*.py[co] | *.py[co] |
*.py~ | *.py~ |
.gitignore | .gitignore |
ckan.log | |
# Packages | # Packages |
*.egg | *.egg |
*.egg-info | *.egg-info |
dist | dist |
build | build |
eggs | eggs |
parts | parts |
bin | bin |
var | var |
sdist | sdist |
develop-eggs | develop-eggs |
.installed.cfg | .installed.cfg |
# Private info | # Private info |
credentials.json | credentials.json |
token.dat | token.dat |
# Installer logs | # Installer logs |
pip-log.txt | pip-log.txt |
# Unit test / coverage reports | # Unit test / coverage reports |
.coverage | .coverage |
.tox | .tox |
#Translations | #Translations |
*.mo | *.mo |
#Mr Developer | #Mr Developer |
.mr.developer.cfg | .mr.developer.cfg |
import logging | import logging |
import datetime | import datetime |
import os | import os |
from pylons import config | from pylons import config |
from ckan.lib.cli import CkanCommand | from ckan.lib.cli import CkanCommand |
# No other CKAN imports allowed until _load_config is run, | # No other CKAN imports allowed until _load_config is run, |
# or logging is disabled | # or logging is disabled |
class InitDB(CkanCommand): | class InitDB(CkanCommand): |
"""Initialise the extension's database tables | """Initialise the extension's database tables |
""" | """ |
summary = __doc__.split('\n')[0] | summary = __doc__.split('\n')[0] |
usage = __doc__ | usage = __doc__ |
max_args = 0 | max_args = 0 |
min_args = 0 | min_args = 0 |
def command(self): | def command(self): |
self._load_config() | self._load_config() |
import ckan.model as model | import ckan.model as model |
model.Session.remove() | model.Session.remove() |
model.Session.configure(bind=model.meta.engine) | model.Session.configure(bind=model.meta.engine) |
log = logging.getLogger('ckanext.ga-report') | log = logging.getLogger('ckanext.ga_report') |
import ga_model | import ga_model |
ga_model.init_tables() | ga_model.init_tables() |
log.info("DB tables are setup") | log.info("DB tables are setup") |
class GetAuthToken(CkanCommand): | class GetAuthToken(CkanCommand): |
""" Get's the Google auth token | """ Get's the Google auth token |
Usage: paster getauthtoken <credentials_file> | Usage: paster getauthtoken <credentials_file> |
Where <credentials_file> is the file name containing the details | Where <credentials_file> is the file name containing the details |
for the service (obtained from https://code.google.com/apis/console). | for the service (obtained from https://code.google.com/apis/console). |
By default this is set to credentials.json | By default this is set to credentials.json |
""" | """ |
summary = __doc__.split('\n')[0] | summary = __doc__.split('\n')[0] |
usage = __doc__ | usage = __doc__ |
max_args = 0 | max_args = 0 |
min_args = 0 | min_args = 0 |
def command(self): | def command(self): |
""" | """ |
In this case we don't want a valid service, but rather just to | 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 | 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 | act as a form of verification instead of just getting the token and |
assuming it is correct. | assuming it is correct. |
""" | """ |
from ga_auth import init_service | from ga_auth import init_service |
init_service('token.dat', | init_service('token.dat', |
self.args[0] if self.args | self.args[0] if self.args |
else 'credentials.json') | 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): | class LoadAnalytics(CkanCommand): |
"""Get data from Google Analytics API and save it | """Get data from Google Analytics API and save it |
in the ga_model | in the ga_model |
Usage: paster loadanalytics <time-period> | Usage: paster loadanalytics <time-period> |
Where <time-period> is: | Where <time-period> is: |
all - data for all time | all - data for all time |
latest - (default) just the 'latest' data | latest - (default) just the 'latest' data |
YYYY-MM - just data for the specific month | YYYY-MM - just data for the specific month |
""" | """ |
summary = __doc__.split('\n')[0] | summary = __doc__.split('\n')[0] |
usage = __doc__ | usage = __doc__ |
max_args = 1 | max_args = 1 |
min_args = 0 | min_args = 0 |
def __init__(self, name): | def __init__(self, name): |
super(LoadAnalytics, self).__init__(name) | super(LoadAnalytics, self).__init__(name) |
self.parser.add_option('-d', '--delete-first', | self.parser.add_option('-d', '--delete-first', |
action='store_true', | action='store_true', |
default=False, | default=False, |
dest='delete_first', | dest='delete_first', |
help='Delete data for the period first') | help='Delete data for the period first') |
self.parser.add_option('-s', '--skip_url_stats', | self.parser.add_option('-s', '--skip_url_stats', |
action='store_true', | action='store_true', |
default=False, | default=False, |
dest='skip_url_stats', | dest='skip_url_stats', |
help='Skip the download of URL data - just do site-wide stats') | help='Skip the download of URL data - just do site-wide stats') |
def command(self): | def command(self): |
self._load_config() | self._load_config() |
from download_analytics import DownloadAnalytics | from download_analytics import DownloadAnalytics |
from ga_auth import (init_service, get_profile_id) | from ga_auth import (init_service, get_profile_id) |
ga_token_filepath = os.path.expanduser(config.get('googleanalytics.token.filepath', '')) | ga_token_filepath = os.path.expanduser(config.get('googleanalytics.token.filepath', '')) |
if not ga_token_filepath: | if not ga_token_filepath: |
print 'ERROR: In the CKAN config you need to specify the filepath of the ' \ | print 'ERROR: In the CKAN config you need to specify the filepath of the ' \ |
'Google Analytics token file under key: googleanalytics.token.filepath' | 'Google Analytics token file under key: googleanalytics.token.filepath' |
return | return |
try: | try: |
svc = init_service(ga_token_filepath, None) | svc = init_service(ga_token_filepath, None) |
except TypeError: | except TypeError: |
print ('Have you correctly run the getauthtoken task and ' | print ('Have you correctly run the getauthtoken task and ' |
'specified the correct token file in the CKAN config under ' | 'specified the correct token file in the CKAN config under ' |
'"googleanalytics.token.filepath"?') | '"googleanalytics.token.filepath"?') |
return | return |
downloader = DownloadAnalytics(svc, profile_id=get_profile_id(svc), | downloader = DownloadAnalytics(svc, profile_id=get_profile_id(svc), |
delete_first=self.options.delete_first, | delete_first=self.options.delete_first, |
skip_url_stats=self.options.skip_url_stats) | skip_url_stats=self.options.skip_url_stats) |
time_period = self.args[0] if self.args else 'latest' | time_period = self.args[0] if self.args else 'latest' |
if time_period == 'all': | if time_period == 'all': |
downloader.all_() | downloader.all_() |
elif time_period == 'latest': | elif time_period == 'latest': |
downloader.latest() | downloader.latest() |
else: | else: |
# The month to use | # The month to use |
for_date = datetime.datetime.strptime(time_period, '%Y-%m') | for_date = datetime.datetime.strptime(time_period, '%Y-%m') |
downloader.specific_month(for_date) | downloader.specific_month(for_date) |
import re | import re |
import csv | import csv |
import sys | import sys |
import json | |
import logging | import logging |
import operator | import operator |
import collections | import collections |
from ckan.lib.base import (BaseController, c, g, render, request, response, abort) | from ckan.lib.base import (BaseController, c, g, render, request, response, abort) |
import sqlalchemy | import sqlalchemy |
from sqlalchemy import func, cast, Integer | from sqlalchemy import func, cast, Integer |
import ckan.model as model | import ckan.model as model |
from ga_model import GA_Url, GA_Stat, GA_ReferralStat, GA_Publisher | from ga_model import GA_Url, GA_Stat, GA_ReferralStat, GA_Publisher |
log = logging.getLogger('ckanext.ga-report') | log = logging.getLogger('ckanext.ga-report') |
DOWNLOADS_AVAILABLE_FROM = '2012-12' | |
def _get_month_name(strdate): | def _get_month_name(strdate): |
import calendar | import calendar |
from time import strptime | from time import strptime |
d = strptime(strdate, '%Y-%m') | d = strptime(strdate, '%Y-%m') |
return '%s %s' % (calendar.month_name[d.tm_mon], d.tm_year) | return '%s %s' % (calendar.month_name[d.tm_mon], d.tm_year) |
def _get_unix_epoch(strdate): | |
def _month_details(cls): | 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 | 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 | knows too much about the type of the cls being passed as GA_Url has a |
more complex query | more complex query |
This may need extending if we add a period_name to the stats | This may need extending if we add a period_name to the stats |
''' | ''' |
months = [] | months = [] |
day = None | day = None |
vals = model.Session.query(cls.period_name,cls.period_complete_day)\ | q = model.Session.query(cls.period_name,cls.period_complete_day)\ |
.filter(cls.period_name!='All').distinct(cls.period_name)\ | .filter(cls.period_name!='All').distinct(cls.period_name) |
.order_by("period_name desc").all() | 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]: | if vals and vals[0][1]: |
day = int(vals[0][1]) | day = int(vals[0][1]) |
ordinal = 'th' if 11 <= day <= 13 \ | ordinal = 'th' if 11 <= day <= 13 \ |
else {1:'st',2:'nd',3:'rd'}.get(day % 10, 'th') | else {1:'st',2:'nd',3:'rd'}.get(day % 10, 'th') |
day = "{day}{ordinal}".format(day=day, ordinal=ordinal) | day = "{day}{ordinal}".format(day=day, ordinal=ordinal) |
for m in vals: | for m in vals: |
months.append( (m[0], _get_month_name(m[0]))) | months.append( (m[0], _get_month_name(m[0]))) |
return months, day | return months, day |
class GaReport(BaseController): | class GaReport(BaseController): |
def csv(self, month): | def csv(self, month): |
import csv | import csv |
q = model.Session.query(GA_Stat) | q = model.Session.query(GA_Stat).filter(GA_Stat.stat_name!='Downloads') |
if month != 'all': | if month != 'all': |
q = q.filter(GA_Stat.period_name==month) | q = q.filter(GA_Stat.period_name==month) |
entries = q.order_by('GA_Stat.period_name, GA_Stat.stat_name, GA_Stat.key').all() | entries = q.order_by('GA_Stat.period_name, GA_Stat.stat_name, GA_Stat.key').all() |
response.headers['Content-Type'] = "text/csv; charset=utf-8" | response.headers['Content-Type'] = "text/csv; charset=utf-8" |
response.headers['Content-Disposition'] = str('attachment; filename=stats_%s.csv' % (month,)) | response.headers['Content-Disposition'] = str('attachment; filename=stats_%s.csv' % (month,)) |
writer = csv.writer(response) | writer = csv.writer(response) |
writer.writerow(["Period", "Statistic", "Key", "Value"]) | writer.writerow(["Period", "Statistic", "Key", "Value"]) |
for entry in entries: | for entry in entries: |
writer.writerow([entry.period_name.encode('utf-8'), | writer.writerow([entry.period_name.encode('utf-8'), |
entry.stat_name.encode('utf-8'), | entry.stat_name.encode('utf-8'), |
entry.key.encode('utf-8'), | entry.key.encode('utf-8'), |
entry.value.encode('utf-8')]) | entry.value.encode('utf-8')]) |
def index(self): | def index(self): |
# Get the month details by fetching distinct values and determining the | # Get the month details by fetching distinct values and determining the |
# month names from the values. | # month names from the values. |
c.months, c.day = _month_details(GA_Stat) | c.months, c.day = _month_details(GA_Stat) |
# Work out which month to show, based on query params of the first item | # Work out which month to show, based on query params of the first item |
c.month_desc = 'all months' | c.month_desc = 'all months' |
c.month = request.params.get('month', '') | c.month = request.params.get('month', '') |
if c.month: | if c.month: |
c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month]) | c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month]) |
q = model.Session.query(GA_Stat).\ | q = model.Session.query(GA_Stat).\ |
filter(GA_Stat.stat_name=='Totals') | filter(GA_Stat.stat_name=='Totals') |
if c.month: | if c.month: |
q = q.filter(GA_Stat.period_name==c.month) | q = q.filter(GA_Stat.period_name==c.month) |
entries = q.order_by('ga_stat.key').all() | entries = q.order_by('ga_stat.key').all() |
def clean_key(key, val): | def clean_key(key, val): |
if key in ['Average time on site', 'Pages per visit', 'New visits', 'Bounce rate (home page)']: | if key in ['Average time on site', 'Pages per visit', 'New visits', 'Bounce rate (home page)']: |
val = "%.2f" % round(float(val), 2) | val = "%.2f" % round(float(val), 2) |
if key == 'Average time on site': | if key == 'Average time on site': |
mins, secs = divmod(float(val), 60) | mins, secs = divmod(float(val), 60) |
hours, mins = divmod(mins, 60) | hours, mins = divmod(mins, 60) |
val = '%02d:%02d:%02d (%s seconds) ' % (hours, mins, secs, val) | val = '%02d:%02d:%02d (%s seconds) ' % (hours, mins, secs, val) |
if key in ['New visits','Bounce rate (home page)']: | if key in ['New visits','Bounce rate (home page)']: |
val = "%s%%" % val | val = "%s%%" % val |
if key in ['Total page views', 'Total visits']: | if key in ['Total page views', 'Total visits']: |
val = int(val) | val = int(val) |
return key, val | 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 = [] | c.global_totals = [] |
if c.month: | if c.month: |
for e in entries: | for e in entries: |
key, val = clean_key(e.key, e.value) | 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: | else: |
d = collections.defaultdict(list) | d = collections.defaultdict(list) |
for e in entries: | for e in entries: |
d[e.key].append(float(e.value)) | d[e.key].append(float(e.value)) |
for k, v in d.iteritems(): | for k, v in d.iteritems(): |
if k in ['Total page views', 'Total visits']: | if k in ['Total page views', 'Total visits']: |
v = sum(v) | v = sum(v) |
else: | else: |
v = float(sum(v))/len(v) | v = float(sum(v))/float(len(v)) |
sparkline = sparkline_data[k] | |
key, val = clean_key(k,v) | key, val = clean_key(k,v) |
c.global_totals.append((key, val)) | 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 = { | keys = { |
'Browser versions': 'browser_versions', | 'Browser versions': 'browser_versions', |
'Browsers': 'browsers', | 'Browsers': 'browsers', |
'Operating Systems versions': 'os_versions', | 'Operating Systems versions': 'os_versions', |
'Operating Systems': 'os', | 'Operating Systems': 'os', |
'Social sources': 'social_networks', | 'Social sources': 'social_networks', |
'Languages': 'languages', | 'Languages': 'languages', |
'Country': 'country' | 'Country': 'country' |
} | } |
def shorten_name(name, length=60): | def shorten_name(name, length=60): |
return (name[:length] + '..') if len(name) > 60 else name | return (name[:length] + '..') if len(name) > 60 else name |
def fill_out_url(url): | def fill_out_url(url): |
import urlparse | import urlparse |
return urlparse.urljoin(g.site_url, url) | return urlparse.urljoin(g.site_url, url) |
c.social_referrer_totals, c.social_referrers = [], [] | c.social_referrer_totals, c.social_referrers = [], [] |
q = model.Session.query(GA_ReferralStat) | q = model.Session.query(GA_ReferralStat) |
q = q.filter(GA_ReferralStat.period_name==c.month) if c.month else q | q = q.filter(GA_ReferralStat.period_name==c.month) if c.month else q |
q = q.order_by('ga_referrer.count::int desc') | q = q.order_by('ga_referrer.count::int desc') |
for entry in q.all(): | for entry in q.all(): |
c.social_referrers.append((shorten_name(entry.url), fill_out_url(entry.url), | c.social_referrers.append((shorten_name(entry.url), fill_out_url(entry.url), |
entry.source,entry.count)) | entry.source,entry.count)) |
q = model.Session.query(GA_ReferralStat.url, | q = model.Session.query(GA_ReferralStat.url, |
func.sum(GA_ReferralStat.count).label('count')) | func.sum(GA_ReferralStat.count).label('count')) |
q = q.filter(GA_ReferralStat.period_name==c.month) if c.month else q | q = q.filter(GA_ReferralStat.period_name==c.month) if c.month else q |
q = q.order_by('count desc').group_by(GA_ReferralStat.url) | q = q.order_by('count desc').group_by(GA_ReferralStat.url) |
for entry in q.all(): | for entry in q.all(): |
c.social_referrer_totals.append((shorten_name(entry[0]), fill_out_url(entry[0]),'', | c.social_referrer_totals.append((shorten_name(entry[0]), fill_out_url(entry[0]),'', |
entry[1])) | entry[1])) |
for k, v in keys.iteritems(): | for k, v in keys.iteritems(): |
q = model.Session.query(GA_Stat).\ | q = model.Session.query(GA_Stat).\ |
filter(GA_Stat.stat_name==k) | filter(GA_Stat.stat_name==k).\ |
order_by(GA_Stat.period_name) | |
# Run the query on all months to gather graph data | |
graph = {} | |
for stat in q: | |
graph[ stat.key ] = graph.get(stat.key,{ | |
'name':stat.key, | |
'data': [] | |
}) | |
graph[ stat.key ]['data'].append({ | |
'x':_get_unix_epoch(stat.period_name), | |
'y':float(stat.value) | |
}) | |
setattr(c, v+'_graph', json.dumps( _to_rickshaw(graph.values(),percentageMode=True) )) | |
# Buffer the tabular data | |
if c.month: | if c.month: |
entries = [] | entries = [] |
q = q.filter(GA_Stat.period_name==c.month).\ | q = q.filter(GA_Stat.period_name==c.month).\ |
order_by('ga_stat.value::int desc') | order_by('ga_stat.value::int desc') |
d = collections.defaultdict(int) | d = collections.defaultdict(int) |
for e in q.all(): | for e in q.all(): |
d[e.key] += int(e.value) | d[e.key] += int(e.value) |
entries = [] | entries = [] |
for key, val in d.iteritems(): | for key, val in d.iteritems(): |
entries.append((key,val,)) | entries.append((key,val,)) |
entries = sorted(entries, key=operator.itemgetter(1), reverse=True) | entries = sorted(entries, key=operator.itemgetter(1), reverse=True) |
# Get the total for each set of values and then set the value as | # Get the total for each set of values and then set the value as |
# a percentage of the total | # a percentage of the total |
if k == 'Social sources': | 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: | else: |
total = sum([num for _,num in entries]) | total = sum([num for _,num in entries]) |
setattr(c, v, [(k,_percent(v,total)) for k,v in entries ]) | setattr(c, v, [(k,_percent(v,total)) for k,v in entries ]) |
return render('ga_report/site/index.html') | return render('ga_report/site/index.html') |
class GaDatasetReport(BaseController): | class GaDatasetReport(BaseController): |
""" | """ |
Displays the pageview and visit count for datasets | Displays the pageview and visit count for datasets |
with options to filter by publisher and time period. | with options to filter by publisher and time period. |
""" | """ |
def publisher_csv(self, month): | def publisher_csv(self, month): |
''' | ''' |
Returns a CSV of each publisher with the total number of dataset | Returns a CSV of each publisher with the total number of dataset |
views & visits. | views & visits. |
''' | ''' |
c.month = month if not month == 'all' else '' | c.month = month if not month == 'all' else '' |
response.headers['Content-Type'] = "text/csv; charset=utf-8" | response.headers['Content-Type'] = "text/csv; charset=utf-8" |
response.headers['Content-Disposition'] = str('attachment; filename=publishers_%s.csv' % (month,)) | response.headers['Content-Disposition'] = str('attachment; filename=publishers_%s.csv' % (month,)) |
writer = csv.writer(response) | writer = csv.writer(response) |
writer.writerow(["Publisher Title", "Publisher Name", "Views", "Visits", "Period Name"]) | 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'), | writer.writerow([publisher.title.encode('utf-8'), |
publisher.name.encode('utf-8'), | publisher.name.encode('utf-8'), |
view, | view, |
visit, | visit, |
month]) | month]) |
def dataset_csv(self, id='all', month='all'): | def dataset_csv(self, id='all', month='all'): |
''' | ''' |
Returns a CSV with the number of views & visits for each dataset. | Returns a CSV with the number of views & visits for each dataset. |
:param id: A Publisher ID or None if you want for all | :param id: A Publisher ID or None if you want for all |
:param month: The time period, or 'all' | :param month: The time period, or 'all' |
''' | ''' |
c.month = month if not month == 'all' else '' | c.month = month if not month == 'all' else '' |
if id != 'all': | if id != 'all': |
c.publisher = model.Group.get(id) | c.publisher = model.Group.get(id) |
if not c.publisher: | if not c.publisher: |
abort(404, 'A publisher with that name could not be found') | abort(404, 'A publisher with that name could not be found') |
packages = self._get_packages(c.publisher) | packages = self._get_packages(c.publisher) |
response.headers['Content-Type'] = "text/csv; charset=utf-8" | response.headers['Content-Type'] = "text/csv; charset=utf-8" |
response.headers['Content-Disposition'] = \ | response.headers['Content-Disposition'] = \ |
str('attachment; filename=datasets_%s_%s.csv' % (c.publisher_name, month,)) | str('attachment; filename=datasets_%s_%s.csv' % (c.publisher_name, month,)) |
writer = csv.writer(response) | writer = csv.writer(response) |
writer.writerow(["Dataset Title", "Dataset Name", "Views", "Visits", "Period Name"]) | writer.writerow(["Dataset Title", "Dataset Name", "Views", "Visits", "Resource downloads", "Period Name"]) |
for package,view,visit in packages: | for package,view,visit,downloads in packages: |
writer.writerow([package.title.encode('utf-8'), | writer.writerow([package.title.encode('utf-8'), |
package.name.encode('utf-8'), | package.name.encode('utf-8'), |
view, | view, |
visit, | visit, |
downloads, | |
month]) | month]) |
def publishers(self): | def publishers(self): |
'''A list of publishers and the number of views/visits for each''' | '''A list of publishers and the number of views/visits for each''' |
# Get the month details by fetching distinct values and determining the | # Get the month details by fetching distinct values and determining the |
# month names from the values. | # month names from the values. |
c.months, c.day = _month_details(GA_Url) | c.months, c.day = _month_details(GA_Url) |
# Work out which month to show, based on query params of the first item | # Work out which month to show, based on query params of the first item |
c.month = request.params.get('month', '') | c.month = request.params.get('month', '') |
c.month_desc = 'all months' | c.month_desc = 'all months' |
if c.month: | if c.month: |
c.month_desc = ''.join([m[1] for m in c.months if m[0]==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') | return render('ga_report/publisher/index.html') |
def _get_packages(self, publisher=None, count=-1): | def _get_packages(self, publisher=None, count=-1): |
'''Returns the datasets in order of views''' | '''Returns the datasets in order of views''' |
if count == -1: | have_download_data = True |
count = sys.maxint | |
month = c.month or 'All' | month = c.month or 'All' |
if month != 'All': | |
have_download_data = month >= DOWNLOADS_AVAILABLE_FROM | |
q = model.Session.query(GA_Url,model.Package)\ | q = model.Session.query(GA_Url,model.Package)\ |
.filter(model.Package.name==GA_Url.package_id)\ | .filter(model.Package.name==GA_Url.package_id)\ |
.filter(GA_Url.url.like('/dataset/%')) | .filter(GA_Url.url.like('/dataset/%')) |
if publisher: | if publisher: |
q = q.filter(GA_Url.department_id==publisher.name) | q = q.filter(GA_Url.department_id==publisher.name) |
q = q.filter(GA_Url.period_name==month) | q = q.filter(GA_Url.period_name==month) |
q = q.order_by('ga_url.pageviews::int desc') | q = q.order_by('ga_url.pageviews::int desc') |
top_packages = [] | 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: | 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: | else: |
log.warning('Could not find package associated package') | log.warning('Could not find package associated package') |
return top_packages | return top_packages |
def read(self): | def read(self): |
''' | ''' |
Lists the most popular datasets across all publishers | Lists the most popular datasets across all publishers |
''' | ''' |
return self.read_publisher(None) | return self.read_publisher(None) |
def read_publisher(self, id): | def read_publisher(self, id): |
''' | ''' |
Lists the most popular datasets for a publisher (or across all publishers) | Lists the most popular datasets for a publisher (or across all publishers) |
''' | ''' |
count = 20 | count = 20 |
c.publishers = _get_publishers() | c.publishers = _get_publishers() |
id = request.params.get('publisher', id) | id = request.params.get('publisher', id) |
if id and id != 'all': | if id and id != 'all': |
c.publisher = model.Group.get(id) | c.publisher = model.Group.get(id) |
if not c.publisher: | if not c.publisher: |
abort(404, 'A publisher with that name could not be found') | abort(404, 'A publisher with that name could not be found') |
c.publisher_name = c.publisher.name | c.publisher_name = c.publisher.name |
c.top_packages = [] # package, dataset_views in c.top_packages | c.top_packages = [] # package, dataset_views in c.top_packages |
# Get the month details by fetching distinct values and determining the | # Get the month details by fetching distinct values and determining the |
# month names from the values. | # month names from the values. |
c.months, c.day = _month_details(GA_Url) | c.months, c.day = _month_details(GA_Url) |
# Work out which month to show, based on query params of the first item | # Work out which month to show, based on query params of the first item |
c.month = request.params.get('month', '') | c.month = request.params.get('month', '') |
if not c.month: | if not c.month: |
c.month_desc = 'all months' | c.month_desc = 'all months' |
else: | else: |
c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month]) | c.month_desc = ''.join([m[1] for m in c.months if m[0]==c.month]) |
month = c.month or 'All' | month = c.month or 'All' |
c.publisher_page_views = 0 | c.publisher_page_views = 0 |
q = model.Session.query(GA_Url).\ | q = model.Session.query(GA_Url).\ |
filter(GA_Url.url=='/publisher/%s' % c.publisher_name) | filter(GA_Url.url=='/publisher/%s' % c.publisher_name) |
entry = q.filter(GA_Url.period_name==c.month).first() | entry = q.filter(GA_Url.period_name==c.month).first() |
c.publisher_page_views = entry.pageviews if entry else 0 | 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(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') | return render('ga_report/publisher/read.html') |
def _to_rickshaw(data, percentageMode=False): | |
if data==[]: | |
return data | |
# Create a consistent x-axis | |
num_points = [ len(package['data']) for package in data ] | |
ideal_index = num_points.index( max(num_points) ) | |
x_axis = [ point['x'] for point in data[ideal_index]['data'] ] | |
for package in data: | |
xs = [ point['x'] for point in package['data'] ] | |
assert set(xs).issubset( set(x_axis) ), (xs, x_axis) | |
# Zero pad any missing values | |
for x in set(x_axis).difference(set(xs)): | |
package['data'].append( {'x':x, 'y':0} ) | |
assert len(package['data'])==len(x_axis), (len(package['data']),len(x_axis),package['data'],x_axis,set(x_axis).difference(set(xs))) | |
if percentageMode: | |
# Transform data into percentage stacks | |
totals = {} | |
for x in x_axis: | |
for package in data: | |
for point in package['data']: | |
totals[ point['x'] ] = totals.get(point['x'],0) + point['y'] | |
# Roll insignificant series into a catch-all | |
THRESHOLD = 0.01 | |
significant_series = [] | |
for package in data: | |
for point in package['data']: | |
fraction = float(point['y']) / totals[point['x']] | |
if fraction>THRESHOLD and not (package in significant_series): | |
significant_series.append(package) | |
temp = {} | |
for package in data: | |
if package in significant_series: continue | |
for point in package['data']: | |
temp[point['x']] = temp.get(point['x'],0) + point['y'] | |
catch_all = { 'name':'Other','data': [ {'x':x,'y':y} for x,y in temp.items() ] } | |
# Roll insignificant series into one | |
data = significant_series | |
data.append(catch_all) | |
# Turn each point into a percentage | |
for package in data: | |
for point in package['data']: | |
point['y'] = (point['y']*100) / totals[point['x']] | |
# Sort the points | |
for package in data: | |
package['data'] = sorted( package['data'], key=lambda x:x['x'] ) | |
# Strip the latest month's incomplete analytics | |
package['data'] = package['data'][:-1] | |
return data | |
def _get_top_publishers(limit=20): | def _get_top_publishers(limit=20): |
''' | ''' |
Returns a list of the top 20 publishers by dataset visits. | Returns a list of the top 20 publishers by dataset visits. |
(The number to show can be varied with 'limit') | (The number to show can be varied with 'limit') |
''' | ''' |
month = c.month or 'All' | month = c.month or 'All' |
connection = model.Session.connection() | connection = model.Session.connection() |
q = """ | q = """ |
select department_id, sum(pageviews::int) views, sum(visits::int) visits | select department_id, sum(pageviews::int) views, sum(visits::int) visits |
from ga_url | from ga_url |
where department_id <> '' | where department_id <> '' |
and package_id <> '' | and package_id <> '' |
and url like '/dataset/%%' | and url like '/dataset/%%' |
and period_name=%s | and period_name=%s |
group by department_id order by views desc | group by department_id order by views desc |
""" | """ |
if limit: | if limit: |
q = q + " limit %s;" % (limit) | q = q + " limit %s;" % (limit) |
top_publishers = [] | top_publishers = [] |
res = connection.execute(q, month) | res = connection.execute(q, month) |
department_ids = [] | |
for row in res: | for row in res: |
g = model.Group.get(row[0]) | g = model.Group.get(row[0]) |
if g: | if g: |
department_ids.append(row[0]) | |
top_publishers.append((g, row[1], row[2])) | top_publishers.append((g, row[1], row[2])) |
return top_publishers | |
graph = {} | |
if limit is not None: | |
# Query for a history graph of these publishers | |
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.period_name!='All' )\ | |
.filter( GA_Url.url.like('/dataset/%') )\ | |
.filter( GA_Url.package_id!='' )\ | |
.group_by( GA_Url.department_id, GA_Url.period_name ) | |
for dept_id,period_name,views in q: | |
graph[dept_id] = graph.get( dept_id, { | |
'name' : model.Group.get(dept_id).title, | |
'data' : [] | |
}) | |
graph[dept_id]['data'].append({ | |
'x': _get_unix_epoch(period_name), | |
'y': views | |
}) | |
return top_publishers, graph | |
def _get_publishers(): | def _get_publishers(): |
''' | ''' |
Returns a list of all publishers. Each item is a tuple: | Returns a list of all publishers. Each item is a tuple: |
(name, title) | (name, title) |
''' | ''' |
publishers = [] | publishers = [] |
for pub in model.Session.query(model.Group).\ | for pub in model.Session.query(model.Group).\ |
filter(model.Group.type=='publisher').\ | filter(model.Group.type=='publisher').\ |
filter(model.Group.state=='active').\ | filter(model.Group.state=='active').\ |
order_by(model.Group.name): | order_by(model.Group.name): |
publishers.append((pub.name, pub.title)) | publishers.append((pub.name, pub.title)) |
return publishers | return publishers |
def _percent(num, total): | def _percent(num, total): |
p = 100 * float(num)/float(total) | p = 100 * float(num)/float(total) |
return "%.2f%%" % round(p, 2) | return "%.2f%%" % round(p, 2) |
import os | import os |
import logging | import logging |
import datetime | import datetime |
import collections | import collections |
from pylons import config | from pylons import config |
from ga_model import _normalize_url | from ga_model import _normalize_url |
import ga_model | import ga_model |
#from ga_client import GA | #from ga_client import GA |
log = logging.getLogger('ckanext.ga-report') | log = logging.getLogger('ckanext.ga-report') |
FORMAT_MONTH = '%Y-%m' | FORMAT_MONTH = '%Y-%m' |
MIN_VIEWS = 50 | MIN_VIEWS = 50 |
MIN_VISITS = 20 | MIN_VISITS = 20 |
MIN_DOWNLOADS = 10 | |
class DownloadAnalytics(object): | class DownloadAnalytics(object): |
'''Downloads and stores analytics info''' | '''Downloads and stores analytics info''' |
def __init__(self, service=None, profile_id=None, delete_first=False, | def __init__(self, service=None, profile_id=None, delete_first=False, |
skip_url_stats=False): | skip_url_stats=False): |
self.period = config['ga-report.period'] | self.period = config['ga-report.period'] |
self.service = service | self.service = service |
self.profile_id = profile_id | self.profile_id = profile_id |
self.delete_first = delete_first | self.delete_first = delete_first |
self.skip_url_stats = skip_url_stats | self.skip_url_stats = skip_url_stats |
def specific_month(self, date): | def specific_month(self, date): |
import calendar | import calendar |
first_of_this_month = datetime.datetime(date.year, date.month, 1) | first_of_this_month = datetime.datetime(date.year, date.month, 1) |
_, last_day_of_month = calendar.monthrange(int(date.year), int(date.month)) | _, 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) | 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), | periods = ((date.strftime(FORMAT_MONTH), |
last_day_of_month, | last_day_of_month, |
first_of_this_month, last_of_this_month),) | first_of_this_month, last_of_this_month),) |
self.download_and_store(periods) | self.download_and_store(periods) |
def latest(self): | def latest(self): |
if self.period == 'monthly': | if self.period == 'monthly': |
# from first of this month to today | # from first of this month to today |
now = datetime.datetime.now() | now = datetime.datetime.now() |
first_of_this_month = datetime.datetime(now.year, now.month, 1) | first_of_this_month = datetime.datetime(now.year, now.month, 1) |
periods = ((now.strftime(FORMAT_MONTH), | periods = ((now.strftime(FORMAT_MONTH), |
now.day, | now.day, |
first_of_this_month, now),) | first_of_this_month, now),) |
else: | else: |
raise NotImplementedError | raise NotImplementedError |
self.download_and_store(periods) | self.download_and_store(periods) |
def for_date(self, for_date): | def for_date(self, for_date): |
assert isinstance(since_date, datetime.datetime) | assert isinstance(since_date, datetime.datetime) |
periods = [] # (period_name, period_complete_day, start_date, end_date) | periods = [] # (period_name, period_complete_day, start_date, end_date) |
if self.period == 'monthly': | if self.period == 'monthly': |
first_of_the_months_until_now = [] | first_of_the_months_until_now = [] |
year = for_date.year | year = for_date.year |
month = for_date.month | month = for_date.month |
now = datetime.datetime.now() | now = datetime.datetime.now() |
first_of_this_month = datetime.datetime(now.year, now.month, 1) | first_of_this_month = datetime.datetime(now.year, now.month, 1) |
while True: | while True: |
first_of_the_month = datetime.datetime(year, month, 1) | first_of_the_month = datetime.datetime(year, month, 1) |
if first_of_the_month == first_of_this_month: | if first_of_the_month == first_of_this_month: |
periods.append((now.strftime(FORMAT_MONTH), | periods.append((now.strftime(FORMAT_MONTH), |
now.day, | now.day, |
first_of_this_month, now)) | first_of_this_month, now)) |
break | break |
elif first_of_the_month < first_of_this_month: | elif first_of_the_month < first_of_this_month: |
in_the_next_month = first_of_the_month + datetime.timedelta(40) | in_the_next_month = first_of_the_month + datetime.timedelta(40) |
last_of_the_month = datetime.datetime(in_the_next_month.year, | last_of_the_month = datetime.datetime(in_the_next_month.year, |
in_the_next_month.month, 1)\ | in_the_next_month.month, 1)\ |
