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

[fix] putting the neighborhoods check after the dupe threshold check, as it's not really needed until then anyway

Browse Source
This commit is contained in:
Al
2016-12-18 03:00:40 -05:00
parent 4314a6822d
commit e5657c5612
1 changed files with 22 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
scripts/geodata/neighborhoods/reverse_geocode.py
View File

@@ -366,35 +366,32 @@ class NeighborhoodReverseGeocoder(RTreePolygonIndex):
if arg_max: if arg_max:
ranks.append(arg_max) ranks.append(arg_max)
if not ranks:
continue
existing_osm_boundaries = osm_admin_rtree.point_in_poly(lat, lon, return_all=True)
skip_node = False
for i, props in enumerate(existing_osm_boundaries):
containing_component = None
name = props.get('name')
# Only exact name matches here since we're comparins OSM to OSM
if name and name.lower() != attrs.get('name', ''.lower()):
continue
containing_ids = [(boundary['type'], boundary['id']) for boundary in existing_osm_boundaries[i + 1:]]
containing_component = osm_address_components.component_from_properties(country, props, containing=containing_ids)
if containing_component and containing_component != component_name and AddressFormatter.component_order[containing_component] <= AddressFormatter.component_order[AddressFormatter.CITY]:
skip_node = True
break
# Skip this element
if skip_node:
continue
ranks.sort(key=operator.itemgetter(0), reverse=True) ranks.sort(key=operator.itemgetter(0), reverse=True)
if ranks and ranks[0][0] >= cls.DUPE_THRESHOLD: if ranks and ranks[0][0] >= cls.DUPE_THRESHOLD:
score, props, poly, idx, i = ranks[0] score, props, poly, idx, i = ranks[0]
existing_osm_boundaries = osm_admin_rtree.point_in_poly(lat, lon, return_all=True)
skip_node = False
for i, props in enumerate(existing_osm_boundaries):
containing_component = None
name = props.get('name')
# Only exact name matches here since we're comparins OSM to OSM
if name and name.lower() != attrs.get('name', ''.lower()):
continue
containing_ids = [(boundary['type'], boundary['id']) for boundary in existing_osm_boundaries[i + 1:]]
containing_component = osm_address_components.component_from_properties(country, props, containing=containing_ids)
if containing_component and containing_component != component_name and AddressFormatter.component_order[containing_component] <= AddressFormatter.component_order[AddressFormatter.CITY]:
skip_node = True
break
# Skip this element
if skip_node:
continue
containing_ids = [(boundary['type'], boundary['id']) for boundary in existing_osm_boundaries] containing_ids = [(boundary['type'], boundary['id']) for boundary in existing_osm_boundaries]
component = osm_address_components.component_from_properties(country, props, containing=containing_ids) component = osm_address_components.component_from_properties(country, props, containing=containing_ids)
attrs['component'] = component attrs['component'] = component