Packages up the data from Analytics.
- Makes sure we get the right account by requiring a new googleanalytics.account
config option which will be used to look up the correct account details.
- Changed the model to use the sqlalchemy orm and made sure it does the updates correctly.
- Still need to collect further info (such as the group details)
--- a/README.rst
+++ b/README.rst
@@ -31,6 +31,7 @@
2. Ensure you development.ini (or similar) contains the info about your Google Analytics account and configuration::
googleanalytics.id = UA-1010101-1
+ googleanalytics.account = Account name (i.e. data.gov.uk, see top level item at https://www.google.com/analytics)
googleanalytics.username = googleaccount@gmail.com
googleanalytics.password = googlepassword
ga-report.period = monthly
@@ -72,7 +73,7 @@
Once you have set up your credentials.json file you can generate an oauth token file by using the
following command, which will store your oauth token in a file called token.dat once you have finished
-giving permission in the browser.
+giving permission in the browser::
$ paster getauthtoken --config=../ckan/development.ini
@@ -84,6 +85,15 @@
$ paster loadanalytics token.dat latest --config=../ckan/development.ini
+The value after the token file is how much data you want to retrieve, this can be
+
+* **all** - data for all time (since 2010)
+
+* **latest** - (default) just the 'latest' data
+
+* **YYYY-MM-DD** - just data for all time periods going back to (and including) this date
+
+
Software Licence
================
--- a/ckanext/ga_report/command.py
+++ b/ckanext/ga_report/command.py
@@ -1,7 +1,10 @@
import logging
+import datetime
from ckan.lib.cli import CkanCommand
-# No other CKAN imports allowed until _load_config is run, or logging is disabled
+# No other CKAN imports allowed until _load_config is run,
+# or logging is disabled
+
class InitDB(CkanCommand):
"""Initialise the extension's database tables
@@ -26,6 +29,12 @@
class GetAuthToken(CkanCommand):
""" Get's the Google auth token
+
+ Usage: paster getauthtoken <credentials_file>
+
+ Where <credentials_file> is the file name containing the details
+ for the service (obtained from https://code.google.com/apis/console).
+ By default this is set to credentials.json
"""
summary = __doc__.split('\n')[0]
usage = __doc__
@@ -33,10 +42,17 @@
min_args = 0
def command(self):
- from ga_auth import initialize_service
- initialize_service('token.dat',
- self.args[0] if self.args
- else 'credentials.json')
+ """
+ In this case we don't want a valid service, but rather just to
+ force the user through the auth flow. We allow this to complete to
+ act as a form of verification instead of just getting the token and
+ assuming it is correct.
+ """
+ from ga_auth import init_service
+ init_service('token.dat',
+ self.args[0] if self.args
+ else 'credentials.json')
+
class LoadAnalytics(CkanCommand):
"""Get data from Google Analytics API and save it
@@ -61,18 +77,20 @@
def command(self):
self._load_config()
- from ga_auth import initialize_service
+ from download_analytics import DownloadAnalytics
+ from ga_auth import (init_service, get_profile_id)
+
try:
- svc = initialize_service(self.args[0], None)
+ svc = init_service(self.args[0], None)
except TypeError:
- print 'Have you correctly run the getauthtoken task and specified the correct file here'
+ print ('Have you correctly run the getauthtoken task and '
+ 'specified the correct file here')
return
- from download_analytics import DownloadAnalytics
- from ga_auth import get_profile_id
downloader = DownloadAnalytics(svc, profile_id=get_profile_id(svc))
- time_period = self.args[1] if self.args and len(self.args) > 1 else 'latest'
+ time_period = self.args[1] if self.args and len(self.args) > 1 \
+ else 'latest'
if time_period == 'all':
downloader.all_()
elif time_period == 'latest':
@@ -81,4 +99,3 @@
since_date = datetime.datetime.strptime(time_period, '%Y-%m-%d')
downloader.since_date(since_date)
-
--- a/ckanext/ga_report/download_analytics.py
+++ b/ckanext/ga_report/download_analytics.py
@@ -97,8 +97,7 @@
