Added a class for spacing to More link
--- a/README.rst
+++ b/README.rst
@@ -26,7 +26,7 @@
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::
--- a/ckanext/ga_report/command.py
+++ b/ckanext/ga_report/command.py
@@ -66,8 +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__
@@ -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
@@ -1,15 +1,94 @@
import logging
import operator
-from ckan.lib.base import BaseController, c, render,request
+from ckan.lib.base import BaseController, c, render, request, response, abort
+import sqlalchemy
+from sqlalchemy import func, cast, Integer
import ckan.model as model
-from ga_model import GA_Url
+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):
+
+ # 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)
+ if e.key == 'Average time on site':
+ mins, secs = divmod(float(val), 60)
+ hours, mins = divmod(mins, 60)
+ val = '%02d:%02d:%02d (%s seconds) ' % (hours, mins, secs, val)
+ 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')
@@ -19,26 +98,46 @@
the datasets associated with the publisher.
"""
- def _get_month_name(self, str):
- import calendar
- from time import strptime
- d = strptime('2012-10', '%Y-%m')
- return '%s %s' % (calendar.month_name[d.tm_mon], d.tm_year)
+ 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 index(self, id):
+ def read(self, id):
+
c.publisher = model.Group.get(id)
+ if not c.publisher:
+ abort(404, 'A publisher with that name could not be found')
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 = []
- vals = model.Session.query(GA_Url.period_name).distinct().all()
- for m in vals:
- c.months.append( (m[0],self._get_month_name(m)))
-
- # Sort the months, so most recent is at the head of our list
- c.months = sorted(c.months, key=operator.itemgetter(0), reverse=True)
+ 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 '')
@@ -52,11 +151,11 @@
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 desc').all()
+ 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/index.html')
+ return render('ga_report/publisher/read.html')
--- 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:
@@ -85,7 +94,7 @@
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, '~/dataset/[a-z0-9-_]+')
log.info('Storing Dataset Analytics for period "%s"',
self.get_full_period_name(period_name, period_complete_day))
@@ -95,7 +104,7 @@
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 )
--- /dev/null
+++ b/ckanext/ga_report/helpers.py
@@ -1,1 +1,41 @@
+import logging
+import operator
+import ckan.lib.base as base
+import ckan.model as model
+_log = logging.getLogger(__name__)
+
+def most_popular_datasets(publisher, count=20):
+ from ckanext.ga_report.ga_model import GA_Url
+
+ if not publisher:
+ _log.error("No valid publisher passed to 'most_popular_datasets'")
+ return ""
+
+ datasets = {}
+ entries = model.Session.query(GA_Url).\
+ filter(GA_Url.department_id==publisher.name).\
+ filter(GA_Url.url.like('/dataset/%')).\
+ order_by('ga_url.pageviews::int desc')[:count]
+ for entry in entries:
+ p = model.Package.get(entry.url[len('/dataset/'):])
+ 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.visitors)
+
+ results = []
+ for k, v in datasets.iteritems():
+ results.append((k,v['views'],v['visits']))
+
+ results = sorted(results, key=operator.itemgetter(1), reverse=True)
+
+ ctx = {
+ 'dataset_count': len(datasets),
+ 'datasets': results,
+
+ 'publisher': publisher
+ }
+
+ return base.render_snippet('ga_report/publisher/popular.html', **ctx)
+
--- a/ckanext/ga_report/plugin.py
+++ b/ckanext/ga_report/plugin.py
@@ -2,30 +2,49 @@
import ckan.lib.helpers as h
import ckan.plugins as p
from ckan.plugins import implements, toolkit
-#import gasnippet
-#import commands
-#import dbutil
log = logging.getLogger('ckanext.ga-report')
class GAReportPlugin(p.SingletonPlugin):
implements(p.IConfigurer, inherit=True)
implements(p.IRoutes, inherit=True)
+ implements(p.ITemplateHelpers, inherit=True)
def update_config(self, config):
toolkit.add_template_directory(config, 'templates')
toolkit.add_public_directory(config, 'public')
+ def get_helpers(self):
+ """
+ A dictionary of extra helpers that will be available to provide
+ ga report info to templates.
