#instantbird log on 07 07 2014

All times are UTC.

00:00:31 <-- Hadi has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
00:29:22 --> rosonline has joined #instantbird
00:46:16 <-- Tonnes has quit (Quit: ChatZilla [Firefox 30.0/20140605174243])
00:52:51 <-- Rym has quit (Ping timeout)
00:53:14 --> Rym has joined #instantbird
01:17:27 <-- Rym has quit (Ping timeout)
01:18:00 --> Rym has joined #instantbird
01:50:46 <-- micahg has quit (Input/output error)
01:53:49 --> micahg has joined #instantbird
02:07:21 <-- Armada has quit (Connection reset by peer)
02:20:11 --> mconley has joined #instantbird
02:29:39 <-- rosonline has quit (Quit: Instantbird 1.5 -- http://www.instantbird.com)
02:54:19 <-- Rym has quit (Ping timeout)
03:13:36 <-- mconley has quit (Input/output error)
03:14:36 <-- EionRobb has quit (Ping timeout)
03:16:20 --> EionRobb has joined #instantbird
03:42:46 --> mayanktg has joined #instantbird
04:50:42 --> gerard-majax_ has joined #instantbird
04:56:08 <-- gerard-majax_ has quit (Ping timeout)
04:57:12 <-- mayanktg has quit (Ping timeout)
05:20:03 <-- EionRobb has quit (Quit: Leaving.)
05:32:31 --> mayanktg has joined #instantbird
06:06:37 <-- mayanktg has quit (Ping timeout)
06:23:06 <sawrubh> howdy folks
06:38:52 * Fallen|away is now known as Fallen
06:57:13 --> mayanktg has joined #instantbird
07:04:22 --> mayanktg-ph has joined #instantbird
07:07:36 --> flo-retina has joined #instantbird
07:07:37 * ChanServ sets mode +qo flo-retina flo-retina 
07:08:41 <flo-retina> nhnt11: are system messages (eg. "<contact name> is now Available") expected to be logged?
07:32:12 --> Mic|web has joined #instantbird
07:36:41 <-- mayanktg has quit (Ping timeout)
07:39:36 <Mic|web> Didn't we have a bug about logging joins/leaves/status changes in a different way? (i.e. not as text in a system messages but as some sort of event that happened in a chat?)
07:41:45 <flo-retina> Mic|web: sure.
07:42:03 <flo-retina> Mic|web: but I thought until that we were logging them in plain text as system messages
07:42:15 <flo-retina> btw, anybody's got any idea about the cause of the current mac packaging bustage?
07:43:11 <Mic|web> Sure, I wasn't suggesting scope creep, it was only a question when I saw the example for the system message.
07:43:45 <flo-retina> Mic|web: my question is because a private conversation I had yesterday (that's still open in a tab) doesn't have any system message when viewed in the log viewer
07:44:15 <Mic|web> Ah
07:45:05 --> jb has joined #instantbird
07:54:14 --> gerard-majax_ has joined #instantbird
07:56:43 <nhnt11> flo-retina: They're not, because they don't fire a new-text notification
07:57:32 <nhnt11> or they do, but they have the noLog flag, but I highly doubt that. Let me check really quickly..
07:57:56 --> chrisccoulson has joined #instantbird
08:13:49 <flo-retina> hmm, looks like it's not new anyway. I was just concerned it might be a recent regression
08:14:24 <flo-retina> but we may still want to start logging them in some way, otherwise when discarding messages automatically and redisplaying them with infinite scroll, it may look strange to not have them :-S
08:15:44 <flo-retina> nhnt11: or maybe these messages are added in the imIConversation rather than in the prplIConversation?
08:19:44 <-- flo-retina has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
08:26:34 <nhnt11> flo-retina: That's interesting to think about...
08:26:38 * nhnt11 ponders
08:31:35 <nhnt11> flo-retina: So, they have the noLog flag set.
08:39:04 <-- Mic|web has quit (Quit: http://www.mibbit.com ajax IRC Client)
08:45:14 --> flo-retina has joined #instantbird
08:45:14 * ChanServ sets mode +qo flo-retina flo-retina 
09:06:00 --> sonny1 has joined #instantbird
09:06:09 <-- sonny1 has quit (Client exited)
09:11:22 <-- Even has quit (Ping timeout)
09:20:54 --> Even has joined #instantbird
09:20:54 * ChanServ sets mode +o Even 
09:36:01 --> mpmc has joined #instantbird
09:58:09 --> arlolra has joined #instantbird
10:14:24 <-- arlolra has quit (Quit: arlolra)
10:14:31 --> arlolra has joined #instantbird
10:29:36 --> BWMerlin has joined #instantbird
10:36:24 * Fallen is now known as Fallen|away
10:39:34 <-- nhnt11 has quit (Ping timeout)
10:39:37 --> nhnt11 has joined #instantbird
10:41:56 <flo-retina> nhnt11, sawrubh, mayanktg-ph: so what are you all working on today/this week? :)
10:42:51 --> mayanktg has joined #instantbird
10:43:43 <nhnt11> flo-retina: So last week was pretty terrible, sorry for that. Today I'm going to look at the twitter code to get delayed messages working, fix up bug 1034967, and start looking at the indexing patch again.
10:43:47 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1034967 nor, --, ---, nhnt11, ASSI, Regression: Context menu broken over blank space in contact list
10:44:07 <nhnt11> I'm going to write a blog post first
10:44:13 * nhnt11 is eating at the moment
10:44:34 <-- mayanktg has quit (Connection reset by peer)
10:45:29 --> mayanktg has joined #instantbird
10:45:30 <sawrubh> flo-retina: today (in an hour I plan to send a review request) having taken all comments into account, then move on to integrating the Downloads panel (which should be done by end of this week, I'm just being conservative given how much time it took for FileLinkt to be integrated), after that just writing the purplexpcom wrapper should be left in my project
10:45:46 <sawrubh> flo-retina: given clokep's on a holiday, should I send it to you instead?
10:46:03 <flo-retina> I can certainly give early feedback
10:46:31 <flo-retina> " just writing the purplexpcom wrapper should be left in my project" I really doubt that.
10:46:49 <flo-retina> unless there's much more in the patches already attached than I thought there was.
10:47:54 <flo-retina> sawrubh: Is http://xmpp.org/extensions/xep-0065.html implemented already?
10:48:26 <flo-retina> per http://xmpp.org/extensions/xep-0096.html#protocol-tech both SOCKS5 Bytestreams (XEP-0065) and In-Band Bytestreams (XEP-0047) are required to claim we support XMPP file transfer.
10:50:09 <sawrubh> so after today's patch and Downloads panel (which I aim to finish by end of week) purplexpcom wrapper is all that's left (according to https://etherpad.mozilla.org/ib-filelink since I think deciding between which method to choose would be a heuristic)
10:50:33 <sawrubh> flo-retina: I remember clokep saying that we shouldn't think about implementing SOCKS5 Bytestream
10:50:35 * sawrubh checks logs
10:50:54 <flo-retina> He probably said that you shouldn't do that before we have any UI.
10:55:54 <sawrubh> bah, can't find it in the logs
10:56:02 <sawrubh> ok, will implement it then
10:56:10 <flo-retina> doesn't change that it has to be done if we want to support file transfer over XMPP
10:56:23 <flo-retina> what's the UI you are implementing?
10:56:40 <sawrubh> we decided on drag and drop for initiating and downloads panel for feedback about progress
10:57:02 <sawrubh> right now for landing it's a notification bar
10:57:21 <flo-retina> ok, so that's just the first steps, right?
10:58:10 <sawrubh> notification bar for feedback is a first step, yes. We plan on implementing downloads Panel in the final thing
10:58:28 <sawrubh> but we were final on drag and drop aiui
10:58:40 <flo-retina> final on it being a first step, right? :-P
10:59:00 <flo-retina> there's no way any real user would discover the feature that way ;)
10:59:02 <sawrubh> do you want a file picker instead of drag and drop in the final version?