# url
#query = 'ga:pagePath=~^%s,ga:pagePath=~^%s' % \
# (PACKAGE_URL, self.resource_url_tag)
- query = 'ga:pagePath=~^/dataset/'
- #query = 'ga:pagePath=~^/User/'
+ query = 'ga:pagePath=~/dataset/[a-z0-9-]+$'
metrics = 'ga:uniquePageviews'
sort = '-ga:uniquePageviews'
@@ -110,30 +109,25 @@
start_date=start_date,
metrics=metrics,
sort=sort,
+ dimensions="ga:pagePath",
end_date=end_date).execute()
- self.print_results(results)
-# for entry in GA.ga_query(query_filter=query,
-# from_date=start_date,
-# metrics=metrics,
-# sort=sort,
-# to_date=end_date):
-# print entry, type(entry)
-# import pdb; pdb.set_trace()
-# for dim in entry.dimension:
-# if dim.name == "ga:pagePath":
-# package = dim.value
-# count = entry.get_metric(
-# 'ga:uniquePageviews').value or 0
-# packages[package] = int(count)
- return []
+ import pprint
+ pprint.pprint(results)
+ print 'Total results: %s' % results.get('totalResults')
+
+ packages = []
+ for entry in results.get('rows'):
+ (loc,size,) = entry
+ packages.append( ('http:/' + loc,size, '',) )
+ return dict(url=packages)
def print_results(self, results):
import pprint
pprint.pprint(results)
if results:
print 'Profile: %s' % results.get('profileInfo').get('profileName')
- print 'Total results: %s' % results.get('totalResults')
+
print 'Total Visits: %s' % results.get('rows', [[-1]])[0][0]
else:
print 'No results found'
--- a/ckanext/ga_report/ga_auth.py
+++ b/ckanext/ga_report/ga_auth.py
@@ -1,3 +1,4 @@
+import os
import httplib2
from apiclient.discovery import build
from oauth2client.client import flow_from_clientsecrets
@@ -7,8 +8,12 @@
from pylons import config
-def _prepare_credentials( token_filename, credentials_filename ):
- storage = Storage( token_filename )
+def _prepare_credentials(token_filename, credentials_filename):
+ """
+ Either returns the user's oauth credentials or uses the credentials
+ file to generate a token (by forcing the user to login in the browser)
+ """
+ storage = Storage(token_filename)
credentials = storage.get()
if credentials is None or credentials.invalid:
@@ -19,7 +24,13 @@
return credentials
-def initialize_service( token_file, credentials_file ):
+
+def init_service(token_file, credentials_file):
+ """
+ Given a file containing the user's oauth token (and another with
+ credentials in case we need to generate the token) will return a
+ service object representing the analytics API.
+ """
http = httplib2.Http()
credentials = _prepare_credentials(token_file, credentials_file)
@@ -27,19 +38,33 @@
return build('analytics', 'v3', http=http)
+
def get_profile_id(service):
- # Get a list of all Google Analytics accounts for this user
+ """
+ Get the profile ID for this user and the service specified by the
+ 'googleanalytics.id' configuration option. This function iterates
+ over all of the accounts available to the user who invoked the
+ service to find one where the account name matches (in case the
+ user has several).
+ """
accounts = service.management().accounts().list().execute()
- if accounts.get('items'):
- firstAccountId = accounts.get('items')[0].get('id')
- webPropertyId = config.get('googleanalytics.id')
- profiles = service.management().profiles().list(
- accountId=firstAccountId,
- webPropertyId=webPropertyId).execute()
+ if not accounts.get('items'):
+ return None
- if profiles.get('items'):
- # return the first Profile ID
- return profiles.get('items')[0].get('id')
+ accountName = config.get('googleanalytics.account')
+ webPropertyId = config.get('googleanalytics.id')
+ for acc in accounts.get('items'):
+ if acc.get('name') == accountName:
+ accountId = acc.get('id')
+
+ webproperties = service.management().webproperties().list(accountId=accountId).execute()
+
+ profiles = service.management().profiles().list(
+ accountId=accountId, webPropertyId=webPropertyId).execute()
+
+ if profiles.get('items'):
+ return profiles.get('items')[0].get('id')
return None
+
--- a/ckanext/ga_report/ga_model.py
+++ b/ckanext/ga_report/ga_model.py
@@ -3,30 +3,44 @@
from sqlalchemy import Table, Column, MetaData
from sqlalchemy import types
-from sqlalchemy.sql import select, text
