def genderize(args):
print(args)
#File initialization
dir_path = os.path.dirname(os.path.realpath(__file__))
logging.basicConfig(
filename=dir_path + os.sep + "log.txt",
level=logging.DEBUG,
format='%(asctime)s %(levelname)s %(name)s %(message)s')
logger = logging.getLogger(__name__)
ofilename, ofile_extension = os.path.splitext(args.output)
ofile = ofilename + "_" + time.strftime("%Y%m%d-%H%M%S") + ".csv"
ifile = args.input
if os.path.isabs(ifile):
print("\n--- Input file: " + ifile)
else:
print("\n--- Input file: " + dir_path + os.sep + ifile)
if os.path.isabs(ofile):
print("--- Output file: " + ofile)
else:
print("--- Output file: " + dir_path + os.sep + ofile + "\n")
#File integrity checking
if not os.path.exists(ifile):
print("--- Input file does not exist. Exiting.\n")
sys.exit()
if not os.path.exists(os.path.dirname(ofile)):
print("--- Error! Invalid output file path. Exiting.\n")
sys.exit()
#Some set up stuff
##csv.field_size_limit(sys.maxsize)
#Initialize API key
if not args.key == "NO_API":
print("--- API key: " + args.key + "\n")
genderize = Genderize(user_agent='GenderizeDocs/0.0', api_key=args.key)
key_present = True
else:
print("--- No API key provided.\n")
key_present = False
#Open ifile
with open(ifile, 'r', encoding="utf8") as csvfile:
readCSV = csv.reader(csvfile, delimiter=',', skipinitialspace=True)
headers = next(readCSV) # take the first row
# [Chen]: use re.search to find the specific column name, match string "name"
specific_header_index = 0
for header in headers:
if re.search("name", header, re.I) != None:
break
specific_header_index += 1
# [Chen]: store rows without headers as a list
rest = list(readCSV)
# [Chen]: extract the first_name column
first_name = [row[specific_header_index].strip() for row in rest]
o_first_name = list()
for l in first_name:
for b in l:
o_first_name.append(b)
if args.auto == True:
uniq_first_name = list(set(o_first_name))
chunks = list(jpyh.splitlist(uniq_first_name, 10))
print("--- Read CSV with " + str(len(first_name)) +
" first_name. " + str(len(uniq_first_name)) + " unique.")
else:
chunks = list(jpyh.splitlist(first_name, 10))
print("--- Read CSV with " + str(len(first_name)) + " first_name")
print("--- Processed into " + str(len(chunks)) + " chunks")
if jpyh.query_yes_no(
"\n---! Ready to send to Genderdize. Proceed?") == False:
print("Exiting...\n")
sys.exit()
if os.path.isfile(ofile):
if jpyh.query_yes_no(
"---! Output file exists, overwrite?") == False:
print("Exiting...\n")
sys.exit()
print("\n")
if args.auto == True:
ofile = ofile + ".tmp"
response_time = []
gender_responses = list()
with open(ofile, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
# [Chen]: generate different headers by'-ORV' argument
if args.override == True:
headers.extend(["female", "male"])
else:
headers.extend(["gender", "probability", "count"])
writer.writerow(headers)
chunks_len = len(chunks)
stopped = False
for index, chunk in enumerate(chunks):
if stopped:
break
success = False
while not success:
try:
start = time.time()
if key_present:
dataset = genderize.get(chunk)
else:
dataset = Genderize().get(chunk)
gender_responses.append(dataset)
success = True
except GenderizeException as e:
#print("\n" + str(e))
logger.error(e)
#Error handling
if "response not in JSON format" in str(
e) and args.catch == True:
if jpyh.query_yes_no(
"\n---!! 502 detected, try again?"
) == True:
success = False
continue
elif "Invalid API key" in str(
e) and args.catch == True:
print(
"\n---!! Error, invalid API key! Check log file for details.\n"
)
else:
print(
"\n---!! GenderizeException - You probably exceeded the request limit, please add or purchase a API key. Check log file for details.\n"
)
stopped = True
break
response_time.append(time.time() - start)
print("Processed chunk " + str(index + 1) + " of " + str(chunks_len) + " -- Time remaining (est.): " + \
str( round( (sum(response_time) / len(response_time) * (chunks_len - index - 1)), 3)) + "s")
# [Chen]: Combine original data and the response: if go with '-OVR', convert gender to 0/1; if go without '-OVR', pick up gender, probability, count
if args.override == True:
for data in dataset:
append_response = []
if data.get('gender'): # null check
female = 1 if data.get(
'gender') == 'female' else 0
append_response = [female, female ^ 1
] # xor for 0 <=> 1
else:
append_response = [0, 0]
writer.writerow([*rest[0], *append_response])
rest.pop(0)
else:
for data in dataset:
writer.writerow(
[*rest[0], *list(data.values())[1:]])
rest.pop(0)
break
if args.auto == True:
print("\nCompleting identical first_name...\n")
#AUTOCOMPLETE first_name
#Create master dict
gender_dict = dict()
print(gender_dict)
for response in gender_responses:
for d in response:
gender_dict[d.get("name")] = [
d.get("gender"),
d.get("probability"),
d.get("count")
]
filename, file_extension = os.path.splitext(ofile)
with open(filename, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
writer.writerow(
list(["first_name", "gender", "probability", "count"]))
for name in o_first_name:
data = gender_dict.get(name)
writer.writerow([name, data[0], data[1], data[2]])
print("Done!\n")
def genderize(args):
print(args)
#File initialization
dir_path = os.path.dirname(os.path.realpath(__file__))
logging.basicConfig(
filename=dir_path + os.sep + "log.txt",
level=logging.DEBUG,
format='%(asctime)s %(levelname)s %(name)s %(message)s')
logger = logging.getLogger(__name__)
ofilename, ofile_extension = os.path.splitext(args.output)
ofile = ofilename + "_" + time.strftime("%Y%m%d-%H%M%S") + ".csv"
ifile = args.input
if os.path.isabs(ifile):
print("\n--- Input file: " + ifile)
else:
print("\n--- Input file: " + dir_path + os.sep + ifile)
if os.path.isabs(ofile):
print("--- Output file: " + ofile)
else:
print("--- Output file: " + dir_path + os.sep + ofile + "\n")
#File integrity checking
if not os.path.exists(ifile):
print("--- Input file does not exist. Exiting.\n")
sys.exit()
if not os.path.exists(os.path.dirname(ofile)):
print("--- Error! Invalid output file path. Exiting.\n")
sys.exit()
#Some set up stuff
csv.field_size_limit(sys.maxsize)
#Initialize API key
if not args.key == "NO_API":
print("--- API key: " + args.key + "\n")
genderize = Genderize(user_agent='GenderizeDocs/0.0',
api_key='169f6a8e933dcec15a57235d7fde49d6')
key_present = True
else:
print("--- No API key provided.\n")
key_present = False
#Open ifile
with open(ifile, 'r', encoding="utf8") as csvfile:
readCSV = csv.reader(csvfile,
delimiter=',',
quotechar='"',
skipinitialspace=True)
line_count = 0
first_name = []
users_id = []
input_fields = []
rows = []
input_fields = next(readCSV)
for row in readCSV: #Read CSV into first_name list
rows.append(row)
line_count += 1
if len(input_fields) > 1:
for row in rows:
first_name.append(row[1])
users_id.append(row[0])
else:
for row in rows:
first_name.append(row[0])
if args.noheader == False and first_name[0] in input_fields:
first_name.pop(0) #Remove header
o_first_name = list()
for l in first_name:
for b in l:
o_first_name.append(b)
if args.auto == True:
# uniq_first_name = list(set(o_first_name))
chunks = list(jpyh.splitlist(first_name, 10))
print("--- Read CSV with " + str(len(first_name)) +
" first_name. " + str(len(first_name)) + " unique.")
else:
chunks = list(jpyh.splitlist(first_name, 10))
print("--- Read CSV with " + str(len(first_name)) + " first_name")
print("--- Processed into " + str(len(chunks)) + " chunks")
if jpyh.query_yes_no(
"\n---! Ready to send to Genderdize. Proceed?") == False:
print("Exiting...\n")
sys.exit()
if os.path.isfile(ofile):
if jpyh.query_yes_no(
"---! Output file exists, overwrite?") == False:
print("Exiting...\n")
sys.exit()
print("\n")
if args.auto == True:
ofile = ofile + ".tmp"
response_time = []
gender_responses = list()
output_fields = [*input_fields, "gender", "probability", "count"]
if args.override == False:
with open(ofile, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
writer.writerow(output_fields)
chunks_len = len(chunks)
stopped = False
for index, chunk in enumerate(chunks):
if stopped:
break
success = False
while not success:
try:
start = time.time()
if key_present:
dataset = genderize.get(chunk)
else:
dataset = Genderize(
user_agent='GenderizeDocs/0.0',
api_key='169f6a8e933dcec15a57235d7fde49d6'
).get(chunk)
gender_responses.append(dataset)
success = True
except GenderizeException as e:
print("\n" + str(e))
logger.error(e)
#Error handling
if "response not in JSON format" in str(
e) and args.catch == True:
if jpyh.query_yes_no(
"\n---!! 502 detected, try again?"
) == True:
success = False
continue
elif "Invalid API key" in str(
e) and args.catch == True:
print(
"\n---!! Error, invalid API key! Check log file for details.\n"
)
else:
print(
"\n---!! GenderizeException - You probably exceeded the request limit, please add or purchase a API key. Check log file for details.\n"
)
stopped = True
break
response_time.append(time.time() - start)
print("Processed chunk " + str(index + 1) + " of " + str(chunks_len) + " -- Time remaining (est.): " + \
str( round( (sum(response_time) / len(response_time) * (chunks_len - index - 1)), 3)) + "s")
break
j = 0
response_arr = list()
for data in gender_responses:
for el in data:
response_arr.append(el)
for row in rows:
row.append(response_arr[j]['gender'])
row.append(response_arr[j]['probability'])
row.append(response_arr[j]['count'])
j += 1
for row in rows:
writer.writerow(row)
if args.auto == True:
print("\nCompleting identical first_name...\n")
#AUTOCOMPLETE first_namey
#Create master dict
gender_dict = dict()
for response in gender_responses:
for d in response:
gender_dict[d.get("name")] = [
d.get("count"),
d.get("gender"),
d.get("probability")
]
filename, file_extension = os.path.splitext(ofile)
with open(filename, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f, delimiter='-')
writer.writerow(
list(["name", "count", "gender", "probability"]))
for name in o_first_name:
data = gender_dict.get(name)
writer.writerow([name, data[1], data[1], data[2]])
if args.override == True:
output_fields = [*input_fields]
with open(ofile, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
writer.writerow(output_fields)
chunks_len = len(chunks)
stopped = False
for index, chunk in enumerate(chunks):
if stopped:
break
success = False
while not success:
try:
start = time.time()
if key_present:
dataset = genderize.get(chunk)
else:
dataset = Genderize(
user_agent='GenderizeDocs/0.0',
api_key='169f6a8e933dcec15a57235d7fde49d6'
).get(chunk)
gender_responses.append(dataset)
success = True
except GenderizeException as e:
print("\n" + str(e))
logger.error(e)
#Error handling
if "response not in JSON format" in str(
e) and args.catch == True:
if jpyh.query_yes_no(
"\n---!! 502 detected, try again?"
) == True:
success = False
continue
elif "Invalid API key" in str(
e) and args.catch == True:
print(
"\n---!! Error, invalid API key! Check log file for details.\n"
)
else:
print(
"\n---!! GenderizeException - You probably exceeded the request limit, please add or purchase a API key. Check log file for details.\n"
)
stopped = True
break
response_time.append(time.time() - start)
print("Processed chunk " + str(index + 1) + " of " + str(chunks_len) + " -- Time remaining (est.): " + \
str( round( (sum(response_time) / len(response_time) * (chunks_len - index - 1)), 3)) + "s")
break
i = 0
response_arr = list()
for data in gender_responses:
for el in data:
response_arr.append(el)
index = len(rows)
for row in rows:
# print(rows[-1][-2])
if response_arr[i]['gender'] == 'male':
rows[i][-1] = "1"
rows[i][-2] = "0"
else:
rows[i][-1] = "0"
rows[i][-2] = "1"
i += 1
for row in rows:
writer.writerow(row)
print("Done!\n")
def genderize(args):
print(args)
#File initialization
dir_path = os.path.dirname(os.path.realpath(__file__))
logging.basicConfig(filename=dir_path + os.sep + "log.txt", level=logging.DEBUG,
format='%(asctime)s %(levelname)s %(name)s %(message)s')
logger=logging.getLogger(__name__)
ofilename, ofile_extension = os.path.splitext(args.output)
#ofile = ofilename + "_" + time.strftime("%Y%m%d-%H%M%S") + ".csv"
ofile = ofilename + ofile_extension
ifile = args.input
if os.path.isabs(ifile):
print("\n--- Input file: " + ifile)
else:
print("\n--- Input file: " + dir_path + os.sep + ifile)
if os.path.isabs(ofile):
print("--- Output file: " + ofile)
else:
print("--- Output file: " + dir_path + os.sep + ofile + "\n")
#File integrity checking
if not os.path.exists(ifile):
print("--- Input file does not exist. Exiting.\n")
sys.exit()
#if not os.path.exists(os.path.dirname(ofile)):
if not os.path.exists(ofile):
print("--- Error! Invalid output file path. Exiting.\n")
sys.exit()
#Some set up stuff
##csv.field_size_limit(sys.maxsize)
#Initialize API key
if not args.key == "NO_API":
print("--- API key: " + args.key + "\n")
genderize = Genderize(
user_agent='GenderizeDocs/0.0',
api_key=args.key)
key_present = True
else:
print("--- No API key provided.\n")
key_present = False
#Open ifile
with open(ifile, 'r', encoding="utf8") as csvfile:
readCSV = csv.reader(csvfile, delimiter=',', skipinitialspace=True)
first_name = []
inputData = []
for row in readCSV: #Read CSV into first_name list
if len(row) > 1:
row[1] = row[1].strip()
first_name.append(row[1])
else:
row[0] = row[0].strip()
first_name.append(row[0])
inputData.append(row)
headers = inputData[0]
inputData = inputData[1:]
data_iterator = iter(inputData)
if args.noheader == False:
first_name.pop(0) #Remove header
o_first_name = list()
for l in first_name:
for b in l:
o_first_name.append(b)
if args.auto == True:
uniq_first_name = list(set(first_name))
chunks = list(jpyh.splitlist(uniq_first_name, 10));
print("--- Read CSV with " + str(len(first_name)) + " first_name. " + str(len(uniq_first_name)) + " unique.")
else:
chunks = list(jpyh.splitlist(first_name, 10));
print("--- Read CSV with " + str(len(first_name)) + " first_name")
print("--- Processed into " + str(len(chunks)) + " chunks")
if jpyh.query_yes_no("\n---! Ready to send to Genderdize. Proceed?") == False:
print("Exiting...\n")
sys.exit()
if os.path.isfile(ofile):
if jpyh.query_yes_no("---! Output file exists, overwrite?") == False:
print("Exiting...\n")
sys.exit()
print("\n")
if args.auto == True:
ofile = ofile + ".tmp"
if "gender" not in headers:
headers.append("gender")
gender_index = headers.index("gender")
if "probability" not in headers:
headers.append("probability")
prob_index = headers.index("probability")
if "count" not in headers:
headers.append("count")
count_index = headers.index("count")
response_time = [];
gender_responses = list()
with open(ofile, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
#writer.writerow(list(["first_name", "gender", "probability", "count"]))
chunks_len = len(chunks)
stopped = False
#print(inputData)
writer.writerow(headers)
for index, chunk in enumerate(chunks):
if stopped:
break
success = False
while not success:
try:
start = time.time()
if key_present:
dataset = genderize.get(chunk)
else:
dataset = Genderize().get(chunk)
gender_responses.append(dataset)
success = True
except GenderizeException as e:
#print("\n" + str(e))
logger.error(e)
#Error handling
if "response not in JSON format" in str(e) and args.catch == True:
if jpyh.query_yes_no("\n---!! 502 detected, try again?") == True:
success = False
continue
elif "Invalid API key" in str(e) and args.catch == True:
print("\n---!! Error, invalid API key! Check log file for details.\n")
else:
print("\n---!! GenderizeException - You probably exceeded the request limit, please add or purchase a API key. Check log file for details.\n")
stopped = True
break
response_time.append(time.time() - start)
print("Processed chunk " + str(index + 1) + " of " + str(chunks_len) + " -- Time remaining (est.): " + \
str( round( (sum(response_time) / len(response_time) * (chunks_len - index - 1)), 3)) + "s")
#print(dataset)
for data in dataset:
next_row = next(data_iterator)
next_row_index = inputData.index(next_row)
entry_row = inputData[next_row_index]
entry = [data["gender"], data["probability"], data["count"]]
entry_row += entry
writer.writerow(entry_row)
break
if args.auto == True:
print("\nCompleting identical first_name...\n")
#AUTOCOMPLETE first_name
#Create master dict
gender_dict = dict()
for response in gender_responses:
for d in response:
gender_dict[d.get("name")] = [d.get("gender"), d.get("probability"), d.get("count")]
#names seen
seen_names = []
with open(ofilename + "_auto" + ofile_extension, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
writer.writerow(headers)
for row in inputData:
if len(row) > 4:
name = row[1]
if name not in seen_names:
writer.writerow(row)
seen_names.append(name)
else:
name = row[0]
if name not in seen_names:
writer.writerow(row)
seen_names.append(name)
if args.override == True:
print("\nExercising override \n")
with open(ofilename + "_override" + ofile_extension, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
#add headers
headers += ["female", "male"]
writer.writerow(headers)
for row in inputData:
gender = row[gender_index]
if gender == "male":
row += [0, 1]
writer.writerow(row)
else:
row += [1, 0]
writer.writerow(row)
master_dict = dict()
for response in gender_responses:
for line in response:
master_dict[line.get("name")] = [line.get("gender"), line.get("probability"), line.get("count")]
userResponse = True
while userResponse == True:
if jpyh.query_yes_no("Would you like to search the output file for the gender of a name") == False:
userResponse = False
sys.exit()
sys.stdout.write("Enter a name you would like to search. \n")
user_input = input()
if user_input == '':
sys.stdout.write("You MUST enter a name \n")
elif user_input not in master_dict:
sys.stdout.write("The name you entered is not in the output file \n")
if jpyh.query_yes_no("Would you like to genderize {}?".format(user_input)) == False:
userResponse = False
else:
if key_present:
api_response = genderize.get([user_input])
else:
api_response = Genderize().get([user_input])
for line in api_response:
name, gender, prob, count = line.get("name"), line.get("gender"), line.get("probability"), line.get("count")
print("Name: {}, Gender: {}, Probability: {}, Count: {}".format(name, gender, prob, count))
else:
response = master_dict[user_input]
sys.stdout.write("Name: {}, Gender: {}, Probability: {}, Count: {}".format(user_input, response[0], response[1], response[2]))
if jpyh.query_yes_no("\nWould you like to search for another name?") == False:
userResponse = False
print("Done!\n")
def genderize(args):
print(args)
#File initialization
dir_path = os.path.dirname(os.path.realpath(__file__))
logging.basicConfig(filename=dir_path + os.sep + "log.txt", level=logging.DEBUG,
format='%(asctime)s %(levelname)s %(name)s %(message)s')
logger=logging.getLogger(__name__)
ofilename, ofile_extension = os.path.splitext(args.output)
ofile = ofilename + "_" + time.strftime("%Y%m%d-%H%M%S") + ".csv"
ifile = args.input
if os.path.isabs(ifile):
print("\n--- Input file: " + ifile)
else:
print("\n--- Input file: " + dir_path + os.sep + ifile)
if os.path.isabs(ofile):
print("--- Output file: " + ofile)
else:
print("--- Output file: " + dir_path + os.sep + ofile + "\n")
#File integrity checking
if not os.path.exists(ifile):
print("--- Input file does not exist. Exiting.\n")
sys.exit()
if not os.path.exists(os.path.dirname(ofile)):
print("--- Error! Invalid output file path. Exiting.\n")
sys.exit()
#Some set up stuff
##csv.field_size_limit(sys.maxsize)
#Initialize API key
if not args.key == "NO_API":
print("--- API key: " + args.key + "\n")
genderize = Genderize(
user_agent='GenderizeDocs/0.0',
api_key=args.key)
key_present = True
else:
print("--- No API key provided.\n")
key_present = False
# Modify this section to take into account what the user wants to use through the command line
#Open ifile
with open(ifile, 'r', encoding="utf8") as csvfile:
readCSV = csv.reader(csvfile, delimiter=',', skipinitialspace=True)
first_name = []
raw = []
original_headers = [];
is_override = False
column_number = -1
# we are attempting to override the column that the names are stored in
# Easier to check a boolean than to constantly check if args.override is equal to 'NO_OVERRIDE'
if args.override != 'NO_OVERRIDE':
is_override = True
for row in readCSV: #Read CSV into first_name list
# if we are overriding the search column
if is_override:
# ugly nested mess but it works.
# if we have not found the list position of the desired override column
if column_number == -1:
# get the first row from the reader (assumed to be the first row)
first_name.append(row)
# also save this to the raw list for later use
raw.append(row)
# iterate through each item in the row we just saved and keep track of the for loop index
for index, column in enumerate(first_name[0]):
# if our column name is equal to the override name, we found the index number we need to proceed. Break from the loop
if column == args.override:
column_number = index
break
# error detection if the user override is not found in the header of the input csv.
if index == len(first_name[0])-1:
print("User Override '" + args.override + "' not found in input CSV file, Exiting...")
