Scrapling — адаптивный веб-скрейпинг на Python

★ 84.4K

Scrapling is an open-source Python framework for web scraping, from a single request to a full crawl; its parser remembers elements and relocates them when a page layout changes. Reach for it when you monitor competitor prices on a site that keeps redesigning, when scrapers break after every frontend update, when Cloudflare Turnstile blocks plain requests, or when you need a large crawl with several sessions and pause and resume. Inside are an adaptive parser (css with adaptive=True and auto_save), the fetchers Fetcher, AsyncFetcher, StealthyFetcher and DynamicFetcher, a Spider framework with concurrent sessions, automatic proxy rotation and a crawl speed that adapts to how the site responds and backs off when blocked, streaming output with live stats, a CLI and an MCP server. Install it with pip. It extracts data from third-party sites, so use it only lawfully, on resources where collection is permitted and in line with their terms of service. Best for Python developers who want one library for every scraping task; it has no GUI and is not a no-code tool. BSD-3-Clause license.