From: Ross Jones Date: Thu, 13 Jun 2013 09:02:37 +0000 Subject: [601] Make sure only active datasets are shown in popular datasets on publisher homepage X-Git-Url: https://maxious.lambdacomplex.org/git/?p=ckanext-ga-report.git&a=commitdiff&h=e1cf709a3d29cfdac40e8dcee1d67e703c4e70fa --- [601] Make sure only active datasets are shown in popular datasets on publisher homepage --- --- a/.gitignore +++ b/.gitignore @@ -1,6 +1,7 @@ *.py[co] *.py~ .gitignore +ckan.log # Packages *.egg --- a/ckanext/ga_report/command.py +++ b/ckanext/ga_report/command.py @@ -23,7 +23,7 @@ import ckan.model as model model.Session.remove() model.Session.configure(bind=model.meta.engine) - log = logging.getLogger('ckanext.ga-report') + log = logging.getLogger('ckanext.ga_report') import ga_model ga_model.init_tables() @@ -55,6 +55,36 @@ init_service('token.dat', self.args[0] if self.args else 'credentials.json') + +class FixTimePeriods(CkanCommand): + """ + Fixes the 'All' records for GA_Urls + + It is possible that older urls that haven't recently been visited + do not have All records. This command will traverse through those + records and generate valid All records for them. + """ + summary = __doc__.split('\n')[0] + usage = __doc__ + max_args = 0 + min_args = 0 + + def __init__(self, name): + super(FixTimePeriods, self).__init__(name) + + def command(self): + import ckan.model as model + from ga_model import post_update_url_stats + self._load_config() + model.Session.remove() + model.Session.configure(bind=model.meta.engine) + + log = logging.getLogger('ckanext.ga_report') + + log.info("Updating 'All' records for old URLs") + post_update_url_stats() + log.info("Processing complete") + class LoadAnalytics(CkanCommand): --- a/ckanext/ga_report/controller.py +++ b/ckanext/ga_report/controller.py @@ -1,6 +1,7 @@ import re import csv import sys +import json import logging import operator import collections @@ -13,6 +14,7 @@ log = logging.getLogger('ckanext.ga-report') +DOWNLOADS_AVAILABLE_FROM = '2012-12' def _get_month_name(strdate): import calendar @@ -20,8 +22,12 @@ d = strptime(strdate, '%Y-%m') return '%s %s' % (calendar.month_name[d.tm_mon], d.tm_year) - -def _month_details(cls): +def _get_unix_epoch(strdate): + from time import strptime,mktime + d = strptime(strdate, '%Y-%m') + return int(mktime(d)) + +def _month_details(cls, stat_key=None): ''' Returns a list of all the periods for which we have data, unfortunately knows too much about the type of the cls being passed as GA_Url has a @@ -32,9 +38,13 @@ months = [] day = None - vals = model.Session.query(cls.period_name,cls.period_complete_day)\ - .filter(cls.period_name!='All').distinct(cls.period_name)\ - .order_by("period_name desc").all() + q = model.Session.query(cls.period_name,cls.period_complete_day)\ + .filter(cls.period_name!='All').distinct(cls.period_name) + if stat_key: + q= q.filter(cls.stat_name==stat_key) + + vals = q.order_by("period_name desc").all() + if vals and vals[0][1]: day = int(vals[0][1]) ordinal = 'th' if 11 <= day <= 13 \ @@ -52,7 +62,7 @@ def csv(self, month): import csv - q = model.Session.query(GA_Stat) + q = model.Session.query(GA_Stat).filter(GA_Stat.stat_name!='Downloads') if month != 'all': q = q.filter(GA_Stat.period_name==month) entries = q.order_by('GA_Stat.period_name, GA_Stat.stat_name, GA_Stat.key').all() @@ -68,6 +78,7 @@ entry.stat_name.encode('utf-8'), entry.key.encode('utf-8'), entry.value.encode('utf-8')]) + def index(self): @@ -101,11 +112,26 @@ return key, val + # Query historic values for sparkline rendering + sparkline_query = model.Session.query(GA_Stat)\ + .filter(GA_Stat.stat_name=='Totals')\ + .order_by(GA_Stat.period_name) + 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) + sparkline_data[x.key].append( (tooltip,x.value) ) + # Trim the latest month, as it looks like a huge dropoff + 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) - c.global_totals.append((key, val)) + sparkline = sparkline_data[e.key] + c.global_totals.append((key, val, sparkline)) else: d = collections.defaultdict(list) for e in entries: @@ -114,11 +140,19 @@ if k in ['Total page views', 'Total visits']: v = sum(v) else: - v = float(sum(v))/len(v) + v = float(sum(v))/float(len(v)) + sparkline = sparkline_data[k] key, val = clean_key(k,v) - c.global_totals.append((key, val)) - c.global_totals = sorted(c.global_totals, key=operator.itemgetter(0)) + c.global_totals.append((key, val, sparkline)) + # 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', @@ -155,12 +189,13 @@ for k, v in keys.iteritems(): q = model.Session.query(GA_Stat).\ - filter(GA_Stat.stat_name==k) + filter(GA_Stat.stat_name==k).\ + order_by(GA_Stat.period_name) + # Buffer the tabular data if c.month: entries = [] q = q.filter(GA_Stat.period_name==c.month).