[1280] Change use of group.get_groups() to group.get_parent_groups()
[1280] Change use of group.get_groups() to group.get_parent_groups()

import os import os
import logging import logging
import datetime import datetime
import httplib import httplib
import collections import collections
import requests import requests
import json import json
from pylons import config from pylons import config
from ga_model import _normalize_url from ga_model import _normalize_url
import ga_model import ga_model
   
#from ga_client import GA #from ga_client import GA
   
log = logging.getLogger('ckanext.ga-report') log = logging.getLogger('ckanext.ga-report')
   
FORMAT_MONTH = '%Y-%m' FORMAT_MONTH = '%Y-%m'
MIN_VIEWS = 50 MIN_VIEWS = 50
MIN_VISITS = 20 MIN_VISITS = 20
MIN_DOWNLOADS = 10 MIN_DOWNLOADS = 10
   
class DownloadAnalytics(object): class DownloadAnalytics(object):
'''Downloads and stores analytics info''' '''Downloads and stores analytics info'''
   
def __init__(self, service=None, token=None, profile_id=None, delete_first=False, def __init__(self, service=None, token=None, profile_id=None, delete_first=False,
skip_url_stats=False): skip_url_stats=False):
self.period = config['ga-report.period'] self.period = config['ga-report.period']
self.service = service self.service = service
self.profile_id = profile_id self.profile_id = profile_id
self.delete_first = delete_first self.delete_first = delete_first
self.skip_url_stats = skip_url_stats self.skip_url_stats = skip_url_stats
self.token = token self.token = token
   
def specific_month(self, date): def specific_month(self, date):
import calendar import calendar
   
first_of_this_month = datetime.datetime(date.year, date.month, 1) first_of_this_month = datetime.datetime(date.year, date.month, 1)
_, last_day_of_month = calendar.monthrange(int(date.year), int(date.month)) _, last_day_of_month = calendar.monthrange(int(date.year), int(date.month))
last_of_this_month = datetime.datetime(date.year, date.month, last_day_of_month) last_of_this_month = datetime.datetime(date.year, date.month, last_day_of_month)
# if this is the latest month, note that it is only up until today # if this is the latest month, note that it is only up until today
now = datetime.datetime.now() now = datetime.datetime.now()
if now.year == date.year and now.month == date.month: if now.year == date.year and now.month == date.month:
last_day_of_month = now.day last_day_of_month = now.day
last_of_this_month = now last_of_this_month = now
periods = ((date.strftime(FORMAT_MONTH), periods = ((date.strftime(FORMAT_MONTH),
last_day_of_month, last_day_of_month,
first_of_this_month, last_of_this_month),) first_of_this_month, last_of_this_month),)
self.download_and_store(periods) self.download_and_store(periods)
   
   
def latest(self): def latest(self):
if self.period == 'monthly': if self.period == 'monthly':
# from first of this month to today # from first of this month to today
now = datetime.datetime.now() now = datetime.datetime.now()
first_of_this_month = datetime.datetime(now.year, now.month, 1) first_of_this_month = datetime.datetime(now.year, now.month, 1)
periods = ((now.strftime(FORMAT_MONTH), periods = ((now.strftime(FORMAT_MONTH),
now.day, now.day,
first_of_this_month, now),) first_of_this_month, now),)
else: else:
raise NotImplementedError raise NotImplementedError
self.download_and_store(periods) self.download_and_store(periods)
   
   
def for_date(self, for_date): def for_date(self, for_date):
assert isinstance(since_date, datetime.datetime) assert isinstance(since_date, datetime.datetime)
periods = [] # (period_name, period_complete_day, start_date, end_date) periods = [] # (period_name, period_complete_day, start_date, end_date)
if self.period == 'monthly': if self.period == 'monthly':
first_of_the_months_until_now = [] first_of_the_months_until_now = []
year = for_date.year year = for_date.year
month = for_date.month month = for_date.month
now = datetime.datetime.now() now = datetime.datetime.now()
first_of_this_month = datetime.datetime(now.year, now.month, 1) first_of_this_month = datetime.datetime(now.year, now.month, 1)
while True: while True:
first_of_the_month = datetime.datetime(year, month, 1) first_of_the_month = datetime.datetime(year, month, 1)
if first_of_the_month == first_of_this_month: if first_of_the_month == first_of_this_month:
periods.append((now.strftime(FORMAT_MONTH), periods.append((now.strftime(FORMAT_MONTH),
now.day, now.day,
first_of_this_month, now)) first_of_this_month, now))
break break
elif first_of_the_month < first_of_this_month: elif first_of_the_month < first_of_this_month:
in_the_next_month = first_of_the_month + datetime.timedelta(40) in_the_next_month = first_of_the_month + datetime.timedelta(40)
last_of_the_month = datetime.datetime(in_the_next_month.year, last_of_the_month = datetime.datetime(in_the_next_month.year,
in_the_next_month.month, 1)\ in_the_next_month.month, 1)\
- datetime.timedelta(1) - datetime.timedelta(1)
periods.append((now.strftime(FORMAT_MONTH), 0, periods.append((now.strftime(FORMAT_MONTH), 0,
first_of_the_month, last_of_the_month)) first_of_the_month, last_of_the_month))
else: else:
# first_of_the_month has got to the future somehow # first_of_the_month has got to the future somehow
break break
month += 1 month += 1
if month > 12: if month > 12:
year += 1 year += 1
month = 1 month = 1
else: else:
raise NotImplementedError raise NotImplementedError
self.download_and_store(periods) self.download_and_store(periods)
   
@staticmethod @staticmethod
def get_full_period_name(period_name, period_complete_day): def get_full_period_name(period_name, period_complete_day):
if period_complete_day: if period_complete_day:
return period_name + ' (up to %ith)' % period_complete_day return period_name + ' (up to %ith)' % period_complete_day
else: else:
return period_name return period_name
   
   
def download_and_store(self, periods): def download_and_store(self, periods):
for period_name, period_complete_day, start_date, end_date in periods: for period_name, period_complete_day, start_date, end_date in periods:
log.info('Period "%s" (%s - %s)', log.info('Period "%s" (%s - %s)',
self.get_full_period_name(period_name, period_complete_day), self.get_full_period_name(period_name, period_complete_day),
start_date.strftime('%Y-%m-%d'), start_date.strftime('%Y-%m-%d'),
end_date.strftime('%Y-%m-%d')) end_date.strftime('%Y-%m-%d'))
   
if self.delete_first: if self.delete_first:
log.info('Deleting existing Analytics for this period "%s"', log.info('Deleting existing Analytics for this period "%s"',
period_name) period_name)
ga_model.delete(period_name) ga_model.delete(period_name)
   
if not self.skip_url_stats: if not self.skip_url_stats:
# Clean out old url data before storing the new # Clean out old url data before storing the new
ga_model.pre_update_url_stats(period_name) ga_model.pre_update_url_stats(period_name)
   
accountName = config.get('googleanalytics.account') accountName = config.get('googleanalytics.account')
   
log.info('Downloading analytics for dataset views') log.info('Downloading analytics for dataset views')
data = self.download(start_date, end_date, '~/%s/dataset/[a-z0-9-_]+' % accountName) data = self.download(start_date, end_date, '~/%s/dataset/[a-z0-9-_]+' % accountName)
   
log.info('Storing dataset views (%i rows)', len(data.get('url'))) log.info('Storing dataset views (%i rows)', len(data.get('url')))
self.store(period_name, period_complete_day, data, ) self.store(period_name, period_complete_day, data, )
   
log.info('Downloading analytics for publisher views') log.info('Downloading analytics for publisher views')
data = self.download(start_date, end_date, '~/%s/publisher/[a-z0-9-_]+' % accountName) data = self.download(start_date, end_date, '~/%s/publisher/[a-z0-9-_]+' % accountName)
   
log.info('Storing publisher views (%i rows)', len(data.get('url'))) log.info('Storing publisher views (%i rows)', len(data.get('url')))
self.store(period_name, period_complete_day, data,) self.store(period_name, period_complete_day, data,)
   
# Make sure the All records are correct. # Make sure the All records are correct.
ga_model.post_update_url_stats() ga_model.post_update_url_stats()
   
log.info('Associating datasets with their publisher') log.info('Associating datasets with their publisher')
ga_model.update_publisher_stats(period_name) # about 30 seconds. ga_model.update_publisher_stats(period_name) # about 30 seconds.
   
   
log.info('Downloading and storing analytics for site-wide stats') log.info('Downloading and storing analytics for site-wide stats')
self.sitewide_stats( period_name, period_complete_day ) self.sitewide_stats( period_name, period_complete_day )
   
log.info('Downloading and storing analytics for social networks') log.info('Downloading and storing analytics for social networks')
self.update_social_info(period_name, start_date, end_date) self.update_social_info(period_name, start_date, end_date)
   
   
def update_social_info(self, period_name, start_date, end_date): def update_social_info(self, period_name, start_date, end_date):
start_date = start_date.strftime('%Y-%m-%d') start_date = start_date.strftime('%Y-%m-%d')
end_date = end_date.strftime('%Y-%m-%d') end_date = end_date.strftime('%Y-%m-%d')
query = 'ga:hasSocialSourceReferral=~Yes$' query = 'ga:hasSocialSourceReferral=~Yes$'
metrics = 'ga:entrances' metrics = 'ga:entrances'
sort = '-ga:entrances' sort = '-ga:entrances'
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = dict(ids='ga:' + self.profile_id, args = dict(ids='ga:' + self.profile_id,
filters=query, filters=query,
metrics=metrics, metrics=metrics,
sort=sort, sort=sort,
dimensions="ga:landingPagePath,ga:socialNetwork", dimensions="ga:landingPagePath,ga:socialNetwork",
max_results=10000) max_results=10000)
   
args['start-date'] = start_date args['start-date'] = start_date
args['end-date'] = end_date args['end-date'] = end_date
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
   
data = collections.defaultdict(list) data = collections.defaultdict(list)
rows = results.get('rows',[]) rows = results.get('rows',[])
for row in rows: for row in rows:
url = _normalize_url('http:/' + row[0]) url = _normalize_url('http:/' + row[0])
data[url].append( (row[1], int(row[2]),) ) data[url].append( (row[1], int(row[2]),) )
ga_model.update_social(period_name, data) ga_model.update_social(period_name, data)
   
   
def download(self, start_date, end_date, path=None): def download(self, start_date, end_date, path=None):
'''Get data from GA for a given time period''' '''Get data from GA for a given time period'''
start_date = start_date.strftime('%Y-%m-%d') start_date = start_date.strftime('%Y-%m-%d')
end_date = end_date.strftime('%Y-%m-%d') end_date = end_date.strftime('%Y-%m-%d')
query = 'ga:pagePath=%s$' % path query = 'ga:pagePath=%s$' % path
metrics = 'ga:pageviews, ga:visits' metrics = 'ga:pageviews, ga:visits'
sort = '-ga:pageviews' sort = '-ga:pageviews'
   
# Supported query params at # Supported query params at
# https://developers.google.com/analytics/devguides/reporting/core/v3/reference # https://developers.google.com/analytics/devguides/reporting/core/v3/reference
try: try:
# Because of issues of invalid responses, we are going to make these requests  
# ourselves.  
headers = {'authorization': 'Bearer ' + self.token}  
   
args = {} args = {}
args["sort"] = "-ga:pageviews" args["sort"] = "-ga:pageviews"
args["max-results"] = 100000 args["max-results"] = 100000
args["dimensions"] = "ga:pagePath" args["dimensions"] = "ga:pagePath"
args["start-date"] = start_date args["start-date"] = start_date
args["end-date"] = end_date args["end-date"] = end_date
args["metrics"] = metrics args["metrics"] = metrics
args["ids"] = "ga:" + self.profile_id args["ids"] = "ga:" + self.profile_id
args["filters"] = query args["filters"] = query
args["alt"] = "json" args["alt"] = "json"
   
r = requests.get("https://www.googleapis.com/analytics/v3/data/ga", params=args, headers=headers) results = self._get_json(args)
if r.status_code != 200:  
raise Exception("Request with params: %s failed" % args) except Exception, e:
  log.exception(e)
results = json.loads(r.content) return dict(url=[])
print len(results.keys())  
except Exception, e:  
log.exception(e)  
#return dict(url=[])  
raise e  
   
packages = [] packages = []
log.info("There are %d results" % results['totalResults']) log.info("There are %d results" % results['totalResults'])
for entry in results.get('rows'): for entry in results.get('rows'):
(loc,pageviews,visits) = entry (loc,pageviews,visits) = entry
url = _normalize_url('http:/' + loc) # strips off domain e.g. www.data.gov.uk or data.gov.uk url = _normalize_url('http:/' + loc) # strips off domain e.g. www.data.gov.uk or data.gov.uk
   
if not url.startswith('/dataset/') and not url.startswith('/publisher/'): if not url.startswith('/dataset/') and not url.startswith('/publisher/'):
# filter out strays like: # filter out strays like:
# /data/user/login?came_from=http://data.gov.uk/dataset/os-code-point-open # /data/user/login?came_from=http://data.gov.uk/dataset/os-code-point-open
# /403.html?page=/about&from=http://data.gov.uk/publisher/planning-inspectorate # /403.html?page=/about&from=http://data.gov.uk/publisher/planning-inspectorate
continue continue
packages.append( (url, pageviews, visits,) ) # Temporary hack packages.append( (url, pageviews, visits,) ) # Temporary hack
return dict(url=packages) return dict(url=packages)
   
def store(self, period_name, period_complete_day, data): def store(self, period_name, period_complete_day, data):
if 'url' in data: if 'url' in data:
ga_model.update_url_stats(period_name, period_complete_day, data['url']) ga_model.update_url_stats(period_name, period_complete_day, data['url'])
   
