import re |
import re |
import uuid |
import uuid |
|
|
from sqlalchemy import Table, Column, MetaData |
from sqlalchemy import Table, Column, MetaData |
from sqlalchemy import types |
from sqlalchemy import types |
from sqlalchemy.sql import select |
from sqlalchemy.sql import select |
from sqlalchemy.orm import mapper |
from sqlalchemy.orm import mapper |
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 * |
|
|
def make_uuid(): |
def make_uuid(): |
return unicode(uuid.uuid4()) |
return unicode(uuid.uuid4()) |
|
|
|
|
|
|
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) |
|
|
|
|
metadata = MetaData() |
metadata = MetaData() |
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('metric', types.UnicodeText), |
Column('metric', types.UnicodeText), |
Column('value', types.UnicodeText), |
Column('value', types.UnicodeText), |
Column('url', types.UnicodeText), |
Column('url', types.UnicodeText), |
Column('department_id', types.UnicodeText), |
Column('department_id', types.UnicodeText), |
) |
) |
mapper(GA_Url, url_table) |
mapper(GA_Url, url_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' |
''' |
''' |
url = re.sub('https?://(www\.)?data.gov.uk', '', url) |
url = re.sub('https?://(www\.)?data.gov.uk', '', url) |
return url |
return url |
|
|
|
|
def _get_department_id_of_url(url): |
def _get_department_id_of_url(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 publisher_groups[0].name |
return publisher_groups[0].name |
|
|
|
|
def update_url_stats(period_name, period_complete_day, url_data): |
def update_url_stats(period_name, period_complete_day, url_data): |
table = get_table('ga_url') |
|
for url, views, next_page in url_data: |
for url, views, next_page in url_data: |
url = _normalize_url(url) |
url = _normalize_url(url) |
department_id = _get_department_id_of_url(url) |
department_id = _get_department_id_of_url(url) |
|
|
# see if the row for this url & month is in the table already |
# see if the row for this url & month is in the table already |
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).\ |
filter(GA_Url.url==url).\ |
filter(GA_Url.metric == 'Total views').first() |
filter(GA_Url.metric == 'Total views').first() |
if item: |
if item: |
item.period_name = period_complete_day = period_complete_day |
item.period_name = period_complete_day = period_complete_day |
item.value = views |
item.value = views |
item.department_id = department_id |
item.department_id = department_id |
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, |
'url': url, |
'url': url, |
'value': views, |
'value': views, |
'metric': 'Total views', |
'metric': 'Total views', |
'department_id': department_id |
'department_id': department_id |
} |
} |
model.Session.add(GA_Url(**values)) |
model.Session.add(GA_Url(**values)) |
model.Session.commit() |
model.Session.commit() |
|
|