+from sqlalchemy.sql import select
+from sqlalchemy.orm import mapper
from sqlalchemy import func
import ckan.model as model
-from ckan.model.types import JsonType
from ckan.lib.base import *
def make_uuid():
return unicode(uuid.uuid4())
+
+
+class GA_Url(object):
+
+ def __init__(self, **kwargs):
+ for k,v in kwargs.items():
+ setattr(self, k, v)
+
+
+metadata = MetaData()
+url_table = Table('ga_url', metadata,
+ Column('id', types.UnicodeText, primary_key=True,
+ default=make_uuid),
+ Column('period_name', types.UnicodeText),
+ Column('period_complete_day', types.Integer),
+ Column('visits', types.Integer),
+ Column('url', types.UnicodeText),
+ Column('next_page', types.UnicodeText),
+ )
+mapper(GA_Url, url_table)
+
+
def init_tables():
- metadata = MetaData()
- package_stats = Table('ga_url', metadata,
- Column('id', types.UnicodeText, primary_key=True, default=make_uuid),
- Column('period_name', types.UnicodeText),
- Column('period_complete_day', types.Integer),
- Column('visits', types.Integer),
- Column('group_id', types.String(60)),
- Column('next_page', JsonType),
- )
metadata.create_all(model.meta.engine)
cached_tables = {}
+
def get_table(name):
if name not in cached_tables:
@@ -46,6 +60,7 @@
url = re.sub('https?://(www\.)?data.gov.uk', '', url)
return url
+
def _get_department_id_of_url(url):
# e.g. /dataset/fuel_prices
# e.g. /dataset/fuel_prices/resource/e63380d4
@@ -58,34 +73,30 @@
if publisher_groups:
return publisher_groups[0].id
+
def update_url_stats(period_name, period_complete_day, url_data):
table = get_table('ga_url')
- connection = model.Session.connection()
for url, views, next_page in url_data:
url = _normalize_url(url)
department_id = _get_department_id_of_url(url)
# see if the row for this url & month is in the table already
- s = select([func.count(id_col)],
- table.c.period_name == period_name,
- table.c.url == url)
- count = connection.execute(s).fetchone()
- if count and count[0]:
- # update the row
- connection.execute(table.update()\
- .where(table.c.period_name == period_name,
- table.c.url == url)\
- .values(period_complete_day=period_complete_day,
- views=views,
- department_id=department_id,
- next_page=next_page))
+ 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_complete_day=period_complete_day
+ item.views = views
+ item.department_id = department_id
+ item.next_page = next_page
+ model.Session.add(item)
else:
# create the row
- values = {'period_name': period_name,
+ values = {'id': make_uuid(),
+ 'period_name': period_name,
'period_complete_day': period_complete_day,
'url': url,
'views': views,
'department_id': department_id,
'next_page': next_page}
- connection.execute(stats.insert()\
- .values(**values))
+ obj = GA_Url(**values)
+ model.Session.add(obj)
+ model.Session.commit()
--- /dev/null
+++ b/ckanext/ga_report/tests/test_auth.py
@@ -1,1 +1,41 @@
+import os
+from nose.tools import assert_equal
+from ckanext.ga_report.ga_auth import (init_service, get_profile_id)
+class TestAuth:
+
+ @classmethod
+ def setup_class(cls):
+ if not os.path.exists("token.dat") or not os.path.exists("credentials.json"):
+ print '*' * 60
+ print "Tests may not run without first having run the auth process"
+ print '*' * 60
+
+ @classmethod
+ def teardown_class(cls):
+ pass
+
+ def test_init(self):
+ try:
+ res = init_service(None, None)
+ assert False, "Init service worked without credentials or tokens"
+ except TypeError:
+ pass
+
+ def test_init_with_token(self):
+ res = init_service("token.dat", None)
+ assert res is not None, "Init service worked without credentials"
+
+ def test_init_with_token_and_credentials(self):
+ res = init_service("token.dat", "credentials.json")
+ assert res is not None, "Unable to create service with valid details"
+
+ def test_init_with_redentials(self):
+ #res = init_service("", "credentials.json")
+ # Triggers the auth flow via the browser
+ pass
+
+ def test_get_profile(self):
+ svc = init_service("token.dat", "credentials.json")
+ profile = get_profile_id(svc)
+ assert profile is not None, "Unable to find a profile given configured UA id and user details"