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

[fix] city name in OSM formatting

Browse Source
This commit is contained in:
Al
2015-12-07 02:33:12 -05:00
parent a066ee9aad
commit f8a3081d0f
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
scripts/geodata/osm/osm_address_training_data.py
View File

@@ -589,6 +589,8 @@ def build_address_format_training_data(admin_rtree, language_rtree, neighborhood
if osm_components: if osm_components:
poly_components = defaultdict(list) poly_components = defaultdict(list)
existing_city_name = address_components.get(AddressFormatter.CITY)
for component, components_values in osm_components.iteritems(): for component, components_values in osm_components.iteritems():
seen = set() seen = set()
@@ -610,12 +612,9 @@ def build_address_format_training_data(admin_rtree, language_rtree, neighborhood
name = component_value[simple_name_key] name = component_value[simple_name_key]
elif international_name_key in component_value: elif international_name_key in component_value:
name = component_value[international_name_key] name = component_value[international_name_key]
if not name: if not name:
name = component_value.get(key, component_value.get(raw_key)) name = component_value.get(key, component_value.get(raw_key))
existing_city_name = address_components.get(AddressFormatter.CITY)
if not name or (component != AddressFormatter.CITY and name == existing_city_name): if not name or (component != AddressFormatter.CITY and name == existing_city_name):
name = component_value.get(name_key, component_value.get(raw_name_key)) name = component_value.get(name_key, component_value.get(raw_name_key))
@@ -654,6 +653,7 @@ def build_address_format_training_data(admin_rtree, language_rtree, neighborhood
if not names or lang not in names: if not names or lang not in names:
continue continue
city = None
if 'abbr' not in names or non_local_language: if 'abbr' not in names or non_local_language:
# Use the common city name in the target language # Use the common city name in the target language
city = names[lang][0][0] city = names[lang][0][0]