def dataset(name):
if not name in DATASETS:
DATASETS[name] = Dataset(name, api_key=app.config['NOMENKLATURA_API_KEY'])
if app.config['NOMENKLATURA_PRELOAD']:
for entity in DATASETS[name].entities():
ENTITIES[(name, entity.name.lower())] = entity
for alias in DATASETS[name].aliases():
if alias.is_invalid:
ENTITIES[(name, alias.name.lower())] = \
Dataset.Invalid({'dataset': alias.dataset, 'name': alias.name})
elif alias.is_matched:
ENTITIES[(name, alias.name.lower())] = \
Entity(DATASETS[name], alias.entity)
return DATASETS[name]
def extract_sections(party):
doc = load_doc(party)
nomenklatura = Dataset('btw13-titles',
api_key=os.environ.get('NOMENKLATURA_API_KEY'))
current = Section(party)
for i, h in enumerate(doc.findall('.//*')):
if h.tag in ['h1', 'h2']:
if current.valid:
yield current
current = Section(party)
current.title = h.text
current.level = h.tag[1:]
fp = '[%s:%s] %s' % (party, h.tag, h.text)
try:
entity = nomenklatura.lookup(fp)
current.topic = entity.name
except Exception, e:
print [fp]
print e
#if h.getparent() == doc:
# #print "XXX", h
current.texts.append(h.text)
from nomenklatura import Dataset, NomenklaturaException
dataset = Dataset('iso-countries')
print [dataset.label]
#print list(dataset.entities())
#cyp = dataset.entity_by_name('cyprus')
#print cyp.__data__
#cyp.reviewed = False
#print cyp.__data__
#cyp.update()
e = dataset.entity_by_name('christma island')
print repr(e)
print repr(e.dereference())
from nomenklatura import Dataset
from utils import KorboDataset
NOMENKLATURA_URL = 'http://nomenklatura.venturi.fbk.eu/'
NOMENKLATURA_PROJ = 'pois'
KORBO_URL = 'http://korbo.netseven.it/'
KORBO_PROJ = '101'
nk_dataset = Dataset(NOMENKLATURA_PROJ, host=NOMENKLATURA_URL)
ko_dataset = KorboDataset(KORBO_PROJ, host=KORBO_URL)
nk_values = sorted(nk_dataset.values(), key=lambda x: x.value)
ko_values = sorted(ko_dataset.values(), key=lambda x: x.value)
nk_values_set = frozenset(nk_values)
ko_values_set = frozenset(ko_values)
if len(nk_values) != len(nk_values_set):
print "FOUND DUPLICATES ON NOMENKLATURA: ", nk_values - nk_values_set
if len(ko_values) != len(ko_values_set):
print "FOUND DUPLICATES ON KORBO: ", nk_values - nk_values_set
cnt = 0
for nk_value in nk_values:
if nk_value.value not in ko_values:
cnt += 1
print nk_value, "NOT FOUND ON KORBO"
print "KORBO PROJ:", cnt, "issues found"
from nomenklatura import Dataset
from utils import KorboDataset
NOMENKLATURA_URL = 'http://nomenklatura.venturi.fbk.eu/'
NOMENKLATURA_PROJ = 'pois'
KORBO_URL = 'http://korbo.netseven.it/'
KORBO_PROJ = '101'
nk_dataset = Dataset(NOMENKLATURA_PROJ, host=NOMENKLATURA_URL)
ko_dataset = KorboDataset(KORBO_PROJ, host=KORBO_URL)
nk_values = sorted(nk_dataset.values(), key=lambda x: x.value)
ko_values = sorted(ko_dataset.values(), key=lambda x: x.value)
nk_values_set = frozenset(nk_values)
ko_values_set = frozenset(ko_values)
if len(nk_values) != len(nk_values_set):
print "FOUND DUPLICATES ON NOMENKLATURA: ", nk_values - nk_values_set
if len(ko_values) != len(ko_values_set):
print "FOUND DUPLICATES ON KORBO: ", nk_values - nk_values_set
cnt = 0
for nk_value in nk_values:
if nk_value.value not in ko_values:
cnt += 1
print nk_value, "NOT FOUND ON KORBO"
print "KORBO PROJ:", cnt, "issues found"
def validate_label(label):
if not label[0:1].isupper():