def _get_geoip_data_json(self):
"""
Cleans out any dirty unicode characters to make the geoip data safe for
JSON encoding.
"""
clean = {}
if not self.geoip_data: return {}
for key,value in self.geoip_data.items():
clean[key] = utils.u_clean(value)
return clean
def _get_geoip_data_json(self):
"""
Cleans out any dirty unicode characters to make the geoip data safe for
JSON encoding.
"""
clean = {}
if not self.geoip_data: return {}
for key,value in self.geoip_data.items():
clean[key] = utils.u_clean(value)
return clean
def process_request(self, request):
# don't process AJAX requests
if request.is_ajax(): return
# create some useful variables
ip_address = utils.get_ip(request)
user_agent = unicode(request.META.get('HTTP_USER_AGENT', '')[:255], errors='ignore')
# retrieve untracked user agents from cache
ua_key = '_tracking_untracked_uas'
untracked = cache.get(ua_key)
if untracked is None:
log.info('Updating untracked user agent cache')
untracked = UntrackedUserAgent.objects.all()
cache.set(ua_key, untracked, 3600)
# see if the user agent is not supposed to be tracked
for ua in untracked:
# if the keyword is found in the user agent, stop tracking
if user_agent.find(ua.keyword) != -1:
log.debug('Not tracking UA "%s" because of keyword: %s' % (user_agent, ua.keyword))
return
if hasattr(request, 'session') and request.session.session_key:
# use the current session key if we can
session_key = request.session.session_key
else:
# otherwise just fake a session key
session_key = '%s:%s' % (ip_address, user_agent)
session_key = session_key[:40]
# ensure that the request.path does not begin with any of the prefixes
for prefix in self.prefixes:
if request.path.startswith(prefix):
log.debug('Not tracking request to: %s' % request.path)
return
# if we get here, the URL needs to be tracked
# determine what time it is
now = datetime.now()
attrs = {
'session_key': session_key,
'ip_address': ip_address
}
# for some reason, Visitor.objects.get_or_create was not working here
try:
visitor = Visitor.objects.get(**attrs)
except Visitor.DoesNotExist:
# see if there's a visitor with the same IP and user agent
# within the last 5 minutes
cutoff = now - timedelta(minutes=5)
visitors = Visitor.objects.filter(
ip_address=ip_address,
user_agent=user_agent,
last_update__gte=cutoff
)
if len(visitors):
visitor = visitors[0]
visitor.session_key = session_key
log.debug('Using existing visitor for IP %s / UA %s: %s' % (ip_address, user_agent, visitor.id))
else:
# it's probably safe to assume that the visitor is brand new
visitor = Visitor(**attrs)
log.debug('Created a new visitor: %s' % attrs)
except:
return
# determine whether or not the user is logged in
user = request.user
if isinstance(user, AnonymousUser):
user = None
# update the tracking information
visitor.user = user
visitor.user_agent = user_agent
# if the visitor record is new, or the visitor hasn't been here for
# at least an hour, update their referrer URL
one_hour_ago = now - timedelta(hours=1)
if not visitor.last_update or visitor.last_update <= one_hour_ago:
visitor.referrer = utils.u_clean(request.META.get('HTTP_REFERER', 'unknown')[:255])
# reset the number of pages they've been to
visitor.page_views = 0
visitor.session_start = now
visitor.url = request.path
visitor.page_views += 1
visitor.last_update = now
try:
visitor.save()
except DatabaseError:
log.error('There was a problem saving visitor information:\n%s\n\n%s' % (traceback.format_exc(), locals()))
def process_request(self, request):
# don't process AJAX requests
if request.path.startswith("/s/") or request.path.startswith("/static/") or request.path.startswith("/admin/")\
or request.path.startswith("/favicon.ico") or (request.is_ajax() and not request.path.startswith('/o/')):
return
# create some useful variables
ip_address = utils.get_ip(request)
user_agent = unicode(request.META.get('HTTP_USER_AGENT', '')[:255],
errors='ignore')
# retrieve untracked user agents from cache
ua_key = '_tracking_untracked_uas'
untracked = cache.get(ua_key)
if untracked is None:
log.info('Updating untracked user agent cache')
untracked = UntrackedUserAgent.objects.all()
cache.set(ua_key, untracked, 3600)
# see if the user agent is not supposed to be tracked
for ua in untracked:
# if the keyword is found in the user agent, stop tracking
if user_agent.find(ua.keyword) != -1:
log.debug('Not tracking UA "%s" because of keyword: %s' %
(user_agent, ua.keyword))
return
if hasattr(request, 'session') and request.session.session_key:
# use the current session key if we can
session_key = request.session.session_key
else:
# otherwise just fake a session key
session_key = '%s:%s' % (ip_address, user_agent)
session_key = session_key[:40]
# ensure that the request.path does not begin with any of the prefixes
for prefix in self.prefixes:
if request.path.startswith(prefix):
log.debug('Not tracking request to: %s' % request.path)
return
# if we get here, the URL needs to be tracked
# determine what time it is
now = datetime.now()
attrs = {'session_key': session_key, 'ip_address': ip_address}
visitor_id = request.session.get('visitor_id', None)
if not visitor_id:
# for some reason, Visitor.objects.get_or_create was not working here
try:
visitor = Visitor.objects.only('id').get(**attrs)
except Visitor.DoesNotExist:
request.session.set_test_cookie()
# see if there's a visitor with the same IP and user agent
# within the last 5 minutes
cutoff = now - timedelta(minutes=5)
visitors = Visitor.objects.only('id').filter(
ip_address=ip_address,
user_agent=user_agent,
last_update__gte=cutoff)
if len(visitors):
visitor = visitors[0]
visitor.session_key = session_key
log.debug('Using existing visitor for IP %s / UA %s: %s' %
(ip_address, user_agent, visitor.id))
else:
# it's probably safe to assume that the visitor is brand new
visitor = Visitor(**attrs)
log.debug('Created a new visitor: %s' % attrs)
try:
visitor.save()
except DatabaseError:
print_stack_trace()
log.error(
'There was a problem saving visitor information:\n%s\n\n%s'
% (traceback.format_exc(), locals()))
except:
return
request.session['visitor_id'] = visitor_id = visitor.id
redis_data = redis.get('visitor_data_%s' % visitor_id) or '{}'
visitor_data = json.loads(redis_data)
visitor_data['visitor_id'] = visitor_id
# update the tracking information
visitor_data['user_agent'] = user_agent
# if the visitor record is new, or the visitor hasn't been here for
# at least an hour, update their referrer URL
one_hour_ago = pytz.UTC.localize(now - timedelta(hours=1))
# TODO: ensure that we are on the same time zone - I just put UTC for now
# to get it working
last_update = visitor_data.get('last_update', None)
if not last_update or last_update <= time.mktime(
one_hour_ago.timetuple()):
visitor_data['referrer'] = utils.u_clean(
request.META.get('HTTP_REFERER', 'unknown')[:255])
# reset the number of pages they've been to
visitor_data['page_views'] = 0
visitor_data['session_start'] = time.mktime(now.timetuple())
visitor_data['url'] = request.path
page_views = visitor_data.get('page_views', 0) + 1
visitor_data['page_views'] = page_views
visitor_data['last_update'] = time.mktime(now.timetuple())
try:
# Extracting visitor data from GA cookie
cookie = request.COOKIES.get('__utmz')
if cookie:
try:
data = cookie.split('.', 4)[-1]
data = dict(match.groups() for match in re.finditer(
r'(utm(?:csr|ccn|cmd|ctr))=([^\|]*)', data))
except (ValueError, IndexError):
log.error('Malformed GA cookie: {0!r}'.format(cookie))
else:
visitor_data['source'] = normalize_ga_value(
data.get('utmcsr'))
visitor_data['medium'] = normalize_ga_value(
data.get('utmcmd'))
visitor_data['campaign'] = normalize_ga_value(
data.get('utmccn'))
visitor_data['keywords'] = normalize_ga_value(
data.get('utm.ctr'))
utm_source = request.GET.get("utm_source", "unknown")
request.session['acquisition_source_name'] = utm_source
if utm_source != "unknown":
# utm_source: Identify the advertiser, site, publication, etc. that is sending traffic to your property, e.g. google, citysearch, newsletter4, billboard.
# utm_medium: The advertising or marketing medium, e.g.: cpc, banner, email newsletter.
# utm_campaign: The individual campaign name, slogan, promo code, etc. for a product.
# utm_term: Identify paid search keywords. If you're manually tagging paid keyword campaigns, you should also use utm_term to specify the keyword.
# utm_content: Used to differentiate similar content, or links within the same ad. For example, if you have two call-to-action links within the same email message, you can use utm_content and set different values for each so you can tell which version is more effective.
