timestamp.py

#!/usr/bin/env python
# timestamp.py
# Author: Zachary Cutlip
#         twitter: @zcutlip
# Purpose: A program to fix sorting of mail messages that have been POPed or IMAPed in the wrong order
#          Compares time stamp sent and timestamp received on an RFC822-formatted email message,
#          and renames the message file using the most recent timestamp that is no more than 24 hours
#          after the date sent.  Updates the file's atime/mtime with the timestamp, as well. Does not
#          modify the headers or contents of the message.

import sys
import os
import email
import email.utils
import re
regex=re.compile("([0-9]+)(\..*$)")

msg_filename=sys.argv[1]
print msg_filename
msg=email.message_from_file(open(msg_filename,"r"))

datesent=email.utils.mktime_tz(email.utils.parsedate_tz(msg.get("Date")))

#print datesent
date_rx=0
for rx_hdr in msg.get_all("Received"):
    rx_date_str=rx_hdr.split(";")[-1]
    #print rx_date_str
    tmp_rx=email.utils.mktime_tz(email.utils.parsedate_tz(rx_date_str))
    if(tmp_rx - datesent) > (24*60*60):
        continue

    date_rx=tmp_rx
    break

#print date_rx


#received header missing?
if(date_rx < 1):
    date_rx = datesent

m=regex.match(msg_filename)

if len(m.groups()) == 2:
    newfilename=str(int(date_rx))+m.groups()[1]
    print newfilename
    if os.path.isfile(newfilename):
        print "File exists!: "+newfilename
    else:
        os.rename(msg_filename,newfilename)
        os.utime(newfilename,(date_rx,date_rx))