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

[osm] Remove boundary names with trailing commas

Browse Source
This commit is contained in:
Al
2016-08-02 03:13:05 -04:00
parent 12466b12dc
commit 4ab60cd4fc
1 changed files with 14 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
scripts/geodata/addresses/components.py
View File

@@ -982,6 +982,19 @@ class AddressComponents(object):
if value.isdigit(): if value.isdigit():
address_components.pop(component) address_components.pop(component)
def cleanup_boundary_names(self, address_components):
'''
Boundary name cleanup
---------------------
Cleanup things like addr:city=Rockport,
'''
for component in list(address_components):
if component not in self.BOUNDARY_COMPONENTS:
continue
address_components[component] = address_components[component].strip(six.u(', '))
def prune_duplicate_names(self, address_components): def prune_duplicate_names(self, address_components):
''' '''
Name deduping Name deduping
@@ -1199,6 +1212,7 @@ class AddressComponents(object):
self.cleanup_house_number(address_components) self.cleanup_house_number(address_components)
self.remove_numeric_boundary_names(address_components) self.remove_numeric_boundary_names(address_components)
self.cleanup_boundary_names(address_components)
self.add_house_number_phrase(address_components, language, country=country) self.add_house_number_phrase(address_components, language, country=country)
self.add_postcode_phrase(address_components, language, country=country) self.add_postcode_phrase(address_components, language, country=country)