Trimming long float numbers
Trimming long float numbers

file:a/README.rst -> file:b/README.rst
--- a/README.rst
+++ b/README.rst
@@ -26,16 +26,15 @@
 1. Activate you CKAN python environment and install this extension's software::
 
     $ pyenv/bin/activate
-    $ pip install -e  git+https://github.com/okfn/ckanext-ga-report.git#egg=ckanext-ga-report
+    $ pip install -e  git+https://github.com/datagovuk/ckanext-ga-report.git#egg=ckanext-ga-report
 
 2. Ensure you development.ini (or similar) contains the info about your Google Analytics account and configuration::
 
       googleanalytics.id = UA-1010101-1
-      googleanalytics.username = googleaccount@gmail.com
-      googleanalytics.password = googlepassword
+      googleanalytics.account = Account name (i.e. data.gov.uk, see top level item at https://www.google.com/analytics)
       ga-report.period = monthly
 
-   Note that your password will be readable by system administrators on your server. Rather than use sensitive account details, it is suggested you give access to the GA account to a new Google account that you create just for this purpose.
+   Note that your credentials will be readable by system administrators on your server. Rather than use sensitive account details, it is suggested you give access to the GA account to a new Google account that you create just for this purpose.
 
 3. Set up this extension's database tables using a paster command. (Ensure your CKAN pyenv is still activated, run the command from ``src/ckanext-ga-report``, alter the ``--config`` option to point to your site config file)::
 

--- a/ckanext/ga_report/command.py
+++ b/ckanext/ga_report/command.py
@@ -1,7 +1,10 @@
 import logging
+import datetime
 
 from ckan.lib.cli import CkanCommand
-# No other CKAN imports allowed until _load_config is run, or logging is disabled
+# No other CKAN imports allowed until _load_config is run,
+# or logging is disabled
+
 
 class InitDB(CkanCommand):
     """Initialise the extension's database tables
@@ -46,9 +49,10 @@
         assuming it is correct.
         """
         from ga_auth import init_service
-        initialize_service('token.dat',
-                           self.args[0] if self.args
-                                        else 'credentials.json')
+        init_service('token.dat',
+                      self.args[0] if self.args
+                                   else 'credentials.json')
+
 
 class LoadAnalytics(CkanCommand):
     """Get data from Google Analytics API and save it
@@ -62,8 +66,7 @@
     And where <time-period> is:
         all         - data for all time
         latest      - (default) just the 'latest' data
-        YYYY-MM-DD  - just data for all time periods going
-                      back to (and including) this date
+        YYYY-MM     - just data for the specific month
     """
     summary = __doc__.split('\n')[0]
     usage = __doc__
@@ -73,24 +76,26 @@
     def command(self):
         self._load_config()
 
-        from ga_auth import init_service
+        from download_analytics import DownloadAnalytics
+        from ga_auth import (init_service, get_profile_id)
+
         try:
             svc = init_service(self.args[0], None)
         except TypeError:
-            print 'Have you correctly run the getauthtoken task and specified the correct file here'
+            print ('Have you correctly run the getauthtoken task and '
+                   'specified the correct file here')
             return
 
-        from download_analytics import DownloadAnalytics
-        from ga_auth import get_profile_id
         downloader = DownloadAnalytics(svc, profile_id=get_profile_id(svc))
 
-        time_period = self.args[1] if self.args and len(self.args) > 1 else 'latest'
+        time_period = self.args[1] if self.args and len(self.args) > 1 \
+            else 'latest'
         if time_period == 'all':
             downloader.all_()
         elif time_period == 'latest':
             downloader.latest()
         else:
-            since_date = datetime.datetime.strptime(time_period, '%Y-%m-%d')
-            downloader.since_date(since_date)
+            # The month to use
+            for_date = datetime.datetime.strptime(time_period, '%Y-%m')
+            downloader.specific_month(for_date)
 
