#instantbird log on 05 15 2016

All times are UTC.

00:12:06 <-- mpmc has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
00:58:44 --> nhnt11 has joined #instantbird
00:58:45 * ChanServ sets mode +h nhnt11 
01:00:34 <-- unghost has quit (A TLS packet with unexpected length was received.)
01:12:47 <-- Bollebib has quit (Connection closed)
02:00:15 --> dx has joined #instantbird
02:10:26 <-- nhnt11 has quit (Ping timeout: 121 seconds)
02:25:25 <-- bogdan_maris has quit (Ping timeout: 121 seconds)
02:25:35 <instant-buildbot> build #3169 of macosx-nightly-default is complete: Failure [4failed compile]  Build details are at http://buildbot.instantbird.org/builders/macosx-nightly-default/builds/3169
02:26:08 <-- bgmCoder has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
04:04:28 <dx> hi
04:04:42 <dx> i'm seeing double http://hg.instantbird.org/instantbird/file/1d0601149149/purple/libpurple/proxy.h#l190
04:05:35 <-- EionRobb has quit (Quit: Leaving.)
04:10:34 <instant-buildbot> build #699 of linux64-nightly-default is complete: Failure [4failed shell_6]  Build details are at http://buildbot.instantbird.org/builders/linux64-nightly-default/builds/699
05:05:53 <dx> what's the point of this? i don't get it http://hg.instantbird.org/instantbird/rev/56f92aea42fe
05:19:32 --> EionRobb has joined #instantbird
05:20:14 <-- EionRobb has quit (Connection closed)
05:20:35 --> EionRobb has joined #instantbird
05:20:41 <-- EionRobb has quit (Connection closed)
05:23:40 --> EionRobb has joined #instantbird
05:23:47 <-- EionRobb has quit (Connection closed)
05:31:57 <flo-retina> dx: the duplicated purple_global_proxy_set_info is most likely because we upstreamed the change, and then when we updated libpurple, our patch applied with fuzz and we didn't notice
05:32:48 <dx> yeah, that was indeed upstreamed by you
05:33:33 <flo-retina> dx: for http://hg.instantbird.org/instantbird/rev/56f92aea42fe it's a workaround for a bug on our side.
05:33:54 <flo-retina> dx: we hacked purple_prefs_get_* functions to return values coming from the mozilla preferences system.
05:34:13 <flo-retina> dx: it turns out the Mozilla preferences always return a copy when we get a string value.
05:34:29 <flo-retina> (which makes a lot of sense because it prevents the caller from corrupting the internal preference storage)
05:34:46 <flo-retina> purple_prefs_get_string returns a const char* and doesn't expect its caller to ever free the result
05:35:09 <flo-retina> so, each time purple_prefs_get_string is called, we end up leaking a string, because there's no way from libpurple to free that string once it's no longer used.
05:35:43 <flo-retina> to mitigate the issue, we added a warning printed to stderr I think whenever our implementation of purple_prefs_get_string is called, so that we can ensure it doesn't happen too often
05:36:15 <flo-retina> so... now to answer your question: the patch you looked at dramatically reduces the number of calls to purple_prefs_get_string, without affecting any behavior.
05:36:17 <dx> huh, that was more serious than expected. glad i asked
05:37:00 <flo-retina> it took me a couple minutes to remember... It was 2010 ;).
05:38:48 <dx> that really sucks.
05:39:21 <flo-retina> the libpurple API is wrong IMHO. There's no way to ensure a broken prpl won't mess with the string it gets, and break the internal preference storage.
05:39:28 <flo-retina> But I assume fixing that would be a lot of effort.
05:39:47 <dx> well, purple_prefs_get_string() returns a const char*
05:39:58 <flo-retina> and if I remember correctly, that function is rarely called more than 5 times per process, after that fix you looked at.