10:59:12 <flo-retina> not instead; in addition to! :)
10:59:16 <sawrubh> ah
10:59:25 <sawrubh> ok, will look at that too then :)
10:59:43 <sawrubh> shouldn't be difficult (hopefully)
10:59:45 <flo-retina> there has to be some visible UI
11:00:00 <flo-retina> I also don't think drag&drop is usable at all by blind users.
11:00:41 <sawrubh> so I had thought of a paper clip icon in the textbox for the visible UI
11:00:59 <sawrubh> but would that be found by the accessibility feature used by blind users
11:01:06 <sawrubh> or do we need a menu item for this?
11:02:33 <flo-retina> I was thinking you would put an icon at the top of the conversation UI, using bug 1004930
11:02:36 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1004930 enh, --, ---, mayanktg, ASSI, Generic way to add buttons for actions to a conversation
11:02:43 <flo-retina> and the download panel would be attached to that icon
11:02:47 <-- mayanktg has quit (Ping timeout)
11:03:07 <flo-retina> you could likely add a "send file" button/action somewhere inside that panel
11:04:19 * sawrubh checks the bug
11:06:22 <sawrubh> wow, so mayanktg-ph already prototyped about 'Transfer file' in one of the screenshots
11:06:56 <-- jb has quit (Ping timeout)
11:13:30 <sawrubh> flo-retina: now that we're already talking about UI, where do you want me to add the downward pointing arrow clicking which opens the downloads panel?
11:13:55 --> jb has joined #instantbird
11:14:04 <sawrubh> or "and the download panel would be attached to that icon" means that icon mayanktg-ph added will open the panle
11:14:19 <flo-retina> mayanktg didn't add any icon
11:14:31 <flo-retina> he's just giving you a place where it's trivial to add one
11:15:17 * Fallen|away is now known as Fallen
11:16:57 <sawrubh> ok
11:18:00 <sawrubh> flo-retina: ah so I see, you want me add an icon there, clicking which would open the Downloads Panel, inside the Panel, there'll be an option to send file?
11:18:11 <flo-retina> yes
11:19:29 <sawrubh> I'm just a bit wary of that extra (navigation and) selection to actually being able to send a file (I mean first Opening the panel and sending the file rather than directly clicking on that icon which would open a picker and you could send a file)
11:19:48 <sawrubh> do you think it's a wee bit bad UX?
11:20:33 <flo-retina> what's the problem?
11:21:03 <flo-retina> If you are just worried about the extra click... the time it takes to click depends significantly on the size of the target, so the second click shouldn't take too long
11:21:16 --> mayanktg has joined #instantbird
11:29:49 <instantbot> New Chat Core - XMPP bug 1035132 filed by saurabhanandiit@gmail.com.
11:29:50 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1035132 nor, --, ---, nobody, NEW, Implement SOCKS5 Bytestreams in XMPP (XEP-0065)
11:39:24 <-- mayanktg has quit (Ping timeout)
11:39:27 --> mayanktg has joined #instantbird
11:42:00 <-- mayanktg has quit (Ping timeout)
11:42:20 --> mayanktg has joined #instantbird
11:44:51 --> rosonline has joined #instantbird
11:45:46 <-- mayanktg has quit (Ping timeout)
11:46:06 --> mayanktg has joined #instantbird
11:47:30 <-- arlolra has quit (Quit: arlolra)
11:49:10 --> aleth has joined #instantbird
11:49:10 * ChanServ sets mode +o aleth 
11:49:52 <aleth> I see flo-retina's already asked the questions I wanted to ask ;)
11:50:24 <sawrubh> aleth: related to me? :)
11:50:55 <aleth> all round (nhnt11 mostly ;) )
11:51:14 <mayanktg> I wrote the test for the function to create the ver attribute.
11:52:15 <-- jb has quit (Connection reset by peer)
11:52:27 --> jb has joined #instantbird
11:52:48 <aleth> still no nightly? :(
11:52:52 <-- mayanktg-ph has quit (Quit: Connection closed for inactivity)
11:53:25 <mayanktg> aleth: Hello :) . Regarding implementing the Entity Discovery I have set a flag for the entities whether available or not as per the Query stanza received. How to send this data to the conversation binding to enable the respective buttons when a user starts the conversation? :-o
11:53:32 * nhnt11 is confused
11:53:49 <nhnt11> My clobber succeeded but I don't have an app
11:53:54 <nhnt11> is this related to the packaging bug?
11:54:00 <nhnt11> aleth: ^
11:54:23 <aleth> No, it builds fine
11:54:30 <aleth> At least for me...
11:54:40 <nhnt11> Okay this is weird.
11:54:45 <nhnt11> It looks like I built Thunderbird
11:54:58 <aleth> Did you delete your mozconfig?
11:55:03 <nhnt11> No
11:55:09 <nhnt11> My mozconfig is the same as it always was
11:55:12 <aleth> Weird
11:55:20 <nhnt11> I use a $THUNDERBIRD flag to set what I want to build
11:55:26 <nhnt11> It used the correct objdir
11:55:28 <nhnt11> (obj-instantbird)
11:55:31 <nhnt11> but I have a Daily.app
11:57:50 <aleth> nhnt11: Did you also update the purple repo?
11:58:03 <aleth> Assuming you pulled recently
11:58:08 <nhnt11> hmm, no, I'll do that
11:58:28 <aleth> mayanktg: Didn't we discuss this already and I asked you to confirm with flo-retina?
11:58:45 <aleth> Did you do this?
12:00:26 * nhnt11 still doesn't get why/how thunderbird could get built but with the correct objdir set
12:00:31 <-- mayanktg has quit (Ping timeout)
12:01:33 <aleth> I don't know, but as there are configure changes you are missing, strangeness may occur.
12:02:14 --> mayanktg has joined #instantbird
12:03:34 <aleth> mayanktg: http://log.bezut.info/instantbird/140701/#m424
12:04:05 <nhnt11> Is there a way to check from the initial parts of the build log if I'm building Instantbird vs. Thunderbird?
12:05:52 <aleth> It dumps enable-application somewhere I think
12:08:10 <nhnt11> "Building Thunderbird by default. Set --enable-application to build a different application."
12:08:11 <nhnt11> :S
12:08:52 <nhnt11> This is my mozconfig: http://pastebin.instantbird.com/747582
12:09:05 <-- mayanktg has quit (Ping timeout)
12:09:12 <nhnt11> $THUNDERBIRD is not set
12:09:17 <nhnt11> And the correct objdir is being used
12:09:24 <nhnt11> (obj-instantbird)
12:09:26 --> mayanktg has joined #instantbird
12:09:47 <aleth> Have you checked you've pulled everything, and run mach configure && mach build?
12:10:01 <mayanktg> aleth: Thanks...I was finding this in 2nd July.
12:10:01 <mayanktg> flo-retina: Is sending a buddy notification regarding the entities the appropriate way to disable/enable a capability in a conversation as aleth suggested?
12:10:10 <nhnt11> Aha. I just found this:
12:10:12 <nhnt11>  0:03.02 Adding configure options from /Users/nhnt11/Dev/comm-central/mozconfig
12:10:12 <nhnt11> ...
12:10:15 <nhnt11>  0:03.02   --enable-application=im
12:10:37 <nhnt11> And yet... I have mail/ and ldap/ in my objdir
12:10:44 <flo-retina> aleth: any idea of the mac failure on buildbot?
12:10:54 <aleth> flo-retina: Haven't taken a look yet
12:12:21 <mayanktg> Also flo-retina whenever you're free could you take a look at the Bug 1004930 .
12:12:24 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1004930 enh, --, ---, mayanktg, ASSI, Generic way to add buttons for actions to a conversation
12:13:45 <flo-retina> mayanktg: It's probably one of the next patches I'll review
12:14:19 <flo-retina> mayanktg: which buddy notification are you talking about
12:14:22 <sawrubh> flo-retina: and then my FileLink one? :)
12:14:53 <flo-retina> mayanktg: maybe look at account-buddy-availability-changed
12:16:27 <mayanktg> flo-retina: OK. I'm taking a look at it. I will have to notify the user regarding the available "capabilities" of the client. So aleth suggested to send a buddy notification about the available entities.
