A Better Python GeoEncoding Script to Use with NYS SAM
I rewrote the short little geocoder that I posted last night in Python, using the multiple address function in NY’s Street and Address Maintenance (SAM) Program. This should be able to encode up to 1,000 lines in a text file. You would use this script by typing python sam-geocode.csv with one address per line.
By running python sam-geocode.py address.txt, with a file like this:
5 Winding Brook Drive Guilderland NY 12084
1224 Union Street Schenectady NY 12308
6239 Randomwood Drive Schenectady NY 12030
1529 Western Avenue Suite 102 Albany NY 12203
4017b State Street Schenectady NY 12304
4017b State Street Schenectady NY 12304
1529 Western Avenue Suite 102 Albany NY 12203
1529 Western Avenue Suite 102 Albany NY 12203
2842 W. Lydius Street Schenectady NY 12306
200 Bloomingdale Road Altamont NY 12009
Gives you out a CSV file named address-geocoded.csv.
"1224 Union St, Schenectady, NY, 12308",42.81103508600006,-73.92220575399995
"6239 Randomwood Dr, Schenectady, NY, 12303",42.74713565800005,-73.93299995099994
"1529 Western Ave, Albany, NY, 12203",42.68293244500006,-73.84222237599994
"4017 B State St, Schenectady, NY, 12304",42.76677680800003,-73.88794539599996
"4017 B State St, Schenectady, NY, 12304",42.76677680800003,-73.88794539599996
"1529 Western Ave, Albany, NY, 12203",42.68293244500006,-73.84222237599994
"1529 Western Ave, Albany, NY, 12203",42.68293244500006,-73.84222237599994
"2842 W Lydius St, Schenectady, NY, 12306",42.75003186400005,-73.95573429299998
"23 Dresden Ct, Albany, NY, 12203",42.69198706100008,-73.87999566599996
"629 Salvia Ln, Schenectady, NY, 12303",42.73869920900006,-73.93268336399996
"5 Winding Brook Dr, Guilderland, NY, 12084",42.69401880400005,-73.90855096599995
"3 Feeney Ln, Schenectady, NY, 12303",42.72653497400006,-73.90924777399994
"730 Sachem Cir, Slingerlands, NY, 12159",42.67422884300004,-73.88860130999996
"200 Bloomingdale Ln, Altamont, NY, 12009",42.69634668300006,-73.94903687499993
Now, I could improve error capture on this script but for my purposes this is very simple and works wells for well-formed addresses. Maybe not as simple as the PHP script I posted last night, but this way of Geocoding is much faster and written in a language that’s more appropriate for such purposes.
#!/usr/bin/python
import requests,sys,json,csv,os
query = '{"records": ['
f = open(sys.argv[-1], 'r+')
with open(sys.argv[-1]) as f:
i = 0
for line in f:
query += '{ "attributes": { "OBJECTID":'+str(i)+', "SINGLELINE": "'+line.rstrip()+'"} },'+"n"
i+=1
query += ']}'
post = { 'f':'pjson', 'outSR': 4326, 'addresses': query }
url = 'https://gisservices.its.ny.gov/arcgis/rest/services/Locators/Street_and_Address_Composite/GeocodeServer/geocodeAddresses'
req = requests.post(url, data = post)
addresses = json.loads(req.text)['locations']
csv = open(os.path.splitext(sys.argv[-1])[0]+'-geocode.csv', "w")
for place in addresses:
csv.write('"'+place['address'].replace('"', '\"')+'",'+str(place['location']['y'])+','+str(place['location']['x'])+"n")
csv.close()
This also works for most other states, as most states have geocoding endpoints, as otherwise the fire department wouldn’t necessarily be able to find your house when it’s on fire when you call 911. To find your state, try searching for StateName "GeocodeServer/geocodeAddresses". For example, replace URL with:
Massachusetts – https://gis.massdot.state.ma.us/arcgis/rest/services/General/MassDOTStreetAddressLocator/GeocodeServer/geocodeAddresses
Pennsylvania – https://maps.pasda.psu.edu/arcgis/rest/services/apps/AddressLocator/GeocodeServer/geocodeAddresses