\ order_by('ga_stat.value::int desc') - d = collections.defaultdict(int) for e in q.all(): d[e.key] += int(e.value) @@ -169,10 +204,27 @@ entries.append((key,val,)) entries = sorted(entries, key=operator.itemgetter(1), reverse=True) + # Run a query on all months to gather graph data + graph_query = model.Session.query(GA_Stat).\ + filter(GA_Stat.stat_name==k).\ + order_by(GA_Stat.period_name) + graph_dict = {} + for stat in graph_query: + graph_dict[ stat.key ] = graph_dict.get(stat.key,{ + 'name':stat.key, + 'raw': {} + }) + graph_dict[ stat.key ]['raw'][stat.period_name] = float(stat.value) + stats_in_table = [x[0] for x in entries] + stats_not_in_table = set(graph_dict.keys()) - set(stats_in_table) + stats = stats_in_table + sorted(list(stats_not_in_table)) + graph = [graph_dict[x] for x in stats] + setattr(c, v+'_graph', json.dumps( _to_rickshaw(graph,percentageMode=True) )) + # Get the total for each set of values and then set the value as # a percentage of the total if k == 'Social sources': - total = sum([x for n,x in c.global_totals if n == 'Total visits']) + total = sum([x for n,x,graph in c.global_totals if n == 'Total visits']) else: total = sum([num for _,num in entries]) setattr(c, v, [(k,_percent(v,total)) for k,v in entries ]) @@ -197,7 +249,9 @@ 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 = _get_top_publishers(limit=None) + + for publisher,view,visit in top_publishers: writer.writerow([publisher.title.encode('utf-8'), publisher.name.encode('utf-8'), view, @@ -217,19 +271,20 @@ if not c.publisher: abort(404, 'A publisher with that name could not be found') - packages = self._get_packages(c.publisher) + packages = self._get_packages(publisher=c.publisher, month=c.month) response.headers['Content-Type'] = "text/csv; charset=utf-8" response.headers['Content-Disposition'] = \ str('attachment; filename=datasets_%s_%s.csv' % (c.publisher_name, month,)) writer = csv.writer(response) - writer.writerow(["Dataset Title", "Dataset Name", "Views", "Visits", "Period Name"]) - - for package,view,visit in packages: + writer.writerow(["Dataset Title", "Dataset Name", "Views", "Visits", "Resource downloads", "Period Name"]) + + for package,view,visit,downloads in packages: writer.writerow([package.title.encode('utf-8'), package.name.encode('utf-8'), view, visit, + downloads, month]) def publishers(self): @@ -246,14 +301,17 @@ c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month]) c.top_publishers = _get_top_publishers() + graph_data = _get_top_publishers_graph() + c.top_publishers_graph = json.dumps( _to_rickshaw(graph_data) ) + return render('ga_report/publisher/index.html') - def _get_packages(self, publisher=None, count=-1): + def _get_packages(self, publisher=None, month='', count=-1): '''Returns the datasets in order of views''' - if count == -1: - count = sys.maxint - - month = c.month or 'All' + have_download_data = True + month = month or 'All' + if month != 'All': + have_download_data = month >= DOWNLOADS_AVAILABLE_FROM q = model.Session.query(GA_Url,model.Package)\ .filter(model.Package.name==GA_Url.package_id)\ @@ -263,9 +321,26 @@ q = q.filter(GA_Url.period_name==month) q = q.order_by('ga_url.pageviews::int desc') top_packages = [] - for entry,package in q.limit(count): + if count == -1: + entries = q.all() + else: + entries = q.limit(count) + + for entry,package in entries: if package: - top_packages.append((package, entry.pageviews, entry.visits)) + # Downloads .... + if have_download_data: + dls = model.Session.query(GA_Stat).\ + filter(GA_Stat.stat_name=='Downloads').\ + 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 = 0 + for x in dls: + downloads += int(x.value) + else: + downloads = 'No data' + top_packages.append((package, entry.pageviews, entry.visits, downloads)) else: log.warning('Could not find package associated package') @@ -311,9 +386,73 @@ entry = q.filter(GA_Url.period_name==c.month).first() c.publisher_page_views = entry.pageviews if entry else 0 - c.top_packages = self._get_packages(c.publisher, 20) + c.top_packages = self._get_packages(publisher=c.publisher, count=20, month=c.month) + + # Graph query + top_packages_all_time = self._get_packages(publisher=c.publisher, count=20, month='All') + top_package_names = [ x[0].name for x in top_packages_all_time ] + 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)) + all_series = {} + for entry,package in graph_query: + if not package: continue + if entry.period_name=='All': continue + all_series[package.name] = all_series.get(package.name,{ + 'name':package.title, + 'raw': {} + }) + all_series[package.name]['raw'][entry.period_name] = int(entry.pageviews) + graph = [ all_series[series_name] for series_name in top_package_names ] + c.graph_data = json.dumps( _to_rickshaw(graph) ) return render('ga_report/publisher/read.html') + +def _to_rickshaw(data, percentageMode=False): + if data==[]: + return data + # x-axis is every month in c.months. Note that data might not exist + # for entire history, eg. for recently-added datasets + x_axis = [x[0] for x in c.months] + x_axis.reverse() # Ascending order + x_axis = x_axis[:-1] # Remove latest month + totals = {} + for series in data: + series['data'] = [] + for x_string in x_axis: + x = _get_unix_epoch( x_string ) + y = series['raw'].get(x_string,0) + series['data'].append({'x':x,'y':y}) + totals[x] = totals.get(x,0)+y + if not percentageMode: + return data + # Turn all data into percentages + # Roll insignificant series into a catch-all + THRESHOLD = 1 + raw_data = data + data = [] + for series in raw_data: + for point in series['data']: + percentage = (100*float(point['y'])) / totals[point['x']] + if not (series in data) and percentage>THRESHOLD: + data.append(series) + point['y'] = percentage + others = [ x for x in raw_data if not (x in data) ] + if len(others): + data_other = [] + for i in range(len(x_axis)): + x = _get_unix_epoch(x_axis[i]) + y = 0 + for series in others: + y += series['data'][i]['y'] + data_other.append({'x':x,'y':y}) + data.append({ + 'name':'Other', + 'data': data_other + }) + return data + def _get_top_publishers(limit=20): ''' @@ -343,6 +482,46 @@ return top_publishers +def _get_top_publishers_graph(limit=20): + ''' + Returns a list of the top 20 publishers by dataset visits. + (The number to show can be varied with 'limit') + ''' + connection = model.Session.connection() + q = """ + select department_id, sum(pageviews::int) views + from ga_url + where department_id <> '' + and package_id <> '' + and url like '/dataset/%%' + and period_name='All' + group by department_id order by views desc + """ + if limit: + q = q + " limit %s;" % (limit) + + res = connection.execute(q) + department_ids = [ row[0] for row in res ] + + # Query for a history graph of these department ids + q = model.Session.query( + GA_Url.department_id, + GA_Url.period_name, + func.sum(cast(GA_Url.pageviews,sqlalchemy.types.INT)))\ + .filter( GA_Url.department_id.in_(department_ids) )\ + .filter( GA_Url.url.like('/dataset/%') )\ + .filter( GA_Url.package_id!='' )\ + .group_by( GA_Url.department_id, GA_Url.period_name ) + graph_dict = {} + for dept_id,period_name,views in q: + graph_dict[dept_id] = graph_dict.get( dept_id, { + 'name' : model.Group.get(dept_id).title, + 'raw' : {} + }) + graph_dict[dept_id]['raw'][period_name] = views + return [ graph_dict[id] for id in department_ids ] + + def _get_publishers(): ''' Returns a list of all publishers. Each item is a tuple: --- a/ckanext/ga_report/download_analytics.py +++ b/ckanext/ga_report/download_analytics.py @@ -13,6 +13,7 @@ FORMAT_MONTH = '%Y-%m' MIN_VIEWS = 50 MIN_VISITS = 20 +MIN_DOWNLOADS = 10 class DownloadAnalytics(object): '''Downloads and stores analytics info''' @@ -31,6 +32,11 @@ 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) + # if this is the latest month, note that it is only up until today + now = datetime.datetime.now() + if now.year == date.year and now.month == date.month: + last_day_of_month = now.day + last_of_this_month = now periods = ((date.strftime(FORMAT_MONTH), last_day_of_month, first_of_this_month, last_of_this_month),) @@ -122,8 +128,12 @@ log.info('Storing publisher views (%i rows)', len(data.get('url'))) self.store(period_name, period_complete_day, data,) - log.info('Aggregating datasets by publisher') + # Make sure the All records are correct. + ga_model.post_update_url_stats() + + log.info('Associating datasets with their 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, period_complete_day ) @@ -179,6 +189,7 @@ end_date=end_date).execute() packages = [] + log.info("There are %d results" % results['totalResults']) for entry in results.get('rows'): (loc,pageviews,visits) = entry url = _normalize_url('http:/' + loc) # strips off domain e.g. www.data.gov.uk or data.gov.uk @@ -203,7 +214,7 @@ start_date = '%s-01' % period_name end_date = '%s-%s' % (period_name, last_day_of_month) funcs = ['_totals_stats', '_social_stats', '_os_stats', - '_locale_stats', '_browser_stats', '_mobile_stats'] + '_locale_stats', '_browser_stats', '_mobile_stats', '_download_stats'] for f in funcs: log.info('Downloading analytics for %s' % f.split('_')[1]) getattr(self, f)(start_date, end_date, period_name, period_complete_day) @@ -250,7 +261,7 @@ ids='ga:' + self.profile_id, filters='ga:pagePath==%s' % (path,), start_date=start_date, - metrics='ga:bounces,ga:pageviews', + metrics='ga:visitBounceRate', dimensions='ga:pagePath', max_results=10000, end_date=end_date).execute() @@ -260,10 +271,10 @@ path, result_data) return results = result_data[0] - bounces, total = [float(x) for x in result_data[0][1:]] - pct = 100 * bounces/total - log.info('%d bounces from %d total == %s', bounces, total, pct) - ga_model.update_sitewide_stats(period_name, "Totals", {'Bounce rate (home page)': pct}, + bounces = float(results[1]) + # visitBounceRate is already a % + log.info('Google reports visitBounceRate as %s', bounces) + ga_model.update_sitewide_stats(period_name, "Totals", {'Bounce rate (home page)': float(bounces)}, period_complete_day) @@ -290,6 +301,74 @@ self._filter_out_long_tail(data, MIN_VIEWS) 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): + """ Fetches stats about data downloads """ + import ckan.model as model + + data = {} + + results = self.service.data().ga().get( + ids='ga:' + self.profile_id, + start_date=start_date, + filters='ga:eventAction==download', + metrics='ga:totalEvents', + sort='-ga:totalEvents', + dimensions="ga:eventLabel", + max_results=10000, + end_date=end_date).execute() + result_data = results.get('rows') + if not result_data: + # We may not have data for this time period, so we need to bail + # early. + log.info("There is no download data for this time period") + return + + def process_result_data(result_data, cached=False): + progress_total = len(result_data) + progress_count = 0 + resources_not_matched = [] + for result in result_data: + progress_count += 1 + if progress_count % 100 == 0: + log.debug('.. %d/%d done so far', progress_count, progress_total) + + url = result[0].strip() + + # Get package id associated with the resource that has this URL. + q = model.Session.query(model.Resource) + if cached: + r = q.filter(model.Resource.cache_url.like("%s%%" % url)).first() + else: + r = q.filter(model.Resource.url.like("%s%%" % url)).first() + + package_name = r.resource_group.package.name if r else "" + if package_name: + data[package_name] = data.get(package_name, 0) + int(result[1]) + else: + resources_not_matched.append(url) + continue + if resources_not_matched: + 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]) + + log.info('Associating downloads of resource URLs with their respective datasets') + process_result_data(results.get('rows')) + + results = self.service.data().ga().get( + ids='ga:' + self.profile_id, + start_date=start_date, + filters='ga:eventAction==download-cache', + metrics='ga:totalEvents', + sort='-ga:totalEvents', + dimensions="ga:eventLabel", + max_results=10000, + end_date=end_date).execute() + log.info('Associating downloads of cache resource URLs with their respective datasets') + process_result_data(results.get('rows'), cached=False) + + self._filter_out_long_tail(data, MIN_DOWNLOADS) + 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): """ Finds out which social sites people are referred from """ --- a/ckanext/ga_report/ga_model.py +++ b/ckanext/ga_report/ga_model.py @@ -161,21 +161,64 @@ def pre_update_url_stats(period_name): - log.debug("Deleting '%s' records" % period_name) - model.Session.query(GA_Url).\ - filter(GA_Url.period_name==period_name).delete() - - count = model.Session.query(GA_Url).\ - filter(GA_Url.period_name == 'All').count() - log.debug("Deleting %d 'All' records" % count) - model.Session.query(GA_Url).\ - filter(GA_Url.period_name == 'All').delete() - log.debug("Query: %s" % model.Session.query(GA_Url). - filter(GA_Url.period_name == 'All')) + q = model.Session.query(GA_Url).\ + filter(GA_Url.period_name==period_name) + log.debug("Deleting %d '%s' records" % (q.count(), period_name)) + q.delete() + + q = model.Session.query(GA_Url).\ + filter(GA_Url.period_name == 'All') + log.debug("Deleting %d 'All' records..." % q.count()) + q.delete() model.Session.flush() model.Session.commit() model.repo.commit_and_remove() + log.debug('...done') + +def post_update_url_stats(): + + """ Check the distinct url field in ga_url and make sure + it has an All record. If not then create one. + + After running this then every URL should have an All + record regardless of whether the URL has an entry for + the month being currently processed. + """ + log.debug('Post-processing "All" records...') + query = """select url, pageviews::int, visits::int + from ga_url + where url not in (select url from ga_url where period_name ='All')""" + connection = model.Session.connection() + res = connection.execute(query) + + views, visits = {}, {} + # url, views, visits + for row in res: + views[row[0]] = views.get(row[0], 0) + row[1] + visits[row[0]] = visits.get(row[0], 0) + row[2] + + progress_total = len(views.keys()) + progress_count = 0 + for key in views.keys(): + progress_count += 1 + if progress_count % 100 == 0: + log.debug('.. %d/%d done so far', progress_count, progress_total) + + package, publisher = _get_package_and_publisher(key) + + values = {'id': make_uuid(), + 'period_name': "All", + 'period_complete_day': 0, + 'url': key, + 'pageviews': views[key], + 'visits': visits[key], + 'department_id': publisher, + 'package_id': package + } + model.Session.add(GA_Url(**values)) + model.Session.commit() + log.debug('..done') def update_url_stats(period_name, period_complete_day, url_data): @@ -184,9 +227,14 @@ stores them in GA_Url under the period and recalculates the totals for the 'All' period. ''' + progress_total = len(url_data) + progress_count = 0 for url, views, visits in url_data: + progress_count += 1 + if progress_count % 100 == 0: + log.debug('.. %d/%d done so far', progress_count, progress_total) + package, publisher = _get_package_and_publisher(url) - item = model.Session.query(GA_Url).\ filter(GA_Url.period_name==period_name).\ --- a/ckanext/ga_report/helpers.py +++ b/ckanext/ga_report/helpers.py @@ -80,7 +80,7 @@ return base.render_snippet('ga_report/ga_popular_single.html', **context) -def most_popular_datasets(publisher, count=20): +def most_popular_datasets(publisher, count=20, preview_image=None): if not publisher: _log.error("No valid publisher passed to 'most_popular_datasets'") @@ -92,7 +92,8 @@ 'dataset_count': len(results), 'datasets': results, - 'publisher': publisher + 'publisher': publisher, + 'preview_image': preview_image } return base.render_snippet('ga_report/publisher/popular.html', **ctx) @@ -106,8 +107,18 @@ for entry in entries: if len(datasets) < count: p = model.Package.get(entry.url[len('/dataset/'):]) + + if not p: + _log.warning("Could not find Package for {url}".format(url=entry.url)) + continue + + if not p.state == 'active': + _log.warning("Package {0} is not active, it is {1}".format(p.name, p.state)) + continue + if not p in datasets: datasets[p] = {'views':0, 'visits': 0} + datasets[p]['views'] = datasets[p]['views'] + int(entry.pageviews) datasets[p]['visits'] = datasets[p]['visits'] + int(entry.visits) @@ -117,3 +128,17 @@ return sorted(results, key=operator.itemgetter(1), reverse=True) +def month_option_title(month_iso, months, day): + month_isos = [ iso_code for (iso_code,name) in months ] + try: + index = month_isos.index(month_iso) + except ValueError: + _log.error('Month "%s" not found in list of months.' % month_iso) + return month_iso + month_name = months[index][1] + if index==0: + return month_name + (' (up to %s)'%day) + return month_name + + + --- a/ckanext/ga_report/plugin.py +++ b/ckanext/ga_report/plugin.py @@ -5,7 +5,8 @@ from ckanext.ga_report.helpers import (most_popular_datasets, popular_datasets, - single_popular_dataset) + single_popular_dataset, + month_option_title) log = logging.getLogger('ckanext.ga-report') @@ -27,7 +28,8 @@ 'ga_report_installed': lambda: True, 'popular_datasets': popular_datasets, 'most_popular_datasets': most_popular_datasets, - 'single_popular_dataset': single_popular_dataset + 'single_popular_dataset': single_popular_dataset, + 'month_option_title': month_option_title } def after_map(self, map): @@ -41,6 +43,16 @@ '/data/site-usage/data_{month}.csv', controller='ckanext.ga_report.controller:GaReport', action='csv' + ) + map.connect( + '/data/site-usage/downloads', + controller='ckanext.ga_report.controller:GaReport', + action='downloads' + ) + map.connect( + '/data/site-usage/downloads_{month}.csv', + controller='ckanext.ga_report.controller:GaReport', + action='csv_downloads' ) # GaDatasetReport --- /dev/null +++ b/ckanext/ga_report/public/css/ga_report.css @@ -1,1 +1,69 @@ +.table-condensed td.sparkline-cell { + padding: 1px 0 0 0; + width: 108px; + text-align: center; + /* Hack to hide the momentary flash of text + * before sparklines are fully rendered */ + font-size: 1px; + color: transparent; + overflow: hidden; +} +.rickshaw_chart_container { + position: relative; + height: 350px; + margin: 0 auto 20px auto; +} +.rickshaw_chart { + position: absolute; + left: 40px; + width: 500px; + top: 0; + bottom: 0; +} +.rickshaw_legend { + background: transparent; + width: 100%; + padding-top: 4px; +} +.rickshaw_y_axis { + position: absolute; + top: 0; + bottom: 0; + width: 40px; +} +.rickshaw_legend .label { + background: transparent !important; + color: #000000 !important; + font-weight: normal !important; +} +.rickshaw_legend .instructions { + color: #000; + margin-bottom: 6px; +} +.rickshaw_legend .line .action { + display: none; +} +.rickshaw_legend .line .swatch { + display: block; + float: left; +} +.rickshaw_legend .line .label { + display: block; + white-space: normal; + float: left; + width: 200px; +} +.rickshaw_legend .line .label:hover { + text-decoration: underline; +} + +.ga-reports-table .td-numeric { + text-align: center; +} +.ga-reports-heading { + padding-right: 10px; + margin-top: 4px; + float: left; +} + --- /dev/null +++ b/ckanext/ga_report/public/scripts/ckanext_ga_reports.js @@ -1,1 +1,132 @@ +var CKAN = CKAN || {}; +CKAN.GA_Reports = {}; +CKAN.GA_Reports.render_rickshaw = function( css_name, data, mode, colorscheme ) { + var graphLegends = $('#graph-legend-container'); + + function renderError(alertClass,alertText,legendText) { + $("#chart_"+css_name) + .html( '