-

--- a/ckanext/ga_report/controller.py
+++ b/ckanext/ga_report/controller.py
@@ -1,10 +1,154 @@
 import logging
-from ckan.lib.base import BaseController, c, render
-import report_model
+import operator
+from ckan.lib.base import BaseController, c, render, request, response
+
+import sqlalchemy
+from sqlalchemy import func, cast, Integer
+import ckan.model as model
+from ga_model import GA_Url, GA_Stat
 
 log = logging.getLogger('ckanext.ga-report')
 
+
+def _get_month_name(strdate):
+    import calendar
+    from time import strptime
+    d = strptime(strdate, '%Y-%m')
+    return '%s %s' % (calendar.month_name[d.tm_mon], d.tm_year)
+
+
+def _month_details(cls):
+    months = []
+    vals = model.Session.query(cls.period_name).distinct().all()
+    for m in vals:
+        months.append( (m[0], _get_month_name(m[0])))
+    return sorted(months, key=operator.itemgetter(0), reverse=True)
+
+
 class GaReport(BaseController):
+
+    def csv(self, month):
+        import csv
+
+        entries = model.Session.query(GA_Stat).\
+            filter(GA_Stat.period_name==month).\
+            order_by('GA_Stat.stat_name, GA_Stat.key').all()
+
+        response.headers['Content-Type'] = "text/csv; charset=utf-8"
+
+        writer = csv.writer(response)
+        writer.writerow(["Period", "Statistic", "Key", "Value"])
+
+        for entry in entries:
+            writer.writerow([entry.period_name.encode('utf-8'),
+                             entry.stat_name.encode('utf-8'),
+                             entry.key.encode('utf-8'),
+                             entry.value.encode('utf-8')])
+
     def index(self):
-        return render('index.html')
 
+        # Get the month details by fetching distinct values and determining the
+        # month names from the values.
+        c.months = _month_details(GA_Stat)
+
+        # Work out which month to show, based on query params of the first item
+        c.month = request.params.get('month', c.months[0][0] if c.months else '')
+        c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month])
+
+        entries = model.Session.query(GA_Stat).\
+            filter(GA_Stat.stat_name=='Totals').\
+            filter(GA_Stat.period_name==c.month).\
+            order_by('ga_stat.key').all()
+
+        c.global_totals = []
+        for e in entries:
+            val = e.value
+            if e.key in ['Average time on site', 'Pages per visit', 'Percent new visits']:
+                val =  "%.2f" % round(float(e.value), 2)
+                e.key = '%s *' % e.key
+            c.global_totals.append((e.key, val))
+
+
+        keys = {
+            'Browser versions': 'browsers',
+            'Operating Systems versions': 'os',
+            'Social sources': 'social_networks',
+            'Languages': 'languages',
+            'Country': 'country'
+        }
+
+        for k, v in keys.iteritems():
+            entries = model.Session.query(GA_Stat).\
+                filter(GA_Stat.stat_name==k).\
+                filter(GA_Stat.period_name==c.month).\
+                order_by('ga_stat.value::int desc').all()
+            setattr(c, v, [(s.key, s.value) for s in entries ])
+
+
+        return render('ga_report/site/index.html')
+
+
+class GaPublisherReport(BaseController):
+    """
+    Displays the pageview and visit count for specific publishers based on
+    the datasets associated with the publisher.
+    """
+
+    def index(self):
+
+        # Get the month details by fetching distinct values and determining the
+        # month names from the values.
+        c.months = _month_details(GA_Url)
+
+        # Work out which month to show, based on query params of the first item
+        c.month = request.params.get('month', c.months[0][0] if c.months else '')
+        c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month])
+
+        connection = model.Session.connection()
+        q = """
+            select department_id, sum(pageviews::int) views, sum(visitors::int) visits
+            from ga_url
+            where department_id <> ''
+                and period_name=%s
+            group by department_id order by views desc limit 20;
+        """
+        # Add this back (before and period_name =%s) if you want to ignore publisher
+        # homepage views
+        # and not url like '/publisher/%%'
+
+        c.top_publishers = []
+        res = connection.execute(q, c.month)
+        for row in res:
+            c.top_publishers.append((model.Group.get(row[0]), row[1], row[2]))
+
+        return render('ga_report/publisher/index.html')
+
+
+    def read(self, id):
+        c.publisher = model.Group.get(id)
+        c.top_packages = [] # package, dataset_views in c.top_packages
+
+        # Get the month details by fetching distinct values and determining the
+        # month names from the values.
+        c.months = _month_details(GA_Url)
+
+        # Work out which month to show, based on query params of the first item
+        c.month = request.params.get('month', c.months[0][0] if c.months else '')
+        c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month])
+
+        entry = model.Session.query(GA_Url).\
+            filter(GA_Url.url=='/publisher/%s' % c.publisher.name).\
+            filter(GA_Url.period_name==c.month).first()
+        c.publisher_page_views = entry.pageviews if entry else 0
+
+        entries = model.Session.query(GA_Url).\
+            filter(GA_Url.department_id==c.publisher.name).\
+            filter(GA_Url.period_name==c.month).\
+            order_by('ga_url.pageviews::int desc')[:20]
+        for entry in entries:
+            if entry.url.startswith('/dataset/'):
+                p = model.Package.get(entry.url[len('/dataset/'):])
+                c.top_packages.append((p,entry.pageviews,entry.visitors))
+
+        return render('ga_report/publisher/read.html')
+