def sitewide_stats(self, period_name, period_complete_day): def sitewide_stats(self, period_name, period_complete_day):
import calendar import calendar
year, month = period_name.split('-') year, month = period_name.split('-')
_, last_day_of_month = calendar.monthrange(int(year), int(month)) _, last_day_of_month = calendar.monthrange(int(year), int(month))
   
start_date = '%s-01' % period_name start_date = '%s-01' % period_name
end_date = '%s-%s' % (period_name, last_day_of_month) end_date = '%s-%s' % (period_name, last_day_of_month)
funcs = ['_totals_stats', '_social_stats', '_os_stats', funcs = ['_totals_stats', '_social_stats', '_os_stats',
'_locale_stats', '_browser_stats', '_mobile_stats', '_download_stats'] '_locale_stats', '_browser_stats', '_mobile_stats', '_download_stats']
for f in funcs: for f in funcs:
log.info('Downloading analytics for %s' % f.split('_')[1]) log.info('Downloading analytics for %s' % f.split('_')[1])
getattr(self, f)(start_date, end_date, period_name, period_complete_day) getattr(self, f)(start_date, end_date, period_name, period_complete_day)
   
def _get_results(result_data, f): def _get_results(result_data, f):
data = {} data = {}
for result in result_data: for result in result_data:
key = f(result) key = f(result)
data[key] = data.get(key,0) + result[1] data[key] = data.get(key,0) + result[1]
return data return data
   
def _get_json(self, params, prev_fail=False): def _get_json(self, params, prev_fail=False):
if prev_fail: ga_token_filepath = os.path.expanduser(config.get('googleanalytics.token.filepath', ''))
import os if not ga_token_filepath:
ga_token_filepath = os.path.expanduser(config.get('googleanalytics.token.filepath', '')) print 'ERROR: In the CKAN config you need to specify the filepath of the ' \
if not ga_token_filepath: 'Google Analytics token file under key: googleanalytics.token.filepath'
print 'ERROR: In the CKAN config you need to specify the filepath of the ' \ return
'Google Analytics token file under key: googleanalytics.token.filepath'  
return log.info("Trying to refresh our OAuth token")
  try:
try: from ga_auth import init_service
self.token, svc = init_service(ga_token_filepath, None) self.token, svc = init_service(ga_token_filepath, None)
except TypeError: log.info("OAuth token refreshed")
print ('Have you correctly run the getauthtoken task and ' except Exception, auth_exception:
'specified the correct token file in the CKAN config under ' log.error("Oauth refresh failed")
'"googleanalytics.token.filepath"?') log.exception(auth_exception)
  return
try:  
# Because of issues of invalid responses, we are going to make these requests try:
# ourselves.  
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
r = requests.get("https://www.googleapis.com/analytics/v3/data/ga", params=params, headers=headers) r = requests.get("https://www.googleapis.com/analytics/v3/data/ga", params=params, headers=headers)
if r.status_code != 200: if r.status_code != 200:
log.info("STATUS: %s" % (r.status_code,)) log.info("STATUS: %s" % (r.status_code,))
log.info("CONTENT: %s" % (r.content,)) log.info("CONTENT: %s" % (r.content,))
raise Exception("Request with params: %s failed" % params) raise Exception("Request with params: %s failed" % params)
   
return json.loads(r.content) return json.loads(r.content)
except Exception, e: except Exception, e:
if not prev_fail:  
print e  
results = self._get_json(self, params, prev_fail=True)  
else:  
log.exception(e) log.exception(e)
   
return dict(url=[]) return dict(url=[])
   
def _totals_stats(self, start_date, end_date, period_name, period_complete_day): def _totals_stats(self, start_date, end_date, period_name, period_complete_day):
""" Fetches distinct totals, total pageviews etc """ """ Fetches distinct totals, total pageviews etc """
try: try:
args = {} args = {}
args["max-results"] = 100000 args["max-results"] = 100000
args["start-date"] = start_date args["start-date"] = start_date
args["end-date"] = end_date args["end-date"] = end_date
args["ids"] = "ga:" + self.profile_id args["ids"] = "ga:" + self.profile_id
   
args["metrics"] = "ga:pageviews" args["metrics"] = "ga:pageviews"
args["sort"] = "-ga:pageviews" args["sort"] = "-ga:pageviews"
args["alt"] = "json" args["alt"] = "json"
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
result_data = results.get('rows') result_data = results.get('rows')
ga_model.update_sitewide_stats(period_name, "Totals", {'Total page views': result_data[0][0]}, ga_model.update_sitewide_stats(period_name, "Totals", {'Total page views': result_data[0][0]},
period_complete_day) period_complete_day)
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = {} args = {}
args["max-results"] = 100000 args["max-results"] = 100000
args["start-date"] = start_date args["start-date"] = start_date
args["end-date"] = end_date args["end-date"] = end_date
args["ids"] = "ga:" + self.profile_id args["ids"] = "ga:" + self.profile_id
   
args["metrics"] = "ga:pageviewsPerVisit,ga:avgTimeOnSite,ga:percentNewVisits,ga:visits" args["metrics"] = "ga:pageviewsPerVisit,ga:avgTimeOnSite,ga:percentNewVisits,ga:visits"
args["alt"] = "json" args["alt"] = "json"
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
result_data = results.get('rows') result_data = results.get('rows')
data = { data = {
'Pages per visit': result_data[0][0], 'Pages per visit': result_data[0][0],
'Average time on site': result_data[0][1], 'Average time on site': result_data[0][1],
'New visits': result_data[0][2], 'New visits': result_data[0][2],
'Total visits': result_data[0][3], 'Total visits': result_data[0][3],
} }
ga_model.update_sitewide_stats(period_name, "Totals", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Totals", data, period_complete_day)
   
# Bounces from / or another configurable page. # Bounces from / or another configurable page.
path = '/%s%s' % (config.get('googleanalytics.account'), path = '/%s%s' % (config.get('googleanalytics.account'),
config.get('ga-report.bounce_url', '/')) config.get('ga-report.bounce_url', '/'))
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = {} args = {}
args["max-results"] = 100000 args["max-results"] = 100000
args["start-date"] = start_date args["start-date"] = start_date
args["end-date"] = end_date args["end-date"] = end_date
args["ids"] = "ga:" + self.profile_id args["ids"] = "ga:" + self.profile_id
   
args["filters"] = 'ga:pagePath==%s' % (path,) args["filters"] = 'ga:pagePath==%s' % (path,)
args["dimensions"] = 'ga:pagePath' args["dimensions"] = 'ga:pagePath'
args["metrics"] = "ga:visitBounceRate" args["metrics"] = "ga:visitBounceRate"
args["alt"] = "json" args["alt"] = "json"
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
result_data = results.get('rows') result_data = results.get('rows')
if not result_data or len(result_data) != 1: if not result_data or len(result_data) != 1:
log.error('Could not pinpoint the bounces for path: %s. Got results: %r', log.error('Could not pinpoint the bounces for path: %s. Got results: %r',
path, result_data) path, result_data)
return return
results = result_data[0] results = result_data[0]
bounces = float(results[1]) bounces = float(results[1])
# visitBounceRate is already a % # visitBounceRate is already a %
log.info('Google reports visitBounceRate as %s', bounces) log.info('Google reports visitBounceRate as %s', bounces)
ga_model.update_sitewide_stats(period_name, "Totals", {'Bounce rate (home page)': float(bounces)}, ga_model.update_sitewide_stats(period_name, "Totals", {'Bounce rate (home page)': float(bounces)},
period_complete_day) period_complete_day)
   
   
def _locale_stats(self, start_date, end_date, period_name, period_complete_day): def _locale_stats(self, start_date, end_date, period_name, period_complete_day):
""" Fetches stats about language and country """ """ Fetches stats about language and country """
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = {} args = {}
args["max-results"] = 100000 args["max-results"] = 100000
args["start-date"] = start_date args["start-date"] = start_date
args["end-date"] = end_date args["end-date"] = end_date
args["ids"] = "ga:" + self.profile_id args["ids"] = "ga:" + self.profile_id
   
args["dimensions"] = "ga:language,ga:country" args["dimensions"] = "ga:language,ga:country"
args["metrics"] = "ga:pageviews" args["metrics"] = "ga:pageviews"
args["sort"] = "-ga:pageviews" args["sort"] = "-ga:pageviews"
args["alt"] = "json" args["alt"] = "json"
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
result_data = results.get('rows') result_data = results.get('rows')
data = {} data = {}
for result in result_data: for result in result_data:
data[result[0]] = data.get(result[0], 0) + int(result[2]) data[result[0]] = data.get(result[0], 0) + int(result[2])
self._filter_out_long_tail(data, MIN_VIEWS) self._filter_out_long_tail(data, MIN_VIEWS)
ga_model.update_sitewide_stats(period_name, "Languages", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Languages", data, period_complete_day)
   
data = {} data = {}
for result in result_data: for result in result_data:
data[result[1]] = data.get(result[1], 0) + int(result[2]) data[result[1]] = data.get(result[1], 0) + int(result[2])
self._filter_out_long_tail(data, MIN_VIEWS) self._filter_out_long_tail(data, MIN_VIEWS)
ga_model.update_sitewide_stats(period_name, "Country", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Country", data, period_complete_day)
   
   
def _download_stats(self, start_date, end_date, period_name, period_complete_day): def _download_stats(self, start_date, end_date, period_name, period_complete_day):
""" Fetches stats about data downloads """ """ Fetches stats about data downloads """
import ckan.model as model import ckan.model as model
   
data = {} data = {}
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = {} args = {}
args["max-results"] = 100000 args["max-results"] = 100000
args["start-date"] = start_date args["start-date"] = start_date
args["end-date"] = end_date args["end-date"] = end_date
args["ids"] = "ga:" + self.profile_id args["ids"] = "ga:" + self.profile_id
   
args["filters"] = 'ga:eventAction==download' args["filters"] = 'ga:eventAction==download'
args["dimensions"] = "ga:eventLabel" args["dimensions"] = "ga:eventLabel"
args["metrics"] = "ga:totalEvents" args["metrics"] = "ga:totalEvents"
args["alt"] = "json" args["alt"] = "json"
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
result_data = results.get('rows') result_data = results.get('rows')
if not result_data: if not result_data:
# We may not have data for this time period, so we need to bail # We may not have data for this time period, so we need to bail
# early. # early.
log.info("There is no download data for this time period") log.info("There is no download data for this time period")
return return
   
def process_result_data(result_data, cached=False): def process_result_data(result_data, cached=False):
progress_total = len(result_data) progress_total = len(result_data)
progress_count = 0 progress_count = 0
resources_not_matched = [] resources_not_matched = []
for result in result_data: for result in result_data:
progress_count += 1 progress_count += 1
if progress_count % 100 == 0: if progress_count % 100 == 0:
log.debug('.. %d/%d done so far', progress_count, progress_total) log.debug('.. %d/%d done so far', progress_count, progress_total)
   
url = result[0].strip() url = result[0].strip()
   
# Get package id associated with the resource that has this URL. # Get package id associated with the resource that has this URL.
q = model.Session.query(model.Resource) q = model.Session.query(model.Resource)
if cached: if cached:
r = q.filter(model.Resource.cache_url.like("%s%%" % url)).first() r = q.filter(model.Resource.cache_url.like("%s%%" % url)).first()
else: else:
r = q.filter(model.Resource.url.like("%s%%" % url)).first() r = q.filter(model.Resource.url.like("%s%%" % url)).first()
   
package_name = r.resource_group.package.name if r else "" package_name = r.resource_group.package.name if r else ""
if package_name: if package_name:
data[package_name] = data.get(package_name, 0) + int(result[1]) data[package_name] = data.get(package_name, 0) + int(result[1])
else: else:
resources_not_matched.append(url) resources_not_matched.append(url)
continue continue
if resources_not_matched: if resources_not_matched:
log.debug('Could not match %i or %i resource URLs to datasets. e.g. %r', log.debug('Could not match %i or %i resource URLs to datasets. e.g. %r',
len(resources_not_matched), progress_total, resources_not_matched[:3]) len(resources_not_matched), progress_total, resources_not_matched[:3])
   
log.info('Associating downloads of resource URLs with their respective datasets') log.info('Associating downloads of resource URLs with their respective datasets')
process_result_data(results.get('rows')) process_result_data(results.get('rows'))
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = dict( ids='ga:' + self.profile_id, args = dict( ids='ga:' + self.profile_id,
filters='ga:eventAction==download-cache', filters='ga:eventAction==download-cache',
metrics='ga:totalEvents', metrics='ga:totalEvents',
sort='-ga:totalEvents', sort='-ga:totalEvents',
dimensions="ga:eventLabel", dimensions="ga:eventLabel",
max_results=10000) max_results=10000)
args['start-date'] = start_date args['start-date'] = start_date
args['end-date'] = end_date args['end-date'] = end_date
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
log.info('Associating downloads of cache resource URLs with their respective datasets') log.info('Associating downloads of cache resource URLs with their respective datasets')
process_result_data(results.get('rows'), cached=False) process_result_data(results.get('rows'), cached=False)
   
self._filter_out_long_tail(data, MIN_DOWNLOADS) self._filter_out_long_tail(data, MIN_DOWNLOADS)
ga_model.update_sitewide_stats(period_name, "Downloads", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Downloads", data, period_complete_day)
   
def _social_stats(self, start_date, end_date, period_name, period_complete_day): def _social_stats(self, start_date, end_date, period_name, period_complete_day):
""" Finds out which social sites people are referred from """ """ Finds out which social sites people are referred from """
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = dict( ids='ga:' + self.profile_id, args = dict( ids='ga:' + self.profile_id,
metrics='ga:pageviews', metrics='ga:pageviews',
sort='-ga:pageviews', sort='-ga:pageviews',
dimensions="ga:socialNetwork,ga:referralPath", dimensions="ga:socialNetwork,ga:referralPath",
max_results=10000) max_results=10000)
args['start-date'] = start_date args['start-date'] = start_date
args['end-date'] = end_date args['end-date'] = end_date
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
result_data = results.get('rows') result_data = results.get('rows')
data = {} data = {}
for result in result_data: for result in result_data:
if not result[0] == '(not set)': if not result[0] == '(not set)':
data[result[0]] = data.get(result[0], 0) + int(result[2]) data[result[0]] = data.get(result[0], 0) + int(result[2])
self._filter_out_long_tail(data, 3) self._filter_out_long_tail(data, 3)
ga_model.update_sitewide_stats(period_name, "Social sources", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Social sources", data, period_complete_day)
   