+ """
+ from ckanext.ga_report.helpers import most_popular_datasets
+ return {
+ 'ga_report_installed': lambda: True,
+ 'most_popular_datasets': most_popular_datasets,
+ }
+
def after_map(self, map):
map.connect(
- '/data/analytics/',
- controller='ckanext.ga_report.controller:GaReport',
+ '/data/analytics/publisher',
+ controller='ckanext.ga_report.controller:GaPublisherReport',
action='index'
)
map.connect(
'/data/analytics/publisher/{id}',
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
@@ -3,15 +3,20 @@
xmlns:xi="http://www.w3.org/2001/XInclude"
py:strip="">
- <py:def function="page_title">Analytics for ${g.site_title}</py:def>
+ <py:def function="page_title">Publisher Analytics for ${g.site_title}</py:def>
+
+ <py:match path="primarysidebar">
+ <li class="widget-container boxed widget_text">
+ <h4>Publishers</h4>
+ <p>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.</p>
+ </li>
+ </py:match>
<div py:match="content">
- <h1>Analytics for ${c.publisher.title}</h1>
+ <h1>Publisher Analytics</h1>
+ <h2>The top 20 publishers</h2>
- <h2>Most viewed datasets</h2>
- <p><em>Note: this data does not include API calls</em></p>
-
- <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaPublisherReport',action='index',id=c.publisher.name)}" method="get">
+ <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaPublisherReport',action='index')}" method="get">
<div class="controls">
<select name="month">
<py:for each="val,desc in c.months">
@@ -22,17 +27,15 @@
</div>
</form>
- <p><a href="${h.url_for(controller='ckanext.dgu.controllers.publisher:PublisherController', action='read', id=c.publisher.name)}">${c.publisher.title} dataset list</a> was viewed ${c.publisher_page_views} times in ${c.month_desc}</p>
-
<table class="table table-condensed table-bordered table-striped">
<tr>
- <th>Dataset</th>
- <th>Views</th>
+ <th>Publisher</th>
+ <th>Dataset Views</th>
<th>Visits</th>
</tr>
- <py:for each="package, views, visits in c.top_packages">
+ <py:for each="publisher, views, visits in c.top_publishers">
<tr>
- <td>${h.link_to(package.title or package.name, h.url_for(controller='package', action='read', id=package.name))}
+ <td>${h.link_to(publisher.title, h.url_for(controller='ckanext.ga_report.controller:GaPublisherReport', action='read', id=publisher.name))}
</td>
<td>${views}</td>
<td>${visits}</td>
@@ -44,6 +47,16 @@
</div>
<xi:include href="../../layout.html" />
+
+ <py:def function="optional_footer">
+ <script type='text/javascript'>
+ $('.nav-tabs li a').click(function (e) {
+ e.preventDefault();
+ $(this).tab('show');
+ })
+ </script>
+ </py:def>
+
</html>
--- /dev/null
+++ b/ckanext/ga_report/templates/ga_report/publisher/popular.html
@@ -1,1 +1,25 @@
+<html xmlns:py="http://genshi.edgewall.org/"
+ xmlns:i18n="http://genshi.edgewall.org/i18n"
+ xmlns:xi="http://www.w3.org/2001/XInclude"
+ py:strip="">
+ <py:if test="dataset_count == 0">
+ <p>We do not currently have analytics data for ${publisher.title}</p>
+ </py:if>
+ <py:if test="dataset_count > 0">
+ <div class="popular_datasets">
+ <ul>
+ <py:for each="dataset,views,visits in datasets">
+ <li>
+ ${h.link_to(dataset.title, h.url_for(controller='package', action='read', id=dataset.name))}
+ </li>
+ </py:for>
+ </ul>
+ <p class="">${h.link_to("More analytics for " + publisher.title, h.url_for(controller='ckanext.ga_report.controller:GaPublisherReport',action='read',id=publisher.name))}</p>
+ </div>
+ </py:if>
+</html>
+
+
+
+
--- /dev/null
+++ b/ckanext/ga_report/templates/ga_report/publisher/read.html
@@ -1,1 +1,65 @@
+<html xmlns:py="http://genshi.edgewall.org/"
+ xmlns:i18n="http://genshi.edgewall.org/i18n"
+ xmlns:xi="http://www.w3.org/2001/XInclude"
+ py:strip="">
+ <py:def function="page_title">Analytics for ${g.site_title}</py:def>
+
+ <py:match path="primarysidebar">
+
+ <li class="widget-container boxed widget_text">
+ <h4>${c.publisher.title}</h4>
+ <p>
+ The table shows the top 20 most viewed datasets belonging to ${c.publisher.title}.
