#instantbird log on 11 10 2012

All times are UTC.

00:08:30 <-- meh has quit (Quit: I don't want to live on this planet anymore.)
00:09:55 <Mic> flo-retina: my second ICQ account fails to work, so I can't test the typing notifications (nor can I ask anyone at this time to send my messages to try). I can confirm that your patch makes the "Interruptions is not defined" error go away though.
00:10:14 <-- unghost has quit (Quit: Ухожу я от вас (xchat 2.4.5 или старше))
00:10:37 <Mic> I can register a new account if you like...
00:10:44 <flo-retina> I know it works (ie I can reproduce the bug before, and not any more after applying the patch)
00:11:04 <flo-retina> I was more asking for code inspection than testing
00:11:28 <flo-retina> ie are there corner cases that I may have missed but are visible for a fresh pair of eyes, etc... :)
00:15:52 <flo-retina> Mic: don't spend too much time on that patch, it's pretty trivial :)
00:15:58 <Mic> It looks OK.
00:16:33 <-- gerard-majax has quit (Ping timeout)
00:17:23 <Mic> _showFirstMessages (which imported it first) wasn't called for detached conversations, because "loaded" was set to true in finishImport, so it never got imported on the detached conversation
00:17:51 <flo-retina> right
00:19:04 <instantbot> benediktp@ymail.com granted review for attachment 2078 on bug 1783.
00:19:06 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1783 nor, --, ---, florian, NEW, Typing notification never goes away for tabs in a detached window
00:19:16 <Mic> I know why I always avoided this file :P
00:20:03 <flo-retina> is it that awful?
00:20:34 <flo-retina> it's tabbrowser.xml that I tend to avoid.
00:20:57 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/8725800af23f - Florian Quèze - Bug 1783 - Typing notification never goes away for tabs in a detached window, r=Mic.
00:20:58 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/ebce4e225799 - aleth - Bug 1784 - Atheme Nickserv auth success message not recognized when using grouped nicks, r=fqueze.
00:20:59 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/e0b125ac72a9 - aleth - Bug 1779 - Focusing Instantbird by clicking an unselected conversation tab doesn't mark the current conversation as read, r=fqueze.
00:21:00 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/992c82d47cc1 - aleth - Bug 1787 - [Bubbles] Change Show nick border radius to better match the border radius of the bubbles, r=fqueze.
00:21:59 <instantbot> florian@instantbird.org set the Resolution field on bug 1783 to FIXED.
00:22:01 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1783 nor, --, 1.3, florian, RESO FIXED, Typing notification never goes away for tabs in a detached window
00:22:29 <instantbot> florian@instantbird.org set the Resolution field on bug 1784 to FIXED.
00:22:31 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1784 nor, --, 1.3, aletheia2, RESO FIXED, Atheme Nickserv auth success message not recognized when using grouped nicks
00:22:59 <instantbot> florian@instantbird.org set the Resolution field on bug 1779 to FIXED.
00:23:08 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1779 nor, --, 1.3, aletheia2, RESO FIXED, Focusing Instantbird by clicking an unselected conversation tab doesn't mark the current conversatio
00:24:11 <instantbot> florian@instantbird.org set the Resolution field on bug 1787 to FIXED.
00:24:20 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1787 min, --, 1.3, aletheia2, RESO FIXED, [Bubbles] Change Show nick border radius to better match the border radius of the bubbles
00:24:28 <flo-retina> Good night! :)
00:24:34 <Mic> Good night.
00:24:41 <Mic> One moment, flo-retina
00:25:21 <flo-retina> is a new patch for bug 1511 coming any minute from now?
00:25:25 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1511 nor, --, ---, benediktp, ASSI, Possible to add duplicate accounts
00:25:52 <Mic> Is the gmail.com domain optional for GTalk account (if one has an @gmail.com" adress?)?
00:26:14 <flo-retina> it is
00:26:31 <flo-retina> an account foobar and foobar@gmail.com are the same
00:26:39 <flo-retina> foobar and foobar@googlemail.com are the same too
00:26:57 <Mic> And that's the only difference? I've done a similar check for facebook.com and the russian networks already..
00:27:15 <flo-retina> foobar@gmail.com and foobar@googlemail.com are more or less the same, but one of them is necessarily invalid, and will fail to connect.
00:28:09 <flo-retina> Mic: for gtalk I think I would remove the "@g[oogle]mail.com" suffix when normalizing for the comparison (and only for the comparison of course)
00:29:26 <Mic> Here's my current WIP: http://pastebin.instantbird.com/99889
00:29:44 <Mic> For GTalk, see line 195 for the XMPPAccount part and ...
00:30:05 <Mic> ... and line 100 for the gtalk protocol part.
00:30:40 <Mic> It's still absolutely untested though :(
00:31:53 <flo-retina> that doesn't handle the googlemail.com case
00:32:41 <flo-retina> did you mean map instead of forEach?
00:32:54 <flo-retina> I don't see what it would do to return something from a forEach callback
00:32:58 <Mic> Yes :(
00:33:04 <flo-retina> and I'm confused by these split/indexOf calls
00:33:18 <Mic> On which line?
00:33:32 <-- Optimizer has quit (Ping timeout)
00:34:10 <Mic> In the map callback? Yes, that could need an explanation.
00:34:59 <flo-retina> 7 locales ready already :)
00:35:06 <flo-retina> it and es-ES were finished today
00:35:37 <flo-retina> Mic: especially, I don't understand why you need to do anything before calling _parseJID
00:35:38 <Mic> x.split("/")[0] will fetch the part before the resource, indexOf("@") checks if there's a node and domain and adds the default server is there's no domain. After that jid's are created from this "corrected name" and compared for the return value.
00:36:25 <flo-retina> I would have more confidence in that code if it used the exact same logic that is already used when connecting the accounts
00:36:28 <flo-retina> http://mxr.mozilla.org/comm-central/source/chat/protocols/gtalk/gtalk.js#45 for example
00:36:33 <Mic> Should we postpone this to tomorrow? I'm having trouble concentrating..
00:36:56 --> Optimizer has joined #instantbird
00:38:05 <flo-retina> Mic: Yes. I'm too tired to give final review comments (ie I wouldn't r+ anything like that this evening)
00:38:37 <Mic> OK, sorry for holding you up..
00:39:55 <Mic> "Mic: especially, I don't understand why you need to do anything before calling _parseJID", I see, btw
00:40:40 <Mic> Good night! :)
00:40:46 <flo-retina> Good night
00:40:47 <flo-retina> :)
00:41:45 <-- Mic has quit (Quit: Instantbird 1.3a1pre -- http://www.instantbird.com)
00:49:58 <instant-buildbot> build #347 of linux-onCommit is complete: Success [build successful]  Build details are at http://buildbot.instantbird.org/builders/linux-onCommit/builds/347
00:51:03 <-- Optimizer has quit (Connection reset by peer)
00:51:10 --> Optimizer has joined #instantbird
00:57:13 --> wesj1 has joined #instantbird
00:58:55 <-- wesj1 has quit (Ping timeout)
01:05:54 --> mconley has joined #instantbird
01:30:25 <-- rosonline has quit (Ping timeout)
01:47:27 <-- mconley has quit (Input/output error)
02:01:44 <instant-buildbot> build #334 of win32-onCommit is complete: Success [build successful]  Build details are at http://buildbot.instantbird.org/builders/win32-onCommit/builds/334
02:22:35 <-- Mook_as has quit (Quit: Mook_as)
03:05:42 --> meh has joined #instantbird
03:06:27 <instant-buildbot> build #315 of macosx-onCommit is complete: Success [build successful]  Build details are at http://buildbot.instantbird.org/builders/macosx-onCommit/builds/315
03:17:04 <-- skeledrew has quit (Ping timeout)
03:17:18 --> skeledrew has joined #instantbird
03:35:03 <instant-buildbot> build #681 of linux-nightly-default is complete: Success [build successful]  Build details are at http://buildbot.instantbird.org/builders/linux-nightly-default/builds/681
03:48:04 --> Mook has joined #instantbird
04:08:54 <-- meh has quit (Quit: I don't want to live on this planet anymore.)
05:15:15 <instant-buildbot> build #769 of win32-nightly-default is complete: Success [build successful]  Build details are at http://buildbot.instantbird.org/builders/win32-nightly-default/builds/769
05:59:40 <-- Optimizer has quit (Ping timeout)
06:37:42 <-- Mook has quit (Quit: Mook)
07:41:17 --> Optimizer has joined #instantbird
07:41:35 --> jb has joined #instantbird
07:46:01 <instant-buildbot> build #682 of macosx-nightly-default is complete: Success [build successful]  Build details are at http://buildbot.instantbird.org/builders/macosx-nightly-default/builds/682
08:08:12 <-- jb has quit (Ping timeout)
08:09:17 --> jb has joined #instantbird
08:32:17 <-- Optimizer has quit (Ping timeout)
08:33:17 --> rosonline has joined #instantbird
09:09:57 <-- jb has quit (Ping timeout)
09:11:53 --> deltafalcon has joined #instantbird
09:12:13 <-- rosonline has quit (Quit: Experimentem Instantbird 1.2 -- http://www.instantbird.com)
09:22:32 --> Optimizer has joined #instantbird
10:11:44 <-- Optimizer has quit (Ping timeout)
10:15:11 --> Optimizer has joined #instantbird
10:26:24 <-- deltafalcon has quit (Quit: Instantbird 1.2 -- http://www.instantbird.com)
10:38:03 <-- Even has quit (Input/output error)
10:42:59 --> Even has joined #instantbird
10:42:59 * ChanServ sets mode +o Even 
11:05:14 --> jb has joined #instantbird
11:06:02 <-- Optimizer has quit (Ping timeout)
11:12:10 --> Optimizer has joined #instantbird
11:13:50 --> FeuerFliege has joined #instantbird
11:27:18 <-- jb has quit (Ping timeout)
11:39:19 <-- Even has quit (Input/output error)
11:42:27 --> Even has joined #instantbird
11:42:28 * ChanServ sets mode +o Even 
11:52:51 --> jb has joined #instantbird
11:54:42 <-- jb has quit (Ping timeout)
12:00:37 --> jb has joined #instantbird
12:33:18 --> rosonline has joined #instantbird
12:48:59 --> Mic|web has joined #instantbird
12:50:40 --> aleth has joined #instantbird
12:50:40 * ChanServ sets mode +h aleth 
12:51:01 <Mic|web> flo-retina: don't bother reviewing/commenting on what I pastebined yesterday night.
12:51:21 <Mic|web> gtg again
12:51:31 <-- Mic|web has quit (Quit: http://www.mibbit.com ajax IRC Client)
12:53:22 <-- aleth has quit (Quit: Au revoir)
12:53:31 --> aleth has joined #instantbird
12:53:32 * ChanServ sets mode +h aleth 
12:54:37 <aleth> flo: sorry about the tabs - my editor lost its some of its config during a distro switch
12:55:11 <flo-retina> it's ok, thanks for patching that so quickly :)
12:58:11 <aleth> Did you file a bug with the SASL auth failure log? I can't find it
12:58:27 <flo-retina> I don't think so
12:58:49 * flo-retina is trying to reach the hotline of his phone carrier.
12:59:06 <aleth> There's a couple things about it I'd like to discuss with clokep
13:02:11 * aleth wonders why IB is uglier on KDE than FF, and whether that will go away with one of the next few gecko updates
13:02:51 <aleth> The other possibility is something to do with FF being a 64b build
13:07:57 <flo-retina> aleth: I think the reason I didn't file it is I don't know what's wrong, and I just know I'll need clokep to look at the logs
13:08:28 <instantbot> New Core - IRC bug 1788 filed by aletheia2@fastmail.fm.
