def test_compare(self):
from geoid.acs import Tract
t1 = Tract.parse('14000US06001442800')
t2 = Tract.parse('14000US06037205110')
print(t1 < t2)
def test_compare(self):
from geoid.acs import Tract
t1 = Tract.parse('14000US06001442800')
t2 = Tract.parse('14000US06037205110')
print(t1 < t2)
def row_generator(resource, doc, env, *args, **kwargs):
""" An example row generator function.
Reference this function in a Metatab file as the value of a Datafile:
Datafile: python:pylib#row_generator
The function must yield rows, with the first being headers, and subsequenct rows being data.
:param resource: The Datafile term being processed
:param doc: The Metatab document that contains the term being processed
:param args: Positional arguments passed to the generator
:param kwargs: Keyword arguments passed to the generator
:return:
The env argument is a dict with these environmental keys:
* CACHE_DIR
* RESOURCE_NAME
* RESOLVED_URL
* WORKING_DIR
* METATAB_DOC
* METATAB_WORKING_DIR
* METATAB_PACKAGE
It also contains key/valu pairs for all of the properties of the resource.
"""
yield 'int float prop ratio str acs_tract census_tract tiger_tract date time datetime'.split()
for i in range(10):
tract = Tract(6, 72, randint(1, 10))
yield [
i,
float(i) / pi,
random(),
float(randint(1, 1000)) / float(randint(1, 1000)),
'string-' + str(i),
tract,
tract.convert(CensusGeoid),
tract.convert(TigerGeoid),
date(randint(2000, 2020), randint(1, 12), randint(1, 27)),
time(randint(0, 23), randint(0, 59), randint(0, 59)),
datetime(randint(2000, 2020), randint(1, 12), randint(1, 27), randint(0, 23), randint(0, 59),
randint(0, 59))
]
def tract_geoid(row):
"""COnvert doted notation into a geoid """
try:
a, b = row['tract'].split('.')
except ValueError:
a = row['tract']
b = '0'
return Tract(6, 73, int(a) * 100 + int(b.ljust(2, '0')))
def parse_tract(v, row, row_n, i_s, i_d, header_s, header_d, scratch, errors,
accumulator):
""" Parse the tract number, within Los Angeles County
"""
from geoid.acs import Tract
# 6 == CA, 037 == Los Angeles County
return Tract(6, 37, v)
def lookup(zip, n):
try:
# The map will return a Census geoid, which has 11 charasters, but it is often missing
# the leading 0, so we have to put it back. Then it much be converted to an
# ACS Tract
census_tract_str = str(f_map[int(zip)](int(n)%100 / 100.0)).zfill(11)
return str(AcsTract.parse(census_tract_str))
except KeyError:
return None
def row_generator(resource, doc, env, *args, **kwargs):
""" An example row generator function.
Reference this function in a Metatab file as the value of a Datafile:
Datafile: python:pylib#row_generator
The function must yield rows, with the first being headers, and subsequenct rows being data.
:param resource: The Datafile term being processed
:param doc: The Metatab document that contains the term being processed
:param args: Positional arguments passed to the generator
:param kwargs: Keyword arguments passed to the generator
:return:
The env argument is a dict with these environmental keys:
* CACHE_DIR
* RESOURCE_NAME
* RESOLVED_URL
* WORKING_DIR
* METATAB_DOC
* METATAB_WORKING_DIR
* METATAB_PACKAGE
It also contains key/value pairs for all of the properties of the resource.
"""
from operator import itemgetter
from geoid.acs import Tract
st = resource.schema_term
header = []
columns = []
for c in st.children:
header.append(c.name)
if c.get('col_pos'):
columns.append(int(c.col_pos) - 1)
yield header
ig = itemgetter(*columns)
for ref in doc.references():
for row in ref:
yield (Tract(row[2], row[3], row[4]), ) + ig(row)
def extract_county(resource, doc, env, *args, **kwargs):
from geoid.acs import Tract
r = doc.reference('census_planning_database')
state_id = int(resource.state)
county_id = int(resource.county)
for i, row in enumerate(r.iterdict):
state = row['State']
county = row['County']
tract = row['Tract']
del row['GIDTR']
del row['State']
del row['County']
del row['State_name']
del row['County_name']
del row['Tract']
del row['Num_BGs_in_Tract']
if i == 0:
yield ['geoid'] + list(row.keys())
if int(state) == state_id and int(county) == county_id:
for col in ['Med_House_Value_ACS_12_16', 'Med_House_Value_ACSMOE_12_16',
'Aggr_House_Value_ACS_12_16', 'Aggr_House_Value_ACSMOE_12_16',
'avg_Agg_HH_INC_ACS_12_16', 'avg_Agg_HH_INC_ACSMOE_12_16',
'Med_HHD_Inc_ACS_12_16', 'Med_HHD_Inc_ACSMOE_12_16',
'Aggregate_HH_INC_ACS_12_16', 'Aggregate_HH_INC_ACSMOE_12_16',
'avg_Agg_House_Value_ACS_12_16','avg_Agg_House_Value_ACSMOE_12_16']:
row[col] = clean_currency(row[col])
tract = Tract(state, county, tract )
yield [str(tract)] + list(row.values())
for row_n, was_geocoded, row in chunked_geocode(doc):
if not row.get('tract_geoid'):
row['tract_geoid'] = zip_to_tract(fac_zip[int(row['unique_id'])], int(row['unique_id']))
row['side_of_street'] = None
row['tiger_id'] = None
if row['tract_geoid']:
if len(row['tract_geoid']) != 18:
# It's probably still a Census Tract, so convert it to an Acs tract.
row['tract_geoid'] = str(CensusTract.parse(row['tract_geoid'].zfill(11)).convert(AcsTract))
assert(len(row['tract_geoid'])) == 18, row['tract_geoid']
t = AcsTract.parse(row['tract_geoid'])
#print(str(t), file=sys.stderr)
row['state_fips'] = t.state
row['county_fips'] = t.county
row['tract_fips'] = t.tract
if row.get('state_fips'):
row['state_fips'] = str(row['state_fips']).zfill(2)
if row.get('county_fips'):
row['county_fips'] = str(row.get('county_fips')).zfill(3)
try:
def mk_tract_geoid(v):
if v is None or v is np.nan:
return None
else:
return str(Tract(6, 73, int(v)))