- datetime.timedelta(1) | - datetime.timedelta(1) |
periods.append((now.strftime(FORMAT_MONTH), 0, | periods.append((now.strftime(FORMAT_MONTH), 0, |
first_of_the_month, last_of_the_month)) | first_of_the_month, last_of_the_month)) |
else: | else: |
# first_of_the_month has got to the future somehow | # first_of_the_month has got to the future somehow |
break | break |
month += 1 | month += 1 |
if month > 12: | if month > 12: |
year += 1 | year += 1 |
month = 1 | month = 1 |
else: | else: |
raise NotImplementedError | raise NotImplementedError |
self.download_and_store(periods) | self.download_and_store(periods) |
@staticmethod | @staticmethod |
def get_full_period_name(period_name, period_complete_day): | def get_full_period_name(period_name, period_complete_day): |
if period_complete_day: | if period_complete_day: |
return period_name + ' (up to %ith)' % period_complete_day | return period_name + ' (up to %ith)' % period_complete_day |
else: | else: |
return period_name | return period_name |
def download_and_store(self, periods): | def download_and_store(self, periods): |
for period_name, period_complete_day, start_date, end_date in periods: | for period_name, period_complete_day, start_date, end_date in periods: |
log.info('Period "%s" (%s - %s)', | log.info('Period "%s" (%s - %s)', |
self.get_full_period_name(period_name, period_complete_day), | self.get_full_period_name(period_name, period_complete_day), |
start_date.strftime('%Y-%m-%d'), | start_date.strftime('%Y-%m-%d'), |
end_date.strftime('%Y-%m-%d')) | end_date.strftime('%Y-%m-%d')) |
if self.delete_first: | if self.delete_first: |
log.info('Deleting existing Analytics for this period "%s"', | log.info('Deleting existing Analytics for this period "%s"', |
period_name) | period_name) |
ga_model.delete(period_name) | ga_model.delete(period_name) |
if not self.skip_url_stats: | if not self.skip_url_stats: |
# Clean out old url data before storing the new | # Clean out old url data before storing the new |
ga_model.pre_update_url_stats(period_name) | ga_model.pre_update_url_stats(period_name) |
accountName = config.get('googleanalytics.account') | accountName = config.get('googleanalytics.account') |
log.info('Downloading analytics for dataset views') | log.info('Downloading analytics for dataset views') |
data = self.download(start_date, end_date, '~/%s/dataset/[a-z0-9-_]+' % accountName) | data = self.download(start_date, end_date, '~/%s/dataset/[a-z0-9-_]+' % accountName) |
log.info('Storing dataset views (%i rows)', len(data.get('url'))) | log.info('Storing dataset views (%i rows)', len(data.get('url'))) |
self.store(period_name, period_complete_day, data, ) | self.store(period_name, period_complete_day, data, ) |
log.info('Downloading analytics for publisher views') | log.info('Downloading analytics for publisher views') |
data = self.download(start_date, end_date, '~/%s/publisher/[a-z0-9-_]+' % accountName) | data = self.download(start_date, end_date, '~/%s/publisher/[a-z0-9-_]+' % accountName) |
log.info('Storing publisher views (%i rows)', len(data.get('url'))) | log.info('Storing publisher views (%i rows)', len(data.get('url'))) |
self.store(period_name, period_complete_day, data,) | 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. | ga_model.update_publisher_stats(period_name) # about 30 seconds. |
log.info('Downloading and storing analytics for site-wide stats') | log.info('Downloading and storing analytics for site-wide stats') |
self.sitewide_stats( period_name, period_complete_day ) | self.sitewide_stats( period_name, period_complete_day ) |
log.info('Downloading and storing analytics for social networks') | log.info('Downloading and storing analytics for social networks') |
self.update_social_info(period_name, start_date, end_date) | self.update_social_info(period_name, start_date, end_date) |
def update_social_info(self, period_name, start_date, end_date): | def update_social_info(self, period_name, start_date, end_date): |
start_date = start_date.strftime('%Y-%m-%d') | start_date = start_date.strftime('%Y-%m-%d') |
end_date = end_date.strftime('%Y-%m-%d') | end_date = end_date.strftime('%Y-%m-%d') |
query = 'ga:hasSocialSourceReferral=~Yes$' | query = 'ga:hasSocialSourceReferral=~Yes$' |
metrics = 'ga:entrances' | metrics = 'ga:entrances' |
sort = '-ga:entrances' | sort = '-ga:entrances' |
# Supported query params at | # Supported query params at |
# https://developers.google.com/analytics/devguides/reporting/core/v3/reference | # https://developers.google.com/analytics/devguides/reporting/core/v3/reference |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
filters=query, | filters=query, |
start_date=start_date, | start_date=start_date, |
metrics=metrics, | metrics=metrics, |
sort=sort, | sort=sort, |
dimensions="ga:landingPagePath,ga:socialNetwork", | dimensions="ga:landingPagePath,ga:socialNetwork", |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
data = collections.defaultdict(list) | data = collections.defaultdict(list) |
rows = results.get('rows',[]) | rows = results.get('rows',[]) |
for row in rows: | for row in rows: |
url = _normalize_url('http:/' + row[0]) | url = _normalize_url('http:/' + row[0]) |
data[url].append( (row[1], int(row[2]),) ) | data[url].append( (row[1], int(row[2]),) ) |
ga_model.update_social(period_name, data) | ga_model.update_social(period_name, data) |
def download(self, start_date, end_date, path=None): | def download(self, start_date, end_date, path=None): |
'''Get data from GA for a given time period''' | '''Get data from GA for a given time period''' |
start_date = start_date.strftime('%Y-%m-%d') | start_date = start_date.strftime('%Y-%m-%d') |
end_date = end_date.strftime('%Y-%m-%d') | end_date = end_date.strftime('%Y-%m-%d') |
query = 'ga:pagePath=%s$' % path | query = 'ga:pagePath=%s$' % path |
metrics = 'ga:pageviews, ga:visits' | metrics = 'ga:pageviews, ga:visits' |
sort = '-ga:pageviews' | sort = '-ga:pageviews' |
# Supported query params at | # Supported query params at |
# https://developers.google.com/analytics/devguides/reporting/core/v3/reference | # https://developers.google.com/analytics/devguides/reporting/core/v3/reference |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
filters=query, | filters=query, |
start_date=start_date, | start_date=start_date, |
metrics=metrics, | metrics=metrics, |
sort=sort, | sort=sort, |
dimensions="ga:pagePath", | dimensions="ga:pagePath", |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
packages = [] | packages = [] |
log.info("There are %d results" % results['totalResults']) | |
for entry in results.get('rows'): | for entry in results.get('rows'): |
(loc,pageviews,visits) = entry | (loc,pageviews,visits) = entry |
url = _normalize_url('http:/' + loc) # strips off domain e.g. www.data.gov.uk or data.gov.uk | url = _normalize_url('http:/' + loc) # strips off domain e.g. www.data.gov.uk or data.gov.uk |
if not url.startswith('/dataset/') and not url.startswith('/publisher/'): | if not url.startswith('/dataset/') and not url.startswith('/publisher/'): |
# filter out strays like: | # filter out strays like: |
# /data/user/login?came_from=http://data.gov.uk/dataset/os-code-point-open | # /data/user/login?came_from=http://data.gov.uk/dataset/os-code-point-open |
# /403.html?page=/about&from=http://data.gov.uk/publisher/planning-inspectorate | # /403.html?page=/about&from=http://data.gov.uk/publisher/planning-inspectorate |
continue | continue |
packages.append( (url, pageviews, visits,) ) # Temporary hack | packages.append( (url, pageviews, visits,) ) # Temporary hack |
return dict(url=packages) | return dict(url=packages) |
def store(self, period_name, period_complete_day, data): | def store(self, period_name, period_complete_day, data): |
if 'url' in data: | if 'url' in data: |
ga_model.update_url_stats(period_name, period_complete_day, data['url']) | ga_model.update_url_stats(period_name, period_complete_day, data['url']) |
def sitewide_stats(self, period_name, period_complete_day): | def sitewide_stats(self, period_name, period_complete_day): |
import calendar | import calendar |
year, month = period_name.split('-') | year, month = period_name.split('-') |
_, last_day_of_month = calendar.monthrange(int(year), int(month)) | _, last_day_of_month = calendar.monthrange(int(year), int(month)) |
start_date = '%s-01' % period_name | start_date = '%s-01' % period_name |
end_date = '%s-%s' % (period_name, last_day_of_month) | end_date = '%s-%s' % (period_name, last_day_of_month) |
funcs = ['_totals_stats', '_social_stats', '_os_stats', | 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: | for f in funcs: |
log.info('Downloading analytics for %s' % f.split('_')[1]) | log.info('Downloading analytics for %s' % f.split('_')[1]) |
getattr(self, f)(start_date, end_date, period_name, period_complete_day) | getattr(self, f)(start_date, end_date, period_name, period_complete_day) |
def _get_results(result_data, f): | def _get_results(result_data, f): |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
key = f(result) | key = f(result) |
data[key] = data.get(key,0) + result[1] | data[key] = data.get(key,0) + result[1] |
return data | return data |
def _totals_stats(self, start_date, end_date, period_name, period_complete_day): | def _totals_stats(self, start_date, end_date, period_name, period_complete_day): |
""" Fetches distinct totals, total pageviews etc """ | """ Fetches distinct totals, total pageviews etc """ |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
start_date=start_date, | start_date=start_date, |
metrics='ga:pageviews', | metrics='ga:pageviews', |
sort='-ga:pageviews', | sort='-ga:pageviews', |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
result_data = results.get('rows') | result_data = results.get('rows') |
ga_model.update_sitewide_stats(period_name, "Totals", {'Total page views': result_data[0][0]}, | ga_model.update_sitewide_stats(period_name, "Totals", {'Total page views': result_data[0][0]}, |
period_complete_day) | period_complete_day) |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
start_date=start_date, | start_date=start_date, |
metrics='ga:pageviewsPerVisit,ga:avgTimeOnSite,ga:percentNewVisits,ga:visits', | metrics='ga:pageviewsPerVisit,ga:avgTimeOnSite,ga:percentNewVisits,ga:visits', |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
result_data = results.get('rows') | result_data = results.get('rows') |
data = { | data = { |
'Pages per visit': result_data[0][0], | 'Pages per visit': result_data[0][0], |
'Average time on site': result_data[0][1], | 'Average time on site': result_data[0][1], |
'New visits': result_data[0][2], | 'New visits': result_data[0][2], |
'Total visits': result_data[0][3], | 'Total visits': result_data[0][3], |
} | } |
ga_model.update_sitewide_stats(period_name, "Totals", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Totals", data, period_complete_day) |
# Bounces from / or another configurable page. | # Bounces from / or another configurable page. |
path = '/%s%s' % (config.get('googleanalytics.account'), | path = '/%s%s' % (config.get('googleanalytics.account'), |
config.get('ga-report.bounce_url', '/')) | config.get('ga-report.bounce_url', '/')) |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
filters='ga:pagePath==%s' % (path,), | filters='ga:pagePath==%s' % (path,), |
start_date=start_date, | start_date=start_date, |
metrics='ga:bounces,ga:pageviews', | metrics='ga:visitBounceRate', |
dimensions='ga:pagePath', | dimensions='ga:pagePath', |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
result_data = results.get('rows') | result_data = results.get('rows') |
if not result_data or len(result_data) != 1: | if not result_data or len(result_data) != 1: |
log.error('Could not pinpoint the bounces for path: %s. Got results: %r', | log.error('Could not pinpoint the bounces for path: %s. Got results: %r', |
path, result_data) | path, result_data) |
return | return |
results = result_data[0] | results = result_data[0] |
bounces, total = [float(x) for x in result_data[0][1:]] | bounces = float(results[1]) |
pct = 100 * bounces/total | # visitBounceRate is already a % |
log.info('%d bounces from %d total == %s', bounces, total, pct) | log.info('Google reports visitBounceRate as %s', bounces) |
ga_model.update_sitewide_stats(period_name, "Totals", {'Bounce rate (home page)': pct}, | ga_model.update_sitewide_stats(period_name, "Totals", {'Bounce rate (home page)': float(bounces)}, |
period_complete_day) | period_complete_day) |
def _locale_stats(self, start_date, end_date, period_name, period_complete_day): | def _locale_stats(self, start_date, end_date, period_name, period_complete_day): |
""" Fetches stats about language and country """ | """ Fetches stats about language and country """ |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
start_date=start_date, | start_date=start_date, |
metrics='ga:pageviews', | metrics='ga:pageviews', |
sort='-ga:pageviews', | sort='-ga:pageviews', |
dimensions="ga:language,ga:country", | dimensions="ga:language,ga:country", |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
result_data = results.get('rows') | result_data = results.get('rows') |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
data[result[0]] = data.get(result[0], 0) + int(result[2]) | data[result[0]] = data.get(result[0], 0) + int(result[2]) |
self._filter_out_long_tail(data, MIN_VIEWS) | self._filter_out_long_tail(data, MIN_VIEWS) |
ga_model.update_sitewide_stats(period_name, "Languages", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Languages", data, period_complete_day) |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
data[result[1]] = data.get(result[1], 0) + int(result[2]) | data[result[1]] = data.get(result[1], 0) + int(result[2]) |
self._filter_out_long_tail(data, MIN_VIEWS) | self._filter_out_long_tail(data, MIN_VIEWS) |
ga_model.update_sitewide_stats(period_name, "Country", data, period_complete_day) | 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): | def _social_stats(self, start_date, end_date, period_name, period_complete_day): |
""" Finds out which social sites people are referred from """ | """ Finds out which social sites people are referred from """ |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
start_date=start_date, | start_date=start_date, |
metrics='ga:pageviews', | metrics='ga:pageviews', |
sort='-ga:pageviews', | sort='-ga:pageviews', |
dimensions="ga:socialNetwork,ga:referralPath", | dimensions="ga:socialNetwork,ga:referralPath", |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
result_data = results.get('rows') | result_data = results.get('rows') |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
if not result[0] == '(not set)': | if not result[0] == '(not set)': |
data[result[0]] = data.get(result[0], 0) + int(result[2]) | data[result[0]] = data.get(result[0], 0) + int(result[2]) |
self._filter_out_long_tail(data, 3) | self._filter_out_long_tail(data, 3) |
ga_model.update_sitewide_stats(period_name, "Social sources", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Social sources", data, period_complete_day) |
def _os_stats(self, start_date, end_date, period_name, period_complete_day): | def _os_stats(self, start_date, end_date, period_name, period_complete_day): |
""" Operating system stats """ | """ Operating system stats """ |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
start_date=start_date, | start_date=start_date, |
metrics='ga:pageviews', | metrics='ga:pageviews', |
sort='-ga:pageviews', | sort='-ga:pageviews', |
dimensions="ga:operatingSystem,ga:operatingSystemVersion", | dimensions="ga:operatingSystem,ga:operatingSystemVersion", |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
result_data = results.get('rows') | result_data = results.get('rows') |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
data[result[0]] = data.get(result[0], 0) + int(result[2]) | data[result[0]] = data.get(result[0], 0) + int(result[2]) |
self._filter_out_long_tail(data, MIN_VIEWS) | self._filter_out_long_tail(data, MIN_VIEWS) |
ga_model.update_sitewide_stats(period_name, "Operating Systems", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Operating Systems", data, period_complete_day) |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
if int(result[2]) >= MIN_VIEWS: | if int(result[2]) >= MIN_VIEWS: |
key = "%s %s" % (result[0],result[1]) | key = "%s %s" % (result[0],result[1]) |
data[key] = result[2] | data[key] = result[2] |
ga_model.update_sitewide_stats(period_name, "Operating Systems versions", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Operating Systems versions", data, period_complete_day) |
def _browser_stats(self, start_date, end_date, period_name, period_complete_day): | def _browser_stats(self, start_date, end_date, period_name, period_complete_day): |
""" Information about browsers and browser versions """ | """ Information about browsers and browser versions """ |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
start_date=start_date, | start_date=start_date, |
metrics='ga:pageviews', | metrics='ga:pageviews', |
sort='-ga:pageviews', | sort='-ga:pageviews', |
dimensions="ga:browser,ga:browserVersion", | dimensions="ga:browser,ga:browserVersion", |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
result_data = results.get('rows') | result_data = results.get('rows') |
# e.g. [u'Firefox', u'19.0', u'20'] | # e.g. [u'Firefox', u'19.0', u'20'] |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
data[result[0]] = data.get(result[0], 0) + int(result[2]) | data[result[0]] = data.get(result[0], 0) + int(result[2]) |
self._filter_out_long_tail(data, MIN_VIEWS) | self._filter_out_long_tail(data, MIN_VIEWS) |