13:08:30 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1788 nor, --, ---, nobody, NEW, SASL auth fails when using a grouped nick (but not the first of the group)
13:08:35 <aleth> Well, here's the bug ;)
13:08:42 <flo-retina> aleth: it's ugly on KDE 64bit if you are using a 32bit build because some theming libraries aren't compiled in 32bit. It's the reason why Even compiles his own 64 bit builds for using on his kubuntu rather than using the nightlies
13:09:19 <aleth> flo-retina: OK, that explains it :)
13:10:09 <aleth> Hopefully one day we can offer 64b builds... maybe in the post-PPC era ;)
13:10:18 <flo> possibly, yes :)
13:10:28 <flo> we will change quite a few things in our infrastructure anyway
13:12:25 <aleth> flo-retina: Could an add-on add external JS (loaded from some URL) to IB conversations?
13:13:05 <flo> by "external" do you mean from the add-on or from the web?
13:13:12 <aleth> From the web.
13:14:17 <aleth> I would like to add MathJax support, but it will be a real pain if people have to install MathJax locally rather than just using the CDN
13:14:22 <flo> they would need to workaround a few security restrictions, but they can more or less do whatever they wan't
13:14:32 <flo> such an add-on shouldn't pass a review on AIO though
13:14:56 <-- FeuerFliege has quit (Ping timeout)
13:15:13 <aleth> That's the problem, I can't figure out a "legal" way to do it
13:17:56 <flo> what's mathjax exactly, and why can't you ship its js code in the add-on?
13:19:43 <aleth> http://www.mathjax.org/
13:20:03 <aleth> I suppose I could ship a whole MathJax installation with the add-on, that's true
13:20:18 <aleth> Then I'd have to keep it updated though ;)
14:11:51 <instantbot> aletheia2@fastmail.fm requested review from florian@instantbird .org for attachment 2079 on bug 1786.
14:11:53 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1786 nor, --, ---, aletheia2, ASSI, Don't log password during SASL auth
14:40:07 <aleth> Why is BIO indexed as a Google group? Is that intentional? https://groups.google.com/forum/?fromgroups=#!topic/instantbird-bugs/Z4aa2VFO1zE
14:47:05 <-- Kaishi has quit (Quit: Kaishi)
14:47:10 --> Kaishi has joined #instantbird
14:48:03 <-- Kaishi has quit (Quit: Kaishi)
14:48:09 --> Kaishi has joined #instantbird
14:54:27 <flo> aleth: is it a problem to you?
14:54:43 <aleth> No, I was merely surprised.
14:55:02 <aleth> Does anyone access BIO that way?
14:55:02 <flo> we made a mailing list (instantbird-bugs@ google groups) receiving all bug mail, so that people would be free to watch all bug changes without having to request it specifically
14:56:02 <aleth> Right.
14:56:02 <aleth> It's just odd to look for a SASL IRC spec and find the BIO bug the top search result ;)
14:56:08 <flo> I think clokep used it at some point, and then went back to being a global watcher, because in the bugmail that's sent directly to his account there are additional headers that can be used in Thunderbird for filtering (like: mailed to you because you are the assignee, etc...)
14:57:09 <flo> aleth: btw, nickserv gives us the name of the first nick of the group in the "password recognized" message, so we could cache it and use it again for the next attempts
14:57:53 <aleth> right... fastmail is good at threading them apart from review-xxx mails which somehow break its expectations
14:58:14 <aleth> flo: That's not much use to someone who can only login with SASL though
14:58:35 <flo> aleth: :-|
14:58:53 <flo> aleth: although if we cache it, it would be in a hidden account pref, so it would be possible to set that pref from about:config
14:59:09 <flo> if it turns out to be a very common situation we may want to add some UI later
14:59:21 <aleth> flo: That's true, that would be a possibility
14:59:59 <aleth> Is possible for the user to change the first grouped nick (via Nickserv)?
15:03:59 <flo> I don't know
15:04:09 <flo> aleth: but why would the user do that?
15:04:23 <aleth> Who knows why users do things? ;)
15:04:44 <flo> if you have grouped nicks, it's likely that you connect them with different computers, so they can't all be the first of the group
15:04:45 <aleth> How about just adding a hidden pref atm so if anyone runs into problems they can use that?
15:04:54 <aleth> We could take that for 1.3
15:05:36 <aleth> (without automatically setting that pref from the Nickserv message)
15:05:52 <flo> if you want
15:06:25 <flo> but I don't see that bug as being super urgent, so I think it can wait for post 1.3 and let clokep tell us what he thinks about it
15:06:59 <aleth> I agree.
15:09:58 <aleth> After all he's read all the spec and not just some snippets like me ;)
15:12:05 <flo> if you want to do something for 1.3, bug 1750 seems an easy target
15:12:08 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1750 min, --, ---, nobody, NEW, Warning: Use of Mutation Events is deprecated. Use MutationObserver instead.
15:13:39 <aleth> I was just looking at the IRC bug to make sure there was no further blocker hidden there.
15:14:04 <flo-retina> seems like a good idea :)
15:14:17 <flo-retina> thanks for checking :)
15:18:23 <flo-retina> the rate at which we receive keep-alive whitespace on the twitter stream seems consistent
15:18:34 <flo-retina> always between 29936 and 29938ms
15:18:44 <flo-retina> (ie almost 30s, but slightly less)
15:18:59 <flo-retina> ah, 30040 this time
15:19:01 <aleth> That's very precise ;)
15:19:53 <flo-retina> so I think it would make sense to make the twitter prpl restart the stream connection if we haven't received anything for 1 (or maybe 2 if we want to be really conservative) minute(s)
15:20:15 <flo-retina> but I would like to find the lowerlevel bug that makes us not notice when necko stops believing the http connection is alive
15:20:58 <flo> I'll put flo-retina's computer to sleep and wake it up several times, so I'm possibly more likely to reliably receive messages here
15:21:07 <aleth> Yes, I was logging the keep-alives too for a day a while back and it was very reliable, so for workaround it seems good
15:21:58 <-- flo-retina has quit (Ping timeout)
15:22:23 --> flo-retina has joined #instantbird
15:22:23 * ChanServ sets mode +qo flo-retina flo-retina 
15:24:06 <flo-retina> I'm surprised by http://i.minus.com/iDNUoaXyBuCLX.png
15:24:25 <flo-retina> not even sure what's the surprising part exactly
15:24:42 <flo-retina> but the messages "received" while the computer was sleeping arrived
15:24:56 <flo-retina> but displayed with "flo-retina" as not being in the room
15:25:36 <flo-retina> is it the router that kept resending the packets until the computer woke up?
15:25:40 <aleth> It does look odd...
15:27:22 <aleth> A race condition between removeBuddy and the message display...
15:28:00 <flo-retina> ah, messages are displayed asynchronously, right? :)
15:28:09 <aleth> Yup.
15:28:43 <flo-retina> so after putting the computer to sleep for 2 minutes, I no longer receive the keep-alive whitespace from twitter, but necko still thinks it has an alive connection
15:28:56 <aleth> At least you can reproduce :)
15:28:57 <flo-retina> (that doesn't match the behavior I observed the last time I tried to reproduce that bug)
15:29:04 <aleth> oh.
15:30:02 <flo-retina> well, yes
15:30:09 <flo-retina> but not in the way I reproduced last time
15:30:18 <flo-retina> I wonder if I should try putting the computer to sleep for longer
15:30:30 <aleth> Just not consistenly reproducible.
15:30:42 <flo-retina> (or just decide that hoping to have an event from necko is a dead end, and add a prpl-level timer)
15:30:59 <flo-retina> would be nice to fix the bug for XMPP too :(
15:31:34 <-- jb has quit (Ping timeout)
15:31:37 <aleth> Right... There were some gtalk bug reports that sounded suspiciously similar...
15:32:28 <aleth> When it happens, does netstat still see an alive connection?
15:32:52 <flo-retina> I haven't tried to use netstat
15:32:56 <flo-retina> (does it exist on Mac?)
15:33:02 <flo-retina> it does!
15:33:33 <flo-retina> any idea of which parameters I should give it to see something useful?
15:33:49 <aleth> No, it's too long since I last used it.
15:39:22 <flo> the machine is sleeping again
15:40:37 <-- flo-retina has quit (Ping timeout)
15:46:17 --> flo-retina has joined #instantbird
15:46:18 * ChanServ sets mode +qo flo-retina flo-retina 
15:47:45 <flo-retina> the twitter https connection is still active, and netstat still shows it as "ESTABLISHED"
15:52:31 * flo-retina isn't sure what http://mxr.mozilla.org/comm-central/source/mozilla/content/base/public/nsIXMLHttpRequest.idl#272 is
15:56:31 <flo> NickServ keeps talking to me
15:56:38 <flo> "16:46:32 - NickServ: fqueze!~Instantbi@til62-1-78-219-212-18.fbx.proxad.net has just authenticated as you (fqueze_)"
15:57:43 <aleth> Is that on a machine using a different nick in the same group?
15:57:57 <flo> it's on the machine using the main nick of the group
15:58:30 <aleth> So, it's telling you someone logged into the group using your password ;)
15:58:37 <flo> (I'm not sure if it's only delivered to the machine that has the main nick connected, or if it would be to any other nick of the group)
15:58:39 <-- rosonline has quit (Quit: Experimentem Instantbird 1.2 -- http://www.instantbird.com)
15:58:44 <flo> aleth: right
15:58:50 <aleth> I think the first group counts as the authorizing identity
15:58:58 <flo> aleth: and I'm afraid it's also telling me that I need to update the nickserv killer add-on :(
15:58:58 <aleth> s/first group/first group nick
15:59:40 <aleth> flo: either that or come up with a way to better handle grouped nicks ;)
16:02:25 <-- Optimizer has quit (Quit: Instantbird 1.3a1pre -- http://www.instantbird.com)
16:02:39 <flo-retina> so that timeout of XHR doesn't seem of any use for chunked data
16:02:58 <flo-retina> it's a timeout from the start of the request, not from the last time we received data
16:06:31 <flo-retina> the timeout on the XMPP socket seems to be of 18*hours*
16:07:44 --> jb has joined #instantbird
16:07:50 <flo-retina> (2^16 seconds)
16:11:45 --> Mnyromyr has joined #instantbird
16:12:26 <flo-retina> bah, infinite timeout for read/write operations after the connection has been successfully established: http://mxr.mozilla.org/comm-central/source/chat/protocols/xmpp/xmpp-session.jsm#69 :(
16:14:54 --> Kaishi1 has joined #instantbird
16:15:48 <-- Kaishi has quit (Ping timeout)
16:16:36 <-- Kaishi1 has quit (Quit: Kaishi1)
16:21:50 <-- jb has quit (Ping timeout)
16:22:33 <flo-retina> how does 5 minutes sound as a readWriteTimer for JS-XMPP?
16:22:42 <flo-retina> s/Timer/Timeout/
16:25:33 <aleth> reasonable, unless that means we reconnect every 5 minutes when idle?
16:26:23 <-- flo-retina has quit (Ping timeout)
16:29:37 <flo> if we want to be a bit more clever, we can send a ping when we think we are disconnected (http://xmpp.org/extensions/xep-0199.html)
16:38:07 --> Kaishi has joined #instantbird
16:40:56 --> flo-retina has joined #instantbird
16:40:56 * ChanServ sets mode +qo flo-retina flo-retina 
16:41:39 <flo-retina> after waking from sleep, the "connected from a few seconds" of the account manager is completely wrong
16:41:52 <flo-retina> and it's strange to see that jump directly to "connected for about 16 minutes"
16:47:59 <-- flo-retina has quit (Ping timeout)
16:48:20 --> flo-retina has joined #instantbird
16:48:20 * ChanServ sets mode +qo flo-retina flo-retina 
17:04:41 <flo-retina> so I just looked at what libpurple does
17:04:56 <flo-retina> it sends a ping every 30s if nothing has been received on the XMPP stream
17:05:07 --> meh has joined #instantbird
17:05:18 <flo-retina> and it disconnects with a ping timeout 120s after sending the ping if it hasn't received a pong.