'+alertText+'
') + .closest('.rickshaw_chart_container').css('height',50); + var myLegend = $('
') + .html(legendText) + .appendTo(graphLegends); + } + + if (!Modernizr.svg) { + renderError('','Your browser does not support vector graphics. No graphs can be rendered.','(Graph cannot be rendered)'); + return; + } + if (data.length==0) { + renderError('alert-info','There is not enough data to render a graph.','(No graph available)'); + return + } + var myLegend = $('
').appendTo(graphLegends); + + var palette = new Rickshaw.Color.Palette( { scheme: colorscheme } ); + $.each(data, function(i, object) { + object['color'] = palette.color(); + }); + // Rickshaw renders the legend in reverse order... + data.reverse(); + + var graphElement = document.querySelector("#chart_"+css_name); + + var graph = new Rickshaw.Graph( { + element: document.querySelector("#chart_"+css_name), + renderer: mode, + series: data , + height: 328 + }); + var x_axis = new Rickshaw.Graph.Axis.Time( { + graph: graph + } ); + var y_axis = new Rickshaw.Graph.Axis.Y( { + graph: graph, + orientation: 'left', + tickFormat: Rickshaw.Fixtures.Number.formatKMBT, + element: document.getElementById('y_axis_'+css_name) + } ); + var legend = new Rickshaw.Graph.Legend( { + element: document.querySelector('#legend_'+css_name), + graph: graph + } ); + var shelving = new Rickshaw.Graph.Behavior.Series.Toggle( { + graph: graph, + legend: legend + } ); + myLegend.prepend('
Click on a series below to isolate its graph:
'); + graph.render(); +}; + +CKAN.GA_Reports.bind_sparklines = function() { + /* + * Bind to the 'totals' tab being on screen, when the + * Sparkline graphs should be drawn. + * Note that they cannot be drawn sooner. + */ + var created = false; + $('a[href="#totals"]').on( + 'shown', + function() { + if (!created) { + var sparkOptions = { + enableTagOptions: true, + type: 'line', + width: 100, + height: 26, + chartRangeMin: 0, + spotColor: '', + maxSpotColor: '', + minSpotColor: '', + highlightSpotColor: '#000000', + lineColor: '#3F8E6D', + fillColor: '#B7E66B' + }; + $('.sparkline').sparkline('html',sparkOptions); + created = true; + } + $.sparkline_display_visible(); + } + ); +}; + +CKAN.GA_Reports.bind_sidebar = function() { + /* + * Bind to changes in the tab behaviour: + * Show the correct rickshaw graph in the sidebar. + * Not to be called before all graphs load. + */ + $('a[data-toggle="hashtab"]').on( + 'shown', + function(e) { + var href = $(e.target).attr('href'); + var pane = $(href); + if (!pane.length) { console.err('bad href',href); return; } + var legend_name = "none"; + var graph = pane.find('.rickshaw_chart'); + if (graph.length) { + legend_name = graph.attr('id').replace('chart_',''); + } + legend_name = '#legend_'+legend_name; + $('#graph-legend-container > *').hide(); + $('#graph-legend-container .instructions').show(); + $(legend_name).show(); + } + ); + /* The first tab might already have been shown */ + $('li.active > a[data-toggle="hashtab"]').trigger('shown'); +}; + +CKAN.GA_Reports.bind_month_selector = function() { + var handler = function(e) { + var target = $(e.delegateTarget); + var form = target.closest('form'); + var url = form.attr('action')+'?month='+target.val()+window.location.hash; + window.location = url; + }; + var selectors = $('select[name="month"]'); + assert(selectors.length>0); + selectors.bind('change', handler); +}; + --- /dev/null +++ b/ckanext/ga_report/public/scripts/rickshaw_ie7_shim.js @@ -1,1 +1,109 @@ +/* + * Collection of shims to allow d3 and Rickshaw to load, error-free + * (but ultimately unusable) on Internet Explorer 7. The browser's + * API lacks several crucial functions which these libraries depend + * upon to load; we try to hide these errors from the user. + * + * With thanks to Array functions from: + * http://stackoverflow.com/questions/2790001/fixing-javascript-array-functions-in-internet-explorer-indexof-foreach-etc + * + * Use (Modernizr.svg==true) to detect whether it's okay to draw a graph. + */ +'use strict'; +window.Element = window.Element || {'prototype': {}}; +window.CSSStyleDeclaration = window.CSSStyleDeclaration || {'prototype':{}}; + +// Add ECMA262-5 method binding if not supported natively +// +if (!('bind' in Function.prototype)) { + Function.prototype.bind= function(owner) { + var that= this; + if (arguments.length<=1) { + return function() { + return that.apply(owner, arguments); + }; + } else { + var args= Array.prototype.slice.call(arguments, 1); + return function() { + return that.apply(owner, arguments.length===0? args : args.concat(Array.prototype.slice.call(arguments))); + }; + } + }; +} + +// Add ECMA262-5 string trim if not supported natively +// +if (!