Bulk move messages from All Mail to Trash

README.md

I hate GMail, but sometimes you gotta. They're notoriously awful at deleting lots of messages at once, as their own search engine will tell you. I had more than 1,000,000 messages I needed to purge and couldn't do it using their own tools. So I wrote my own.

purge_gmail.py

#!/usr/bin/env python
# -*- encoding: utf-8 -*-

## http://nickblogus.tumblr.com/post/73553186015/gmail-deleting-big-amount-of-old-messages-problem

import imaplib
from email.parser import HeaderParser
from email.utils import parsedate_tz, mktime_tz
from datetime import datetime, timedelta


def chunks(arr, num):
    for i in xrange(0, len(arr), num):
        yield arr[i:i + num]


def check_result(command, resp, result):
    assert resp == "OK", "%s failed with %s: %s" % (command, resp, result)
    return result


def main(host, userid, password, days):
    age_threshold = timedelta(days=int(days))
    now = datetime.now()
    
    print "connecting"
    box = imaplib.IMAP4_SSL(host, 993)
    
    print "logging in"
    check_result("logging in", *box.login(userid, password))
    
    print "selecting"
    check_result("selecting", *box.select("[Gmail]/All Mail"))
    
    print "searching"
    result = check_result("searching", *box.search(None, '(BEFORE "%s")' % (now - age_threshold).strftime("%d-%b-%Y")))
    
    uids = result[0].split()
    print "%d messages in result" % len(uids)

    total_deleted = 0
    chunk_size = 250
    for chunk in chunks(uids, chunk_size):
        deleted = False
        for attempt in range(0, 3):
            try:
                print "deleting %d, attempt %d" % (len(chunk), attempt)
                check_result("storing", *box.store(",".join(chunk), "+X-GM-LABELS", "\\Trash"))
                deleted = True
                total_deleted += chunk_size
                break
            except AssertionError, e:
                print "attempt %d failed: %s" % (attempt, e)

        assert deleted, "unable to delete after multiple attempts"
        print "deleted %6.2f%%, %d of %d" % ((total_deleted * 100.0 / len(uids)), total_deleted, len(uids))
        # assert box.expunge()[0] == "OK"


if __name__ == "__main__":
    import sys
    
    host, userid, password, days = sys.argv[1:]
    
    main(host, userid, password, days)