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

[osm] adding country-specific cleanups to OSM place training data

Browse Source
This commit is contained in:
Al
2016-10-05 17:13:13 -04:00
parent 6b0186782d
commit 66af532850
1 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
scripts/geodata/osm/formatter.py
View File

@@ -651,13 +651,16 @@ class OSMAddressFormatter(object):
for address_components, language, is_default in place_tags:
revised_address_components = place_config.dropout_components(address_components, osm_components, country=country, population=population)
revised_address_components[component_name] = address_components[component_name]
self.components.cleanup_boundary_names(revised_address_components)
self.components.country_specific_cleanup(revised_address_components, country)
self.components.drop_invalid_components(revised_address_components, country)
self.components.replace_name_affixes(revised_address_components, language)
self.components.replace_names(revised_address_components)
self.components.remove_numeric_boundary_names(revised_address_components)
self.components.cleanup_boundary_names(revised_address_components)
if (AddressFormatter.COUNTRY in address_components or place_config.include_component(AddressFormatter.COUNTRY, containing_ids, country=country)) and random.random() < cldr_country_prob:
address_country = self.components.cldr_country_name(country, language)