def load_2002_file(self, mapping, contest):
headers = [
'office',
'fill',
'jurisdiction',
'last',
'middle',
'first',
'party',
'winner',
'vote_type',
'votes',
'fill2'
]
# Store result instances for bulk loading
results = []
with open(join(self.cache_dir, mapping['generated_filename']), 'rU') as csvfile:
reader = unicodecsv.DictReader(csvfile, fieldnames = headers, delimiter='|', encoding='latin-1')
reporting_level = 'county'
#TODO: replace jurisidctions with datasource.mappings
#jurisdictions = self.jurisdiction_mappings(('ocd','fips','urlname','name'))
for row in reader:
if row['jurisdiction'].strip() == 'Baltimore City':
jurisdiction = [x for x in jurisdictions if x['name'] == "Baltimore City"][0]
else:
jurisdiction = [x for x in jurisdictions if x['name']+" County" == row['jurisdiction'].strip()][0]
if row['winner'] == '1':
winner = True
else:
winner = False
if row['last'].strip() == 'zz998':
name = row['last'].strip()
candidate = Candidate(state=self.state.upper(), family_name=name)
write_in=True
else:
cand_name = self.combine_name_parts([row['first'], row['middle'], row['last']])
name = self.parse_name(cand_name)
write_in=False
candidate = Candidate(state=self.state.upper(), given_name=name.first, additional_name=name.middle, family_name=name.last, suffix=name.suffix, name=name.full_name)
result_kwargs = {
'ocd_id': ocd_id + "/precinct:" + str(row['Election District']) + "-" + str(row['Election Precinct']),
'jurisdiction': jurisdiction,
'raw_office': row['Office Name'] + ' ' + row['Office District'],
'reporting_level': reporting_level,
'candidate': candidate,
'party': row['Party'],
'write_in': write_in,
'total_votes': total_votes,
'vote_breakdowns': vote_breakdowns
}
result = Result(**result_kwargs)
#self.load_county_2002(row, jurisdiction['ocd'], jurisdiction['name'], candidate, reporting_level, write_in, winner)))
Result.insert(results)
def validate_obama_primary_candidacy_2012():
"""Should only be one Obama primary candidacy for 2012"""
elec_id= 'md-2012-04-03-primary'
kwargs = {
'election_id': elec_id,
'contest_slug': 'president-d',
'slug': 'barack-obama',
}
try:
cand = Candidate.objects.get(**kwargs)
print "PASS: 1 obama primary candidacy found for 2012: %s" % "-".join(cand.key)
except Candidate.DoesNotExist:
raise Candidate.DoesNotExist("zero obama primary candidacies found for 2012")
except Candidate.MultipleObjectsReturned as e:
raise Candidate.MultipleObjectsReturned("multiple obama primary candidacies found for 2012: %s" % e)
def __call__(self):
candidates = []
seen = set()
for rr in self.get_rawresults():
key = (rr.election_id, rr.contest_slug, rr.candidate_slug)
if key not in seen:
fields = self.get_candidate_fields(rr)
fields['contest'] = self.get_contest(rr)
if "other" in fields['full_name'].lower():
if fields['full_name'] == "Other Write-Ins":
fields['flags'] = [
'aggregate',
]
else:
# As far as I can tell the value should always be
# "Other Write-Ins", but output a warning to let us
# know about some cases we may be missing.
logging.warn("'other' found in candidate name field"
"value: '%s'" % rr.full_name)
candidate = Candidate(**fields)
candidates.append(candidate)
seen.add(key)
Candidate.objects.insert(candidates, load_bulk=False)
print "Created %d candidates." % len(candidates)
def load_non2002_file(self, mapping, contest):
with open(join(self.cache_dir, mapping['generated_filename']), 'rU') as csvfile:
reader = unicodecsv.DictReader(csvfile, encoding='latin-1')
if 'state_legislative' in mapping['generated_filename']:
reporting_level = 'state_legislative'
districts = [j for j in reader.fieldnames if j not in ['Vote Type','Vote Type (FOR=1,AGAINST=2)', 'County', 'Candidate Name', 'Party', 'Office Name', 'Office District', 'Winner', 'Write-In?']]
