#instantbird log on 06 12 2014

All times are UTC.

00:01:43 <sawrubh|ib> yes, I should have followed my hunch http://log.bezut.info/instantbird/140611/#m421 but Mic figured it out thankfully!
00:05:31 <sawrubh|ib> btw how do I enable a sound notification for *only* mentions?
00:06:12 * nhnt11 can finally asnwer "Is there a clever way around this?" (aleth asked in review comments for bug 955292) with "Yes"
00:06:14 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=955292 enh, --, ---, nhnt11, ASSI, Read/write chat logs asynchronously
00:06:15 <sawrubh|ib> Also a nice feature to have would be notification popups like HTML5 notifications that come when a new email comes
00:06:49 <-- nhnt11 has quit (Input/output error)
00:06:58 --> nhnt11 has joined #instantbird
00:07:11 <nhnt11> Wow, my firefox and Ib just crashed at the same time :S
00:07:33 <-- Mic has quit (Quit: Instantbird -- http://www.instantbird.com)
00:07:33 <nhnt11> sawrubh: send me a ping please?
00:07:46 <sawrubh|ib> nhnt11: bling
00:07:58 <nhnt11> sawrubh|ib: Yeah so I have all sounds turned off in prefs and still got a sound for that
00:08:02 <nhnt11> so there's your answer ;)
00:08:14 <nhnt11> sawrubh|ib: OH! That may be an OS X feature
00:08:28 <nhnt11> I get a sound for every notification that goes through the OS X notifications API
00:08:41 <nhnt11> (That's controlled via an OS system preference)
00:09:29 <nhnt11> Yeah it is: http://puu.sh/9pAaf/f0354b344f.png
00:11:03 <sawrubh|ib> you mean you get those banner alerts on mentions, sweet!
00:12:01 <sawrubh|ib> but these 2 features should be available for Linux people too, I think
00:12:33 <nhnt11> sawrubh|ib: Does ubuntu/unity/whatever have an integrated notifications system?
00:12:39 <nhnt11> (I really don't know)
00:13:06 <sawrubh|ib> I think Unity, not sure about Gnome
00:13:22 <sawrubh|ib> but Gnome also has this little popup thingy which comes at the bottom
00:13:48 <sawrubh|ib> anyhoo, I should get back to work, I've posted teh two features requests here so someone interested might take a look
00:14:13 <nhnt11> Hmm. Make it work? :)
00:14:27 <nhnt11> sawrubh|ib: File a bug if it doesn't exist
00:14:41 <nhnt11> lest your requests be forgotten
00:21:58 <clokep> flo-retina, nhnt11: Another idea for the database is figuring out what dictionary to use when talking in a MUC.
00:22:25 <nhnt11> clokep: Dictionary? For spell check you mean?
00:23:06 <instantbot> New Chat Core - General bug 1024235 filed by saurabhanandiit@gmail.com.
00:23:08 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1024235 nor, --, ---, nobody, NEW, Ability to turn on sound alert only for mentions
00:24:08 <clokep> nhnt11: Yes. . .
00:26:29 <instantbot> New Chat Core - General bug 1024237 filed by saurabhanandiit@gmail.com.
00:26:30 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1024237 nor, --, ---, nobody, NEW, HTML5 like notification toasts
00:27:17 <clokep> sawrubh|ib: https://addons.instantbird.org/en-US/instantbird/addon/306 might do what you want
00:28:02 <nhnt11> clokep: Looks like that needs to be updated for 1.6a1pre?
00:28:16 <nhnt11> Seems ancient.. october 2011
00:28:26 <nhnt11> Maybe it'll work though :)
00:29:29 <sawrubh|ib> seems that still doesn't provide sound *just* for the mentions, it has incoming/outgoing messages, not mentions
00:30:00 <-- mayanktg has quit (Ping timeout)
00:30:02 * sawrubh|ib plans on going to sleep soon
00:30:09 <-- nhnt11 has quit (Ping timeout)
00:30:12 --> nhnt11 has joined #instantbird
00:31:09 --> mayanktg has joined #instantbird
00:32:37 <clokep> nhnt11: Depends how much the APIs have changed. ;)
00:32:53 <clokep> So you want to disable sounds for incoming messages but keep it for mentions? Interesting.
00:37:45 <sawrubh|ib> yes, right now I have sound turned off, and I've to keep checking IB for bolded out lines (telling me I was mentioned)
00:38:09 <sawrubh|ib> also another feature which would be cool to have is the ability to have these little popups like how many unread mentions you have
00:38:21 <clokep> "these little popups?
00:38:40 <sawrubh|ib> and as soon as you focus on the window (of the conversation) where you were mentioned, it would go
00:38:43 <clokep> Do you have the flashing in the taskbar disabled? Is that a Windows only feature?
00:39:01 <clokep> You can also turn on pop ups that show the message.s
00:39:05 <sawrubh|ib> popups == like the red numbers we have on Google telling us about notifications
00:39:19 <clokep> "we have on Google" . .. I don't know what you're talking about. :P
00:39:27 * sawrubh|ib sends a screenshot
00:40:58 <nhnt11> sawrubh|ib: That addon does what you want
00:41:04 <nhnt11> There's a checkbox for "alerts"
00:41:56 <nhnt11> According to a review from March 2014 it "doesn't work at all" though
00:42:01 <sawrubh|ib> nhnt11: alert messages means mentions, well if yes then that addon author needs to make it more clear ;)
00:42:20 <nhnt11> sawrubh|ib: Not everyone speaks English fluently, you have to extrapolate ;)
00:42:56 <clokep> nhnt11: I suspect that means it doesn't install.
00:42:59 <nhnt11> Er, that's an awkward thing to say if the author's first language is english and he indeed just phrased it badly
00:43:09 <nhnt11> clokep: Yeah, that's what I think too
00:43:23 <sawrubh|ib> clokep: I meant the little red bubble (with numbers) that comes up besides the bell icon http://imgur.com/ZWTXz1l
00:43:26 * clokep starts a TB build
00:43:32 <sawrubh|ib> which tells you the number of unread notification
00:43:37 <clokep> That has like no context. :-S
00:43:37 <nhnt11> sawrubh|ib: So you meant Google Plus
00:43:48 <clokep> Ah, well where would you want that to show up?
00:43:55 <clokep> Mac has a badge on the icon that shows unread messages.
00:43:58 <clokep> We should add one on Windows too.
00:44:08 <nhnt11> clokep: Isn't that a Mac feature?
00:44:15 <nhnt11> Does Windows allow such things?
00:44:30 <clokep> nhnt11: Didn't I just say that we have it on Mac? :-S
00:44:39 <clokep> nhnt11: Windows >= 7 let you put overlays on top of icons.
00:44:59 <nhnt11> clokep: No I meant, isn't that an OS level feature. And I didn't know that about windows, good to know.
00:45:08 <sawrubh|ib> don't you forget Linux in all of this! :P
00:45:16 <clokep> nhnt11: Yes, it is. I'm saying we should support Windows.
00:45:21 <clokep> sawrubh|ib: Patches accepted. ;)
00:45:35 <clokep> sawrubh|ib: The issue with Linux is that...all window managers do this stuff differently.
00:45:37 <-- Rym has quit (Ping timeout)
00:45:43 <clokep> Like there's some Ubuntu thingy for IM clients, some Debian thing, etc.
00:45:47 <clokep> s/Debain/Gnome/
00:46:23 * sawrubh|ib wonders if people have written little abstraction layers to deal with this on Linux
00:47:02 <Mook_as> as far as I can tell the gnome thing is, err, telepathy these days? which is more of a IM client/framework than just UI...
00:47:43 * sawrubh|ib just realized the Wold Cup is starting soon
00:47:58 <-- nhnt11 has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
00:48:04 --> nhnt11 has joined #instantbird
00:48:15 <-- Mook_as has quit (Quit: Mook_as)
00:48:21 <nhnt11> sawrubh|ib: The addon works if you change the version limit in install.rd
00:48:22 <nhnt11> rdf*
00:48:46 <nhnt11> Well, appears to work
00:51:24 --> Rym has joined #instantbird
00:52:52 <nhnt11> Here's the repackaged version if anyone wants: http://puu.sh/9pCVO/f8ad3ec683.xpi
00:52:56 <nhnt11> sawrubh|ib: ^
00:53:03 <nhnt11> Not sure how long that link will last though
00:53:44 <clokep> I can probably upload that someh oww....
00:54:17 <nhnt11> Tested it, seems to work as expected
00:55:50 <clokep> nhnt11: Did you bump the version # too?
00:55:54 <nhnt11> No
00:55:57 <nhnt11> Should I?
00:57:22 <clokep> I'll do it.
00:57:30 <nhnt11> ok
00:57:32 <clokep> Oh wait.
00:57:37 <clokep> I'll just bump compatibility.
00:57:45 <nhnt11> Ok
00:58:07 <clokep> Yay.
00:58:08 <nhnt11> clokep: The compatibility on AIO says 1.5a1pre, but the install.rdf said 1.2something
00:58:14 <nhnt11> Not sure if that matters
00:58:15 <clokep> nhnt11: Refresh? :P
00:58:22 <clokep> I think it queries the website.
00:58:25 <clokep> I don't know exactly how it all works.
00:58:25 <nhnt11> ok
00:58:30 <clokep> Mic probably knows better.
00:58:39 <clokep> I try to stay away from the inner workings of AIO. ;)
00:58:47 <clokep> (Although I apparently was the last one to modify the code. . .)
00:59:01 <nhnt11> 👍
00:59:07 * nhnt11 missed that ^ :')
00:59:10 <sawrubh|ib> nhnt11: what was that?
00:59:32 <nhnt11> sawrubh|ib: Emoji thumbs up. You need to install a package on ubuntu to use it
00:59:45 <nhnt11> The thumbs up that you can use from WhatsApp and so on
01:00:08 <sawrubh|ib> I like the IB feature to grey out nicks mentioned who are offline
01:00:13 <nhnt11> sawrubh|ib: According to Wikipedia you need to install the ttf-ancient-fonts package
01:00:18 * sawrubh|ib looks up the emoji-package
01:03:38 * sawrubh|ib building IB after cobbler
01:04:21 <-- sawrubh|ib has quit (Quit: sawrubh|ib)
01:09:28 <clokep> sawrubh: bug 1024237...you can definitely do that already.
01:09:30 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1024237 nor, --, ---, nobody, NEW, HTML5 like notification toasts
01:09:54 <clokep> Oh hmm....maybe it's only for inactive windows.
01:13:38 --> mconley has joined #instantbird
01:25:55 <-- wnayes has quit (Ping timeout)
01:29:13 --> wnayes has joined #instantbird
01:32:27 <-- mconley has quit (Connection timed out)
01:32:58 --> mconley has joined #instantbird
01:47:11 <-- mayanktg has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
02:00:15 <-- flo-retina has quit (Ping timeout)
02:30:40 <-- mconley has quit (Input/output error)
02:46:39 <-- wnayes has quit (Quit: wnayes)
03:39:16 <-- clokep has quit (Ping timeout)
04:07:45 <-- nhnt11 has quit (Ping timeout)
04:08:43 --> nhnt11 has joined #instantbird
04:44:07 <-- nhnt11 has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
04:44:14 --> nhnt11 has joined #instantbird
04:46:32 --> mconley has joined #instantbird
04:48:21 <-- mconley has quit (Ping timeout)
04:52:11 <-- Rym has quit (Ping timeout)
04:56:23 --> jb has joined #instantbird
05:25:08 --> flo-retina has joined #instantbird
05:25:09 * ChanServ sets mode +qo flo-retina flo-retina 
05:25:38 <flo-retina> clokep: I'm not sure we need a database to guess the language of a MUC.
05:25:40 <flo-retina> wouldn't the latest log be enough?
05:25:45 <flo-retina> using CLD we usually get a confident result as soon as we have ~100 characters
05:30:02 <-- EionRobb has quit (Quit: Leaving.)
05:30:46 <-- flo-retina has quit (Ping timeout)
05:57:29 <-- jb has quit (Ping timeout)
05:57:36 --> jb has joined #instantbird
06:17:53 <-- jb has quit (Ping timeout)
07:09:42 --> EionRobb has joined #instantbird
07:18:20 --> mpmc has joined #instantbird
07:22:12 --> jb has joined #instantbird
07:32:33 <-- EionRobb has quit (Ping timeout)
07:35:15 --> EionRobb has joined #instantbird
08:00:18 --> sawrubh|ib has joined #instantbird
08:05:43 --> Mic has joined #instantbird
08:05:44 * ChanServ sets mode +o Mic 
08:12:15 --> gerard-majax__ has joined #instantbird
08:28:17 <sawrubh|ib> Mic: hey
08:28:43 <-- Mic has quit (Connection reset by peer)
08:30:23 <sawrubh|ib> http://i.imgur.com/A5gjAVL.png is how my notification box looks like
08:36:24 <sawrubh|ib> does the accessKey shown next to the Box and Hightail buttons look good (I don't like it personally), seems like accessKey (https://developer.mozilla.org/en-US/docs/Mozilla/Tech/XUL/Method/appendNotification) for a notificationbox can't be null, I mean even if it's null, 'N' shows up on these buttons
08:37:18 <sawrubh|ib> also pressing Shift+N or N does nothing so I'm not sure what accesskeys are meant to do, I thought it would be sort of like keyboard bindings/shortcuts
08:40:18 <-- chrisccoulson has quit (Quit: OSError: [Errno 130] Owner died)
08:40:36 --> chrisccoulson has joined #instantbird
08:42:23 --> sonny has joined #instantbird
08:47:54 --> Rym has joined #instantbird
09:09:41 --> Mic has joined #instantbird
09:09:41 * ChanServ sets mode +o Mic 
09:12:57 <sawrubh|ib> hmm, interesting so with http://pastebin.instantbird.com/734342 as the patch, my upload works when I click on the notificationbox button for a provider, while when I comment out L112 and L113, it stops working and gives an error on line L200 that 'file is null'
09:14:01 <sawrubh|ib> how can adding a Cu.reportError() call get it to work :s
09:18:57 <Mic> It's a notification bar by the way ;) A notification box is the type of xul element that can contain a notification bar.
09:21:24 <sawrubh|ib> ah, ok :)
09:22:23 <sawrubh|ib> Mic: thoughts on http://log.bezut.info/instantbird/today#m168 ?
09:23:11 <Mic> Beside "Why do the provider buttons have the same access key" ? ;)
09:24:26 <Mic> The bar is shown when you drop a file on the conversation?
09:24:32 <sawrubh|ib> yes
09:24:50 <sawrubh|ib> the code is the pastebin I gave above
09:24:52 <Mic> I'd use a different text for the bar. Less technical and maybe more specific to the situation.
09:25:32 <Mic> "Upload file '<the files leafname, without path>' using: <list of buttons>" or so.
09:26:47 <sawrubh|ib> Mic: can you tell me what accesskeys are used for? pressing Shift+N or N does nothing so I'm not sure what accesskeys are meant to do, I thought it would be sort of like keyboard bindings/shortcuts
09:27:09 <Mic> Access keys are the keys that you can use in menus and the like to access an element directly.
09:27:53 <Mic> Like "Alt+F" on Windows to open a file menu. They're usually underlined in the label of the element, except if the letter isn't contained in it. In this case it's shown in brackets behind the label.
09:27:53 * sawrubh|ib tries pressing C and checks whether that presses that button directly
09:34:09 <-- mpmc has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
09:34:30 <sawrubh|ib> Mic: so pressing 'c' which I set as the accesskey for the cancel button, does nothing for me, it just starts typing that in the textbox in the conversation window
09:34:50 --> BWMerlin has joined #instantbird
09:35:11 <sawrubh|ib> also I guess I'll just set the accesskeys as the first alphabet of the capitalized provider names ('B' for Box, 'H' for Hightail)
09:35:35 <sawrubh|ib> (although I'm not yet sure about the use or working of these accesskeys, pressing them does nothing for me)
09:36:50 --> flo-retina has joined #instantbird
09:36:50 * ChanServ sets mode +qo flo-retina flo-retina 
09:39:22 <Mic> sawrubh|ib: you might have clashing access keys if you always use the first letter. There aren't so many providers for FileLink, though, are they? Maybe the problem is rather theoretical then...
09:42:38 <sawrubh|ib> there shouldn't be many providers so using the first letter should be fine (and easier),  otherwise I could keep a simple Map between provider name and accesskey we want and then do it like that
09:43:03 <sawrubh|ib> but in that case the provider name would be hardcoded
09:43:51 <sawrubh|ib> Mic: btw I still don't understand why pressing 'c' doesn't do anything, am I supposed to remove the focus on the textbox and then click 'c' to activate/use the Cancel button
09:44:10 <Mic> Have you tried Alt+C for example?
09:45:08 <sawrubh|ib> Ah, Alt+C worked :)
09:45:24 <sawrubh|ib> err Alt+c
09:46:10 <sawrubh|ib> I guess I'll just keep it lowercase and for now use the first letter as accesskeys for the providers
09:47:12 <Mic> Access keys should match the label also in upper/lowercase sense fyi...
09:47:43 <Mic> I'm not sure if it really makes a difference but we have that convention at least.
09:48:43 <sawrubh|ib> you mean for Box, I should keep it 'B', otherwise if I keep it 'b', it'll show up in a bracket besides it?
09:51:35 <Mic> Does uploading already work by the way?
09:51:54 <sawrubh|ib> yes, it's been working for a long time
09:52:23 <sawrubh|ib> it's the preference pane and the don't break TB part which has been a pain
09:58:50 <sawrubh|ib> Mic: any idea about why adding that Cu.reportError makes it work while without it, it doesn't
10:03:36 <Mic> sawrubh|ib: on the phone.
10:05:58 <flo-retina> I don't think the UI proposed in bug 1024017 is correct :(
10:06:01 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1024017 nor, --, ---, syshagarwal, ASSI, Allow disabling chat message preview in desktop notifications
10:18:39 --> Armada has joined #instantbird
10:21:09 --> clokep has joined #instantbird
10:21:09 * ChanServ sets mode +o clokep 
10:21:37 <clokep> sawrubh|ib: I figured out your Dropbox issue.