17:06:01 <flo-retina> it's approximately the values I wanted to use, but I was considering sending a ping only every 2 minutes, and disconnecting faster (20-30s) after sending a ping when not receiving any reply.
17:22:03 --> wnayes has joined #instantbird
17:43:52 <-- flo has quit (Ping timeout)
17:43:57 <-- flo-retina has quit (Ping timeout)
17:47:27 --> flo-retina has joined #instantbird
17:47:28 * ChanServ sets mode +qo flo-retina flo-retina 
18:01:46 --> jb has joined #instantbird
18:02:21 --> jb1 has joined #instantbird
18:03:18 --> jb2 has joined #instantbird
18:03:38 <-- jb has quit (Ping timeout)
18:03:41 <-- jb1 has quit (Connection reset by peer)
18:25:49 --> mconley has joined #instantbird
18:26:00 --> flo has joined #instantbird
18:26:01 * ChanServ sets mode +qo flo flo 
18:26:38 <flo-retina> ahah, "19:26:06 - SaslServ: fqueze_!~Instantbi@ks368640.kimsufi.com has just authenticated as you (fqueze_)"
18:26:55 <flo-retina> so NickServ is no longer the only noisy "server"
18:44:23 <flo-retina> I would need a localized string for the twitter error message :(
18:45:45 <flo-retina> bah, I'll just reopen the stream instead of disconnecting the account
18:59:50 <-- flo-retina has quit (Ping timeout)
19:03:09 --> flo-retina has joined #instantbird
19:03:09 * ChanServ sets mode +qo flo-retina flo-retina 
19:11:58 * flo should have a patch for timeouts on twitter soon
19:12:26 <-- mconley has quit (Ping timeout)
19:13:02 <-- flo-retina has quit (Ping timeout)
19:16:08 <-- jb2 has quit (Ping timeout)
19:16:33 --> jb has joined #instantbird
19:16:52 --> flo-retina has joined #instantbird
19:16:53 * ChanServ sets mode +qo flo-retina flo-retina 
19:27:57 <aleth> flo++ :)
19:33:19 <instantbot> florian@instantbird.org requested review from aletheia2@fastmail. fm for attachment 2080 on bug 1613.
19:33:20 <flo-retina> testing that kind of patches takes a long time :(
19:33:21 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1613 nor, --, ---, florian, REOP, Twitter feed stops receiving incoming tweets (without disconnecting)
19:37:11 <instantbot> New Core - XMPP bug 1789 filed by florian@instantbird.org.
19:37:13 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1789 maj, --, ---, nobody, NEW, Google Talk (JS-XMPP) accounts should timeout when the connection to the server has stalled
19:37:22 <-- jb has quit (Ping timeout)
19:37:37 --> jb has joined #instantbird
19:38:04 <flo-retina> aleth: what would you think of using "base64 encoded password, not logged" instead of just "password" for bug 1786?
19:38:07 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1786 nor, --, ---, aletheia2, ASSI, Don't log password during SASL auth
19:39:25 <aleth> flo-retina: we use <password> everywhere else in the code when we send the password
19:39:46 <aleth> Feel free to change it though if you think it's clearer.
19:40:54 <flo-retina> aleth: http://lxr.instantbird.org/instantbird/search?string=not%20logg
19:41:14 <aleth> I stand corrected.
19:41:43 <aleth> Definitely change it then :D
19:41:57 <flo-retina> ok
19:42:00 <flo-retina> the patch is fine otherwise :)
19:42:15 <flo-retina> (there's not much I could want to change outside of that string anyway :-D)
19:42:28 <aleth> not really ;)
19:54:04 --> mconley has joined #instantbird
19:58:48 <-- jb has quit (Ping timeout)
20:02:16 --> jb has joined #instantbird
20:04:52 <-- jb has quit (Ping timeout)
20:05:17 <instantbot> aletheia2@fastmail.fm granted review for attachment 2080 on bug 1613.
20:05:19 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1613 nor, --, ---, florian, REOP, Twitter feed stops receiving incoming tweets (without disconnecting)
20:05:47 --> jb has joined #instantbird
20:06:28 <-- wnayes has quit (Quit: Instantbird 1.3a1pre -- http://www.instantbird.com)
20:06:29 <aleth> Thanks for dealing with that hard-to-track-down bug flo :)
20:08:19 <-- jb has quit (Ping timeout)
20:11:48 --> jb has joined #instantbird
20:14:35 <-- jb has quit (Ping timeout)
20:19:23 --> jb has joined #instantbird
20:22:17 <-- jb has quit (Ping timeout)
20:26:02 --> jb has joined #instantbird
20:29:02 <-- jb has quit (Ping timeout)
20:32:20 --> jb has joined #instantbird
20:35:27 <-- jb has quit (Ping timeout)
20:37:22 <-- aleth has quit (Quit: Au revoir)
20:37:25 --> aleth has joined #instantbird
20:37:25 * ChanServ sets mode +h aleth 
20:38:56 --> jb has joined #instantbird
20:42:07 <-- jb has quit (Ping timeout)
20:45:32 --> jb has joined #instantbird
20:53:00 <-- jb has quit (Ping timeout)
20:56:46 --> jb has joined #instantbird
21:00:19 <-- jb has quit (Ping timeout)
21:02:40 --> jb has joined #instantbird
21:06:46 <-- jb has quit (Ping timeout)
21:11:49 --> jb has joined #instantbird
21:15:39 <-- jb has quit (Ping timeout)
21:18:51 --> jb has joined #instantbird
21:19:40 <-- aleth has quit (Quit: Au revoir)
21:19:43 --> aleth has joined #instantbird
21:19:43 * ChanServ sets mode +h aleth 
21:23:01 <-- jb has quit (Ping timeout)
21:27:40 --> jb has joined #instantbird
21:38:08 <-- jb has quit (Ping timeout)
21:42:13 --> jb has joined #instantbird
21:44:38 <flo-retina> I'm afraid we will lose tweets. I thought mostly about the case reported in the bug (ie "The feed stops receiving incoming tweets because the server dropped it"), but if the reason why the connection is dead is that the machine has been sleeping for a few hours, just reopening the stream won't get the tweets that were posted while the machine was sleeping
21:45:24 <flo-retina> so in that case it will work only if necko went offline before coming back online (likely if we are switching from a wifi network to another)
21:46:42 <-- jb has quit (Ping timeout)
21:47:22 <aleth> Hmm... that's the part I didn't understand: what causes the bug, ie how does the twitter server know the machine is hibernating? Do we usually ping back?
