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

[osm] checking for valid street names in OSM street-only training data so e.g. the street name is not just a simple number like "831"

Browse Source
This commit is contained in:
Al
2017-01-19 02:34:29 -05:00
parent 7c04627f60
commit a931c5ddc9
2 changed files with 5 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
scripts/geodata/addresses/components.py
View File

@@ -1502,9 +1502,12 @@ class AddressComponents(object):
invalid_street_regex = re.compile('^\s*(?:none|null|not applicable|n\s*/\s*a)\s*$', re.I)
def street_name_is_valid(self, street):
return street is not None and not (self.invalid_street_regex.match(street) or not any((c.isalnum() for c in street)))
def cleanup_street(self, address_components):
street = address_components.get(AddressFormatter.ROAD)
if street is not None and (self.invalid_street_regex.match(street) or not any((c.isalnum() for c in street))):
if not self.street_name_is_valid(street):
address_components.pop(AddressFormatter.ROAD)
newline_regex = re.compile('[\n]+')