From: Tom Rees Date: Thu, 17 Jan 2013 00:20:53 +0000 Subject: Done integrating graphs onto site_usage/publishers and site_usage/datasets. Including some interesting queries. X-Git-Url: http://maxious.lambdacomplex.org/git/?p=ckanext-ga-report.git&a=commitdiff&h=5277d6bac60f0eb0cf88a550b77b15e3110dfba9 --- Done integrating graphs onto site_usage/publishers and site_usage/datasets. Including some interesting queries. --- --- a/ckanext/ga_report/controller.py +++ b/ckanext/ga_report/controller.py @@ -113,24 +113,24 @@ return key, val # Query historic values for sparkline rendering - graph_query = model.Session.query(GA_Stat)\ + sparkline_query = model.Session.query(GA_Stat)\ .filter(GA_Stat.stat_name=='Totals')\ .order_by(GA_Stat.period_name) - graph_data = {} - for x in graph_query: - graph_data[x.key] = graph_data.get(x.key,[]) + sparkline_data = {} + for x in sparkline_query: + sparkline_data[x.key] = sparkline_data.get(x.key,[]) key, val = clean_key(x.key,float(x.value)) tooltip = '%s: %s' % (_get_month_name(x.period_name), val) - graph_data[x.key].append( (tooltip,x.value) ) + sparkline_data[x.key].append( (tooltip,x.value) ) # Trim the latest month, as it looks like a huge dropoff - for key in graph_data: - graph_data[key] = graph_data[key][:-1] + for key in sparkline_data: + sparkline_data[key] = sparkline_data[key][:-1] c.global_totals = [] if c.month: for e in entries: key, val = clean_key(e.key, e.value) - sparkline = graph_data[e.key] + sparkline = sparkline_data[e.key] c.global_totals.append((key, val, sparkline)) else: d = collections.defaultdict(list) @@ -141,11 +141,18 @@ v = sum(v) else: v = float(sum(v))/float(len(v)) - sparkline = graph_data[k] + sparkline = sparkline_data[k] key, val = clean_key(k,v) c.global_totals.append((key, val, sparkline)) - c.global_totals = sorted(c.global_totals, key=operator.itemgetter(0)) + # Sort the global totals into a more pleasant order + def sort_func(x): + key = x[0] + total_order = ['Total page views','Total visits','Pages per visit'] + if key in total_order: + return total_order.index(key) + return 999 + c.global_totals = sorted(c.global_totals, key=sort_func) keys = { 'Browser versions': 'browser_versions', @@ -246,7 +253,9 @@ writer = csv.writer(response) writer.writerow(["Publisher Title", "Publisher Name", "Views", "Visits", "Period Name"]) - for publisher,view,visit in _get_top_publishers(None): + top_publishers, top_publishers_graph = _get_top_publishers(None) + + for publisher,view,visit in top_publishers: writer.writerow([publisher.title.encode('utf-8'), publisher.name.encode('utf-8'), view, @@ -295,7 +304,9 @@ if c.month: c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month]) - c.top_publishers = _get_top_publishers() + c.top_publishers, graph_data = _get_top_publishers() + c.top_publishers_graph = json.dumps( _to_rickshaw(graph_data.values()) ) + return render('ga_report/publisher/index.html') def _get_packages(self, publisher=None, count=-1): @@ -327,8 +338,9 @@ filter(GA_Stat.key==package.name) if month != 'All': # Fetch everything unless the month is specific dls = dls.filter(GA_Stat.period_name==month) - - downloads = sum(int(d.value) for d in dls.all()) + downloads = 0 + for x in dls: + downloads += int(x.value) else: downloads = 'No data' top_packages.append((package, entry.pageviews, entry.visits, downloads)) @@ -379,7 +391,40 @@ c.top_packages = self._get_packages(c.publisher, 20) + # Graph query + top_package_names = [ x[0].name for x in c.top_packages ] + graph_query = model.Session.query(GA_Url,model.Package)\ + .filter(model.Package.name==GA_Url.package_id)\ + .filter(GA_Url.url.like('/dataset/%'))\ + .filter(GA_Url.package_id.in_(top_package_names)) + graph_data = {} + for entry,package in graph_query: + if not package: continue + if entry.period_name=='All': continue + graph_data[package.id] = graph_data.get(package.id,{ + 'name':package.title, + 'data':[] + }) + graph_data[package.id]['data'].append({ + 'x':_get_unix_epoch(entry.period_name), + 'y':int(entry.pageviews), + }) + + c.graph_data = json.dumps( _to_rickshaw(graph_data.values()) ) + return render('ga_report/publisher/read.html') + +def _to_rickshaw(data): + num_points = [] + for package in data: + package['data'] = sorted( package['data'], key=lambda x:x['x'] ) + num_points.append( len(package['data']) ) + if len(set(num_points))>1: + example = num_points[ num_points.index(max(num_points)) ] + for package in data: + while len(package['data']) -
-
-
-
+
+
+
+
+ + + +
@@ -34,6 +42,7 @@
+ ${rickshaw_graph(c.top_publishers_graph,'publishers')} --- a/ckanext/ga_report/templates/ga_report/publisher/read.html +++ b/ckanext/ga_report/templates/ga_report/publisher/read.html @@ -6,6 +6,15 @@ Usage by Dataset + + + + + + + + +
  • @@ -41,21 +50,26 @@

    ${c.publisher.title}

    No page views in this period

    -
    • Publisher
    - - - - - - - - - - - + + + ${rickshaw_graph(c.graph_data,'dataset-downloads',debug=True)} +
    DatasetViewsDownloads
    ${h.link_to(package.title or package.name, h.url_for(controller='package', action='read', id=package.name))} - ${views}${downloads}
    + + + + + + + + + + + -
    DatasetViewsDownloads
    + ${h.link_to(package.title or package.name, h.url_for(controller='package', action='read', id=package.name))} + ${views}${downloads}
    + +
    --- a/ckanext/ga_report/templates/ga_report/site/index.html +++ b/ckanext/ga_report/templates/ga_report/site/index.html @@ -9,44 +9,11 @@ + - @@ -160,13 +127,8 @@ ${rickshaw_graph(c.country_graph,'country')} ${stat_table(c.country)}
    - - - - -