10:30:36 <clokep> flo-retina: What UI would you lik?
10:30:54 <-- EionRobb has quit (Quit: Leaving.)
10:31:23 <sawrubh|ib> clokep++
10:32:04 <sawrubh|ib> I
10:32:14 <sawrubh|ib> (ignore that, sorry)
10:32:24 <clokep> sawrubh|ib: We no longer include the CSS for the management interface in the preferences document, it's included directely in each management.xhtml file.
10:32:29 <clokep> Which means extensions don't get it for free.
10:32:57 <sawrubh|ib> but I couldn't have included it in the preference document coz it won't get inheited by an iframe
10:33:05 <clokep> But it used to wokr.
10:33:08 <clokep> Right?
10:33:11 <clokep> that's how it used to be done.
10:33:31 <clokep> I'm not saying the change you made was wrong or bad, just identifying the issue. :)
10:33:36 <sawrubh|ib> yeah, that's true :s
10:33:53 <flo-retina> clokep: I don't know. I just feel adding yet another checkbox that address only part of the issue is suboptimal.
10:34:31 --> mpmc has joined #instantbird
10:35:09 <clokep> flo-retina: I kind of complained about adding options in #maildev yesterday and I got funny looks.
10:35:16 <clokep> Well I assume, I couldn't see their eyes. ;)
10:35:33 <clokep> By "part of the issue" you mean showing the names still and such?
10:35:41 <flo-retina> yeah
10:35:56 --> mayanktg has joined #instantbird
10:39:36 <clokep> sawrubh|ib: Btw that UI with the notificationbar looks reasonable as a first cut. :) I won't say it's super pretty, but it shuld be usable.
10:42:03 <sawrubh|ib> :)
10:42:27 <sawrubh|ib> btw I'm still facing the issue I highlighted http://log.bezut.info/instantbird/today#m177 but you probably want to take a look once you reach work
10:45:37 <clokep> sawrubh|ib: Is it actually both of them? My guess is you can comment out line 112 and 113 might actually be doing something
10:46:24 <clokep> Hmm.... :-S
10:46:25 <clokep> I'm not sure.
10:46:42 <clokep> Try moving the reportError calls around first?
10:47:00 * clokep wonders if it's a race condition.
10:47:27 --> mconley has joined #instantbird
10:47:29 <-- Rym has quit (Ping timeout)
10:47:32 <sawrubh|ib> ok
10:48:27 --> Rym has joined #instantbird
10:49:14 <-- mconley has quit (Ping timeout)
10:52:05 <flo-retina> is bug 1024235 not something that already has a hidden pref?
10:52:07 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1024235 nor, --, ---, nobody, NEW, Ability to turn on sound alert only for mentions
10:54:17 <-- sawrubh|ib has quit (Quit: sawrubh|ib)
10:54:34 --> sawrubh|ib has joined #instantbird
10:59:15 <sawrubh|ib> so "Notify of messages received in inactive windows" seems to be what I wanted when I filed bug 1024237
10:59:18 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1024237 nor, --, ---, nobody, NEW, HTML5 like notification toasts
10:59:28 <sawrubh|ib> it notifies me when I'm mentioned, so I'll go close the bug now
11:00:04 <-- clokep has quit (Ping timeout)
11:01:11 <-- sawrubh|ib has quit (Ping timeout)
11:02:05 --> sawrubh|ib has joined #instantbird
11:02:37 <sawrubh|ib> ok, so now if we just get support for a)number of unread messages in a scrollback b)sound notification only when mentioned c)a little bubble somewhere telling me how many unread mentions I have in the scrollback, I think Instantbird would be my bestest IRC client
11:07:29 <instantbot> saurabhanandiit@gmail.com changed the Resolution on bug 1024237 from --- to WORKSFORME.
11:07:30 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1024237 nor, --, ---, nobody, RESO WORKSFORME, HTML5 like notification toasts
11:32:10 <-- jb has quit (Ping timeout)
11:36:14 <sawrubh|ib> hmm, that error seems to be some race condition or something but something is surely wrong, because now that I've changed the code to http://pastebin.instantbird.com/734407 the code works even when I've removed the Cu.reportError calls, however one interesting thing to note is that in the place of the dataTransfer.mozDataAt being called in the Cu.reportError earlier, now it's getting called on L214 so that might be why even after
11:36:14 <sawrubh|ib> removing the Cu.reportError calls it works
11:42:43 <-- sawrubh|ib has quit (Ping timeout)
11:43:29 --> sawrubh|ib has joined #instantbird
11:45:24 <sawrubh|ib> also just wondering, the link in the topic on irc isn't clickable, is that intentional or a bug?
11:45:48 <sawrubh|ib> nor am I able to select and copy it to clipboard
11:55:50 --> clokep_work has joined #instantbird
11:55:51 * ChanServ sets mode +o clokep_work 
11:56:19 <clokep_work> sawrubh|ib: Those are bugs, probably both on file.
11:56:47 <-- mpmc has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
11:57:49 <-- BWMerlin has quit (Quit: BWMerlin)
12:00:10 --> jb has joined #instantbird
12:02:24 <-- mayanktg has quit (Ping timeout)
12:04:28 <sawrubh|ib> clokep_work: I think I understand why that Dropbox issue is coming and how I *can* fix it :P So earlier, the management.xhtml's *didn't* get styled by inheriting the styling from what was included in the preferences document, they still used to get styled by explicitly including the CSS file (which turns out to be the same one as the one included in the preferences document)
12:04:34 <clokep_work> That new patch from arlo is definitely an r- from this line: let UIConv = Services.conversations.getUIConversation(this); inside of irc.js :-\
12:04:52 <clokep_work> sawrubh|ib: Alright.
12:05:02 <sawrubh|ib> what I did was correct until the point that I name my file cloudfile.css
12:05:09 <clokep_work> So it needs to be fixed on the Dropbox side? That's OK w/ me.
12:05:28 <sawrubh|ib> that's why it's breaking the addons, right now the path is 'chrome://messenger/skin/preferences/preferences.css', with my change the path is 'chrome://messenger/skin/preferences/cloudfile.css'
12:05:48 <sawrubh|ib> only the name is different, the addons expect it to be preferences.css, so I rename my file, things should work
12:05:54 <flo-retina> clokep_work: that line looks pretty bad indeed
12:07:03 * sawrubh|ib tries to rename and css and see
12:07:11 <sawrubh|ib> s/and/the
12:13:54 --> mayanktg has joined #instantbird
12:17:51 <-- jb has quit (Ping timeout)
12:19:56 <-- mayanktg has quit (Ping timeout)
12:20:50 --> mayanktg has joined #instantbird
12:26:46 --> rosonline has joined #instantbird
12:29:14 <sawrubh|ib> gah, renaming cloudfile.css to preferences.css broke everything now
12:29:30 * sawrubh|ib checks if he's done anything stupid
12:39:51 <sawrubh|ib> ah, ok, so when I've renamed cloudfile.css to preferences.css, it's path clashes with the preferences.css mail has in it's themes directory, and hence it doesn't get applied
12:43:42 <sawrubh|ib> ok, so the problem is that we need to have the styling in the preferences.css in mail's themes directory, A) if we do that then *everything* (Dropbox, Box etc) in TB would work but *nothing* in IB would work since it won't be able to access it and B) if we don't do that and keep the situation as it is now, Dropbox addon will break in both TB and IB, however the rest of the providers will work in *both* TB and IB
12:43:48 --> Guido has joined #instantbird
12:44:16 <sawrubh|ib> clokep_work: seems like we'll have to break Dropbox then, as you said
12:46:10 <sawrubh|ib> if we agree (which looks like we do) then I'll address your comments in https://bugzilla.mozilla.org/show_bug.cgi?id=1014644#c31 and then send it for the review this patch has been waiting for :)
12:46:13 <instantbot> Bug 1014644 nor, --, ---, saurabhanandiit, ASSI, Make FileLink work in Instantbird
12:47:10 * sawrubh|ib comments on the bug about the status of Dropbox breakage
12:48:12 --> qheaden has joined #instantbird
12:51:09 <clokep_work> sawrubh|ib: My inclination is to fix the add-ons in this case.
12:51:45 <sawrubh|ib> should I talk to mconley about it?
12:52:01 <sawrubh|ib> or is the addon's source code available somewhere?
12:52:01 <clokep_work> No.
12:52:05 <clokep_work> It's on github.
12:52:15 <clokep_work> You should do nothing for now.
12:52:24 <clokep_work> Well, besides identify what's broken.
12:52:32 <clokep_work> But once it lands we'll need to fix the add-on, yeah.
12:53:04 <sawrubh|ib> https://github.com/mikeconley/thunderbird-filelink-dropbox aha!
12:53:12 <clokep_work> Yes.
12:53:38 <sawrubh|ib> so are we good to go ahead with the review then?
12:53:42 <sawrubh|ib> (for now)
12:55:36 <sawrubh|ib> https://github.com/mikeconley/thunderbird-filelink-dropbox/blob/master/content/chrome/management.xhtml#L19 aha!
12:57:23 * sawrubh|ib sent a PR to the repo with the fix
12:58:35 <clokep_work> sawrubh|ib: Please put it for review, yes.
12:59:26 <clokep_work> I'd wait until Mike reviews this patch and says it's OK to move that stuff out of preferences.css.
13:00:06 * sawrubh|ib puts it up for flo's review first
13:00:24 * flo-retina thinks flo's reviews may take a while
13:01:57 <clokep_work> sawrubh|ib: You should put it for MY review first. :P
13:02:01 <sawrubh|ib> flo-retina: do you get pinged if I write flo, just curious? I hope you don't
13:02:09 <sawrubh|ib> clokep_work: ah, sorry
13:02:09 <flo-retina> I was reading
13:02:13 <flo-retina> no, I don't
13:03:16 --> aleth_web has joined #instantbird
13:03:43 <sawrubh|ib> yay, aleth_web is here!
13:05:27 <-- Guido has quit (Quit: Instantbird 1.5 -- http://www.instantbird.com)
13:06:21 <aleth_web> nhnt11: http://log.bezut.info/instantbird/140611/#m656 hey, that's the one database thing I did tell you about! ;)
13:12:23 <sawrubh|ib> clokep_work: btw what about the other addons (https://addons.mozilla.org/en-us/thunderbird/search/?q=filelink&appver=&platform=) we break
13:12:31 <sawrubh|ib> do we email them to fix it?
13:12:41 <clokep_work> sawrubh|ib: Do you even know if they're broken? :P
13:12:58 <sawrubh|ib> they will be, they /should/ be :)
13:13:04 <sawrubh|ib> unless they package their own styling code
13:13:08 <clokep_work> sawrubh|ib: Why? Only if they use the same management.xhtml.
13:13:10 <clokep_work> Yes.
13:13:18 <clokep_work> Which they probably will, but "should" is the wrong modifier. :P
13:13:47 <clokep_work> sawrubh|ib: As I've said (twice now, I think) we should get Mike to look at the code before emailing a bucnh of people say that something MIGHT change.
13:18:10 <aleth_web> sawrubh|ib: What actually breaks for the addons?
13:18:18 <aleth_web> What don't they find?
13:18:35 <sawrubh|ib> the styling of the management iframe inside the preference pane
13:19:03 <sawrubh|ib> they're not able to find the management.xhtml styling code (as I've said in the bug comment) which I moved to cloudfile.css
13:20:28 <aleth_web> Ask clokep whether it is preferable to duplicate the CSS in TB to avoid the breakage (i.e. keep a copy of the CSS in TB's preferences.css where it is now, if that's all that the addons need to work)
13:20:56 <sawrubh|ib> clokep_work: ^2
13:21:08 <clokep_work> aleth_web: We already had this conversation.
13:21:18 <aleth_web> OK, sorry! 
13:21:19 <clokep_work> I've already told sawrubh|ib what I prefer he do.
13:21:23 * aleth_web should read the logs
13:23:16 <sawrubh|ib> gah DOMi, y u no work with TB
13:23:59 <aleth_web> It does work.
13:25:11 <sawrubh|ib> I'm trying to inspect the DOM node in the preference panel on the TB I've compiled and it doesn't show the preference panel in the list of available Chorme Documents
13:25:21 --> mconley has joined #instantbird
13:25:50 <sawrubh|ib> moreover for the ones it does list, it doesn't recognize the DOM nodes, only their stylesheets
13:26:43 <aleth_web> Isn't the pane contents in an iframa and therefore content?
13:27:28 <aleth_web> Are you using the lates version of DOMi (2.0.14)?
13:30:28 <aleth_web> nhnt11: http://log.bezut.info/instantbird/140612/#m2 sounds good! :)
13:32:09 <sawrubh|ib> yes, I'm using 2.0.14 and previously in IB I've inspected the pane contents under Chrome documents itself
13:32:59 <aleth_web> Seems to work in my TB build at least.
13:33:05 <sawrubh|ib> btw I'm trying to debug the font and font size different between http://i.imgur.com/A6CpUgt.png?1 (without my changes) and http://i.imgur.com/LxsEx5h.png?1 (with my changes)
13:33:14 <sawrubh|ib> s/different/difference
13:33:59 <clokep_work> sawrubh|ib: preferences.css probably has a rule that cloudfile.css is missing.
13:35:49 <-- sawrubh|ib has quit (Connection reset by peer)
13:42:08 <mayanktg> aleth_web: The SDP creation is much more complex :-/ . I've to change the structure of |m=video 33680 RTP/SAVPF 120| when there are no "rel-port" attr present in *any* of the candidate :'( . The port address is otherwise the "port" attribute of the first candidate.
13:42:24 <mayanktg> Ahh..I can do this.
13:42:29 <mayanktg> :)
13:42:55 <aleth_web> mayanktg: Seems like the sdp <-> xml is really the hard part of getting this to work ;)
13:43:29 <aleth_web> Make sure you put lots of comments in the code so you can remember why you did what in a couple months.
13:44:16 <mayanktg> Yeah..I'm pretty sure I'll have to change the code as we start to get more and more SDP offer samples.
13:46:06 --> sawrubh|ib has joined #instantbird
13:46:25 <aleth_web> Just add the new samples to the test and work from there.
13:52:25 * sawrubh|ib moves on to fixing that weird race condition in the drag and drop patch
13:52:49 <sawrubh|ib> aleth_web: interested in having a try at a weird bug?
13:53:12 <aleth_web> You'd have to be more specific.
13:56:39 <sawrubh|ib> aleth_web: http://log.bezut.info/instantbird/today#m177
13:58:20 <aleth_web> Unlikely to be a race condition, this is all synchronous, isn't it?
13:59:08 <sawrubh|ib> dunno, I mean I drop a file, a notification bar pops up, it waits until something is clicked and then uploads the file if you chose a provider
13:59:15 <sawrubh|ib> sounds synchronous to me
14:00:29 <sawrubh|ib> aleth_web: then here come the next piece of information : http://log.bezut.info/instantbird/today#m268
14:00:49 <Mic> aleth_web, sawrubh|ib: other option would be that a function call changes the state of an object but it doesn't look like that either.
14:02:25 <aleth_web> It's possible... I don't have any experience with DataTransfer unfortunately.
14:02:37 <sawrubh|ib> :s
14:02:37 <aleth_web> Mic: Neat idea to optimize the SVG :)
14:04:47 <aleth_web> sawrubh|ib: Work on something else for a bit then come back to it, to avoid just staring at the code.
14:05:50 <sawrubh|ib> ok
14:06:49 <aleth_web> Then maybe comment out all the complicating filelink factors and just see if you can read the data you need on drop.
14:06:54 --> jb has joined #instantbird
14:08:21 <-- mayanktg has quit (Ping timeout)
14:10:58 <clokep_work> sawrubh|ib: Did you re-read the docs on that object?
14:11:46 <sawrubh|ib> clokep_work: you mean DataTransfer?
14:16:42 --> mayanktg has joined #instantbird
14:18:24 <clokep_work> sawrubh|ib: Yes.
14:26:19 <sawrubh|ib> No, I didn't. I'll do that.
14:33:52 <-- clokep_work has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
14:33:53 --> clokep_work has joined #instantbird
14:33:53 * ChanServ sets mode +o clokep_work 
14:39:49 <-- sawrubh|ib has quit (Ping timeout)
14:40:43 --> sawrubh|ib has joined #instantbird
14:41:56 <-- mconley has quit (Ping timeout)
14:46:03 <-- sawrubh|ib has quit (Ping timeout)
14:46:12 --> sawrubh|ib has joined #instantbird
14:48:46 <-- sawrubh|ib has quit (Ping timeout)
14:49:58 --> sawrubh|ib has joined #instantbird
14:51:37 <-- sawrubh|ib has quit (Connection reset by peer)
14:55:45 --> mconley has joined #instantbird
14:55:49 --> sawrubh|ib has joined #instantbird
15:17:11 <-- sonny has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
15:17:14 --> sonny has joined #instantbird
15:25:05 <-- Mic has quit (Quit: Instantbird -- http://www.instantbird.com)
15:30:18 <nhnt11> aleth_web: "I don't think this prevents trying to write the header multiple times if we are
15:30:18 <nhnt11> unlucky and logMessage is called rapidly in succession." What?
15:30:25 <nhnt11> How would the header be written multiple times?
15:30:40 <nhnt11> If it failed once, closing the log writer results in a new file being created next time
15:31:06 <nhnt11> If logMessage is called multiple times quickly, still shouldn't be a problem
15:31:23 <nhnt11> Because the calls will be in order and the first one would have written the header
15:31:30 <nhnt11> Or did you mean if they're queued, and the first one faild?
15:31:32 <nhnt11> fails*
15:32:07 <aleth_web> Until the header is written (succes or fail), initialized is false, and each Task.async call will add a "write header" call to the queue.
