From: Ross Jones Date: Thu, 25 Oct 2012 14:52:26 +0000 Subject: Change the rules for top datasets X-Git-Url: http://maxious.lambdacomplex.org/git/?p=ckanext-ga-report.git&a=commitdiff&h=63e0fea8902deac8c94761974bf16c809d6d8efd --- Change the rules for top datasets --- --- a/ckanext/ga_report/command.py +++ b/ckanext/ga_report/command.py @@ -66,8 +66,7 @@ And where 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__ @@ -96,6 +95,7 @@ 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 @@ -10,10 +10,10 @@ log = logging.getLogger('ckanext.ga-report') -def _get_month_name(str): +def _get_month_name(strdate): import calendar from time import strptime - d = strptime('2012-10', '%Y-%m') + d = strptime(strdate, '%Y-%m') return '%s %s' % (calendar.month_name[d.tm_mon], d.tm_year) @@ -21,7 +21,7 @@ months = [] vals = model.Session.query(cls.period_name).distinct().all() for m in vals: - months.append( (m[0], _get_month_name(m))) + months.append( (m[0], _get_month_name(m[0]))) return sorted(months, key=operator.itemgetter(0), reverse=True) @@ -34,7 +34,6 @@ filter(GA_Stat.period_name==month).\ order_by('GA_Stat.stat_name, GA_Stat.key').all() - response.headers['Content-disposition'] = 'attachment; filename=dgu_analytics_%s.csv' % (month) response.headers['Content-Type'] = "text/csv; charset=utf-8" writer = csv.writer(response) @@ -58,7 +57,8 @@ entries = model.Session.query(GA_Stat).\ filter(GA_Stat.stat_name=='Totals').\ - filter(GA_Stat.period_name==c.month).all() + filter(GA_Stat.period_name==c.month).\ + order_by('ga_stat.key').all() c.global_totals = [(s.key, s.value) for s in entries ] keys = { @@ -87,6 +87,7 @@ """ 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) @@ -94,13 +95,12 @@ # 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]) - +# and not url like '/publisher/%%' connection = model.Session.connection() q = """ select department_id, sum(pageviews::int) views, sum(visitors::int) visits from ga_url where department_id <> '' - and not url like '/publisher/%%' and period_name=%s group by department_id order by views desc limit 20; """ --- a/ckanext/ga_report/download_analytics.py +++ b/ckanext/ga_report/download_analytics.py @@ -21,8 +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': @@ -37,13 +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: --- a/ckanext/ga_report/plugin.py +++ b/ckanext/ga_report/plugin.py @@ -18,17 +18,7 @@ def after_map(self, map): map.connect( - '/data/analytics', - controller='ckanext.ga_report.controller:GaReport', - action='index' - ) - map.connect( - '/data/analytics_{month}.csv', - controller='ckanext.ga_report.controller:GaReport', - action='csv' - ) - map.connect( - '/data/analytics/publisher/', + '/data/analytics/publisher', controller='ckanext.ga_report.controller:GaPublisherReport', action='index' ) @@ -37,6 +27,16 @@ controller='ckanext.ga_report.controller:GaPublisherReport', action='read' ) + map.connect( + '/data/analytics', + controller='ckanext.ga_report.controller:GaReport', + action='index' + ) + map.connect( + '/data/analytics/data_{month}.csv', + controller='ckanext.ga_report.controller:GaReport', + action='csv' + ) return map --- a/ckanext/ga_report/templates/ga_report/publisher/index.html +++ b/ckanext/ga_report/templates/ga_report/publisher/index.html @@ -8,6 +8,7 @@

  • Publishers

    +

    The table shows the top 20 publishers as recorded by page views of datasets owned by that publisher, and the number of visits to each publisher's home page.

    •  --- a/ckanext/ga_report/templates/ga_report/publisher/read.html +++ b/ckanext/ga_report/templates/ga_report/publisher/read.html @@ -15,9 +15,8 @@ As well as showing the number of views within ${c.month_desc}, it will also show the number of visitors that viewed each dataset.

    -

    The dataset list page for ${c.publisher.title} was viewed ${c.publisher_page_views} times during ${c.month_desc}

    -

    +

    View the publishers leaderboard

    --- a/ckanext/ga_report/tests/test_api.py +++ b/ckanext/ga_report/tests/test_api.py @@ -32,7 +32,7 @@ svc = init_service("token.dat", "credentials.json") downloader = DownloadAnalytics(svc, profile_id=get_profile_id(svc)) try: - downloader.since_date(datetime.datetime.now() - datetime.timedelta(days=-30)) + downloader.for_date(datetime.datetime.now() - datetime.timedelta(days=-30)) except Exception as e: assert False, e