scrapy-playwright-cloud-browser

A Scrapy extension that integrates with surfsky.io for web scraping using Playwright. This extension allows you to use Chrome anti-detection browsers through Surfsky.io's cloud browser service with Playwright integration, helping you avoid detection while scraping challenging websites.

Installation

pip install scrapy-playwright-cloud-browser

Usage

Setup environment variables in settings.py in CLOUD_BROWSER namespace:

CLOUD_BROWSER = {
    "API_HOST": <HOST>,
    "API_TOKEN": <API_TOKEN>,
    "NUM_BROWSERS": <NUM_BROWSERS>,
    "PROXIES": [<proxy>],
    "INIT_HANDLER": <INIT_HANDLER>,
    "PAGES_PER_BROWSER": <PAGES_PER_BROWSER>,
    "START_SEMAPHORES": <START_SEMAPHORES>,
    "PROXY_ORDERING": <PROXY_ORDERING>,
    "BROWSER_SETTINGS": <BROWSER_SETTINGS>,
    "FINGERPRINT": <FINGERPRINT>
}

Configuration Parameters

• API_HOST: The URL of the Surfsky.io API host.
• API_TOKEN: Your authentication token for the Surfsky.io service.
• NUM_BROWSERS (default: 1): Number of browser instances to run in parallel. Increase this value to improve throughput for large-scale scraping.
• PROXIES: A list of proxy URLs to use with the browsers. Each browser will be assigned a proxy from this list according to the PROXY_ORDERING strategy.
• INIT_HANDLER: Custom initialization handler for browser setup.
• PAGES_PER_BROWSER (default: 100): Maximum number of pages a browser instance will process before being recycled.
• START_SEMAPHORES (default: 10): Controls how many browsers can be started simultaneously. This prevents overwhelming the system with too many concurrent browser startups.
• PROXY_ORDERING (default: 'random'): Strategy for assigning proxies to browsers. Options are:
  • 'random': Randomly select a proxy from the list for each browser
  • 'round-robin': Cycle through the proxy list in order

For detailed browser settings (BROWSER_SETTINGS) and fingerprint configuration options (FINGERPRINT), please refer to the Surfsky API Reference.

Add cloud browser handlers and change reactor in settings.py:

TWISTED_REACTOR = "twisted.internet.asyncioreactor.AsyncioSelectorReactor"

EXTENSIONS = {
    'scrapy_playwright_cloud_browser.CloudBrowserExtension': 100,
}

License

scrapy-playwright-cloud-browser is distributed under the terms of the MIT license.