   
def _os_stats(self, start_date, end_date, period_name, period_complete_day): def _os_stats(self, start_date, end_date, period_name, period_complete_day):
""" Operating system stats """ """ Operating system stats """
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = dict( ids='ga:' + self.profile_id, args = dict( ids='ga:' + self.profile_id,
metrics='ga:pageviews', metrics='ga:pageviews',
sort='-ga:pageviews', sort='-ga:pageviews',
dimensions="ga:operatingSystem,ga:operatingSystemVersion", dimensions="ga:operatingSystem,ga:operatingSystemVersion",
max_results=10000) max_results=10000)
args['start-date'] = start_date args['start-date'] = start_date
args['end-date'] = end_date args['end-date'] = end_date
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
result_data = results.get('rows') result_data = results.get('rows')
data = {} data = {}
for result in result_data: for result in result_data:
data[result[0]] = data.get(result[0], 0) + int(result[2]) data[result[0]] = data.get(result[0], 0) + int(result[2])
self._filter_out_long_tail(data, MIN_VIEWS) self._filter_out_long_tail(data, MIN_VIEWS)
ga_model.update_sitewide_stats(period_name, "Operating Systems", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Operating Systems", data, period_complete_day)
   
data = {} data = {}
for result in result_data: for result in result_data:
if int(result[2]) >= MIN_VIEWS: if int(result[2]) >= MIN_VIEWS:
key = "%s %s" % (result[0],result[1]) key = "%s %s" % (result[0],result[1])
data[key] = result[2] data[key] = result[2]
ga_model.update_sitewide_stats(period_name, "Operating Systems versions", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Operating Systems versions", data, period_complete_day)
   
   
def _browser_stats(self, start_date, end_date, period_name, period_complete_day): def _browser_stats(self, start_date, end_date, period_name, period_complete_day):
""" Information about browsers and browser versions """ """ Information about browsers and browser versions """
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = dict( ids='ga:' + self.profile_id, args = dict( ids='ga:' + self.profile_id,
metrics='ga:pageviews', metrics='ga:pageviews',
sort='-ga:pageviews', sort='-ga:pageviews',
dimensions="ga:browser,ga:browserVersion", dimensions="ga:browser,ga:browserVersion",
max_results=10000) max_results=10000)
   
args['start-date'] = start_date args['start-date'] = start_date
args['end-date'] = end_date args['end-date'] = end_date
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
   
result_data = results.get('rows') result_data = results.get('rows')
# e.g. [u'Firefox', u'19.0', u'20'] # e.g. [u'Firefox', u'19.0', u'20']
   
data = {} data = {}
for result in result_data: for result in result_data:
data[result[0]] = data.get(result[0], 0) + int(result[2]) data[result[0]] = data.get(result[0], 0) + int(result[2])
self._filter_out_long_tail(data, MIN_VIEWS) self._filter_out_long_tail(data, MIN_VIEWS)
ga_model.update_sitewide_stats(period_name, "Browsers", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Browsers", data, period_complete_day)
   
data = {} data = {}
for result in result_data: for result in result_data:
key = "%s %s" % (result[0], self._filter_browser_version(result[0], result[1])) key = "%s %s" % (result[0], self._filter_browser_version(result[0], result[1]))
data[key] = data.get(key, 0) + int(result[2]) data[key] = data.get(key, 0) + int(result[2])
self._filter_out_long_tail(data, MIN_VIEWS) self._filter_out_long_tail(data, MIN_VIEWS)
ga_model.update_sitewide_stats(period_name, "Browser versions", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Browser versions", data, period_complete_day)
   
@classmethod @classmethod
def _filter_browser_version(cls, browser, version_str): def _filter_browser_version(cls, browser, version_str):
''' '''
Simplifies a browser version string if it is detailed. Simplifies a browser version string if it is detailed.
i.e. groups together Firefox 3.5.1 and 3.5.2 to be just 3. i.e. groups together Firefox 3.5.1 and 3.5.2 to be just 3.
This is helpful when viewing stats and good to protect privacy. This is helpful when viewing stats and good to protect privacy.
''' '''
ver = version_str ver = version_str
parts = ver.split('.') parts = ver.split('.')
if len(parts) > 1: if len(parts) > 1:
if parts[1][0] == '0': if parts[1][0] == '0':
ver = parts[0] ver = parts[0]
else: else:
ver = "%s" % (parts[0]) ver = "%s" % (parts[0])
# Special case complex version nums # Special case complex version nums
if browser in ['Safari', 'Android Browser']: if browser in ['Safari', 'Android Browser']:
ver = parts[0] ver = parts[0]
if len(ver) > 2: if len(ver) > 2:
num_hidden_digits = len(ver) - 2 num_hidden_digits = len(ver) - 2
ver = ver[0] + ver[1] + 'X' * num_hidden_digits ver = ver[0] + ver[1] + 'X' * num_hidden_digits
return ver return ver
   
def _mobile_stats(self, start_date, end_date, period_name, period_complete_day): def _mobile_stats(self, start_date, end_date, period_name, period_complete_day):
""" Info about mobile devices """ """ Info about mobile devices """
   
try: try:
# Because of issues of invalid responses, we are going to make these requests # Because of issues of invalid responses, we are going to make these requests
# ourselves. # ourselves.
headers = {'authorization': 'Bearer ' + self.token} headers = {'authorization': 'Bearer ' + self.token}
   
args = dict( ids='ga:' + self.profile_id, args = dict( ids='ga:' + self.profile_id,
metrics='ga:pageviews', metrics='ga:pageviews',
sort='-ga:pageviews', sort='-ga:pageviews',
dimensions="ga:mobileDeviceBranding, ga:mobileDeviceInfo", dimensions="ga:mobileDeviceBranding, ga:mobileDeviceInfo",
max_results=10000) max_results=10000)
args['start-date'] = start_date args['start-date'] = start_date
args['end-date'] = end_date args['end-date'] = end_date
   
results = self._get_json(args) results = self._get_json(args)
except Exception, e: except Exception, e:
log.exception(e) log.exception(e)
results = dict(url=[]) results = dict(url=[])
   
   
result_data = results.get('rows') result_data = results.get('rows')
data = {} data = {}
for result in result_data: for result in result_data:
data[result[0]] = data.get(result[0], 0) + int(result[2]) data[result[0]] = data.get(result[0], 0) + int(result[2])
self._filter_out_long_tail(data, MIN_VIEWS) self._filter_out_long_tail(data, MIN_VIEWS)
ga_model.update_sitewide_stats(period_name, "Mobile brands", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Mobile brands", data, period_complete_day)
   
data = {} data = {}
for result in result_data: for result in result_data:
data[result[1]] = data.get(result[1], 0) + int(result[2]) data[result[1]] = data.get(result[1], 0) + int(result[2])
self._filter_out_long_tail(data, MIN_VIEWS) self._filter_out_long_tail(data, MIN_VIEWS)
ga_model.update_sitewide_stats(period_name, "Mobile devices", data, period_complete_day) ga_model.update_sitewide_stats(period_name, "Mobile devices", data, period_complete_day)
   
@classmethod @classmethod
def _filter_out_long_tail(cls, data, threshold=10): def _filter_out_long_tail(cls, data, threshold=10):
''' '''
Given data which is a frequency distribution, filter out Given data which is a frequency distribution, filter out
results which are below a threshold count. This is good to protect results which are below a threshold count. This is good to protect
privacy. privacy.
''' '''
for key, value in data.items(): for key, value in data.items():
if value < threshold: if value < threshold:
del data[key] del data[key]
   
import re import re
import uuid import uuid
   
from sqlalchemy import Table, Column, MetaData, ForeignKey from sqlalchemy import Table, Column, MetaData, ForeignKey
from sqlalchemy import types from sqlalchemy import types
from sqlalchemy.sql import select from sqlalchemy.sql import select
from sqlalchemy.orm import mapper, relation from sqlalchemy.orm import mapper, relation
from sqlalchemy import func from sqlalchemy import func
   
import ckan.model as model import ckan.model as model
from ckan.lib.base import * from ckan.lib.base import *
   
log = __import__('logging').getLogger(__name__) log = __import__('logging').getLogger(__name__)
   
def make_uuid(): def make_uuid():
return unicode(uuid.uuid4()) return unicode(uuid.uuid4())
   
metadata = MetaData() metadata = MetaData()
   
class GA_Url(object): class GA_Url(object):
   
def __init__(self, **kwargs): def __init__(self, **kwargs):
for k,v in kwargs.items(): for k,v in kwargs.items():
setattr(self, k, v) setattr(self, k, v)
   
url_table = Table('ga_url', metadata, url_table = Table('ga_url', metadata,
Column('id', types.UnicodeText, primary_key=True, Column('id', types.UnicodeText, primary_key=True,
default=make_uuid), default=make_uuid),
Column('period_name', types.UnicodeText), Column('period_name', types.UnicodeText),
Column('period_complete_day', types.Integer), Column('period_complete_day', types.Integer),
Column('pageviews', types.UnicodeText), Column('pageviews', types.UnicodeText),
Column('visits', types.UnicodeText), Column('visits', types.UnicodeText),
Column('url', types.UnicodeText), Column('url', types.UnicodeText),
Column('department_id', types.UnicodeText), Column('department_id', types.UnicodeText),
Column('package_id', types.UnicodeText), Column('package_id', types.UnicodeText),
) )
mapper(GA_Url, url_table) mapper(GA_Url, url_table)
   
   
class GA_Stat(object): class GA_Stat(object):
   
def __init__(self, **kwargs): def __init__(self, **kwargs):
for k,v in kwargs.items(): for k,v in kwargs.items():
setattr(self, k, v) setattr(self, k, v)
   
stat_table = Table('ga_stat', metadata, stat_table = Table('ga_stat', metadata,
Column('id', types.UnicodeText, primary_key=True, Column('id', types.UnicodeText, primary_key=True,
default=make_uuid), default=make_uuid),
Column('period_name', types.UnicodeText), Column('period_name', types.UnicodeText),
Column('period_complete_day', types.UnicodeText), Column('period_complete_day', types.UnicodeText),
Column('stat_name', types.UnicodeText), Column('stat_name', types.UnicodeText),
Column('key', types.UnicodeText), Column('key', types.UnicodeText),
Column('value', types.UnicodeText), ) Column('value', types.UnicodeText), )
mapper(GA_Stat, stat_table) mapper(GA_Stat, stat_table)
   
   
class GA_Publisher(object): class GA_Publisher(object):
   
def __init__(self, **kwargs): def __init__(self, **kwargs):
for k,v in kwargs.items(): for k,v in kwargs.items():
setattr(self, k, v) setattr(self, k, v)
   
pub_table = Table('ga_publisher', metadata, pub_table = Table('ga_publisher', metadata,
Column('id', types.UnicodeText, primary_key=True, Column('id', types.UnicodeText, primary_key=True,
default=make_uuid), default=make_uuid),
Column('period_name', types.UnicodeText), Column('period_name', types.UnicodeText),
Column('publisher_name', types.UnicodeText), Column('publisher_name', types.UnicodeText),
Column('views', types.UnicodeText), Column('views', types.UnicodeText),
Column('visits', types.UnicodeText), Column('visits', types.UnicodeText),
Column('toplevel', types.Boolean, default=False), Column('toplevel', types.Boolean, default=False),
Column('subpublishercount', types.Integer, default=0), Column('subpublishercount', types.Integer, default=0),
Column('parent', types.UnicodeText), Column('parent', types.UnicodeText),
) )
mapper(GA_Publisher, pub_table) mapper(GA_Publisher, pub_table)
   
   
class GA_ReferralStat(object): class GA_ReferralStat(object):
   
def __init__(self, **kwargs): def __init__(self, **kwargs):
for k,v in kwargs.items(): for k,v in kwargs.items():
setattr(self, k, v) setattr(self, k, v)
   
referrer_table = Table('ga_referrer', metadata, referrer_table = Table('ga_referrer', metadata,
Column('id', types.UnicodeText, primary_key=True, Column('id', types.UnicodeText, primary_key=True,
default=make_uuid), default=make_uuid),
Column('period_name', types.UnicodeText), Column('period_name', types.UnicodeText),
Column('source', types.UnicodeText), Column('source', types.UnicodeText),
Column('url', types.UnicodeText), Column('url', types.UnicodeText),
Column('count', types.Integer), Column('count', types.Integer),
) )
mapper(GA_ReferralStat, referrer_table) mapper(GA_ReferralStat, referrer_table)
   
   
   
def init_tables(): def init_tables():
metadata.create_all(model.meta.engine) metadata.create_all(model.meta.engine)
   
   
cached_tables = {} cached_tables = {}
   
   
def get_table(name): def get_table(name):
if name not in cached_tables: if name not in cached_tables:
meta = MetaData() meta = MetaData()
meta.reflect(bind=model.meta.engine) meta.reflect(bind=model.meta.engine)
table = meta.tables[name] table = meta.tables[name]
cached_tables[name] = table cached_tables[name] = table
return cached_tables[name] return cached_tables[name]
   
   
def _normalize_url(url): def _normalize_url(url):
'''Strip off the hostname etc. Do this before storing it. '''Strip off the hostname etc. Do this before storing it.
   
