X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/98b2c09ef1f1c8288a31517f61423264c6b3291c..f99929a3fa250a4d39388013b990334ae1f7c779:/src/wolnelektury/utils.py diff --git a/src/wolnelektury/utils.py b/src/wolnelektury/utils.py index 8c5ead612..2657a5542 100644 --- a/src/wolnelektury/utils.py +++ b/src/wolnelektury/utils.py @@ -12,6 +12,7 @@ from functools import wraps import pytz from inspect import getargspec +import re from django.core.mail import send_mail from django.http import HttpResponse from django.template import RequestContext @@ -149,3 +150,19 @@ class UnicodeCSVWriter(object): def writerows(self, rows): for row in rows: self.writerow(row) + + +# the original re.escape messes with unicode +def re_escape(s): + return re.sub(r"[(){}\[\].*?|^$\\+-]", r"\\\g<0>", s) + + +BOT_BITS = ['bot', 'slurp', 'spider', 'facebook', 'crawler', 'parser', 'http'] + + +def is_crawler(request): + user_agent = request.META.get('HTTP_USER_AGENT') + if not user_agent: + return True + user_agent = user_agent.lower() + return any(bot_bit in user_agent for bot_bit in BOT_BITS)