12:17:02 <flo-retina> mayanktg: again, what's a "buddy notification" in that sentence?
12:17:20 <flo-retina> (I can likely imagine at least 3 different possible things that could be described with these words)
12:18:03 <aleth> flo-retina: I suggested adding a video boolean to the accountbuddy interface, and firing a notification when it changes.
12:18:16 <flo-retina> uh :(
12:18:28 <flo-retina> aleth: then we would also need one for audio, and one for file transfer? :(
12:19:43 <aleth> We'd probably want to generalize it to a capabilities object at that point.
12:20:15 <aleth> I didn't want to complicate things too much for his WIP
12:20:41 <-- mayanktg has quit (Ping timeout)
12:20:58 <-- BWMerlin has quit (Quit: BWMerlin)
12:21:46 --> mayanktg has joined #instantbird
12:22:32 <aleth> But if you have a better idea... that's why I asked him to check with you ;)
12:24:07 * Fallen is now known as Fallen|away
12:25:49 <flo-retina> aleth: we already have something called "availabilityDetails"
12:26:03 <flo-retina> aleth: I think it's an integer though :-S
12:26:27 <flo-retina> maybe 1 = file transfer, 2 = audio, 4 = video? :-D
12:26:46 <aleth> Isn't that the integer for invisible/dnd/busy...?
12:27:03 * aleth can't remember
12:27:05 <flo-retina> nope
12:27:23 <flo-retina> aleth: http://lxr.instantbird.org/instantbird/source/chat/components/public/imIStatusInfo.idl#38
12:27:27 <mayanktg> http://mxr.mozilla.org/comm-central/source/chat/components/public/imIStatusInfo.idl#41 is this the definition?
12:27:30 <mayanktg> ohh
12:27:34 <mayanktg> :)
12:28:07 <flo-retina> aleth: it was meant to be anything that could influence sorting by status, within a status type.
12:28:33 <aleth> flo-retina: So, really it's for something else than what we need, but it could be redefined? ;)
12:28:44 <flo-retina> while what I just suggested is a small abuse, it's not completely crazy to think that someone with video enabled is more available than someone with only audio
12:30:14 <aleth> ah, the integer is used for picking the preferred buddy
12:30:21 <flo-retina> aleth: http://lxr.instantbird.org/instantbird/source/purple/purplexpcom/src/purpleAccountBuddy.cpp#423 and http://lxr.instantbird.org/instantbird/source/chat/modules/jsProtoHelper.jsm#352 seem to be the only places where it's actually set
12:30:48 <flo-retina> ie. it's always 0 currently
12:31:00 --> Rym has joined #instantbird
12:31:31 <aleth> Using it for capabilities is OK with me, but a bitwise flag seems quite oldschool :D
12:31:33 <flo-retina> I was hoping to use it in JS-XMPP maybe for the priority, or something.
12:32:00 * Fallen|away is now known as Fallen
12:32:21 <flo-retina> aleth: we could put nicer getters on the imIConversation interface?
12:32:33 <aleth> That would help.
12:32:46 <flo-retina> aleth: I'm trying to avoid changing the API too much and requiring addition of plenty of pointless getters in the purplexpcom part
12:32:58 <aleth> Yes
12:33:21 <aleth> It's better than a dozen booleans in prpl interfaces, for sure.
12:35:34 <flo-retina> bug 963519 has a very nice mix of reviewer/reviewee pairs, it's different for each of the parts! Nice collaborative effort! :)
12:35:37 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=963519 min, --, ---, nobody, NEW, Stop using Object.prototype.__proto__ because of bug 948227
12:35:39 <-- Rym has quit (Ping timeout)
12:36:26 <flo-retina> aleth: btw, thanks for the feedback on the message api changes :).
12:37:21 --> Rym has joined #instantbird
12:38:19 <sawrubh> flo-retina: just confirming, in http://xmpp.org/extensions/xep-0047.html it seems the receiver cannot figure out if the stream was closed due to manual closing on sender's side or whether the transfer is complete, am I right?
12:40:13 <aleth> mayanktg: So you know what to do now? Change the comment in http://lxr.instantbird.org/instantbird/source/chat/components/public/imIStatusInfo.idl#38 to reflect how you assign the bits ;)
12:40:57 <flo-retina> sawrubh: what about http://xmpp.org/extensions/xep-0047.html#close ?
12:42:07 <sawrubh> flo-retina: yeah, so a) suppose the sender for some reason wants to close the transfer deliberately, it'll still send a close element in the IQ stanza and b) if the file transfer is complete it would still send the same thing
12:42:38 <aleth> sawrubh: Don't you know how many bytes you are expecting>
12:42:42 <flo-retina> sawrubh: don't you have the size of the file at the beginning of the transfer?
12:43:18 <sawrubh> yeah I could do that, I thought the spec provided some in built way to distinguish these two cases
12:43:32 <mayanktg> aleth: Yes. I'm taking a look at it. I will need to set the "availabilityDetails" to 1,2,3,... as per each of the entity received and then set the capabilities in the conversation binding accordingly. Right?
12:44:07 <flo-retina> sawrubh: having the size seems like enough. Anything more would be duplicated information (and you would be annoyed if it's inconsistent)
12:55:37 <aleth> mayanktg: You think of the long integer as a register where you set bits, i.e. like flo said, 1 for audio, 2 for video, 4 for filetransfer etc. and you put a method on uiConvs to read this for current preferred buddy turning it into a nice JS object, e.g. {audio: true, video: false, filetransfer: true}, that you can then use from conversation.xml
12:56:28 <-- mayanktg has quit (Ping timeout)
12:56:48 --> mayanktg has joined #instantbird
13:04:21 <flo-retina> aleth: video or audio require a much greater availability than file transfer
13:05:02 <aleth> flo-retina: Sure, it was merely an example for using bits ;)
13:05:49 <mayanktg> I didn't understand what much greater availability means here :-|
13:06:11 <mayanktg> aleth: ^
13:06:11 <flo-retina> mayanktg: don't worry too much about it now, it's a detail :)
13:06:20 <mayanktg> flo-retina: Ok 
13:07:08 <aleth> mayanktg: It means that it's an integer so that you can compare the values, bigger values mean "available in more interesting ways". So higher bits should correspond to more exciting things ;)
13:11:05 <mayanktg> aleth: Hmm. Okays :) 
13:11:05 <mayanktg> /me tries...
13:17:34 * nhnt11 doesn't know why he can't build ib
13:21:09 <nhnt11> flo-retina, aleth: Any ideas?
13:21:39 <nhnt11> Is there some recent build system change I'm not aware of?
13:23:39 <aleth> Unless something new broke yesterday, and you pulled everything, I don't know why it wouldnt work for you
13:27:48 <nhnt11> I don't even know what else I can try
13:28:20 <nhnt11> I tried removing the if block for the enable-application line too
13:28:41 <sawrubh> so http://pastebin.instantbird.com/747589 is how my writeChunk looks, this function was earlier opening and closing the file being written to every time
13:29:03 <-- GeekShadow has quit (Ping timeout)
13:29:04 <sawrubh> I've changed it to only open it the first time and close it when transfer closes
13:29:06 --> GeekShadow has joined #instantbird
13:30:12 <sawrubh> http://pastebin.instantbird.com/747590 is inside onIQStanza where I close when file transfer is done
13:30:48 <sawrubh> is there any visible Promise-conceptual issue that you can see in my first pastebin
13:32:00 <sawrubh> what I want to do basically is always do all the operations on 'this._file'
13:35:19 <flo-retina> nhnt11: you don't have a mozconfig file inside the mozilla/ sub folder that could be messing things up, right?
13:35:42 <nhnt11> flo-retina: nope
13:38:48 <aleth> sawrubh: If you're writing to the same file, you should chain all the promises referring to that file.
