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

[osm] Adding new localized country names in anguage training data for formatted addresses

Browse Source
This commit is contained in:
Al
2015-11-24 21:49:10 -05:00
parent e40ca0bb89
commit 8c422a6e61
1 changed files with 13 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
scripts/geodata/osm/osm_address_training_data.py
View File

@@ -692,13 +692,19 @@ def build_address_format_training_data_limited(language_rtree, infile, out_dir):
remove_keys = NAME_KEYS + HOUSE_NUMBER_KEYS + COUNTRY_KEYS + POSTAL_KEYS + OSM_IGNORE_KEYS remove_keys = NAME_KEYS + HOUSE_NUMBER_KEYS + COUNTRY_KEYS + POSTAL_KEYS + OSM_IGNORE_KEYS
country_keys_set = set(COUNTRY_KEYS)
for key, value, deps in parse_osm(infile): for key, value, deps in parse_osm(infile):
try: try:
latitude, longitude = latlon_to_decimal(value['lat'], value['lon']) latitude, longitude = latlon_to_decimal(value['lat'], value['lon'])
except Exception: except Exception:
continue continue
have_country = False
for k in remove_keys: for k in remove_keys:
if k in country_keys_set:
have_country = True
_ = value.pop(k, None) _ = value.pop(k, None)
if not value: if not value:
@@ -709,10 +715,17 @@ def build_address_format_training_data_limited(language_rtree, infile, out_dir):
continue continue
single_language = len(name_language) == 1 single_language = len(name_language) == 1
for lang, val in name_language.iteritems(): for lang, val in name_language.iteritems():
if lang not in languages: if lang not in languages:
continue continue
if have_country:
localized = language_country_names.get(lang, {}).get(country.upper())
if localized:
value['addr:country:{}'.format(lang)] = localized
address_dict = value.copy() address_dict = value.copy()
for k in address_dict.keys(): for k in address_dict.keys():
namespaced_val = u'{}:{}'.format(k, lang) namespaced_val = u'{}:{}'.format(k, lang)