--- a/ckanext/ga_report/download_analytics.py
+++ b/ckanext/ga_report/download_analytics.py
@@ -1,3 +1,4 @@
+import os
 import logging
 import datetime
 
@@ -20,8 +21,17 @@
         self.profile_id = profile_id
 
 
-    def all_(self):
-        self.since_date(datetime.datetime(2010, 1, 1))
+    def specific_month(self, date):
+        import calendar
+
+        first_of_this_month = datetime.datetime(date.year, date.month, 1)
+        _, 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)
+        periods = ((date.strftime(FORMAT_MONTH),
+                    last_day_of_month,
+                    first_of_this_month, last_of_this_month),)
+        self.download_and_store(periods)
+
 
     def latest(self):
         if self.period == 'monthly':
@@ -36,13 +46,13 @@
         self.download_and_store(periods)
 
 
-    def since_date(self, since_date):
+    def for_date(self, for_date):
         assert isinstance(since_date, datetime.datetime)
         periods = [] # (period_name, period_complete_day, start_date, end_date)
         if self.period == 'monthly':
             first_of_the_months_until_now = []
-            year = since_date.year
-            month = since_date.month
+            year = for_date.year
+            month = for_date.month
             now = datetime.datetime.now()
             first_of_this_month = datetime.datetime(now.year, now.month, 1)
             while True:
@@ -84,22 +94,27 @@
                      self.get_full_period_name(period_name, period_complete_day),
                      start_date.strftime('%Y %m %d'),
                      end_date.strftime('%Y %m %d'))
-            data = self.download(start_date, end_date)
-            log.info('Storing Analytics for period "%s"',
+
+            data = self.download(start_date, end_date, '~/dataset/[a-z0-9-_]+')
+            log.info('Storing Dataset Analytics for period "%s"',
                      self.get_full_period_name(period_name, period_complete_day))
-            self.store(period_name, period_complete_day, data)
-
-
-    def download(self, start_date, end_date):
+            self.store(period_name, period_complete_day, data, )
+
+            data = self.download(start_date, end_date, '~/publisher/[a-z0-9-_]+')
+            log.info('Storing Publisher Analytics for period "%s"',
+                     self.get_full_period_name(period_name, period_complete_day))
+            self.store(period_name, period_complete_day, data,)
+
+            ga_model.update_publisher_stats(period_name) # about 30 seconds.
+            self.sitewide_stats( period_name )
+
+
+    def download(self, start_date, end_date, path='~/dataset/[a-z0-9-_]+'):
         '''Get data from GA for a given time period'''
         start_date = start_date.strftime('%Y-%m-%d')
         end_date = end_date.strftime('%Y-%m-%d')
-        # url
-        #query = 'ga:pagePath=~^%s,ga:pagePath=~^%s' % \
-        #        (PACKAGE_URL, self.resource_url_tag)
-        query = 'ga:pagePath=~^/dataset/'
-        #query = 'ga:pagePath=~^/User/'
-        metrics = 'ga:uniquePageviews'
+        query = 'ga:pagePath=%s$' % path
+        metrics = 'ga:uniquePageviews, ga:visitors'
         sort = '-ga:uniquePageviews'
 
