Bug 27672

Summary:	telepathy-butterfly crashed with NotImplementedError in __init__()
Product:	Telepathy	Reporter:	Sebastien Bacher <seb128>
Component:	butterfly	Assignee:	Telepathy bugs list <telepathy-bugs>
Status:	RESOLVED FIXED	QA Contact:	Telepathy bugs list <telepathy-bugs>
Severity:	normal
Priority:	medium
Version:	unspecified
Hardware:	Other
OS:	All
URL:	http://git.collabora.co.uk/?p=user/jonny/papyon.git;a=shortlog;h=refs/heads/27672
Whiteboard:	review+
i915 platform:		i915 features:

Description Sebastien Bacher 2010-04-15 14:23:12 UTC

The bug has been opened on https://bugs.launchpad.net/ubuntu/+source/telepathy-butterfly/+bug/561800

"I installed the upgrade packages from today.
Initialized Empathy and it just dropped this error.

telepathy-butterfly 0.5.8-1ubuntu1

Traceback (most recent call last):
  File "/usr/lib/pymodules/python2.6/papyon/msnp/base.py", line 107, in _dispatch_command
    handler(command)
  File "/usr/lib/pymodules/python2.6/papyon/msnp/notification.py", line 453, in _handle_NLN
    urllib.unquote(command.arguments[idx]))
  File "/usr/lib/pymodules/python2.6/papyon/p2p.py", line 172, in parse
    result = MSNObject(creator, size, type, location, friendly, shad, shac)
  File "/usr/lib/pymodules/python2.6/papyon/p2p.py", line 107, in __init__
    raise NotImplementedError
NotImplementedError

http://launchpadlibrarian.net/44261326/butterfly-log.txt.tar.gz
...
If you think I have deleted too much things I can sent it to you directly,"

Comment 1 Jonny Lamb 2010-04-16 03:49:54 UTC

After putting these details on the wrong bug by accident, I'll try again:

Okay, that log was pretty useful and shows what's wrong with the SHA1D. Here is
msn object XML:

<msnobj Creator="example@hotmail.com" Size="19086" Type="3"
Location="TFRlgrTEF.dat" Friendly="SQBtAGEAZwBlAG4AIAAxADAAAAA=" SHA1D="1h
Jo825av2d/RagyNCPI0V3rTs=" SHA1C="VK4lMv5Ori9LfY1J6THi4o/sOAU="/>

Take the SHA1D attribute:

    SHA1D="1h Jo825av2d/RagyNCPI0V3rTs="

I've no idea what that "1h", but we need to work around it. I had a little look
at the other MSN implementations and couldn't find anything about this. I'll
cook up a patch.

Oh, also, the rogue string isn't always first in the SHA1D attribute:

% grep "Invalid SHA1" butterfly-log.txt
papyon/papyon.p2p-WARNING: 14/04/10 18:16:26.809406: Invalid SHA1D in
MSNObject: /IkRrvJpPvK/ZFEZEM7eRhle k4=
papyon/papyon.p2p-WARNING: 14/04/10 18:16:27.35620: Invalid SHA1D in MSNObject:
1h Jo825av2d/RagyNCPI0V3rTs=
papyon/papyon.p2p-WARNING: 14/04/10 19:18:38.737390: Invalid SHA1D in
MSNObject: 1h Jo825av2d/RagyNCPI0V3rTs=
papyon/papyon.p2p-WARNING: 14/04/10 19:20:31.38971: Invalid SHA1D in MSNObject:
1h Jo825av2d/RagyNCPI0V3rTs=

:-(

Comment 2 Jonny Lamb 2010-04-16 04:11:09 UTC

Check it out.

Comment 3 Riccardo (c10ud) 2010-04-16 05:24:02 UTC

(In reply to comment #2)
> Check it out.

i think those contacts are using a faulty client, i asked on the bug report if the reporter can tell us which client it is (so, if it's oss, we can fix it)

https://bugs.launchpad.net/ubuntu/+source/telepathy-butterfly/+bug/561800/comments/13

Comment 4 Jonny Lamb 2010-04-16 07:55:32 UTC

(In reply to comment #3)
> i think those contacts are using a faulty client, i asked on the bug report if
> the reporter can tell us which client it is (so, if it's oss, we can fix it)

Probably. Is there anything against committing this until the faulty clients are fixed? At that point the commit can be reverted.

Comment 5 Riccardo (c10ud) 2010-04-17 04:39:26 UTC

(In reply to comment #4)
> (In reply to comment #3)
> > i think those contacts are using a faulty client, i asked on the bug report if
> > the reporter can tell us which client it is (so, if it's oss, we can fix it)
> 
> Probably. Is there anything against committing this until the faulty clients
> are fixed? At that point the commit can be reverted.

i didn't try it yet, but seems ok and i'm fine with it. 

strange thing is i never seen anything like this in amsn/emesene (maybe they just discard/ignore the error?)..i hope to know which client it is though, i'm curious :)

Comment 6 Olivier Le Thanh Duong 2010-04-18 01:49:15 UTC

We should be robust against faulty client anyway so let's merge this

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.