ga_model.update_sitewide_stats(period_name, "Browsers", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Browsers", data, period_complete_day) |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
key = "%s %s" % (result[0], self._filter_browser_version(result[0], result[1])) | key = "%s %s" % (result[0], self._filter_browser_version(result[0], result[1])) |
data[key] = data.get(key, 0) + int(result[2]) | data[key] = data.get(key, 0) + int(result[2]) |
self._filter_out_long_tail(data, MIN_VIEWS) | self._filter_out_long_tail(data, MIN_VIEWS) |
ga_model.update_sitewide_stats(period_name, "Browser versions", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Browser versions", data, period_complete_day) |
@classmethod | @classmethod |
def _filter_browser_version(cls, browser, version_str): | def _filter_browser_version(cls, browser, version_str): |
''' | ''' |
Simplifies a browser version string if it is detailed. | Simplifies a browser version string if it is detailed. |
i.e. groups together Firefox 3.5.1 and 3.5.2 to be just 3. | i.e. groups together Firefox 3.5.1 and 3.5.2 to be just 3. |
This is helpful when viewing stats and good to protect privacy. | This is helpful when viewing stats and good to protect privacy. |
''' | ''' |
ver = version_str | ver = version_str |
parts = ver.split('.') | parts = ver.split('.') |
if len(parts) > 1: | if len(parts) > 1: |
if parts[1][0] == '0': | if parts[1][0] == '0': |
ver = parts[0] | ver = parts[0] |
else: | else: |
ver = "%s" % (parts[0]) | ver = "%s" % (parts[0]) |
# Special case complex version nums | # Special case complex version nums |
if browser in ['Safari', 'Android Browser']: | if browser in ['Safari', 'Android Browser']: |
ver = parts[0] | ver = parts[0] |
if len(ver) > 2: | if len(ver) > 2: |
num_hidden_digits = len(ver) - 2 | num_hidden_digits = len(ver) - 2 |
ver = ver[0] + ver[1] + 'X' * num_hidden_digits | ver = ver[0] + ver[1] + 'X' * num_hidden_digits |
return ver | return ver |
def _mobile_stats(self, start_date, end_date, period_name, period_complete_day): | def _mobile_stats(self, start_date, end_date, period_name, period_complete_day): |
""" Info about mobile devices """ | """ Info about mobile devices """ |
results = self.service.data().ga().get( | results = self.service.data().ga().get( |
ids='ga:' + self.profile_id, | ids='ga:' + self.profile_id, |
start_date=start_date, | start_date=start_date, |
metrics='ga:pageviews', | metrics='ga:pageviews', |
sort='-ga:pageviews', | sort='-ga:pageviews', |
dimensions="ga:mobileDeviceBranding, ga:mobileDeviceInfo", | dimensions="ga:mobileDeviceBranding, ga:mobileDeviceInfo", |
max_results=10000, | max_results=10000, |
end_date=end_date).execute() | end_date=end_date).execute() |
result_data = results.get('rows') | result_data = results.get('rows') |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
data[result[0]] = data.get(result[0], 0) + int(result[2]) | data[result[0]] = data.get(result[0], 0) + int(result[2]) |
self._filter_out_long_tail(data, MIN_VIEWS) | self._filter_out_long_tail(data, MIN_VIEWS) |
ga_model.update_sitewide_stats(period_name, "Mobile brands", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Mobile brands", data, period_complete_day) |
data = {} | data = {} |
for result in result_data: | for result in result_data: |
data[result[1]] = data.get(result[1], 0) + int(result[2]) | data[result[1]] = data.get(result[1], 0) + int(result[2]) |
self._filter_out_long_tail(data, MIN_VIEWS) | self._filter_out_long_tail(data, MIN_VIEWS) |
ga_model.update_sitewide_stats(period_name, "Mobile devices", data, period_complete_day) | ga_model.update_sitewide_stats(period_name, "Mobile devices", data, period_complete_day) |
@classmethod | @classmethod |
def _filter_out_long_tail(cls, data, threshold=10): | def _filter_out_long_tail(cls, data, threshold=10): |
''' | ''' |
Given data which is a frequency distribution, filter out | Given data which is a frequency distribution, filter out |
results which are below a threshold count. This is good to protect | results which are below a threshold count. This is good to protect |
privacy. | privacy. |
''' | ''' |
for key, value in data.items(): | for key, value in data.items(): |
if value < threshold: | if value < threshold: |
del data[key] | del data[key] |
import re | import re |
import uuid | import uuid |
from sqlalchemy import Table, Column, MetaData, ForeignKey | from sqlalchemy import Table, Column, MetaData, ForeignKey |
from sqlalchemy import types | from sqlalchemy import types |
from sqlalchemy.sql import select | from sqlalchemy.sql import select |
from sqlalchemy.orm import mapper, relation | from sqlalchemy.orm import mapper, relation |
from sqlalchemy import func | from sqlalchemy import func |
import ckan.model as model | import ckan.model as model |
from ckan.lib.base import * | from ckan.lib.base import * |
log = __import__('logging').getLogger(__name__) | log = __import__('logging').getLogger(__name__) |
def make_uuid(): | def make_uuid(): |
return unicode(uuid.uuid4()) | return unicode(uuid.uuid4()) |
metadata = MetaData() | metadata = MetaData() |
class GA_Url(object): | class GA_Url(object): |
def __init__(self, **kwargs): | def __init__(self, **kwargs): |
for k,v in kwargs.items(): | for k,v in kwargs.items(): |
setattr(self, k, v) | setattr(self, k, v) |
url_table = Table('ga_url', metadata, | url_table = Table('ga_url', metadata, |
Column('id', types.UnicodeText, primary_key=True, | Column('id', types.UnicodeText, primary_key=True, |
default=make_uuid), | default=make_uuid), |
Column('period_name', types.UnicodeText), | Column('period_name', types.UnicodeText), |
Column('period_complete_day', types.Integer), | Column('period_complete_day', types.Integer), |
Column('pageviews', types.UnicodeText), | Column('pageviews', types.UnicodeText), |
Column('visits', types.UnicodeText), | Column('visits', types.UnicodeText), |
Column('url', types.UnicodeText), | Column('url', types.UnicodeText), |
Column('department_id', types.UnicodeText), | Column('department_id', types.UnicodeText), |
Column('package_id', types.UnicodeText), | Column('package_id', types.UnicodeText), |
) | ) |
mapper(GA_Url, url_table) | mapper(GA_Url, url_table) |
class GA_Stat(object): | class GA_Stat(object): |
def __init__(self, **kwargs): | def __init__(self, **kwargs): |
for k,v in kwargs.items(): | for k,v in kwargs.items(): |
setattr(self, k, v) | setattr(self, k, v) |
stat_table = Table('ga_stat', metadata, | stat_table = Table('ga_stat', metadata, |
Column('id', types.UnicodeText, primary_key=True, | Column('id', types.UnicodeText, primary_key=True, |
default=make_uuid), | default=make_uuid), |
Column('period_name', types.UnicodeText), | Column('period_name', types.UnicodeText), |
Column('period_complete_day', types.UnicodeText), | Column('period_complete_day', types.UnicodeText), |
Column('stat_name', types.UnicodeText), | Column('stat_name', types.UnicodeText), |
Column('key', types.UnicodeText), | Column('key', types.UnicodeText), |
Column('value', types.UnicodeText), ) | Column('value', types.UnicodeText), ) |
mapper(GA_Stat, stat_table) | mapper(GA_Stat, stat_table) |
class GA_Publisher(object): | class GA_Publisher(object): |
def __init__(self, **kwargs): | def __init__(self, **kwargs): |
for k,v in kwargs.items(): | for k,v in kwargs.items(): |
setattr(self, k, v) | setattr(self, k, v) |
pub_table = Table('ga_publisher', metadata, | pub_table = Table('ga_publisher', metadata, |
Column('id', types.UnicodeText, primary_key=True, | Column('id', types.UnicodeText, primary_key=True, |
default=make_uuid), | default=make_uuid), |
Column('period_name', types.UnicodeText), | Column('period_name', types.UnicodeText), |
Column('publisher_name', types.UnicodeText), | Column('publisher_name', types.UnicodeText), |
Column('views', types.UnicodeText), | Column('views', types.UnicodeText), |
Column('visits', types.UnicodeText), | Column('visits', types.UnicodeText), |
Column('toplevel', types.Boolean, default=False), | Column('toplevel', types.Boolean, default=False), |
Column('subpublishercount', types.Integer, default=0), | Column('subpublishercount', types.Integer, default=0), |
Column('parent', types.UnicodeText), | Column('parent', types.UnicodeText), |
) | ) |
mapper(GA_Publisher, pub_table) | mapper(GA_Publisher, pub_table) |
class GA_ReferralStat(object): | class GA_ReferralStat(object): |
def __init__(self, **kwargs): | def __init__(self, **kwargs): |
for k,v in kwargs.items(): | for k,v in kwargs.items(): |
setattr(self, k, v) | setattr(self, k, v) |
referrer_table = Table('ga_referrer', metadata, | referrer_table = Table('ga_referrer', metadata, |
Column('id', types.UnicodeText, primary_key=True, | Column('id', types.UnicodeText, primary_key=True, |
default=make_uuid), | default=make_uuid), |
Column('period_name', types.UnicodeText), | Column('period_name', types.UnicodeText), |
Column('source', types.UnicodeText), | Column('source', types.UnicodeText), |
Column('url', types.UnicodeText), | Column('url', types.UnicodeText), |
Column('count', types.Integer), | Column('count', types.Integer), |
) | ) |
mapper(GA_ReferralStat, referrer_table) | mapper(GA_ReferralStat, referrer_table) |
def init_tables(): | def init_tables(): |
metadata.create_all(model.meta.engine) | metadata.create_all(model.meta.engine) |
cached_tables = {} | cached_tables = {} |
def get_table(name): | def get_table(name): |
if name not in cached_tables: | if name not in cached_tables: |
meta = MetaData() | meta = MetaData() |
meta.reflect(bind=model.meta.engine) | meta.reflect(bind=model.meta.engine) |
table = meta.tables[name] | table = meta.tables[name] |
cached_tables[name] = table | cached_tables[name] = table |
return cached_tables[name] | return cached_tables[name] |
def _normalize_url(url): | def _normalize_url(url): |
'''Strip off the hostname etc. Do this before storing it. | '''Strip off the hostname etc. Do this before storing it. |
>>> normalize_url('http://data.gov.uk/dataset/weekly_fuel_prices') | >>> normalize_url('http://data.gov.uk/dataset/weekly_fuel_prices') |
'/dataset/weekly_fuel_prices' | '/dataset/weekly_fuel_prices' |
''' | ''' |
return '/' + '/'.join(url.split('/')[3:]) | return '/' + '/'.join(url.split('/')[3:]) |
def _get_package_and_publisher(url): | def _get_package_and_publisher(url): |
# e.g. /dataset/fuel_prices | # e.g. /dataset/fuel_prices |
# e.g. /dataset/fuel_prices/resource/e63380d4 | # e.g. /dataset/fuel_prices/resource/e63380d4 |
dataset_match = re.match('/dataset/([^/]+)(/.*)?', url) | dataset_match = re.match('/dataset/([^/]+)(/.*)?', url) |
if dataset_match: | if dataset_match: |
dataset_ref = dataset_match.groups()[0] | dataset_ref = dataset_match.groups()[0] |
dataset = model.Package.get(dataset_ref) | dataset = model.Package.get(dataset_ref) |
if dataset: | if dataset: |
publisher_groups = dataset.get_groups('publisher') | publisher_groups = dataset.get_groups('publisher') |
if publisher_groups: | if publisher_groups: |
return dataset_ref,publisher_groups[0].name | return dataset_ref,publisher_groups[0].name |
return dataset_ref, None | return dataset_ref, None |
else: | else: |
publisher_match = re.match('/publisher/([^/]+)(/.*)?', url) | publisher_match = re.match('/publisher/([^/]+)(/.*)?', url) |
if publisher_match: | if publisher_match: |
return None, publisher_match.groups()[0] | return None, publisher_match.groups()[0] |
return None, None | return None, None |
def update_sitewide_stats(period_name, stat_name, data, period_complete_day): | def update_sitewide_stats(period_name, stat_name, data, period_complete_day): |
for k,v in data.iteritems(): | for k,v in data.iteritems(): |
item = model.Session.query(GA_Stat).\ | item = model.Session.query(GA_Stat).\ |
filter(GA_Stat.period_name==period_name).\ | filter(GA_Stat.period_name==period_name).\ |
filter(GA_Stat.key==k).\ | filter(GA_Stat.key==k).\ |
filter(GA_Stat.stat_name==stat_name).first() | filter(GA_Stat.stat_name==stat_name).first() |
if item: | if item: |
item.period_name = period_name | item.period_name = period_name |
item.key = k | item.key = k |
item.value = v | item.value = v |
item.period_complete_day = period_complete_day | item.period_complete_day = period_complete_day |
model.Session.add(item) | model.Session.add(item) |
else: | else: |
# create the row | # create the row |
values = {'id': make_uuid(), | values = {'id': make_uuid(), |
'period_name': period_name, | 'period_name': period_name, |
'period_complete_day': period_complete_day, | 'period_complete_day': period_complete_day, |
'key': k, | 'key': k, |
'value': v, | 'value': v, |
'stat_name': stat_name | 'stat_name': stat_name |
} | } |
model.Session.add(GA_Stat(**values)) | model.Session.add(GA_Stat(**values)) |
model.Session.commit() | model.Session.commit() |
def pre_update_url_stats(period_name): | def pre_update_url_stats(period_name): |
model.Session.query(GA_Url).\ | q = model.Session.query(GA_Url).\ |
filter(GA_Url.period_name==period_name).delete() | filter(GA_Url.period_name==period_name) |
model.Session.query(GA_Url).\ | log.debug("Deleting %d '%s' records" % (q.count(), period_name)) |
filter(GA_Url.period_name=='All').delete() | 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): | def update_url_stats(period_name, period_complete_day, url_data): |
''' | ''' |
Given a list of urls and number of hits for each during a given period, | Given a list of urls and number of hits for each during a given period, |
stores them in GA_Url under the period and recalculates the totals for | stores them in GA_Url under the period and recalculates the totals for |
the 'All' period. | the 'All' period. |
''' | ''' |
progress_total = len(url_data) | |
progress_count = 0 | |
for url, views, visits in url_data: | 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) | package, publisher = _get_package_and_publisher(url) |
item = model.Session.query(GA_Url).\ | item = model.Session.query(GA_Url).\ |
filter(GA_Url.period_name==period_name).\ | filter(GA_Url.period_name==period_name).\ |
filter(GA_Url.url==url).first() | filter(GA_Url.url==url).first() |
if item: | if item: |
item.pageviews = item.pageviews + views | item.pageviews = item.pageviews + views |
item.visits = item.visits + visits | item.visits = item.visits + visits |
if not item.package_id: | if not item.package_id: |
item.package_id = package | item.package_id = package |
if not item.department_id: | if not item.department_id: |
item.department_id = publisher | item.department_id = publisher |
model.Session.add(item) | model.Session.add(item) |
else: | else: |
values = {'id': make_uuid(), | values = {'id': make_uuid(), |
'period_name': period_name, | 'period_name': period_name, |
'period_complete_day': period_complete_day, | 'period_complete_day': period_complete_day, |
'url': url, | 'url': url, |
'pageviews': views, | 'pageviews': views, |
'visits': visits, | 'visits': visits, |
'department_id': publisher, | 'department_id': publisher, |
'package_id': package | 'package_id': package |
} | } |
model.Session.add(GA_Url(**values)) | model.Session.add(GA_Url(**values)) |
model.Session.commit() | model.Session.commit() |
if package: | if package: |
old_pageviews, old_visits = 0, 0 | old_pageviews, old_visits = 0, 0 |
old = model.Session.query(GA_Url).\ | old = model.Session.query(GA_Url).\ |
filter(GA_Url.period_name=='All').\ | filter(GA_Url.period_name=='All').\ |
filter(GA_Url.url==url).all() | filter(GA_Url.url==url).all() |
old_pageviews = sum([int(o.pageviews) for o in old]) | old_pageviews = sum([int(o.pageviews) for o in old]) |
old_visits = sum([int(o.visits) for o in old]) | old_visits = sum([int(o.visits) for o in old]) |
entries = model.Session.query(GA_Url).\ | entries = model.Session.query(GA_Url).\ |
filter(GA_Url.period_name!='All').\ | filter(GA_Url.period_name!='All').\ |
filter(GA_Url.url==url).all() | filter(GA_Url.url==url).all() |
values = {'id': make_uuid(), | values = {'id': make_uuid(), |
'period_name': 'All', | 'period_name': 'All', |
'period_complete_day': 0, | 'period_complete_day': 0, |
'url': url, | 'url': url, |
'pageviews': sum([int(e.pageviews) for e in entries]) + old_pageviews, | 'pageviews': sum([int(e.pageviews) for e in entries]) + int(old_pageviews), |
'visits': sum([int(e.visits or 0) for e in entries]) + old_visits, | 'visits': sum([int(e.visits or 0) for e in entries]) + int(old_visits), |
'department_id': publisher, | 'department_id': publisher, |
'package_id': package | 'package_id': package |
} | } |
model.Session.add(GA_Url(**values)) | model.Session.add(GA_Url(**values)) |
model.Session.commit() | model.Session.commit() |
def update_social(period_name, data): | def update_social(period_name, data): |
# Clean up first. | # Clean up first. |
model.Session.query(GA_ReferralStat).\ | model.Session.query(GA_ReferralStat).\ |
filter(GA_ReferralStat.period_name==period_name).delete() | filter(GA_ReferralStat.period_name==period_name).delete() |
for url,data in data.iteritems(): | for url,data in data.iteritems(): |
for entry in data: | for entry in data: |
source = entry[0] | source = entry[0] |
count = entry[1] | count = entry[1] |
item = model.Session.query(GA_ReferralStat).\ | item = model.Session.query(GA_ReferralStat).\ |
filter(GA_ReferralStat.period_name==period_name).\ | filter(GA_ReferralStat.period_name==period_name).\ |
filter(GA_ReferralStat.source==source).\ | filter(GA_ReferralStat.source==source).\ |
filter(GA_ReferralStat.url==url).first() | filter(GA_ReferralStat.url==url).first() |
if item: | if item: |
item.count = item.count + count | item.count = item.count + count |
model.Session.add(item) | model.Session.add(item) |
else: | else: |
# create the row | # create the row |
values = {'id': make_uuid(), | values = {'id': make_uuid(), |
'period_name': period_name, | 'period_name': period_name, |
'source': source, | 'source': source, |
'url': url, | 'url': url, |
'count': count, | 'count': count, |
} | } |
model.Session.add(GA_ReferralStat(**values)) | model.Session.add(GA_ReferralStat(**values)) |
model.Session.commit() | model.Session.commit() |
def update_publisher_stats(period_name): | def update_publisher_stats(period_name): |
""" | """ |
Updates the publisher stats from the data retrieved for /dataset/* | Updates the publisher stats from the data retrieved for /dataset/* |
and /publisher/*. Will run against each dataset and generates the | and /publisher/*. Will run against each dataset and generates the |
totals for the entire tree beneath each publisher. | totals for the entire tree beneath each publisher. |
""" | """ |
toplevel = get_top_level() | toplevel = get_top_level() |
publishers = model.Session.query(model.Group).\ | publishers = model.Session.query(model.Group).\ |
filter(model.Group.type=='publisher').\ | filter(model.Group.type=='publisher').\ |
filter(model.Group.state=='active').all() | filter(model.Group.state=='active').all() |
for publisher in publishers: | for publisher in publishers: |
views, visits, subpub = update_publisher(period_name, publisher, publisher.name) | views, visits, subpub = update_publisher(period_name, publisher, publisher.name) |
parent, parents = '', publisher.get_groups('publisher') | parent, parents = '', publisher.get_groups('publisher') |
if parents: | if parents: |
parent = parents[0].name | parent = parents[0].name |
item = model.Session.query(GA_Publisher).\ | item = model.Session.query(GA_Publisher).\ |
filter(GA_Publisher.period_name==period_name).\ | filter(GA_Publisher.period_name==period_name).\ |
filter(GA_Publisher.publisher_name==publisher.name).first() | filter(GA_Publisher.publisher_name==publisher.name).first() |
if item: | if item: |
item.views = views | item.views = views |
item.visits = visits | item.visits = visits |
item.publisher_name = publisher.name | item.publisher_name = publisher.name |
item.toplevel = publisher in toplevel | item.toplevel = publisher in toplevel |
item.subpublishercount = subpub | item.subpublishercount = subpub |
item.parent = parent | item.parent = parent |
model.Session.add(item) | model.Session.add(item) |
else: | else: |
# create the row | # create the row |
values = {'id': make_uuid(), | values = {'id': make_uuid(), |
'period_name': period_name, | 'period_name': period_name, |
'publisher_name': publisher.name, | 'publisher_name': publisher.name, |
'views': views, | 'views': views, |
'visits': visits, | 'visits': visits, |
'toplevel': publisher in toplevel, | 'toplevel': publisher in toplevel, |
'subpublishercount': subpub, | 'subpublishercount': subpub, |
'parent': parent | 'parent': parent |
} | } |
model.Session.add(GA_Publisher(**values)) | model.Session.add(GA_Publisher(**values)) |
model.Session.commit() | model.Session.commit() |
def update_publisher(period_name, pub, part=''): | def update_publisher(period_name, pub, part=''): |
views,visits,subpub = 0, 0, 0 | views,visits,subpub = 0, 0, 0 |
for publisher in go_down_tree(pub): | for publisher in go_down_tree(pub): |
subpub = subpub + 1 | subpub = subpub + 1 |
items = model.Session.query(GA_Url).\ | items = model.Session.query(GA_Url).\ |
filter(GA_Url.period_name==period_name).\ | filter(GA_Url.period_name==period_name).\ |
filter(GA_Url.department_id==publisher.name).all() | filter(GA_Url.department_id==publisher.name).all() |
for item in items: | for item in items: |
views = views + int(item.pageviews) | views = views + int(item.pageviews) |
visits = visits + int(item.visits) | visits = visits + int(item.visits) |
return views, visits, (subpub-1) | return views, visits, (subpub-1) |
def get_top_level(): | def get_top_level(): |
'''Returns the top level publishers.''' | '''Returns the top level publishers.''' |
return model.Session.query(model.Group).\ | return model.Session.query(model.Group).\ |
outerjoin(model.Member, model.Member.table_id == model.Group.id and \ | outerjoin(model.Member, model.Member.table_id == model.Group.id and \ |
model.Member.table_name == 'group' and \ | model.Member.table_name == 'group' and \ |
model.Member.state == 'active').\ | model.Member.state == 'active').\ |
filter(model.Member.id==None).\ | filter(model.Member.id==None).\ |
filter(model.Group.type=='publisher').\ | filter(model.Group.type=='publisher').\ |
order_by(model.Group.name).all() | order_by(model.Group.name).all() |
def get_children(publisher): | def get_children(publisher): |
'''Finds child publishers for the given publisher (object). (Not recursive)''' | '''Finds child publishers for the given publisher (object). (Not recursive)''' |
from ckan.model.group import HIERARCHY_CTE | from ckan.model.group import HIERARCHY_CTE |
return model.Session.query(model.Group).\ | return model.Session.query(model.Group).\ |
from_statement(HIERARCHY_CTE).params(id=publisher.id, type='publisher').\ | from_statement(HIERARCHY_CTE).params(id=publisher.id, type='publisher').\ |
all() | all() |
def go_down_tree(publisher): | def go_down_tree(publisher): |
'''Provided with a publisher object, it walks down the hierarchy and yields each publisher, | '''Provided with a publisher object, it walks down the hierarchy and yields each publisher, |
including the one you supply.''' | including the one you supply.''' |
yield publisher | yield publisher |
for child in get_children(publisher): | for child in get_children(publisher): |
for grandchild in go_down_tree(child): | for grandchild in go_down_tree(child): |
yield grandchild | yield grandchild |
def delete(period_name): | def delete(period_name): |
''' | ''' |
Deletes table data for the specified period, or specify 'all' | Deletes table data for the specified period, or specify 'all' |
for all periods. | for all periods. |
''' | ''' |
for object_type in (GA_Url, GA_Stat, GA_Publisher, GA_ReferralStat): | for object_type in (GA_Url, GA_Stat, GA_Publisher, GA_ReferralStat): |
q = model.Session.query(object_type) | q = model.Session.query(object_type) |
if period_name != 'all': | if period_name != 'All': |
q = q.filter_by(period_name=period_name) | q = q.filter_by(period_name=period_name) |
q.delete() | q.delete() |
model.Session.commit() | model.repo.commit_and_remove() |
def get_score_for_dataset(dataset_name): | def get_score_for_dataset(dataset_name): |
''' | ''' |
Returns a "current popularity" score for a dataset, | Returns a "current popularity" score for a dataset, |
based on how many views it has had recently. | based on how many views it has had recently. |
''' | ''' |
import datetime | import datetime |
now = datetime.datetime.now() | now = datetime.datetime.now() |
last_month = now - datetime.timedelta(days=30) | last_month = now - datetime.timedelta(days=30) |
period_names = ['%s-%02d' % (last_month.year, last_month.month), | period_names = ['%s-%02d' % (last_month.year, last_month.month), |
'%s-%02d' % (now.year, now.month), | '%s-%02d' % (now.year, now.month), |
] | ] |
score = 0 | score = 0 |
for period_name in period_names: | for period_name in period_names: |
score /= 2 # previous periods are discounted by 50% | score /= 2 # previous periods are discounted by 50% |
entry = model.Session.query(GA_Url)\ | entry = model.Session.query(GA_Url)\ |
.filter(GA_Url.period_name==period_name)\ | .filter(GA_Url.period_name==period_name)\ |
.filter(GA_Url.package_id==dataset_name).first() | .filter(GA_Url.package_id==dataset_name).first() |
# score | # score |
if entry: | if entry: |
views = float(entry.pageviews) | views = float(entry.pageviews) |
if entry.period_complete_day: | if entry.period_complete_day: |
views_per_day = views / entry.period_complete_day | views_per_day = views / entry.period_complete_day |
else: | else: |
views_per_day = views / 15 # guess | views_per_day = views / 15 # guess |
score += views_per_day | score += views_per_day |
score = int(score * 100) | score = int(score * 100) |
log.debug('Popularity %s: %s', score, dataset_name) | log.debug('Popularity %s: %s', score, dataset_name) |
return score | return score |
import logging | import logging |
import operator | import operator |
import ckan.lib.base as base | import ckan.lib.base as base |
import ckan.model as model | import ckan.model as model |
from ckan.logic import get_action | from ckan.logic import get_action |
from ckanext.ga_report.ga_model import GA_Url, GA_Publisher | from ckanext.ga_report.ga_model import GA_Url, GA_Publisher |
from ckanext.ga_report.controller import _get_publishers | from ckanext.ga_report.controller import _get_publishers |
_log = logging.getLogger(__name__) | _log = logging.getLogger(__name__) |
def popular_datasets(count=10): | def popular_datasets(count=10): |
import random | import random |
publisher = None | publisher = None |
publishers = _get_publishers(30) | publishers = _get_publishers(30) |
total = len(publishers) | total = len(publishers) |
while not publisher or not datasets: | while not publisher or not datasets: |
rand = random.randrange(0, total) | rand = random.randrange(0, total) |
publisher = publishers[rand][0] | publisher = publishers[rand][0] |
if not publisher.state == 'active': | if not publisher.state == 'active': |
publisher = None | publisher = None |
continue | continue |
datasets = _datasets_for_publisher(publisher, 10)[:count] | datasets = _datasets_for_publisher(publisher, 10)[:count] |
ctx = { | ctx = { |
'datasets': datasets, | 'datasets': datasets, |
'publisher': publisher | 'publisher': publisher |
} | } |
return base.render_snippet('ga_report/ga_popular_datasets.html', **ctx) | return base.render_snippet('ga_report/ga_popular_datasets.html', **ctx) |
def single_popular_dataset(top=20): | def single_popular_dataset(top=20): |
'''Returns a random dataset from the most popular ones. | '''Returns a random dataset from the most popular ones. |
:param top: the number of top datasets to select from | :param top: the number of top datasets to select from |
''' | ''' |
import random | import random |
top_datasets = model.Session.query(GA_Url).\ | top_datasets = model.Session.query(GA_Url).\ |
filter(GA_Url.url.like('/dataset/%')).\ | filter(GA_Url.url.like('/dataset/%')).\ |
order_by('ga_url.pageviews::int desc') | order_by('ga_url.pageviews::int desc') |
num_top_datasets = top_datasets.count() | num_top_datasets = top_datasets.count() |
dataset = None | dataset = None |
if num_top_datasets: | if num_top_datasets: |
count = 0 | count = 0 |
while not dataset: | while not dataset: |
rand = random.randrange(0, min(top, num_top_datasets)) | rand = random.randrange(0, min(top, num_top_datasets)) |
ga_url = top_datasets[rand] | ga_url = top_datasets[rand] |
dataset = model.Package.get(ga_url.url[len('/dataset/'):]) | dataset = model.Package.get(ga_url.url[len('/dataset/'):]) |
if dataset and not dataset.state == 'active': | if dataset and not dataset.state == 'active': |
dataset = None | dataset = None |
# When testing, it is possible that top datasets are not available | # When testing, it is possible that top datasets are not available |
# so only go round this loop a few times before falling back on | # so only go round this loop a few times before falling back on |
# a random dataset. | # a random dataset. |
count += 1 | count += 1 |
if count > 10: | if count > 10: |
break | break |
if not dataset: | if not dataset: |
# fallback | # fallback |
dataset = model.Session.query(model.Package)\ | dataset = model.Session.query(model.Package)\ |
.filter_by(state='active').first() | .filter_by(state='active').first() |
if not dataset: | if not dataset: |
return None | return None |
dataset_dict = get_action('package_show')({'model': model, | dataset_dict = get_action('package_show')({'model': model, |
'session': model.Session, | 'session': model.Session, |
'validate': False}, | 'validate': False}, |
{'id':dataset.id}) | {'id':dataset.id}) |
return dataset_dict | return dataset_dict |
def single_popular_dataset_html(top=20): | def single_popular_dataset_html(top=20): |
dataset_dict = single_popular_dataset(top) | dataset_dict = single_popular_dataset(top) |
groups = package.get('groups', []) | groups = package.get('groups', []) |
publishers = [ g for g in groups if g.get('type') == 'publisher' ] | publishers = [ g for g in groups if g.get('type') == 'publisher' ] |
publisher = publishers[0] if publishers else {'name':'', 'title': ''} | publisher = publishers[0] if publishers else {'name':'', 'title': ''} |
context = { | context = { |
'dataset': dataset_dict, | 'dataset': dataset_dict, |
'publisher': publisher_dict | 'publisher': publisher_dict |
} | } |
return base.render_snippet('ga_report/ga_popular_single.html', **context) | 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): |
if not publisher: | if not publisher: |
_log.error("No valid publisher passed to 'most_popular_datasets'") | _log.error("No valid publisher passed to 'most_popular_datasets'") |
return "" | return "" |
results = _datasets_for_publisher(publisher, count) | results = _datasets_for_publisher(publisher, count) |
ctx = { | ctx = { |
'dataset_count': len(results), | 'dataset_count': len(results), |
'datasets': results, | 'datasets': results, |
'publisher': publisher | 'publisher': publisher |
} | } |
return base.render_snippet('ga_report/publisher/popular.html', **ctx) | return base.render_snippet('ga_report/publisher/popular.html', **ctx) |
def _datasets_for_publisher(publisher, count): | def _datasets_for_publisher(publisher, count): |
datasets = {} | datasets = {} |
entries = model.Session.query(GA_Url).\ | entries = model.Session.query(GA_Url).\ |
filter(GA_Url.department_id==publisher.name).\ | filter(GA_Url.department_id==publisher.name).\ |
filter(GA_Url.url.like('/dataset/%')).\ | filter(GA_Url.url.like('/dataset/%')).\ |
order_by('ga_url.pageviews::int desc').all() | order_by('ga_url.pageviews::int desc').all() |
for entry in entries: | for entry in entries: |
if len(datasets) < count: | if len(datasets) < count: |
p = model.Package.get(entry.url[len('/dataset/'):]) | 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 in datasets: | if not p in datasets: |
datasets[p] = {'views':0, 'visits': 0} | datasets[p] = {'views':0, 'visits': 0} |
datasets[p]['views'] = datasets[p]['views'] + int(entry.pageviews) | datasets[p]['views'] = datasets[p]['views'] + int(entry.pageviews) |
datasets[p]['visits'] = datasets[p]['visits'] + int(entry.visits) | datasets[p]['visits'] = datasets[p]['visits'] + int(entry.visits) |
results = [] | results = [] |
for k, v in datasets.iteritems(): | for k, v in datasets.iteritems(): |
results.append((k,v['views'],v['visits'])) | results.append((k,v['views'],v['visits'])) |
return sorted(results, key=operator.itemgetter(1), reverse=True) | return sorted(results, key=operator.itemgetter(1), reverse=True) |
import logging | import logging |
import ckan.lib.helpers as h | import ckan.lib.helpers as h |
import ckan.plugins as p | import ckan.plugins as p |
from ckan.plugins import implements, toolkit | from ckan.plugins import implements, toolkit |
from ckanext.ga_report.helpers import (most_popular_datasets, | from ckanext.ga_report.helpers import (most_popular_datasets, |
popular_datasets, | popular_datasets, |
single_popular_dataset) | single_popular_dataset) |
log = logging.getLogger('ckanext.ga-report') | log = logging.getLogger('ckanext.ga-report') |
class GAReportPlugin(p.SingletonPlugin): | class GAReportPlugin(p.SingletonPlugin): |
implements(p.IConfigurer, inherit=True) | implements(p.IConfigurer, inherit=True) |
implements(p.IRoutes, inherit=True) | implements(p.IRoutes, inherit=True) |
implements(p.ITemplateHelpers, inherit=True) | implements(p.ITemplateHelpers, inherit=True) |
def update_config(self, config): | def update_config(self, config): |
toolkit.add_template_directory(config, 'templates') | toolkit.add_template_directory(config, 'templates') |
toolkit.add_public_directory(config, 'public') | toolkit.add_public_directory(config, 'public') |
def get_helpers(self): | def get_helpers(self): |
""" | """ |
A dictionary of extra helpers that will be available to provide | A dictionary of extra helpers that will be available to provide |
ga report info to templates. | ga report info to templates. |
""" | """ |
return { | return { |
'ga_report_installed': lambda: True, | 'ga_report_installed': lambda: True, |
'popular_datasets': popular_datasets, | 'popular_datasets': popular_datasets, |
'most_popular_datasets': most_popular_datasets, | 'most_popular_datasets': most_popular_datasets, |
'single_popular_dataset': single_popular_dataset | 'single_popular_dataset': single_popular_dataset |
} | } |
def after_map(self, map): | def after_map(self, map): |
# GaReport | # GaReport |
map.connect( | map.connect( |
'/data/site-usage', | '/data/site-usage', |
controller='ckanext.ga_report.controller:GaReport', | controller='ckanext.ga_report.controller:GaReport', |
action='index' | action='index' |
) | ) |
map.connect( | map.connect( |
'/data/site-usage/data_{month}.csv', | '/data/site-usage/data_{month}.csv', |
controller='ckanext.ga_report.controller:GaReport', | controller='ckanext.ga_report.controller:GaReport', |
action='csv' | 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 | # GaDatasetReport |
map.connect( | map.connect( |
'/data/site-usage/publisher', | '/data/site-usage/publisher', |
controller='ckanext.ga_report.controller:GaDatasetReport', | controller='ckanext.ga_report.controller:GaDatasetReport', |
action='publishers' | action='publishers' |
) | ) |
map.connect( | map.connect( |
'/data/site-usage/publishers_{month}.csv', | '/data/site-usage/publishers_{month}.csv', |
controller='ckanext.ga_report.controller:GaDatasetReport', | controller='ckanext.ga_report.controller:GaDatasetReport', |
