On Mon, Feb 15, 2010 at 2:47 PM, BJ Swope <bigbluesw...@gmail.com> wrote:
> On Mon, Feb 15, 2010 at 2:31 PM, Stephen Hansen <apt.shan...@gmail.com> wrote:
>> On Mon, Feb 15, 2010 at 10:53 AM, BJ Swope <bigbluesw...@gmail.com> wrote:
>>>
>>> File "/usr/lib/python2.5/email/_parseaddr.py", line 142, in mktime_tz
>>> if data[9] is None:
>>> TypeError: 'NoneType' object is unsubscriptable
>>>
>>> I'm parsing a bunch of spam and using the date field from the spams
>>> for a date-time stamp.
>>>
>>> I've fixed the lib on my box to place the call inside a try/except
>>> clause to catch the exception now, but it seems the module has a bug
>>> in it.
>>
>> While there may or may not be a bug in the library, I don't think its where
>> you're fixing. Just because an exception occurs in a function doesn't mean
>> that function is broken: its documented as accepting a 10 item tuple, only.
>> Functions in the stdlib generally -should- throw exceptions on invalid
>> input.
>> Someone's passing None into it, which its not allowed to do. So -that's-
>> where the bug probably is, I think. (Although it may not be the immediate of
>> mktime_tz; it could be happening higher up on the stack)
>> Advice: Always post complete tracebacks to c.p.l/python-list :)
>> --S
>
>
> From the module:
>
> def mktime_tz(data):
> """Turn a 10-tuple as returned by parsedate_tz() into a UTC timestamp."""
> if data[9] is None:
> # No zone info, so localtime is better assumption than GMT
> return time.mktime(data[:8] + (-1,))
> else:
> t = time.mktime(data[:8] + (0,))
> return t - data[9] - time.timezone
>
>
> It appears that the module is trying to accommodate the potential
> missing TZ data because poorly written emails are missing the TZ data.
>
> I discarded all the crontab emails that had the full traceback in
> them. I took out the try/except clause in the hopes that I'll get
> another exception soon.
>
> If I do I'll post the entire exception traceback.
>
Speak of the devil and demons appear...
/logs/python/imap_fetcher/spam_serv1.bigbluenetworks.com.py
Parsing of emails for spam at serv1.bigbluenetworks.com failed.
Traceback (most recent call last):
File "/logs/python/imap_fetcher/spam_serv1.bigbluenetworks.com.py",
line 81, in <module>
clean_stale_mail()
File "/logs/python/imap_fetcher/spam_serv1.bigbluenetworks.com.py",
line 24, in clean_stale_mail
utc_msg_date = email.utils.mktime_tz(msg_date2)
File "/usr/lib/python2.5/email/_parseaddr.py", line 142, in mktime_tz
if data[9] is None:
TypeError: 'NoneType' object is unsubscriptable
def clean_stale_mail():
msg_date1= the_email.get('Date')
msg_date2 = email.utils.parsedate_tz(msg_date1)
try:
utc_msg_date = email.utils.mktime_tz(msg_date2)
except OverflowError:
M.store(msg_id, '+FLAGS.SILENT', '\\Deleted')
return
utc_stale_date = time.time() - (86000*stale_days)
if utc_msg_date <= utc_stale_date:
M.store(msg_id, '+FLAGS.SILENT', '\\Deleted')
try:
#M = imaplib.IMAP4(HOST)
M = imaplib.IMAP4_SSL(HOST)
M.login(USER, PASSWD)
M.select(MAILBOX)
response, msg_list = M.search(None, 'ALL') # response is IMAP
response, msg_list is list of Message IDs
for msg_id in msg_list[0].split(): #msg_list[0] is a space
separated string of message ids
response, message = M.fetch(msg_id, '(RFC822)') # response is
the IMAP response, message is an RFC222 message
msg = email.FeedParser.FeedParser( )
msg.feed(message[0][1])
the_email = msg.close()
clean_stale_mail()
if the_email.is_multipart():
for part in the_email.walk():
if part.get_content_type() == 'text/html':
decoded_part = part.get_payload(decode=1)
soup_parse(decoded_part)
elif the_email.get_content_type() == 'text/html':
decoded_part = the_email.get_payload(decode=1)
soup_parse(decoded_part)
elif the_email.get_content_type() == 'text/plain':
msg_payload = the_email.get_payload()
manual_parse(msg_payload)
else:
continue
#print msg_id, "Did't match any defined content types..."
#print the_email
M.expunge()
M.close()
M.logout()
--
http://mail.python.org/mailman/listinfo/python-list
