Logs for jdev

Show join/part/nick changes:

[00:00:07] * Tobias left the chat.
[00:00:27] * joachim.lindborg joined the chat.
[00:00:28] * joachim.lindborg left the chat.
[00:00:38] * joachim.lindborg joined the chat.
[00:00:38] * joachim.lindborg left the chat.
[00:05:40] * joachim.lindborg joined the chat.
[00:56:34] * Lance_ joined the chat.
[00:57:19] * Lance left the chat.
[01:02:49] * Florob left the chat.
[01:03:20] * Lance_ left the chat.
[02:18:26] * Treebilou left the chat.
[02:21:03] * Treebilou joined the chat.
[04:05:54] * darkrain left the chat.
[04:08:30] * darkrain joined the chat.
[04:12:53] * Lance_ joined the chat.
[08:16:08] * Tobias joined the chat.
[08:20:12] * joachim.lindborg left the chat.
[08:24:09] * joachim.lindborg joined the chat.
[09:05:22] * Lance_ left the chat.
[10:08:35] * Alex joined the chat.
[11:02:23] * Alex left the chat.
[11:24:21] * naw joined the chat.
[11:37:34] * smoku joined the chat.
[11:41:12] * smoku left the chat.
[11:48:01] * smoku joined the chat.
[11:58:49] * smoku left the chat.
[12:09:12] * Flow joined the chat.
[12:11:36] * Flow left the chat.
[12:21:28] * joachim.lindborg left the chat.
[13:36:32] * Kev left the chat.
[14:02:59] * naw left the chat.
[14:41:30] * jonkri joined the chat.
[14:45:31] <jonkri> why is it up to xmpp clients to send invalid-namespace, invalid-xml, and so on? it adds some in mind some unnecessary complexity to clients just because some other systems might be buggy. i see that some clients are ignoring the "SHOULD" and sends bad-format instead. for me, that seems like a more attractive options, but i'm not entirely happy ignoring the "SHOULD". what are your thoughts on this?
[15:09:01] * Flow joined the chat.
[15:14:03] * Flow left the chat.
[15:58:48] * harrykar left the chat.
[16:37:15] * harrykar joined the chat.
[16:51:29] * Tobias left the chat.
[17:27:26] * Tobias joined the chat.
[17:38:34] * Tobias left the chat.
[17:38:45] * Flow joined the chat.
[17:38:58] * Tobias joined the chat.
[17:39:01] * Kev joined the chat.
[17:48:48] <Kev> jonkri: Because any entity that may receive data may receive invalid data.
[17:57:57] <jonkri> hi Kev, and thanks for you reply. i'm sorry to say that i don't quite understand what you mean, though. i'm making a client xmpp library, and i'm all for sending bad-format, but i don't see why i should need to be more informative than that
[17:58:15] <Kev> Oh, I see what you mean.
[17:58:26] <Kev> I thought you were complaining about having to send errors when receiving bad data.
[17:58:35] <jonkri> ah, no
[17:59:16] <Kev> So, it probably doesn't really matter what error you send when receiving bad XML.
[17:59:52] <Kev> It'll help people debug problems if you send the right one, but ultimately you've already received bad data, so ...
[18:00:10] <jonkri> exactly. i think it should be a MAY
[18:00:10] <Kev> That said - there's a spec and you should try to follow it :)
[18:03:24] <jonkri> it may seem a little arrogant, but in my opinion, debugging faulty clients shouldn't really be my problem (not when it makes the implementation significantly more complex, with all these special cases)
[18:09:07] * Alex joined the chat.
[18:18:35] * Alex left the chat.
[19:07:14] * Asterix joined the chat.
[19:43:56] * Florob joined the chat.
[20:02:09] <jonkri> "Interoperability Note: In RFC 3920, the text regarding inclusion of the 'id' attribute was ambiguous, leading some implementations to leave the attribute off the response stream header."
[20:02:10] <jonkri> should a client xmpp library be liberal in accepting response stream headers without an id attribute?
[20:14:43] <jonkri> if i report a stream error, can i then close the stream and terminate the underlying tcp connection immediately without waiting for the other entity to finish processing? (since i'm not going to accept anything that the other entity is going to send anyway)
[20:41:09] * Flow left the chat.
[20:44:58] * Tobias left the chat.
[20:54:42] * jonkri left the chat.
[20:54:43] * jon joined the chat.
[21:08:30] * Flow joined the chat.
[21:13:06] * Flow left the chat.
[21:28:35] * Kev left the chat.
[21:34:54] * Asterix left the chat.
[21:54:11] * Florob left the chat.
[21:54:12] * Lance_ joined the chat.
[21:54:22] * Florob joined the chat.
[21:56:02] <Lance_> jonkri: yes, that is what you're supposed to do. See http://tools.ietf.org/html/rfc6120#section-4.9.1.1
[21:56:47] <Lance_> er, well, I didn't read references far back enough
[21:57:20] <Lance_> you are supposed to wait for the other side to close the stream, within a reasonable period of time
[21:58:30] <Lance_> that said, ending immediately generally works and is how some servers operate anyway
[22:35:18] * zash_ joined the chat.
[22:39:31] * zash_ left the chat.
[22:39:36] * zash_ joined the chat.
[22:46:44] * zash_ left the chat.
[22:54:10] * Minos left the chat.
[22:54:21] <jon> thanks Lance_
[23:20:43] * Zash joined the chat.
[23:27:31] * Zash left the chat.
[23:28:38] * Zash joined the chat.
[23:36:41] * Zash left the chat.