Extract all email addresses in from/to/cc fields of every msg in one Gmail folder

gistfile1.py

"""Create a connection to Gmail and do something with the results

References:
http://www.voidynullness.net/blog/2013/07/25/gmail-email-with-python-via-imap/
and
https://yuji.wordpress.com/2011/06/22/python-imaplib-imap-example-with-gmail/
"""
__author__ = 'abought'

import email
import imaplib
import getpass
import sys
import re

from pprint import pprint as pp

# User may want to change these parameters if running script as-is
SEARCH_FOLDER = '[Gmail]Trash'  # TODO: A user will want to change this
    # Other folders: "INBOX", "[Gmail]/All Mail"
DEFAULT_MAIL_SERVER = 'imap.gmail.com'

# No user parameters below this line
ADDR_PATTERN = re.compile('<(.*?)>')  # Finds email as <[email protected]>


def connect(user, pwd, server=DEFAULT_MAIL_SERVER):
    """Connect to [the specified] mail server. Return an open connection"""
    conn = imaplib.IMAP4_SSL(server)
    try:
        conn.login(user, pwd)
    except imaplib.IMAP4.error:
        print "Failed to login"
        sys.exit(1)
    return conn


def print_folders(conn):
    """Print a list of open mailbox folders"""
    for f in conn.list():
        print "\t", f


def get_folder(conn, folder_name):
    """Fetch a specific folder (or label) from server"""

    if conn.state == "SELECTED":
        # Explicitly close any previously opened folders; may not be necessary
        conn.close()

    rv, data = conn.select(folder_name)
    if rv != 'OK':
        print "Could not open specified folder. Known labels:"
        print_folders(conn)
    return conn


def get_email_ids(conn, query='ALL'):
    """Get the numeric IDs for all emails in a given folder"""
    if conn.state != "SELECTED":
        raise imaplib.IMAP4.error("Cannot search without selecting a folder")

    rv, data = conn.uid('search', None, query)
    if rv != 'OK':
        print "Could not fetch email ids"  # for some reason...
        return []

    return data[0].split()


def fetch_message(conn, msg_uid ):
    """
    Fetch a specific message uid (not sequential id!) from the given folder;
    return the parsed message. User must ensure that specified
    message ID exists in that folder.
    """
    # TODO: Could we fetch just the envelope of the response to save bandwidth?
    rv, data = conn.uid('fetch', msg_uid, "(RFC822)")
    if rv != 'OK':
        print "ERROR fetching message #", msg_uid
        return {}

    return email.message_from_string(data[0][1])  # dict-like object


def get_recipients(msg_parsed):
    """Given a parsed message, extract and return recipient list"""
    recipients = []
    addr_fields = ['From', 'To', 'Cc', 'Bcc']

    for f in addr_fields:
        rfield = msg_parsed.get(f, "") # Empty string if field not present
        rlist = re.findall(ADDR_PATTERN, rfield)
        recipients.extend(rlist)

    return recipients

if __name__ == "__main__":
    username = raw_input("Full email address: ")
    password = getpass.getpass()

    # Connect
    mail_conn = connect(username, password)

    # Open a specific folder and get list of email message uids
    mail_conn = get_folder(mail_conn, SEARCH_FOLDER)
    msg_uid_list = get_email_ids(mail_conn)

    # Fetch a list of recipients
    all_recipients = []
    for msg_uid in msg_uid_list:
        msg = fetch_message(mail_conn, msg_uid)
        recip_list = get_recipients(msg)
        all_recipients.extend(recip_list)

    # Very unsophisticated way of showing the recipient list
    print "List of all recipients:"
    print "------------"
    pp(all_recipients)

    print "\n\n List of all UNIQUE recipients:"
    print "-------------------------------"
    pp(set(all_recipients))

    try:
        mail_conn.close()  # Close currently selected folder (if any)
    finally:
        mail_conn.logout()

Thanks! That's what I was looking for. Is that a way of limiting by date?