Qwiklabs Assessment: Log Analysis Using Regular Expression github code example
Example: Qwiklabs Assessment: Working with Regular Expressions
#!/usr/bin/env python3
import re
import csv
def contains_domain(address, domain):
"""Returns True if the email address contains the given,domain,in the domain position, false if not."""
domain = r'[\w\.-]+@'+domain+'$'
if re.match(domain,address):
return True
return False
def replace_domain(address, old_domain, new_domain):
"""Replaces the old domain with the new domain in the received address."""
old_domain_pattern = r'' + old_domain + '$'
address = re.sub(old_domain_pattern, new_domain, address)
return address
def main():
"""Processes the list of emails, replacing any instances of the old domain with the new domain."""
old_domain, new_domain = 'abc.edu', 'xyz.edu'
csv_file_location = '<csv_file_location>'
report_file = '<path_to_home_directory>' + '/updated_user_emails.csv'
user_email_list = []
old_domain_email_list = []
new_domain_email_list = []
with open(csv_file_location, 'r') as f:
user_data_list = list(csv.reader(f))
user_email_list = [data[1].strip() for data in user_data_list[1:]]
for email_address in user_email_list:
if contains_domain(email_address, old_domain):
old_domain_email_list.append(email_address)
replaced_email = replace_domain(email_address,old_domain,new_domain)
new_domain_email_list.append(replaced_email)
email_key = ' ' + 'Email Address'
email_index = user_data_list[0].index(email_key)
for user in user_data_list[1:]:
for old_domain, new_domain in zip(old_domain_email_list, new_domain_email_list):
if user[email_index] == ' ' + old_domain:
user[email_index] = ' ' + new_domain
f.close()
with open(report_file, 'w+') as output_file:
writer = csv.writer(output_file)
writer.writerows(user_data_list)
output_file.close()
main()