15:32:57 <nhnt11> aleth_web: But the _initialized variable will be checked when the task actually runs
15:33:29 <aleth_web> And there's nothing stopping that task from starting to run.
15:33:42 <aleth_web> Anyway, isn't my suggestion simpler?
15:33:53 <nhnt11> My point is that tasks will be run in sequence, so it doesn't matter
15:33:56 <nhnt11> Yeah okay.
15:34:30 <aleth_web> nhnt11: The tasks are run in sequence, but the initial header write may take half a second, so the second task might run well before that finishes ;)
15:35:00 <nhnt11> ....Ew.
15:35:01 <aleth_web> You're not chaining those tasks after all.
15:35:11 <nhnt11> I just realized yielding => the next task can run
15:35:18 <aleth_web> Right!
15:35:47 <aleth_web> The idea behind my suggestion is simply "let's just use the queue we have for this"
15:35:54 <nhnt11> Yeah, cool
15:36:03 <aleth_web> Now it seems possible to do so after your simplifications in the last patch.
15:36:30 <nhnt11> aleth_web: Btw, I seem to have forgotten or never seen that you suggested I use sqlite.jsm :(
15:36:43 <aleth_web> Just the other day in a PM ;)
15:37:12 <aleth_web> Anyway, it doesn't matter, I'm glad you are having those discussions with flo now.
15:37:38 <nhnt11> Oh yeah, you mentioned it :]
15:37:38 <nhnt11> Yeah
15:38:19 <nhnt11> Btw flo-retina, we should probably store the entire contents of the log in one row like you mentioned you did with gloda yesterday.
15:39:07 <nhnt11> If we want a query to give us a log file, We can just read every message from the log, join them with "\n", and store that against the filename
15:39:22 <nhnt11> We don't need any metadata since we'll be reading the logfile anyway
15:40:09 <nhnt11> So yeah, it should be pretty simple and work well with the existing API. What it won't really do is help with infinite scroll.
15:40:38 <flo-retina> infinite scroll needs to read log files
15:41:01 <flo-retina> and I think both for the database and for the infinite scroll, you'll want to limit the size of log files, and make them smaller than they currently are
15:41:40 <nhnt11> "infinite scroll needs to read log files." ..I know :S
15:41:50 <-- gerard-majax__ has quit (Ping timeout)
15:42:00 <nhnt11> I just meant it can't quickly query the database for the next set of messages
15:42:37 <nhnt11> Limiting the size of new log files shouldn't be too difficult, but maybe it'll be tricky to re-organize existing logs
15:43:08 <aleth_web> Let's not reorganize existing logs unless it's really unavoidable.
15:43:33 <nhnt11> Yeah, scary stuff
15:44:47 <aleth_web> nhnt11: Could you put your ideas for how the database and its API will look in an etherpad when they clarify for you? (before implementing more than experiments)
15:44:59 <nhnt11> Ok
15:53:34 <-- rosonline has quit (Quit: Instantbird 1.5 -- http://www.instantbird.com)
15:53:41 --> rosonline has joined #instantbird
15:55:22 <nhnt11> bbiab
16:06:19 <mayanktg> aleth_web: I have sent the offer, but how should I make the callee send the answer SDP by calling __.webrtcCallRequest(aData)? http://pastebin.instantbird.com/734484 
16:07:00 <aleth_web> I need more context
16:07:15 <mayanktg> Ok. 
16:07:43 <aleth_web> Isn't the notification handler there something that was already working?
16:08:34 <mayanktg> http://pastebin.instantbird.com/734523
16:09:35 <aleth_web> Doesn't this.webrtcCallRequest work? The method you are trying to call is in the same binding.
16:10:07 <mayanktg> Wait.. I tried that before. I'm trying again ..
16:11:44 <mayanktg> aleth_web: Using this.webrtcCallRequest gives the error |body is null| 
16:12:02 <aleth_web> Where does it give that error?
16:12:34 <mayanktg> In the initPeerConnection method.
16:13:03 <mayanktg> Should I share the diff of the patch where I defined initPeerConnection too?
16:13:04 <aleth_web> So that means webrtCallRequest is being called correctly, doesn't it.
16:13:15 <mayanktg> Yes. 
16:13:54 <aleth_web> You'll have to look at initPeerConnection to figure out what's going wrong.
16:14:57 <mayanktg> But when we call the startCall method..when  creating, I receive no such error. Its only when I try to create an answer I'm receiving the error.
16:15:08 <mayanktg> *when creating offer
16:15:38 <aleth_web> Does the error still happen if you make sure there is already a conversation open with that contact? (exchange a few messages before sending the call request)
16:16:15 <mayanktg> Ok.Let me try.
16:17:56 <sawrubh|ib> when does one an IDL file imI*.idl vs prplI*.idl, I can see interface names starting with 'prplI' even in those imI*.idl files
16:18:02 <sawrubh|ib> +call
16:18:12 <mayanktg> aleth_web: The error is coming even after I'm having a text conversation between the two users --> starting the call again.
16:18:21 <-- Rym has quit (Ping timeout)
16:18:34 <mayanktg> http://pastebin.instantbird.com/734524 pastebin of the patch that contains initPeerConnection.
16:18:57 <clokep_work> prpl = stuff to be implemented by protocols, im = instantbird stuff
16:19:57 <aleth_web> mayanktg: Put some dumps after each line in initPeerConnection to find oumt exactly what goes wrong. e.g. on line one, dump(this.browser) to see if it's defined as it should be, etc
16:20:04 <sawrubh|ib> ok, then I'll call my idl prpl then, prplIFileTransfers.idl :)
16:20:37 <mayanktg> aleth_web: The message stanza is properly exchanged btw. I just verified from the debug log. 
16:20:46 <mayanktg> Ok. I'm adding the dumps.
16:20:50 <aleth_web> Yeah, this will have nothing to do with your backend code.
16:21:00 <sawrubh|ib> aleth_web: talking to me?
16:21:08 <aleth_web> sawrubh|ib: no
16:21:25 <sawrubh|ib> pfff, this has happened twice to me in two days
16:21:30 * sawrubh|ib goes away
16:21:38 <aleth_web> mayanktg: Your goal is to find out what is different between the two times you call initPeerConnection
16:22:27 <clokep_work> sawrubh|ib: What does it do?
16:22:38 --> Rym has joined #instantbird
16:23:53 <sawrubh|ib> clokep_work: I'm not sure, defines http://xmpp.org/extensions/xep-0096.html#registrar-querytypes-sendfile probably
16:24:09 <sawrubh|ib> I'm still a bit confused about the starting point here:s
16:25:47 <sawrubh|ib> I've read XEP96 and I understand what's happening in it
16:27:10 <sawrubh|ib> but how do I code it up?
16:27:25 <mayanktg> aleth_web: Ahh. The first time initPeerConnection is called this.browser.contentDocument is |[object HTMLDocument]| but the next time it is |[object XrayWrapper [object HTMLDocument]]|
16:29:27 <aleth_web> mayanktg: and then? where does it go wrong?
16:30:37 <mayanktg> aleth_web: We are calling startCall method (when we create an offer) from the <command> defined in instantbird.xul.
16:31:05 <aleth_web> Yes, I want to know which line in initPeerConnection doesn't do what it should.
16:31:30 <mayanktg> which uses getTabBrowser().selectedConversation.startCall().
16:31:31 * sawrubh|ib will read up XEP95 after dinner
16:32:08 <aleth_web> mayanktg: So try dump(this) as well in initPeerConnection if you are concerned about what "this" is pointing to.
16:32:09 <mayanktg> |body   = doc.getElementById("ibcontent");| body is null in the second time.
16:32:15 <mayanktg> Ok.
16:36:20 <mayanktg> aleth_web: "this" is pointing to conversation. 
16:37:03 <-- jb has quit (Ping timeout)
16:37:18 --> iamjayakumars has joined #instantbird
16:39:25 <aleth_web> That's a strange error, I don't see what the problem is either.
16:39:37 <mayanktg> aleth_web: the error disappears if I already have a video stream running in the second conversation. 
16:39:46 <mayanktg> ie..
16:40:03 <aleth_web> So having some messages displayed in the second conversation is not enough?
16:41:09 <mayanktg> Yeah. Its enough!
16:41:43 <mayanktg> It starts the local stream on both the user conversations :)
16:42:14 <aleth_web> What changed since I asked you earlier? (http://log.bezut.info/instantbird/140612/#m472) :-S
16:44:52 <mayanktg> aleth_web: I'm sorry. Only the conversation was left open :-/ I only added this.webrtcCallRequest() . 
16:44:57 <aleth_web> Anyway, in that case the answer is easy, it's an edge case: if the prpl has only just asked for the conversation binding to be created when it sends the webrtc notification, then the browser doesn't have time to initialize before initPeerconnection is called. This isn't a problem if the conversation already existed.
16:46:02 <aleth_web> Put a comment in webrtcCallRequest pointing out this is a problem, and move on (lets not worry about this edge case now)
16:46:20 --> jb has joined #instantbird
16:46:35 <aleth_web> Just send a couple messages before calling when testing.
16:46:48 --> jb1 has joined #instantbird
16:46:52 <-- jb has quit (Connection reset by peer)
16:46:57 <mayanktg> Ok. I'm doing that.
16:50:30 --> arlolra has joined #instantbird
16:52:44 <aleth_web> This problem should go away later when the video call UI uses its own browser element rather than the conversation browser element.
16:54:44 --> Mook_as has joined #instantbird
16:55:46 <-- jb1 has quit (Ping timeout)
16:58:18 <-- iamjayakumars has quit (Ping timeout)
17:05:38 <-- mayanktg has quit (Ping timeout)
17:05:44 --> mayanktg has joined #instantbird
17:09:56 <-- mayanktg has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
17:09:58 --> mayanktg has joined #instantbird
17:10:19 <nhnt11> flo-retina: Just to be very clear, your reasoning for giving log files a maximum size is so that we don't have to read a ton of data whenever the user scrolls up, right?
17:10:44 <nhnt11> Is there any way having a maximum size will affect the database in a positive way?
17:11:00 <clokep_work> sawrubh|ib: Not sure if you're back yet, lemme know when you are.
17:11:08 <sawrubh|ib> I'm back
17:11:58 <flo-retina> nhnt11: I think it will avoid rows of infinite size in your table
17:12:14 <flo-retina> nhnt11: it will also let you search by time
17:12:24 <flo-retina> if you have a single log file that covers several days, you can't do that
17:12:30 <sawrubh|ib> clokep_work: could we talk about exactly how should I go about this (I know I should be leading the effort) but I just wanted to have some sort of clear way (liek I did while making filelink work in IB)
17:12:33 <nhnt11> Ah, that way!
17:12:45 <nhnt11> I never thought about if a log covers several days
17:12:53 <sawrubh|ib> clokep_work: I think I'll write my understanding (which is very little at this point) in an etherpad and then we can track that
17:12:58 <clokep_work> sawrubh|ib: OK, so first of all...what's the entry point into this? We have to notify the prpls that we want to send a file, right?
17:13:03 <sawrubh|ib> like mayanktg used to do for his SDP offer or something
17:13:08 <flo-retina> nhnt11: currently that's rare... only because we use nightlies and restart everyday ;)
17:13:11 <clokep_work> sawrubh|ib: OK, that works.
17:13:18 <clokep_work> sawrubh|ib: And yes, it's very similar to the SDP stuff!
17:13:32 <nhnt11> flo-retina: But for the sake of discussion, will having large rows really have a negative impact on the database?
17:14:04 <sawrubh|ib> clokep_work: https://etherpad.mozilla.org/ib-filelink-week4
17:14:10 <aleth_web> Starting a new log file every day was in one of the early etherpads iirc, if only so as to make the current log tree give sensible results.
17:14:50 <nhnt11> I'm trying to think of the best time to update the index - when a conversation closes (and the log writer for it is closed) or for every new text?
17:16:02 <nhnt11> Doing it at every new text would mean we're guaranteed to have indexed every message and won't have to look for stuff at startup and so on. But it also means we'd have to store messages vs. filename rather than file contents vs filename
17:16:13 <nhnt11> messages vs file id, and file id vs filename of course as Mic said
17:18:10 <sawrubh|ib> clokep_work: sorry my internet's acting a bit weird
17:19:52 <clokep_work> sawrubh|ib: I only see one bullet pt there.
17:20:12 <sawrubh|ib> yeah, I got disconnected, writing up now
17:22:31 --> iamjayakumars has joined #instantbird
17:24:15 <sawrubh|ib> clokep_work: thanks, yeah questions will help me think what needs to be done :)
17:27:38 <clokep_work> aleth_web: I was about to ask who taht was in that etherpad. ;)
17:29:36 <sawrubh|ib> bbiab
17:31:17 <-- iamjayakumars has quit (Ping timeout)
17:34:07 <sawrubh|ib> back
17:40:30 <sawrubh|ib> wow, you're typing really quick
17:40:37 <clokep_work> sawrubh|ib: I type around 60 - 80 wpm.
17:43:15 <-- flo-retina has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
17:43:28 <sawrubh|ib> clokep_work: ok, so should I started sequentially or keep that magic request part later and write the offerStreamInitiation and acceptStreamInitation functions now
17:43:36 <sawrubh|ib> s/started/start
17:44:23 <clokep_work> sawrubh|ib: Either way is OK w/ me. You should answer the questions I asked first though. :P
17:44:35 <clokep_work> sawrubh|ib: Btw the "magic" is the interface.
17:45:18 <-- sawrubh|ib has quit (Ping timeout)
17:45:47 * mayanktg ... and we the first call from XMPP clients :)
17:45:53 --> sawrubh|ib has joined #instantbird
17:46:35 * aleth_web suspects mayanktg is so excited he is skipping words ;)
17:46:45 <mayanktg> *have :D
17:47:19 <mayanktg> it works!
17:47:44 <aleth_web> mayanktg: excellent news :)
17:47:48 <sawrubh|ib> mayanktg++
17:49:06 <mayanktg> The remote stream stops when net is disconnected and automatically resumes thereafter.
17:49:51 <-- qheaden has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
17:51:47 <aleth_web> Sounds like webrtc does some magic there?
17:55:24 <mayanktg> Idk why did it automatically resumed. Maybe because the connection was not disconencted between the XMPP clients . http://i.imgur.com/pBJ9Byb.png Here's the image.
17:56:55 <-- aleth_web has left #instantbird ()
17:57:41 <mayanktg> sawrubh|ib: Do you still need to faster way to upload images? You can try Shutter for Ubuntu. It crops only the selected region..add effects...uploads it to your desired image hosting client.
18:03:49 <-- sonny has quit (Ping timeout)
18:06:00 --> aleth_web has joined #instantbird
18:11:00 --> mpmc has joined #instantbird
18:14:10 <-- aleth_web has left #instantbird ()
18:20:48 <nhnt11> flo-retina, aleth: I've updated the etherpad: https://etherpad.mozilla.org/indexed-logs-infinite-scrollback
18:20:59 <nhnt11> clokep_work: Maybe you want to take a look too^ :)
18:21:36 <nhnt11> And Mic, while I'm at it^ ;)
18:21:59 <-- sawrubh|ib has quit (Ping timeout)
18:22:56 --> sawrubh|ib has joined #instantbird
18:23:33 <-- Rym has quit (Ping timeout)
18:25:30 <-- mayanktg has quit (Ping timeout)
18:30:46 --> mayanktg has joined #instantbird
18:32:50 --> Rym has joined #instantbird
18:35:17 --> sonny has joined #instantbird
18:36:57 <nhnt11> clokep_work: Can modules have interfaces?
18:37:29 <clokep_work> nhnt11: Components have interfaces.
18:37:38 <nhnt11> clokep_work: Are modules components?
18:37:42 <nhnt11> I guess they are
18:37:45 <clokep_work> No.
18:37:45 <nhnt11> Or can be
18:37:49 <clokep_work> Modules are modules.
18:37:50 <-- sonny has quit (Ping timeout)
18:37:52 <nhnt11> Alright
18:38:01 <nhnt11> Oh yeah, modules is a different folder :]
18:38:04 <clokep_work> Modules can export objects that implement interfaces.
18:38:10 <nhnt11> That works.
18:38:31 <nhnt11> If we move the log sweeping code to its own file, it will want an interface.
18:38:42 <sawrubh|ib> clokep_work: let me wrap my head around all that's been written :s
18:38:54 <clokep_work> sawrubh|ib: OK, ask if you have questions.
18:39:12 <clokep_work> nhnt11: I don't understand the difference betwen "log sweeping" and "indexing service"
18:39:17 <nhnt11> clokep_work: 
18:39:19 <nhnt11> sorry
18:39:47 <nhnt11> clokep_work: Both the stats service and the indexing service need to parse all existing logs in certain scenarios
18:40:05 <nhnt11> So I'm saying we can move the code that iterates through all the log files to a module
18:40:57 <nhnt11> brb, I need to eat.
18:46:26 --> gerard-majax__ has joined #instantbird
18:47:17 --> flo-retina has joined #instantbird
18:47:17 * ChanServ sets mode +qo flo-retina flo-retina 
19:25:09 <nhnt11> flo-retina: http://log.bezut.info/instantbird/today/#m588
19:26:17 <-- mpmc has quit (Connection reset by peer)
19:26:19 --> mpmc has joined #instantbird
19:26:47 <-- mconley has quit (Ping timeout)
19:28:01 <flo-retina> nhnt11: you don't need an idl interface just because it's a separate file
19:28:40 <flo-retina> nhnt11: could the stats service get its data from the database?
19:29:14 <nhnt11> flo-retina: 
19:29:21 <nhnt11> damn, why does that keep happening (sorry
19:29:39 <nhnt11> flo-retina: I know that, I figured it would need an interface because it would be an API
19:30:02 <nhnt11> It wouldn't really be a "utility" module
19:30:32 <nhnt11> flo-retina: It could, and in that case we'd need to have another table in the database for that.