13:39:19 <flo-retina> nhnt11: so what's the problem exactly?
13:40:21 <aleth> sawrubh: That code looks like it won't work, and might be easier with Tasks
13:40:55 <nhnt11> flo-retina: Looks like enable-application=im isn't working. Thunderbird is getting built instead of Ib
13:41:32 <flo-retina> nhnt11: that part I knew. Is there stuff we should look out to see if we can understand what's happening?
13:43:07 <aleth> sawrubh: I don't understand the double .then in you pastebin
13:43:11 <nhnt11> flo-retina: Do you want a build log? I already pastebin'd my mozconfig..
13:43:25 <flo-retina> yeah, whatever you have that could be useful
13:43:37 <flo-retina> it's not that I "want" it. it's just that I currently can't help you
13:43:47 <flo-retina> so if you want us to help, we need more information
13:43:56 <nhnt11> Okay.
13:44:29 <flo-retina> hmm, if you suspect a very recent breakage, we could maybe start a new mac nightly to see if we have the same failure on buildbot
13:45:10 <nhnt11> One minute..
13:46:05 <aleth> sawrubh: look again at the "copy file by chunk" examples here https://developer.mozilla.org/en-US/docs/JavaScript_OS.File/OS.File_for_the_main_thread
13:46:51 <sawrubh> aleth: that code was working before I had inserted that if check and used to close in the chain itself
13:47:06 <aleth> sawrubh: It may have been working by accident
13:47:27 <aleth> It's *async*,  you have to chain things to make them happen in the right order.
13:47:31 <sawrubh> aleth: I don't understand why I should chain all the promises and what's wrong with the current code
13:48:19 <aleth> sawrubh: You're chaining a write on a file open but subsequent writes aren't chained on anything so the file might not even be open yet
13:48:32 <sawrubh> aleth: yeah so to make them work in the right order, I'm chaining the two thens (one on L8 and one on L16 of http://pastebin.instantbird.com/747589)
13:48:44 <sawrubh> ok
13:49:04 <aleth> sawrubh: You are doing absolutely nothing with the promise returned by write()
13:50:00 <sawrubh> ok so this is what I wanna do, I want to chain the write on a file open and for the subsequent writes I wanna chain them on this open file (or promise)
13:50:04 * sawrubh checks the example
13:50:32 <aleth> Yeah, you need to look at some more os.file/promises/tasks examples and understand them ;)
13:50:44 <aleth> It's not trivial the first time you encounter them.
13:51:32 <nhnt11> flo-retina: Here's my build log up to reticulating splines: http://pastebin.instantbird.com/747615
13:51:40 <nhnt11> line 35 looks like mail is getting enabled 
13:51:54 <nhnt11> line 38 is adding im though
13:53:02 <nhnt11> line 794 looks odd to me
13:53:20 <nhnt11> Seems like --enable-application is an "unrecognized option"? {S
13:53:22 <nhnt11> :S *
13:55:44 <-- mayanktg has quit (Ping timeout)
13:56:58 --> mayanktg has joined #instantbird
13:59:43 <flo-retina> nhnt11: are you 100% sure that both m-c and c-c were pulled at the same time?
14:00:36 <nhnt11> flo-retina: I did python client.py checkout first, and after it built thunderbird, did it again, went into mozilla/ and did hg pull -u there. I also pulled purple
14:01:04 <nhnt11> I'm not sure what you mean by "at the same time". If you mean are both up to date, I'm fairly certain, yes.
14:01:28 <flo-retina> yeah
14:02:10 <aleth> I'm just building in the background to see if something broke since yesterday
14:02:43 <nhnt11> aleth: Just mentioning, you can tell if it's building tb if your object has ldap/ and mail/ in it..
14:02:57 <nhnt11> (so you don't have to wait for it to finish)
14:03:13 --> iamjayakumars has joined #instantbird
14:03:17 * Fallen is now known as Fallen|away
14:04:54 <sawrubh> so aleth, I still don't get it, according to my understanding the subsequent writes will be chained to the file which is open already and the promise is represented by this._file
14:07:37 --> clokep_work has joined #instantbird
14:07:37 * ChanServ sets mode +o clokep_work 
14:07:57 <aleth> this._file may be a promise, but as soon as the file is open that promise is resolved and stays resolved whatever happens after that.
14:08:20 <aleth> Writing to that file will give you a new promise, that resolves when the write has happened, etc
14:08:44 <sawrubh> so I should assign that latest promise to this._file to write to it in future?
14:08:53 <aleth> Please read some more of https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Promise until you see how it works.
14:08:55 <sawrubh> because I'm not closing it so it'll remain open
14:09:17 <clokep_work> Or ask questions if you're unsure. ;)
14:09:33 <aleth> That will only happen after the write has completed, which is too late
14:09:52 <aleth> If I understand you correctly.
14:10:30 <sawrubh> let me rephrase (and that might make things clear for me :) )
14:11:46 * clokep_work hasn't read the log yet.
14:12:41 <aleth> nhnt11: My build worked fine.
14:13:27 <-- clokep_work has quit (Ping timeout)
14:14:21 --> mconley has joined #instantbird
14:16:50 <aleth> flo-retina: Can you retrigger a nightly? make package works for me, locally, fwiw.
14:22:48 <aleth> Maybe it's distribution that's broken, like on win?
14:26:09 <-- mpmc has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
14:28:05 <flo-retina> aleth: have you tried on an universal build?
14:28:12 <aleth> sawrubh: btw it should be fine to open/close the file for each chunk if that makes your life easier, as long as the chunks aren't ridiculously tiny.
14:28:24 <flo-retina> aleth: it fails in the "compile" step, that's before we even try to run |make distribution|
14:28:33 <aleth> flo-retina: ah :(
14:28:44 * aleth mumbles something about universal builds
14:29:55 <flo-retina> aleth: and the windows failure is 100% unlikely to happen on Unix
14:30:01 <-- iamjayakumars has quit (Client exited)
14:30:02 <sawrubh> aleth: I was doing that before (and I'll do that eventually if I fail in this attempt or waste too much time) but I want to do this since this is just good practice. Btw the chunk size can be max 4096 bytes, is that ridiculously tiny?
14:30:18 <aleth> flo-retina: I didn't mean the same failure, I meant a failure in the same step ;)
14:30:34 <flo-retina> the windows machine currently needs a reboot btw
14:30:59 <aleth> sawrubh: no it's reasonable, and yes it's better if you don't keep closing the file of course
14:31:01 <flo-retina> awful OS that breaks when we reboot and breaks every other week if we DON'T reboot :(
14:36:58 <-- mayanktg has quit (Ping timeout)
14:39:31 --> mayanktg has joined #instantbird
14:41:43 --> Armada has joined #instantbird
14:48:00 * nhnt11 isn't sure what to do next
14:48:21 <nhnt11> build log isn't revealing anything else after rereading it..
14:49:10 --> iamjayakumars has joined #instantbird
14:49:33 <flo-retina> nhnt11: what's in /Users/nhnt11/Dev/comm-central/mozconfig ?
14:49:55 <nhnt11> flo-retina: http://pastebin.instantbird.com/747582
14:50:12 --> mayanktg-ph has joined #instantbird
14:50:40 <nhnt11> I also tried this, just to be very sure: http://pastebin.instantbird.com/747620
14:51:14 <flo-retina> is the $ at the end of line 3 expected?
14:51:46 <nhnt11> flo-retina: That was copied from nano, sorry.
14:51:52 <nhnt11> (there's not really a $)
14:52:16 <nhnt11> This is what it looks like at the moment: http://pastebin.instantbird.com/747621
14:52:40 <nhnt11> I've been using this mozconfig for ages now, no problems.
14:53:41 <flo-retina> nhnt11: stuff doesn't seem broken on the new nightly that I started: http://buildbot.instantbird.org/builders/macosx-nightly-default/builds/2232/steps/compile/logs/stdio
14:53:49 <nhnt11> I saw that.
