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

[addresses] remove Quattroshapes/GeoNames cities as they may have problematic names, and in any case we have point-based cities from OSM now

Browse Source
This commit is contained in:
Al
2016-12-10 02:08:33 -05:00
parent 18c5fd0855
commit 5098599ed6
3 changed files with 8 additions and 106 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
scripts/geodata/openaddresses/openaddresses_training_data.py
View File

@@ -12,11 +12,10 @@ import os
from geodata.openaddresses.formatter import OpenAddressesFormatter
from geodata.addresses.components import AddressComponents
from geodata.geonames.db import GeoNamesDB
from geodata.polygons.language_polys import LanguagePolygonIndex
from geodata.neighborhoods.reverse_geocode import NeighborhoodReverseGeocoder
from geodata.places.reverse_geocode import PlaceReverseGeocoder
from geodata.polygons.reverse_geocode import OSMReverseGeocoder, OSMCountryReverseGeocoder, QuattroshapesReverseGeocoder
from geodata.polygons.reverse_geocode import OSMReverseGeocoder, OSMCountryReverseGeocoder
if __name__ == '__main__':
@@ -44,18 +43,10 @@ if __name__ == '__main__':
default=None,
help='OSM reverse geocoder RTree directory')
parser.add_argument('--quattroshapes-rtree-dir',
default=None,
help='Quattroshapes reverse geocoder RTree directory')
parser.add_argument('--places-index-dir',
default=None,
help='Places index directory')
parser.add_argument('--geonames-db',
default=None,
help='GeoNames db file')
parser.add_argument('--neighborhoods-rtree-dir',
default=None,
help='Neighborhoods reverse geocoder RTree directory')
@@ -85,17 +76,8 @@ if __name__ == '__main__':
if args.places_index_dir:
places_index = PlaceReverseGeocoder.load(args.places_index_dir)
quattroshapes_rtree = None
if args.quattroshapes_rtree_dir:
quattroshapes_rtree = QuattroshapesReverseGeocoder.load(args.quattroshapes_rtree_dir)
geonames = None
if args.geonames_db:
geonames = GeoNamesDB(args.geonames_db)
if args.openaddresses_dir and args.format:
components = AddressComponents(osm_rtree, neighborhoods_rtree, places_index, quattroshapes_rtree, geonames)
components = AddressComponents(osm_rtree, neighborhoods_rtree, places_index)
oa_formatter = OpenAddressesFormatter(components, country_rtree, debug=args.debug)
oa_formatter.build_training_data(args.openaddresses_dir, args.out_dir, tag_components=not args.untagged)