elif 'precinct' in mapping['generated_filename']:
reporting_level = 'precinct'
jurisdiction = mapping['name']
else:
reporting_level = 'county'
jurisdiction = mapping['name']
for row in reader:
if not row['Office Name'].strip() in ['President - Vice Pres', 'U.S. Senator', 'U.S. Congress', 'Governor / Lt. Governor', 'Comptroller', 'Attorney General', 'State Senator', 'House of Delegates']:
continue
# parse candidate - we get full names
name = self.parse_name(row['Candidate Name'])
# if office is president, then skip state in lookup, otherwise use
if row['Office Name'] == 'President - Vice Pres':
candidate = Candidate(state='US', given_name=name.first, additional_name=name.middle, family_name=name.last, suffix=name.suffix, name=name.full_name)
else:
candidate = Candidate(state=self.state.upper(), given_name=name.first, additional_name=name.middle, family_name=name.last, suffix=name.suffix, name=name.full_name)
# sometimes write-in field not present
try:
if row['Write-In?'] == 'Y':
write_in = True
else:
write_in = False
except KeyError:
write_in = None
if row['Winner'] == 'Y':
winner = True
else:
winner = False
if reporting_level == 'state_legislative':
contest.update(push_all__results=(self.load_state_legislative(row, districts, candidate, reporting_level, write_in, winner)))
elif reporting_level == 'county':
contest.update(push__results=(self.load_county(row, mapping['ocd_id'], jurisdiction, candidate, reporting_level, write_in, winner)))
elif reporting_level == 'precinct':
contest.update(push__results=(self.load_precinct(row, mapping['ocd_id'], jurisdiction, candidate, reporting_level, write_in, winner)))
def test_get_list_filter_by_level(self):
level = 'precinct'
# Make sure there's at least one result at this level
if Result.objects(reporting_level=level).count() == 0:
candidate = Candidate.objects()[0]
contest = candidate.contest
ResultFactory(candidate=candidate, contest=contest,
reporting_level=level)
data = self.roller.get_list(state="MD", reporting_level=level)
self.assertEqual(len(data),
Result.objects(state="MD", reporting_level=level).count())
def test_get_list_filter_by_level(self):
level = 'precinct'
# Make sure there's at least one result at this level
if Result.objects(reporting_level=level).count() == 0:
candidate = Candidate.objects()[0]
contest = candidate.contest
ResultFactory(candidate=candidate,
contest=contest,
reporting_level=level)
data = self.roller.get_list(state="MD", reporting_level=level)
self.assertEqual(
len(data),
Result.objects(state="MD", reporting_level=level).count())
def __call__(self):
candidates = []
seen = set()
for rr in self.get_raw_results():
key = (rr.election_id, rr.contest_slug, rr.candidate_slug)
if key not in seen:
fields = self.get_candidate_fields(rr)
if not fields['full_name']:
quit(fields)
fields['contest'] = self.get_contest(rr)
#print fields
candidate = Candidate(**fields)
candidates.append(candidate)
seen.add(key)
Candidate.objects.insert(candidates, load_bulk=False)
logger.info("Created {} candidates.".format(len(candidates)))
def load_2000_primary_file(self, mapping, contest):
candidates = {}
results = []
with open(join(self.cache_dir, mapping['generated_filename']), 'rU') as csvfile:
lines = csvfile.readlines()
#TODO: use Datasource.mappings instead of jurisdiction_mappings
#import ipdb;ipdb.set_trace()
#jurisdictions = self.jurisdiction_mappings(('ocd','fips','urlname','name'))
#jurisdictions = self.datasource.mappings()
for line in lines[0:377]:
if line.strip() == '':
continue # skip blank lines
else: # determine if this is a row with an office
cols = line.strip().split(',')
if cols[0][1:29] == 'President and Vice President':
office_name = 'President - Vice Pres'
if 'Democratic' in cols[0]:
party = 'DEM'
else:
party = 'REP'
continue
elif cols[0][1:13] == 'U.S. Senator':
office_name = 'U.S. Senator'
if 'Democratic' in cols[0]:
party = 'DEM'
else:
party = 'REP'
continue
elif cols[0][1:27] == 'Representative in Congress':
district = int(cols[0][71:73])
office_name = 'U.S. Congress ' + str(district)
if 'Democratic' in cols[0]:
party = 'DEM'
else:
party = 'REP'
continue
# skip offices we don't want
elif cols[0][1:21] == 'Judge of the Circuit':
continue
elif cols[0][1:31] == 'Female Delegates and Alternate':
continue
elif cols[0][1:29] == 'Male Delegates and Alternate':
continue
elif cols[0][1:28] == 'Delegates to the Republican':
continue
elif cols[0] == '""':
candidates = [x.replace('"','').strip() for x in cols if x.replace('"','') != '']
winner = [i for i, j in enumerate(candidates) if 'Winner' in j][0] # index of winning candidate
candidates[winner] = candidates[winner].split(' Winner')[0]
# handle name_parse and Uncommitted candidate
else: # county results
result = [x.replace('"','').strip() for x in cols if x != '']
#juris = [j for j in jurisdictions if j['name'] == result[0].strip()][0]
juris = [j for j in self.datasource.mappings() if j['name'] == result[0].strip()][0]
cand_results = zip(candidates, result[1:])
for cand, votes in cand_results:
name = self.parse_name(cand)
#cand_kwargs = {
#'given_name': name.first,
#'additional_name': name.middle,
#'family_name': name.last,
#'suffix': name.suffix,
#'name': name.full_name
#}
if office_name == 'President - Vice Pres':
cand_kwargs['state'] = 'US'
else:
cand_kwargs['state'] = self.state.upper()
#TODO Get or create candidate
candidate = Candidate(**cand_kwargs)
result_kwargs = {
'candidate': candidate,
'ocd_id': juris['ocd'],
'jurisdiction': juris['name'],
'raw_office': office_name,
'reporting_level': 'county',
'party': party,
'write_in': False,
'total_votes': int(votes),
'vote_breakdowns': {}
}
result = Result(**result_kwargs)
results.append(result)