14:53:57 <nhnt11> :(
14:55:00 <flo-retina> nhnt11: to unblock you, you can poke at http://mxr.mozilla.org/comm-central/source/configure.in#112
14:55:07 <flo-retina> either export MOZ_BUILD_APP=im from the mozconfig
14:55:24 <flo-retina> or just replace "mail" with "im" as the default value on line 113
14:55:26 <nhnt11> flo-retina: Thanks!
14:55:40 <flo-retina> but you'll really need to figure it out at some point :-s
14:55:50 <nhnt11> Yeah, but I'd rather not do it now :-/
14:55:56 <flo-retina> sure
14:56:00 <flo-retina> just hardcode the values you want
14:59:11 <nhnt11> exporting it from mozconfig didn't work, but hardcoding it in configure.in did.
14:59:13 <nhnt11> Thanks
14:59:19 <nhnt11> I really suspect something weird is up with my mozconfig
14:59:41 <aleth> The way mozconfigs are handled /has/ changed recently.
14:59:55 <nhnt11> aleth: Hmm, can you paste your mozconfig please?
14:59:58 <aleth> May or may not be related
15:00:06 <nhnt11> (just curious)
15:01:02 <aleth> much simpler than yours http://pastebin.instantbird.com/747622
15:01:30 <flo-retina> ah you are also using the halloween branding :)
15:03:10 <aleth> Very pretty alternative icon :)
15:12:25 <-- mayanktg has quit (Ping timeout)
15:13:25 --> mayanktg has joined #instantbird
15:16:04 <-- mayanktg-ph has quit (Quit: )
15:21:02 <-- flo-retina has quit (Ping timeout)
15:21:04 --> flo-retina has joined #instantbird
15:21:05 * ChanServ sets mode +qo flo-retina flo-retina 
15:23:03 <-- mayanktg has quit (Ping timeout)
15:23:21 --> mayanktg has joined #instantbird
15:25:11 <-- aleth has quit (Ping timeout)
15:26:29 --> aleth has joined #instantbird
15:26:29 * ChanServ sets mode +o aleth 
15:26:43 <sawrubh> I still can't figure out what I need to do
15:27:17 * sawrubh re-reads aleth's suggestions
15:27:36 <flo-retina> aleth: mac still broken :(
15:30:50 * nhnt11 gets to work finally
15:33:19 <nhnt11> aleth: I have two ideas for bug 1034967. Either add a null check here: https://mxr.mozilla.org/comm-central/source/im/content/blist.js#178, or simply return Promise.reject() instead of null 3 lines above.
15:33:21 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1034967 nor, --, ---, nhnt11, ASSI, Regression: Context menu broken over blank space in contact list
15:33:42 <nhnt11> I like the latter idea.
15:34:17 <nhnt11> Hmm, on second thought, no
15:35:18 <nhnt11> (Promise.reject() would seem to imply something went wrong getting logs, whereas in this case we don't have a target for which to get logs in the first place)
15:35:44 <aleth> nhnt11: Right, there shouldn't be a show logs item in the context menu over blank space anyway
15:36:05 <aleth> https://mxr.mozilla.org/comm-central/source/im/content/blist.js#79 should be an else if
15:36:44 <nhnt11> aleth: else if (this._getLogs()) ?
15:36:46 <aleth> note https://mxr.mozilla.org/comm-central/source/im/content/blist.js#73
15:37:12 <nhnt11> um
15:37:16 <nhnt11> give me a minute
15:37:20 <aleth> no
15:37:27 <nhnt11> Yeah okay
15:37:29 <nhnt11> Got it
15:37:44 <aleth> I mean, it would work, but we can do better ;)
15:37:52 <-- mayanktg has quit (Ping timeout)
15:37:58 <nhnt11> else if (hide && !this.onConv)
15:38:06 <nhnt11> er, the not of that^
15:38:11 <aleth> yeah.
15:38:14 --> mayanktg has joined #instantbird
15:38:33 <aleth> You can drop the else too.
15:39:57 <-- mayanktg has quit (Ping timeout)
15:40:36 --> mayanktg has joined #instantbird
15:43:08 <-- mayanktg has quit (Ping timeout)
15:43:29 --> mayanktg has joined #instantbird
15:44:37 <nhnt11> brb
15:47:36 <aleth> flo-retina: http://mxr.mozilla.org/comm-central/source/im/config/mozconfigs/macosx/mozconfig#6 and http://mxr.mozilla.org/comm-central/source/im/config/mozconfigs/macosx/mozconfig-release#6
15:47:58 <flo-retina> ahah!
15:48:04 <flo-retina> aleth: <3
15:48:29 <flo-retina> (it's sad that this doesn't fail with a much more descriptive error)
15:48:44 <flo-retina> r=me for the obvious fix if you are fixing it
15:49:26 <aleth> No problem
15:53:37 <aleth> flo-retina: done
15:53:50 <-- gerard-majax_ has quit (Ping timeout)
15:54:42 <flo-retina> aleth: thanks!
15:54:44 * flo-retina started a new nightly again
15:54:52 <aleth> Lets hope this it it.
15:57:11 <-- nhnt11 has quit (Ping timeout)
15:57:30 <-- mconley has quit (Ping timeout)
15:57:58 --> mconley has joined #instantbird
15:58:05 --> nhnt11 has joined #instantbird
16:09:53 <flo-retina> aleth: the Windows issue is at http://mxr.mozilla.org/comm-central/source/im/build.mk#79
16:10:12 <flo-retina> topsrcdir is an absolute path, and I'm afraid we do need to keep it that way
16:12:31 <flo-retina> oh! http://hg.mozilla.org/comm-central/rev/9e17f6ef6620
16:13:07 <flo-retina> http://hg.mozilla.org/comm-central/diff/9e17f6ef6620/Makefile.in is a change we don't have
16:13:13 <flo-retina> it looks a lot like it could be related to our issue
16:14:01 <aleth> excellent!
16:15:02 <aleth> Though it's a tiny change which I don't understand how it would make any difference
16:15:12 <aleth> but maybe it's a Windows thing
16:25:51 --> gerard-majax_ has joined #instantbird
16:30:54 <aleth> flo-retina: http://buildbot.instantbird.org/builders/macosx-nightly-default/builds/2233/steps/compile/logs/stdio :(
16:31:47 <flo-retina> looks like it wasn't the right fix then :-/
16:33:03 <flo-retina> aleth: http://mxr.mozilla.org/comm-central/source/mail/config/mozconfigs/macosx-universal/nightly#10
16:33:32 <aleth> Yeah, it clearly worked better before the "fix" too ;)
16:37:49 <aleth> flo-retina: Is there a neat hg command to backout the last patch on remote?
16:38:00 <flo-retina> hg backout ?
16:38:12 * Fallen|away is now known as Fallen
16:38:23 <aleth> ah, thanks
16:38:37 <flo-retina> aleth: is it worth backing out right away until we have a working fix?
16:38:54 <flo-retina> (probably doesn't matter much as it won't trigger new builds anyway :))
16:39:40 <aleth> Either way works but that way I can forget about it.
16:39:57 <aleth> Nice how in hg commands have the names they should :D
16:40:51 <-- gerard-majax_ has quit (Ping timeout)
16:51:36 --> gerard-majax_ has joined #instantbird
16:54:21 --> Mook_as has joined #instantbird
17:04:45 <-- aleth has quit (Ping timeout)
17:08:32 <-- mayanktg has quit (Ping timeout)
17:08:55 --> mayanktg has joined #instantbird
17:10:40 <-- mayanktg has quit (Ping timeout)
17:11:03 --> mayanktg has joined #instantbird
17:12:37 <-- gerard-majax_ has quit (Ping timeout)
17:17:40 <-- jb has quit (Ping timeout)
17:18:43 --> gerard-majax_ has joined #instantbird
17:22:14 --> jb has joined #instantbird
17:25:29 <-- rosonline has quit (Quit: Instantbird 1.5 -- http://www.instantbird.com)
17:27:33 --> rosonline has joined #instantbird
17:34:22 <-- flo-retina has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
17:38:54 <-- jb has quit (Ping timeout)
17:46:35 <-- gerard-majax_ has quit (Ping timeout)
17:50:09 --> clokep_work has joined #instantbird
17:50:09 * ChanServ sets mode +o clokep_work 
17:57:48 <clokep_work> sawrubh: "then move on to integrating the Downloads panel" weren't we supposed to have a discussion first about hwo we want this UI to look?