('trim' in String.prototype)) { + String.prototype.trim= function() { + return this.replace(/^\s+/, '').replace(/\s+$/, ''); + }; +} + +// Add ECMA262-5 Array methods if not supported natively +// +if (!('indexOf' in Array.prototype)) { + Array.prototype.indexOf= function(find, i /*opt*/) { + if (i===undefined) i= 0; + if (i<0) i+= this.length; + if (i<0) i= 0; + for (var n= this.length; ithis.length-1) i= this.length-1; + for (i++; i-->0;) /* i++ because from-argument is sadly inclusive */ + if (i in this && this[i]===find) + return i; + return -1; + }; +} +if (!('forEach' in Array.prototype)) { + Array.prototype.forEach= function(action, that /*opt*/) { + for (var i= 0, n= this.length; id&&(c=b,d=e);return c}function u(a){return a.reduce(v,0)}function v(a,b){return a+b[1]}function w(a,b){return x(a,Math.ceil(Math.log(b.length)/Math.LN2+1))}function x(a,b){var c=-1,d=+a[0],e=(a[1]-d)/b,f=[];while(++c<=b)f[c]=e*c+d;return f}function y(a){return[d3.min(a),d3.max(a)]}function z(a,b){return a.sort=d3.rebind(a,b.sort),a.children=d3.rebind(a,b.children),a.links=D,a.value=d3.rebind(a,b.value),a.nodes=function(b){return E=!0,(a.nodes=a)(b)},a}function A(a){return a.children}function B(a){return a.value}function C(a,b){return b.value-a.value}function D(a){return d3.merge(a.map(function(a){return(a.children||[]).map(function(b){return{source:a,target:b}})}))}function F(a,b){return a.value-b.value}function G(a,b){var c=a._pack_next;a._pack_next=b,b._pack_prev=a,b._pack_next=c,c._pack_prev=b}function H(a,b){a._pack_next=b,b._pack_prev=a}function I(a,b){var c=b.x-a.x,d=b.y-a.y,e=a.r+b.r;return e*e-c*c-d*d>.001}function J(a){function l(a){b=Math.min(a.x-a.r,b),c=Math.max(a.x+a.r,c),d=Math.min(a.y-a.r,d),e=Math.max(a.y+a.r,e)}var b=Infinity,c=-Infinity,d=Infinity,e=-Infinity,f=a.length,g,h,i,j,k;a.forEach(K),g=a[0],g.x=-g.r,g.y=0,l(g);if(f>1){h=a[1],h.x=h.r,h.y=0,l(h);if(f>2){i=a[2],O(g,h,i),l(i),G(g,i),g._pack_prev=i,G(i,h),h=g._pack_next;for(var m=3;m0?(H(g,j),h=j,m--):(H(j,h),g=j,m--)}}}var q=(b+c)/2,r=(d+e)/2,s=0;for(var m=0;m0&&(a=d)}return a}function X(a,b){return a.x-b.x}function Y(a,b){return b.x-a.x}function Z(a,b){return a.depth-b.depth}function $(a,b){function c(a,d){var e=a.children;if(e&&(i=e.length)){var f,g=null,h=-1,i;while(++h=0)f=d[e]._tree,f.prelim+=b,f.mod+=b,b+=f.shift+(c+=f.change)}function ba(a,b,c){a=a._tree,b=b._tree;var d=c/(b.number-a.number);a.change+=d,b.change-=d,b.shift+=c,b.prelim+=c,b.mod+=c}function bb(a,b,c){return a._tree.ancestor.parent==b.parent?a._tree.ancestor:c}function bc(a){return{x:a.x,y:a.y,dx:a.dx,dy:a.dy}}function bd(a,b){var c=a.x+b[3],d=a.y+b[0],e=a.dx-b[1]-b[3],f=a.dy-b[0]-b[2];return e<0&&(c+=e/2,e=0),f<0&&(d+=f/2,f=0),{x:c,y:d,dx:e,dy:f}}d3.layout={},d3.layout.bundle=function(){return function(b){var c=[],d=-1,e=b.length;while(++de&&(e=h),d.push(h)}for(g=0;g=i[0]&&o<=i[1]&&(k=g[d3.bisect(j,o,1,m)-1],k.y+=n,k.push(e[f]));return g}var a=!0,b=Number,c=y,d=w;return e.value=function(a){return arguments.length?(b=a,e):b},e.range=function(a){return arguments.length?(c=d3.functor(a),e):c},e.bins=function(a){return arguments.length?(d=typeof a=="number"?function(b){return x(b,a)}:d3.functor(a),e):d},e.frequency=function(b){return arguments.length?(a=!!b,e):a},e},d3.layout.hierarchy=function(){function e(f,h,i){var j=b.call(g,f,h),k=E?f:{data:f};k.depth=h,i.push(k);if(j&&(m=j.length)){var l=-1,m,n=k.children=[],o=0,p=h+1;while(++l0&&(ba(bb(g,a,d),a,m),i+=m,j+=m),k+=g._tree.mod,i+=e._tree.mod,l+=h._tree.mod,j+=f._tree.mod;g&&!V(f)&&(f._tree.thread=g,f._tree.mod+=k-j),e&&!U(h)&&(h._tree.thread=e,h._tree.mod+=i-l,d=a)}return d}var f=a.call(this,d,e),g=f[0];$(g,function(a,b){a._tree={ancestor:a,prelim:0,mod:0,change:0,shift:0,number:b?b._tree.number+1:0}}),h(g),i(g,-g._tree.prelim);var k=W(g,Y),l=W(g,X),m=W(g,Z),n=k.x-b(k,l)/2,o=l.x+b(l,k)/2,p=m.depth||1;return $(g,function(a){a.x=(a.x-n)/(o-n)*c[0],a.y=a.depth/p*c[1],delete a._tree}),f}var a=d3.layout.hierarchy().sort(null).value(null),b=T,c=[1,1];return d.separation=function(a){return arguments.length?(b=a,d):b},d.size=function(a){return arguments.length?(c=a,d):c},z(d,a)},d3.layout.treemap=function(){function i(a,b){var c=-1,d=a.length,e,f;while(++c0)d.push(g=f[o-1]),d.area+=g.area,(k=l(d,n))<=h?(f.pop(),h=k):(d.area-=d.pop().area,m(d,n,c,!1),n=Math.min(c.dx,c.dy),d.length=d.area=0,h=Infinity);d.length&&(m(d,n,c,!0),d.length=d.area=0),b.forEach(j)}}function k(a){var b=a.children;if(b&&b.length){var c=e(a),d=b.slice(),f,g=[];i(d,c.dx*c.dy/a.value),g.area=0;while(f=d.pop())g.push(f),g.area+=f.area,f.z!