19:31:16 <-- nhnt11 has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
19:31:23 --> nhnt11 has joined #instantbird
19:31:25 <flo-retina> so I'm not exactly sure what you are asking in that pastebin/how I can comment without making the whole document confusing
19:31:31 <clokep_work> nhnt11: I think Florian is asking if you can get it from the database w/o adding to it.
19:32:22 <nhnt11> flo-retina: I just wrote down what I currently envision for the API, and figured it would be good to have you guys approve before I proceed
19:32:33 --> mconley has joined #instantbird
19:33:06 <nhnt11> clokep_work: I have no idea what that means or how that would work.
19:33:26 <clokep_work> nhnt11: It seems to me like you're over engineering whatever you're trying to do. ;)
19:33:30 <flo-retina> nhnt11: you are talking about a log indexing service, and logger.js like if they were 2 separate things
19:33:53 <nhnt11> Ah. So you want logger.js to maintain the index as well.
19:34:06 <nhnt11> That would be pretty useless to thunderbird, no? :S
19:34:17 * nhnt11 shrugs
19:34:32 <nhnt11> If you think that's best, great. 
19:34:43 <flo-retina> pref'ed off, it wouldn't matter to TB
19:35:12 <flo-retina> nhnt11: have you seen Mic's comment yesterday about not duplicating the file paths over and over?
19:35:28 <nhnt11> flo-retina: Yes. I've mentioned it in the etherpad too.
19:36:05 <nhnt11> flo-retina: If we store the entire contents of the file in a single row, there won't be any duplication and hence no need for a second table mapping filenames to unique ids
19:36:20 <flo-retina> I disagree :)
19:36:31 <flo-retina> I would see a table with the folder paths
19:36:39 <nhnt11> Oh.
19:36:46 <nhnt11> Right.
19:36:54 * nhnt11 was thinking of filenames, bah
19:36:57 <flo-retina> and the large table with the content would have the folder id (int), the filename, and the indexed text content
19:37:13 <nhnt11> Right right.
19:38:07 <-- mpmc has quit (Connection reset by peer)
19:38:09 --> mpmc has joined #instantbird
19:38:14 <-- Rym has quit (Ping timeout)
19:38:30 --> Rym has joined #instantbird
19:38:39 <nhnt11> flo-retina: Do you have an opinion on how we update the index for new incoming messages?
19:39:00 <flo-retina> yeah
19:39:13 <flo-retina> updating when closing the log file seems fine
19:39:35 <flo-retina> and you should also update the ongoing conversation just when you are about to search it
19:39:41 <flo-retina> updating for each new message is pointless
19:39:44 <nhnt11> Right
19:39:46 <nhnt11> Cool.
19:40:09 <nhnt11> flo-retina: What about in the scenario of a crash?
19:40:11 <flo-retina> you should also have a way to detect that a log wasn't fully indexed, when you restart after a crash
19:40:14 <nhnt11> Yeah
19:40:15 <flo-retina> ahah :-
19:40:16 <flo-retina> )
19:40:28 <flo-retina> isn't this something you've already looked at before?
19:41:13 <nhnt11> With the stats service?
19:41:22 <flo-retina> indeed
19:41:30 <nhnt11> I don't think so.
19:41:43 <flo-retina> how is the stats service handling log files?
19:41:45 <flo-retina> I don't remember
19:41:54 <flo-retina> is it doing a full resweep at each startup?
19:42:23 <nhnt11> flo-retina: No. It does a full sweep if either the cache JSON file isn't present or is corrupt, or when the version has been updated.
19:42:34 <flo-retina> errr
19:42:38 <flo-retina> I meant full directory traversal
19:42:40 <nhnt11> After that, stats are updated in real time using observer notifictions, and cached at regular intervals
19:42:42 <nhnt11> Yes.
19:43:12 <flo-retina> I have a suggestion to avoid that full directory traversal, but it may be slightly overkill :-]
19:43:13 <nhnt11> It does full directory traversal only if the JSON cache is corrupt or non-existent
19:43:23 * nhnt11 is curious
19:43:47 <flo-retina> idea: whenever you are done indexing a file, add it inside a zip archive of previously indexed logs. Then delete the .json file.
19:44:07 <flo-retina> if your database ever needs to be rebuilt (corrupted; schema upgrade; whatever...) you index all the logs inside .zip files
19:44:17 <nhnt11> Ah, you mean avoid full directory traversal after a crash
19:44:21 <flo-retina> otherwise you index at startup only the .json files
19:44:34 <nhnt11> flo-retina: Right now after a crash, the stats cache JSON file isn't corrupted or anything, so we just lose stats.
19:44:36 <flo-retina> and if there's a .json file around, you know you crashed
19:45:18 <nhnt11> We lose a maximum of 10 minutes worth of stats if there's a crash. which isn't a lot
19:45:22 <nhnt11> Cool.
19:45:41 <flo-retina> btw, I'm saying zip file because there's the jar: protocol handler that would let you read them asynchronously using XHR
19:46:06 <flo-retina> without having to extract them first
19:46:13 <sawrubh|ib> is there a bug on file to open the last closed tab/conversation by doing Ctrl+Shift+T?
19:46:14 <nhnt11> flo-retina: So if the stats cache json file disappears or gets corrupt, we open up the zip file and read its contents?
19:46:23 <nhnt11> sawrubh|ib: I've got an addon for it
19:46:30 <nhnt11> sawrubh|ib: https://addons.instantbird.org/en-US/instantbird/addon/364
19:46:34 <nhnt11> There's a bug too
19:46:47 <nhnt11> bug 954981
19:46:51 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=954981 enh, --, ---, nhnt11, ASSI, Reopen recently closed tabs
19:47:13 <flo-retina> nhnt11: not sure it makes sense for you to work on that zip file idea at this point, but you were curious :)
19:47:26 <flo-retina> it could be a future improvement
19:47:30 <nhnt11> flo-retina: It's definitely something nice we could add at a later stage
19:47:30 <nhnt11> yea
19:47:31 <nhnt11> h
19:47:46 <flo-retina> and I suspect it would need to have a hidden pref to disable it, as people may want to still be able to grep
19:48:05 * nhnt11 is saddened by those people who want to ignore his gsoc project :P
19:48:12 <nhnt11> (kidding, I understand there may be use cases)
19:48:28 <sawrubh|ib> nhnt11: is the source code for these online somewhere?
19:48:45 <flo-retina> note: having the files compressed on disk would reduce the amount of disk I/O you would need to display an old conversation
19:48:53 <nhnt11> sawrubh|ib: https://bitbucket.org/nhnt11/instantbird-addons/src/4c8cc3e79ccc74e3835a67e556ba22a58488dc7e/ReopenClosedTab/?at=default
19:49:30 <nhnt11> flo-retina: Because we would unzip the whole file, we wouldn't need to do "actual" directory traversal?
19:49:37 <nhnt11> I don't know how a zip-file-reading API would work
19:49:54 <nhnt11> You can extract only the files you want, right?
19:50:15 <flo-retina> "we would unzip the whole file" uh?
19:50:27 <nhnt11> Uh, I meant, we would only need to read one file
19:50:30 <flo-retina> nhnt11: a zip file can work almost like a normal directory
19:50:37 <flo-retina> it's just that each file is compressed
19:50:40 <nhnt11> okay
19:51:01 <flo-retina> I don't think OS.File handles zip files though, that's why I said you could use XHR to access logs
19:51:10 --> jb has joined #instantbird
19:51:15 <nhnt11> Okay
19:51:26 <flo-retina> nhnt11: play a bit with the jar: protocol in Firefox's url bar :)
19:51:48 * nhnt11 now works on getting the async logs bug to land asap
19:52:07 <flo-retina> one thing to ponder: how likely would we be to lose the whole zip file (with all the previous logs with someone) if we crash while updating it?
19:52:20 <nhnt11> flo-retina: Shouldn't be a problem if we have a journal
19:52:35 * flo-retina isn't sure what that means
19:52:53 <nhnt11> flo-retina: Make a copy of the file, update it, and then rename it to the original one
19:53:01 <nhnt11> Write atomic
19:56:06 <flo-retina> nhnt11: so in that case we would need to copy the whole zip file each time we add a tiny piece of data
19:56:13 <flo-retina> that may be worth it
19:56:46 <flo-retina> would also have the interesting benefit of avoiding disk fragmentation :)
19:56:50 <-- chrisccoulson has quit (Quit: OSError: [Errno 130] Owner died)
19:57:13 <nhnt11> Hmm, yeah
19:58:03 --> chrisccoulson has joined #instantbird
20:01:34 <-- Tonnes has quit (Connection reset by peer)
20:03:10 --> Tonnes has joined #instantbird
20:15:48 <-- sawrubh|ib has quit (Quit: sawrubh|ib)
20:18:56 --> EionRobb has joined #instantbird
20:22:26 <nhnt11> Hmm, I can't find Firefox's scratchpad anymore :S
20:22:45 <nhnt11> Hmm, it's in the web developer menu in Tools
20:22:50 <nhnt11> There used to be a button on the console...