18:01:18 <sawrubh> clokep_work: which UI are we talking about? the feedback UI?
18:01:32 <clokep_work> All of it.
18:02:15 <sawrubh> I thought drag and drop was good enough but yeah flo-retina pointed out we wanted something more prominent
18:02:36 <sawrubh> and we kinda settled on having a button on the toolbar mayanktg is adding
18:02:57 <sawrubh> clicking on that button would open up the downloads panel where you would have a button to send a file
18:02:57 <clokep_work> sawrubh: I'm talking about what you said about the Download panel.
18:03:28 --> flo-retina has joined #instantbird
18:03:28 * ChanServ sets mode +qo flo-retina flo-retina 
18:03:33 --> arlolra has joined #instantbird
18:03:47 <sawrubh> clokep_work: sorry it got off my mind that we wanted to discuss about it
18:04:27 <sawrubh> (and we weren't final on drag and drop being it, I had thought of adding the Downloads Panel anyway so didn't really occur to me we had a discussion pending)
18:11:30 <clokep_work> Phone, one second.
18:19:54 <-- rosonline has quit (Quit: Instantbird 1.5 -- http://www.instantbird.com)
18:22:10 --> rosonline has joined #instantbird
18:30:12 <-- iamjayakumars has quit (Client exited)
18:33:17 <clokep_work> sawrubh: So we had agree that DND is the reasonable thing for an initial UI, but never discussed a more "final" UI, IIRC.
18:33:38 <sawrubh> yes
18:34:02 <sawrubh> think now's the right time :)
18:34:24 <clokep_work> I'm available.
18:35:25 <sawrubh> so flo-retina's suggestion sounds good to me except the part where we have to click the button->download panel opens->click on button to send file
18:35:42 <sawrubh> I would personally like it to a click button->send file flow
18:35:50 <clokep_work> sawrubh: Can you please reword his suggestion since 1. I don't know what you mean and 2. I want to be sure you understand it. :)
18:37:25 <sawrubh> So he said, we'll have a button (like mayanktg's planning to add in bug 1004930) clicking on which will open/display the Downloads Panel, inside the Downloads Panel you would have a relatively large click target/button, clicking on which will probably open the File Picker and then you send the file
18:37:28 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1004930 enh, --, ---, mayanktg, ASSI, Generic way to add buttons for actions to a conversation
18:38:03 <clokep_work> sawrubh: OK, so you odn't like the extra click.
18:38:08 <sawrubh> yes
18:38:10 <clokep_work> If your proposal, how will status of downloads be shown?
18:38:32 <sawrubh> In the Downloads Panel, like in Firefox
18:38:49 <clokep_work> How would you bring up that Downloads panel then? :_S
18:39:55 <sawrubh> a downward pointing arrow (like the one in Firefox) in the line of where the tabs are in IB : http://i.imgur.com/U09C6g9.png
18:40:41 <-- chrisccoulson has quit (Ping timeout)
18:41:10 <clokep_work> In the tabbar?
18:41:16 <clokep_work> That's weird (and no where it is in Fx)
18:41:46 <mayanktg> sawrubh: Won't it look odd if we have a feature near the right side of tabbar and then three buttons just below it (video call, audio call, TS) ?
18:42:49 <nhnt11> Suggestion: A button on mayanktg's toolbar for the current conversation's downloads, and a special tab to show "all downloads" (this isn't too bad since /most/ people have relatively small chat windows (1/6th of the screen?))
18:43:00 * nhnt11 shrugs
18:43:17 <nhnt11> That duplicates stuff though, so probably bad ^
18:43:21 --> gerard-majax_ has joined #instantbird
18:43:45 <sawrubh> mayanktg: my concern with the buttons you're adding is that they're (a wee bit) small click targets?
18:43:50 * nhnt11 would personally like just a tab ;)
18:44:55 <-- Rym has quit (Ping timeout)
18:45:14 <clokep_work> sawrubh: If the'yre too small to click that's an issue we need to fix, but doesn't mean we need radically different UI for a similar thing.
18:46:11 <sawrubh> but yes after seeing mayanktg's work, I think it makes perfect sense to be aligned with what he's doing (a button like he has for video calls and such)
18:46:37 <sawrubh> that would be homogenous 
18:46:38 <flo-retina> sawrubh: how many times per hours do you expect people to want to click the "send file" button?
18:47:36 --> Tonnes has joined #instantbird
18:48:04 <sawrubh> flo-retina: not a lot :) but I'm not too concerned if you guys are ok with it 
18:48:13 <mayanktg> sawrubh: They are about 24px toolbarbuttons. https://bug1004930.bugzilla.mozilla.org/attachment.cgi?id=8445456 idk much, we should go with the best possilbe UI though :)
18:48:16 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1004930 enh, --, ---, mayanktg, ASSI, Generic way to add buttons for actions to a conversation
18:49:05 --> Rym has joined #instantbird
18:49:40 <flo-retina> sawrubh: I think I don't even understand your point. A few hours ago you wanted to have only the drag & drop UI. That's not discoverable at all. Now you are saying that a button in the primary UI isn't enough. I'm confused.
18:51:31 <sawrubh> a few hours ago I hadn't realized the point that a drag and drop might not be so obvious, once I realized it I wanted to make it amply clear (easy to click and hence the click target size argument) and that's why I was proposing what I was proposing
18:53:05 <sawrubh> I had finalized everything while talking to you, I was just discussing it now with clokep_work, and as I've said above going with mayanktg's UI (combined with your send file UX) makes the most sense in terms of aligning with existing experience and given that people won't be sending files too often during an hour
18:53:53 <clokep_work> sawrubh: OK, sounds like we'rea ll agreed then.
18:54:12 <sawrubh> yes we are.
18:56:57 --> chrisccoulson has joined #instantbird
18:57:23 --> nhnt11-testing has joined #instantbird
18:58:10 <-- nhnt11-testing has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
18:59:27 <-- mayanktg has quit (Ping timeout)
19:00:13 --> mayanktg has joined #instantbird
19:03:38 <-- mayanktg has quit (Ping timeout)
19:03:52 <sawrubh> clokep_work: can you explain your review comment about "ClassInfo?" I'm still not clear about that.
19:04:00 --> mayanktg has joined #instantbird
19:04:22 <clokep_work> sawrubh: The comment means "Doesnt it make more sense to use ClassInfo here?"
19:04:39 <clokep_work> Although really you should first answer whether it makes sense to add a Generic prototype to jsProtoHElper or not.
19:05:41 <-- mayanktg has quit (Ping timeout)
19:05:48 <sawrubh> I've removed that Generic prototype from jsProtoHelper
19:05:57 <sawrubh> I didn't see any use for it
19:06:08 --> mayanktg has joined #instantbird
19:06:08 <clokep_work> OK.
19:06:39 <sawrubh> btw I didn't know I could simply replace 'QueryInterface: XPCOMUtils.generateQI([Ci.prplIFileTransfer])' with ClassInfo
19:09:03 <clokep_work> Did you look at what ClassInfo does?
19:09:19 <sawrubh> it defines what interfaces can be used
19:09:34 <clokep_work> That didn't answer my question.
19:10:05 <sawrubh> No I didn't, I remembered this from flo's reply sometime back
19:11:08 <clokep_work> OK.
19:11:22 <sawrubh> https://developer.mozilla.org/en/docs/Mozilla_DOM_Hacking_Guide#The_nsDOMClassInfo_class
19:11:27 <clokep_work> It does more than just the QI call, but generally it's a better way cause it gets rid of more boiler-plate code.
