Tool to extract emails from body and FROM field

# -*- coding: utf-8 -*-
# Python Tool to extract emails from body and FROM field from gmail (imap) account
# author: pymen # [email protected]
# dot forget to enter your login and password in user vaviables 'userlogin' and 'userpass'

import sys
import re
import poplib
import email
import imaplib

in_server = 'pop.gmail.com'
username = '[email protected]'
userpass = ""

class gmail():
    def __init__(self):
        self.emaildata_from = []
        self.emaildata_body = []
        self.regex_from = re.compile(r'[\w\-][\w\-\.]+@[\w\-][\w\-\.]+[a-zA-Z]{1,4}')
        self.regex_html = re.compile(r""".*<a href="mailto:(.*)">""")
        self.regex_txt = re.compile(r"""(?:[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*|"(?:[\x01-\x08\x0b\x0c\x0e-\x1f\x21\x23-\x5b\x5d-\x7f]|\\[\x01-\x09\x0b\x0c\x0e-\x7f])*")@(?:(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?|\[(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?|[a-z0-9-]*[a-z0-9]:(?:[\x01-\x08\x0b\x0c\x0e-\x1f\x21-\x5a\x53-\x7f]|\\[\x01-\x09\x0b\x0c\x0e-\x7f])+)\])""")

    def connect(self):
        try:
            self.m = imaplib.IMAP4_SSL(in_server)
            self.m.login(username, userpass)
            self.m.select()
            print ("connecting to account %s" %username)
        except imaplib.IMAP4.error:
            print ("Error. Please check your login and passors in user vaviables 'userlogin' and 'userpass' above")
            sys.exit()

    def receive_imap(self):
        typ, data = self.m.search(None, 'ALL')
        number = data[0].split()[-1]
        print ("There %s number of messages in this account " %number)
        print ("Processing message: "),
        for num in data[0].split():
            print num,
            typ, data = self.m.fetch(num, '(RFC822)')
            raw_email_string = data[0][1]
            self.msg = email.message_from_string(raw_email_string)
            self.get_email_address_from()
            for part in self.msg.walk():
                if part.get_content_type() == "text/plain":
                    self.body = part.get_payload(decode=True)
                    self.get_email_address_html()
                if part.get_content_type() == "text/html":
                    self.body = part.get_payload(decode=True)
                    self.get_email_address_txt()

    def get_email_address_html(self):
        self.email_address = self.regex_html.findall(self.body)
        self.emaildata_body += self.email_address
    def get_email_address_txt(self):
        self.email_address = self.regex_txt.findall(self.body)
        self.emaildata_body += self.email_address
    def get_email_address_from(self):
        email_address = self.msg.get("From")
        self.email_address =  self.regex_from.findall(email_address)
        self.emaildata_from += self.email_address

    def print_results(self):
        self.m.close()
        self.m.logout()
        print ("\n\nResult emails from body:")
        print (list(set(self.emaildata_body)))
        print ("\nResult emails from FROM:")
        print (list(set(self.emaildata_from)))

if __name__ == "__main__":
    gm = gmail()
    gm.connect()
    gm.receive_imap()
    gm.print_results()