tommy/libpostal
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[openaddresses] some postcodes in OpenAddresses are stored as floats, convert to int and then to string if that's the case

Browse Source
This commit is contained in:
Al
2016-08-26 19:12:48 -04:00
parent 155d46f46c
commit 56329439af
1 changed files with 10 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
scripts/geodata/openaddresses/formatter.py
View File

@@ -11,7 +11,7 @@ from geodata.address_expansions.gazetteers import street_types_gazetteer, unit_t
from geodata.address_formatting.formatter import AddressFormatter from geodata.address_formatting.formatter import AddressFormatter
from geodata.addresses.components import AddressComponents from geodata.addresses.components import AddressComponents
from geodata.countries.names import country_names from geodata.countries.names import country_names
from geodata.encoding import safe_decode from geodata.encoding import safe_decode, safe_encode
from geodata.math.sampling import cdf, weighted_choice from geodata.math.sampling import cdf, weighted_choice
from geodata.text.utils import is_numeric, is_numeric_strict from geodata.text.utils import is_numeric, is_numeric_strict
@@ -209,6 +209,15 @@ class OpenAddressesFormatter(object):
postcode = components.get(AddressFormatter.POSTCODE, None) postcode = components.get(AddressFormatter.POSTCODE, None)
if postcode: if postcode:
postcode = postcode.strip() postcode = postcode.strip()
try:
postcode = int(postcode)
except (ValueError, TypeError):
try:
postcode = float(postcode)
postcode = safe_decode(int(postcode))
except (ValueError, TypeError):
postcode = safe_decode(postcode)
if postcode_strip_non_digit_chars: if postcode_strip_non_digit_chars:
postcode = six.u('').join((c for c in postcode if c.isdigit())) postcode = six.u('').join((c for c in postcode if c.isdigit()))