From: Ross Jones Date: Wed, 07 Nov 2012 14:11:35 +0000 Subject: Handling duplicate urls in ga_url processing X-Git-Url: http://maxious.lambdacomplex.org/git/?p=ckanext-ga-report.git&a=commitdiff&h=6a953b2c1630aabdd8d7ea629b3f06c8c2f6eb47 --- Handling duplicate urls in ga_url processing --- --- a/ckanext/ga_report/controller.py +++ b/ckanext/ga_report/controller.py @@ -9,7 +9,7 @@ import sqlalchemy from sqlalchemy import func, cast, Integer import ckan.model as model -from ga_model import GA_Url, GA_Stat, GA_ReferralStat +from ga_model import GA_Url, GA_Stat, GA_ReferralStat, GA_Publisher log = logging.getLogger('ckanext.ga-report') @@ -229,7 +229,6 @@ c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month]) c.top_publishers = _get_top_publishers() - return render('ga_report/publisher/index.html') def _get_packages(self, publisher=None, count=-1): @@ -247,7 +246,6 @@ q = q.filter(GA_Url.period_name==month) q = q.order_by('ga_url.visitors::int desc') top_packages = [] - for entry,package in q.limit(count): if package: top_packages.append((package, entry.pageviews, entry.visitors)) @@ -289,7 +287,7 @@ else: c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month]) - month = c.mnth or 'All' + month = c.month or 'All' c.publisher_page_views = 0 q = model.Session.query(GA_Url).\ filter(GA_Url.url=='/publisher/%s' % c.publisher_name) @@ -305,33 +303,28 @@ Returns a list of the top 20 publishers by dataset visits. (The number to show can be varied with 'limit') ''' + month = c.month or 'All' connection = model.Session.connection() q = """ select department_id, sum(pageviews::int) views, sum(visitors::int) visits from ga_url - where department_id <> ''""" - if c.month: - q = q + """ - and period_name=%s - """ - q = q + """ - group by department_id order by visits desc + where department_id <> '' + and package_id <> '' + and url like '/dataset/%%' + and period_name=%s + group by department_id order by visits desc """ if limit: q = q + " limit %s;" % (limit) - # Add this back (before and period_name =%s) if you want to ignore publisher - # homepage views - # and not url like '/publisher/%%' - top_publishers = [] - res = connection.execute(q, c.month) - + res = connection.execute(q, month) for row in res: g = model.Group.get(row[0]) if g: top_publishers.append((g, row[1], row[2])) return top_publishers + def _get_publishers(): ''' --- a/ckanext/ga_report/download_analytics.py +++ b/ckanext/ga_report/download_analytics.py @@ -3,7 +3,7 @@ import datetime import collections from pylons import config - +from ga_model import _normalize_url import ga_model #from ga_client import GA @@ -92,35 +92,35 @@ def download_and_store(self, periods): for period_name, period_complete_day, start_date, end_date in periods: - log.info('Period "%s" (%s - %s)', - self.get_full_period_name(period_name, period_complete_day), - start_date.strftime('%Y-%m-%d'), - end_date.strftime('%Y-%m-%d')) - if self.delete_first: - log.info('Deleting existing Analytics for this period "%s"', + log.info('Deleting existing Analytics for period "%s"', period_name) ga_model.delete(period_name) - - log.info('Downloading analytics for dataset views') - data = self.download(start_date, end_date, '~/dataset/[a-z0-9-_]+') - - log.info('Storing dataset views (%i rows)', len(data.get('url'))) + log.info('Downloading Analytics for period "%s" (%s - %s)', + self.get_full_period_name(period_name, period_complete_day), + start_date.strftime('%Y %m %d'), + end_date.strftime('%Y %m %d')) + + # Clean up the entries before we run this + ga_model.pre_update_url_stats(period_name) + + accountName = config.get('googleanalytics.account') + + data = self.download(start_date, end_date, '~/%s/dataset/[a-z0-9-_]+' % accountName) + 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, ) - log.info('Downloading analytics for publisher views') - data = self.download(start_date, end_date, '~/publisher/[a-z0-9-_]+') - log.info('Storing publisher views (%i rows)', len(data.get('url'))) + data = self.download(start_date, end_date, '~/%s/publisher/[a-z0-9-_]+' % accountName) + 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,) - log.info('Aggregating datasets by publisher') ga_model.update_publisher_stats(period_name) # about 30 seconds. - - log.info('Downloading and storing analytics for site-wide stats') self.sitewide_stats( period_name ) - log.info('Downloading and storing analytics for social networks') self.update_social_info(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') @@ -143,12 +143,11 @@ data = collections.defaultdict(list) rows = results.get('rows',[]) for row in rows: - from ga_model import _normalize_url data[_normalize_url(row[0])].append( (row[1], int(row[2]),) ) ga_model.update_social(period_name, data) - def download(self, start_date, end_date, path='~/dataset/[a-z0-9-_]+'): + def download(self, start_date, end_date, path=None): '''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') @@ -171,7 +170,10 @@ packages = [] for entry in results.get('rows'): (loc,pageviews,visits) = entry - packages.append( ('http:/' + loc, pageviews, visits,) ) # Temporary hack + url = _normalize_url('http:/' + loc) + if not url.startswith('/dataset/') and not url.startswith('/publisher/'): + continue + packages.append( (url, pageviews, visits,) ) # Temporary hack return dict(url=packages) def store(self, period_name, period_complete_day, data): @@ -185,10 +187,12 @@ 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: - log.info('Downloading analytics for %s' % f.split('_')[1]) + print ' + Fetching %s stats' % f.split('_')[1] getattr(self, f)(start_date, end_date, period_name) def _get_results(result_data, f): --- a/ckanext/ga_report/ga_model.py +++ b/ckanext/ga_report/ga_model.py @@ -116,7 +116,7 @@ return '/' + '/'.join(url.split('/')[2:]) -def _get_department_id_of_url(url): +def _get_package_and_publisher(url): # e.g. /dataset/fuel_prices # e.g. /dataset/fuel_prices/resource/e63380d4 dataset_match = re.match('/dataset/([^/]+)(/.*)?', url) @@ -126,12 +126,13 @@ if dataset: publisher_groups = dataset.get_groups('publisher') if publisher_groups: - return publisher_groups[0].name + return dataset_ref,publisher_groups[0].name + return dataset_ref, None else: publisher_match = re.match('/publisher/([^/]+)(/.*)?', url) if publisher_match: - return publisher_match.groups()[0] - + return None, publisher_match.groups()[0] + return None, None def update_sitewide_stats(period_name, stat_name, data): for k,v in data.iteritems(): @@ -156,46 +157,9 @@ model.Session.commit() - -def update_url_stats(period_name, period_complete_day, url_data): - for url, views, visitors in url_data: - url = _normalize_url(url) - department_id = _get_department_id_of_url(url) - - package = None - if url.startswith('/dataset/'): - package = url[len('/dataset/'):] - - # see if the row for this url & month is in the table already - 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 - item.package_id = package - model.Session.add(item) - else: - # create the row - values = {'id': make_uuid(), - 'period_name': period_name, - 'period_complete_day': period_complete_day, - 'url': url, - 'pageviews': views, - 'visitors': visitors, - 'department_id': department_id, - 'package_id': package - } - model.Session.add(GA_Url(**values)) - - # We now need to recaculate the ALL time_period from the data we have - # Delete the old 'All' - old = model.Session.query(GA_Url).\ - filter(GA_Url.period_name == "All").\ - filter(GA_Url.url==url).delete() - +def update_url_stat_totals(period_name): + + """ items = model.Session.query(GA_Url).\ filter(GA_Url.period_name != "All").\ filter(GA_Url.url==url).all() @@ -209,8 +173,60 @@ 'package_id': package } model.Session.add(GA_Url(**values)) - model.Session.commit() + """ + +def pre_update_url_stats(period_name): + model.Session.query(GA_Url).\ + filter(GA_Url.period_name==period_name).delete() + model.Session.query(GA_Url).\ + filter(GA_Url.period_name=='All').delete() + + +def update_url_stats(period_name, period_complete_day, url_data): + + for url, views, visitors in url_data: + package, publisher = _get_package_and_publisher(url) + + item = model.Session.query(GA_Url).\ + filter(GA_Url.period_name==period_name).\ + filter(GA_Url.url==url).first() + if item: + item.pageviews = item.pageviews + views + item.visitors = item.visitors + visitors + model.Session.add(item) + else: + values = {'id': make_uuid(), + 'period_name': period_name, + 'period_complete_day': period_complete_day, + 'url': url, + 'pageviews': views, + 'visitors': visitors, + 'department_id': publisher, + 'package_id': package + } + model.Session.add(GA_Url(**values)) + model.Session.commit() + + if package: + entries = model.Session.query(GA_Url).\ + filter(GA_Url.period_name!='All').\ + filter(GA_Url.url==url).all() + values = {'id': make_uuid(), + 'period_name': 'All', + 'period_complete_day': 0, + 'url': url, + 'pageviews': sum([int(e.pageviews) for e in entries]), + 'visitors': sum([int(e.visitors) for e in entries]), + 'department_id': publisher, + 'package_id': package + } + + model.Session.add(GA_Url(**values)) + model.Session.commit() + + + def update_social(period_name, data): --- a/ckanext/ga_report/templates/ga_report/notes.html +++ b/ckanext/ga_report/templates/ga_report/notes.html @@ -6,11 +6,11 @@

  • Notes

    • --- a/ckanext/ga_report/templates/ga_report/publisher/index.html +++ b/ckanext/ga_report/templates/ga_report/publisher/index.html @@ -41,15 +41,15 @@ + - + -
    PublisherDataset Visits Dataset ViewsDataset Visits
    ${h.link_to(publisher.title, h.url_for(controller='ckanext.ga_report.controller:GaDatasetReport', action='read_publisher', id=publisher.name))} ${visits} ${views}${visits}
    --- a/ckanext/ga_report/templates/ga_report/publisher/read.html +++ b/ckanext/ga_report/templates/ga_report/publisher/read.html @@ -47,15 +47,15 @@ + - + -
    DatasetVisits ViewsVisits
    ${h.link_to(package.title or package.name, h.url_for(controller='package', action='read', id=package.name))} ${visits} ${views}${visits}