20:22:52 * nhnt11 shrugs
20:23:55 <nhnt11> Ah, there's a checkbox under "Available Toolbox Buttons"
20:33:12 <-- mconley has quit (Connection timed out)
20:33:50 --> mconley has joined #instantbird
20:35:16 <-- jb has quit (Ping timeout)
20:42:34 <nhnt11> flo-retina: In a scenario where a log writer has not yet written the header of the log file, and then multiple logMessage calls are queued, what should the behavior be if writing the header fails?
20:42:47 <nhnt11> I'm unsure about what to do with those pending logMessage calls
20:42:56 <nhnt11> Should we try to write the header in every logMessage call?
20:43:22 <nhnt11> Dropping those messages to the floor sounds wrong, but if something went so wrong that the header couldn't be written... we'd have bigger problems
20:46:53 <nhnt11> Hmm, we should try to write the header in every logMessage call. Never mind.
20:47:02 <nhnt11> (I had a reason this wouldn't work, but found a way around it)
20:47:55 <-- mayanktg has quit (Ping timeout)
20:49:12 <nhnt11> Eh, more questions. What happens if we fail to write the header in the logMessage call? Should that message be dropped? :S
20:49:39 <nhnt11> I think it's not worth worrying about, in that if we failed to write the header right before the message, then writing the message will likely fail too
20:49:48 <flo-retina> what are situations that would cause header-writing to fail?
20:50:04 <nhnt11> Some weird I/O error that I'm not aware of :]
20:50:08 <nhnt11> Maybe no write permissions
20:50:15 <nhnt11> which would mean writing the message will fail too
20:50:24 <flo-retina> a log file with header isn't parsable, right?
20:50:53 <nhnt11> without you mean? No, we don't parse log files with no header.
20:50:59 <flo-retina> if you can't write anything to the folder because of invalid permissions, there's not much you can do
20:51:04 <nhnt11> Yeah
20:51:05 <flo-retina> yeah, without
20:58:30 --> mayanktg has joined #instantbird
21:01:07 <nhnt11> I think I've been using Task.async wrongly :(
21:01:37 <nhnt11> Hmm, maybe not.
21:03:02 <flo-retina> if in doubt, ask Yoric for feedback on the patch
21:03:14 <nhnt11> I'm not using it wrongly :)
21:03:37 <nhnt11> The other async stuff, I've already asked Yoric where it was unclear
21:03:48 <nhnt11> But I can still f? him if you think it's required.
21:03:57 <nhnt11> It's a long patch to read :-/
21:04:53 <flo-retina> if you ask him for feedback, be specific in what you want feedback about
21:04:57 <nhnt11> Yeah
21:05:10 <-- clokep_work has quit (Ping timeout)
21:05:12 <nhnt11> At the moment I don't think I need any more feedback from him
21:06:39 <-- mconley has quit (Connection timed out)
21:07:43 --> mconley has joined #instantbird
21:12:19 <-- mayanktg has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
21:19:34 <nhnt11> Heh
21:19:47 <nhnt11> I just wrote some code that ends up in two promises waiting on each other forever
21:19:50 <nhnt11> :-]
21:20:10 <flo-retina> are you promising a dead lock?
21:20:55 <nhnt11> I'm swearing it :D
21:23:08 <nhnt11> This makes my head hurt: https://pastebin.mozilla.org/5398649
21:23:34 <nhnt11> (The problem there is the return)
21:28:37 * flo-retina doesn't understand what that does
21:28:46 <flo-retina> and I don't need to make my head hurt tonight
21:28:49 * nhnt11 neither :P
21:28:55 <flo-retina> I'm still hoping to review one of aleth's patches
21:29:20 <nhnt11> Okay, I kind of know what it does. It queues a promise that resolves only when the next promise in the queue resolves
21:29:42 <nhnt11> Which never happens because the next one won't be fired until the first one complete
21:29:42 <nhnt11> s
21:29:45 <nhnt11> => deadlock!
21:31:20 <nhnt11> It's equivalent to this: https://pastebin.mozilla.org/5398676
21:31:50 <nhnt11> Not /really/ a deadlock I guess. Just bad code :)
21:39:31 --> Mic has joined #instantbird
21:39:31 * ChanServ sets mode +o Mic 
21:42:57 <-- Rym has quit (Ping timeout)
21:53:57 --> Rym has joined #instantbird
21:54:30 <Mic> mayanktg: cool :)
21:55:18 <Mic> nhnt11: you can access the Scratchpad from the Web development sub-menu as well.
21:55:25 <nhnt11> Yeah, I found it :)
21:55:54 <Mic> I don't know how easily you can access the (old) menu bar on OSX.
21:56:10 <nhnt11> It's always there :)
22:04:52 <flo-retina> aleth: apologies if some of the comments I wrote aren't good... I can't really claim to be fully awake :-/.
22:05:53 <flo-retina> Mic: you can't remove the menubar on mac; unless the application is in fullscreen mode
22:06:44 <-- mconley has quit (Ping timeout)
22:07:35 <Mic> That makes the Fx-start-menu - pardon, hamburger menu - kind of pointless, doesn't it?
22:09:53 <nhnt11> Mic: I don't think I've ever used the hamburger menu :)
22:10:11 <nhnt11> Then again, I never really use the menu bar either.
22:10:40 <Mic> To be honest I haven't found out yet if it is keyboard accessible.
22:11:10 <flo-retina> given the amount of energy that went into Australis, I would be surprised if it wasn't accessible
22:11:18 <Mic> That's why I'm always using the menu bar...
22:12:31 * nhnt11 has no idea how to access the OS X menubar via keybaord
22:12:33 <nhnt11> board*
22:12:55 <flo-retina> nhnt11: I think you need to google it ;)
22:13:18 <nhnt11> Just did
22:13:25 <-- rosonline has quit (Ping timeout)
22:13:27 <nhnt11> Turns out it's ctrl+F2
22:13:36 <nhnt11> Which sucks because I need to hold mey Fn key as well
22:13:36 <flo-retina> yeah
22:13:51 <nhnt11> s/mey/my
22:14:07 <flo-retina> you can toggle that
22:14:22 <flo-retina> so that when not pressing fn the f<N> keys work
22:14:45 <nhnt11> That will mean I need to hold fn to adjust brightness, right? I use my brightness controls far more than "F2" :)
22:15:02 <flo-retina> nhnt11: it does.
22:15:09 <flo-retina> nhnt11: I never use the brightness control while typing something.
22:15:27 <flo-retina> I sometimes need the f<n> keys for shortcuts that require my other hand pressing something else
22:16:04 <nhnt11> Well something that I use a /lot/ are the playback controls, and if I have to hold Fn for those, then I need to use two hands. So I'm fine the way it is now :)
22:20:30 --> mconley has joined #instantbird
22:22:49 <nhnt11> Uh, when an interface changes the uuid needs to be updated right?
22:23:14 <flo-retina> yes
22:23:33 * flo-retina isn't too strict on that usually
22:26:50 <Mook_as> probably doesn't matter nearly as much for /chat, since there are probably few things that would be useful for a binary extension?
22:27:07 <flo-retina> Mook_as: exactly my point :)
22:27:13 <flo-retina> Mook_as: there's purplexpcom though
22:27:23 <flo-retina> but I don't think it uses the logs service
22:27:27 <Mook_as> ah, right, that's an extension of sorts
22:30:18 <nhnt11> instantbot: uuid
22:30:19 <instantbot> 27712ece-ad2c-4504-87d5-9e2c16d40fef (/msg instantbot cid for CID form)
22:31:17 <nhnt11> instantbot: uuid
22:31:18 <instantbot> b9d5701a-df53-4e0e-99b7-706e0118e075 (/msg instantbot cid for CID form)
22:31:59 <Mic> I had googled by the way and there was no answer that suggested that the hamburger menu has a keyboard shortcut. On top of that, nothing happened when the panel was open and I tried to use <tab> or the arrow keys.
22:32:08 <-- flo-retina has quit (Ping timeout)
22:32:12 <Mic> It seems to me that it isn't keyboard accessible.
22:34:14 --> rosonline has joined #instantbird
22:49:38 <Mic> Here's the answer by the way: bug 881937, bug 946395
22:49:40 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=881937 nor, --, Firefox 28, gijskruitbosch+bugs, RESO FIXED, The Australis panel menu should be keyboard accessible
22:49:41 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=946395 nor, --, Firefox 29, dao, RESO FIXED, Back out bug 881937 which made the panel menu keyboard accessible
22:49:55 <Mic> That means "no, it's not keyboard accessible (anymore)".
22:54:18 <-- Even has quit (Ping timeout)
22:59:40 --> Even has joined #instantbird
22:59:40 * ChanServ sets mode +o Even 
23:00:40 <-- Mic has quit (Quit: Instantbird -- http://www.instantbird.com)
23:11:56 <-- Armada has quit (Connection reset by peer)
23:14:19 --> clokep has joined #instantbird
23:14:19 * ChanServ sets mode +o clokep 
23:16:14 <-- mconley has quit (Input/output error)
23:16:42 --> mconley has joined #instantbird
23:18:32 <-- mconley has quit (Ping timeout)