19:11:38 <clokep_work> sawrubh: https://mxr.mozilla.org/comm-central/source/chat/modules/imXPCOMUtils.jsm#164
19:52:20 <-- clokep_work has quit (Ping timeout)
19:57:50 --> clokep_work has joined #instantbird
19:57:50 * ChanServ sets mode +o clokep_work 
20:12:26 --> aleth has joined #instantbird
20:12:26 * ChanServ sets mode +o aleth 
20:14:57 <-- mayanktg has quit (Ping timeout)
20:15:48 --> mayanktg has joined #instantbird
20:22:55 <-- chrisccoulson has quit (Quit: OSError: [Errno 130] Owner died)
20:23:29 * aleth is out of ideas for diagnosing the mac distribution failure
20:23:48 <nhnt11> Call Dr. House!
20:25:08 <flo-retina> aleth: have you tried reproducing it locally with a universal build?
20:25:50 <aleth> nhnt11: you mean we should send a team to France to look for clues?
20:26:14 <nhnt11> The buildbot must be lying somehow.
20:26:41 <aleth> flo-retina: no, but I checked for recent universal build mozconfig changes (none)
20:26:45 <flo-retina> that must be a French bot, always on strike ;)
20:27:21 --> EionRobb has joined #instantbird
20:29:15 <-- aleth has quit (Ping timeout)
20:29:18 --> aleth has joined #instantbird
20:29:18 * ChanServ sets mode +o aleth 
20:29:56 <aleth> flo-retina: I'm right in thinking none of this needs porting, right? https://hg.mozilla.org/comm-central/rev/915a09757a18
20:31:05 <clokep_work> aleth: Possibly https://hg.mozilla.org/comm-central/rev/915a09757a18#l4.1
20:31:19 <clokep_work> And I'm unsure what https://hg.mozilla.org/comm-central/rev/915a09757a18#l3.12 is
20:32:02 <-- EionRobb has quit (Ping timeout)
20:34:16 --> EionRobb has joined #instantbird
20:35:47 <aleth> clokep_work: I /think/ it's for the try server?
20:39:19 <clokep_work> I think so too.
20:41:21 <-- aleth has quit (Ping timeout)
20:41:26 --> aleth has joined #instantbird
20:41:26 * ChanServ sets mode +o aleth 
20:43:18 <-- mayanktg has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
20:43:35 <-- mconley has quit (Ping timeout)
20:44:04 --> mconley has joined #instantbird
20:45:58 <flo-retina> if it's in build/ rather than in mail/ we probably inherit all of that already without any active porting effort on our side
20:48:51 <-- Rym has quit (Ping timeout)
20:49:10 --> Rym has joined #instantbird
20:50:42 <-- mconley has quit (Input/output error)
20:53:15 <-- aleth has quit (Ping timeout)
20:55:16 --> aleth has joined #instantbird
20:55:16 * ChanServ sets mode +o aleth 
20:56:47 <nhnt11> aleth: You reviewed faster than I could change the patch name :D
20:57:07 <nhnt11> also you changed the patch name back to a backslash 
20:57:09 <nhnt11> anyways
20:57:13 * nhnt11 fixes the nit
20:57:15 <aleth> heh, sorry
20:57:16 --> mconley has joined #instantbird
20:57:38 <nhnt11> Editors should highlight double spaces :-/
21:00:32 <aleth> nhnt11: You can push it yourself ;)
21:00:38 <nhnt11> okay
21:00:50 <aleth> Or is the tree closed again?
21:01:07 <nhnt11> Nah.
21:01:54 <aleth> Ah, don't worry about it, I think Ryan is going to do some checkins soon anyway.
21:02:07 <aleth> Faster than we fix nightlies probably :-/
21:02:30 <nhnt11> i just pushed :]
21:02:37 <nhnt11> Um, it failed, nvm
21:04:19 <instantbot> nhnt11@gmail.com changed the Resolution on bug 1034967 from --- to FIXED.
21:04:21 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1034967 nor, --, 1.6, nhnt11, RESO FIXED, Regression: Context menu broken over blank space in contact list
21:06:02 <nhnt11> flo-retina, aleth: I'm looking at twitter, but can I fix it in a separate bug so that log file splitting can land? Is it important that twitter be fixed before we land it?
21:06:37 <nhnt11> Well not quite "fixed" but "updated"
21:07:07 <nhnt11> Also, we're all clear on the splitting conditions right?
21:07:08 <clokep_work> nhnt11: Also clear checkin-needed?
21:07:10 <-- aleth has quit (Ping timeout)
21:07:38 <nhnt11> I didn't see any other bug there...
21:07:42 <nhnt11> Oh, missed it.
21:07:48 <clokep_work> nhnt11: I haven't been following that closely, but it scares me slightly that Twitter needs to be updated. Why is this?
21:08:06 <nhnt11> clokep_work: Twitter currently does not mark tweets it receives at reconnect as "delayed"
21:08:23 <clokep_work> I don't understand what that means. :)
21:08:48 <nhnt11> clokep_work: Basically like offline messages, the tweets were received some time after they were sent
21:09:06 <nhnt11> And we want to use the delayed flag to prevent splitting log files in a couple of cases
21:09:19 <nhnt11> i.e. offline messages shouldn't trigger new sessions
21:09:29 <clokep_work> Hmm...alright.
21:09:42 <clokep_work> Couldn't we just split based on the "now" time vs. the time in the message?
21:10:26 <nhnt11> clokep_work: What would our delay tolerance be? what if our I/O queue gets delayed and we end up processing stuff late?
21:10:40 <nhnt11> Using the delayed flag would be the "right" way to do it...
21:10:55 <clokep_work> I don't see how this is related to I/O at all, it'd be on when you say to log things, not when you actually write to disk.
21:11:15 <clokep_work> nhnt11: Funny I see it as the exact opposite.
21:11:47 <nhnt11> clokep_work: Right, I just meant, if logMessage was called 1k times in a second, our tolerances would vary...
21:12:46 <nhnt11> clokep_work: Why? An offline message isn't really the same case as when we process a message later than it arrived
21:13:00 <clokep_work> Why?
21:13:19 <-- Rym has quit (Ping timeout)
21:13:24 <nhnt11> clokep_work: Processing a message late doesn't mean it should trigger a new session in the log file.
21:13:46 <clokep_work> Yes, isn't that what we just discussed? :-S
21:14:05 --> Rym has joined #instantbird
21:14:22 * nhnt11 is now confused
21:14:39 <clokep_work> nhnt11: I'm confused at why we need a special flag for it.
21:14:45 <clokep_work> It seems like an over-engineered API.
21:14:53 <nhnt11> We already have one, so I was going to use it...
21:15:09 * nhnt11 doesn't get why it's over-engineered
21:15:21 <flo-retina> clokep_work: we already have that flag
21:15:51 <flo-retina> clokep_work: currently only JS-XMPP (and probably libpurple-xmpp, and other libpurple prpls receiving offline messages) set it
21:15:53 <clokep_work> flo-retina: We do? Why?
21:15:58 <clokep_work> Interesting.
21:16:13 <flo-retina> but twitter.js doesn't set it
21:16:47 <clokep_work> Alright, so it's an unrelated bug, essentially.
21:16:48 <flo-retina> clokep_work: so IIRC the way Pidgin used that flag was that it showed the full timestamp (including the day in addition to the hour) for these messages.
21:17:05 <nhnt11> clokep_work: That's why I wanted to do it in a different BMO bug
21:17:17 <flo-retina> clokep_work: yeah, it's an unrelated bug. But if we don't fix it, connecting a twitter account for the first time will cause plenty of small log files to be created
21:17:25 <clokep_work> Yeah, got it.
21:17:28 <clokep_work> OK.
21:17:28 <flo-retina> nhnt11: yeah, go file a different bug :)
21:17:37 <nhnt11> Great
21:17:44 <flo-retina> clokep_work: your suggestion of comparing the time and not splitting the file if the message is far in the past made sense though.
