• The app supports: web UI, scheduled jobs (daily pushes or on-demand), email delivery, online reader, library sharing, browser extension for recipe creation, and integration with AI summarization/translation/TTS via multiple third-party APIs.