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

[osm] Using street for language disambiguation in training data

Browse Source
This commit is contained in:
Al
2015-09-21 04:09:15 -04:00
parent ccac4a5a90
commit 134cf616d6
1 changed files with 9 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

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

@@ -396,10 +396,15 @@ def build_address_format_training_data(language_rtree, infile, out_dir, tag_comp
for key in remove_keys: for key in remove_keys:
_ = value.pop(key, None) _ = value.pop(key, None)
if len(candidate_languages) == 1: language = None
language = candidate_languages[0]['lang'] if tag_components:
else: if len(candidate_languages) == 1:
language = disambiguate_language(v, [(l['lang'], l['default']) for l in candidate_languages]) language = candidate_languages[0]['lang']
else:
street = value.get('addr:street', None)
if street is None:
continue
language = disambiguate_language(street, [(l['lang'], l['default']) for l in candidate_languages])
formatted_address = formatter.format_address(country, value, tag_components=tag_components) formatted_address = formatter.format_address(country, value, tag_components=tag_components)
if formatted_address is not None: if formatted_address is not None: