def before_request():
"""
Store API anonymous cookie in session or if it exists, check if it has expired
"""
g.request_start_time = time.time()
g.request_time = lambda: "{:.3f}s".format(
(time.time() - g.request_start_time))
if 'cookies' not in session:
session['cookies'] = {}
if 'auth' not in session or is_expired(session['auth']):
session['auth'] = api.bootstrap()
def before_request():
"""
Store API anonymous cookie in session or if it exists, check if it has expired
"""
if request.path in ('/ready', '/alive'):
# Do not bootstrap readiness/liveness probes
return
g.request_start_time = time.time()
g.request_time = lambda: "{:.3f}s".format(
(time.time() - g.request_start_time))
if 'cookies' not in session:
session['cookies'] = {}
if request.cookies.get('session'):
# Re-use BBB session, if it is valid, the same BBB token will be returned by bootstrap
# thus if the user was authenticated, it will use the user token
session['cookies']['session'] = request.cookies.get('session')
if 'auth' not in session or is_expired(session['auth']):
user_agent = request.headers.get('User-Agent')
remote_ip = get_remote_address()
#### For testing purposes:
#user_agent = "Googlebot"
#remote_ip = "66.249.66.1" # crawl-66-249-66-1.googlebot.com.
#user_agent = "DuckDuckBot"
#remote_ip = "50.16.241.117"
#remote_ip = "127.0.0.1"
evaluation = crawlers.evaluate(remote_ip, user_agent)
if evaluation == crawlers.VERIFIED_BOT:
# Extremely high rate limit
session['auth'] = {
'access_token': app.config['VERIFIED_BOTS_ACCESS_TOKEN'],
'expire_in': "2050-01-01T00:00:00",
'bot': True
}
elif evaluation == crawlers.UNVERIFIABLE_BOT:
# Slightly higher rate limit
session['auth'] = {
'access_token': app.config['UNVERIFIABLE_BOTS_ACCESS_TOKEN'],
'expire_in': "2050-01-01T00:00:00",
'bot': True
}
elif evaluation == crawlers.POTENTIAL_MALICIOUS_BOT:
# Rate limits as a regular user with the advantage that there is no bootstrap
session['auth'] = {
'access_token': app.config['MALICIOUS_BOTS_ACCESS_TOKEN'],
'expire_in': "2050-01-01T00:00:00",
'bot': True
}
else:
session['auth'] = api.bootstrap()
def before_request():
"""
Store API anonymous cookie in session or if it exists, check if it has expired
"""
if request.path in ('/ready', '/alive'):
# Do not bootstrap readiness/liveness probes
return
g.request_start_time = time.time()
g.request_time = lambda: "{:.3f}s".format((time.time() - g.request_start_time))
if 'cookies' not in session:
session['cookies'] = {}
if 'auth' not in session or is_expired(session['auth']):
user_agent = request.headers.get('User-Agent')
remote_ip = get_remote_address()
#### For testing purposes:
#user_agent = "Googlebot"
#remote_ip = "66.249.66.1" # crawl-66-249-66-1.googlebot.com.
#user_agent = "DuckDuckBot"
#remote_ip = "50.16.241.117"
#remote_ip = "127.0.0.1"
evaluation = crawlers.evaluate(remote_ip, user_agent)
if evaluation == crawlers.VERIFIED_BOT:
# Extremely high rate limit
RequestsManager.init(auth={'access_token': app.config['VERIFIED_BOTS_ACCESS_TOKEN'], 'expire_in': "2050-01-01T00:00:00", 'bot': True}, cookies={})
elif evaluation == crawlers.UNVERIFIABLE_BOT:
# Slightly higher rate limit
RequestsManager.init(auth={'access_token': app.config['UNVERIFIABLE_BOTS_ACCESS_TOKEN'], 'expire_in': "2050-01-01T00:00:00", 'bot': True}, cookies={})
elif evaluation == crawlers.POTENTIAL_MALICIOUS_BOT:
# Rate limits as a regular user with the advantage that there is no bootstrap
RequestsManager.init(auth={'access_token': app.config['MALICIOUS_BOTS_ACCESS_TOKEN'], 'expire_in': "2050-01-01T00:00:00", 'bot': True}, cookies={})
if not RequestsManager.is_initialized():
if request.cookies.get('session'):
# - Re-use BBB session, if it is valid, the same BBB token will be returned by bootstrap
# thus if the user was authenticated, it will use the user token
# - Always bootstrap, otherwise the browser may end up logged in with different
# users in BBB and core
# - Ignore any previous bootstrapped access token
RequestsManager.init(auth={}, cookies={'session': request.cookies.get('session')})
elif 'auth' not in session:
# No BBB or core session, API will bootstrap
RequestsManager.init(auth={}, cookies={})
else:
# We have a core session and no BBB session, this is the only situation
# API will not bootstrap
RequestsManager.init(auth=session['auth'], cookies={})