#update the tracking info with the latest and bump the old one to be stored in the history
#visitor.bump_past_acquisition_info()
past_acquisition_info = visitor_data.get(
'past_acquisition_info', [])
if visitor_data.get('acquisition_source', None):
old_visitor_data = {'date_valid_until': time.time()}
for k in VISITOR_PARAMS_MAPPING.keys():
old_visitor_data[k] = visitor_data.get(k, None)
past_acquisition_info.append(old_visitor_data)
visitor_data[
'past_acquisition_info'] = past_acquisition_info
for k, v in VISITOR_PARAMS_MAPPING.items():
value = request.GET.get(v, 'unknown')[:255]
visitor_data[k] = value
except:
print_stack_trace()
redis.set('visitor_data_%s' % visitor_id, json.dumps(visitor_data))
def process_request(self, request):
# create some useful variables
ip_address = utils.get_ip(request)
user_agent = unicode(request.META.get('HTTP_USER_AGENT', '')[:255], errors='ignore')
# retrieve untracked user agents from cache
ua_key = '_tracking_untracked_uas'
untracked = cache.get(ua_key)
if untracked is None:
log.info('Updating untracked user agent cache')
untracked = UntrackedUserAgent.objects.all()
cache.set(ua_key, untracked, 3600)
# see if the user agent is not supposed to be tracked
for ua in untracked:
# if the keyword is found in the user agent, stop tracking
if user_agent.find(ua.keyword) != -1:
log.debug('Not tracking UA "%s" because of keyword: %s' % (user_agent, ua.keyword))
return
if not request.session.session_key:
request.session.save()
session_key = request.session.session_key
# ensure that the request.path does not begin with any of the prefixes
for prefix in self.prefixes:
if request.path.startswith(prefix):
log.debug('Not tracking request to: %s' % request.path)
return
# if we get here, the URL needs to be tracked
# determine what time it is
# For now we have made a fork that works with MySQL backed Django that
# also has the setting USE_TZ set to False (we are no longer calling `localtime()`)
now=timezone.now()
attrs = {
'session_key': session_key,
'ip_address': ip_address
}
# for some reason, Visitor.objects.get_or_create was not working here
try:
visitor = Visitor.objects.get(**attrs)
except Visitor.DoesNotExist:
# see if there's a visitor with the same IP and user agent
# within the last 5 minutes
cutoff = now - timedelta(minutes=5)
visitors = Visitor.objects.filter(
ip_address=ip_address,
user_agent=user_agent,
last_update__gte=cutoff
)
if len(visitors):
visitor = visitors[0]
visitor.session_key = session_key
log.debug('Using existing visitor for IP %s / UA %s: %s' % (ip_address, user_agent, visitor.id))
else:
visitor, created = Visitor.objects.get_or_create(**attrs)
if created:
log.debug('Created a new visitor: %s' % attrs)
except:
return
# determine whether or not the user is logged in
user = request.user
if isinstance(user, AnonymousUser):
user = None
# update the tracking information
visitor.user = user
visitor.user_agent = user_agent
# if the visitor record is new, or the visitor hasn't been here for
# at least an hour, update their referrer URL
one_hour_ago = now - timedelta(hours=1)
if not visitor.last_update or visitor.last_update <= one_hour_ago:
visitor.referrer = utils.u_clean(request.META.get('HTTP_REFERER', 'unknown')[:255])
# reset the number of pages they've been to
visitor.page_views = 0
visitor.session_start = now
visitor.url = request.path
visitor.page_views += 1
visitor.last_update = now
try:
sid = transaction.savepoint()
visitor.save()
transaction.savepoint_commit(sid)
except IntegrityError:
transaction.savepoint_rollback(sid)
except DatabaseError:
log.error('There was a problem saving visitor information:\n%s\n\n%s' % (traceback.format_exc(), locals()))
def process_request(self, request):
# don't process AJAX requests
if request.is_ajax(): return
# create some useful variables
ip_address = utils.get_ip(request)
user_agent = unicode(request.META.get('HTTP_USER_AGENT', '')[:255], errors='ignore')
# retrieve untracked user agents from cache
ua_key = '_tracking_untracked_uas'
untracked = cache.get(ua_key)
if untracked is None:
log.info('Updating untracked user agent cache')
untracked = UntrackedUserAgent.objects.all()
cache.set(ua_key, untracked, 3600)
# see if the user agent is not supposed to be tracked
for ua in untracked:
# if the keyword is found in the user agent, stop tracking
if user_agent.find(ua.keyword) != -1:
log.debug('Not tracking UA "%s" because of keyword: %s' % (user_agent, ua.keyword))
return
if hasattr(request, 'session') and request.session.session_key:
# use the current session key if we can
session_key = request.session.session_key
else:
# otherwise just fake a session key
session_key = '%s:%s' % (ip_address, user_agent)
session_key = session_key[:40]
# ensure that the request.path does not begin with any of the prefixes
for prefix in self.prefixes:
if request.path.startswith(prefix):
log.debug('Not tracking request to: %s' % request.path)
return
# if we get here, the URL needs to be tracked
# determine what time it is
now = right_now()
attrs = {
'session_key': session_key,
'ip_address': ip_address
}
# for some reason, Visitor.objects.get_or_create was not working here
try:
visitor = Visitor.objects.get(**attrs)
except Visitor.DoesNotExist:
# see if there's a visitor with the same IP and user agent
# within the last 5 minutes
cutoff = now - timedelta(minutes=5)
visitors = Visitor.objects.filter(
ip_address=ip_address,
user_agent=user_agent,
last_update__gte=cutoff
)
if len(visitors):
visitor = visitors[0]
visitor.session_key = session_key
log.debug('Using existing visitor for IP %s / UA %s: %s' % (ip_address, user_agent, visitor.id))
else:
# it's probably safe to assume that the visitor is brand new
visitor = Visitor(**attrs)
log.debug('Created a new visitor: %s' % attrs)
except:
return
# determine whether or not the user is logged in
user = request.user
if isinstance(user, AnonymousUser):
user = None
# update the tracking information
visitor.user = user
visitor.user_agent = user_agent
# if the visitor record is new, or the visitor hasn't been here for
# at least an hour, update their referrer URL
one_hour_ago = now - timedelta(hours=1)
if not visitor.last_update or visitor.last_update <= one_hour_ago:
visitor.referrer = utils.u_clean(request.META.get('HTTP_REFERER', 'unknown')[:255])
# reset the number of pages they've been to
visitor.page_views = 0
visitor.session_start = now
visitor.url = request.path
visitor.page_views += 1
visitor.last_update = now
try:
visitor.save()
except DatabaseError:
log.error('There was a problem saving visitor information:\n%s\n\n%s' % (traceback.format_exc(), locals()))
def process_request(self, request):
# don't process AJAX requests
if request.is_ajax(): return
# create some useful variables
ip_address = utils.get_ip(request)
user_agent = request.META.get('HTTP_USER_AGENT', '')[:255]
# retrieve untracked user agents from cache
ua_key = '_tracking_untracked_uas'
untracked = cache.get(ua_key)
if untracked is None:
log.info('Updating untracked user agent cache')
untracked = list(UntrackedUserAgent.objects.all())
cache.set(ua_key, untracked, 3600)
# see if the user agent is not supposed to be tracked
for ua in untracked:
# if the keyword is found in the user agent, stop tracking
if unicode(user_agent, errors='ignore').find(ua.keyword) != -1:
log.debug('Not tracking UA "%s" because of keyword: %s' % (user_agent, ua.keyword))
return
if hasattr(request, 'session'):
if not request.session.session_key:
request.session.save()
# use the current session key if we can
session_key = request.session.session_key
else:
# otherwise just fake a session key
session_key = '%s:%s' % (ip_address, user_agent)
# ensure that the request.path does not begin with any of the prefixes
for prefix in self.prefixes:
if request.path.startswith(prefix):
log.debug('Not tracking request to: %s' % request.path)
return
# if we get here, the URL needs to be tracked
# determine what time it is
now = datetime.now()
# Attributes we use when creating a new user
new_attrs = {
'session_key': session_key,
'ip_address': ip_address
}
# If we have a visitor_id cookie, use it
visitor_id = request.COOKIES.get('visitor_id')
if visitor_id:
attrs = {'id': visitor_id}
else:
attrs = new_attrs
# for some reason, Visitor.objects.get_or_create was not working here
try:
visitor = Visitor.objects.get(**attrs)
except Visitor.DoesNotExist:
# add tracking ID to model if specified in the URL
tid = request.GET.get('tid') or request.GET.get('fb_source')
if tid:
get = request.GET.copy()
attrs['tid'] = tid
request.GET = get
visitor = Visitor(**new_attrs)
