# The Google Maps library
import googlemaps
# Date time for easy computations between dates
from datetime import datetime
# JSON handling
import json
# Pandas
import pandas as pd
# Regular expressions
import re
# TQDM for fancy loading bars
from tqdm import tqdm
import time
import random
# !!! Define the main access point to the Google APIs.
# !!! This object will contain all the functions needed
geolocator = googlemaps.Client(key="<YOUR API KEY>")
WORK_LAT_LNG = (<LATITUDE>, <LONGITUDE>)
# You can set this parameter to decide the time from which
# Google needs to calculate the directions
# Different times affect public transport
DEPARTURE_TIME = datetime.now
# Load the source data
data = pd.read_csv("/path/to/raw/data/data.csv")
# Define the columns that we want in the geocoded dataframe
geo_columns = [
               "_link",
               "lat",
               "lng",
               "_time_to_work_seconds_transit",
               "_time_to_work_seconds_walking"
               ]
# Create an array where we'll store the geocoded data
geo_data = []
# For each element of the raw dataframe, start the geocoding
for index,
in tqdm(data.iterrows()):
    # Google Geo coding
    _location = ""
    _location_json = ""
    try:
        # Try to retrieve the base location,
        # i.e. the Latitude and Longitude given the address
        _location = geolocator.geocode(row._address)
        _location_json = json.dumps(_location[0])
    except:
        pass
    _time_to_work_seconds_transit = 0
    _directions_json = ""
    _lat_lon = {"lat": 0, "lng": 0}
    try:
        # Given the work latitude and longitude, plus the property latitude and longitude,
        # retrieve the distance with PUBLIC TRANSPORT (`mode=transit`)
        _lat_lon = _location[0]["geometry"]["location"]
        _directions = geolocator.directions(WORK_LAT_LNG,
              (_lat_lon["lat"], _lat_lon["lng"]), mode="transit")
        _time_to_work_seconds_transit = _directions[0]["legs"][0]["duration"]["value"]
        _directions_json = json.dumps(_directions[0])
    except:
        pass
    _time_to_work_seconds_walking = 0
    try:
        # Given the work latitude and longitude, plus the property latitude and longitude,
        # retrieve the WALKING distance (`mode=walking`)
        _lat_lon = _location[0]["geometry"]["location"]
        _directions = geolocator.directions(WORK_LAT_LNG, (_lat_lon["lat"], _lat_lon["lng"]), mode="walking")
        _time_to_work_seconds_walking = _directions[0]["legs"][0]["duration"]["value"]
    except:
        pass
    #  This block retrieves the number of SUPERMARKETS arount the property
    '''
    _supermarket_nr = 0
    _supermarket = ""
    try:
        # _supermarket = geolocator.places_nearby((_lat_lon["lat"],_lat_lon["lng"]), radius=750, type="supermarket")
        _supermarket_nr = len(_supermarket["results"])
    except:
        pass
    '''
     #  This block retrieves the number of PHARMACIES arount the property
    '''
    _pharmacy_nr = 0
    _pharmacy = ""
    try:
        # _pharmacy = geolocator.places_nearby((_lat_lon["lat"],_lat_lon["lng"]), radius=750, type="pharmacy")
        _pharmacy_nr = len(_pharmacy["results"])
    except:
        pass
    '''
    #  This block retrieves the number of RESTAURANTS arount the property
    '''
    _restaurant_nr = 0
    _restaurant = ""
    try:
        # _restaurant = geolocator.places_nearby((_lat_lon["lat"],_lat_lon["lng"]), radius=750, type="restaurant")
        _restaurant_nr = len(_restaurant["results"])
    except:
        pass
    '''
    geo_data.append([row._link, _lat_lon["lat"], _lat_lon["lng"], _time_to_work_seconds_transit,
                     _time_to_work_seconds_walking])
    geo_data_df = pd.DataFrame(geo_data)
    geo_data_df.columns = geo_columns
    geo_data_df.to_csv("geo_data_houses.csv", index=False)

import pandas as pd
from sklearn.cluster import DBSCAN
data = pd.read_csv("data.csv")
data["labels"] = DBSCAN(eps=0.01, min_samples=3).fit(data[["lat","lng"]].values).labels_
print(data["labels"].unique())
data.to_csv("out.csv")