action='publisher_csv' | action='publisher_csv' |
) | ) |
map.connect( | map.connect( |
'/data/site-usage/dataset/datasets_{id}_{month}.csv', | '/data/site-usage/dataset/datasets_{id}_{month}.csv', |
controller='ckanext.ga_report.controller:GaDatasetReport', | controller='ckanext.ga_report.controller:GaDatasetReport', |
action='dataset_csv' | action='dataset_csv' |
) | ) |
map.connect( | map.connect( |
'/data/site-usage/dataset', | '/data/site-usage/dataset', |
controller='ckanext.ga_report.controller:GaDatasetReport', | controller='ckanext.ga_report.controller:GaDatasetReport', |
action='read' | action='read' |
) | ) |
map.connect( | map.connect( |
'/data/site-usage/dataset/{id}', | '/data/site-usage/dataset/{id}', |
controller='ckanext.ga_report.controller:GaDatasetReport', | controller='ckanext.ga_report.controller:GaDatasetReport', |
action='read_publisher' | action='read_publisher' |
) | ) |
return map | return map |
.table-condensed td.sparkline-cell { | |
padding: 1px 0 0 0; | |
width: 108px; | |
text-align: center; | |
} | |
.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%; | |
} | |
.rickshaw_y_axis { | |
position: absolute; | |
top: 0; | |
bottom: 0; | |
width: 40px; | |
} | |
.rickshaw_legend .label { | |
background: transparent !important; | |
color: #000000 !important; | |
font-weight: normal !important; | |
} | |
var CKAN = CKAN || {}; | |
CKAN.GA_Reports = {}; | |
CKAN.GA_Reports.render_rickshaw = function( css_name, data, mode, colorscheme ) { | |
var graphLegends = $('#graph-legend-container'); | |
if (!Modernizr.svg) { | |
$("#chart_"+css_name) | |
.html( '<div class="alert">Your browser does not support vector graphics. No graphs can be rendered.</div>') | |
.closest('.rickshaw_chart_container').css('height',50); | |
var myLegend = $('<div id="legend_'+css_name+'"/>') | |
.html('(Graph cannot be rendered)') | |
.appendTo(graphLegends); | |
return; | |
} | |
var myLegend = $('<div id="legend_'+css_name+'"/>').appendTo(graphLegends); | |
var palette = new Rickshaw.Color.Palette( { scheme: colorscheme } ); | |
$.each(data, function(i, object) { | |
object['color'] = palette.color(); | |
}); | |
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 hoverDetail = new Rickshaw.Graph.HoverDetail( { | |
graph: graph, | |
formatter: function(series, x, y) { | |
var date = '<span class="date">' + new Date(x * 1000).toUTCString() + '</span>'; | |
var swatch = '<span class="detail_swatch" style="background-color: ' + series.color + '"></span>'; | |
var content = swatch + series.name + ": " + parseInt(y) + '<br>' + date; | |
return content; | |
} | |
} ); | |
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. | |
*/ | |
$('a[href="#totals"]').on( | |
'shown', | |
function() { | |
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); | |
} | |
); | |
}; | |
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="hashchange"]').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(); | |
$(legend_name).show(); | |
} | |
); | |
}; | |
/* | |
* Custom bootstrap plugin for handling data-toggle="hashchange". | |
* Behaves like data-toggle="tab" but I respond to the hashchange. | |
* Page state is memo-ized in the URL this way. Why doesn't Bootstrap do this? | |
*/ | |
$(function() { | |
var mapping = {}; | |
$('a[data-toggle="hashchange"]').each( | |
function(i,link) { | |
link = $(link); | |
mapping[link.attr('href')] = link; | |
} | |
); | |
$(window).hashchange(function() { | |
var link = mapping[window.location.hash]; | |
if (link) { link.tab('show'); } | |
}); | |
}); | |
/* | |
* Collection of shims to allow d3 and Rickshaw to load, error-free | |
* (but ultimately unusable) on Internet Explorer 7. The browser's | |
* API lacks several crucial functions which these libraries depend | |
* upon to load; we try to hide these errors from the user. | |
* | |
* With thanks to Array functions from: | |
* http://stackoverflow.com/questions/2790001/fixing-javascript-array-functions-in-internet-explorer-indexof-foreach-etc | |
* | |
* Use (Modernizr.svg==true) to detect whether it's okay to draw a graph. | |
*/ | |
'use strict'; | |
window.Element = window.Element || {'prototype': {}}; | |
window.CSSStyleDeclaration = window.CSSStyleDeclaration || {'prototype':{}}; | |
// Add ECMA262-5 method binding if not supported natively | |
// | |
if (!('bind' in Function.prototype)) { | |
Function.prototype.bind= function(owner) { | |
var that= this; | |
if (arguments.length<=1) { | |
return function() { | |
return that.apply(owner, arguments); | |
}; | |
} else { | |
var args= Array.prototype.slice.call(arguments, 1); | |
return function() { | |
return that.apply(owner, arguments.length===0? args : args.concat(Array.prototype.slice.call(arguments))); | |
}; | |
} | |
}; | |
} | |
// Add ECMA262-5 string trim if not supported natively | |
// | |
if (!('trim' in String.prototype)) { | |
String.prototype.trim= function() { | |
return this.replace(/^\s+/, '').replace(/\s+$/, ''); | |
}; | |
} | |
// 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; i<n; i++) | |
if (i in this && this[i]===find) | |
return i; | |
return -1; | |
}; | |
} | |
if (!('lastIndexOf' in Array.prototype)) { | |
Array.prototype.lastIndexOf= function(find, i /*opt*/) { | |
if (i===undefined) i= this.length-1; | |
if (i<0) i+= this.length; | |
if (i>this.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; i<n; i++) | |
if (i in this) | |
action.call(that, this[i], i, this); | |
}; | |
} | |
if (!('map' in Array.prototype)) { | |
Array.prototype.map= function(mapper, that /*opt*/) { | |
var other= new Array(this.length); | |
for (var i= 0, n= this.length; i<n; i++) | |
if (i in this) | |
other[i]= mapper.call(that, this[i], i, this); | |
return other; | |
}; | |
} | |
if (!('filter' in Array.prototype)) { | |
Array.prototype.filter= function(filter, that /*opt*/) { | |
var other= [], v; | |
for (var i=0, n= this.length; i<n; i++) | |
if (i in this && filter.call(that, v= this[i], i, this)) | |
other.push(v); | |
return other; | |
}; | |
} | |
if (!('every' in Array.prototype)) { | |
Array.prototype.every= function(tester, that /*opt*/) { | |
for (var i= 0, n= this.length; i<n; i++) | |
if (i in this && !tester.call(that, this[i], i, this)) | |
return false; | |
return true; | |
}; | |
} | |
if (!('some' in Array.prototype)) { | |
Array.prototype.some= function(tester, that /*opt*/) { | |
for (var i= 0, n= this.length; i<n; i++) | |
if (i in this && tester.call(that, this[i], i, this)) | |
return true; | |
return false; | |
}; | |
} | |
(function(){function a(a){var b=a.source,d=a.target,e=c(b,d),f=[b];while(b!==e)b=b.parent,f.push(b);var g=f.length;while(d!==e)f.splice(g,0,d),d=d.parent;return f}function b(a){var b=[],c=a.parent;while(c!=null)b.push(a),a=c,c=c.parent;return b.push(a),b}function c(a,c){if(a===c)return a;var d=b(a),e=b(c),f=d.pop(),g=e.pop(),h=null;while(f===g)h=f,f=d.pop(),g=e.pop();return h}function g(a){a.fixed|=2}function h(a){a!==f&&(a.fixed&=1)}function i(){j(),f.fixed&=1,e=f=null}function j(){f.px+=d3.event.dx,f.py+=d3.event.dy,e.resume()}function k(a,b,c){var d=0,e=0;a.charge=0;if(!a.leaf){var f=a.nodes,g=f.length,h=-1,i;while(++h<g){i=f[h];if(i==null)continue;k(i,b,c),a.charge+=i.charge,d+=i.charge*i.cx,e+=i.charge*i.cy}}if(a.point){a.leaf||(a.point.x+=Math.random()-.5,a.point.y+=Math.random()-.5);var j=b*c[a.point.index];a.charge+=a.pointCharge=j,d+=j*a.point.x,e+=j*a.point.y}a.cx=d/a.charge,a.cy=e/a.charge}function l(a){return 20}function m(a){return 1}function o(a){return a.x}function p(a){return a.y}function q(a,b,c){a.y0=b,a.y=c}function t(a){var b=1,c=0,d=a[0][1],e,f=a.length;for(;b<f;++b)(e=a[b][1])>d&&(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;m<f;m++){O(g,h,i=a[m]);var n=0,o=1,p=1;for(j=h._pack_next;j!==h;j=j._pack_next,o++)if(I(j,i)){n=1;break}if(n==1)for(k=g._pack_prev;k!==j._pack_prev;k=k._pack_prev,p++)if(I(k,i)){p<o&&(n=-1,j=k);break}n==0?(G(g,i),h=i,l(i)):n>0?(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;m<f;m++){var t=a[m];t.x-=q,t.y-=r,s=Math.max(s,t.r+Math.sqrt(t.x*t.x+t.y*t.y))}return a.forEach(L),s}function K(a){a._pack_next=a._pack_prev=a}function L(a){delete a._pack_next,delete a._pack_prev}function M(a){var b=a.children;b&&b.length?(b.forEach(M),a.r=J(b)):a.r=Math.sqrt(a.value)}function N(a,b,c,d){var e=a.children;a.x=b+=d*a.x,a.y=c+=d*a.y,a.r*=d;if(e){var f=-1,g=e.length;while(++f<g)N(e[f],b,c,d)}}function O(a,b,c){var d=a.r+c.r,e=b.x-a.x,f=b.y-a.y;if(d&&(e||f)){var g=b.r+c.r,h=Math.sqrt(e*e+f*f),i=Math.max(-1,Math.min(1,(d*d+h*h-g*g)/(2*d*h))),j=Math.acos(i),k=i*(d/=h),l=Math.sin(j)*d;c.x=a.x+k*e+l*f,c.y=a.y+k*f-l*e}else c.x=a.x+d,c.y=a.y}function P(a){return 1+d3.max(a,function(a){return a.y})}function Q(a){return a.reduce(function(a,b){return a+b.x},0)/a.length}function R(a){var b=a.children;return b&&b.length?R(b[0]):a}function S(a){var b=a.children,c;return b&&(c=b.length)?S(b[c-1]):a}function T(a,b){return a.parent==b.parent?1:2}function U(a){var b=a.children;return b&&b.length?b[0]:a._tree.thread}function V(a){var b=a.children,c;return b&&(c=b.length)?b[c-1]:a._tree.thread}function W(a,b){var c=a.children;if(c&&(e=c.length)){var d,e,f=-1;while(++f<e)b(d=W(c[f],b),a)>0&&(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<i)f=e[h],c(f,g),g=f}b(a,d)}c(a,null)}function _(a){var b=0,c=0,d=a.children,e=d.length,f;while(--e>=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(++d<e)c.push(a(b[d]));return c}},d3.layout.chord=function(){function j(){var a={},j=[],l=d3.range(e),m=[],n,o,p,q,r;b=[],c=[],n=0,q=-1;while(++q<e){o=0,r=-1;while(++r<e)o+=d[q][r];j.push(o),m.push(d3.range(e)),n+=o}g&&l.sort(function(a,b){return g(j[a],j[b])}),h&&m.forEach(function(a,b){a.sort(function(a,c){return h(d[b][a],d[b][c])})}),n=(2*Math.PI-f*e)/n,o=0,q=-1;while(++q<e){p=o,r=-1;while(++r<e){var s=l[q],t=m[s][r],u=d[s][t],v=o,w=o+=u*n;a[s+"-"+t]={index:s,subindex:t,startAngle:v,endAngle:w,value:u}}c.push({index:s,startAngle:p,endAngle:o,value:(o-p)/n}),o+=f}q=-1;while(++q<e){r=q-1;while(++r<e){var x=a[q+"-"+r],y=a[r+"-"+q];(x.value||y.value)&&b.push(x.value<y.value?{source:y,target:x}:{source:x,target:y})}}i&&k()}function k(){b.sort(function(a,b){return i((a.source.value+a.target.value)/2,(b.source.value+b.target.value)/2)})}var a={},b,c,d,e,f=0,g,h,i;return a.matrix=function(f){return arguments.length?(e=(d=f)&&d.length,b=c=null,a):d},a.padding=function(d){return arguments.length?(f=d,b=c=null,a):f},a.sortGroups=function(d){return arguments.length?(g=d,b=c=null,a):g},a.sortSubgroups=function(c){return arguments.length?(h=c,b=null,a):h},a.sortChords=function(c){return arguments.length?(i=c,b&&k(),a):i},a.chords=function(){return b||j(),b},a.groups=function(){return c||j(),c},a},d3.layout.force=function(){function A(a){return function(b,c,d,e,f){if(b.point!==a){var g=b.cx-a.x,h=b.cy-a.y,i=1/Math.sqrt(g*g+h*h);if((e-c)*i<t){var j=b.charge*i*i;return a.px-=g*j,a.py-=h*j,!0}if(b.point&&isFinite(i)){var j=b.pointCharge*i*i;a.px-=g*j,a.py-=h*j}}return!b.charge}}function B(){var a=v.length,d=w.length,e,f,g,h,i,j,l,m,p;for(f=0;f<d;++f){g=w[f],h=g.source,i=g.target,m=i.x-h.x,p=i.y-h.y;if(j=m*m+p*p)j=n*y[f]*((j=Math.sqrt(j))-x[f])/j,m*=j,p*=j,i.x-=m*(l=h.weight/(i.weight+h.weight)),i.y-=p*l,h.x+=m*(l=1-l),h.y+=p*l}if(l=n*s){m=c[0]/2,p=c[1]/2,f=-1;if(l)while(++f<a)g=v[f],g.x+=(m-g.x)*l,g.y+=(p-g.y)*l}if(r){k(e=d3.geom.quadtree(v),n,z),f=-1;while(++f<a)(g=v[f]).fixed||e.visit(A(g))}f=-1;while(++f<a)g=v[f],g.fixed?(g.x=g.px,g.y=g.py):(g.x-=(g.px-(g.px=g.x))*o,g.y-=(g.py-(g.py=g.y))*o);return b.tick({type:"tick",alpha:n}),(n*=.99)<.005}function C(b){g(f=b),e=a}var a={},b=d3.dispatch("tick"),c=[1,1],d,n,o=.9,p=l,q=m,r=-30,s=.1,t=.8,u,v=[],w=[],x,y,z;return a.on=function(c,d){return b.on(c,d),a},a.nodes=function(b){return arguments.length?(v=b,a):v},a.links=function(b){return arguments.length?(w=b,a):w},a.size=function(b){return arguments.length?(c=b,a):c},a.linkDistance=function(b){return arguments.length?(p=d3.functor(b),a):p},a.distance=a.linkDistance,a.linkStrength=function(b){return arguments.length?(q=d3.functor(b),a):q},a.friction=function(b){return arguments.length?(o=b,a):o},a.charge=function(b){return arguments.length?(r=typeof b=="function"?b:+b,a):r},a.gravity=function(b){return arguments.length?(s=b,a):s},a.theta=function(b){return arguments.length?(t=b,a):t},a.start=function(){function k(a,c){var d=l(b),e=-1,f=d.length,g;while(++e<f)if(!isNaN(g=d[e][a]))return g;return Math.random()*c}function l(){if(!i){i=[];for(d=0;d<e;++d)i[d]=[];for(d=0;d<f;++d){var a=w[d];i[a.source.index].push(a.target),i[a.target.index].push(a.source)}}return i[b]}var b,d,e=v.length,f=w.length,g=c[0],h=c[1],i,j;for(b=0;b<e;++b)(j=v[b]).index=b,j.weight=0;x=[],y=[];for(b=0;b<f;++b)j=w[b],typeof j.source=="number"&&(j.source=v[j.source]),typeof j.target=="number"&&(j.target=v[j.target]),x[b]=p.call(this,j,b),y[b]=q.call(this,j,b),++j.source.weight,++j.target.weight;for(b=0;b<e;++b)j=v[b],isNaN(j.x)&&(j.x=k("x",g)),isNaN(j.y)&&(j.y=k("y",h)),isNaN(j.px)&&(j.px=j.x),isNaN(j.py)&&(j.py=j.y);z=[];if(typeof r=="function")for(b=0;b<e;++b)z[b]=+r.call(this,v[b],b);else for(b=0;b<e;++b)z[b]=r;return a.resume()},a.resume=function(){return n=.1,d3.timer(B),a},a.stop=function(){return n=0,a},a.drag=function(){d||(d=d3.behavior.drag().on("dragstart",C).on("drag",j).on("dragend",i)),this.on("mouseover.force",g).on("mouseout.force",h).call(d)},a};var e,f;d3.layout.partition=function(){function c(a,b,d,e){var f=a.children;a.x=b,a.y=a.depth*e,a.dx=d,a.dy=e;if(f&&(h=f.length)){var g=-1,h,i,j;d=a.value?d/a.value:0;while(++g<h)c(i=f[g],b,j=i.value*d,e),b+=j}}function d(a){var b=a.children,c=0;if(b&&(f=b.length)){var e=-1,f;while(++e<f)c=Math.max(c,d(b[e]))}return 1+c}function e(e,f){var g=a.call(this,e,f);return c(g[0],0,b[0],b[1]/d(g[0])),g}var a=d3.layout.hierarchy(),b=[1,1];return e.size=function(a){return arguments.length?(b=a,e):b},z(e,a)},d3.layout.pie=function(){function f(g,h){var i=g.map(function(b,c){return+a.call(f,b,c)}),j=+(typeof c=="function"?c.apply(this,arguments):c),k=((typeof e=="function"?e.apply(this,arguments):e)-c)/d3.sum(i),l=d3.range(g.length);b!=null&&l.sort(b===n?function(a,b){return i[b]-i[a]}:function(a,c){return b(g[a],g[c])});var m=l.map(function(a){return{data:g[a],value:d=i[a],startAngle:j,endAngle:j+=d*k}});return g.map(function(a,b){return m[l[b]]})}var a=Number,b=n,c=0,e=2*Math.PI;return f.value=function(b){return arguments.length?(a=b,f):a},f.sort=function(a){return arguments.length?(b=a,f):b},f.startAngle=function(a){return arguments.length?(c=a,f):c},f.endAngle=function(a){return arguments.length?(e=a,f):e},f};var n={};d3.layout.stack=function(){function g(h,i){var j=h.map(function(b,c){return a.call(g,b,c)}),k=j.map(function(a,b){return a.map(function(a,b){return[e.call(g,a,b),f.call(g,a,b)]})}),l=b.call(g,k,i);j=d3.permute(j,l),k=d3.permute(k,l);var m=c.call(g,k,i),n=j.length,o=j[0].length,p,q,r;for(q=0;q<o;++q){d.call(g,j[0][q],r=m[q],k[0][q][1]);for(p=1;p<n;++p)d.call(g,j[p][q],r+=k[p-1][q][1],k[p][q][1])}return h}var a=Object,b=r["default"],c=s.zero,d=q,e=o,f=p;return g.values=function(b){return arguments.length?(a=b,g):a},g.order=function(a){return arguments.length?(b=typeof a=="function"?a:r[a],g):b},g.offset=function(a){return arguments.length?(c=typeof a=="function"?a:s[a],g):c},g.x=function(a){return arguments.length?(e=a,g):e},g.y=function(a){return arguments.length?(f=a,g):f},g.out=function(a){return arguments.length?(d=a,g):d},g};var r={"inside-out":function(a){var b=a.length,c,d,e=a.map(t),f=a.map(u),g=d3.range(b).sort(function(a,b){return e[a]-e[b]}),h=0,i=0,j=[],k=[];for(c=0;c<b;++c)d=g[c],h<i?(h+=f[d],j.push(d)):(i+=f[d],k.push(d));return k.reverse().concat(j)},reverse:function(a){return d3.range(a.length).reverse()},"default":function(a){return d3.range(a.length)}},s={silhouette:function(a){var b=a.length,c=a[0].length,d=[],e=0,f,g,h,i=[];for(g=0;g<c;++g){for(f=0,h=0;f<b;f++)h+=a[f][g][1];h>e&&(e=h),d.push(h)}for(g=0;g<c;++g)i[g]=(e-d[g])/2;return i},wiggle:function(a){var b=a.length,c=a[0],d=c.length,e=0,f,g,h,i,j,k,l,m,n,o=[];o[0]=m=n=0;for(g=1;g<d;++g){for(f=0,i=0;f<b;++f)i+=a[f][g][1];for(f=0,j=0,l=c[g][0]-c[g-1][0];f<b;++f){for(h=0,k=(a[f][g][1]-a[f][g-1][1])/(2*l);h<f;++h)k+=(a[h][g][1]-a[h][g-1][1])/l;j+=k*a[f][g][1]}o[g]=m-=i?j/i*l:0,m<n&&(n=m)}for(g=0;g<d;++g)o[g]-=n;return o},expand:function(a){var b=a.length,c=a[0].length,d=1/b,e,f,g,h=[];for(f=0;f<c;++f){for(e=0,g=0;e<b;e++)g+=a[e][f][1];if(g)for(e=0;e<b;e++)a[e][f][1]/=g;else for(e=0;e<b;e++)a[e][f][1]=d}for(f=0;f<c;++f)h[f]=0;return h},zero:function(a){var b=-1,c=a[0].length,d=[];while(++b<c)d[b]=0;return d}};d3.layout.histogram=function(){function e(e,f){var g=[],h=e.map(b,this),i=c.call(this,h,f),j=d.call(this,i,h,f),k,f=-1,l=h.length,m=j.length-1,n=a?1:1/l,o;while(++f<m)k=g[f]=[],k.dx=j[f+1]-(k.x=j[f]),k.y=0;f=-1;while(++f<l)o=h[f],o>=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(++l<m)d=e(j[l],p,i),d.parent=k,n.push(d),o+=d.value;a&&n.sort(a),c&&(k.value=o)}else c&&(k.value=+c.call(g,f,h)||0);return k}function f(a,b){var d=a.children,e=0;if(d&&(i=d.length)){var h=-1,i,j=b+1;while(++h<i)e+=f(d[h],j)}else c&&(e=+c.call(g,E?a:a.data,b)||0);return c&&(a.value=e),e}function g(a){var b=[];return e(a,0,b),b}var a=C,b=A,c=B;return g.sort=function(b){return arguments.length?(a=b,g):a},g.children=function(a){return arguments.length?(b=a,g):b},g.value=function(a){return arguments.length?(c=a,g):c},g.revalue=function(a){return f(a,0),a},g};var E=!1;d3.layout.pack=function(){function c(c,d){var e=a.call(this,c,d),f=e[0];f.x=0,f.y=0,M(f);var g=b[0],h=b[1],i=1/Math.max(2*f.r/g,2*f.r/h);return N(f,g/2,h/2,i),e}var a=d3.layout.hierarchy().sort(F),b=[1,1];return c.size=function(a){return arguments.length?