I need a small, reliable scraper that pulls the GIS (JSON/GeoJSON) layers showing every district (quartiere/municipio) for each of Italy’s major cities from a well-known real-estate website. The site exposes its map tiles through client-side calls, so the script will have to locate the endpoint that serves the district geometry and capture the full polygon data—not just centre-points. My ideal workflow is straightforward: run the script once, get a single GeoJSON file (or separate files per city) containing the district name, the geometry, and any internal ID the site already provides. Python with requests + aiohttp, pandas, or similar would be perfect, but I am open to another language if it achieves the same clean result and is easy for me to rerun later. Please make sure the scraper: • Navigates any lazy-loaded or JavaScript layers (headless browser or direct API calls—whichever is lighter). • Handles rate limits gracefully and retries without losing data. • Outputs valid, deduplicated GeoJSON so I can import it into QGIS or PostGIS immediately. Deliverables: 1. The complete GeoJSON (district boundaries for Rome, Milan, Naples, Turin, Bologna, Florence, Genoa, Bari, Palermo, Catania, and Venice). 2. Source code with clear instructions in a README so I can rerun the job when boundaries change. 3. Brief log or CSV that maps the original site IDs to the district names, just for reference. If part of the site is protected behind a token, please include the token retrieval step in the script rather than hard-coding it. Feel free to use any open-source libraries that keep dependencies minimal. Let me know your proposed approach and the timeframe you need; the sooner I can load these districts into my GIS workflow, the better.