sys.exit()
# Our column number should be found by now, so continue to import the specific data that we want.
else:
# IMPORTANT: we need to remove all leading and trailing whitespaces to ensure that the genderizer responds with correct information
stripped = row[column_number].strip()
# append our stripped string onto the first_name list
first_name.append(stripped)
# save the entire row to the raw list
raw.append(row)
# if no override, continue like normal
else:
first_name.append(row)
# if we have a header, we need to remove it so it is not included in the submission
if args.noheader == False:
if is_override:
# Before we pop the first list item in first_name, save it to be our original headers so we can write them later
original_headers = first_name[0]
# We also need to pop the for item in the raw list or we will end up with extra data
raw.pop(0)
first_name.pop(0) #Remove header
o_first_name = list()
# We dont need to strip the first name list if we are overriding because it has already been taken care of
if is_override:
o_first_name = first_name
# Removes the [''] on each list item so we just end up with names when iterating through the list
else:
for l in first_name:
for b in l:
o_first_name.append(b)
# moved uniq_first_name outside of the if statement for later use.
uniq_first_name = []
if args.auto == True:
uniq_first_name = list(dict.fromkeys(o_first_name))
chunks = list(jpyh.splitlist(uniq_first_name, 10));
print("--- Read CSV with " + str(len(first_name)) + " first_name. " + str(len(uniq_first_name)) + " unique.")
else:
# splitting the name list into chunks of 10 due to api restrictions
chunks = list(jpyh.splitlist(first_name, 10));
print("--- Read CSV with " + str(len(first_name)) + " first_name")
print("--- Processed into " + str(len(chunks)) + " chunks")
if jpyh.query_yes_no("\n---! Ready to send to Genderdize. Proceed?") == False:
print("Exiting...\n")
sys.exit()
if os.path.isfile(ofile):
if jpyh.query_yes_no("---! Output file exists, overwrite?") == False:
print("Exiting...\n")
sys.exit()
print("\n")
if args.auto == True:
ofile = ofile + ".tmp"
response_time = [];
gender_responses = list()
with open(ofile, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
## TODO Add new system to write all rows of the original file. Done
# If we are overriding, we need to write different headers into the output csv file. We call the write_headers function to keep the code clean
if is_override:
write_headers(writer, original_headers)
# else, continue as expected
else:
writer.writerow(list(["first_name","gender", "probability", "count"]))
chunks_len = len(chunks)
stopped = False
for index, chunk in enumerate(chunks):
if stopped:
break
success = False
while not success:
try:
start = time.time()
if key_present:
dataset = genderize.get(chunk)
else:
dataset = Genderize().get(chunk)
gender_responses.append(dataset)
success = True
except GenderizeException as e:
#print("\n" + str(e))
logger.error(e)
#Error handling
if "response not in JSON format" in str(e) and args.catch == True:
if jpyh.query_yes_no("\n---!! 502 detected, try again?") == True:
success = False
continue
elif "Invalid API key" in str(e) and args.catch == True:
print("\n---!! Error, invalid API key! Check log file for details.\n")
else:
print("\n---!! GenderizeException - You probably exceeded the request limit, please add or purchase a API key. Check log file for details.\n")
stopped = True
break
response_time.append(time.time() - start)
print("Processed chunk " + str(index + 1) + " of " + str(chunks_len) + " -- Time remaining (est.): " + \
str( round( (sum(response_time) / len(response_time) * (chunks_len - index - 1)), 3)) + "s")
gender_dict = dict()
# Moved this function out of the autocomplete function to allow us to use it for the non-autocomplete writing as well
for response in gender_responses:
for d in response:
gender_dict[d.get("name")] = [d.get("gender"), d.get("probability"), d.get("count")]
# we need to iterate over all of our "cleaned" first names
for index, name in enumerate(o_first_name):
data = gender_dict.get(name)
# If we are overriding, we need to print our raw data plus our genderize information.
if is_override:
data_list = [data[0], data[1], data[2]]
writer.writerow(raw[index] + data_list)
# If we are not overriding, we print the standard information
else:
writer.writerow([name, data[0], data[1], data[2]])
# if we have the autocomplete enabled, we need to allow overriding in this mode as well.
if args.auto == True:
print("\nCompleting identical first_name...\n")
filename, file_extension = os.path.splitext(ofile)
with open(filename, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
# Before we enter the for loop, we need to print the correct headers into the output csv file.
# If we are overriding, we need to print out saved original headers as well as the new headers. We call our write_headers function to keep the code clean
if is_override:
write_headers(writer, original_headers)
# we need to remove duplicate items in our raw file for proper file writing.
raw_cleaned = remove_dupes(raw, column_number)
# If we are not overriding, we can print the standard headers.
else:
writer.writerow(list(["first_name","gender", "probability", "count"]))
# We need to iterate over our uniq_first_name list inorder to write the correct names
for index, name in enumerate(uniq_first_name):
# If we are overriding, we need to combine the data recieved from the genderize api and combine it with our clean raw list inorder to write the correct information
if is_override:
data = gender_dict.get(name)
data_list = [data[0], data[1], data[2]]
writer.writerow(raw_cleaned[index] + data_list)
