Sonoma map: Difference between revisions

From Wildsong
Jump to navigationJump to search
Brian Wilson (talk | contribs)
mNo edit summary
Brian Wilson (talk | contribs)
mNo edit summary
 
(One intermediate revision by the same user not shown)
Line 57: Line 57:


  sudo pip install simplejson
  sudo pip install simplejson
<pre>
python
from nominatim import ReverseGeocoder
client = ReverseGeocoder("http://nominatim.openstreetmap.org/reverse?format=json")
response = client.geocode(44.5454, -123.2626)
print response['full_address']
</pre>
Gag me! Does not give house number only street, useless! :-)
Trying this one
sudo pip install pygeocoder
Peets Coffee, the sidewalk outside anyway
N 38 26.459
W 122 42.737
which the Garmin reports as 276 D Street.
38.440983 -122.712283
from pygeocoder import Geocoder
results = Geocoder.reverse_geocode(38.440983,-122.71245)
print(results.formatted_address)

Latest revision as of 05:49, 27 May 2014

I am building a map to help me find a house to live in.

Address points

I have parcels but I don't have an official source for address points so I decided to create one.

I want to walk through the parcel layer, grab the centroid, and geocode it, and generate a new shapefile that contains the centroid with address data as attributes.

I can't use ArcMap right now because I am on vacation and have only this little MacBook. (Yes, this is what I do on vacation.)

I think the easiest way to do this is to use shapely to read and write the file and I have to throttle the geocode requests. I have to spread them out over a few days.

I ran across Fiona, which I wanted to try, so I had to install Homebrew. Installing Homebrew

ruby -e "$(curl -fsSL https://raw.github.com/Homebrew/homebrew/go/install)"

Installing gdal then becomes

brew install gdal

I still need pip though and brew recommends

sudo easy_install pip

Then I can do

pip install Fiona

which fails on the Mac so maybe I will not use Fiona at the moment. :-)

The steps

# Read shapefile and load into shapely geometries
# For each shape, 
## Reverse geocode the centroid
## Write geometry to shapefile

Step one is covered in Finding Intersections with Python but don't forget

import sys, os
from osgeo import ogr
import shapely.wkt

Once I do step one I have a list of features in both shapely and ogr format (in tuples). I store only the centroid, not the complete shapely polygon.

I wanted to use the geopy module but reverse geocoding is not supported in the trunk and the branch would not build on the Mac. So I am looking at Openstreetmap Nominatum as a possibility. Works for my old house anyway. REST example

http://nominatim.openstreetmap.org/reverse?format=json&lat=44.5454&lon=-123.2626&zoom=18&addressdetails=1&pretty=1

{"place_id":"86732458","licence":"Data \u00a9 OpenStreetMap contributors, ODbL 1.0. http:\/\/www.openstreetmap.org\/copyright","osm_type":"way","osm_id":"154822790","lat":"44.545242","lon":"-123.259303","display_name":"Southeast Alexander Avenue, Corvallis, Benton County, Oregon, 97333, United States of America","address":{"road":"Southeast Alexander Avenue","city":"Corvallis","county":"Benton County","state":"Oregon","postcode":"97333","country":"United States of America","country_code":"us"}}

sudo pip install simplejson
python
from nominatim import ReverseGeocoder
client = ReverseGeocoder("http://nominatim.openstreetmap.org/reverse?format=json")
response = client.geocode(44.5454, -123.2626)
print response['full_address']

Gag me! Does not give house number only street, useless! :-)

Trying this one

sudo pip install pygeocoder

Peets Coffee, the sidewalk outside anyway N 38 26.459 W 122 42.737 which the Garmin reports as 276 D Street.

38.440983 -122.712283

from pygeocoder import Geocoder results = Geocoder.reverse_geocode(38.440983,-122.71245) print(results.formatted_address)