How cjsladek2/scanadillo works
Project type and purpose
• From the README and file‐tree, we can see this is a full‐stack “Scanadillo” application combining:
• A Next.js front-end (in DilloScan/)
• A Node.js OCR microservice (in image-ocr-app/)
• A Python NLP/ingredients microservice (ingredx/ + server.py)
• Static/public assets (DilloScan/public/)
• The overall user flow is: user uploads/scans an image → OCR extracts text → NLP engine summarizes/translates ingredients or drives a chatbot.
File structure
• Top level:
• DilloScan/ – Next.js app
• image-ocr-app/ – Node.js server (Express) for handling uploads & Tesseract OCR
• ingredx/ + server.py – Python package and entrypoint for summarization/translation via OpenAI
• DilloScan/app/components – React UI components (ImageGallery, chatbot, ingredients, NavBar)
• DilloScan/public – static assets and Tesseract WASM if used client-side
• image-ocr-app/server.js – Express endpoints (/upload, /ocr)
• ingredx/api_server.py or server.py – Python HTTP endpoints for summarizer & translator
• Configuration files: next.config.mjs, package.json, requirements.txt
README for additional insights
• We infer the rest of the system from adjacent folders.