log.debug('Created a new visitor: %s' % new_attrs)
except:
return
# determine whether or not the user is logged in
user = request.user
if isinstance(user, AnonymousUser):
user = None
# update the tracking information
visitor.user = user
visitor.user_agent = user_agent
# if the visitor record is new, update their referrer URL
if not visitor.last_update:
visitor.referrer = utils.u_clean(request.META.get('HTTP_REFERER', 'unknown')[:255])
# reset the number of pages they've been to
visitor.page_views = 0
visitor.session_start = now
visitor.url = request.path
visitor.page_views += 1
visitor.last_update = now
try:
visitor.save()
except DatabaseError:
log.error('There was a problem saving visitor information:\n%s\n\n%s' % (traceback.format_exc(), locals()))
request.visitor = visitor
request.session['visitor_id'] = visitor.pk
def process_request(self, request):
# don't process AJAX requests
if request.is_ajax(): return
# create some useful variables
ip_address = utils.get_ip(request)
user_agent = request.META.get('HTTP_USER_AGENT', '')
if utils.user_agent_is_untracked( user_agent ):
log.debug('Not tracking UA "%s" because of keyword: %s' % (user_agent, ua.keyword))
return
if hasattr(request, 'session'):
# use the current session key if we can
session_key = request.session.session_key
else:
# otherwise just fake a session key
session_key = md5( '%s:%s' % ( ip_address, user_agent ) ).hexdigest()
# ensure that the request.path does not begin with any of the prefixes
for prefix in NO_TRACKING_PREFIXES:
if request.path.startswith(prefix):
log.debug('Not tracking request to: %s' % request.path)
return
# if we get here, the URL needs to be tracked
# determine what time it is
now = datetime.now()
attrs = {
'session_key': session_key,
'ip_address': ip_address
}
# for some reason, Visitor.objects.get_or_create was not working here
try:
visitor = Visitor.objects.get(**attrs)
except Visitor.DoesNotExist:
# see if there's a visitor with the same IP and user agent
# within the last 5 minutes
cutoff = now - timedelta(minutes=5)
visitors = Visitor.objects.filter(
ip_address=ip_address,
user_agent=user_agent[:255],
last_update__gte=cutoff
)
if len(visitors):
visitor = visitors[0]
visitor.session_key = session_key
log.debug('Using existing visitor for IP %s / UA %s: %s' % (ip_address, user_agent, visitor.id))
else:
# it's probably safe to assume that the visitor is brand new
visitor = Visitor(**attrs)
log.debug('Created a new visitor: %s' % attrs)
except:
return
# determine whether or not the user is logged in
user = request.user
if isinstance(user, AnonymousUser):
user = None
# update the tracking information
visitor.user = user
visitor.user_agent = user_agent[:255]
# if the visitor record is new, or the visitor hasn't been here for
# at least an hour, update their referrer URL
if not visitor.last_update or visitor.last_update <= ( now - timedelta( hours = 1 ) ) :
visitor.referrer = utils.u_clean(request.META.get('HTTP_REFERER', 'unknown')[:255])
# reset the number of pages they've been to
visitor.page_views = 0
visitor.session_start = now
visitor.url = request.path[:255]
visitor.page_views += 1
visitor.last_update = now
visitor.save()
def log_click_track(request, coupon=None):
try:
referer = utils.u_clean(request.META.get('HTTP_REFERER', 'unknown')[:255])
clicked_link = request.path
source_url_type='landing'
if re.search('/coupons/[a-z0-9-]+/[a-z0-9-]+/[\d]+/', referer):
source_url_type = 'coupon'
elif re.search('/coupons/[a-z0-9-]+/', referer):
source_url_type = 'company'
elif re.search('/categories/[A-z0-9-]+/', referer):
source_url_type = 'category'
merchant=None
if "/o/" in clicked_link:
source_url = clicked_link
merchant = coupon.merchant
target_url = coupon.get_retailer_link()
else:
source_url = referer
target_url = clicked_link
merchant = None
if 'go.redirectingat.com' in target_url:
target_url = _remove_skimlinks(target_url)
merchant_domain = shorten_to_domain(target_url)
visitor = Visitor.objects.get(pk=request.session['visitor_id'])
click_track = ClickTrack(visitor=visitor, user_agent=visitor.user_agent[:255], referer=referer[:255],
target_url=target_url[:255], source_url_type=source_url_type[:255],
source_url=source_url[:255], merchant=merchant, coupon=coupon,
merchant_domain=merchant_domain[:255])