>>> normalize_url('http://data.gov.uk/dataset/weekly_fuel_prices') >>> normalize_url('http://data.gov.uk/dataset/weekly_fuel_prices')
'/dataset/weekly_fuel_prices' '/dataset/weekly_fuel_prices'
''' '''
return '/' + '/'.join(url.split('/')[3:]) return '/' + '/'.join(url.split('/')[3:])
   
   
def _get_package_and_publisher(url): def _get_package_and_publisher(url):
# e.g. /dataset/fuel_prices # e.g. /dataset/fuel_prices
# e.g. /dataset/fuel_prices/resource/e63380d4 # e.g. /dataset/fuel_prices/resource/e63380d4
dataset_match = re.match('/dataset/([^/]+)(/.*)?', url) dataset_match = re.match('/dataset/([^/]+)(/.*)?', url)
if dataset_match: if dataset_match:
dataset_ref = dataset_match.groups()[0] dataset_ref = dataset_match.groups()[0]
dataset = model.Package.get(dataset_ref) dataset = model.Package.get(dataset_ref)
if dataset: if dataset:
publisher_groups = dataset.get_groups('publisher') publisher_groups = dataset.get_groups('organization')
if publisher_groups: if publisher_groups:
return dataset_ref,publisher_groups[0].name return dataset_ref,publisher_groups[0].name
return dataset_ref, None return dataset_ref, None
else: else:
publisher_match = re.match('/publisher/([^/]+)(/.*)?', url) publisher_match = re.match('/publisher/([^/]+)(/.*)?', url)
if publisher_match: if publisher_match:
return None, publisher_match.groups()[0] return None, publisher_match.groups()[0]
return None, None return None, None
   
def update_sitewide_stats(period_name, stat_name, data, period_complete_day): def update_sitewide_stats(period_name, stat_name, data, period_complete_day):
for k,v in data.iteritems(): for k,v in data.iteritems():
item = model.Session.query(GA_Stat).\ item = model.Session.query(GA_Stat).\
filter(GA_Stat.period_name==period_name).\ filter(GA_Stat.period_name==period_name).\
filter(GA_Stat.key==k).\ filter(GA_Stat.key==k).\
filter(GA_Stat.stat_name==stat_name).first() filter(GA_Stat.stat_name==stat_name).first()
if item: if item:
item.period_name = period_name item.period_name = period_name
item.key = k item.key = k
item.value = v item.value = v
item.period_complete_day = period_complete_day item.period_complete_day = period_complete_day
model.Session.add(item) model.Session.add(item)
else: else:
# create the row # create the row
values = {'id': make_uuid(), values = {'id': make_uuid(),
'period_name': period_name, 'period_name': period_name,
'period_complete_day': period_complete_day, 'period_complete_day': period_complete_day,
'key': k, 'key': k,
'value': v, 'value': v,
'stat_name': stat_name 'stat_name': stat_name
} }
model.Session.add(GA_Stat(**values)) model.Session.add(GA_Stat(**values))
model.Session.commit() model.Session.commit()
   
   
def pre_update_url_stats(period_name): def pre_update_url_stats(period_name):
q = model.Session.query(GA_Url).\ q = model.Session.query(GA_Url).\
filter(GA_Url.period_name==period_name) filter(GA_Url.period_name==period_name)
log.debug("Deleting %d '%s' records" % (q.count(), period_name)) log.debug("Deleting %d '%s' records" % (q.count(), period_name))
q.delete() q.delete()
   
q = model.Session.query(GA_Url).\ q = model.Session.query(GA_Url).\
filter(GA_Url.period_name == 'All') filter(GA_Url.period_name == 'All')
log.debug("Deleting %d 'All' records..." % q.count()) log.debug("Deleting %d 'All' records..." % q.count())
q.delete() q.delete()
   
model.Session.flush() model.Session.flush()
model.Session.commit() model.Session.commit()
model.repo.commit_and_remove() model.repo.commit_and_remove()
log.debug('...done') log.debug('...done')
   
def post_update_url_stats(): def post_update_url_stats():
   
""" Check the distinct url field in ga_url and make sure """ Check the distinct url field in ga_url and make sure
it has an All record. If not then create one. it has an All record. If not then create one.
   
After running this then every URL should have an All After running this then every URL should have an All
record regardless of whether the URL has an entry for record regardless of whether the URL has an entry for
the month being currently processed. the month being currently processed.
""" """
log.debug('Post-processing "All" records...') log.debug('Post-processing "All" records...')
query = """select url, pageviews::int, visits::int query = """select url, pageviews::int, visits::int
from ga_url from ga_url
where url not in (select url from ga_url where period_name ='All')""" where url not in (select url from ga_url where period_name ='All')"""
connection = model.Session.connection() connection = model.Session.connection()
res = connection.execute(query) res = connection.execute(query)
   
views, visits = {}, {} views, visits = {}, {}
# url, views, visits # url, views, visits
for row in res: for row in res:
views[row[0]] = views.get(row[0], 0) + row[1] views[row[0]] = views.get(row[0], 0) + row[1]
visits[row[0]] = visits.get(row[0], 0) + row[2] visits[row[0]] = visits.get(row[0], 0) + row[2]
   
progress_total = len(views.keys()) progress_total = len(views.keys())
progress_count = 0 progress_count = 0
for key in views.keys(): for key in views.keys():
progress_count += 1 progress_count += 1
if progress_count % 100 == 0: if progress_count % 100 == 0:
log.debug('.. %d/%d done so far', progress_count, progress_total) log.debug('.. %d/%d done so far', progress_count, progress_total)
   
package, publisher = _get_package_and_publisher(key) package, publisher = _get_package_and_publisher(key)
   
values = {'id': make_uuid(), values = {'id': make_uuid(),
'period_name': "All", 'period_name': "All",
'period_complete_day': 0, 'period_complete_day': 0,
'url': key, 'url': key,
'pageviews': views[key], 'pageviews': views[key],
'visits': visits[key], 'visits': visits[key],
'department_id': publisher, 'department_id': publisher,
'package_id': package 'package_id': package
} }
model.Session.add(GA_Url(**values)) model.Session.add(GA_Url(**values))
model.Session.commit() model.Session.commit()
log.debug('..done') log.debug('..done')
   
   
def update_url_stats(period_name, period_complete_day, url_data): def update_url_stats(period_name, period_complete_day, url_data):
''' '''
Given a list of urls and number of hits for each during a given period, Given a list of urls and number of hits for each during a given period,
stores them in GA_Url under the period and recalculates the totals for stores them in GA_Url under the period and recalculates the totals for
the 'All' period. the 'All' period.
''' '''
progress_total = len(url_data) progress_total = len(url_data)
progress_count = 0 progress_count = 0
for url, views, visits in url_data: for url, views, visits in url_data:
progress_count += 1 progress_count += 1
if progress_count % 100 == 0: if progress_count % 100 == 0:
log.debug('.. %d/%d done so far', progress_count, progress_total) log.debug('.. %d/%d done so far', progress_count, progress_total)
   
package, publisher = _get_package_and_publisher(url) package, publisher = _get_package_and_publisher(url)
   
item = model.Session.query(GA_Url).\ item = model.Session.query(GA_Url).\
filter(GA_Url.period_name==period_name).\ filter(GA_Url.period_name==period_name).\
filter(GA_Url.url==url).first() filter(GA_Url.url==url).first()
if item: if item:
item.pageviews = item.pageviews + views item.pageviews = item.pageviews + views
item.visits = item.visits + visits item.visits = item.visits + visits
if not item.package_id: if not item.package_id:
item.package_id = package item.package_id = package
if not item.department_id: if not item.department_id:
item.department_id = publisher item.department_id = publisher
model.Session.add(item) model.Session.add(item)
else: else:
values = {'id': make_uuid(), values = {'id': make_uuid(),
'period_name': period_name, 'period_name': period_name,
'period_complete_day': period_complete_day, 'period_complete_day': period_complete_day,
'url': url, 'url': url,
'pageviews': views, 'pageviews': views,
'visits': visits, 'visits': visits,
'department_id': publisher, 'department_id': publisher,
'package_id': package 'package_id': package
} }
model.Session.add(GA_Url(**values)) model.Session.add(GA_Url(**values))
model.Session.commit() model.Session.commit()
   
if package: if package:
old_pageviews, old_visits = 0, 0 old_pageviews, old_visits = 0, 0
old = model.Session.query(GA_Url).\ old = model.Session.query(GA_Url).\
filter(GA_Url.period_name=='All').\ filter(GA_Url.period_name=='All').\
filter(GA_Url.url==url).all() filter(GA_Url.url==url).all()
old_pageviews = sum([int(o.pageviews) for o in old]) old_pageviews = sum([int(o.pageviews) for o in old])
old_visits = sum([int(o.visits) for o in old]) old_visits = sum([int(o.visits) for o in old])
   
entries = model.Session.query(GA_Url).\ entries = model.Session.query(GA_Url).\
filter(GA_Url.period_name!='All').\ filter(GA_Url.period_name!='All').\
filter(GA_Url.url==url).all() filter(GA_Url.url==url).all()
values = {'id': make_uuid(), values = {'id': make_uuid(),
'period_name': 'All', 'period_name': 'All',
'period_complete_day': 0, 'period_complete_day': 0,
'url': url, 'url': url,
'pageviews': sum([int(e.pageviews) for e in entries]) + int(old_pageviews), 'pageviews': sum([int(e.pageviews) for e in entries]) + int(old_pageviews),
'visits': sum([int(e.visits or 0) for e in entries]) + int(old_visits), 'visits': sum([int(e.visits or 0) for e in entries]) + int(old_visits),
'department_id': publisher, 'department_id': publisher,
'package_id': package 'package_id': package
} }
   
model.Session.add(GA_Url(**values)) model.Session.add(GA_Url(**values))
model.Session.commit() model.Session.commit()
   
   
   
   
def update_social(period_name, data): def update_social(period_name, data):
# Clean up first. # Clean up first.
model.Session.query(GA_ReferralStat).\ model.Session.query(GA_ReferralStat).\
filter(GA_ReferralStat.period_name==period_name).delete() filter(GA_ReferralStat.period_name==period_name).delete()
   
for url,data in data.iteritems(): for url,data in data.iteritems():
for entry in data: for entry in data:
source = entry[0] source = entry[0]
count = entry[1] count = entry[1]
   
item = model.Session.query(GA_ReferralStat).\ item = model.Session.query(GA_ReferralStat).\
filter(GA_ReferralStat.period_name==period_name).\ filter(GA_ReferralStat.period_name==period_name).\
filter(GA_ReferralStat.source==source).\ filter(GA_ReferralStat.source==source).\
filter(GA_ReferralStat.url==url).first() filter(GA_ReferralStat.url==url).first()
if item: if item:
item.count = item.count + count item.count = item.count + count
model.Session.add(item) model.Session.add(item)
else: else:
# create the row # create the row
values = {'id': make_uuid(), values = {'id': make_uuid(),
'period_name': period_name, 'period_name': period_name,
'source': source, 'source': source,
'url': url, 'url': url,
'count': count, 'count': count,
} }
model.Session.add(GA_ReferralStat(**values)) model.Session.add(GA_ReferralStat(**values))
model.Session.commit() model.Session.commit()
   
def update_publisher_stats(period_name): def update_publisher_stats(period_name):
""" """
Updates the publisher stats from the data retrieved for /dataset/* Updates the publisher stats from the data retrieved for /dataset/*
and /publisher/*. Will run against each dataset and generates the and /publisher/*. Will run against each dataset and generates the
totals for the entire tree beneath each publisher. totals for the entire tree beneath each publisher.
""" """
toplevel = get_top_level() toplevel = get_top_level()
publishers = model.Session.query(model.Group).\ publishers = model.Session.query(model.Group).\
filter(model.Group.type=='publisher').\ filter(model.Group.type=='organization').\
filter(model.Group.state=='active').all() filter(model.Group.state=='active').all()
for publisher in publishers: for publisher in publishers:
views, visits, subpub = update_publisher(period_name, publisher, publisher.name) views, visits, subpub = update_publisher(period_name, publisher, publisher.name)
parent, parents = '', publisher.get_groups('publisher') parent, parents = '', publisher.get_parent_groups(type='organization')
if parents: if parents:
parent = parents[0].name parent = parents[0].name
item = model.Session.query(GA_Publisher).\ item = model.Session.query(GA_Publisher).\
filter(GA_Publisher.period_name==period_name).\ filter(GA_Publisher.period_name==period_name).\
filter(GA_Publisher.publisher_name==publisher.name).first() filter(GA_Publisher.publisher_name==publisher.name).first()
if item: if item:
item.views = views item.views = views
item.visits = visits item.visits = visits
item.publisher_name = publisher.name item.publisher_name = publisher.name
item.toplevel = publisher in toplevel item.toplevel = publisher in toplevel
item.subpublishercount = subpub item.subpublishercount = subpub
item.parent = parent item.parent = parent
model.Session.add(item) model.Session.add(item)
else: else:
# create the row # create the row
values = {'id': make_uuid(), values = {'id': make_uuid(),
'period_name': period_name, 'period_name': period_name,
'publisher_name': publisher.name, 'publisher_name': publisher.name,
'views': views, 'views': views,
'visits': visits, 'visits': visits,
'toplevel': publisher in toplevel, 'toplevel': publisher in toplevel,
'subpublishercount': subpub, 'subpublishercount': subpub,
'parent': parent 'parent': parent
} }
model.Session.add(GA_Publisher(**values)) model.Session.add(GA_Publisher(**values))
model.Session.commit() model.Session.commit()
   
   
def update_publisher(period_name, pub, part=''): def update_publisher(period_name, pub, part=''):
views,visits,subpub = 0, 0, 0 views,visits,subpub = 0, 0, 0
for publisher in go_down_tree(pub): for publisher in go_down_tree(pub):
subpub = subpub + 1 subpub = subpub + 1
items = model.Session.query(GA_Url).\ items = model.Session.query(GA_Url).\
filter(GA_Url.period_name==period_name).\ filter(GA_Url.period_name==period_name).\
filter(GA_Url.department_id==publisher.name).all() filter(GA_Url.department_id==publisher.name).all()
for item in items: for item in items:
views = views + int(item.pageviews) views = views + int(item.pageviews)
visits = visits + int(item.visits) visits = visits + int(item.visits)
   
return views, visits, (subpub-1) return views, visits, (subpub-1)
   
   
def get_top_level(): def get_top_level():
'''Returns the top level publishers.''' '''Returns the top level publishers.'''
return model.Session.query(model.Group).\ return model.Session.query(model.Group).\
outerjoin(model.Member, model.Member.table_id == model.Group.id and \ outerjoin(model.Member, model.Member.table_id == model.Group.id and \
model.Member.table_name == 'group' and \ model.Member.table_name == 'group' and \
model.Member.state == 'active').\ model.Member.state == 'active').\
filter(model.Member.id==None).\ filter(model.Member.id==None).\
filter(model.Group.type=='publisher').\ filter(model.Group.type=='organization').\
order_by(model.Group.name).all() order_by(model.Group.name).all()
   
def get_children(publisher): def get_children(publisher):
'''Finds child publishers for the given publisher (object). (Not recursive i.e. returns one level)''' '''Finds child publishers for the given publisher (object). (Not recursive i.e. returns one level)'''
return publisher.get_children_groups(type='organization') return publisher.get_children_groups(type='organization')
   
def go_down_tree(publisher): def go_down_tree(publisher):
'''Provided with a publisher object, it walks down the hierarchy and yields each publisher, '''Provided with a publisher object, it walks down the hierarchy and yields each publisher,
including the one you supply.''' including the one you supply.'''
yield publisher yield publisher
for child in get_children(publisher): for child in get_children(publisher):
for grandchild in go_down_tree(child): for grandchild in go_down_tree(child):
yield grandchild yield grandchild
   
def delete(period_name): def delete(period_name):
''' '''
Deletes table data for the specified period, or specify 'all' Deletes table data for the specified period, or specify 'all'
for all periods. for all periods.
''' '''
for object_type in (GA_Url, GA_Stat, GA_Publisher, GA_ReferralStat): for object_type in (GA_Url, GA_Stat, GA_Publisher, GA_ReferralStat):
q = model.Session.query(object_type) q = model.Session.query(object_type)
if period_name != 'All': if period_name != 'All':
q = q.filter_by(period_name=period_name) q = q.filter_by(period_name=period_name)
q.delete() q.delete()
model.repo.commit_and_remove() model.repo.commit_and_remove()
   
def get_score_for_dataset(dataset_name): def get_score_for_dataset(dataset_name):
''' '''
Returns a "current popularity" score for a dataset, Returns a "current popularity" score for a dataset,
based on how many views it has had recently. based on how many views it has had recently.
''' '''
import datetime import datetime
now = datetime.datetime.now() now = datetime.datetime.now()
last_month = now - datetime.timedelta(days=30) last_month = now - datetime.timedelta(days=30)
period_names = ['%s-%02d' % (last_month.year, last_month.month), period_names = ['%s-%02d' % (last_month.year, last_month.month),
'%s-%02d' % (now.year, now.month), '%s-%02d' % (now.year, now.month),
] ]
   
score = 0 score = 0
for period_name in period_names: for period_name in period_names:
score /= 2 # previous periods are discounted by 50% score /= 2 # previous periods are discounted by 50%
entry = model.Session.query(GA_Url)\ entry = model.Session.query(GA_Url)\
.filter(GA_Url.period_name==period_name)\ .filter(GA_Url.period_name==period_name)\
.filter(GA_Url.package_id==dataset_name).first() .filter(GA_Url.package_id==dataset_name).first()
# score # score
if entry: if entry:
views = float(entry.pageviews) views = float(entry.pageviews)
if entry.period_complete_day: if entry.period_complete_day:
views_per_day = views / entry.period_complete_day views_per_day = views / entry.period_complete_day
else: else:
views_per_day = views / 15 # guess views_per_day = views / 15 # guess
score += views_per_day score += views_per_day
   
score = int(score * 100) score = int(score * 100)
log.debug('Popularity %s: %s', score, dataset_name) log.debug('Popularity %s: %s', score, dataset_name)
return score return score
   
import logging import logging
import operator import operator
   
import ckan.lib.base as base import ckan.lib.base as base
import ckan.model as model import ckan.model as model
from ckan.logic import get_action from ckan.logic import get_action
   
from ckanext.ga_report.ga_model import GA_Url, GA_Publisher from ckanext.ga_report.ga_model import GA_Url, GA_Publisher
from ckanext.ga_report.controller import _get_publishers from ckanext.ga_report.controller import _get_publishers
_log = logging.getLogger(__name__) _log = logging.getLogger(__name__)
   
def popular_datasets(count=10): def popular_datasets(count=10):
import random import random
   
publisher = None publisher = None
publishers = _get_publishers(30) publishers = _get_publishers(30)
total = len(publishers) total = len(publishers)
while not publisher or not datasets: while not publisher or not datasets:
rand = random.randrange(0, total) rand = random.randrange(0, total)
publisher = publishers[rand][0] publisher = publishers[rand][0]
if not publisher.state == 'active': if not publisher.state == 'active':
publisher = None publisher = None
continue continue
datasets = _datasets_for_publisher(publisher, 10)[:count] datasets = _datasets_for_publisher(publisher, 10)[:count]
   
ctx = { ctx = {
'datasets': datasets, 'datasets': datasets,
'publisher': publisher 'publisher': publisher
} }
return base.render_snippet('ga_report/ga_popular_datasets.html', **ctx) return base.render_snippet('ga_report/ga_popular_datasets.html', **ctx)
   
def single_popular_dataset(top=20): def single_popular_dataset(top=20):
'''Returns a random dataset from the most popular ones. '''Returns a random dataset from the most popular ones.
   
:param top: the number of top datasets to select from :param top: the number of top datasets to select from
''' '''
import random import random
   
top_datasets = model.Session.query(GA_Url).\ top_datasets = model.Session.query(GA_Url).\
filter(GA_Url.url.like('/dataset/%')).\ filter(GA_Url.url.like('/dataset/%')).\
order_by('ga_url.pageviews::int desc') order_by('ga_url.pageviews::int desc')
num_top_datasets = top_datasets.count() num_top_datasets = top_datasets.count()
   
dataset = None dataset = None
if num_top_datasets: if num_top_datasets:
count = 0 count = 0
while not dataset: while not dataset:
rand = random.randrange(0, min(top, num_top_datasets)) rand = random.randrange(0, min(top, num_top_datasets))
ga_url = top_datasets[rand] ga_url = top_datasets[rand]
dataset = model.Package.get(ga_url.url[len('/dataset/'):]) dataset = model.Package.get(ga_url.url[len('/dataset/'):])
if dataset and not dataset.state == 'active': if dataset and not dataset.state == 'active':
dataset = None dataset = None
# When testing, it is possible that top datasets are not available # When testing, it is possible that top datasets are not available
# so only go round this loop a few times before falling back on # so only go round this loop a few times before falling back on
# a random dataset. # a random dataset.
count += 1 count += 1
if count > 10: if count > 10:
break break
if not dataset: if not dataset:
# fallback # fallback
dataset = model.Session.query(model.Package)\ dataset = model.Session.query(model.Package)\
.filter_by(state='active').first() .filter_by(state='active').first()
if not dataset: if not dataset:
return None return None
dataset_dict = get_action('package_show')({'model': model, dataset_dict = get_action('package_show')({'model': model,
'session': model.Session, 'session': model.Session,
'validate': False}, 'validate': False},
{'id':dataset.id}) {'id':dataset.id})
return dataset_dict return dataset_dict
   
def single_popular_dataset_html(top=20): def single_popular_dataset_html(top=20):
dataset_dict = single_popular_dataset(top) dataset_dict = single_popular_dataset(top)
groups = package.get('groups', []) groups = package.get('groups', [])
publishers = [ g for g in groups if g.get('type') == 'publisher' ] publishers = [ g for g in groups if g.get('type') == 'organization' ]
publisher = publishers[0] if publishers else {'name':'', 'title': ''} publisher = publishers[0] if publishers else {'name':'', 'title': ''}
context = { context = {
'dataset': dataset_dict, 'dataset': dataset_dict,
'publisher': publisher_dict 'publisher': publisher_dict
} }
return base.render_snippet('ga_report/ga_popular_single.html', **context) return base.render_snippet('ga_report/ga_popular_single.html', **context)
   
   
def most_popular_datasets(publisher, count=20, preview_image=None): def most_popular_datasets(publisher, count=20, preview_image=None):
   
if not publisher: if not publisher:
_log.error("No valid publisher passed to 'most_popular_datasets'") _log.error("No valid publisher passed to 'most_popular_datasets'")
return "" return ""
   
results = _datasets_for_publisher(publisher, count) results = _datasets_for_publisher(publisher, count)
   
ctx = { ctx = {
'dataset_count': len(results), 'dataset_count': len(results),
'datasets': results, 'datasets': results,
   
'publisher': publisher, 'publisher': publisher,
'preview_image': preview_image 'preview_image': preview_image
} }
   
return base.render_snippet('ga_report/publisher/popular.html', **ctx) return base.render_snippet('ga_report/publisher/popular.html', **ctx)
   
def _datasets_for_publisher(publisher, count): def _datasets_for_publisher(publisher, count):
datasets = {} datasets = {}
entries = model.Session.query(GA_Url).\ entries = model.Session.query(GA_Url).\
filter(GA_Url.department_id==publisher.name).\ filter(GA_Url.department_id==publisher.name).\
filter(GA_Url.url.like('/dataset/%')).\ filter(GA_Url.url.like('/dataset/%')).\
order_by('ga_url.pageviews::int desc').all() order_by('ga_url.pageviews::int desc').all()
for entry in entries: for entry in entries:
if len(datasets) < count: if len(datasets) < count:
p = model.Package.get(entry.url[len('/dataset/'):]) p = model.Package.get(entry.url[len('/dataset/'):])
   
if not p: if not p:
_log.warning("Could not find Package for {url}".format(url=entry.url)) _log.warning("Could not find Package for {url}".format(url=entry.url))
continue continue
   
if not p.state == 'active': if not p.state == 'active':
_log.warning("Package {0} is not active, it is {1}".format(p.name, p.state)) _log.warning("Package {0} is not active, it is {1}".format(p.name, p.state))
continue continue
   
if not p in datasets: if not p in datasets:
datasets[p] = {'views':0, 'visits': 0} datasets[p] = {'views':0, 'visits': 0}
   
datasets[p]['views'] = datasets[p]['views'] + int(entry.pageviews) datasets[p]['views'] = datasets[p]['views'] + int(entry.pageviews)
datasets[p]['visits'] = datasets[p]['visits'] + int(entry.visits) datasets[p]['visits'] = datasets[p]['visits'] + int(entry.visits)
   
results = [] results = []
for k, v in datasets.iteritems(): for k, v in datasets.iteritems():
results.append((k,v['views'],v['visits'])) results.append((k,v['views'],v['visits']))
   
return sorted(results, key=operator.itemgetter(1), reverse=True) return sorted(results, key=operator.itemgetter(1), reverse=True)
   
def month_option_title(month_iso, months, day): def month_option_title(month_iso, months, day):
month_isos = [ iso_code for (iso_code,name) in months ] month_isos = [ iso_code for (iso_code,name) in months ]
try: try:
index = month_isos.index(month_iso) index = month_isos.index(month_iso)
except ValueError: except ValueError:
_log.error('Month "%s" not found in list of months.' % month_iso) _log.error('Month "%s" not found in list of months.' % month_iso)
return month_iso return month_iso
month_name = months[index][1] month_name = months[index][1]
if index==0: if index==0:
return month_name + (' (up to %s)'%day) return month_name + (' (up to %s)'%day)
return month_name return month_name
   
   
   
.table-condensed td.sparkline-cell { .table-condensed td.sparkline-cell {
padding: 1px 0 0 0; padding: 1px 0 0 0;
width: 108px; width: 108px;
text-align: center; text-align: center;
/* Hack to hide the momentary flash of text /* Hack to hide the momentary flash of text
* before sparklines are fully rendered */ * before sparklines are fully rendered */
font-size: 1px; font-size: 1px;
color: transparent; color: transparent;
overflow: hidden; overflow: hidden;
} }
.rickshaw_chart_container { .rickshaw_chart_container {
position: relative; position: relative;
height: 350px; height: 350px;
margin: 0 auto 20px auto; margin: 0 auto 20px auto;
} }
.rickshaw_chart { .rickshaw_chart {
position: absolute; position: absolute;
left: 40px; left: 40px;
width: 500px; width: 500px;
top: 0; top: 0;
bottom: 0; bottom: 0;
} }
.rickshaw_legend { .rickshaw_legend {
background: transparent; background: transparent;
width: 100%; width: 100%;
padding-top: 4px; padding-top: 4px;
} }
.rickshaw_y_axis { .rickshaw_y_axis {
position: absolute; position: absolute;
top: 0; top: 0;
bottom: 0; bottom: 0;
width: 40px; width: 40px;
} }
.rickshaw_legend .label { .rickshaw_legend .label {
background: transparent !important; background: transparent !important;
color: #000000 !important; color: #000000 !important;
font-weight: normal !important; font-weight: normal !important;
} }
.rickshaw_legend .instructions { .rickshaw_legend .instructions {
color: #000; color: #000;
margin-bottom: 6px; margin-bottom: 6px;
} }
   
.rickshaw_legend .line .action { .rickshaw_legend .line .action {
display: none; display: none;
} }
.rickshaw_legend .line .swatch { .rickshaw_legend .line .swatch {
display: block; display: block;
float: left; float: left;
} }
.rickshaw_legend .line .label { .rickshaw_legend .line .label {
display: block; display: block;
white-space: normal; white-space: normal;
float: left; float: left;
width: 200px; width: 200px;
} }
.rickshaw_legend .line .label:hover { .rickshaw_legend .line .label:hover {
text-decoration: underline; text-decoration: underline;
} }
   
.ga-reports-table .td-numeric { .ga-reports-table .td-numeric {
text-align: center; text-align: center;
} }
.ga-reports-heading { .ga-reports-heading {
padding-right: 10px; padding-right: 10px;
margin-top: 4px; margin-top: 4px;
float: left; float: left;
} }
  .tab-content {
  padding-top: 12px;
  }
   
var CKAN = CKAN || {}; var CKAN = CKAN || {};
CKAN.GA_Reports = {}; CKAN.GA_Reports = {};
   
CKAN.GA_Reports.render_rickshaw = function( css_name, data, mode, colorscheme ) { CKAN.GA_Reports.render_rickshaw = function( css_name, data, mode, colorscheme ) {
var graphLegends = $('#graph-legend-container'); var graphLegends = $('#graph-legend-container');
   
function renderError(alertClass,alertText,legendText) { function renderError(alertClass,alertText,legendText) {
$("#chart_"+css_name) $("#chart_"+css_name)
.html( '<div class="alert '+alertClass+'">'+alertText+'</div>') .html( '<div class="alert '+alertClass+'">'+alertText+'</div>')
.closest('.rickshaw_chart_container').css('height',50); .closest('.rickshaw_chart_container').css('height',50);
var myLegend = $('<div id="legend_'+css_name+'"/>') var myLegend = $('<div id="legend_'+css_name+'"/>')
.html(legendText) .html(legendText)
.appendTo(graphLegends); .appendTo(graphLegends);
} }
   
if (!Modernizr.svg) { if (!Modernizr.svg) {
renderError('','Your browser does not support vector graphics. No graphs can be rendered.','(Graph cannot be rendered)'); renderError('','Your browser does not support vector graphics. No graphs can be rendered.','(Graph cannot be rendered)');
return; return;
} }
if (data.length==0) { if (data.length==0) {
renderError('alert-info','There is not enough data to render a graph.','(No graph available)'); renderError('alert-info','There is not enough data to render a graph.','(No graph available)');
return return
} }
var myLegend = $('<div id="legend_'+css_name+'"/>').appendTo(graphLegends); var myLegend = $('<div id="legend_'+css_name+'"/>').appendTo(graphLegends);
   
var palette = new Rickshaw.Color.Palette( { scheme: colorscheme } ); var palette = new Rickshaw.Color.Palette( { scheme: colorscheme } );
$.each(data, function(i, object) { $.each(data, function(i, object) {
object['color'] = palette.color(); object['color'] = palette.color();
}); });
// Rickshaw renders the legend in reverse order... // Rickshaw renders the legend in reverse order...
data.reverse(); data.reverse();
   
var graphElement = document.querySelector("#chart_"+css_name); var graphElement = document.querySelector("#chart_"+css_name);
   
var graph = new Rickshaw.Graph( { var graph = new Rickshaw.Graph( {
element: document.querySelector("#chart_"+css_name), element: document.querySelector("#chart_"+css_name),
renderer: mode, renderer: mode,
series: data , series: data ,
height: 328 height: 328
}); });
var x_axis = new Rickshaw.Graph.Axis.Time( { var x_axis = new Rickshaw.Graph.Axis.Time( {
graph: graph graph: graph
} ); } );
var y_axis = new Rickshaw.Graph.Axis.Y( { var y_axis = new Rickshaw.Graph.Axis.Y( {
graph: graph, graph: graph,
orientation: 'left', orientation: 'left',
tickFormat: Rickshaw.Fixtures.Number.formatKMBT, tickFormat: Rickshaw.Fixtures.Number.formatKMBT,
element: document.getElementById('y_axis_'+css_name) element: document.getElementById('y_axis_'+css_name)
} ); } );
var legend = new Rickshaw.Graph.Legend( { var legend = new Rickshaw.Graph.Legend( {
element: document.querySelector('#legend_'+css_name), element: document.querySelector('#legend_'+css_name),
graph: graph graph: graph
} ); } );
var shelving = new Rickshaw.Graph.Behavior.Series.Toggle( { var shelving = new Rickshaw.Graph.Behavior.Series.Toggle( {
graph: graph, graph: graph,
legend: legend legend: legend
} ); } );
myLegend.prepend('<div class="instructions">Click on a series below to isolate its graph:</div>'); myLegend.prepend('<div class="instructions">Click on a series below to isolate its graph:</div>');
graph.render(); graph.render();
}; };
   
CKAN.GA_Reports.bind_sparklines = function() { CKAN.GA_Reports.bind_sparklines = function() {
/* /*
* Bind to the 'totals' tab being on screen, when the * Bind to the 'totals' tab being on screen, when the
* Sparkline graphs should be drawn. * Sparkline graphs should be drawn.
* Note that they cannot be drawn sooner. * Note that they cannot be drawn sooner.
*/ */
var created = false; var created = false;
$('a[href="#totals"]').on( $('a[href="#totals"]').on(
'shown', 'shown',
function() { function() {
if (!created) { if (!created) {
var sparkOptions = { var sparkOptions = {
enableTagOptions: true, enableTagOptions: true,
type: 'line', type: 'line',
width: 100, width: 100,
height: 26, height: 26,
chartRangeMin: 0, chartRangeMin: 0,
spotColor: '', spotColor: '',
maxSpotColor: '', maxSpotColor: '',
minSpotColor: '', minSpotColor: '',
highlightSpotColor: '#000000', highlightSpotColor: '#000000',
lineColor: '#3F8E6D', lineColor: '#3F8E6D',
fillColor: '#B7E66B' fillColor: '#B7E66B'
}; };
$('.sparkline').sparkline('html',sparkOptions); $('.sparkline').sparkline('html',sparkOptions);
created = true; created = true;
} }
$.sparkline_display_visible(); $.sparkline_display_visible();
} }
); );
}; };
   
CKAN.GA_Reports.bind_sidebar = function() { CKAN.GA_Reports.bind_sidebar = function() {
/* /*
* Bind to changes in the tab behaviour: * Bind to changes in the tab behaviour:
* Show the correct rickshaw graph in the sidebar. * Show the correct rickshaw graph in the sidebar.
* Not to be called before all graphs load. * Not to be called before all graphs load.
*/ */
$('a[data-toggle="hashtab"]').on( $('a[data-toggle="tab"]').on(
'shown', 'shown',
function(e) { function(e) {
var href = $(e.target).attr('href'); var href = $(e.target).attr('href');
var pane = $(href); var pane = $(href);
if (!pane.length) { console.err('bad href',href); return; } if (!pane.length) { console.err('bad href',href); return; }
var legend_name = "none"; var legend_name = "none";
var graph = pane.find('.rickshaw_chart'); var graph = pane.find('.rickshaw_chart');
if (graph.length) { if (graph.length) {
legend_name = graph.attr('id').replace('chart_',''); legend_name = graph.attr('id').replace('chart_','');
} }
legend_name = '#legend_'+legend_name; legend_name = '#legend_'+legend_name;
$('#graph-legend-container > *').hide(); $('#graph-legend-container > *').hide();
$('#graph-legend-container .instructions').show(); $('#graph-legend-container .instructions').show();
$(legend_name).show(); $(legend_name).show();
} }
); );
/* The first tab might already have been shown */ /* The first tab might already have been shown */
$('li.active > a[data-toggle="hashtab"]').trigger('shown'); $('li.active > a[data-toggle="tab"]').trigger('shown');
}; };
   
CKAN.GA_Reports.bind_month_selector = function() { CKAN.GA_Reports.bind_month_selector = function() {
var handler = function(e) { var handler = function(e) {
var target = $(e.delegateTarget); var target = $(e.delegateTarget);
var form = target.closest('form'); var form = target.closest('form');
var url = form.attr('action')+'?month='+target.val()+window.location.hash; var url = form.attr('action')+'?month='+target.val()+window.location.hash;
window.location = url; window.location = url;
}; };
var selectors = $('select[name="month"]'); var selectors = $('select[name="month"]');
assert(selectors.length>0); assert(selectors.length>0);
selectors.bind('change', handler); selectors.bind('change', handler);
}; };
   
<html <html
xmlns="http://www.w3.org/1999/xhtml" xmlns="http://www.w3.org/1999/xhtml"
xmlns:i18n="http://genshi.edgewall.org/i18n" xmlns:i18n="http://genshi.edgewall.org/i18n"
xmlns:py="http://genshi.edgewall.org/" xmlns:py="http://genshi.edgewall.org/"
xmlns:xi="http://www.w3.org/2001/XInclude" xmlns:xi="http://www.w3.org/2001/XInclude"
py:strip="" py:strip=""
> >
   
<select name="month" py:def="month_selector(current_month, months, day)"> <select name="month" py:def="month_selector(current_month, months, day)">
<option value='' py:attrs="{'selected': 'selected' if not current_month else None}">All months</option> <option value='' py:attrs="{'selected': 'selected' if not current_month else None}">All months</option>
<py:for each="(iso_code,string_name) in months"> <py:for each="(iso_code,string_name) in months">
<option value='${iso_code}' py:attrs="{'selected': 'selected' if current_month == iso_code else None}">${h.month_option_title(iso_code,months,day)}</option> <option value='${iso_code}' py:attrs="{'selected': 'selected' if current_month == iso_code else None}">${h.month_option_title(iso_code,months,day)}</option>
</py:for> </py:for>
</select> </select>
   
   
<table py:def="social_table(items, with_source=False)" class="ga-reports-table table table-condensed table-bordered table-striped"> <table py:def="social_table(items, with_source=False)" class="ga-reports-table table table-condensed table-bordered table-striped">
<tr> <tr>
<th>Name</th> <th>Name</th>
<th py:if="with_source">Source</th> <th py:if="with_source">Source</th>
<th>Visits</th> <th>Visits</th>
</tr> </tr>
<py:for each="name, url, source, count in items"> <py:for each="name, url, source, count in items">
<tr> <tr>
<td><a href="${url}">${name}</a></td> <td><a href="${url}">${name}</a></td>
<td py:if="with_source">${source}</td> <td py:if="with_source">${source}</td>
<td class="td-numeric">${count}</td> <td class="td-numeric">${count}</td>
</tr> </tr>
</py:for> </py:for>
</table> </table>
   
   
<div py:def="rickshaw_graph(items_json,id,mode='line',colorscheme='munin')"> <div py:def="rickshaw_graph(items_json,id,mode='line',colorscheme='munin')">
<div id="chart_container_$id" class="rickshaw_chart_container"> <div id="chart_container_$id" class="rickshaw_chart_container">
<div id="y_axis_$id" class="rickshaw_y_axis"></div> <div id="y_axis_$id" class="rickshaw_y_axis"></div>
<div id="chart_$id" class="rickshaw_chart"></div> <div id="chart_$id" class="rickshaw_chart"></div>
<script type="text/javascript"> <script type="text/javascript">
$(function() { $(function() {
var items = $items_json; var items = $items_json;
CKAN.GA_Reports.render_rickshaw('$id', items, '$mode', '$colorscheme'); CKAN.GA_Reports.render_rickshaw('$id', items, '$mode', '$colorscheme');
}); });
</script> </script>
</div> </div>
</div> </div>
   
   
<table py:def="stat_table(items, title='Views')" class="ga-reports-table table table-condensed table-bordered table-striped"> <table py:def="stat_table(items, title='Views')" class="ga-reports-table table table-condensed table-bordered table-striped">
<tr> <tr>
<th>Name</th> <th>Name</th>
<th class="td-numeric">% ${title}</th> <th class="td-numeric">% ${title}</th>
</tr> </tr>
<py:for each="name, value in items"> <py:for each="name, value in items">
<tr> <tr>
<td>${name}</td> <td>${name}</td>
<td class="td-numeric">${value}</td> <td class="td-numeric">${value}</td>
</tr> </tr>
</py:for> </py:for>
</table> </table>
   
  <py:def function="ga_footer()">
<div py:def="usage_nav(active_name)"> <div class="panel panel-default">
<span class="subheading"> <div class="panel-heading"><strong>Notes</strong></div>
<img src="/images/arrow-down-right.png" /> <div class="panel-body">
<a class="${if_(active_name=='Site-wide','active')}" href="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}${if_(c.month,'?month='+c.month)}">Site-wide</a>  
<span class="divider">|</span>  
<a class="${if_(active_name=='Publishers','active')}" href="${h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='publishers')}${if_(c.month,'?month='+c.month)}">Publishers</a>  
<span class="divider">|</span>  
<a class="${if_(active_name=='Datasets','active')}" href="${h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='read')}${if_(c.month,'?month='+c.month)}">Datasets</a>  
</span>  
</div>  
   
<div py:def="ga_sidebar(download_link)">  
<div class="boxed">  
<div class="widget-container widget_text">  
<h4>Download</h4>  
<p><center>  
<a class="btn button btn-primary" href="${download_link}">Download as CSV</a></center>  
</p>  
</div>  
<div class="widget-container widget_text">  
<h4>Graph Legend</h4>  
<div id="graph-legend-container">  
<div style="display: none;" id="legend_none">(No graph is loaded)</div>  
</div>  
</div>  
<div class="widget-container widget_text">  
<h4>Notes</h4>  
<ul> <ul>
<li>"Views" is the number of times a page was loaded in users' browsers.</li> <li>"Views" is the number of times a page was loaded in users' browsers.</li>
<li>"Downloads" is the number of times a user has clicked to download either an original or cached resource for a particular dataset. Download information is only available from 2nd December 2012; 'No data' is shown for records before that date.</li> <li>"Downloads" is the number of times a user has clicked to download either an original or cached resource for a particular dataset. Download information is only available from 2nd December 2012; 'No data' is shown for records before that date.</li>
<li>These usage statistics are confined to users with javascript enabled, which excludes web crawlers and API calls.</li> <li>These usage statistics are confined to users with javascript enabled, which excludes web crawlers and API calls.</li>
<li>The results are not shown when the number of views/visits is tiny. Where these relate to site pages, results are available in full in the CSV download. Where these relate to users' web browser information, results are not disclosed, for privacy reasons.</li> <li>The results are not shown when the number of views/visits is tiny. Where these relate to site pages, results are available in full in the CSV download. Where these relate to users' web browser information, results are not disclosed, for privacy reasons.</li>
</ul> </ul>
</div> </div>
</div> </div>
</div> </py:def>
   