21:49:07 <flo-retina> pff, there's a spec
21:49:13 <flo-retina> https://dev.twitter.com/docs/streaming-apis/connecting
21:49:25 <flo-retina> "Set a timer, either a 90 second TCP level socket timeout, or a 90 second application level timer on the receipt of new data. If 90 seconds pass with no data received, including newlines, disconnect and reconnect immediately according to the backoff strategies in the next section."
21:49:55 <aleth> heh :D
21:50:06 <aleth> That would have saved a lot of time, finding that earlier...
21:50:07 <flo-retina> "pff" meant "how come I missed that" in this case ;)
21:50:22 <flo-retina> aleth: I'm sure I read that at least 3 times when I implemented the twitter code ;)
21:50:50 --> jb has joined #instantbird
21:51:16 <aleth> Btw why don't we simply disconnect and reconnect, without a new error message string? We just get the usual disconnected/reconnected system messages, no harm in that.
21:53:10 <aleth> Plus we then already have the "If the reconnect fails, slow down your reconnect attempts..." logic
21:53:52 --> wnayes has joined #instantbird
21:53:59 <-- wnayes has quit (Quit: wnayes)
21:54:00 --> wnayes has joined #instantbird
21:54:08 <flo-retina> so the timeout should be 90s
21:54:19 <flo-retina> and I think we should really disconnect the account and do a full reconnect
21:54:21 <aleth> 120s was not too far off ;)
21:54:35 <aleth> It's definitely what the spec calls for.
21:54:45 <aleth> Unless I am missing something.
21:55:08 <flo-retina> the spec only talks about the case where the connection was closed on the twitter side
21:55:25 <flo-retina> I'm more concerned by the connection being closed on our side (laptop sleeping, ...)
21:55:38 <aleth> I thought you had ruled that out (netstat ESTABLISHED)
21:55:55 <aleth> or is that secretly reestablished?
21:56:26 <flo-retina> well. I expressed myself poorly. :-)
21:56:44 <flo-retina> we have no timeout on the connection
21:57:07 <flo-retina> and the IP packet telling us the server has closed the connection "arrived" while the computer was asleep, so it never really arrived
21:57:08 <-- jb has quit (Ping timeout)
21:57:20 <aleth> I was assuming based on this page that what happened was "Twitter will close a streaming connection for the following reasons: ... A client reads data too slowly. Every streaming connection is backed by a queue of messages to be sent to the client. If this queue grows too large over time, the connection will be closed."
21:57:26 <flo-retina> so our computer keeps thinking the connection is "established", just because it's never been closed
21:57:39 --> jb has joined #instantbird
21:58:11 <aleth> That makes sense.
21:58:45 <flo-retina> aleth: btw, the only part of that spec that isn't currently implemented is "Set a timer, either a 90 second TCP level socket timeout, or a 90 second application level timer on the receipt of new data. If 90 seconds pass with no data received, including newlines, disconnect"
21:58:47 <aleth> So the twitter server decides to close the connection because we aren't picking up, and we don't notice.
21:58:49 <flo-retina> everything else is
21:59:02 <flo-retina> when we know the twitter server has closed the connection, we behave correctly
21:59:20 <aleth> :)
22:01:57 <-- jb has quit (Ping timeout)
22:04:03 <flo-retina> aleth: so I really need an error string
22:04:29 <aleth> flo-retina: To replace the disconnect/reconnect pair?
22:04:59 --> jb has joined #instantbird
22:05:03 <flo-retina> to go with the disconnect call
22:05:27 <flo-retina> it will hopefully never be displayed more than one second
22:05:57 <aleth> Ah, for the account manager. Sorry, was being slow.
22:06:31 <flo-retina> I think I'll just use "timeout"
22:06:52 <aleth> I think it's more important the bug is fixed for 1.3 than that anyone watching the account manager is potentially puzzled for a minute
22:07:06 <flo-retina> we already have http://mxr.mozilla.org/comm-central/source/mail/base/modules/http.jsm#31 that's not localized and sometimes appears at that place
22:07:17 <flo-retina> s/minute/second/ ;)
22:07:28 <flo-retina> but I agree, not losing any data is more important
22:10:19 <-- jb has quit (Ping timeout)
22:10:23 <aleth> Maybe there is a vague "connection failed" string somewhere that can be reused
22:10:52 <flo-retina> aleth: here's the interdiff: http://pastebin.instantbird.com/100602
22:11:19 <flo-retina> I'll file a bug to get "offline" and "timeout" replaced for 1.4
22:11:35 <aleth> Great :)
22:11:47 <flo-retina> the interdiff is reversed
22:11:49 <aleth> That sounds like less hassle than trying to send tweets that go missing
22:11:50 <flo-retina> sorry about that
22:11:55 <aleth> np
22:12:22 <flo-retina> is there anything to change, or should I attach that new patch and commit?
22:12:26 <flo-retina> I may need to test again :-/
22:13:43 * flo-retina hopes to have all the blockers resolved tonight
22:13:58 <aleth> Testing it would be good...
22:14:13 <aleth> But it looks good to me.
22:14:17 <flo-retina> I intend to do a first almost-release-candidate tomorrow to check that the l10n repackaging build system code hasn't been broken by the mozilla updates / fix it.
22:14:36 <flo-retina> I'll be quite busy next week, so I won't have time to deal with that mess if it happens again. I prefer dealing with it during the week-end.
