Google_Geocoding.py

import pandas as pd
import httplib2
from lxml import etree
import json
from io import StringIO
import re
import unicodedata
import time


def address_get(place_name, ward):
    request = "http://maps.google.com/maps/api/geocode/json?address=" + ward + "%20" + place_name + "&sensor=false&language=ja"

    try:
        h = httplib2.Http('.cache') #by httplib2
        response, content = h.request(request)
        content = content.decode("utf-8") 
        jdata = json.load(StringIO(content)) 
        results = re.search(r"日本,\s〒(?P<PostalCode>\d\d\d-\d\d\d\d)\s(?P<Address>.+?)\s(?P<Name>.+?)", jdata["results"][0]['formatted_address'])
        PostalCode = results.group("PostalCode")
        Address = unicodedata.normalize('NFKC', results.group("Address")) # 全角数字は半角にしておく
        lat = jdata["results"][0]['geometry']["location"]["lat"]
        lng = jdata["results"][0]['geometry']["location"]["lng"]
    except:
        PostalCode = "Error"
        Address = "Error"
        lat = "Error"
        lng = "Error"

    time.sleep(3.5) #連続での取得を防ぐ
    print (PostalCode, Address, lat, lng)
    
    return (PostalCode, Address, lat, lng)


if __name__ == "__main__":
    data = pd.read_csv("yokohama_PS.csv")
    
    for i in range(len(data)):
        place_name = data.ix[i, "住所"]
        ward = "横浜市港北区" 
        
        PostalCode, Address, lat, lng = address_get(place_name, ward)
        
        data.ix[i, "郵便番号"] = PostalCode
        data.ix[i, "住所"] = Address
        data.ix[i, "Google緯度"] = lat
        data.ix[i, "Google経度"] = lng
    
    data.to_csv( 'yokohama_PS_G.csv', index=False )