05:40:08 <dx> neat
05:40:47 <flo-retina> dx: well, you know giving a const char* value to a char* parameter will in most case just cause a compiler warning, which will be lost in the large set of existing warnings ;)
05:41:25 <dx> heh
05:42:28 <dx> i was thinking of a way to prevent that leak you mentioned, but yeah, very few calls in the source, most of which are voice/video related, not worth bothering
05:43:06 <flo-retina> exactly... It's a known issue, but not one that makes any significant difference
05:43:27 <flo-retina> (until a new prpl starts calling that in a loop... but that's why we have a warning printed to the terminal; we will notice hopefully before releasing)
05:43:32 <dx> anyway i should introduce myself, i am dx, an unpaid intern for pidgin corp, participating in this year's nuclear winter of code.
05:43:53 <dx> i'm here to write a few things then disappear mysteriously
05:44:07 <flo-retina> welcome :)
05:44:21 <flo-retina> I think EionRobb mentioned you would likely join and ask questions ;).
05:44:39 <flo-retina> why do you want to surround your disappearance with mystery? :-P
05:44:57 <dx> that's what my contract says i'm supposed to do
05:45:21 <dx> something about doing the same thing some other person did with instantbird and libpurple
05:45:27 <flo-retina> why do you have a contract if you aren't paid?
05:45:41 <dx> hmmm.. good question
05:46:26 <dx> ok serious mode now, i just want to bring that prefs api to libpurple because i need it in bitlbee
05:46:57 <flo-retina> :)
05:56:26 <dx> flo-retina: what's the difference between the purple hg repo and the purple subdirectory in the instantbird repo?
06:00:02 <dx> or the comm-central repo...
06:05:20 --> bogdan_maris has joined #instantbird
06:09:19 <flo-retina> http://hg.instantbird.org/instantbird/ is dead, the code migrated to comm-central in early 2014.
06:09:52 <flo-retina> there's no purple/ folder in comm-central, because libpurple is GPL, so our C++ code using it is GPL too.
06:09:53 <instantbot> c++ is e-- ah, nevermind.
06:10:00 <flo-retina> but comm-central is MPL'ed.
06:10:24 <flo-retina> dx: the current repository to look at for Instantbird's libpurple integration is https://hg.mozilla.org/users/florian_queze.net/purple
06:13:40 <dx> hmm, okay, i used http://hg.instantbird.org/purple/ where the last commit is 766:18e35a43ad30 from 2015-02-13
06:14:10 <dx> i didn't miss a lot... except the 2.10.11 update. fuck.
06:19:29 <flo-retina> I didn't even know http://hg.instantbird.org/purple/ existed
06:19:32 <flo-retina> I don't know what it is
06:20:47 <flo-retina> dx: looks like a .htaccess file is missing or has disappeared. I don't think that "purple" folder (nor the "comm-central" one) on hg.instantbird.org is supposed to be visible
06:21:11 <flo-retina> I think they are there only as a technical detail of how a script generates the changes to http://hg.instantbird.org/l10n/en-US/
06:21:51 <dx> \o/
06:22:17 <dx> you really need to link these repos somewhere
06:22:50 <dx> i got there from https://wiki.instantbird.org/Instantbird:related_links
06:23:36 <dx> another wiki page mentions the comm-central move, but, well, libpurple is not there
06:29:52 <flo-retina> well, it can't be there. GPL ;)
06:30:38 <flo-retina> I think https://developer.mozilla.org/en-US/docs/Mozilla/Developer_guide/Build_Instructions/Simple_Instantbird_build gives all the info
06:30:43 <flo-retina> that's where we point people currently
06:31:48 --> nhnt11 has joined #instantbird
06:31:49 * ChanServ sets mode +h nhnt11 
07:03:40 <dx> ok rebasing my stuff to 2.10.11 didn't take too long.
07:05:20 <dx> this is how it's looking so far: http://dump.dequis.org/ezcL9.png - 55 commits, terrible commit messages, but i'm just splitting the diff by topics
07:06:56 <dx> a few of those are easy to cherry pick, like the signals. most will be dropped (all the #if 0 ones)
07:21:40 <-- nhnt11 has quit (Ping timeout: 121 seconds)
07:55:52 <dx> https://hg.mozilla.org/users/florian_queze.net/purple/rev/bcf2b012f79a "Fix the compilation of gutils on CE" - what's CE?
08:22:08 --> EionRobb has joined #instantbird
08:22:20 <-- EionRobb has quit (Connection closed)
08:25:02 --> EionRobb has joined #instantbird
08:29:04 <-- EionRobb has quit (Ping timeout: 121 seconds)
08:35:33 --> EionRobb has joined #instantbird
08:35:42 --> mpmc has joined #instantbird
08:41:56 --> Bollebib has joined #instantbird
08:49:52 <dx> starting with the easiest part: https://bitbucket.org/pidgin/main/pull-requests/49/add-a-couple-of-signals-from-instantbird/diff
08:52:42 --> abdelrhman has joined #instantbird
08:54:40 <EionRobb> you go girl!
08:55:16 <dx> \o/
08:57:11 <dx> http://dump.dequis.org/5PPAk.txt
09:13:59 --> unghost has joined #instantbird
09:45:10 <-- abdelrhman has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
09:48:54 --> gerard-majax has joined #instantbird
10:06:46 <-- EionRobb has quit (Connection closed)
10:08:05 --> EionRobb has joined #instantbird
13:32:49 <instantbot> richard.marti@gmail.com changed the Resolution on bug 1272608 from --- to FIXED.
13:32:50 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1272608 nor, --, Instantbird 49, richard.marti, RESO FIXED, IB: convert all in-tree style sheets logical properties from -moz-padding-start etc. to padding-inli
14:01:00 <-- EionRobb has quit (Connection closed)
14:01:36 --> EionRobb has joined #instantbird
14:13:20 <-- mpmc has quit (A TLS packet with unexpected length was received.)
14:25:06 <-- bogdan_maris has quit (Ping timeout: 121 seconds)
15:34:19 <-- unghost has quit (A TLS packet with unexpected length was received.)
17:32:05 --> abdelrhman has joined #instantbird
17:35:37 <-- abdelrhman has quit (Connection closed)
17:36:49 --> abdelrhman has joined #instantbird
18:07:49 <-- abdelrhman has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
18:14:46 --> abdelrhman has joined #instantbird
18:23:35 <-- abdelrhman has quit (Connection closed)
18:23:48 --> abdelrhman has joined #instantbird
18:26:29 <-- abdelrhman has quit (Connection closed)
18:27:04 --> abdelrhman has joined #instantbird
18:31:33 <-- abdelrhman has quit (Connection closed)
18:32:17 --> abdelrhman has joined #instantbird
18:37:00 <-- abdelrhman has quit (Connection closed)
18:37:02 --> bogdan_maris has joined #instantbird
18:37:44 --> abdelrhman has joined #instantbird
18:43:06 <-- abdelrhman has quit (Connection closed)
18:43:32 --> abdelrhman has joined #instantbird
18:55:50 <-- abdelrhman has quit (Ping timeout: 121 seconds)
18:56:11 <-- gerard-majax has quit (Ping timeout: 121 seconds)
18:56:50 --> abdelrhman has joined #instantbird
19:01:32 <-- abdelrhman has quit (Ping timeout: 121 seconds)
19:03:14 --> abdelrhman has joined #instantbird
19:10:53 <-- abdelrhman has quit (Ping timeout: 121 seconds)
19:11:06 --> abdelrhman has joined #instantbird
19:13:37 <-- abdelrhman has quit (Connection closed)
19:13:52 --> abdelrhman has joined #instantbird
19:16:25 --> nhnt11 has joined #instantbird
19:16:25 * ChanServ sets mode +h nhnt11 
19:19:57 <-- abdelrhman has quit (Ping timeout: 121 seconds)
19:41:22 <-- EionRobb has quit (Quit: Leaving.)
19:49:18 --> abdelrhman has joined #instantbird
19:57:12 <abdelrhman> nhnt11: I hope you have seen the initial interface
19:57:47 <nhnt11> abdelrhman: hey, yes, I saw the etherpad
19:58:00 <nhnt11> I only skimmed through it, looking at it again now
20:01:51 <nhnt11> abdelrhman: Have you considered a boolean attribute which will be set to true if the form element has multiple options?
20:02:06 <abdelrhman> nhnt11: This interface for field elements only, and I think dialogRequest interface would be better to be generic and each protocol uses it passes the method that implements its behaviour
20:03:39 <dx> hey, where am i supposed to throw patches at? bugzilla?
20:03:47 <nhnt11> abdelrhman: Can you rephrase the last part?
20:04:33 <nhnt11> dx: We use Mozilla's bugzilla instance, yeah. https://bugzilla.mozilla.org
20:05:32 <nhnt11> I have to go, sorry
20:06:03 <nhnt11> abdelrhman: The interface looks pretty good to me so far. I'll read through the XEP again and give you more feedback if I think of anything!
20:06:05 <-- nhnt11 has quit (Quit: Instantbird 1.6a1pre -- http://www.instantbird.com)
20:08:40 <-- abdelrhman has quit (Ping timeout: 121 seconds)
20:08:56 --> aleth has joined #instantbird
20:08:57 * ChanServ sets mode +o aleth 
20:20:44 <flo-retina> instantbot: review my patch
20:20:48 <instantbot> flo-retina: Sorry, I've no idea what 'review my patch' might be.
20:20:49 <instantbot> flo-retina: Your patch looks good. r+sr+ui-r+a=mconnor
20:21:59 <instantbot> aleth@instantbird.org changed the Resolution on bug 1060891 from --- to FIXED.
20:22:00 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1060891 nor, --, Instantbird 49, aleth, RESO FIXED, Stats service uses too much memory
20:22:29 <instantbot> aleth@instantbird.org changed the Resolution on bug 1042211 from --- to DUPLICATE.
20:22:30 <instantbot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1042211 nor, --, ---, nobody, RESO DUPLICATE, /list crashes instantbird on Windows causing lagging and hanging my machine.
20:25:54 --> EionRobb has joined #instantbird
20:30:45 --> chrisccoulson has joined #instantbird
20:43:28 <-- EionRobb has quit (Ping timeout: 121 seconds)
20:50:40 --> nhnt11 has joined #instantbird
20:50:40 * ChanServ sets mode +h nhnt11 
21:07:54 <-- nhnt11 has quit (Ping timeout: 121 seconds)
21:14:10 <-- Bollebib has quit (Ping timeout: 121 seconds)
21:16:07 --> Bollebib has joined #instantbird
21:20:12 --> EionRobb has joined #instantbird
21:20:19 --> mpmc has joined #instantbird
21:20:27 <-- bogdan_maris has quit (Ping timeout: 121 seconds)
21:26:23 <-- aleth has quit (Quit: Instantbird 49)
21:46:50 <-- mpmc has quit (A TLS packet with unexpected length was received.)
21:46:56 --> mpmc has joined #instantbird
22:40:18 --> nhnt11 has joined #instantbird
22:40:19 * ChanServ sets mode +h nhnt11 
22:42:42 <-- mpmc has quit (A TLS packet with unexpected length was received.)
22:42:42 <-- Suiseiseki has quit (Connection closed)
22:46:29 <dx> flo-retina: i'd appreciate if you could have a look to confirm that i'm getting this right: https://bitbucket.org/pidgin/main/pull-requests/51/add-purplecommandsuiops-api-from/diff
22:50:08 <EionRobb> looking at that now, it seems strange to create a uiops when there's already a signal that could be hooked into?
22:52:00 <EionRobb> although I personally much prefer uiops to signals, in general
22:54:19 <dx> EionRobb: the signal doesn't have the PurpleCmd, only the name, priority and flags
22:54:36 <EionRobb> its the first parameter of the signal?
22:54:46 <EionRobb> oh, no, sorry
22:55:00 <dx> yay short meaningless variable names
22:55:07 <EionRobb> ^
23:36:10 --> Suiseiseki has joined #instantbird