(b=a,c):b},z(c,a)},d3.layout.cluster=function(){function d(d,e){var f=a.call(this,d,e),g=f[0],h,i=0,j,k;$(g,function(a){var c=a.children;c&&c.length?(a.x=Q(c),a.y=P(c)):(a.x=h?i+=b(a,h):0,a.y=0,h=a)});var l=R(g),m=S(g),n=l.x-b(l,m)/2,o=m.x+b(m,l)/2;return $(g,function(a){a.x=(a.x-n)/(o-n)*c[0],a.y=(1-a.y/g.y)*c[1]}),f}var a=d3.layout.hierarchy().sort(null).value(null),b=T,c=[1,1];return d.separation=function(a){return arguments.length?(b=a,d):b},d.size=function(a){return arguments.length?(c=a,d):c},z(d,a)},d3.layout.tree=function(){function d(d,e){function h(a,c){var d=a.children,e=a._tree;if(d&&(f=d.length)){var f,g=d[0],i,k=g,l,m=-1;while(++m<f)l=d[m],h(l,i),k=j(l,i,k),i=l;_(a);var n=.5*(g._tree.prelim+l._tree.prelim);c?(e.prelim=c._tree.prelim+b(a,c),e.mod=e.prelim-n):e.prelim=n}else c&&(e.prelim=c._tree.prelim+b(a,c))}function i(a,b){a.x=a._tree.prelim+b;var c=a.children;if(c&&(e=c.length)){var d=-1,e;b+=a._tree.mod;while(++d<e)i(c[d],b)}}function j(a,c,d){if(c){var e=a,f=a,g=c,h=a.parent.children[0],i=e._tree.mod,j=f._tree.mod,k=g._tree.mod,l=h._tree.mod,m;while(g=V(g),e=U(e),g&&e)h=U(h),f=V(f),f._tree.ancestor=a,m=g._tree.prelim+k-e._tree.prelim-i+b(g,e),m>0&&(ba(bb(g,a,d),a,m),i+=m,j+=m),k+=g._tree.mod,i+=e._tree.mod,l+=h._tree.mod,j+=f._tree.mod;g&&!V(f)&&(f._tree.thread=g,f._tree.mod+=k-j),e&&!U(h)&&(h._tree.thread=e,h._tree.mod+=i-l,d=a)}return d}var f=a.call(this,d,e),g=f[0];$(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(++c<d)f=(e=a[c]).value*(b<0?0:b),e.area=isNaN(f)||f<=0?0:f}function j(a){var b=a.children;if(b&&b.length){var c=e(a),d=[],f=b.slice(),g,h=Infinity,k,n=Math.min(c.dx,c.dy),o;i(f,c.dx*c.dy/a.value),d.area=0;while((o=f.length)>0)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(++g<i){if(!(d=a[g].area))continue;d<f&&(f=d),d>e&&(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(++f<g)k=a[f],k.x=h,k.y=i,k.dy=j,h+=k.dx=j?b(k.area/j):0;k.z=!0,k.dx+=d.x+d.dx-h,d.y+=j,d.dy-=j}else{if(e||j>d.dx)j=j?d.dx:0;while(++f<g)k=a[f],k.x=h,k.y=i,k.dx=j,i+=k.dy=j?b(k.area/j):0;k.z=!1,k.dy+=d.y+d.dy-i,d.x+=j,d.dx-=j}}function n(b){var d=g||a(b),e=d[0];return e.x=0,e.y=0,e.dx=c[0],e.dy=c[1],g&&a.revalue(e),i([e],e.dx*e.dy/e.value),(g?k:j)(e),f&&(g=d),d}var a=d3.layout.hierarchy(),b=Math.round,c=[1,1],d=null,e=bc,f=!1,g,h=.5*(1+Math.sqrt(5));return n.size=function(a){return arguments.length?(c=a,n):c},n.padding=function(a){function b(b){var c=a.call(n,b,b.depth);return c==null?bc(b):bd(b,typeof c=="number"?[c,c,c,c]:c)}function c(b){return bd(b,a)}if(!arguments.length)return d;var f;return e=(d=a)==null?bc:(f=typeof a)==="function"?b:f==="number"?(a=[a,a,a,a],c):c,n},n.round=function(a){return arguments.length?(b=a?Math.round:Number,n):b!=Number},n.sticky=function(a){return arguments.length?(f=a,g=null,n):f},n.ratio=function(a){return arguments.length?(h=a,n):h},z(n,a)}})(); |
(function() { | |
function d3_class(ctor, properties) { | |
try { | |
for (var key in properties) { | |
Object.defineProperty(ctor.prototype, key, { | |
value: properties[key], | |
enumerable: false | |
}); | |
} | |
} catch (e) { | |
ctor.prototype = properties; | |
} | |
} | |
function d3_arrayCopy(pseudoarray) { | |
var i = -1, n = pseudoarray.length, array = []; | |
while (++i < n) array.push(pseudoarray[i]); | |
return array; | |
} | |
function d3_arraySlice(pseudoarray) { | |
return Array.prototype.slice.call(pseudoarray); | |
} | |
function d3_Map() {} | |
function d3_identity(d) { | |
return d; | |
} | |
function d3_this() { | |
return this; | |
} | |
function d3_true() { | |
return true; | |
} | |
function d3_functor(v) { | |
return typeof v === "function" ? v : function() { | |
return v; | |
}; | |
} | |
function d3_rebind(target, source, method) { | |
return function() { | |
var value = method.apply(source, arguments); | |
return arguments.length ? target : value; | |
}; | |
} | |
function d3_number(x) { | |
return x != null && !isNaN(x); | |
} | |
function d3_zipLength(d) { | |
return d.length; | |
} | |
function d3_splitter(d) { | |
return d == null; | |
} | |
function d3_collapse(s) { | |
return s.trim().replace(/\s+/g, " "); | |
} | |
function d3_range_integerScale(x) { | |
var k = 1; | |
while (x * k % 1) k *= 10; | |
return k; | |
} | |
function d3_dispatch() {} | |
function d3_dispatch_event(dispatch) { | |
function event() { | |
var z = listeners, i = -1, n = z.length, l; | |
while (++i < n) if (l = z[i].on) l.apply(this, arguments); | |
return dispatch; | |
} | |
var listeners = [], listenerByName = new d3_Map; | |
event.on = function(name, listener) { | |
var l = listenerByName.get(name), i; | |
if (arguments.length < 2) return l && l.on; | |
if (l) { | |
l.on = null; | |
listeners = listeners.slice(0, i = listeners.indexOf(l)).concat(listeners.slice(i + 1)); | |
listenerByName.remove(name); | |
} | |
if (listener) listeners.push(listenerByName.set(name, { | |
on: listener | |
})); | |
return dispatch; | |
}; | |
return event; | |
} | |
function d3_format_precision(x, p) { | |
return p - (x ? 1 + Math.floor(Math.log(x + Math.pow(10, 1 + Math.floor(Math.log(x) / Math.LN10) - p)) / Math.LN10) : 1); | |
} | |
function d3_format_typeDefault(x) { | |
return x + ""; | |
} | |
function d3_format_group(value) { | |
var i = value.lastIndexOf("."), f = i >= 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); | |
g = format.substring(3, 5); | |
b = format.substring(5, 7); | |
} | |
r = parseInt(r, 16); | |
g = parseInt(g, 16); | |
b = parseInt(b, 16); | |
} | |
return rgb(r, g, b); | |
} | |
function d3_rgb_hsl(r, g, b) { | |
var min = Math.min(r /= 255, g /= 255, b /= 255), max = Math.max(r, g, b), d = max - min, h, s, l = (max + min) / 2; | |
if (d) { | |
s = l < .5 ? d / (max + min) : d / (2 - max - min); | |
if (r == max) h = (g - b) / d + (g < b ? 6 : 0); else if (g == max) h = (b - r) / d + 2; else h = (r - g) / d + 4; | |
h *= 60; | |
} else { | |
s = h = 0; | |
} | |
return d3_hsl(h, s, l); | |
} | |
function d3_rgb_lab(r, g, b) { | |
r = d3_rgb_xyz(r); | |
g = d3_rgb_xyz(g); | |
b = d3_rgb_xyz(b); | |
var x = d3_xyz_lab((.4124564 * r + .3575761 * g + .1804375 * b) / d3_lab_X), y = d3_xyz_lab((.2126729 * r + .7151522 * g + .072175 * b) / d3_lab_Y), z = d3_xyz_lab((.0193339 * r + .119192 * g + .9503041 * b) / d3_lab_Z); | |
return d3_lab(116 * y - 16, 500 * (x - y), 200 * (y - z)); | |
} | |
function d3_rgb_xyz(r) { | |
return (r /= 255) <= .04045 ? r / 12.92 : Math.pow((r + .055) / 1.055, 2.4); | |
} | |
function d3_rgb_parseNumber(c) { | |
var f = parseFloat(c); | |
return c.charAt(c.length - 1) === "%" ? Math.round(f * 2.55) : f; | |
} | |
function d3_hsl(h, s, l) { | |
return new d3_Hsl(h, s, l); | |
} | |
function d3_Hsl(h, s, l) { | |
this.h = h; | |
this.s = s; | |
this.l = l; | |
} | |
function d3_hsl_rgb(h, s, l) { | |
function v(h) { | |
if (h > 360) h -= 360; else if (h < 0) h += 360; | |
if (h < 60) return m1 + (m2 - m1) * h / 60; | |
if (h < 180) return m2; | |
if (h < 240) return m1 + (m2 - m1) * (240 - h) / 60; | |
return m1; | |
} | |
function vv(h) { | |
return Math.round(v(h) * 255); | |
} | |
var m1, m2; | |
h = h % 360; | |
if (h < 0) h += 360; | |
s = s < 0 ? 0 : s > 1 ? 1 : s; | |
l = l < 0 ? 0 : l > 1 ? 1 : l; | |
m2 = l <= .5 ? l * (1 + s) : l + s - l * s; | |
m1 = 2 * l - m2; | |
return d3_rgb(vv(h + 120), vv(h), vv(h - 120)); | |
} | |
function d3_hcl(h, c, l) { | |
return new d3_Hcl(h, c, l); | |
} | |
function d3_Hcl(h, c, l) { | |
this.h = h; | |
this.c = c; | |
this.l = l; | |
} | |
function d3_hcl_lab(h, c, l) { | |
return d3_lab(l, Math.cos(h *= Math.PI / 180) * c, Math.sin(h) * c); | |
} | |
function d3_lab(l, a, b) { | |
return new d3_Lab(l, a, b); | |
} | |
function d3_Lab(l, a, b) { | |
this.l = l; | |
this.a = a; | |
this.b = b; | |
} | |
function d3_lab_rgb(l, a, b) { | |
var y = (l + 16) / 116, x = y + a / 500, z = y - b / 200; | |
x = d3_lab_xyz(x) * d3_lab_X; | |
y = d3_lab_xyz(y) * d3_lab_Y; | |
z = d3_lab_xyz(z) * d3_lab_Z; | |
return d3_rgb(d3_xyz_rgb(3.2404542 * x - 1.5371385 * y - .4985314 * z), d3_xyz_rgb(-.969266 * x + 1.8760108 * y + .041556 * z), d3_xyz_rgb(.0556434 * x - .2040259 * y + 1.0572252 * z)); | |
} | |
function d3_lab_hcl(l, a, b) { | |
return d3_hcl(Math.atan2(b, a) / Math.PI * 180, Math.sqrt(a * a + b * b), l); | |
} | |
function d3_lab_xyz(x) { | |
return x > .206893034 ? x * x * x : (x - 4 / 29) / 7.787037; | |
} | |
function d3_xyz_lab(x) { | |
return x > .008856 ? Math.pow(x, 1 / 3) : 7.787037 * x + 4 / 29; | |
} | |
function d3_xyz_rgb(r) { | |
return Math.round(255 * (r <= .00304 ? 12.92 * r : 1.055 * Math.pow(r, 1 / 2.4) - .055)); | |
} | |
function d3_selection(groups) { | |
d3_arraySubclass(groups, d3_selectionPrototype); | |
return groups; | |
} | |
function d3_selection_selector(selector) { | |
return function() { | |
return d3_select(selector, this); | |
}; | |
} | |
function d3_selection_selectorAll(selector) { | |
return function() { | |
return d3_selectAll(selector, this); | |
}; | |
} | |
function d3_selection_attr(name, value) { | |
function attrNull() { | |
this.removeAttribute(name); | |
} | |
function attrNullNS() { | |
this.removeAttributeNS(name.space, name.local); | |
} | |
function attrConstant() { | |
this.setAttribute(name, value); | |
} | |
function attrConstantNS() { | |
this.setAttributeNS(name.space, name.local, value); | |
} | |
function attrFunction() { | |
var x = value.apply(this, arguments); | |
if (x == null) this.removeAttribute(name); else this.setAttribute(name, x); | |
} | |
function attrFunctionNS() { | |
var x = value.apply(this, arguments); | |
if (x == null) this.removeAttributeNS(name.space, name.local); else this.setAttributeNS(name.space, name.local, x); | |
} | |
name = d3.ns.qualify(name); | |
return value == null ? name.local ? attrNullNS : attrNull : typeof value === "function" ? name.local ? attrFunctionNS : attrFunction : name.local ? attrConstantNS : attrConstant; | |
} | |
function d3_selection_classedRe(name) { | |
return new RegExp("(?:^|\\s+)" + d3.requote(name) + "(?:\\s+|$)", "g"); | |
} | |
function d3_selection_classed(name, value) { | |
function classedConstant() { | |
var i = -1; | |
while (++i < n) name[i](this, value); | |
} | |
function classedFunction() { | |
var i = -1, x = value.apply(this, arguments); | |
while (++i < n) name[i](this, x); | |
} | |
name = name.trim().split(/\s+/).map(d3_selection_classedName); | |
var n = name.length; | |
return typeof value === "function" ? classedFunction : classedConstant; | |
} | |
function d3_selection_classedName(name) { | |
var re = d3_selection_classedRe(name); | |
return function(node, value) { | |
if (c = node.classList) return value ? c.add(name) : c.remove(name); | |
var c = node.className, cb = c.baseVal != null, cv = cb ? c.baseVal : c; | |
if (value) { | |
re.lastIndex = 0; | |
if (!re.test(cv)) { | |
cv = d3_collapse(cv + " " + name); | |
if (cb) c.baseVal = cv; else node.className = cv; | |
} | |
} else if (cv) { | |
cv = d3_collapse(cv.replace(re, " ")); | |
if (cb) c.baseVal = cv; else node.className = cv; | |
} | |
}; | |
} | |
function d3_selection_style(name, value, priority) { | |
function styleNull() { | |
this.style.removeProperty(name); | |
} | |
function styleConstant() { | |
this.style.setProperty(name, value, priority); | |
} | |
function styleFunction() { | |
var x = value.apply(this, arguments); | |
if (x == null) this.style.removeProperty(name); else this.style.setProperty(name, x, priority); | |
} | |
return value == null ? styleNull : typeof value === "function" ? styleFunction : styleConstant; | |
} | |
function d3_selection_property(name, value) { | |
function propertyNull() { | |
delete this[name]; | |
} | |
function propertyConstant() { | |
this[name] = value; | |
} | |
function propertyFunction() { | |
var x = value.apply(this, arguments); | |
if (x == null) delete this[name]; else this[name] = x; | |
} | |
return value == null ? propertyNull : typeof value === "function" ? propertyFunction : propertyConstant; | |
} | |
function d3_selection_dataNode(data) { | |
return { | |
__data__: data | |
}; | |
} | |
function d3_selection_filter(selector) { | |
return function() { | |
return d3_selectMatches(this, selector); | |
}; | |
} | |
function d3_selection_sortComparator(comparator) { | |
if (!arguments.length) comparator = d3.ascending; | |
return function(a, b) { | |
return comparator(a && a.__data__, b && b.__data__); | |
}; | |
} | |
function d3_selection_on(type, listener, capture) { | |
function onRemove() { | |
var wrapper = this[name]; | |
if (wrapper) { | |
this.removeEventListener(type, wrapper, wrapper.$); | |
delete this[name]; | |
} | |
} | |
function onAdd() { | |
function wrapper(e) { | |
var o = d3.event; | |
d3.event = e; | |
args[0] = node.__data__; | |
try { | |
listener.apply(node, args); | |
} finally { | |
d3.event = o; | |
} | |
} | |
var node = this, args = arguments; | |
onRemove.call(this); | |
this.addEventListener(type, this[name] = wrapper, wrapper.$ = capture); | |
wrapper._ = listener; | |
} | |
var name = "__on" + type, i = type.indexOf("."); | |
if (i > 0) type = type.substring(0, i); | |
return listener ? onAdd : onRemove; | |
} | |
function d3_selection_each(groups, callback) { | |
for (var j = 0, m = groups.length; j < m; j++) { | |
for (var group = groups[j], i = 0, n = group.length, node; i < n; i++) { | |
if (node = group[i]) callback(node, i, j); | |
} | |
} | |
return groups; | |
} | |
function d3_selection_enter(selection) { | |
d3_arraySubclass(selection, d3_selection_enterPrototype); | |
return selection; | |
} | |
function d3_transition(groups, id, time) { | |
d3_arraySubclass(groups, d3_transitionPrototype); | |
var tweens = new d3_Map, event = d3.dispatch("start", "end"), ease = d3_transitionEase; | |
groups.id = id; | |
groups.time = time; | |
groups.tween = function(name, tween) { | |
if (arguments.length < 2) return tweens.get(name); | |
if (tween == null) tweens.remove(name); else tweens.set(name, tween); | |
return groups; | |
}; | |
groups.ease = function(value) { | |
if (!arguments.length) return ease; | |
ease = typeof value === "function" ? value : d3.ease.apply(d3, arguments); | |
return groups; | |
}; | |
groups.each = function(type, listener) { | |
if (arguments.length < 2) return d3_transition_each.call(groups, type); | |
event.on(type, listener); | |
return groups; | |
}; | |
d3.timer(function(elapsed) { | |
return d3_selection_each(groups, function(node, i, j) { | |
function start(elapsed) { | |
if (lock.active > id) return stop(); | |
lock.active = id; | |
tweens.forEach(function(key, value) { | |
if (value = value.call(node, d, i)) { | |
tweened.push(value); | |
} | |
}); | |
event.start.call(node, d, i); | |
if (!tick(elapsed)) d3.timer(tick, 0, time); | |
return 1; | |
} | |
function tick(elapsed) { | |
if (lock.active !== id) return stop(); | |
var t = (elapsed - delay) / duration, e = ease(t), n = tweened.length; | |
while (n > 0) { | |
tweened[--n].call(node, e); | |
} | |
if (t >= 1) { | |
stop(); | |
d3_transitionId = id; | |
event.end.call(node, d, i); | |
d3_transitionId = 0; | |
return 1; | |
} | |
} | |
function stop() { | |
if (!--lock.count) delete node.__transition__; | |
return 1; | |
} | |
var tweened = [], delay = node.delay, duration = node.duration, lock = (node = node.node).__transition__ || (node.__transition__ = { | |
active: 0, | |
count: 0 | |
}), d = node.__data__; | |
++lock.count; | |
delay <= elapsed ? start(elapsed) : d3.timer(start, delay, time); | |
}); | |
}, 0, time); | |
return groups; | |
} | |
function d3_transition_each(callback) { | |
var id = d3_transitionId, ease = d3_transitionEase, delay = d3_transitionDelay, duration = d3_transitionDuration; | |
d3_transitionId = this.id; | |
d3_transitionEase = this.ease(); | |
d3_selection_each(this, function(node, i, j) { | |
d3_transitionDelay = node.delay; | |
d3_transitionDuration = node.duration; | |
callback.call(node = node.node, node.__data__, i, j); | |
}); | |
d3_transitionId = id; | |
d3_transitionEase = ease; | |
d3_transitionDelay = delay; | |
d3_transitionDuration = duration; | |
return this; | |
} | |
function d3_tweenNull(d, i, a) { | |
return a != "" && d3_tweenRemove; | |
} | |
function d3_tweenByName(b, name) { | |
return d3.tween(b, d3_interpolateByName(name)); | |
} | |
function d3_timer_step() { | |
var elapsed, now = Date.now(), t1 = d3_timer_queue; | |
while (t1) { | |
elapsed = now - t1.then; | |
if (elapsed >= t1.delay) t1.flush = t1.callback(elapsed); | |
t1 = t1.next; | |
} | |
var delay = d3_timer_flush() - now; | |
if (delay > 24) { | |
if (isFinite(delay)) { | |
clearTimeout(d3_timer_timeout); | |
d3_timer_timeout = setTimeout(d3_timer_step, delay); | |
} | |
d3_timer_interval = 0; | |
} else { | |
d3_timer_interval = 1; | |
d3_timer_frame(d3_timer_step); | |
} | |
} | |
function d3_timer_flush() { | |
var t0 = null, t1 = d3_timer_queue, then = Infinity; | |
while (t1) { | |
if (t1.flush) { | |
delete d3_timer_byId[t1.callback.id]; | |
t1 = t0 ? t0.next = t1.next : d3_timer_queue = t1.next; | |
} else { | |
then = Math.min(then, t1.then + t1.delay); | |
t1 = (t0 = t1).next; | |
} | |
} | |
return then; | |
} | |
function d3_mousePoint(container, e) { | |
var svg = container.ownerSVGElement || container; | |
if (svg.createSVGPoint) { | |
var point = svg.createSVGPoint(); | |
if (d3_mouse_bug44083 < 0 && (window.scrollX || window.scrollY)) { | |
svg = d3.select(document.body).append("svg").style("position", "absolute").style("top", 0).style("left", 0); | |
var ctm = svg[0][0].getScreenCTM(); | |
d3_mouse_bug44083 = !(ctm.f || ctm.e); | |
svg.remove(); | |
} | |
if (d3_mouse_bug44083) { | |
point.x = e.pageX; | |
point.y = e.pageY; | |
} else { | |
point.x = e.clientX; | |
point.y = e.clientY; | |
} | |
point = point.matrixTransform(container.getScreenCTM().inverse()); | |