22:14:41 <aleth> What do the l10n status pages look like?
22:15:01 --> jb has joined #instantbird
22:15:06 <aleth> I suppose it makes more sense to check those after the weekend though
22:15:18 <flo-retina> cs, de, es-ES, it, ru, sk, sv-SE are ready
22:16:14 <flo-retina> et and pt-BR aren't fully translated, but seem mostly shippable
22:16:37 <aleth> Ah, this._streamingrequest is deleted in cleanUp().
22:16:43 <flo-retina> fr, nl, pl, uk really need an update
22:17:20 <flo-retina> Tonnes: ping :) (just checking that you have received the email announcing the string freeze for 1.3)
22:17:38 <flo-retina> aleth: we need to find new localizers for pl
22:17:56 <aleth> fr seems kind of important too ;)
22:18:03 <flo-retina> bah...
22:18:23 <flo-retina> fr seems like I'll be very upset if I have to do it myself the day before the release ;)
22:18:58 <flo-retina> there are only 44 strings that need to be localized, so it shouldn't be a terribly long job...
22:19:12 <aleth> Find an IB-using officemate? ;)
22:19:47 <flo-retina> I'm thinking of just asking in #frenchmoz who wants to take over that localization
22:20:12 <flo-retina> and I think we should really do the same to find one or more new people for the pl team
22:20:59 <-- jb has quit (Ping timeout)
22:21:12 <flo-retina> I don't remember the situation for uk, but 1.2 shipped in uk, so I assume there's an active or at least mostly active localizer. A gentle ping may be a good idea, if only to ensure he has received the email about the string freeze
22:22:49 <aleth> There should be enough mozillians in fr and pl...
22:24:00 * flo-retina has just pinged the current fr localizer by email
22:26:09 --> jb has joined #instantbird
22:28:03 <flo> testing the twitter patch again
22:29:28 <-- flo-retina has quit (Ping timeout)
22:31:00 <-- jb has quit (Ping timeout)
22:32:50 --> flo-retina has joined #instantbird
22:32:50 * ChanServ sets mode +qo flo-retina flo-retina 
22:33:06 <flo-retina> aleth: there's something strange in the position of that unread ruler: http://i.minus.com/ibnUnJxngFRc5w.png
22:33:48 <flo-retina> ah, I know what's causing that
22:33:53 --> jb has joined #instantbird
22:34:02 <flo-retina> the timestamp of the ABC tweet is < the timestamp of the system message
22:34:32 <flo-retina> so the timebubbles code gives up trying to adapt the interval
22:35:17 <aleth> :(
22:35:42 <aleth> So it maybe goes wrong just after reconnects...
22:35:46 <aleth> I'll have to test that
22:38:55 <-- jb has quit (Ping timeout)
22:39:29 <instantbot> florian@instantbird.org requested review from aletheia2@fastmail. fm for attachment 2081 on bug 1613.
22:39:31 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1613 nor, --, ---, florian, REOP, Twitter feed stops receiving incoming tweets (without disconnecting)
22:39:52 <flo-retina> aleth: it's not a problem related to the unread ruler, it's just that time bubbles sucks when we receive "offline messages"
22:40:10 <aleth> Yeah, but I dimly remember fixing that, or something related to it
22:42:12 <instantbot> aletheia2@fastmail.fm granted review for attachment 2081 on bug 1613.
22:42:14 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1613 nor, --, ---, florian, REOP, Twitter feed stops receiving incoming tweets (without disconnecting)
22:42:42 * aleth dislikes looking at Footer.html
22:44:36 --> jb has joined #instantbird
22:45:30 <flo-retina> aleth: well, what do you think would be the correct "expected behavior"?
22:46:10 <aleth> For the gap between messages in that case?
22:46:22 <flo-retina> what would you want in that case?
22:46:30 <aleth> A centered unread ruler ;)
22:46:37 <flo-retina> ah!
22:46:58 <aleth> The fact that it's not centered means something is not being set correctly, as you say
22:47:07 <flo-retina> I'm not sure that's worth the effort for that corner case
22:47:43 <aleth> Probably not, it just bugs me as I thought I had fixed the code for out-of-time-order messages
22:48:19 <aleth> Seems like there's always one more edge case ;)
22:50:34 <flo-retina> aleth: this is what I'm about to commit: http://pastebin.instantbird.com/100645
22:51:00 <aleth> Looks good.
22:51:04 <flo-retina> aleth: it may be different when the previous message is a system message
22:51:17 <flo-retina> (I didn't know why there was an empty line before the return true so I removed it)
22:51:39 <aleth> clokep sometimes likes extra empty lines.
22:52:31 <Tonnes> flo-retina: greetings
22:52:39 <flo-retina> Tonnes: hello :)
22:52:47 <-- jb has quit (Ping timeout)
22:53:24 <Tonnes> just checked the mail but I'm afraid I don;t see it yet..
22:53:44 <Tonnes> checking spam folder
22:53:52 <instantbot> florian@instantbird.org granted review for attachment 2079 on bug 1786.
22:53:54 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1786 nor, --, ---, aletheia2, ASSI, Don't log password during SASL auth
22:54:02 <flo-retina> Tonnes: it was sent 3 days ago
22:54:19 <flo-retina> Tonnes: it said http://pastebin.instantbird.com/100648
22:54:47 <flo-retina> the subject line is "Instantbird 1.3 string frozen" (if that helps finding it in the spam)
22:56:04 <flo-retina> aleth: I think clokep will have a lot of bugmail when getting home/back online ;).
22:56:07 <Tonnes> weird.. neither the spam or the normal flder appear to have it
22:56:29 <flo-retina> well, at least now you know
22:56:31 <flo-retina> :)
22:57:14 <flo-retina> Tonnes: have you received correctly the previous messages sent to the instantbird-l10n list? (the previous email was several months ago; we purposefully keep that mailing list very low traffic)
22:58:26 <Tonnes> the last message I have is from july 5th, is that correct?
