generate random address

random_address.py

import os
import io
import zipfile
import requests
import pandas as pd
import pdb

def download_addresses(path='lane.csv'):
    print('Downloading addresses')
    lane_addresses_zip = requests.get(
        'https://data.openaddresses.io/runs/864834/us/or/lane.zip'
        )

    # decompress
    z = zipfile.ZipFile(
            io.BytesIO(lane_addresses_zip.content)
        )

    with z.open('us/or/lane.csv', 'r') as lane_f:
        lane = lane_f.read()

    # convert to df
    df = pd.read_csv(io.BytesIO(lane))

    abs_path = os.path.abspath(path)
    df.to_csv(abs_path, index=False)

    print(f"Saved addresses to:\n    {os.path.abspath(path)}")

    return df

def pick_address(addresses):
    # filter to eugene
    addresses = addresses.loc[addresses.CITY == "Eugene",:]

    # filter to minimally complete addresses
    addresses = addresses[addresses.NUMBER.notnull()]
    addresses = addresses[addresses.STREET.notnull()]
    addresses = addresses[addresses.POSTCODE.notnull()]

    # pick one
    address = addresses.sample(1)

    # format
    #pdb.set_trace()
    try:
        if address.UNIT.isna().all():
            address_str = " ".join([
                address.NUMBER.values[0],
                address.STREET.values[0],
                ]) + f', {address.POSTCODE.values[0]}'
        else:
            address_str = " ".join([
                address.NUMBER.values[0],
                address.STREET.values[0]
                ])
            address_str = ", ".join([
                address_str,
                address.UNIT.values[0],
                str(address.POSTCODE.values[0])])
    except Exception as e:
        pdb.set_trace()

    print(address_str)
    return address_str


if __name__ == "__main__":
    if os.path.exists('lane.csv'):
        addresses = pd.read_csv('lane.csv')
    else:
        addresses = download_addresses()

    address = pick_address(addresses)