         # Supported query params at
@@ -110,35 +125,184 @@
                                  start_date=start_date,
                                  metrics=metrics,
                                  sort=sort,
-                                 end_date=end_date).execute()
-        self.print_results(results)
-
-#        for entry in GA.ga_query(query_filter=query,
-#                                 from_date=start_date,
-#                                 metrics=metrics,
-#                                 sort=sort,
-#                                 to_date=end_date):
-#            print entry, type(entry)
-#            import pdb; pdb.set_trace()
-#            for dim in entry.dimension:
-#                if dim.name == "ga:pagePath":
-#                    package = dim.value
-#                    count = entry.get_metric(
-#                        'ga:uniquePageviews').value or 0
-#                    packages[package] = int(count)
-        return []
-
-    def print_results(self, results):
-        import pprint
-        pprint.pprint(results)
-        if results:
-            print 'Profile: %s' % results.get('profileInfo').get('profileName')
+                                 dimensions="ga:pagePath",
+                                 max_results=10000,
+                                 end_date=end_date).execute()
+
+        if os.getenv('DEBUG'):
+            import pprint
+            pprint.pprint(results)
             print 'Total results: %s' % results.get('totalResults')
-            print 'Total Visits: %s' % results.get('rows', [[-1]])[0][0]
-        else:
-            print 'No results found'
+
+        packages = []
+        for entry in results.get('rows'):
+            (loc,pageviews,visits) = entry
+            packages.append( ('http:/' + loc, pageviews, visits,) ) # Temporary hack
+        return dict(url=packages)
 
     def store(self, period_name, period_complete_day, data):
         if 'url' in data:
             ga_model.update_url_stats(period_name, period_complete_day, data['url'])
 
+    def sitewide_stats(self, period_name):
+        import calendar
+        year, month = period_name.split('-')
+        _, last_day_of_month = calendar.monthrange(int(year), int(month))
+
+        start_date = '%s-01' % period_name
+        end_date = '%s-%s' % (period_name, last_day_of_month)
+        print 'Sitewide_stats for %s (%s -> %s)' % (period_name, start_date, end_date)
+
+        funcs = ['_totals_stats', '_social_stats', '_os_stats',
+                 '_locale_stats', '_browser_stats', '_mobile_stats']
+        for f in funcs:
+            print ' + Fetching %s stats' % f.split('_')[1]
+            getattr(self, f)(start_date, end_date, period_name)
+
+    def _get_results(result_data, f):
+        data = {}
+        for result in result_data:
+            key = f(result)
+            data[key] = data.get(key,0) + result[1]
+        return data
+
+    def _totals_stats(self, start_date, end_date, period_name):
+        """ Fetches distinct totals, total pageviews etc """
+        results = self.service.data().ga().get(
+                                 ids='ga:' + self.profile_id,
+                                 start_date=start_date,
+                                 metrics='ga:uniquePageviews',
+                                 sort='-ga:uniquePageviews',
+                                 max_results=10000,
+                                 end_date=end_date).execute()
+        result_data = results.get('rows')
+        ga_model.update_sitewide_stats(period_name, "Totals", {'Total pageviews': result_data[0][0]})
+
+        results = self.service.data().ga().get(
+                                 ids='ga:' + self.profile_id,
+                                 start_date=start_date,
+                                 metrics='ga:pageviewsPerVisit,ga:bounces,ga:avgTimeOnSite,ga:percentNewVisits',
+                                 max_results=10000,
+                                 end_date=end_date).execute()
+        result_data = results.get('rows')
+        data = {
+            'Pages per visit': result_data[0][0],
+            'Bounces': result_data[0][1],
+            'Average time on site': result_data[0][2],
+            'Percent new visits': result_data[0][3],
+        }
+        ga_model.update_sitewide_stats(period_name, "Totals", data)
+
+
+    def _locale_stats(self, start_date, end_date, period_name):
+        """ Fetches stats about language and country """
+        results = self.service.data().ga().get(
+                                 ids='ga:' + self.profile_id,
+                                 start_date=start_date,
+                                 metrics='ga:uniquePageviews',
+                                 sort='-ga:uniquePageviews',
+                                 dimensions="ga:language,ga:country",
+                                 max_results=10000,
+                                 end_date=end_date).execute()
+        result_data = results.get('rows')
+        data = {}
+        for result in result_data:
+            data[result[0]] = data.get(result[0], 0) + int(result[2])
+        ga_model.update_sitewide_stats(period_name, "Languages", data)
+
+        data = {}
+        for result in result_data:
+            data[result[1]] = data.get(result[1], 0) + int(result[2])
+        ga_model.update_sitewide_stats(period_name, "Country", data)
+
+
+    def _social_stats(self, start_date, end_date, period_name):
+        """ Finds out which social sites people are referred from """
+        results = self.service.data().ga().get(
+                                 ids='ga:' + self.profile_id,
+                                 start_date=start_date,
+                                 metrics='ga:uniquePageviews',
+                                 sort='-ga:uniquePageviews',
+                                 dimensions="ga:socialNetwork,ga:referralPath",
+                                 max_results=10000,
+                                 end_date=end_date).execute()
+        result_data = results.get('rows')
+        twitter_links = []
+        data = {}
+        for result in result_data:
+            if not result[0] == '(not set)':
+                data[result[0]] = data.get(result[0], 0) + int(result[2])
+                if result[0] == 'Twitter':
+                    twitter_links.append(result[1])
+        ga_model.update_sitewide_stats(period_name, "Social sources", data)
+
+
+    def _os_stats(self, start_date, end_date, period_name):
+        """ Operating system stats """
+        results = self.service.data().ga().get(
+                                 ids='ga:' + self.profile_id,
+                                 start_date=start_date,
+                                 metrics='ga:uniquePageviews',
+                                 sort='-ga:uniquePageviews',
+                                 dimensions="ga:operatingSystem,ga:operatingSystemVersion",
+                                 max_results=10000,
+                                 end_date=end_date).execute()
+        result_data = results.get('rows')
+        data = {}
+        for result in result_data:
+            data[result[0]] = data.get(result[0], 0) + int(result[2])
+        ga_model.update_sitewide_stats(period_name, "Operating Systems", data)
+
+        data = {}
+        for result in result_data:
+            key = "%s (%s)" % (result[0],result[1])
+            data[key] = result[2]
+        ga_model.update_sitewide_stats(period_name, "Operating Systems versions", data)
+
+
+    def _browser_stats(self, start_date, end_date, period_name):
+        """ Information about browsers and browser versions """
+        results = self.service.data().ga().get(
+                                 ids='ga:' + self.profile_id,
+                                 start_date=start_date,
+                                 metrics='ga:uniquePageviews',
+                                 sort='-ga:uniquePageviews',
+                                 dimensions="ga:browser,ga:browserVersion",
+                                 max_results=10000,
+                                 end_date=end_date).execute()
+        result_data = results.get('rows')
+        data = {}
+        for result in result_data:
+            data[result[0]] = data.get(result[0], 0) + int(result[2])
+        ga_model.update_sitewide_stats(period_name, "Browsers", data)
+
+        data = {}
+        for result in result_data:
+            key = "%s (%s)" % (result[0], result[1])
+            data[key] = result[2]
+        ga_model.update_sitewide_stats(period_name, "Browser versions", data)
+
+
+    def _mobile_stats(self, start_date, end_date, period_name):
+        """ Info about mobile devices """
+
+        results = self.service.data().ga().get(
+                                 ids='ga:' + self.profile_id,
+                                 start_date=start_date,
+                                 metrics='ga:uniquePageviews',
+                                 sort='-ga:uniquePageviews',
+                                 dimensions="ga:mobileDeviceBranding, ga:mobileDeviceInfo",
+                                 max_results=10000,
+                                 end_date=end_date).execute()
+
+        result_data = results.get('rows')
+        data = {}
+        for result in result_data:
+            data[result[0]] = data.get(result[0], 0) + int(result[2])
+        ga_model.update_sitewide_stats(period_name, "Mobile brands", data)
+
+        data = {}
+        for result in result_data:
+            data[result[1]] = data.get(result[1], 0) + int(result[2])
+        ga_model.update_sitewide_stats(period_name, "Mobile devices", data)
+

