http_user_agent.py 2.0 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566
  1. # SPDX-License-Identifier: AGPL-3.0-or-later
  2. """
  3. Method ``http_user_agent``
  4. --------------------------
  5. The ``http_user_agent`` method evaluates a request as the request of a bot if
  6. the User-Agent_ header is unset or matches the regular expression
  7. :py:obj:`USER_AGENT`.
  8. .. _User-Agent:
  9. https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/User-Agent
  10. """
  11. from __future__ import annotations
  12. import re
  13. from ipaddress import (
  14. IPv4Network,
  15. IPv6Network,
  16. )
  17. import werkzeug
  18. from searx.extended_types import SXNG_Request
  19. from . import config
  20. from ._helpers import too_many_requests
  21. USER_AGENT = (
  22. r'('
  23. + r'unknown'
  24. + r'|[Cc][Uu][Rr][Ll]|[wW]get|Scrapy|splash|JavaFX|FeedFetcher|python-requests|Go-http-client|Java|Jakarta|okhttp'
  25. + r'|HttpClient|Jersey|Python|libwww-perl|Ruby|SynHttpClient|UniversalFeedParser|Googlebot|GoogleImageProxy'
  26. + r'|bingbot|Baiduspider|yacybot|YandexMobileBot|YandexBot|Yahoo! Slurp|MJ12bot|AhrefsBot|archive.org_bot|msnbot'
  27. + r'|MJ12bot|SeznamBot|linkdexbot|Netvibes|SMTBot|zgrab|James BOT|Sogou|Abonti|Pixray|Spinn3r|SemrushBot|Exabot'
  28. + r'|ZmEu|BLEXBot|bitlybot|HeadlessChrome'
  29. # unmaintained Farside instances
  30. + r'|'
  31. + re.escape(r'Mozilla/5.0 (compatible; Farside/0.1.0; +https://farside.link)')
  32. # other bots and client to block
  33. + '|.*PetalBot.*'
  34. + r')'
  35. )
  36. """Regular expression that matches to User-Agent_ from known *bots*"""
  37. _regexp = None
  38. def regexp_user_agent():
  39. global _regexp # pylint: disable=global-statement
  40. if not _regexp:
  41. _regexp = re.compile(USER_AGENT)
  42. return _regexp
  43. def filter_request(
  44. network: IPv4Network | IPv6Network,
  45. request: SXNG_Request,
  46. cfg: config.Config, # pylint: disable=unused-argument
  47. ) -> werkzeug.Response | None:
  48. user_agent = request.headers.get('User-Agent', 'unknown')
  49. if regexp_user_agent().match(user_agent):
  50. return too_many_requests(network, f"bot detected, HTTP header User-Agent: {user_agent}")
  51. return None