try:
click_track.acquisition_source = visitor.acquisition_source
click_track.acquisition_medium = visitor.acquisition_medium
click_track.acquisition_term = visitor.acquisition_term
click_track.acquisition_content = visitor.acquisition_content
click_track.acquisition_campaign = visitor.acquisition_campaign
click_track.acquisition_gclid = visitor.acquisition_gclid
except:
print_stack_trace()
try:
click_track.save()
except:
try:
print "Visitor ID", click_track.visitor
print "User Agent", click_track.user_agent
print "Referer", click_track.referer
print "target_url", click_track.target_url
print "source_url_type", click_track.source_url_type
print "merchant", click_track.merchant
print "coupon", click_track.coupon
print "merchant_domain", click_track.merchant_domain
print merchant.name, merchant.id
except:
pass
print_stack_trace()
except:
print_stack_trace()
def click_track(request, clicked_link_path=None):
try:
referer = utils.u_clean(request.META.get('HTTP_REFERER', 'unknown')[:255])
clicked_link = request.POST["clicked"][:255]
try:
clicked_link=clicked_link.lower()
except:
print_stack_trace()
source_url_type='landing'
if re.search('/coupons/[a-z0-9-]+/[a-z0-9-]+/[\d]+/', referer):
source_url_type = 'coupon'
elif re.search('/coupons/[a-z0-9-]+/', referer):
source_url_type = 'company'
elif re.search('/categories/[A-z0-9-]+/', referer):
source_url_type = 'category'
coupon=None
merchant=None
if "/coupon/" in clicked_link:
#skimlinks will assume the source url to be the /coupon/ url
if clicked_link.endswith("/"):
coupon_id = clicked_link.split("/")[-2] #assumes trailing '/'
else:
coupon_id = clicked_link.split("/")[-1]
source_url = clicked_link
coupon = Coupon.active_objects.get(id=int(coupon_id))
try:
merchant = Merchant.objects.get(id=coupon.merchant.id)
except:
merchant = None
target_url = coupon.get_retailer_link()
else:
source_url = referer
target_url = clicked_link
merchant = None
if 'go.redirectingat.com' in target_url:
target_url = _remove_skimlinks(target_url)
merchant_domain = shorten_to_domain(target_url)
visitor = Visitor.objects.get(pk=request.session['visitor_id'])
click_track = ClickTrack()
click_track.visitor = visitor
click_track.user_agent = visitor.user_agent[:255]
click_track.referer = referer[:255]
click_track.target_url = target_url[:255]
click_track.source_url_type = source_url_type[:255]
click_track.source_url = source_url[:255]
click_track.merchant = merchant
click_track.coupon = coupon
click_track.merchant_domain = merchant_domain[:255]
try:
click_track.acquisition_source = visitor.acquisition_source
click_track.acquisition_medium = visitor.acquisition_medium
click_track.acquisition_term = visitor.acquisition_term
click_track.acquisition_content = visitor.acquisition_content
click_track.acquisition_campaign = visitor.acquisition_campaign
click_track.acquisition_gclid = visitor.acquisition_gclid
except:
print_stack_trace()
try:
click_track.save()
except:
try:
print "Visitor ID", click_track.visitor
print "User Agent", click_track.user_agent
print "Referer", click_track.referer
print "target_url", click_track.target_url
print "source_url_type", click_track.source_url_type
print "merchant", click_track.merchant
print "coupon", click_track.coupon
print "merchant_domain", click_track.merchant_domain
print merchant.name, merchant.id
except:
pass
print_stack_trace()
except:
print_stack_trace()
return success()
def process_request(self, request):
# don't process AJAX requests
if request.path.startswith("/s/") or request.path.startswith("/static/") or request.path.startswith("/admin/")\
or request.path.startswith("/favicon.ico") or (request.is_ajax() and not request.path.startswith('/o/')):
return
# create some useful variables
ip_address = utils.get_ip(request)
user_agent = unicode(request.META.get('HTTP_USER_AGENT', '')[:255], errors='ignore')
# retrieve untracked user agents from cache
ua_key = '_tracking_untracked_uas'
untracked = cache.get(ua_key)
if untracked is None:
log.info('Updating untracked user agent cache')
untracked = UntrackedUserAgent.objects.all()
cache.set(ua_key, untracked, 3600)
# see if the user agent is not supposed to be tracked
for ua in untracked:
# if the keyword is found in the user agent, stop tracking
if user_agent.find(ua.keyword) != -1:
log.debug('Not tracking UA "%s" because of keyword: %s' % (user_agent, ua.keyword))
return
if hasattr(request, 'session') and request.session.session_key:
# use the current session key if we can
session_key = request.session.session_key
else:
# otherwise just fake a session key
session_key = '%s:%s' % (ip_address, user_agent)
session_key = session_key[:40]
# ensure that the request.path does not begin with any of the prefixes
for prefix in self.prefixes:
if request.path.startswith(prefix):
log.debug('Not tracking request to: %s' % request.path)
return
# if we get here, the URL needs to be tracked
# determine what time it is
now = datetime.now()
attrs = {
'session_key': session_key,
'ip_address': ip_address
}
visitor_id = request.session.get('visitor_id', None)
if not visitor_id:
# for some reason, Visitor.objects.get_or_create was not working here
try:
visitor = Visitor.objects.only('id').get(**attrs)
except Visitor.DoesNotExist:
request.session.set_test_cookie()
# see if there's a visitor with the same IP and user agent
# within the last 5 minutes
cutoff = now - timedelta(minutes=5)
visitors = Visitor.objects.only('id').filter(
ip_address=ip_address,
user_agent=user_agent,
last_update__gte=cutoff
)
if len(visitors):
visitor = visitors[0]
visitor.session_key = session_key
log.debug('Using existing visitor for IP %s / UA %s: %s' % (ip_address, user_agent, visitor.id))
else:
# it's probably safe to assume that the visitor is brand new
visitor = Visitor(**attrs)
log.debug('Created a new visitor: %s' % attrs)
try:
visitor.save()
except DatabaseError:
print_stack_trace()
log.error('There was a problem saving visitor information:\n%s\n\n%s' % (traceback.format_exc(), locals()))
except:
return
request.session['visitor_id'] = visitor_id = visitor.id
redis_data = redis.get('visitor_data_%s' % visitor_id) or '{}'
visitor_data = json.loads(redis_data)
visitor_data['visitor_id'] = visitor_id
# update the tracking information
visitor_data['user_agent'] = user_agent
# if the visitor record is new, or the visitor hasn't been here for
# at least an hour, update their referrer URL
one_hour_ago = pytz.UTC.localize(now - timedelta(hours=1))
# TODO: ensure that we are on the same time zone - I just put UTC for now
# to get it working
last_update = visitor_data.get('last_update', None)
if not last_update or last_update <= time.mktime(one_hour_ago.timetuple()):
visitor_data['referrer'] = utils.u_clean(request.META.get('HTTP_REFERER', 'unknown')[:255])
# reset the number of pages they've been to
visitor_data['page_views'] = 0
visitor_data['session_start'] = time.mktime(now.timetuple())
visitor_data['url'] = request.path
page_views = visitor_data.get('page_views', 0) + 1
visitor_data['page_views'] = page_views
visitor_data['last_update'] = time.mktime(now.timetuple())
try:
# Extracting visitor data from GA cookie
cookie = request.COOKIES.get('__utmz')
if cookie:
try:
data = cookie.split('.', 4)[-1]
data = dict(match.groups() for match in re.finditer(
r'(utm(?:csr|ccn|cmd|ctr))=([^\|]*)', data))
except (ValueError, IndexError):
log.error('Malformed GA cookie: {0!r}'.format(cookie))
else:
visitor_data['source'] = normalize_ga_value(data.get('utmcsr'))
visitor_data['medium'] = normalize_ga_value(data.get('utmcmd'))
visitor_data['campaign'] = normalize_ga_value(data.get('utmccn'))
visitor_data['keywords'] = normalize_ga_value(data.get('utm.ctr'))
utm_source = request.GET.get("utm_source", "unknown")
request.session['acquisition_source_name'] = utm_source
if utm_source != "unknown":
# utm_source: Identify the advertiser, site, publication, etc. that is sending traffic to your property, e.g. google, citysearch, newsletter4, billboard.
# utm_medium: The advertising or marketing medium, e.g.: cpc, banner, email newsletter.
# utm_campaign: The individual campaign name, slogan, promo code, etc. for a product.
# utm_term: Identify paid search keywords. If you're manually tagging paid keyword campaigns, you should also use utm_term to specify the keyword.
# utm_content: Used to differentiate similar content, or links within the same ad. For example, if you have two call-to-action links within the same email message, you can use utm_content and set different values for each so you can tell which version is more effective.