--- a/ckanext/ga_report/ga_auth.py
+++ b/ckanext/ga_report/ga_auth.py
@@ -1,3 +1,4 @@
+import os
 import httplib2
 from apiclient.discovery import build
 from oauth2client.client import flow_from_clientsecrets
@@ -7,12 +8,12 @@
 from pylons import config
 
 
-def _prepare_credentials( token_filename, credentials_filename ):
+def _prepare_credentials(token_filename, credentials_filename):
     """
     Either returns the user's oauth credentials or uses the credentials
     file to generate a token (by forcing the user to login in the browser)
     """
-    storage = Storage( token_filename )
+    storage = Storage(token_filename)
     credentials = storage.get()
 
     if credentials is None or credentials.invalid:
@@ -23,7 +24,8 @@
 
     return credentials
 
-def init_service( token_file, credentials_file ):
+
+def init_service(token_file, credentials_file):
     """
     Given a file containing the user's oauth token (and another with
     credentials in case we need to generate the token) will return a
@@ -40,15 +42,24 @@
 def get_profile_id(service):
     """
     Get the profile ID for this user and the service specified by the
-    'googleanalytics.id' configuration option.
+    'googleanalytics.id' configuration option. This function iterates
+    over all of the accounts available to the user who invoked the
+    service to find one where the account name matches (in case the
+    user has several).
     """
     accounts = service.management().accounts().list().execute()
 
     if not accounts.get('items'):
         return None
 
-    accountId = accounts.get('items')[0].get('id')
+    accountName = config.get('googleanalytics.account')
     webPropertyId = config.get('googleanalytics.id')
+    for acc in accounts.get('items'):
+        if acc.get('name') == accountName:
+            accountId = acc.get('id')
+
+    webproperties = service.management().webproperties().list(accountId=accountId).execute()
+
     profiles = service.management().profiles().list(
         accountId=accountId, webPropertyId=webPropertyId).execute()
 
@@ -56,3 +67,4 @@
         return profiles.get('items')[0].get('id')
 
     return None
+

--- a/ckanext/ga_report/ga_model.py
+++ b/ckanext/ga_report/ga_model.py
@@ -3,30 +3,80 @@
 
 from sqlalchemy import Table, Column, MetaData
 from sqlalchemy import types
-from sqlalchemy.sql import select, text
+from sqlalchemy.sql import select
+from sqlalchemy.orm import mapper
 from sqlalchemy import func
 
 import ckan.model as model
-from ckan.model.types import JsonType
 from ckan.lib.base import *
 
 def make_uuid():
     return unicode(uuid.uuid4())
 