+ </p>
+ <p>
+ As well as showing the number of views within ${c.month_desc}, it will also show the
+ number of visitors that viewed each dataset.
+ </p>
+ <p>The dataset list page for <a href="${h.url_for(controller='ckanext.dgu.controllers.publisher:PublisherController', action='read', id=c.publisher.name)}">${c.publisher.title}</a> was viewed ${c.publisher_page_views} times during ${c.month_desc}</p>
+ <p>View the <a href="${h.url_for(controller='ckanext.ga_report.controller:GaPublisherReport', action='index')}">publishers</a> leaderboard</p>
+ </li>
+ </py:match>
+
+ <div py:match="content">
+ <h1>Analytics for ${c.publisher.title}</h1>
+
+ <h2>Top 20 most viewed datasets</h2>
+ <p><em>Note: this data does not include API calls</em></p>
+
+ <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaPublisherReport',action='read',id=c.publisher.name)}" method="get">
+ <div class="controls">
+ <select name="month">
+ <py:for each="val,desc in c.months">
+ <option value='${val}' py:attrs="{'selected': 'selected' if c.month == val else None}">${desc}</option>
+ </py:for>
+ </select>
+ <input class="btn button" type='submit' value="Update"/>
+ </div>
+ </form>
+
+ <table class="table table-condensed table-bordered table-striped">
+ <tr>
+ <th>Dataset</th>
+ <th>Views</th>
+ <th>Visits</th>
+ </tr>
+ <py:for each="package, views, visits in c.top_packages">
+ <tr>
+ <td>${h.link_to(package.title or package.name, h.url_for(controller='package', action='read', id=package.name))}
+ </td>
+ <td>${views}</td>
+ <td>${visits}</td>
+ </tr>
+ </py:for>
+ </table>
+
+
+ </div>
+
+ <xi:include href="../../layout.html" />
+</html>
+
+
+
+
--- a/ckanext/ga_report/templates/ga_report/site/index.html
+++ b/ckanext/ga_report/templates/ga_report/site/index.html
@@ -1,1 +1,153 @@
-HAI Site
+<html xmlns:py="http://genshi.edgewall.org/"
+ xmlns:i18n="http://genshi.edgewall.org/i18n"
+ xmlns:xi="http://www.w3.org/2001/XInclude"
+ py:strip="">
+
+ <py:def function="page_title">Site analytics</py:def>
+
+ <py:match path="primarysidebar">
+ <li class="widget-container boxed widget_text">
+ <h4>Statistics</h4>
+ <p>It is possible to <a href="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='csv',month=c.month)}">export the analytics data</a> as a CSV file, which contains all of the information for ${c.month_desc}</p>
+ </li>
+ <li class="widget-container boxed widget_text">
+ <h4>Publisher statistics</h4>
+ <p>You can view statistics about specific publishers at the <a href="${h.url_for(controller='ckanext.ga_report.controller:GaPublisherReport',action='index')}">Publisher Analytics</a> page</p>
+
+ </li>
+ </py:match>
+
+ <div py:match="content">
+ <h1>Site statistics</h1>
+
+ <form class="form-inline" action="${h.url_for(controller='ckanext.ga_report.controller:GaReport',action='index')}" method="get">
+ <div class="controls">
+ <select name="month">
+ <py:for each="val,desc in c.months">
+ <option value='${val}' py:attrs="{'selected': 'selected' if c.month == val else None}">${desc}</option>
