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

[osm] OSM untagged formatted addresses try to use language namespaced tags

Browse Source
This commit is contained in:
Al
2015-08-18 16:18:27 -04:00
parent 330002197a
commit 0528d1b578
1 changed files with 23 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
scripts/geodata/osm/osm_address_training_data.py
View File

@@ -638,18 +638,31 @@ def build_address_format_training_data_limited(language_rtree, infile, out_dir):
if not name_language: if not name_language:
continue continue
formatted_address_untagged = formatter.format_address(country, value, tag_components=False) single_language = len(name_language) == 1
if formatted_address_untagged is not None: for lang, val in name_language.iteritems():
formatted_address_untagged = tsv_string(formatted_address_untagged) if lang not in languages:
continue
for k, v in name_language.iteritems(): address_dict = value.copy()
for s in v: for k in address_dict.keys():
if k in languages: namespaced_val = u'{}:{}'.format(k, lang)
writer.writerow((k, country, formatted_address_untagged)) if namespaced_val in address_dict:
address_dict[k] = d[namespaced_val]
elif not single_language:
address_dict.pop(k)
i += 1 if not address_dict:
if i % 1000 == 0 and i > 0: continue
print 'did', i, 'formatted addresses'
formatted_address_untagged = formatter.format_address(country, address_dict, tag_components=False)
if formatted_address_untagged is not None:
formatted_address_untagged = tsv_string(formatted_address_untagged)
writer.writerow((lang, country, formatted_address_untagged))
i += 1
if i % 1000 == 0 and i > 0:
print 'did', i, 'formatted addresses'
def build_address_training_data(langauge_rtree, infile, out_dir, format=False): def build_address_training_data(langauge_rtree, infile, out_dir, format=False):