A live, interactive full-stack prototype demonstrating an end-to-end data pipeline. This project combines high-performance automated data extraction with a dynamic, mobile-optimized split-pane geospatial interface.
Experience the production deployment live in your browser: π Live Web Application
The application is engineered in two completely decoupled layers to ensure maximum performance, speed, and structural flexibility:
Instead of relying on brittle DOM-based scrapers that fail during visual frontend redesigns, this architecture focuses on Network Interception Mechanics.
- Engine: Built using Python.
- Strategy: Intercepts internal XHR/API endpoints directly from target sources to pull raw, unthrottled data streams.
- Resilience: Ready for headless browser integration via Playwright/Selenium to maintain session contexts and execute complex JavaScript widgets or reactive user interactions.
- Output: Normalizes messy frontend data into a clean, standardized, and production-ready
scraped_events.jsonschema.
A beautifully responsive split-pane interface engineered for immediate user engagement and seamless state management.
- Mapping Layer: Integrated Map API displaying dynamic, precise interactive geolocation plot nodes.
- Dynamic Filtering: On-the-fly category filtering (e.g., Music, Food & Drink, Art) with instant rendering state updates.
- Mobile Optimization: Fully fluid breakpoints designed specifically to maintain viewport integrity on mobile, tablet, and desktop viewports.
βββ scraper.py # Custom Python data engineering & API interception logic
βββ scraped_events.json # Normalized, structured target database payload
βββ index.html # Main viewport structure (Split-pane canvas configuration)
βββ css/ # Mobile-first responsive layouts and presentation layers
βββ js/ # Mapping API integration and dynamic state filtering logic