def extract():
""" Extract all data from the respective web sources. """
from lobbyfacts.data import etl_engine
engine = etl_engine()
from lobbyfacts.data.extract.reginterests import extract
extract(engine)
from lobbyfacts.data.extract.regaccredit import extract
extract(engine)
def load():
""" Load the data from ETL into the production database. """
from lobbyfacts.data import etl_engine
engine = etl_engine()
from lobbyfacts.data.load.common import load
load(engine)
from lobbyfacts.data.load.reginterests import load
load(engine)
from lobbyfacts.model import update_index
update_index()
def transform():
""" Apply simple data transformation and cleansing operations. """
from lobbyfacts.data import etl_engine
engine = etl_engine()
from lobbyfacts.data.transform.categories import transform
transform(engine)
#from lobbyfacts.data.transform.pbnetworks import transform
#transform(engine)
from lobbyfacts.data.transform.names import transform
transform(engine)
#from lobbyfacts.data.transform.dedup import transform
#transform(engine)
from lobbyfacts.data.transform.geocode import transform
transform(engine)
def extract():
""" Extract all data from the respective web sources. """
from lobbyfacts.data import etl_engine
engine = etl_engine()
from lobbyfacts.data.extract.reginterests import extract
extract(engine)
from lobbyfacts.data.extract.regaccredit import extract
extract(engine)
#from lobbyfacts.data.extract.regexpert import extract
#extract(engine)
from lobbyfacts.data.extract.unreginterest import extract
extract(engine)
from lobbyfacts.data.extract.tag import extract
extract(engine)
from lobbyfacts.data.extract.meetings import extract
extract(engine)
if out is None:
d['etl_clean'] = False
else:
d[out_column] = out
except ValueError, ve:
d['etl_clean'] = False
sl.upsert(engine, table, d, [source_column])
def transform(engine):
countries_func = lambda v, c: country_by_name(v).get('iso2')
map_names(countries_func, engine, 'representative',
'contact_country', 'country_code')
map_names(countries_func, engine, 'country_of_member',
'country', 'country_code')
map_names(countries_func, engine, 'expertgroup_member_country',
'country', 'country_code')
names_func = lambda v, c: canonical(DATASET, v, context=c)
map_names(names_func, engine, 'representative')
#map_names(names_func, engine, 'person')
map_names(names_func, engine, 'financial_data_turnover')
map_names(names_func, engine, 'organisation')
map_names(names_func, engine, 'network_entity')
map_names(names_func, engine, 'expertgroup_member')
if __name__ == '__main__':
engine = etl_engine()
transform(engine)