#update the tracking info with the latest and bump the old one to be stored in the history
#visitor.bump_past_acquisition_info()
past_acquisition_info = visitor_data.get('past_acquisition_info', [])
if visitor_data.get('acquisition_source', None):
old_visitor_data = {'date_valid_until': time.time()}
for k in VISITOR_PARAMS_MAPPING.keys():
old_visitor_data[k] = visitor_data.get(k, None)
past_acquisition_info.append(old_visitor_data)
visitor_data['past_acquisition_info'] = past_acquisition_info
for k,v in VISITOR_PARAMS_MAPPING.items():
value = request.GET.get(v, 'unknown')[:255]
visitor_data[k] = value
except:
print_stack_trace()
redis.set('visitor_data_%s' % visitor_id, json.dumps(visitor_data))
def log_click_track(request, coupon=None):
try:
referer = utils.u_clean(
request.META.get('HTTP_REFERER', 'unknown')[:255])
clicked_link = request.path
source_url_type = 'landing'
if re.search('/coupons/[a-z0-9-]+/[a-z0-9-]+/[\d]+/', referer):
source_url_type = 'coupon'
elif re.search('/coupons/[a-z0-9-]+/', referer):
source_url_type = 'company'
elif re.search('/categories/[A-z0-9-]+/', referer):
source_url_type = 'category'
merchant = None
if "/o/" in clicked_link:
source_url = clicked_link
merchant = coupon.merchant
target_url = coupon.get_retailer_link()
else:
source_url = referer
target_url = clicked_link
merchant = None
if 'go.redirectingat.com' in target_url:
target_url = _remove_skimlinks(target_url)
merchant_domain = shorten_to_domain(target_url)
visitor = Visitor.objects.get(pk=request.session['visitor_id'])
click_track = ClickTrack(visitor=visitor,
user_agent=visitor.user_agent[:255],
referer=referer[:255],
target_url=target_url[:255],
source_url_type=source_url_type[:255],
source_url=source_url[:255],
merchant=merchant,
coupon=coupon,
merchant_domain=merchant_domain[:255])
try:
click_track.acquisition_source = visitor.acquisition_source
click_track.acquisition_medium = visitor.acquisition_medium
click_track.acquisition_term = visitor.acquisition_term
click_track.acquisition_content = visitor.acquisition_content
click_track.acquisition_campaign = visitor.acquisition_campaign
click_track.acquisition_gclid = visitor.acquisition_gclid
except:
print_stack_trace()
try:
click_track.save()
except:
try:
print "Visitor ID", click_track.visitor
print "User Agent", click_track.user_agent
print "Referer", click_track.referer
print "target_url", click_track.target_url
print "source_url_type", click_track.source_url_type
print "merchant", click_track.merchant
print "coupon", click_track.coupon
print "merchant_domain", click_track.merchant_domain
print merchant.name, merchant.id
except:
pass
print_stack_trace()
except:
print_stack_trace()
def click_track(request, clicked_link_path=None):
try:
referer = utils.u_clean(
request.META.get('HTTP_REFERER', 'unknown')[:255])
clicked_link = request.POST["clicked"][:255]
try:
clicked_link = clicked_link.lower()
except:
print_stack_trace()
source_url_type = 'landing'
if re.search('/coupons/[a-z0-9-]+/[a-z0-9-]+/[\d]+/', referer):
source_url_type = 'coupon'
elif re.search('/coupons/[a-z0-9-]+/', referer):
source_url_type = 'company'
elif re.search('/categories/[A-z0-9-]+/', referer):
source_url_type = 'category'
coupon = None
merchant = None
if "/coupon/" in clicked_link:
#skimlinks will assume the source url to be the /coupon/ url
if clicked_link.endswith("/"):
coupon_id = clicked_link.split("/")[-2] #assumes trailing '/'
else:
coupon_id = clicked_link.split("/")[-1]
source_url = clicked_link
coupon = Coupon.active_objects.get(id=int(coupon_id))
try:
merchant = Merchant.objects.get(id=coupon.merchant.id)
except:
merchant = None
target_url = coupon.get_retailer_link()
else:
source_url = referer
target_url = clicked_link
merchant = None
if 'go.redirectingat.com' in target_url:
target_url = _remove_skimlinks(target_url)
merchant_domain = shorten_to_domain(target_url)
visitor = Visitor.objects.get(pk=request.session['visitor_id'])
click_track = ClickTrack()
click_track.visitor = visitor
click_track.user_agent = visitor.user_agent[:255]
click_track.referer = referer[:255]
click_track.target_url = target_url[:255]
click_track.source_url_type = source_url_type[:255]
click_track.source_url = source_url[:255]
click_track.merchant = merchant
click_track.coupon = coupon
click_track.merchant_domain = merchant_domain[:255]
try:
click_track.acquisition_source = visitor.acquisition_source
click_track.acquisition_medium = visitor.acquisition_medium
click_track.acquisition_term = visitor.acquisition_term
click_track.acquisition_content = visitor.acquisition_content
click_track.acquisition_campaign = visitor.acquisition_campaign
click_track.acquisition_gclid = visitor.acquisition_gclid
except:
print_stack_trace()
try:
click_track.save()
except:
try:
print "Visitor ID", click_track.visitor
print "User Agent", click_track.user_agent
print "Referer", click_track.referer
print "target_url", click_track.target_url
print "source_url_type", click_track.source_url_type
print "merchant", click_track.merchant
print "coupon", click_track.coupon
print "merchant_domain", click_track.merchant_domain
print merchant.name, merchant.id
except:
pass
print_stack_trace()
except:
print_stack_trace()
return success()
