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

[fix] Exclude polygons with > 1 regional language

Browse Source
This commit is contained in:
Al
2015-09-03 20:48:04 -04:00
parent 9a9530c1b9
commit e7cf5000fe
1 changed files with 5 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
scripts/geodata/osm/osm_address_training_data.py
View File

@@ -626,10 +626,12 @@ def build_toponym_training_data(language_rtree, infile, out_dir):
official = official_languages[country] official = official_languages[country]
num_langs = len(official) num_langs = len(candidate_languages)
default_langs = set([l for l, default in official.iteritems() if default]) default_langs = set([l for l, default in official.iteritems() if default])
num_defaults = len(default_langs) num_defaults = len(default_langs)
regional_langs = [p['lang'] for p in language_props if p.get('admin_level', 0) > 0]
top_lang = None top_lang = None
if len(official) > 0: if len(official) > 0:
top_lang = official.iterkeys().next() top_lang = official.iterkeys().next()
@@ -664,8 +666,8 @@ def build_toponym_training_data(language_rtree, infile, out_dir):
have_qualified_names = True have_qualified_names = True
name_language[lang].append(v) name_language[lang].append(v)
if not have_qualified_names and num_langs == 1 and 'name' in value: if not have_qualified_names and len(regional_langs) <= 1 and num_langs == 1 and 'name' in value:
name_language[top_lang].append(value['name']) name_language[candidate_languages[0]['lang']].append(value['name'])
for k, v in name_language.iteritems(): for k, v in name_language.iteritems():
for s in v: for s in v: