#instantbird log on 02 19 2013

All times are UTC.

00:00:35 <clokep> Ah, that's the issue...
00:00:41 <clokep> mozilla-build doesn't have intltool-update
00:01:44 <instantbot> florian@instantbird.org set the Resolution field on bug 1870 to FIXED.
00:01:46 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1870 nor, --, 1.4, aleth, RESO FIXED, Leading whitespace is lost from messages
00:01:55 <flo-retina> it's part of gettext
00:02:28 <flo-retina> maybe follow https://developer.pidgin.im/wiki/BuildingWinPidgin#gettext ?
00:05:19 <clokep> Working on it. :)
00:06:41 <clokep> Ah, that worked...
00:06:46 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/850f08465f41 - Patrick Cloke - Bug 1869 - Fix spacing in account.xml and accounts.xul, r=fqueze.
00:06:47 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/dcd23e84ee7a - aleth - Bug 1870 - Leading whitespace is lost from messages, r=fqueze.
00:06:48 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/168d3b21ad2d - aleth - Bug 898 - Copying text from conversations removes leading indent, r=fqueze.
00:07:39 <instantbot> clokep@gmail.com set the Resolution field on bug 898 to FIXED.
00:07:43 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=898 maj, --, 1.4, aleth, RESO FIXED, Copying text from conversations removes leading indent
00:08:10 <instantbot> clokep@gmail.com set the Resolution field on bug 1869 to FIXED.
00:08:12 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1869 tri, --, 1.4, clokep, RESO FIXED, Fix spacing in account.xml and accounts.xul
00:10:17 <clokep> flo-retina: Do we usually just provide the new properties files or do we normally patch them?
00:10:25 <clokep> (for the l10n repos)
00:11:08 <flo-retina> I don't really understand the question
00:11:35 <flo-retina> the files are in a repository, so the repository only contains the latest version of the files, and to see the patch it's easy (it's mercurial!)
00:11:47 <clokep> What? :-S
00:11:58 <clokep> See what patch?
00:12:06 <flo-retina> between 2 version of a file
00:12:12 <clokep> I vaguely recall you just providing the new l10n files after upgrades on your server.
00:12:22 <flo-retina> I think it will become obvious if you inspect the content of http://hg.instantbird.org/l10n/libpurple/
00:12:29 <clokep> Is that normally whta we do, or should I go through and apply the changes for each repo?
00:12:44 <clokep> Ah, I didn't know that existed.
00:12:50 * clokep was just on hg.instantbird.org/l10n
00:13:13 <clokep> Sorry!
00:13:19 <flo-retina> http://log.bezut.info/instantbird/yesterday#m371
00:13:53 <clokep> Yeah, I misread it. :P
00:13:59 <flo-retina> ok, good night :)
00:14:06 <flo-retina> well, evening for you :)
00:14:39 <clokep> :)
00:14:49 <clokep> Bleh, seems like the line endings are different on Windows.
00:14:50 * clokep sighs.
00:30:51 <instant-buildbot> build #354 of macosx-onCommit is complete: Failure [failed compile]  Build details are at http://buildbot.instantbird.org/builders/macosx-onCommit/builds/354  blamelist: aleth <aleth@instantbird.org>, Patrick Cloke <clokep@gmail.com>
00:44:19 <instant-buildbot> build #380 of linux-onCommit is complete: Failure [failed shell_1]  Build details are at http://buildbot.instantbird.org/builders/linux-onCommit/builds/380  blamelist: aleth <aleth@instantbird.org>, Patrick Cloke <clokep@gmail.com>
01:31:30 <instant-buildbot> build #366 of win32-onCommit is complete: Failure [failed shell_1]  Build details are at http://buildbot.instantbird.org/builders/win32-onCommit/builds/366  blamelist: aleth <aleth@instantbird.org>, Patrick Cloke <clokep@gmail.com>
01:49:03 <-- Mook_as has quit (Quit: Mook_as)
02:33:33 --> mconley has joined #instantbird
03:11:19 <-- mali has quit (Ping timeout)
03:12:28 --> mali has joined #instantbird
03:15:26 --> Mook has joined #instantbird
03:23:01 <-- clokep has quit (Quit: Instantbird 1.4a1pre -- http://www.instantbird.com)
03:30:57 <-- mali has quit (Quit: Instantbird 1.3 -- http://www.instantbird.com)
03:36:40 <instant-buildbot> build #785 of linux-nightly-default is complete: Failure [failed shell_3]  Build details are at http://buildbot.instantbird.org/builders/linux-nightly-default/builds/785
03:59:17 <instant-buildbot> build #781 of macosx-nightly-default is complete: Failure [failed shell_3]  Build details are at http://buildbot.instantbird.org/builders/macosx-nightly-default/builds/781
04:22:51 <-- mconley has quit (Input/output error)
04:23:18 --> mconley has joined #instantbird
04:25:05 <-- mconley has quit (Ping timeout)
05:00:32 <-- Mook has quit (Quit: Mook)
05:18:07 <-- EionRobb has quit (Quit: Leaving.)
05:42:12 <instant-buildbot> build #875 of win32-nightly-default is complete: Failure [failed shell_3]  Build details are at http://buildbot.instantbird.org/builders/win32-nightly-default/builds/875
07:05:41 --> mpmc has joined #instantbird
07:50:44 --> Mic has joined #instantbird
07:50:45 * ChanServ sets mode +h Mic 
07:56:43 --> deltafalcon has joined #instantbird
08:25:03 --> jb has joined #instantbird
08:46:35 <flo-retina> running our xpcshell tests on a debug build crashes :(
08:52:55 <-- Mic has quit (Quit: Output/input error)
08:55:23 <-- gerard-majax__ has quit (Ping timeout)
09:05:07 --> chrisccoulson has joined #instantbird
09:43:03 --> gerard-majax__ has joined #instantbird
10:26:06 --> aleth has joined #instantbird
10:26:06 * ChanServ sets mode +h aleth 
10:40:45 <-- jb has quit (Ping timeout)
11:00:27 --> clokep has joined #instantbird
11:00:27 * ChanServ sets mode +o clokep 
11:12:35 <-- deltafalcon has quit (Quit: Instantbird 1.3 -- http://www.instantbird.com)
11:17:03 --> rosonline has joined #instantbird
11:24:34 <clokep> One of aleth's check-ins caused a test to regress?
11:31:49 <-- rosonline has quit (Client exited)
11:33:12 --> rosonline has joined #instantbird
11:39:37 <aleth> Yes :(
11:59:41 <clokep> aleth: Do we know how to fix it / is someone working on it?
12:03:23 <aleth> clokep: Neither. I'm a bit confused by it to be honest
12:04:04 <-- clokep has quit (Ping timeout)
12:08:08 <aleth> I'm not sure it's a problem because by the time we call it we have already converted < to &lt; etc and so any such tags won't get parsed as tags.
12:09:00 <aleth> And I'm not sure the new behaviour (on the test string) is buggy because the tags do get stripped, just not the text content of the tagss.
12:13:30 <aleth> So currently it seems to me we could simply change the expected result of that test.
12:13:45 --> jb has joined #instantbird
12:49:37 --> clokep_work has joined #instantbird
12:49:37 * ChanServ sets mode +o clokep_work 
12:50:45 <clokep_work> aleth: That might be an OK solution. I'm not positive what the "expected" behavior in that case is.
12:50:52 <clokep_work> (It would seem to me you'd want to keep the text content though.)
12:51:46 <aleth> clokep_work: That's why I am confused, I would also have guessed that to be the expected behaviour.
12:56:48 <aleth> Btw I updated the mathjax addon yesterday - while I have no idea what caused the crash Mic reported, I'm hoping a config change I made might help.
12:57:59 <-- mpmc has quit (Connection reset by peer)
12:59:50 <flo-retina> clokep_work: are you saying that if a message is "blah blah blah <script>alert('hey');</script> blah" we should display "blah blah blah alert('hey'); blah" ?
13:00:25 <clokep_work> flo-retina: No, I think I misunderstood the issue.
13:00:38 <flo-retina> clokep_work: that's the new behavior with aleth's patch
13:00:59 <flo-retina> the previous behavior was to display "blah blah blah  blah"
13:00:59 <aleth> flo-retina: That's certainly what I was saying ;)
13:01:57 <flo-retina> aleth: so what were you saying?
13:02:21 <aleth> That I don't think the new behaviour is wrong.
13:03:23 <-- gerard-majax__ has quit (Ping timeout)
13:05:22 <clokep_work> So if we didn't support the bold tag (<b>), would we still show the text inside of it or no?
13:05:52 --> gerard-majax__ has joined #instantbird
13:14:57 <flo-retina> clokep_work: I think it's a completely different case
13:15:24 <flo-retina> clokep_work: the expected content of <script> and <style> is some code, not some user-readable text
13:17:14 <aleth> I still don't understand why the behaviour changed
13:19:07 <aleth> I also wonder whether in strict mode (I think there's a pref?) we would now display nothing at all.
13:21:38 <clokep_work> flo-retina: That's my point, I'm asking if they're handled differently.
13:22:02 <clokep_work> But I haven't looked at that code in forever, so feel free to ignore me.
13:29:28 <flo-retina> clokep_work: our code doesn't handle them differently, but I suspect <script> is handled differently when it's not inside an inline or block element (ie when it's at the top level of the body or head)
13:29:48 <aleth> flo-retina: Aha! That test is misleading. The previous behaviour only dropped the content of a script tag if there was nothing else in the message. So the example you gave above actually didn't work the way you thought
13:30:28 <flo-retina> aleth: yeah, I'm pretty sure the previous behavior was inconsistent (the <script> tag wouldn't have been handled the same, depending on where it was in the string)
13:47:52 <clokep_work> We should define the behavior we want and write a good test then! :P
13:56:09 * flo-retina agrees
14:02:54 --> FireFly_TB has joined #instantbird
14:03:18 <-- jb has quit (Ping timeout)
14:05:41 <-- Kaishi has quit (Quit: Kaishi)
14:21:28 --> mconley has joined #instantbird
14:22:04 <-- FireFly_TB has quit (Input/output error)
14:29:50 --> Mic|web has joined #instantbird
14:30:19 --> jb has joined #instantbird
14:30:37 <-- jb has quit (Connection reset by peer)
14:31:03 --> jb has joined #instantbird
14:32:44 <Mic|web> I'm not expecting to get script- and style-tags at all in my IM messages!
14:32:51 <-- jb has quit (Connection reset by peer)
14:32:53 --> jb1 has joined #instantbird
14:34:00 <flo-retina> Mic|web: right. That's why I'm wondering if we shouldn't put a message in the error console when filtering out such 'content'
14:35:05 <Mic|web> I'd file a bug...
14:35:19 <Mic|web> "Warn when receiving script and style tags in IM messages" ?
14:41:18 <instantbot> New Core - General bug 1878 filed by benediktp@ymail.com.
14:41:20 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1878 enh, --, ---, nobody, NEW, Warn when receiving script and style tags in IM messages
14:43:06 <Mic|web> bye
14:44:05 <-- Mic|web has quit (Quit: http://www.mibbit.com ajax IRC Client)
14:53:37 --> FireFly_TB has joined #instantbird
14:54:29 <-- jb1 has quit (Ping timeout)
15:20:28 <-- FireFly_TB has quit (Input/output error)
15:27:10 --> jb has joined #instantbird
15:38:28 --> FireFly_TB has joined #instantbird
15:41:53 <clokep_work> Can we implement FT just so we can show other clients the UI doesn't have to be awful? :-D
15:52:17 <-- FireFly_TB has quit (Input/output error)
16:09:33 <flo-retina> clokep_work: feel free!
16:17:49 <clokep_work> I mostly don't want to make the UI. :-D
16:26:32 * aleth wishes the mathjax devs had an IRC channel
16:27:09 <aleth> Every project should have one!
16:27:40 * clokep_work agrees.
16:33:52 <-- gerard-majax__ has quit (Ping timeout)
16:37:00 <flo-retina> aleth: it only makes sense for projects that have developers spending time on it *everyday*
16:41:39 <aleth> It would be enough if someone was lurking everyday...
16:43:09 <flo-retina> lurking is already spending time
16:43:19 <aleth> True
16:43:20 <flo-retina> and lurking in a dead channel where nobody ever talks isn't motivating
16:43:42 * flo-retina remembers the days when he was happy when he joined #instantbird to see that he wasn't recreating the channel by joining it ;)
16:47:19 --> qlum has joined #instantbird
16:49:00 <clokep_work> :)
16:49:12 <clokep_work> I'm surprised MathJax doesn't have one though, that seems to have a pretty good community.
16:49:25 <clokep_work> Maybe they just need a good IRC client? ;)
16:49:41 <flo-retina> clokep_work: do you know a good IRC client that uses their stuff?
16:49:44 <aleth> flo-retina: :)
16:49:44 <flo-retina> ;)
16:50:27 <clokep_work> I do! :P
16:50:44 <aleth> I might send them a mail once my add-on is in a more stationary state ;)
16:51:43 --> mali has joined #instantbird
16:52:50 <mali> Hi guyz, I'd like to ask a simple question
16:53:12 <mali> for example, I have kept e conversation window open for a few days
16:53:13 <clokep_work> mali: Go for it.
16:53:37 <mali> then I scrolled to top of the window yo see an  old message
16:53:46 <mali> am I able to see the date of it?
16:53:56 <clokep_work> Yes, of course.
16:53:57 <mali> currently I can see only time, not date
16:54:04 <clokep_work> It's in the tooltip, IIRC.
16:54:15 <aleth> Not for me :(
16:54:17 * clokep_work isn't on Bubbles right now.
16:54:17 <flo-retina> clokep_work: only the time unfortunately
16:54:38 <mali> yes, only the time
16:54:40 <clokep_work> mali: Then I must amend my statement: "No, of course not." ;)
16:54:44 <flo-retina> clokep_work: I had some experiment at some point to include the date too when the date wasn't the same as today, but that didn't work as well as I hoped
16:54:47 <clokep_work> I thought we fixed that at some point though. :(
16:54:47 <mali> :)
16:55:20 <mali> I 've searched the bug database a few mins ago, and found no bugs related to this
16:55:31 <mali> then I wanted to ask here if there is a way already
16:55:46 <mali> now should I file a biug about this?
16:55:53 <aleth> Yes please :)
16:56:02 <flo-retina> mali: if it's a message completely at the top of the conversation, you can open the conversation in the log viewer, and see when that conversation started
16:56:05 <flo-retina> that's not a general solution though
16:56:40 <flo-retina> clokep_work: the difficulty with this is that there doesn't seem to be any even that the content JS can get when the tooltip popup is about to be shown.
16:56:49 <mali> flo-retina: yep, I noticed the syntax in the logs
16:56:55 <mali> there are many info kept there
16:57:01 <clokep_work> flo-retina: Unfortunate. :(
16:57:31 <flo-retina> clokep_work: so I was relying on mouse events to update the tooltips, and they aren't reliably fired when scrolling with the mouse wheel, or focusing the window with alt+tab and the cursor becomes in the middle of the message without the mouse having moved :(
16:58:02 <aleth> :(
17:03:03 <clokep_work> Someone added "android support" to the GSoC list...
17:03:35 <flo-retina> is that someone volunteering to mentor it
17:03:36 <flo-retina> ?
17:06:40 <clokep_work> No.
17:14:20 <instantbot> New Instantbird (UI) bug 1879 filed by dev@maliayas.com.
17:14:23 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1879 enh, --, ---, nobody, UNCO, Date information should be accessible for messages and status messages
17:14:57 <mali> so it appears here :) great
17:17:04 <mali> android support is a great idea
17:17:16 <mali> I currently use firefox mobile and i love it
17:17:22 <mali> it's quite responsive
17:22:54 <aleth> mali: It's not an easy thing to do though. Android would require a different UI...
17:23:30 <aleth> I don't know enough about FF on Android to know whether it would be feasible in 3 months
17:23:44 <mali> hmm
17:24:52 <-- GeekShadow has quit (Ping timeout)
17:28:17 <-- qlum has quit (Quit: Getting the <censored> out.)
17:28:24 --> qlum has joined #instantbird
17:32:44 <clokep_work> I don't find using IM programs on a phone pleasant at all...
17:35:08 <aleth> Mic: https://i.minus.com/jlelsqwhthnlR.png
17:38:55 <clokep_work> aleth: That's the exact case I want it for. :)
17:39:22 <aleth> clokep_work: Yes :)
17:39:24 --> Mook_as has joined #instantbird
17:42:10 <flo-retina> :)
18:08:14 --> GeekShadow has joined #instantbird
18:10:34 <mali> I'm wondering if Microsoft will shut down the messenger API servers on April 8th
18:10:45 <mali> I am actively using msn messenger
18:13:49 <flo-retina> haven't they announced a date in 2014 for that?
18:31:23 <mali> nope
18:31:32 <mali> it was mart 15th at first
18:31:36 --> EionRobb has joined #instantbird
18:31:40 <mali> then delayed to april 8th
18:31:59 <mali> at least this is the mail I've received
18:34:37 <flo-retina> just seen this code, several time in a file. for (var i = 0;; i++) { var object = null; try { object = array[i].object; } catch (error) { /* No more objects in the list, get out */ break; } /* some more code */
18:37:59 <flo-retina> also, editing one line in that file to fix an error caused my editor to remove trailing whitespace on 1069 other lines :-/
18:48:59 <clokep_work> :( What file is this?
18:50:50 <-- chrisccoulson has quit (Ping timeout)
18:51:06 <aleth> flo-retina: wow, is there an actual reason for that or...?
18:52:44 --> unghost has joined #instantbird
18:55:33 <flo-retina> clokep_work: Code from a partner that I'm looking at / trying to debug. Don't worry, no such code will ever come near Instantbird ;).
18:55:45 <-- Tonnes has quit (Connection reset by peer)
18:55:56 <Mook_as> what editor?
18:56:00 <flo-retina> aleth: I suspect limited knownledge of JavaScript and/or limited computer science skills.
18:56:02 <flo-retina> Mook_as: emacs.
18:56:11 <Mook_as> komodo is _supposed_ to be able to trim only edited lines... but I dunno if I trust it :p
18:56:15 <flo-retina> Mook_as: although I'm more annoyed that there are 1069 lines in the file with trailing whitespace
18:56:26 <Mook_as> hehe
18:56:43 <flo-retina> Mook_as: and I discovered when seeing the diffstat that the file is actually too long. Likely ~7k lines of JS
18:57:03 <flo-retina> then I looked at code at random, and found the example I gave above
18:57:09 <flo-retina> and decided I didn't want to read more!
18:58:09 <clokep_work> Mook_as: It works now, I renenabled it. :)
18:58:33 <EionRobb> I used to have a colleague at uni who used to write code like that, flo-retina, but his for loops were while(true){...} instead
18:58:51 <clokep_work> You don't want to see the code I get daily at work...:(
18:59:03 <flo-retina> EionRobb: I think that kind of code is perfectly correct in some functional languages like OCaml
18:59:12 <flo-retina> EionRobb: but it's definitely not something I would expect to see in JS
18:59:27 <EionRobb> nor in Java, where it was continually used
18:59:28 <flo-retina> clokep_work: I don't want to see matlab code, even correctly written ;)
19:00:11 <mali> :)
19:00:24 <mali> matlab was a headache
19:00:33 <clokep_work> flo-retina: This is c-code.
19:00:39 <clokep_work> MATLAB is very useful for many things.
19:00:54 <clokep_work> We occasionally use while (true)... :)
19:00:59 <clokep_work> Useful for things that converge...
19:01:21 <flo-retina> there's nothing wrong with using while (true) in some cases
19:01:55 <flo-retina> it's throwing an exception to detect the end of a JS array that makes me cringe.
19:02:45 <Mook_as> well, if you assume your array can have holes in it...
19:03:01 <EionRobb> yeah, that's pretty gross... though I suppose its kind of an optimisation so you don't have to recalculate array length
19:03:04 * Mook_as has no idea why you'd assume that, but it's perfectly possible in js
19:03:42 <flo-retina> Mook_as: if there was a valid reason for it, I would expect a comment to be there to explain it ;)
19:04:02 <Mook_as> // crack is tasty
19:05:13 <EionRobb>  // this is a nice comment
19:07:34 --> Tonnes has joined #instantbird
19:10:44 <aleth>  // exceptions are fun!
19:11:02 <flo-retina> // silently ignoring errors is even better!
19:12:35 <-- jb has quit (Ping timeout)
19:17:54 <aleth> Does IB store cookies?
19:18:22 <clokep_work> mmmm cookies.
19:24:18 <flo-retina> aleth: I think we turned that off to avoid conflicts between different twitter accounts connected at once
19:25:19 <aleth> OK, good to know
19:26:04 * clokep_work wonders why.
19:26:31 <aleth> Why it's good to know?
19:26:38 <clokep_work> Why do you want to know?
19:26:52 <aleth> It transpires from the code that mathjax stores some user settings in cookies
19:27:03 <aleth> And I was debugging...
20:21:18 <-- SM0TVI has quit (Ping timeout)
20:26:55 --> Mic has joined #instantbird
20:26:55 * ChanServ sets mode +h Mic 
20:38:44 <Mic> aleth: thank you for the screenshot :)
20:54:04 <-- aleth has left #instantbird ()
20:59:43 --> aleth has joined #instantbird
20:59:44 * ChanServ sets mode +h aleth 
21:04:41 --> DGMurdockIII has joined #instantbird
21:13:39 <-- Mic has quit (Quit: Output/input error)
21:28:44 <-- unghost has quit (Quit: Ухожу я от вас (xchat 2.4.5 или старше))
21:39:53 <-- rosonline has quit (Client exited)
21:45:15 --> Kaishi has joined #instantbird
21:47:09 <-- clokep_work has quit (Quit: Instantbird 1.4a1pre -- http://www.instantbird.com)
22:01:15 <-- aleth has quit (Quit: Au revoir)
22:01:21 --> aleth has joined #instantbird
22:01:21 * ChanServ sets mode +h aleth 
22:24:22 <-- mconley has quit (Input/output error)
22:27:03 --> SM0TVI has joined #instantbird
22:39:07 --> clokep has joined #instantbird
22:39:07 * ChanServ sets mode +o clokep 
22:39:39 <aleth> I uploaded a new MathJax add-on version, for those who are trying it
22:42:58 <-- SM0TVI has quit (Quit: I view things as they are, without regard to place or person; my country is the world, and my religion is to do good. -- Thomas Paine (*1737 †1809))
22:44:31 --> Mic has joined #instantbird
22:44:32 * ChanServ sets mode +h Mic 
22:47:40 <flo-retina> aleth, clokep: so what's the plan to fix the tree?
22:48:27 * clokep was hoping aleth would rewrite the test.
22:48:29 <aleth> I'd say change the expected test result (as it is now at least consistent behaviour) and then fix Mic's new bug later if that is wanted
22:48:38 <clokep> I'm going with the "You break it, you bought it." type of thing here. :-D
22:49:38 <flo-retina> should I just comment out the line of the test that's failing with a FIXME comment?
22:50:02 <aleth> Or change the result to "alert..."
22:50:24 <flo-retina> is that a desirable behavior?
22:50:28 <aleth> I'm not sure what there is to fix beyond that, unless you mean Mic's bug.
22:50:56 <aleth> And that's not part of the bustage - it was broken before too, for script tags not at the beginning of the message
22:52:01 <aleth> But if you prefer to add FIXME and link it to that bug, that would also work of course
22:52:39 <clokep> I would rather just fix it.
22:52:49 <clokep> If it's just changing the test, that should be fairly simple, no?
22:53:26 <aleth> flo-retina is basically wondering whether the current behaviour isn't undesired and so the test should have failed all along
22:53:57 * clokep needs to look at the test again
22:54:40 <flo-retina> http://pastebin.instantbird.com/145096
22:55:00 --> jb has joined #instantbird
22:55:22 <flo-retina> this is what I have in mind
22:55:33 <flo-retina> and I'm annoyed that I can't really run the test because it crashes on my debug build
22:55:36 <aleth> Fixing the bug immediately? :)
22:55:58 <flo-retina> aleth: for now, explaining what I mean
22:56:34 <flo-retina> if others agree that it's the desired behavior, then maybe fixing it... but I'd like to go to bed soon
22:56:39 --> SM0TVI has joined #instantbird
22:56:44 <flo-retina> and I would appreciate having a new nightly soon too :)
22:58:30 <aleth> I'm not sure what the desired behaviour is, in a way it would be good to see if someone is trying to send scripts, not just in the error console. I'd be happy either way
22:58:44 <aleth> Or do some "other clients" send style and script tags by default?
22:59:21 <clokep> flo-retina: I think it's reasonable.
22:59:29 <flo-retina> aleth: I don't think other clients do, no
22:59:42 <flo-retina> aleth: I would expect that to only happen if something malicious is going on
23:00:06 * clokep needs to reember how to run tests. :-[
23:00:13 <flo-retina> fwiw, the change I pastebined is enough to make the test as is pass (I commented out the part that crashes)
23:00:24 <aleth> That's what I would expect too.
23:00:28 <flo-retina> make xpcshell-tests in objdir
23:01:09 <clokep> flo-retina: I get a PROCESS-CRASH result when running that test w/ that patch.
23:01:20 <flo-retina> clokep: actually, it was make xpcs<tab> for me :)
23:01:36 <flo-retina> but if you are on windows your shell likely sucks at completion
23:01:44 <clokep> Yes, it does. :)
23:01:46 <flo-retina> clokep: and you get it without that patch too, don't you?
23:01:57 <clokep> flo-retina: Compiling now...
23:02:10 <flo-retina> compiling what?
23:02:22 <clokep> flo-retina: No, I don't. Just an unexpected fail.
23:02:41 <flo-retina> ah, yes, of course
23:02:49 <flo-retina> the unexpected fail breaks the test before the crash :)
23:02:58 <flo-retina> I meant you crash without aleth's patch
23:03:15 <clokep> Oh.
23:03:18 <flo-retina> the part that crashes is in the test_links function
23:03:34 <clokep> I didn't try without aleth's patch.
23:03:47 <flo-retina> some #ifdef DEBUG code is really unhappy (fatal assertions) because of "something". I suspect it's related to us not compiling Places
23:04:02 <flo-retina> adding "return;" as the first line of that function "fixes" it for me
23:04:34 <flo-retina> (it sucks to not be able to run our tests on debug builds, but that's a completely different issue...)
23:06:45 <clokep> OK.
23:07:02 <clokep> So now it fails still.
23:07:27 <flo-retina> with my patch applied?
23:07:45 <clokep> flo-retina: I lied. I didn't apply it between. :)
23:07:47 <clokep> With your patch it passes.
23:07:55 <clokep> And I'm OK w/ that change. r=me, if you want.
23:08:11 <aleth> Thinking about it, I think stripping the content of those tags is better too.
23:08:50 <aleth> Anything else will just confuse most people, should they see it
23:08:54 <flo-retina> so should we also change the test or is it fine as is?
23:09:04 <flo-retina> I don't have good ideas of what should be tested there
23:09:07 <aleth> In that case the test is fine, right?
23:09:15 <clokep> The test is fine, unless you want to add more.
23:09:26 <flo-retina> aleth: well, I suspect it's still a bit incomplete, as it should have detected a failure even before your patch and it didn't
23:09:56 <aleth> One could add some characters before <script>, but that's just closing stable doors after the horse has bolted
23:10:20 <aleth> But maybe that's OK for a test ;)
23:11:08 <flo-retina> or should we test with <span><script>alert('foo');</script></span> so that if the parsing ever changes, we keep the current behavior keeps being tested?
23:11:34 <aleth> One could test that leading spaces survive too...
23:11:56 * flo-retina should have r-'ed for the lack of test!
23:12:07 <flo-retina> it definitely should be tested
23:12:08 <-- qlum has quit (Quit: Getting the <censored> out.)
23:13:34 <clokep> aleth: Where in the Twitter code do we fetch our own description?
23:15:44 <flo-retina> clokep: in the context menu of a tab (an IRC channel), I clicked "close conversation"
23:15:49 <flo-retina> clokep: the tab disappeared
23:16:19 <flo-retina> clokep: and I just noticed that it reappeared with the "left" icon. It caught my curiosity so I clicked on that new tab
23:16:31 <clokep> That's weird.
23:16:47 <aleth> clokep: I don't remember offhand, let me check tomorrow
23:16:51 <aleth> gtg
23:16:54 <flo-retina> clokep: it contains 2 system messages: "00:04:48 - <nick> [<nick>@2D29657.F60B0462.67AC9B1.IP] entered the room.
23:16:55 <flo-retina> 00:04:48 - You have left the room (Part)."
23:17:17 <Mic> good night
23:17:25 <clokep> Hm...interesting.
23:17:26 <aleth> good night!
23:17:30 <-- aleth has left #instantbird ()
23:17:31 <flo-retina> good night everybody
23:17:37 * flo-retina still needs to fix the tree :(
23:17:44 <clokep> We should probably check if the conversation exists before display those messages?
23:17:52 <Mic> clokep: look for "description" in twitter.js? I think it's also called "description" there...
23:18:10 <clokep> Mic: Yeah....I'm just trying to find the place we request it if we don't get our own tweet...
23:18:29 <flo-retina> clokep: I don't know what should be done. It's likely just a race condition with the server already sending us a QUIT message from a user when receiving our PART
23:18:41 <clokep> Most likely.
23:18:51 <clokep> It's certainly broken though.
23:19:03 <flo-retina> clokep: all of the twitter implementation is in a single file so searching is easy. I just look for ".json" and it gives me all the API calls we do
23:19:22 <clokep> Ah, that's a good idea.
23:19:26 <flo-retina> clokep: there's certainly something that could be improved, but I don't know what exactly
23:21:45 <-- Mic has quit (Quit: Output/input error)
23:22:04 <flo-retina> bah, my idea of testing with a <span> was idiotic. We filter out span nodes
23:22:06 <clokep> Ah...we simply requestBuddyInfo on ourself.
23:25:03 <flo-retina> http://lxr.instantbird.org/instantbird/source/chat/modules/test/test_filtering.js#30 makes me think even more that I should have r-'ed ;)
23:27:41 <clokep> :)
23:31:56 <instantbot> florian@instantbird.org requested review from clokep@gmail.com for attachment 2239 on bug 1878.
23:31:58 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1878 enh, --, ---, florian, NEW, Warn when receiving script and style tags in IM messages
23:33:07 <instantbot> clokep@gmail.com granted review for attachment 2239 on bug 1878.
23:35:09 <-- jb has quit (Ping timeout)
23:44:12 <flo-retina> Good night :)
23:48:19 <instantbot> Check-in: http://hg.instantbird.org/instantbird/rev/e8c8d59028ac - Florian Quèze - Bug 1878 - Warn when receiving script and style tags in IM messages, r=clokep.
23:48:27 <clokep> Thank you flo-retina!
23:49:09 <instantbot> clokep@gmail.com set the Resolution field on bug 1878 to FIXED.
23:49:12 <instantbot> Bug https://bugzilla.instantbird.org/show_bug.cgi?id=1878 enh, --, 1.4, florian, RESO FIXED, Warn when receiving script and style tags in IM messages
23:57:34 <instant-buildbot> build #355 of macosx-onCommit is complete: Failure [failed compile]  Build details are at http://buildbot.instantbird.org/builders/macosx-onCommit/builds/355  blamelist: Florian Qu?ze <florian@instantbird.org>