=null&&(m(g,f.z?c.dx:c.dy,c,!d.length),g.length=g.area=0);b.forEach(k)}}function l(a,b){var c=a.area,d,e=0,f=Infinity,g=-1,i=a.length;while(++ge&&(e=d)}return c*=c,b*=b,c?Math.max(b*e*h/c,c/(b*f*h)):Infinity}function m(a,c,d,e){var f=-1,g=a.length,h=d.x,i=d.y,j=c?b(a.area/c):0,k;if(c==d.dx){if(e||j>d.dy)j=j?d.dy:0;while(++fd.dx)j=j?d.dx:0;while(++f= 0 ? value.substring(i) : (i = value.length, ""), t = []; + while (i > 0) t.push(value.substring(i -= 3, i + 3)); + return t.reverse().join(",") + f; + } + function d3_formatPrefix(d, i) { + var k = Math.pow(10, Math.abs(8 - i) * 3); + return { + scale: i > 8 ? function(d) { + return d / k; + } : function(d) { + return d * k; + }, + symbol: d + }; + } + function d3_ease_clamp(f) { + return function(t) { + return t <= 0 ? 0 : t >= 1 ? 1 : f(t); + }; + } + function d3_ease_reverse(f) { + return function(t) { + return 1 - f(1 - t); + }; + } + function d3_ease_reflect(f) { + return function(t) { + return .5 * (t < .5 ? f(2 * t) : 2 - f(2 - 2 * t)); + }; + } + function d3_ease_identity(t) { + return t; + } + function d3_ease_poly(e) { + return function(t) { + return Math.pow(t, e); + }; + } + function d3_ease_sin(t) { + return 1 - Math.cos(t * Math.PI / 2); + } + function d3_ease_exp(t) { + return Math.pow(2, 10 * (t - 1)); + } + function d3_ease_circle(t) { + return 1 - Math.sqrt(1 - t * t); + } + function d3_ease_elastic(a, p) { + var s; + if (arguments.length < 2) p = .45; + if (arguments.length < 1) { + a = 1; + s = p / 4; + } else s = p / (2 * Math.PI) * Math.asin(1 / a); + return function(t) { + return 1 + a * Math.pow(2, 10 * -t) * Math.sin((t - s) * 2 * Math.PI / p); + }; + } + function d3_ease_back(s) { + if (!s) s = 1.70158; + return function(t) { + return t * t * ((s + 1) * t - s); + }; + } + function d3_ease_bounce(t) { + return t < 1 / 2.75 ? 7.5625 * t * t : t < 2 / 2.75 ? 7.5625 * (t -= 1.5 / 2.75) * t + .75 : t < 2.5 / 2.75 ? 7.5625 * (t -= 2.25 / 2.75) * t + .9375 : 7.5625 * (t -= 2.625 / 2.75) * t + .984375; + } + function d3_eventCancel() { + d3.event.stopPropagation(); + d3.event.preventDefault(); + } + function d3_eventSource() { + var e = d3.event, s; + while (s = e.sourceEvent) e = s; + return e; + } + function d3_eventDispatch(target) { + var dispatch = new d3_dispatch, i = 0, n = arguments.length; + while (++i < n) dispatch[arguments[i]] = d3_dispatch_event(dispatch); + dispatch.of = function(thiz, argumentz) { + return function(e1) { + try { + var e0 = e1.sourceEvent = d3.event; + e1.target = target; + d3.event = e1; + dispatch[e1.type].apply(thiz, argumentz); + } finally { + d3.event = e0; + } + }; + }; + return dispatch; + } + function d3_transform(m) { + var r0 = [ m.a, m.b ], r1 = [ m.c, m.d ], kx = d3_transformNormalize(r0), kz = d3_transformDot(r0, r1), ky = d3_transformNormalize(d3_transformCombine(r1, r0, -kz)) || 0; + if (r0[0] * r1[1] < r1[0] * r0[1]) { + r0[0] *= -1; + r0[1] *= -1; + kx *= -1; + kz *= -1; + } + this.rotate = (kx ? Math.atan2(r0[1], r0[0]) : Math.atan2(-r1[0], r1[1])) * d3_transformDegrees; + this.translate = [ m.e, m.f ]; + this.scale = [ kx, ky ]; + this.skew = ky ? Math.atan2(kz, ky) * d3_transformDegrees : 0; + } + function d3_transformDot(a, b) { + return a[0] * b[0] + a[1] * b[1]; + } + function d3_transformNormalize(a) { + var k = Math.sqrt(d3_transformDot(a, a)); + if (k) { + a[0] /= k; + a[1] /= k; + } + return k; + } + function d3_transformCombine(a, b, k) { + a[0] += k * b[0]; + a[1] += k * b[1]; + return a; + } + function d3_interpolateByName(name) { + return name == "transform" ? d3.interpolateTransform : d3.interpolate; + } + function d3_uninterpolateNumber(a, b) { + b = b - (a = +a) ? 1 / (b - a) : 0; + return function(x) { + return (x - a) * b; + }; + } + function d3_uninterpolateClamp(a, b) { + b = b - (a = +a) ? 1 / (b - a) : 0; + return function(x) { + return Math.max(0, Math.min(1, (x - a) * b)); + }; + } + function d3_Color() {} + function d3_rgb(r, g, b) { + return new d3_Rgb(r, g, b); + } + function d3_Rgb(r, g, b) { + this.r = r; + this.g = g; + this.b = b; + } + function d3_rgb_hex(v) { + return v < 16 ? "0" + Math.max(0, v).toString(16) : Math.min(255, v).toString(16); + } + function d3_rgb_parse(format, rgb, hsl) { + var r = 0, g = 0, b = 0, m1, m2, name; + m1 = /([a-z]+)\((.*)\)/i.exec(format); + if (m1) { + m2 = m1[2].split(","); + switch (m1[1]) { + case "hsl": + { + return hsl(parseFloat(m2[0]), parseFloat(m2[1]) / 100, parseFloat(m2[2]) / 100); + } + case "rgb": + { + return rgb(d3_rgb_parseNumber(m2[0]), d3_rgb_parseNumber(m2[1]), d3_rgb_parseNumber(m2[2])); + } + } + } + if (name = d3_rgb_names.get(format)) return rgb(name.r, name.g, name.b); + if (format != null && format.charAt(0) === "#") { + if (format.length === 4) { + r = format.charAt(1); + r += r; + g = format.charAt(2); + g += g; + b = format.charAt(3); + b += b; + } else if (format.length === 7) { + r = format.substring(1, 3);