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

[osm] splitting streets (way names) on semicolons

Browse Source
This commit is contained in:
Al
2016-01-24 00:42:25 -05:00
parent 3485738c2b
commit bab7a0f961
1 changed files with 7 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

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

@@ -498,12 +498,13 @@ def build_ways_training_data(language_rtree, infile, out_dir):
continue
for lang, val in name_language.iteritems():
for s in val:
if lang in languages:
writer.writerow((lang, country, tsv_string(s)))
abbrev = osm_abbreviate(street_types_gazetteer, s, lang, abbreviate_prob=1.0, separate_prob=0.5)
if abbrev != s:
writer.writerow((lang, country, tsv_string(abbrev)))
for v in val:
for s in v.split(';'):
if lang in languages:
writer.writerow((lang, country, tsv_string(s)))
abbrev = osm_abbreviate(street_types_gazetteer, s, lang, abbreviate_prob=1.0, separate_prob=0.5)
if abbrev != s:
writer.writerow((lang, country, tsv_string(abbrev)))
if i % 1000 == 0 and i > 0:
print('did {} ways'.format(i))
i += 1