   
</html> </html>
   
<html xmlns:py="http://genshi.edgewall.org/" <html xmlns:py="http://genshi.edgewall.org/"
xmlns:i18n="http://genshi.edgewall.org/i18n" xmlns:i18n="http://genshi.edgewall.org/i18n"
xmlns:xi="http://www.w3.org/2001/XInclude" xmlns:xi="http://www.w3.org/2001/XInclude"
py:strip=""> py:strip="">
   
<xi:include href="../ga_util.html" /> <xi:include href="../ga_util.html" />
   
<py:def function="page_title">Usage by Publisher</py:def> <py:def function="page_title">Usage by Publisher</py:def>
   
   
   
<py:def function="sidebar">  
${ga_sidebar(download_link=h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='publisher_csv',month=c.month or 'all'))}  
</py:def>  
   
   
<py:def function="optional_head"> <py:def function="optional_head">
<link rel="stylesheet" type="text/css" href="/scripts/vendor/rickshaw.min.css"/> <link rel="stylesheet" type="text/css" href="/scripts/vendor/rickshaw.min.css"/>
<link rel="stylesheet" type="text/css" href="/css/ga_report.css?1"/> <link rel="stylesheet" type="text/css" href="/css/ga_report.css?1"/>
<script type="text/javascript" src="/scripts/modernizr-2.6.2.custom.js"></script> <script type="text/javascript" src="/scripts/modernizr-2.6.2.custom.js"></script>
<script type="text/javascript" src="/scripts/ckanext_ga_reports.js?1"></script> <script type="text/javascript" src="/scripts/ckanext_ga_reports.js?1"></script>
<script type="text/javascript" src="/scripts/vendor/jquery.sparkline.modified.js"></script> <script type="text/javascript" src="/scripts/vendor/jquery.sparkline.modified.js"></script>
<script type="text/javascript" src="/scripts/rickshaw_ie7_shim.js"></script> <script type="text/javascript" src="/scripts/rickshaw_ie7_shim.js"></script>
<script type="text/javascript" src="/scripts/vendor/d3.v2.js"></script> <script type="text/javascript" src="/scripts/vendor/d3.v2.js"></script>
<script type="text/javascript" src="/scripts/vendor/d3.layout.min.js"></script> <script type="text/javascript" src="/scripts/vendor/d3.layout.min.js"></script>
<script type="text/javascript" src="/scripts/vendor/rickshaw.min.js"></script> <script type="text/javascript" src="/scripts/vendor/rickshaw.min.js"></script>
</py:def> </py:def>
<py:def function="page_heading">Site Usage ${usage_nav('Publishers')}</py:def>  
  <py:match path="breadcrumbs">
  <li><a href="/data/site-usage">Site Analytics</a></li>
  <li><a href="/data/site-usage/publisher">Publishers</a></li>
  </py:match>
   
<div py:match="content"> <div py:match="content">
  <py:with vars="download_link=h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='publisher_csv',month=c.month or 'all')">
  <a class="btn button btn-primary btn-sm pull-right" href="${download_link}"><i class="icon-download"></i>&nbsp; Download as CSV</a>
  </py:with>
  <h1>Site Usage</h1>
   
  <div class="row" style="background: #fff;">
<div class="boxed"> <div class="col-md-4">
  <div class="whitebox">
${rickshaw_graph(c.top_publishers_graph,'publishers')} <strong>Graph Legend</strong>
  <div id="graph-legend-container">
  <div style="display: none;" id="legend_none">(No graph is loaded)</div>
  </div>
  </div>
  </div>
  <div class="col-md-8">
  <div class="whitebox">
  <strong>Publishers</strong>
  ${rickshaw_graph(c.top_publishers_graph,'publishers')}
  </div>
  </div>
  </div>
   
<hr/> <hr/>
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='publishers')}" method="get"> <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='publishers')}" method="get">
<div class="controls"> <div class="controls">
<h4 class="ga-reports-heading">Statistics for</h4> <h2 class="ga-reports-heading">Statistics for</h2>
${month_selector(c.month, c.months, c.day)} ${month_selector(c.month, c.months, c.day)}
</div> </div>
</form> </form>
   
<table class="ga-reports-table table table-condensed table-bordered table-striped"> <table class="ga-reports-table table table-condensed table-bordered table-striped">
<tr> <tr>
<th>Publisher</th> <th>Publisher</th>
<th class="td-numeric">Dataset Views</th> <th class="td-numeric">Dataset Views</th>
</tr> </tr>
<py:for each="publisher, views, visits in c.top_publishers"> <py:for each="publisher, views, visits in c.top_publishers">
<tr> <tr>
<td> <td>
${h.link_to(publisher.title, h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport', action='read_publisher', id=publisher.name) + (("?month=" + c.month) if c.month else ''))} ${h.link_to(publisher.title, h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport', action='read_publisher', id=publisher.name) + (("?month=" + c.month) if c.month else ''))}
</td> </td>
<td class="td-numeric">${views}</td> <td class="td-numeric">${views}</td>
</tr> </tr>
</py:for> </py:for>
</table> </table>
</div><!--/boxed-->  
</div><!--/content--> </div><!--/content-->
   
<py:def function="optional_footer"> <py:def function="optional_footer">
<script type="text/javascript"> <script type="text/javascript">
$(function() { $(function() {
CKAN.GA_Reports.bind_month_selector(); CKAN.GA_Reports.bind_month_selector();
}); });
</script> </script>
</py:def> </py:def>
<xi:include href="../../layout.html" /> <xi:include href="../../layout.html" />
   
</html> </html>
   
   
<html xmlns:py="http://genshi.edgewall.org/" <html xmlns:py="http://genshi.edgewall.org/"
xmlns:i18n="http://genshi.edgewall.org/i18n" xmlns:i18n="http://genshi.edgewall.org/i18n"
xmlns:xi="http://www.w3.org/2001/XInclude" xmlns:xi="http://www.w3.org/2001/XInclude"
py:strip=""> py:strip="">
   
<xi:include href="../ga_util.html" /> <xi:include href="../ga_util.html" />
   
<py:def function="page_title">Usage by Dataset</py:def> <py:def function="page_title">Usage by Dataset</py:def>
   
<py:def function="optional_head"> <py:def function="optional_head">
<link rel="stylesheet" type="text/css" href="/scripts/vendor/rickshaw.min.css"/> <link rel="stylesheet" type="text/css" href="/scripts/vendor/rickshaw.min.css"/>
<link rel="stylesheet" type="text/css" href="/css/ga_report.css?1"/> <link rel="stylesheet" type="text/css" href="/css/ga_report.css?1"/>
<script type="text/javascript" src="/scripts/modernizr-2.6.2.custom.js"></script> <script type="text/javascript" src="/scripts/modernizr-2.6.2.custom.js"></script>
<script type="text/javascript" src="/scripts/ckanext_ga_reports.js?1"></script> <script type="text/javascript" src="/scripts/ckanext_ga_reports.js?1"></script>
<script type="text/javascript" src="/scripts/vendor/jquery.sparkline.modified.js"></script> <script type="text/javascript" src="/scripts/vendor/jquery.sparkline.modified.js"></script>
<script type="text/javascript" src="/scripts/rickshaw_ie7_shim.js"></script> <script type="text/javascript" src="/scripts/rickshaw_ie7_shim.js"></script>
<script type="text/javascript" src="/scripts/vendor/d3.v2.js"></script> <script type="text/javascript" src="/scripts/vendor/d3.v2.js"></script>
<script type="text/javascript" src="/scripts/vendor/d3.layout.min.js"></script> <script type="text/javascript" src="/scripts/vendor/d3.layout.min.js"></script>
<script type="text/javascript" src="/scripts/vendor/rickshaw.min.js"></script> <script type="text/javascript" src="/scripts/vendor/rickshaw.min.js"></script>
</py:def> </py:def>
   
<py:def function="sidebar">  
${ga_sidebar(download_link=h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='dataset_csv',id=c.publisher_name or 'all',month=c.month or 'all'))} <py:match path="breadcrumbs">
</py:def> <li><a href="/data/site-usage">Site Analytics</a></li>
<py:def function="page_heading">Site Usage ${usage_nav('Datasets')}</py:def> <py:if test="c.publisher">
  <li><a href="/data/site-usage/publisher">Publishers</a></li>
  <li py:if="c.publisher"><a href="/data/site-usage/publisher/${c.publisher.name}">${c.publisher.title}</a></li>
  </py:if>
  <py:if test="not c.publisher">
  <li><a href="${request.url}">Usage By Dataset</a></li>
  </py:if>
  </py:match>
   
<div py:match="content"> <div py:match="content">
   
<div class="boxed"> <py:with vars="download_link=h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='dataset_csv',id=c.publisher_name or 'all',month=c.month or 'all')">
<h3 py:if="c.publisher"><a href="${h.url_for(controller='ckanext.dgu.controllers.publisher:PublisherController',action='read',id=c.publisher.name)}">${c.publisher.title}</a></h3> <a class="btn button btn-primary btn-sm pull-right" href="${download_link}"><i class="icon-download"></i>&nbsp; Download as CSV</a>
  </py:with>
  <h1>Site Usage
  <small py:if="c.publisher">${c.publisher.title}</small>
  <small py:if="not c.publisher">All datasets</small>
  </h1>
   
<py:if test="c.graph_data"> <div class="row" style="background: #fff;">
${rickshaw_graph(c.graph_data,'dataset-downloads',debug=True)} <div class="col-md-4">
</py:if> <div class="whitebox">
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='read')}" method="get"> <strong>Graph Legend</strong>
<div class="controls"> <div id="graph-legend-container">
  <div style="display: none;" id="legend_none">(No graph is loaded)</div>
  </div>
  </div>
  </div>
  <div class="col-md-8">
  <div class="whitebox">
  <py:if test="c.graph_data">
  ${rickshaw_graph(c.graph_data,'dataset-downloads',debug=True)}
  </py:if>
  </div>
  </div>
  </div>
  <hr/>
  <py:if test="c.month">
  <h4>Statistics for ${h.month_option_title(c.month,c.months,c.day)}:</h4>
  </py:if>
  <py:if test="not c.month">
  <h2>Statistics for all months</h2>
  </py:if>
  <form style="margin-bottom:10px;" class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport',action='read')}" method="get">
  <div class="controls">
   
${month_selector(c.month, c.months, c.day)} ${month_selector(c.month, c.months, c.day)}
   
<select name="publisher"> <select name="publisher">
<option value='' py:attrs="{'selected': 'selected' if not c.publisher else None}">All publishers</option> <option value='' py:attrs="{'selected': 'selected' if not c.publisher else None}">All publishers</option>
<py:for each="val,desc in c.publishers"> <py:for each="val,desc in c.publishers">
<option value='${val}' py:attrs="{'selected': 'selected' if c.publisher_name == val else None}">${desc}</option> <option value='${val}' py:attrs="{'selected': 'selected' if c.publisher_name == val else None}">${desc}</option>
</py:for> </py:for>
</select> </select>
<input class="btn button btn-primary" type='submit' value="Update"/> <input class="btn button btn-primary btn-xs" type='submit' value="Update"/>
</div> </div>
</form> </form>
<py:if test="c.month"> <div class="alert alert-info" py:if="not c.top_packages">No page views in this period.</div>
<h4>Statistics for ${h.month_option_title(c.month,c.months,c.day)}:</h4> <py:if test="c.top_packages">
</py:if> <table class="ga-reports-table table table-condensed table-bordered">
<py:if test="not c.month"> <tr>
<h4>Statistics for all months:</h4> <th>Dataset</th>
</py:if> <th>Views</th>
<div class="alert alert-info" py:if="not c.top_packages">No page views in this period.</div> <th>Downloads</th>
<py:if test="c.top_packages"> </tr>
<table class="ga-reports-table table table-condensed table-bordered table-striped"> <py:for each="package, views, visits,downloads in c.top_packages">
<tr> <tr>
<th>Dataset</th> <td>
<th>Views</th> ${h.link_to(package.title or package.name, h.url_for(controller='package', action='read', id=package.name))}
<th>Downloads</th> </td>
</tr> <td class="td-numeric">${views}</td>
<py:for each="package, views, visits,downloads in c.top_packages"> <td class="td-numeric">${downloads}</td>
<tr> </tr>
<td> </py:for>
${h.link_to(package.title or package.name, h.url_for(controller='package', action='read', id=package.name))} </table>
</td> </py:if>
<td class="td-numeric">${views}</td>  
<td class="td-numeric">${downloads}</td>  
</tr>  
</py:for>  
</table>  
</py:if>  
   
</div> ${ga_footer()}
   
</div> </div>
   
<xi:include href="../../layout.html" /> <xi:include href="../../layout.html" />
</html> </html>
   
   
   
   
<html xmlns:py="http://genshi.edgewall.org/" <html xmlns:py="http://genshi.edgewall.org/"
xmlns:i18n="http://genshi.edgewall.org/i18n" xmlns:i18n="http://genshi.edgewall.org/i18n"
xmlns:xi="http://www.w3.org/2001/XInclude" xmlns:xi="http://www.w3.org/2001/XInclude"
py:strip=""> py:strip="">
   
<xi:include href="../ga_util.html" /> <xi:include href="../ga_util.html" />
   
<py:def function="page_title">Site usage</py:def> <py:def function="page_title">Site usage</py:def>
   
<py:def function="optional_head"> <py:def function="optional_head">
<link rel="stylesheet" type="text/css" href="/scripts/vendor/rickshaw.min.css"/> <link rel="stylesheet" type="text/css" href="/scripts/vendor/rickshaw.min.css"/>
<link rel="stylesheet" type="text/css" href="/css/ga_report.css?1"/> <link rel="stylesheet" type="text/css" href="/css/ga_report.css?1"/>
<script type="text/javascript" src="/scripts/modernizr-2.6.2.custom.js"></script> <script type="text/javascript" src="/scripts/modernizr-2.6.2.custom.js"></script>
<script type="text/javascript" src="/scripts/ckanext_ga_reports.js?1"></script> <script type="text/javascript" src="/scripts/ckanext_ga_reports.js?1"></script>
<script type="text/javascript" src="/scripts/vendor/jquery.sparkline.modified.js"></script> <script type="text/javascript" src="/scripts/vendor/jquery.sparkline.modified.js"></script>
<script type="text/javascript" src="/scripts/rickshaw_ie7_shim.js"></script> <script type="text/javascript" src="/scripts/rickshaw_ie7_shim.js"></script>
<script type="text/javascript" src="/scripts/vendor/d3.v2.js"></script> <script type="text/javascript" src="/scripts/vendor/d3.v2.js"></script>
<script type="text/javascript" src="/scripts/vendor/d3.layout.min.js"></script> <script type="text/javascript" src="/scripts/vendor/d3.layout.min.js"></script>
<script type="text/javascript" src="/scripts/vendor/rickshaw.min.js"></script> <script type="text/javascript" src="/scripts/vendor/rickshaw.min.js"></script>
</py:def> </py:def>
   
<py:def function="sidebar"> <py:match path="breadcrumbs">
${ga_sidebar(download_link=h.url_for(controller='ckanext.ga_report.controller:GaReport',action='csv',month=c.month or 'all'))} <li><a href="/data/site-usage">Site Analytics</a></li>
</py:def> <li><a href="/data/site-usage">Site-wide</a></li>
  </py:match>
<py:def function="page_heading">Site Usage ${usage_nav('Site-wide')}</py:def>  
   
<div py:match="content"> <div py:match="content">
  <div class="row">
<div class="boxed"> <div class="col-sm-7 col-md-8 col-lg-9">
<div class="tabbable"> <py:with vars="download_link=h.url_for(controller='ckanext.ga_report.controller:GaReport',action='csv',month=c.month or 'all')">
<ul class="nav nav-tabs"> <a class="btn button btn-primary btn-sm pull-right" href="${download_link}"><i class="icon-download"></i>&nbsp; Download as CSV</a>
<li class="active"><a href="#totals" data-toggle="hashtab">Totals</a></li> </py:with>
<li class="dropdown"> <h1>Site Usage</h1>
<a href="#" class="dropdown-toggle" data-toggle="dropdown">Browsers </div>
<b class="caret"></b></a> <div class="col-sm-5 col-md-4 col-lg-3">
<ul class="dropdown-menu"> <div class="panel panel-default">
<li><a href="#browsers_names" data-toggle="hashtab">Browsers</a></li> <div class="panel-heading"><strong>Jump To...</strong></div>
<li><a href="#browsers_versions" data-toggle="hashtab">Versions</a></li> <div class="panel-body">
  <ul>
  <li><a href="/data/site-usage/publisher">Publisher Usage Statistics</a></li>
  <li><a href="/data/site-usage/dataset">Dataset Usage Statistics</a></li>
</ul> </ul>
</li> </div>
<li class="dropdown"> </div>
<a href="#" class="dropdown-toggle" data-toggle="dropdown">Operating Systems </div>
<b class="caret"></b></a> </div>
<ul class="dropdown-menu">  
<li><a href="#os" data-toggle="hashtab">Operating Systems</a></li> <div class="row" style="background: #fff;">
<li><a href="#os_versions" data-toggle="hashtab">Versions</a></li> <div class="col-md-4">
  <div class="whitebox">
  <strong>Graph Legend</strong>
  <div id="graph-legend-container">
  <div style="display: none;" id="legend_none">(No graph is loaded)</div>
  </div>
  </div>
  </div>
  <div class="col-md-8">
  <div class="whitebox">
  <div class="tabbable">
  <ul class="nav nav-tabs">
  <li class="active"><a href="#totals" data-hash="totals" data-toggle="tab">Totals</a></li>
  <li class="dropdown">
  <a href="#" class="dropdown-toggle" data-toggle="dropdown">Browsers
  <b class="caret"></b></a>
  <ul class="dropdown-menu">
  <li><a href="#browsers_names" data-hash="browsers_names" data-toggle="tab">Browsers</a></li>
  <li><a href="#browsers_versions" data-hash="browsers_versions" data-toggle="tab">Versions</a></li>
  </ul>
  </li>
  <li class="dropdown">
  <a href="#" class="dropdown-toggle" data-toggle="dropdown">Operating Systems
  <b class="caret"></b></a>
  <ul class="dropdown-menu">
  <li><a href="#os" data-hash="os" data-toggle="tab">Operating Systems</a></li>
  <li><a href="#os_versions" data-hash="os_versions" data-toggle="tab">Versions</a></li>
  </ul>
  </li>
  <li class="dropdown">
  <a href="#" class="dropdown-toggle" data-toggle="dropdown">Social
  <b class="caret"></b></a>
  <ul class="dropdown-menu">
  <li><a href="#social_networks" data-hash="social_networks" data-toggle="tab">All networks</a></li>
  <li><a href="#social_referrals_totals" data-hash="social_referrals_totals" data-toggle="tab">Referral links</a></li>
  </ul>
  </li>
  <li><a href="#languages" data-hash="languages" data-toggle="tab">Languages</a></li>
  <li><a href="#country" data-hash="country" data-toggle="tab">Country</a></li>
</ul> </ul>
</li> <div class="tab-content">
<li class="dropdown"> <div class="tab-pane active" id="totals">
<a href="#" class="dropdown-toggle" data-toggle="dropdown">Social <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
<b class="caret"></b></a> <h4 class="ga-reports-heading">Show stats table for:</h4>
<ul class="dropdown-menu"> ${month_selector(c.month, c.months, c.day)}
<li><a href="#social_networks" data-toggle="hashtab">All networks</a></li> </form>
<li><a href="#social_referrals_totals" data-toggle="hashtab">Referral links</a></li> <table class="ga-reports-table table table-condensed table-bordered table-striped">
</ul> <tr>
</li> <th>Name</th>
<li><a href="#languages" data-toggle="hashtab">Languages</a></li> <th class="td-numeric">Value</th>
<li><a href="#country" data-toggle="hashtab">Country</a></li> <th>History</th>
</ul> </tr>
<div class="tab-content"> <py:for each="name, value, graph in c.global_totals">
<div class="tab-pane active" id="totals"> <tr>
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> <td>${name}</td>
<h4 class="ga-reports-heading">Show stats table for:</h4> <td class="td-numeric">${value}</td>
${month_selector(c.month, c.months, c.day)} <td class="sparkline-cell">
</form> <span class="sparkline" sparkTooltips="${','.join([x for x,y in graph])}">
<table class="ga-reports-table table table-condensed table-bordered table-striped"> ${','.join([y for x,y in graph])}
<tr> </span>
<th>Name</th> </td>
<th class="td-numeric">Value</th> </tr>
<th>History</th> </py:for>
</tr> </table>
<py:for each="name, value, graph in c.global_totals"> </div>
<tr> <div class="tab-pane" id="browsers_versions">
<td>${name}</td> ${rickshaw_graph(c.browser_versions_graph,'browser-versions',mode='stack')}
<td class="td-numeric">${value}</td> <hr/>
<td class="sparkline-cell"> <p>Note: Where a browser has a large number of versions, these have been grouped together.</p>
<span class="sparkline" sparkTooltips="${','.join([x for x,y in graph])}"> <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
${','.join([y for x,y in graph])} <h4 class="ga-reports-heading">Show stats table for:</h4>
</span> ${month_selector(c.month, c.months, c.day)}
</td> </form>
</tr> <hr/>
</py:for> ${stat_table(c.browser_versions)}
</table> </div>
</div> <div class="tab-pane" id="browsers_names">
<div class="tab-pane" id="browsers_versions"> ${rickshaw_graph(c.browsers_graph,'browsers',mode='stack')}
${rickshaw_graph(c.browser_versions_graph,'browser-versions',mode='stack')} <hr/>
<hr/> <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
<p>Note: Where a browser has a large number of versions, these have been grouped together.</p> <h4 class="ga-reports-heading">Show stats table for:</h4>
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> ${month_selector(c.month, c.months, c.day)}
<h4 class="ga-reports-heading">Show stats table for:</h4> </form>
${month_selector(c.month, c.months, c.day)} <hr/>
</form> ${stat_table(c.browsers)}
${stat_table(c.browser_versions)} </div>
</div> <div class="tab-pane" id="os">
<div class="tab-pane" id="browsers_names"> ${rickshaw_graph(c.os_graph,'os',mode='stack')}
${rickshaw_graph(c.browsers_graph,'browsers',mode='stack')} <hr/>
<hr/> <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> <h4 class="ga-reports-heading">Show stats table for:</h4>
<h4 class="ga-reports-heading">Show stats table for:</h4> ${month_selector(c.month, c.months, c.day)}
${month_selector(c.month, c.months, c.day)} </form>
</form> <hr/>
${stat_table(c.browsers)} ${stat_table(c.os)}
</div> </div>
<div class="tab-pane" id="os"> <div class="tab-pane" id="os_versions">
${rickshaw_graph(c.os_graph,'os',mode='stack')} ${rickshaw_graph(c.os_versions_graph,'os_versions',mode='stack')}
<hr/> <hr/>
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
<h4 class="ga-reports-heading">Show stats table for:</h4> <h4 class="ga-reports-heading">Show stats table for:</h4>
${month_selector(c.month, c.months, c.day)} ${month_selector(c.month, c.months, c.day)}
</form> </form>
${stat_table(c.os)} <hr/>
</div> ${stat_table(c.os_versions)}
<div class="tab-pane" id="os_versions"> </div>
${rickshaw_graph(c.os_versions_graph,'os_versions',mode='stack')} <div class="tab-pane" id="social_referrals_totals">
<hr/> <p>Number of visits that were referred from social networks</p>
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
<h4 class="ga-reports-heading">Show stats table for:</h4> <h4 class="ga-reports-heading">Show stats table for:</h4>
${month_selector(c.month, c.months, c.day)} ${month_selector(c.month, c.months, c.day)}
</form> </form>
${stat_table(c.os_versions)} ${social_table(c.social_referrer_totals)}
</div> </div>
<div class="tab-pane" id="social_referrals_totals"> <div class="tab-pane" id="social_networks">
<p>Number of visits that were referred from social networks</p> ${rickshaw_graph(c.social_networks_graph, 'social_networks',mode='stack')}
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> <hr/>
<h4 class="ga-reports-heading">Show stats table for:</h4> <p>Percentage of visits that were referred from these social networks</p>
${month_selector(c.month, c.months, c.day)} <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
</form> <h4 class="ga-reports-heading">Show stats table for:</h4>
${social_table(c.social_referrer_totals)} ${month_selector(c.month, c.months, c.day)}
</div> </form>
<div class="tab-pane" id="social_networks"> <hr/>
${rickshaw_graph(c.social_networks_graph, 'social_networks',mode='stack')} ${stat_table(c.social_networks, 'Visits')}
<hr/> </div>
<p>Percentage of visits that were referred from these social networks</p> <div class="tab-pane" id="languages">
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> ${rickshaw_graph(c.languages_graph,'languages',mode='stack')}
<h4 class="ga-reports-heading">Show stats table for:</h4> <hr/>
${month_selector(c.month, c.months, c.day)} <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
</form> <h4 class="ga-reports-heading">Show stats table for:</h4>
${stat_table(c.social_networks, 'Visits')} ${month_selector(c.month, c.months, c.day)}
</div> </form>
<div class="tab-pane" id="languages"> <hr/>
${rickshaw_graph(c.languages_graph,'languages',mode='stack')} ${stat_table(c.languages)}
<hr/> </div>
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> <div class="tab-pane" id="country">
<h4 class="ga-reports-heading">Show stats table for:</h4> ${rickshaw_graph(c.country_graph,'country',mode='stack')}
${month_selector(c.month, c.months, c.day)} <hr/>
</form> <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
${stat_table(c.languages)} <h4 class="ga-reports-heading">Show stats table for:</h4>
</div> ${month_selector(c.month, c.months, c.day)}
<div class="tab-pane" id="country"> </form>
${rickshaw_graph(c.country_graph,'country',mode='stack')} <hr/>
<hr/> ${stat_table(c.country)}
<form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get"> </div>
<h4 class="ga-reports-heading">Show stats table for:</h4> </div>
${month_selector(c.month, c.months, c.day)} </div>
</form>  
${stat_table(c.country)}  
</div>  
</div> </div>
</div> </div>
</div> </div>
   
</div> </div>
   
   
<py:def function="optional_footer"> <py:def function="optional_footer">
<script type="text/javascript"> <script type="text/javascript">
$(function() { $(function() {
CKAN.GA_Reports.bind_sparklines(); CKAN.GA_Reports.bind_sparklines();
CKAN.GA_Reports.bind_sidebar(); CKAN.GA_Reports.bind_sidebar();
CKAN.GA_Reports.bind_month_selector(); CKAN.GA_Reports.bind_month_selector();
}); });
</script> </script>
</py:def> </py:def>
   
<xi:include href="../../layout.html" /> <xi:include href="../../layout.html" />
</html> </html>