chase-complete.py

    # import libraries
    from selenium import webdriver
    from selenium.webdriver.common.keys import Keys
    import time
    import pandas as pd

    # retrieve all NY branch/atm location names
    state_link = 'https://locator.chase.com/ut'
    driver = webdriver.Chrome(executable_path="/Users/erikgregorywebb/Downloads/chromedriver 2")
    driver.get(state_link)

    items = driver.find_elements_by_class_name('child')

    locations = []
    for i in range(0, len(items)):
        locations.append(items[i].text)
    driver.close()

    # convert branch/atm location names to urls
    for i in range(0, len(locations)):
        locations[i] = locations[i].lower()
        if ' ' in locations[i]:
            locations[i] = locations[i].replace(" ", "+")

    location_links = []
    for i in range(0, len(locations)):
        link = state_link + "/" + locations[i]
        location_links.append(link)

    # function to extract branch/atm location details
    def getLocations(link):    
        driver = webdriver.Chrome(executable_path="/Users/erikgregorywebb/Downloads/chromedriver 2")
        driver.get(link)

        items = driver.find_elements_by_class_name('details')
        locations = []
        for i in range(0, len(items)):
            locations.append(items[i].text)

        driver.close()
        return locations

    # execute the function for all branch/atms links
    all_locations = []
    for link in location_links:
        time.sleep(3)
        try: locations = getLocations(link)
        except: print(link)
        all_locations.append(locations)    

    # clean the organize the data
    final = pd.DataFrame()
    for location in all_locations:
        for i in range(0, len(location)):
            location[i] = location[i].splitlines()
        data = pd.DataFrame(location)

        data = data[[0, 2, 3, 6]]
        data = data.dropna()
        data.columns = ['city', 'address1', 'address2', 'type']
        data['address'] = data['address1'] + " " + data['address2']
        data = data.drop(['address1', 'address2'], axis=1)
        data['type'] = data.type.str.split().str.get(0)
        data = data[['city', 'address', 'type']] # rearrange column names
        final = pd.concat([final, data])

    # remove duplictates
    ut_locations = final.drop_duplicates(subset = ['city', 'address', 'type'], keep = 'first')

    # export the data
    ut_locations.to_csv("ut-locations.csv", index = False)    
    
    # read in branch/atms locations
    branches = pd.read_csv("/Users/erikgregorywebb/Documents/Python/chase-atm/utah-locations.csv")
    addresses = branches['address']

    # prepare the API Call
    gmaps = googlemaps.Client(key='YOUR-KEY-HERE')

    # geocode addresses to obtain geographic coordinates
    lats = []
    lngs = []

    for i in range(0, len(addresses)):
        time.sleep(1)
        geo = gmaps.geocode(addresses[i])
        lat = geo[0]['geometry']['location']['lat']
        lng = geo[0]['geometry']['location']['lng']
        lats.append(lat)
        lngs.append(lng)

    # format and export results
    data = pd.DataFrame({'address': addresses, 'lat': lats, 'lng': lngs})
    final = pd.merge(branches, data, on='address', how='outer')    
    final.to_csv("utah-chase-branch-google-maps.csv", index = False)