return [ point.x, point.y ]; | |
} | |
var rect = container.getBoundingClientRect(); | |
return [ e.clientX - rect.left - container.clientLeft, e.clientY - rect.top - container.clientTop ]; | |
} | |
function d3_noop() {} | |
function d3_scaleExtent(domain) { | |
var start = domain[0], stop = domain[domain.length - 1]; | |
return start < stop ? [ start, stop ] : [ stop, start ]; | |
} | |
function d3_scaleRange(scale) { | |
return scale.rangeExtent ? scale.rangeExtent() : d3_scaleExtent(scale.range()); | |
} | |
function d3_scale_nice(domain, nice) { | |
var i0 = 0, i1 = domain.length - 1, x0 = domain[i0], x1 = domain[i1], dx; | |
if (x1 < x0) { | |
dx = i0, i0 = i1, i1 = dx; | |
dx = x0, x0 = x1, x1 = dx; | |
} | |
if (nice = nice(x1 - x0)) { | |
domain[i0] = nice.floor(x0); | |
domain[i1] = nice.ceil(x1); | |
} | |
return domain; | |
} | |
function d3_scale_niceDefault() { | |
return Math; | |
} | |
function d3_scale_linear(domain, range, interpolate, clamp) { | |
function rescale() { | |
var linear = Math.min(domain.length, range.length) > 2 ? d3_scale_polylinear : d3_scale_bilinear, uninterpolate = clamp ? d3_uninterpolateClamp : d3_uninterpolateNumber; | |
output = linear(domain, range, uninterpolate, interpolate); | |
input = linear(range, domain, uninterpolate, d3.interpolate); | |
return scale; | |
} | |
function scale(x) { | |
return output(x); | |
} | |
var output, input; | |
scale.invert = function(y) { | |
return input(y); | |
}; | |
scale.domain = function(x) { | |
if (!arguments.length) return domain; | |
domain = x.map(Number); | |
return rescale(); | |
}; | |
scale.range = function(x) { | |
if (!arguments.length) return range; | |
range = x; | |
return rescale(); | |
}; | |
scale.rangeRound = function(x) { | |
return scale.range(x).interpolate(d3.interpolateRound); | |
}; | |
scale.clamp = function(x) { | |
if (!arguments.length) return clamp; | |
clamp = x; | |
return rescale(); | |
}; | |
scale.interpolate = function(x) { | |
if (!arguments.length) return interpolate; | |
interpolate = x; | |
return rescale(); | |
}; | |
scale.ticks = function(m) { | |
return d3_scale_linearTicks(domain, m); | |
}; | |
scale.tickFormat = function(m) { | |
return d3_scale_linearTickFormat(domain, m); | |
}; | |
scale.nice = function() { | |
d3_scale_nice(domain, d3_scale_linearNice); | |
return rescale(); | |
}; | |
scale.copy = function() { | |
return d3_scale_linear(domain, range, interpolate, clamp); | |
}; | |
return rescale(); | |
} | |
function d3_scale_linearRebind(scale, linear) { | |
return d3.rebind(scale, linear, "range", "rangeRound", "interpolate", "clamp"); | |
} | |
function d3_scale_linearNice(dx) { | |
dx = Math.pow(10, Math.round(Math.log(dx) / Math.LN10) - 1); | |
return dx && { | |
floor: function(x) { | |
return Math.floor(x / dx) * dx; | |
}, | |
ceil: function(x) { | |
return Math.ceil(x / dx) * dx; | |
} | |
}; | |
} | |
function d3_scale_linearTickRange(domain, m) { | |
var extent = d3_scaleExtent(domain), span = extent[1] - extent[0], step = Math.pow(10, Math.floor(Math.log(span / m) / Math.LN10)), err = m / span * step; | |
if (err <= .15) step *= 10; else if (err <= .35) step *= 5; else if (err <= .75) step *= 2; | |
extent[0] = Math.ceil(extent[0] / step) * step; | |
extent[1] = Math.floor(extent[1] / step) * step + step * .5; | |
extent[2] = step; | |
return extent; | |
} | |
function d3_scale_linearTicks(domain, m) { | |
return d3.range.apply(d3, d3_scale_linearTickRange(domain, m)); | |
} | |
function d3_scale_linearTickFormat(domain, m) { | |
return d3.format(",." + Math.max(0, -Math.floor(Math.log(d3_scale_linearTickRange(domain, m)[2]) / Math.LN10 + .01)) + "f"); | |
} | |
function d3_scale_bilinear(domain, range, uninterpolate, interpolate) { | |
var u = uninterpolate(domain[0], domain[1]), i = interpolate(range[0], range[1]); | |
return function(x) { | |
return i(u(x)); | |
}; | |
} | |
function d3_scale_polylinear(domain, range, uninterpolate, interpolate) { | |
var u = [], i = [], j = 0, k = Math.min(domain.length, range.length) - 1; | |
if (domain[k] < domain[0]) { | |
domain = domain.slice().reverse(); | |
range = range.slice().reverse(); | |
} | |
while (++j <= k) { | |
u.push(uninterpolate(domain[j - 1], domain[j])); | |
i.push(interpolate(range[j - 1], range[j])); | |
} | |
return function(x) { | |
var j = d3.bisect(domain, x, 1, k) - 1; | |
return i[j](u[j](x)); | |
}; | |
} | |
function d3_scale_log(linear, log) { | |
function scale(x) { | |
return linear(log(x)); | |
} | |
var pow = log.pow; | |
scale.invert = function(x) { | |
return pow(linear.invert(x)); | |
}; | |
scale.domain = function(x) { | |
if (!arguments.length) return linear.domain().map(pow); | |
log = x[0] < 0 ? d3_scale_logn : d3_scale_logp; | |
pow = log.pow; | |
linear.domain(x.map(log)); | |
return scale; | |
}; | |
scale.nice = function() { | |
linear.domain(d3_scale_nice(linear.domain(), d3_scale_niceDefault)); | |
return scale; | |
}; | |
scale.ticks = function() { | |
var extent = d3_scaleExtent(linear.domain()), ticks = []; | |
if (extent.every(isFinite)) { | |
var i = Math.floor(extent[0]), j = Math.ceil(extent[1]), u = pow(extent[0]), v = pow(extent[1]); | |
if (log === d3_scale_logn) { | |
ticks.push(pow(i)); | |
for (; i++ < j; ) for (var k = 9; k > 0; k--) ticks.push(pow(i) * k); | |
} else { | |
for (; i < j; i++) for (var k = 1; k < 10; k++) ticks.push(pow(i) * k); | |
ticks.push(pow(i)); | |
} | |
for (i = 0; ticks[i] < u; i++) {} | |
for (j = ticks.length; ticks[j - 1] > v; j--) {} | |
ticks = ticks.slice(i, j); | |
} | |
return ticks; | |
}; | |
scale.tickFormat = function(n, format) { | |
if (arguments.length < 2) format = d3_scale_logFormat; | |
if (arguments.length < 1) return format; | |
var k = Math.max(.1, n / scale.ticks().length), f = log === d3_scale_logn ? (e = -1e-12, Math.floor) : (e = 1e-12, Math.ceil), e; | |
return function(d) { | |
return d / pow(f(log(d) + e)) <= k ? format(d) : ""; | |
}; | |
}; | |
scale.copy = function() { | |
return d3_scale_log(linear.copy(), log); | |
}; | |
return d3_scale_linearRebind(scale, linear); | |
} | |
function d3_scale_logp(x) { | |
return Math.log(x < 0 ? 0 : x) / Math.LN10; | |
} | |
function d3_scale_logn(x) { | |
return -Math.log(x > 0 ? 0 : -x) / Math.LN10; | |
} | |
function d3_scale_pow(linear, exponent) { | |
function scale(x) { | |
return linear(powp(x)); | |
} | |
var powp = d3_scale_powPow(exponent), powb = d3_scale_powPow(1 / exponent); | |
scale.invert = function(x) { | |
return powb(linear.invert(x)); | |
}; | |
scale.domain = function(x) { | |
if (!arguments.length) return linear.domain().map(powb); | |
linear.domain(x.map(powp)); | |
return scale; | |
}; | |
scale.ticks = function(m) { | |
return d3_scale_linearTicks(scale.domain(), m); | |
}; | |
scale.tickFormat = function(m) { | |
return d3_scale_linearTickFormat(scale.domain(), m); | |
}; | |
scale.nice = function() { | |
return scale.domain(d3_scale_nice(scale.domain(), d3_scale_linearNice)); | |
}; | |
scale.exponent = function(x) { | |
if (!arguments.length) return exponent; | |
var domain = scale.domain(); | |
powp = d3_scale_powPow(exponent = x); | |
powb = d3_scale_powPow(1 / exponent); | |
return scale.domain(domain); | |
}; | |
scale.copy = function() { | |
return d3_scale_pow(linear.copy(), exponent); | |
}; | |
return d3_scale_linearRebind(scale, linear); | |
} | |
function d3_scale_powPow(e) { | |
return function(x) { | |
return x < 0 ? -Math.pow(-x, e) : Math.pow(x, e); | |
}; | |
} | |
function d3_scale_ordinal(domain, ranger) { | |
function scale(x) { | |
return range[((index.get(x) || index.set(x, domain.push(x))) - 1) % range.length]; | |
} | |
function steps(start, step) { | |
return d3.range(domain.length).map(function(i) { | |
return start + step * i; | |
}); | |
} | |
var index, range, rangeBand; | |
scale.domain = function(x) { | |
if (!arguments.length) return domain; | |
domain = []; | |
index = new d3_Map; | |
var i = -1, n = x.length, xi; | |
while (++i < n) if (!index.has(xi = x[i])) index.set(xi, domain.push(xi)); | |
return scale[ranger.t].apply(scale, ranger.a); | |
}; | |
scale.range = function(x) { | |
if (!arguments.length) return range; | |
range = x; | |
rangeBand = 0; | |
ranger = { | |
t: "range", | |
a: arguments | |
}; | |
return scale; | |
}; | |
scale.rangePoints = function(x, padding) { | |
if (arguments.length < 2) padding = 0; | |
var start = x[0], stop = x[1], step = (stop - start) / (Math.max(1, domain.length - 1) + padding); | |
range = steps(domain.length < 2 ? (start + stop) / 2 : start + step * padding / 2, step); | |
rangeBand = 0; | |
ranger = { | |
t: "rangePoints", | |
a: arguments | |
}; | |
return scale; | |
}; | |
scale.rangeBands = function(x, padding, outerPadding) { | |
if (arguments.length < 2) padding = 0; | |
if (arguments.length < 3) outerPadding = padding; | |
var reverse = x[1] < x[0], start = x[reverse - 0], stop = x[1 - reverse], step = (stop - start) / (domain.length - padding + 2 * outerPadding); | |
range = steps(start + step * outerPadding, step); | |
if (reverse) range.reverse(); | |
rangeBand = step * (1 - padding); | |
ranger = { | |
t: "rangeBands", | |
a: arguments | |
}; | |
return scale; | |
}; | |
scale.rangeRoundBands = function(x, padding, outerPadding) { | |
if (arguments.length < 2) padding = 0; | |
if (arguments.length < 3) outerPadding = padding; | |
var reverse = x[1] < x[0], start = x[reverse - 0], stop = x[1 - reverse], step = Math.floor((stop - start) / (domain.length - padding + 2 * outerPadding)), error = stop - start - (domain.length - padding) * step; | |
range = steps(start + Math.round(error / 2), step); | |
if (reverse) range.reverse(); | |
rangeBand = Math.round(step * (1 - padding)); | |
ranger = { | |
t: "rangeRoundBands", | |
a: arguments | |
}; | |
return scale; | |
}; | |
scale.rangeBand = function() { | |
return rangeBand; | |
}; | |
scale.rangeExtent = function() { | |
return d3_scaleExtent(ranger.a[0]); | |
}; | |
scale.copy = function() { | |
return d3_scale_ordinal(domain, ranger); | |
}; | |
return scale.domain(domain); | |
} | |
function d3_scale_quantile(domain, range) { | |
function rescale() { | |
var k = 0, n = domain.length, q = range.length; | |
thresholds = []; | |
while (++k < q) thresholds[k - 1] = d3.quantile(domain, k / q); | |
return scale; | |
} | |
function scale(x) { | |
if (isNaN(x = +x)) return NaN; | |
return range[d3.bisect(thresholds, x)]; | |
} | |
var thresholds; | |
scale.domain = function(x) { | |
if (!arguments.length) return domain; | |
domain = x.filter(function(d) { | |
return !isNaN(d); | |
}).sort(d3.ascending); | |
return rescale(); | |
}; | |
scale.range = function(x) { | |
if (!arguments.length) return range; | |
range = x; | |
return rescale(); | |
}; | |
scale.quantiles = function() { | |
return thresholds; | |
}; | |
scale.copy = function() { | |
return d3_scale_quantile(domain, range); | |
}; | |
return rescale(); | |
} | |
function d3_scale_quantize(x0, x1, range) { | |
function scale(x) { | |
return range[Math.max(0, Math.min(i, Math.floor(kx * (x - x0))))]; | |
} | |
function rescale() { | |
kx = range.length / (x1 - x0); | |
i = range.length - 1; | |
return scale; | |
} | |
var kx, i; | |
scale.domain = function(x) { | |
if (!arguments.length) return [ x0, x1 ]; | |
x0 = +x[0]; | |
x1 = +x[x.length - 1]; | |
return rescale(); | |
}; | |
scale.range = function(x) { | |
if (!arguments.length) return range; | |
range = x; | |
return rescale(); | |
}; | |
scale.copy = function() { | |
return d3_scale_quantize(x0, x1, range); | |
}; | |
return rescale(); | |
} | |
function d3_scale_threshold(domain, range) { | |
function scale(x) { | |
return range[d3.bisect(domain, x)]; | |
} | |
scale.domain = function(_) { | |
if (!arguments.length) return domain; | |
domain = _; | |
return scale; | |
}; | |
scale.range = function(_) { | |
if (!arguments.length) return range; | |
range = _; | |
return scale; | |
}; | |
scale.copy = function() { | |
return d3_scale_threshold(domain, range); | |
}; | |
return scale; | |
} | |
function d3_scale_identity(domain) { | |
function identity(x) { | |
return +x; | |
} | |
identity.invert = identity; | |
identity.domain = identity.range = function(x) { | |
if (!arguments.length) return domain; | |
domain = x.map(identity); | |
return identity; | |
}; | |
identity.ticks = function(m) { | |
return d3_scale_linearTicks(domain, m); | |
}; | |
identity.tickFormat = function(m) { | |
return d3_scale_linearTickFormat(domain, m); | |
}; | |
identity.copy = function() { | |
return d3_scale_identity(domain); | |
}; | |
return identity; | |
} | |
function d3_svg_arcInnerRadius(d) { | |
return d.innerRadius; | |
} | |
function d3_svg_arcOuterRadius(d) { | |
return d.outerRadius; | |
} | |
function d3_svg_arcStartAngle(d) { | |
return d.startAngle; | |
} | |
function d3_svg_arcEndAngle(d) { | |
return d.endAngle; | |
} | |
function d3_svg_line(projection) { | |
function line(data) { | |
function segment() { | |
segments.push("M", interpolate(projection(points), tension)); | |
} | |
var segments = [], points = [], i = -1, n = data.length, d, fx = d3_functor(x), fy = d3_functor(y); | |
while (++i < n) { | |
if (defined.call(this, d = data[i], i)) { | |
points.push([ +fx.call(this, d, i), +fy.call(this, d, i) ]); | |
} else if (points.length) { | |
segment(); | |
points = []; | |
} | |
} | |
if (points.length) segment(); | |
return segments.length ? segments.join("") : null; | |
} | |
var x = d3_svg_lineX, y = d3_svg_lineY, defined = d3_true, interpolate = d3_svg_lineLinear, interpolateKey = interpolate.key, tension = .7; | |
line.x = function(_) { | |
if (!arguments.length) return x; | |
x = _; | |
return line; | |
}; | |
line.y = function(_) { | |
if (!arguments.length) return y; | |
y = _; | |
return line; | |
}; | |
line.defined = function(_) { | |
if (!arguments.length) return defined; | |
defined = _; | |
return line; | |
}; | |
line.interpolate = function(_) { | |
if (!arguments.length) return interpolateKey; | |
if (typeof _ === "function") interpolateKey = interpolate = _; else interpolateKey = (interpolate = d3_svg_lineInterpolators.get(_) || d3_svg_lineLinear).key; | |
return line; | |
}; | |
line.tension = function(_) { | |
if (!arguments.length) return tension; | |
tension = _; | |
return line; | |
}; | |
return line; | |
} | |
function d3_svg_lineX(d) { | |
return d[0]; | |
} | |
function d3_svg_lineY(d) { | |
return d[1]; | |
} | |
function d3_svg_lineLinear(points) { | |
return points.join("L"); | |
} | |
function d3_svg_lineLinearClosed(points) { | |
return d3_svg_lineLinear(points) + "Z"; | |
} | |
function d3_svg_lineStepBefore(points) { | |
var i = 0, n = points.length, p = points[0], path = [ p[0], ",", p[1] ]; | |
while (++i < n) path.push("V", (p = points[i])[1], "H", p[0]); | |
return path.join(""); | |
} | |
function d3_svg_lineStepAfter(points) { | |
var i = 0, n = points.length, p = points[0], path = [ p[0], ",", p[1] ]; | |
while (++i < n) path.push("H", (p = points[i])[0], "V", p[1]); | |
return path.join(""); | |
} | |
function d3_svg_lineCardinalOpen(points, tension) { | |
return points.length < 4 ? d3_svg_lineLinear(points) : points[1] + d3_svg_lineHermite(points.slice(1, points.length - 1), d3_svg_lineCardinalTangents(points, tension)); | |
} | |
function d3_svg_lineCardinalClosed(points, tension) { | |
return points.length < 3 ? d3_svg_lineLinear(points) : points[0] + d3_svg_lineHermite((points.push(points[0]), points), d3_svg_lineCardinalTangents([ points[points.length - 2] ].concat(points, [ points[1] ]), tension)); | |
} | |
function d3_svg_lineCardinal(points, tension, closed) { | |
return points.length < 3 ? d3_svg_lineLinear(points) : points[0] + d3_svg_lineHermite(points, d3_svg_lineCardinalTangents(points, tension)); | |
} | |
function d3_svg_lineHermite(points, tangents) { | |
if (tangents.length < 1 || points.length != tangents.length && points.length != tangents.length + 2) { | |
return d3_svg_lineLinear(points); | |
} | |
var quad = points.length != tangents.length, path = "", p0 = points[0], p = points[1], t0 = tangents[0], t = t0, pi = 1; | |
if (quad) { | |
path += "Q" + (p[0] - t0[0] * 2 / 3) + "," + (p[1] - t0[1] * 2 / 3) + "," + p[0] + "," + p[1]; | |
p0 = points[1]; | |
pi = 2; | |
} | |
if (tangents.length > 1) { | |
t = tangents[1]; | |
p = points[pi]; | |
pi++; | |
path += "C" + (p0[0] + t0[0]) + "," + (p0[1] + t0[1]) + "," + (p[0] - t[0]) + "," + (p[1] - t[1]) + "," + p[0] + "," + p[1]; | |
for (var i = 2; i < tangents.length; i++, pi++) { | |
p = points[pi]; | |
t = tangents[i]; | |
path += "S" + (p[0] - t[0]) + "," + (p[1] - t[1]) + "," + p[0] + "," + p[1]; | |
} | |
} | |
if (quad) { | |
var lp = points[pi]; | |
path += "Q" + (p[0] + t[0] * 2 / 3) + "," + (p[1] + t[1] * 2 / 3) + "," + lp[0] + "," + lp[1]; | |
} | |
return path; | |
} | |
function d3_svg_lineCardinalTangents(points, tension) { | |
var tangents = [], a = (1 - tension) / 2, p0, p1 = points[0], p2 = points[1], i = 1, n = points.length; | |
while (++i < n) { | |
p0 = p1; | |
p1 = p2; | |
p2 = points[i]; | |
tangents.push([ a * (p2[0] - p0[0]), a * (p2[1] - p0[1]) ]); | |
} | |
return tangents; | |
} | |
function d3_svg_lineBasis(points) { | |
if (points.length < 3) return d3_svg_lineLinear(points); | |
var i = 1, n = points.length, pi = points[0], x0 = pi[0], y0 = pi[1], px = [ x0, x0, x0, (pi = points[1])[0] ], py = [ y0, y0, y0, pi[1] ], path = [ x0, ",", y0 ]; | |
d3_svg_lineBasisBezier(path, px, py); | |
while (++i < n) { | |
pi = points[i]; | |
px.shift(); | |
px.push(pi[0]); | |
py.shift(); | |
py.push(pi[1]); | |
d3_svg_lineBasisBezier(path, px, py); | |
} | |
i = -1; | |
while (++i < 2) { | |
px.shift(); | |
px.push(pi[0]); | |
py.shift(); | |
py.push(pi[1]); | |
d3_svg_lineBasisBez |