# If we are not overriding, we can perform everything as expected.
else:
data = gender_dict.get(name)
writer.writerow([name, data[0], data[1], data[2]])
print("Done!\n")
def genderize():
if len(sys.argv) != 3:
print(
"Plrease specify input and output files: python genderize.py [input file path] [output file path]"
)
sys.exit()
dir_path = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
filename, file_extension = os.path.splitext(sys.argv[2])
ofile = filename + "_" + time.strftime("%Y%m%d-%H%M%S") + ".csv"
print("\n--- Input file: " + dir_path + os.sep + sys.argv[1])
print("--- Output file: " + dir_path + os.sep + ofile + "\n")
csv.field_size_limit(sys.maxsize)
#Initialize API key
genderize = Genderize(user_agent='GenderizeDocs/0.0', api_key=API_KEY)
with open(sys.argv[1], encoding="utf8") as csvfile:
readCSV = csv.reader(csvfile, delimiter=',')
names = []
for row in readCSV:
names.append(row)
names.pop(0)
print("--- Read CSV with " + str(len(names)) + " names")
chunks = list(jpyh.splitlist(names, 10))
print("--- Processed into " + str(len(chunks)) + " chunks")
if jpyh.query_yes_no(
"\n---! Ready to send to Genderdize. Proceed?") == False:
print("Exiting...\n")
sys.exit()
if os.path.isfile(ofile):
if jpyh.query_yes_no(
"---! Output file exists, overwrite?") == False:
print("Exiting...\n")
sys.exit()
print("\n")
response_time = []
with open(ofile, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
writer.writerow(list(["names", "gender", "probability", "count"]))
chunks_len = len(chunks)
for index, chunk in enumerate(chunks):
start = time.time()
#while True:
try:
dataset = genderize.get(chunk)
#break
except GenderizeException as e:
print(e)
'''
if "response not in JSON format" in e:
if jpyh.query_yes_no("\n---! 502 detected, try again?") == True:
print("Exiting...\n")
pass
else:
break
'''
print(
"\n---!! GenderizeException - You probably exceeded the request limit, please add or purchase a API key\n"
)
sys.exit()
response_time.append(time.time() - start)
print("Processed chunk " + str(index + 1) + " of " + str(chunks_len) + " -- Time remaining (est.): " + \
str( round( (sum(response_time) / len(response_time) * (chunks_len - index - 1)), 3)) + "s")
for data in dataset:
writer.writerow(data.values())
def genderize(args):
print(args)
# File initialization
dir_path = os.path.dirname(os.path.realpath(__file__))
logging.basicConfig(
filename=dir_path + os.sep + "log.txt",
level=logging.DEBUG,
format='%(asctime)s %(levelname)s %(name)s %(message)s')
logger = logging.getLogger(__name__)
ofilename, ofile_extension = os.path.splitext(args.output)
ofile = ofilename + "_" + time.strftime("%Y%m%d-%H%M%S") + ".csv"
ifile = args.input
if os.path.isabs(ifile):
print("\n--- Input file: " + ifile)
else:
print("\n--- Input file: " + dir_path + os.sep + ifile)
if os.path.isabs(ofile):
print("--- Output file: " + ofile)
else:
print("--- Output file: " + dir_path + os.sep + ofile + "\n")
# File integrity checking
if not os.path.exists(ifile):
print("--- Input file does not exist. Exiting.\n")
sys.exit()
if not os.path.exists(os.path.dirname(ofile)):
print("--- Error! Invalid output file path. Exiting.\n")
sys.exit()
# Some set up stuff
# csv.field_size_limit(sys.maxsize)
# Initialize API key
if not args.key == "NO_API":
print("--- API key: " + args.key + "\n")
genderize = Genderize(user_agent='GenderizeDocs/0.0', api_key=args.key)
key_present = True
else:
print("--- No API key provided.\n")
key_present = False
# Open ifile
with open(ifile, 'r', encoding="utf8") as csvfile:
readCSV = csv.reader(csvfile, delimiter=',', skipinitialspace=True)
i = next(readCSV) # set i to the headers as an array
first_name = []
allData = []
if args.auto == False: # add columns based off auto
i.append('male')
i.append('female')
if args.auto == True:
i.append('gender')
i.append('probability')
i.append('count')
for row in readCSV: # Read CSV into first_name list
allData.append(row)
first_name.append(row[1]) # specify the name column index here
if args.noheader == False:
first_name.pop(0) # Remove header
o_first_name = list()
# Deleted this part of the code because it substrings the name column
# for l in first_name:
# for b in l:
# o_first_name.append(b)
# print(o_first_name)
if args.auto == True:
chunks = list(jpyh.splitlist(first_name, 10))
print("--- Read CSV with " + str(len(first_name)) +
" first_name. " + str(len(first_name)) + " unique.")
