With python2 I'm getting a traceback in the email gateway when
processing an IMAP mailbox:

roundup@tux5:~$ roundup-mailgw /var/lib/tracker/tracker-home/ imap
tracker@example.com:verysecret@tux2
Traceback (most recent call last):
  File "/usr/local/bin/roundup-mailgw", line 6, in <module>
    run()
  File
"/usr/local/lib/python2.7/dist-packages/roundup/scripts/roundup_mailgw.py",
line 226, in run
    sys.exit(main(sys.argv))
  File
"/usr/local/lib/python2.7/dist-packages/roundup/scripts/roundup_mailgw.py",
line 220, in main
    cram)
  File "/usr/local/lib/python2.7/dist-packages/roundup/mailgw.py", line
1363, in do_imap
    self.handle_Message(message_from_bytes(s, RoundupMessage))
  File "/usr/lib/python2.7/email/__init__.py", line 57, in
message_from_string
    return Parser(*args, **kws).parsestr(s)
  File "/usr/lib/python2.7/email/parser.py", line 82, in parsestr
    return self.parse(StringIO(text), headersonly=headersonly)
TypeError: StringIO() argument 1 must be string or buffer, not _io.BytesIO


It looks to me like a BytesIO instance is passed to the email package
message_from_string (aliased in roundup to message_from_bytes for
python2) instead of bytes, so this should use getvalue() on the BytesIO
instance for passing to the email Parser.

Didn't anybody use the email gateway so far?
I wonder how this could last so long without being detected?

Hi Ralf:

Email interfaces other than pipe and mailbox are untested
post python3 conversion.

See: https://issues.roundup-tracker.org/issue2550974

I have that as one of my CTA in the 2.0.0alpha announcements.

-- rouilj

Pushed a fix in 7264b2e79a31
Adding a test is hard as it would have to somehow emulate imap (or
instrument the imap handling somehow).

Anybody could give this a check?
Works for me for now with python2, tested also with python3 in debug mode.

Leaving this open for now, should get some more testing.

I requested assistance in testing this in the 2.0.0alpha0 release notes.

Ralf any ideas on testing this, or are you satisfied that it is fixed at this point?

If you are happy with it can you close it.

Ralf any conclusion on this?

Ping Ralf. Any final thoughts on this?

Closing as abandoned.

Sorry for not following up on this: I have no good idea how to offline test this in a test harness. But after my fix it has not re-occurred. Note that in the meanwhile all my tracker instances are python3 and afair this was originally a python2 issue.

So I've set the resolution to 'fixed'.

Thanks John for the persistence and again sorry for neglecting to answer this.

Ralf