def parse_csv(csv_file, verbose):
""" Builds urls based on street address and city.
Args:
Takes a csv file passed in as cli argument.
Returns:
A list of properly formatted urls for querying
lat lng using Google Geocode.
"""
get_url = BuildURL("https://maps.googleapis.com/maps/api/geocode/json?address=", state="GA")
street = compile("\d+\s\w.*")
msg = "\nWould you like to use an api key this time? If not you will be prompted to add one as the query limit\n" \
"is reached. 2500 Queries will run without one just fine: y/n "
with open(csv_file, 'r') as open_csv:
csv_stream_object = reader(open_csv)
headers = next(csv_stream_object, None)
(city_header_index, tract_header_index) = scrape_headers(headers)
google_api_key = require_api(msg)
print("\nPlease wait, Generating coordinate urls...\n")
sleep(1)
address = {'street': None, 'city': None}
list_geocode_urls, list_of_cities = [], []
for row in csv_stream_object:
try:
list_of_cities.append(row[city_header_index])
# Build list of possible serving territories based on city
# from the index position of Tax City
except UnboundLocalError as err:
print(err)
# logger.warn("Error: {}, You must have at least 'Tax City', 'Census Tract'"
# " and 'Service Address' headers in your file ".format(err))
for field in row:
if search(street, field):
# find street in fields
address['city'] = row[row.index(field)]
for city in set(list_of_cities):
# Assign a city for the query from dynamic list
try:
if city in row:
address['street'] = row[row.index(city)]
except:
raise
list_geocode_urls.append(get_url.geo_url(address, verbose, api_key=google_api_key))
print("Done...\n")
return list_geocode_urls, tract_header_index
def main(csv, verbose=None):
print("Welcome to the 477 report generator")
print("""
TODO: (Does not effect current functionality.)
# Cut out tracts that don't start with 1304.
# Refactor file splitter code as large files are limited by 2500 queries a day or use multiple API keys
# Refactor elements when time permits, shift towards OOP where possible
""")
input("Use -h for more options. Press <Enter> to continue:")
# CREATE TIMESTAMP/LOG FILE NAME THAT'S OS SENSITIVE
date = CheckSystemDate()
log = date.system_date()
# TODO refactor file splitter code as large files are limited by 2500 queries a day or use multiple API keys
# final_csv = split_file(csv)
# PARSE CSV BUILD LOCATION URLS, POP OFF INDEX TO WRITE BACK TO
(geocode_urls, tract_header_index) = parse_csv(csv, verbose)
# GET LOCATION INFORMATION
get_location_date = HttpWorker(geocode_urls, msg="Getting accurate coordinates .. Please wait\n", endpoint="google")
location_data = get_location_date.get(verbose)
# BUILD ACCURATE FCC URL FROM LOCATION INFORMATION
create_fcc_urls = BuildURL("http://data.fcc.gov/api/block/find?format=json&censusYear=2010&")
fcc_urls = create_fcc_urls.fcc_url(location_data, verbose)
# DOCUMENT THIS
(clean_fcc_urls, fcc_error_indices) = filter_url_errors(fcc_urls)
# GET JSON DATA FOR EACH CUSTOMER
get_fcc_data = HttpWorker(clean_fcc_urls, msg="Getting accurate fips codes ...Please wait\n", endpoint="fcc")
tract_data = get_fcc_data.get(verbose)
# FIPS ARE THE 13 DIGIT CODES
fips = [code["Block"]["FIPS"] for code in tract_data]
for error in fcc_error_indices:
fips.insert(error, "unable to retrieve location or tact data")
(with_err, no_err) = write_to_csv(fips, csv, tract_header_index, log)
print("\nWrote updated csv to disk:\n",
with_err + "\n",
no_err)