else:
chunks = list(jpyh.splitlist(first_name, 10))
print("--- Read CSV with " + str(len(first_name)) + " first_name")
print("--- Processed into " + str(len(chunks)) + " chunks")
if jpyh.query_yes_no(
"\n---! Ready to send to Genderdize. Proceed?") == False:
print("Exiting...\n")
sys.exit()
if os.path.isfile(ofile):
if jpyh.query_yes_no(
"---! Output file exists, overwrite?") == False:
print("Exiting...\n")
sys.exit()
print("\n")
if args.auto == True:
ofile = ofile + ".tmp"
response_time = []
gender_responses = list()
with open(ofile, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
writer.writerow(i) # write row header based off input file header
chunks_len = len(chunks)
stopped = False
for index, chunk in enumerate(chunks):
if stopped:
break
success = False
while not success:
try:
start = time.time()
if key_present:
dataset = genderize.get(chunk)
else:
dataset = Genderize().get(chunk)
gender_responses.append(dataset)
success = True
except GenderizeException as e:
#print("\n" + str(e))
logger.error(e)
# zzError handling
if "response not in JSON format" in str(
e) and args.catch == True:
if jpyh.query_yes_no(
"\n---!! 502 detected, try again?"
) == True:
success = False
continue
elif "Invalid API key" in str(
e) and args.catch == True:
print(
"\n---!! Error, invalid API key! Check log file for details.\n"
)
else:
print(
"\n---!! GenderizeException - You probably exceeded the request limit, please add or purchase a API key. Check log file for details.\n"
)
stopped = True
break
response_time.append(time.time() - start)
print("Processed chunk " + str(index + 1) + " of " +
str(chunks_len) + " -- Time remaining (est.): " +
str(
round((sum(response_time) / len(response_time) *
(chunks_len - index - 1)), 3)) + "s")
singleDataDict1 = dict()
newSingleDataDict = dict()
singleDataDictGen = dict()
singleDataDictData = dict()
newDataListWithGen = list()
allDataList = list()
addDataDict = {}
newDataList = list()
newArry = list()
# get each row and make a new dictionatry
for l in allData:
for n in range(0, len(l)):
singleDataDict1[n] = l[n]
dictionary_copy = singleDataDict1.copy()
allDataList.append(dictionary_copy)
# Combined the input file data with the gender api data
for m in allDataList:
for n in dataset:
if n['name'] == m[1]: # specify name column here
singleDataDictGen = n
singleDataDictData = m
singleDataDictData.update(singleDataDictGen)
# delete the name key because of duplication
del singleDataDictData['name']
newDataListWithGen.append(singleDataDictData)
for newData in newDataListWithGen:
# add gender binary at the end of the data dictionary
if newData['gender'] == 'female' and newData[
'probability'] > 0.5:
newData.update({'male': 0, 'female': 1})
elif newData['gender'] == 'male' and newData[
'probability'] > 0.5:
newData.update({'male': 1, 'female': 0})
# since this is without argument auto delete gender, probability, and count
del newData['gender']
del newData['probability']
del newData['count']
# write the data column
writer.writerow(newData.values())
break
if args.auto == True:
print("\nCompleting identical first_name...\n")
# AUTOCOMPLETE first_name
# Create master dict
gender_dict = dict()
singleDataDict = dict()
singleDataDict1 = dict()
newSingleDataDict = dict()
singleDataDictGen = dict()
singleDataDictData = dict()
newDataListWithGen = list()
allDataList = list()
addDataDict = {}
newDataList = list()
newArry = list()
# get each row and make a new dictionatry
for l in allData:
for n in range(0, len(l)):
singleDataDict1[n] = l[n]
dictionary_copy = singleDataDict1.copy()
allDataList.append(dictionary_copy)
# Combined the input file data rows with the gender api data rows
for m in allDataList:
for n in dataset:
if n['name'] == m[1]: # specify name column here
singleDataDictGen = n
singleDataDictData = m
singleDataDictData.update(singleDataDictGen)
del singleDataDictData['name']
newDataListWithGen.append(singleDataDictData)
filename, file_extension = os.path.splitext(ofile)
with open(filename, 'w', newline='', encoding="utf8") as f:
writer = csv.writer(f)
writer.writerow(i) # write the header column
for newData in newDataListWithGen:
# write the data column
writer.writerow(newData.values())
print("Done!\n")