22:58:43 <Tonnes> ..though that is sent to the personal address, not the list
22:59:08 <flo-retina> hmm, I wonder if it's possible that the invitation to the list bounced and has now expired
22:59:11 <flo-retina> I don't see you on that list
22:59:27 <Tonnes> exactly.. I don;t recall subscribing..
22:59:41 <flo-retina> ah, you are in the "pending invitations" list
22:59:45 <flo-retina> and your invitation expired
22:59:53 <Tonnes> aargh
23:00:05 <flo-retina> the invitation was sent on "01/10/11"
23:00:13 <Tonnes> when was the invitation sent?
23:00:16 <Tonnes> ah
23:00:22 <Tonnes> wow
23:00:47 <Tonnes> ah, I see that one
23:00:53 <flo-retina> Tonnes: I resent it
23:01:22 <Tonnes> got it
23:01:54 <flo-retina> and I have 6 more people with expired invitations :-(
23:02:52 <flo-retina> Tonnes: anyway, your locale was up to date for 1.2, so there are only 44 untranslated strings.
23:03:05 <Tonnes> it was marked as read but must either have slipped my mind or I didn;t know it was really necessary.. sorry..
23:03:27 <Tonnes> I just have to click it, right?
23:03:37 <flo-retina> click what?
23:03:45 <Tonnes> the url :)
23:03:56 <flo-retina> I don't kno
23:03:57 <flo-retina> w
23:04:10 <flo-retina> I assumed the google group invitation emails were self-descriptive
23:05:13 <Tonnes> obvious indeed :) - just accepted it
23:05:34 <Tonnes> however, it is not accepted
23:06:16 <Tonnes> unable to accept group invitation, Reason: you are not authorized to accept the invitation
23:06:58 <Tonnes> should I be logged in to Google?
23:07:10 <flo-retina> possibly
23:07:31 <Tonnes> ok, it's accepted
23:07:33 <flo-retina> maybe because it's on an @gmail address?
23:08:04 <flo-retina> all the other translators in the "pending invitations" list seem inactive, so not much damage there
23:08:06 <Tonnes> needed to be logged in allright
23:08:51 --> jb has joined #instantbird
23:10:42 <Tonnes> flo-retina: sorry again, I will likely update them tomorrow, it's been a tough day..
23:11:00 <flo-retina> Tonnes: it's ok, the deadline isn't today :)
23:11:08 <flo-retina> Tonnes: I'm just glad I pinged you before it was too late :)
23:11:18 <Tonnes> me too :)
23:12:54 <flo-retina> still need to figure out what's up with the uk localizer, and find new people for pl
23:13:49 <-- jb has quit (Ping timeout)
23:14:25 <Tonnes> ah
23:15:15 <Tonnes> did you check verbatim or sumo contributors?
23:16:07 <flo-retina> Tonnes: I thought we could ask in the IRC channel of the polish community.
23:16:29 <flo-retina> I assume the current pl localization is abandoned because it's not been updated for 1.2.
23:17:05 <flo-retina> The uk locale was updated for 1.2, so it's possible the translator just hasn't received the email (like in your case; although he's definitely in the mailing list :))
23:17:05 <-- aleth has quit (Input/output error)
23:17:11 <Tonnes> if there is such a channel, I would try so indeed.. 
23:18:11 <Tonnes> you probably know of https://wiki.mozilla.org/L10n:Teams:pl
23:19:18 <Tonnes> Marek (top) is likely the most active one, hence most able to pick it up
23:21:24 <flo-retina> I can try.
23:21:45 <flo-retina> Or maybe the person translating Thunderbird, as we have lots of strings in common now that chat/ ships in Thunderbird.
23:22:34 <flo-retina> "the e-mail address to the whole team: mozilla at aviary.pl " could also be a way to let them decide ;)
23:22:52 <Tonnes> exactly.. :)
23:23:33 <Tonnes> I really need some rest now, I'll ping you when updated (though you'll probably notice it). Thanks again for the update
23:23:52 <flo-retina> I don't really need a ping, except if you have questions
23:23:52 --> jb has joined #instantbird
23:23:57 <flo-retina> thanks, good night! :)
23:24:10 <Tonnes> ok , good night..
23:27:22 <-- jb has quit (Max SendQ exceeded)
23:27:26 --> BlueMaxima has joined #instantbird
23:27:35 --> gerard-majax has joined #instantbird
23:29:43 --> jb has joined #instantbird
23:33:59 * flo-retina has just discovered that "hg ci <filename>" works even if the file hasn't been hg add'ed before invoking hg commit.
23:34:22 <-- jb has quit (Ping timeout)
23:38:41 <-- mconley has quit (Input/output error)
23:43:17 * flo-retina emailed a ping to the uk translator
23:43:59 <flo-retina> and pushed a few updates to the pl repository with the strings I could import from other places, to have a more accurate count of strings that actually need to be translated into pl, before I email the aviary.pl team.
23:44:55 <-- Mnyromyr has quit (Quit: ChatZilla 0.9.86 [SeaMonkey 1.1.19/2010030105])
23:45:46 <flo-retina> ok, 89 strings need to be translated to pl. That's not too horrible :).
23:48:02 <instantbot> florian@instantbird.org set the Resolution field on bug 1786 to FIXED.
23:48:04 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1786 nor, --, 1.3, aletheia2, RESO FIXED, Don't log password during SASL auth
23:48:22 <instantbot> florian@instantbird.org set the Resolution field on bug 1613 to FIXED.
23:48:24 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1613 nor, --, 1.3, florian, RESO FIXED, Twitter feed stops receiving incoming tweets (without disconnecting)
23:49:07 <BlueMaxima> oh yay it's fixed!
23:52:21 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/1566e1bf0c8b - Florian Quèze - Bug 1613 - Twitter feed stops receiving incoming tweets (without disconnecting), r=aleth.
23:52:22 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/ec702866871c - aleth - Bug 1786 - Don't log password during SASL auth, r=fqueze.
23:58:50 <flo-retina> and now (while writing the email for aviary.pl) I'm wondering if it's possible that the pl translators never received the message about the 1.2 string freeze :-S