+ </py:for>
+ </select>
+ <input class="btn button" type='submit' value="Update"/>
+ </div>
+ </form>
+
+ <div class="tabbable">
+ <ul class="nav nav-tabs">
+ <li class="active"><a href="#totals" data-toggle="tab">Totals</a></li>
+ <li><a href="#browsers" data-toggle="tab">Browsers</a></li>
+ <li><a href="#os" data-toggle="tab">Operating Systems</a></li>
+ <li><a href="#social_networks" data-toggle="tab">Social Networks</a></li>
+ <li><a href="#languages" data-toggle="tab">Languages</a></li>
+ <li><a href="#country" data-toggle="tab">Country</a></li>
+ </ul>
+ <div class="tab-content">
+ <div class="tab-pane active" id="totals">
+ <table class="table table-condensed table-bordered table-striped">
+ <tr>
+ <th>Name</th>
+ <th>Value</th>
+ </tr>
+ <py:for each="name, value in c.global_totals">
+ <tr>
+ <td>${name}</td>
+ <td>${value}</td>
+ </tr>
+ </py:for>
+ </table>
+
+ <p>* Values are rounded up to 2 decimal places.</p>
+ </div>
+ <div class="tab-pane" id="browsers">
+ <table class="table table-condensed table-bordered table-striped">
+ <tr>
+ <th>Name</th>
+ <th>Value</th>
+ </tr>
+ <py:for each="name, value in c.browsers">
+ <tr>
+ <td>${name}</td>
+ <td>${value}</td>
+ </tr>
+ </py:for>
+ </table>
+ </div>
+ <div class="tab-pane" id="os">
+ <table class="table table-condensed table-bordered table-striped">
+ <tr>
+ <th>Name</th>
+ <th>Value</th>
+ </tr>
+ <py:for each="name, value in c.os">
+ <tr>
+ <td>${name}</td>
+ <td>${value}</td>
+ </tr>
+ </py:for>
+ </table>
+ </div>
+ <div class="tab-pane" id="social_networks">
+ <table class="table table-condensed table-bordered table-striped">
+ <tr>
+ <th>Name</th>
+ <th>Value</th>
+ </tr>
+ <py:for each="name, value in c.social_networks">
+ <tr>
+ <td>${name}</td>
+ <td>${value}</td>
+ </tr>
+ </py:for>
+ </table>
+ </div>
+ <div class="tab-pane" id="languages">
+ <table class="table table-condensed table-bordered table-striped">
+ <tr>
+ <th>Name</th>
+ <th>Value</th>
+ </tr>
+ <py:for each="name, value in c.languages">
+ <tr>
+ <td>${name}</td>
+ <td>${value}</td>
+ </tr>
+ </py:for>
+ </table>
+ </div>
+ <div class="tab-pane" id="country">
+ <table class="table table-condensed table-bordered table-striped">
+ <tr>
+ <th>Name</th>
+ <th>Value</th>
+ </tr>
+ <py:for each="name, value in c.country">
+ <tr>
+ <td>${name}</td>
+ <td>${value}</td>
+ </tr>
+ </py:for>
+ </table>
+ </div>
+
+
+ </div>
+ </div>
+
+
+
+ </div>
+
+ <xi:include href="../../layout.html" />
+
+ <py:def function="optional_footer">
+ <script type='text/javascript'>
+ $('.nav-tabs li a').click(function (e) {
+ e.preventDefault();
+ $(this).tab('show');
+ })
+ </script>
+ </py:def>
+</html>
+
+
+
+
--- 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
--- a/setup.py
+++ b/setup.py
@@ -32,6 +32,7 @@
[paste.paster_command]
loadanalytics = ckanext.ga_report.command:LoadAnalytics
initdb = ckanext.ga_report.command:InitDB
+ getauthtoken = ckanext.ga_report.command:GetAuthToken
""",
)