+
+
+class GA_Url(object):
+
+    def __init__(self, **kwargs):
+        for k,v in kwargs.items():
+            setattr(self, k, v)
+
+class GA_Stat(object):
+
+    def __init__(self, **kwargs):
+        for k,v in kwargs.items():
+            setattr(self, k, v)
+
+class GA_Publisher(object):
+
+    def __init__(self, **kwargs):
+        for k,v in kwargs.items():
+            setattr(self, k, v)
+
+
+metadata = MetaData()
+url_table = Table('ga_url', metadata,
+                      Column('id', types.UnicodeText, primary_key=True,
+                             default=make_uuid),
+                      Column('period_name', types.UnicodeText),
+                      Column('period_complete_day', types.Integer),
+                      Column('pageviews', types.UnicodeText),
+                      Column('visitors', types.UnicodeText),
+                      Column('url', types.UnicodeText),
+                      Column('department_id', types.UnicodeText),
+                )
+mapper(GA_Url, url_table)
+
+stat_table = Table('ga_stat', metadata,
+                  Column('id', types.UnicodeText, primary_key=True,
+                         default=make_uuid),
+                  Column('period_name', types.UnicodeText),
+                  Column('stat_name', types.UnicodeText),
+                  Column('key', types.UnicodeText),
+                  Column('value', types.UnicodeText), )
+mapper(GA_Stat, stat_table)
+
+
+pub_table = Table('ga_publisher', metadata,
+                  Column('id', types.UnicodeText, primary_key=True,
+                         default=make_uuid),
+                  Column('period_name', types.UnicodeText),
+                  Column('publisher_name', types.UnicodeText),
+                  Column('views', types.UnicodeText),
+                  Column('visitors', types.UnicodeText),
+                  Column('toplevel', types.Boolean, default=False),
+                  Column('subpublishercount', types.Integer, default=0),
+                  Column('parent', types.UnicodeText),
+)
+mapper(GA_Publisher, pub_table)
+
+
 def init_tables():
-    metadata = MetaData()
-    package_stats = Table('ga_url', metadata,
-                          Column('id', types.UnicodeText, primary_key=True, default=make_uuid),
-                          Column('period_name', types.UnicodeText),
-                          Column('period_complete_day', types.Integer),
-                          Column('visits', types.Integer),
-                          Column('group_id', types.String(60)),
-                          Column('next_page', JsonType),
-                          )
     metadata.create_all(model.meta.engine)
 
 
 cached_tables = {}
+
 
 def get_table(name):
     if name not in cached_tables:
@@ -45,6 +95,7 @@
     '''
     url = re.sub('https?://(www\.)?data.gov.uk', '', url)
     return url
+
 
 def _get_department_id_of_url(url):
     # e.g. /dataset/fuel_prices
@@ -56,36 +107,144 @@
         if dataset:
             publisher_groups = dataset.get_groups('publisher')
             if publisher_groups:
-                return publisher_groups[0].id
+                return publisher_groups[0].name
+    else:
+        publisher_match = re.match('/publisher/([^/]+)(/.*)?', url)
+        if publisher_match:
+            return publisher_match.groups()[0]
+
+
+def update_sitewide_stats(period_name, stat_name, data):
+    for k,v in data.iteritems():
+        item = model.Session.query(GA_Stat).\
+            filter(GA_Stat.period_name==period_name).\
+            filter(GA_Stat.key==k).\
+            filter(GA_Stat.stat_name==stat_name).first()
+        if item:
+            item.period_name = period_name
+            item.key = k
+            item.value = v
+            model.Session.add(item)
+        else:
+            # create the row
+            values = {'id': make_uuid(),
+                     'period_name': period_name,
+                     'key': k,
+                     'value': v,
+                     'stat_name': stat_name
+                     }
+            model.Session.add(GA_Stat(**values))
+        model.Session.commit()
+
+
 
 def update_url_stats(period_name, period_complete_day, url_data):
-    table = get_table('ga_url')
-    connection = model.Session.connection()
-    for url, views, next_page in url_data:
+    for url, views, visitors in url_data:
         url = _normalize_url(url)
         department_id = _get_department_id_of_url(url)
+
         # see if the row for this url & month is in the table already
-        s = select([func.count(id_col)],
-                   table.c.period_name == period_name,
-                   table.c.url == url)
-        count = connection.execute(s).fetchone()
-        if count and count[0]:
-            # update the row
-            connection.execute(table.update()\
-                .where(table.c.period_name == period_name,
-                       table.c.url == url)\
-                .values(period_complete_day=period_complete_day,
-                        views=views,
-                        department_id=department_id,
-                        next_page=next_page))
+        item = model.Session.query(GA_Url).\
+            filter(GA_Url.period_name==period_name).\
+            filter(GA_Url.url==url).first()
+        if item:
+            item.period_name = period_name
+            item.pageviews = views
+            item.visitors = visitors
+            item.department_id = department_id
+            model.Session.add(item)
         else:
             # create the row
-            values = {'period_name': period_name,
+            values = {'id': make_uuid(),
+                      'period_name': period_name,
                       'period_complete_day': period_complete_day,
                       'url': url,
-                      'views': views,
-                      'department_id': department_id,
-                      'next_page': next_page}
-            conn