21:18:33 <nhnt11> That would ridiculously mess with my tests :] (I know that's not a valid argument)
21:22:42 <flo-retina> nhnt11: I think you should just file that twitter bug, attach a patch, get an r+, and land everything at once ;)
21:22:49 <flo-retina> nhnt11: that twitter thing can't be difficult, right?
21:23:54 <nhnt11> flo-retina: Not difficult, once I figure out how these tweets are being handled (How many tweets do we retrieve on the first connect? Is there some way we avoid retrieving logged tweets in future reconnects? etc...). I'm doing that now.
21:24:14 <clokep_work> Unknown and yes.
21:24:17 <clokep_work> Are the answers to those. ;)
21:24:21 <nhnt11> I just found lastMessageId
21:24:57 <flo-retina> nhnt11: there are different requests. We have timeline requests, and then a stream.
21:25:10 <flo-retina> you can probably safely assume that anything coming from the timeline requests is delayed
21:25:19 <nhnt11> Hmm, right
21:25:29 <nhnt11> The stream would be the only "realtime" source of tweets
21:30:02 <flo-retina> IIRC the stream also gets some tweets from a few minutes ago
21:30:38 <flo-retina> but ensuring you set the delayed flag correctly on the other requests should be enough to get your logs to not be excessively split
21:32:20 <nhnt11> flo-retina: This looks like it should do the trick, right? http://pastebin.instantbird.com/747726
21:33:03 <nhnt11> The first hunk is the displayTweet function
21:33:24 * nhnt11 wonders how to properly test this.
21:33:31 <flo-retina> nhnt11: I don't think you want to ever set flags.delayed to false
21:33:43 <nhnt11> Hmm
21:33:50 <nhnt11> okay
21:35:08 * nhnt11 thought this would take more effort :)
21:38:01 <nhnt11> uh
21:38:11 <flo-retina> I was also under the impression that you were procrastinating the twitter fix more than it was worth ;)
21:38:19 <nhnt11> looks like Ib/ oauth don't support two-factor auth?
21:39:11 <flo-retina> why would there be something specific to support for that?
21:39:29 <nhnt11> No idea, but the oauth window is completely blank after I enter my password, though i received an auth code on my phone
21:39:46 <flo-retina> any JS error?
21:39:57 <nhnt11> Nope
21:40:29 <nhnt11> This is what I see: http://puu.sh/a1aBG/7b0aac152a.png
21:40:43 <nhnt11> That weird undefined at the corner was there before I entered my password
21:42:15 <nhnt11> I'll test with a clone of my main profile..
21:42:53 <-- nhnt11 has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
21:44:29 --> nhnt11 has joined #instantbird
21:44:36 <flo-retina> that undefined doesn't sound good :-S
21:45:06 <nhnt11> okay, my main profile went completely bonkers on my local build
21:45:16 <nhnt11> "Services.logins" wasn't found or soemthing
21:45:30 <flo-retina> at the same time as the mozconfig? :-D
21:45:36 <nhnt11> "TypeError: Services.logins is undefined (new TypeError("Services.logins is undefined", "file:///Users/nhnt11/Dev/obj-instantbird/mozilla/dist/Instantbird.app/Contents/MacOS/components/imAccounts.js", 493))"
21:45:56 <nhnt11> My testing profile is working fine other than the oauth thing
21:46:16 * nhnt11 checks if twitter allows application-specific passwords or something
21:46:49 <nhnt11> Indeed, that worked :)
21:49:22 <clokep_work> Services.logins stuff? Hmm...I wonder if that's that bug I filed earlier.
21:50:06 <-- arlolra has quit (Quit: arlolra)
21:52:02 <nhnt11> flo-retina: I'm setting the feedback flag to review on the log file splitting patch now that the twitter patch is up.
21:52:24 * clokep_work sighs.
21:52:24 <clokep_work> I hate you CTCP.
21:52:26 <clokep_work> I hate you.
21:52:57 <flo-retina> clokep_work: is that a general statement, or related to the new message APIs?
21:53:11 <clokep_work> flo-retina: Both. :)
21:53:18 <flo-retina> ahah
21:53:18 <clokep_work> Mostly cause I can never remember how it works.
21:53:44 <flo-retina> so for the message api changes, I think it would be better to handle CTCP as a separate patch
21:54:01 <flo-retina> but it would be nice to be sure that the current approach isn't fully incompatible with handling CTCP as a follow-up
21:54:48 <flo-retina> clokep_work: re bug 1034971. I don't know an example of where it does NOT happen :-/.
21:54:50 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1034971 nor, --, ---, nobody, NEW, Regression: IRC topic not always settable on freshly-joined channels
21:55:53 <clokep_work> flo-retina: So I definitely fixed all of this recently. What changed it? What broke it?
21:56:02 <clokep_work> flo-retina: Sure. I want to define WHAT the expecation is with CTCP.
21:56:05 <clokep_work> expectation.
21:56:08 <clokep_work> What do other clients do?
21:57:26 <flo-retina> you mean with CTCP and OTR?
21:58:44 <clokep_work> Yes.
21:59:18 <flo-retina> I'm not sure any of them handle it properly
21:59:26 <flo-retina> wait, are you talking about /me, or about colors?
22:01:31 <clokep_work> Both.
22:01:42 <clokep_work> Although colors aren't "really" CTCP it seems.
22:02:20 <flo-retina> so for colors, I think we shouldn't bother about it
22:02:32 <flo-retina> it seems unlikely that any other IRC client handles both colors and OTR
22:02:58 <flo-retina> for /me, I'm pretty sure it's been "fixed" for Pidgin-otr, so there must be an answer to your question (I don't have it though)
22:06:54 <clokep_work> flo-retina: Does Pidgin just strip colors?
22:07:21 <flo-retina> does pidgin support sending outgoing colors in irc messages?
22:16:10 <clokep_work> What about receiving them?
22:16:17 <clokep_work> I'm mostly concerned we haven't thought about the reciving side enough.
22:17:08 <flo-retina> as far as colors are concerned, we are not sending them, right? So we are only thinking about the receiving side.
22:17:24 <clokep_work> Sure
22:17:39 <clokep_work> By "colors" you also mean bold/italic/etc., right? :)
22:29:47 <flo-retina> yeah
22:45:31 <-- mconley has quit (Broken pipe)
22:50:29 <-- Rym has quit (Ping timeout)
22:50:36 --> mconley has joined #instantbird
22:53:38 --> aleth has joined #instantbird
22:53:38 * ChanServ sets mode +o aleth 
22:54:17 <aleth> clokep_work: Funny, I can reproduce that topic bug every time #instantbird autojoins ;)
22:54:42 <clokep_work> Bleh bit rot...
22:54:54 <clokep_work> aleth: OK, whne the hell did this happen though?
22:54:57 <clokep_work> It WAS working.
22:54:58 <clokep_work> What broke it?
22:55:03 <clokep_work> I know for sure it was.
22:55:16 <aleth> race condition?
22:55:42 <aleth> My guess in the bug means it would depend on when we receive our own nick in NAMES or something like that
22:56:31 <aleth> But I haven't looked at it closely
22:57:08 <aleth> However, I'm sure we used to send an update-topic-state notification after receiving all the nicks.
22:58:01 <clokep_work> aleth: Also is it broken in TB 31?
22:58:14 <aleth> No idea
22:58:17 <clokep_work> aleth: It does not depend on that.
23:12:28 <aleth> clokep_work: so I added some dump()s... http://mxr.mozilla.org/comm-central/source/chat/protocols/irc/irc.js#541 is indeed the problem. When it happens, the participant isnt there when the mode arrives.
23:22:54 --> Rym has joined #instantbird
23:27:49 * Fallen is now known as Fallen|away
23:34:10 * aleth wishes the nightly build issues were as easy to diagnose :-/
23:37:20 <-- aleth has quit (Quit: exit stage left)
23:39:01 <clokep_work> aleth: So what? We need to always send an update message at the end again?
23:40:44 <-- clokep_work has quit (Ping timeout)
23:58:52 <-- Armada has quit (Connection reset by peer)