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

[osm] Removing postal code keys in formatted language training data

Browse Source
This commit is contained in:
Al
2015-08-24 14:08:36 -04:00
parent 8bbcb60aee
commit a2ec8001b0
1 changed files with 9 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
scripts/geodata/osm/osm_address_training_data.py
View File

@@ -491,6 +491,12 @@ COUNTRY_KEYS = (
'country_name', 'country_name',
'addr:country', 'addr:country',
) )
POSTAL_KEYS = (
'postcode',
'postal_code',
'addr:postcode',
'addr:postal_code',
)
def build_address_format_training_data_limited(language_rtree, infile, out_dir): def build_address_format_training_data_limited(language_rtree, infile, out_dir):
@@ -501,13 +507,15 @@ def build_address_format_training_data_limited(language_rtree, infile, out_dir):
f = open(os.path.join(out_dir, ADDRESS_FORMAT_DATA_LANGUAGE_FILENAME), 'w') f = open(os.path.join(out_dir, ADDRESS_FORMAT_DATA_LANGUAGE_FILENAME), 'w')
writer = csv.writer(f, 'tsv_no_quote') writer = csv.writer(f, 'tsv_no_quote')
remove_keys = NAME_KEYS + COUNTRY_KEYS + POSTAL_KEYS
for key, value in parse_osm(infile): for key, value in parse_osm(infile):
try: try:
latitude, longitude = latlon_to_floats(value['lat'], value['lon']) latitude, longitude = latlon_to_floats(value['lat'], value['lon'])
except Exception: except Exception:
continue continue
for k in NAME_KEYS + COUNTRY_KEYS: for k in remove_keys:
_ = value.pop(k, None) _ = value.pop(k, None)
if not value: if not value: