[04:58] <hikiko> hi
[05:09] <pitti> Good morning
[07:30] <seb128> good morning desktopers
[07:30] <pitti> bonjour seb128, ça va ?
[07:30] <seb128> salut pitti, oui, et toi ?
[07:30] <pitti> seb128: ça va bien aussi
[07:31] <pitti> trying to beat systemd on 14.04 into a shape where it's not completely broken..
[07:31] <seb128> I saw people putting efforts into that, is there a specific reason?
[07:31] <seb128> oldLTS better than currentLTS? ;-)
[07:32] <pitti> don't get me started..
[07:32] <seb128> that's what I though :p
[07:51] <willcooke> hey seb128 pitti
[07:51] <seb128> hey willcooke, how is your u.k corner today?
[07:52] <seb128> no more houses on fire?
[07:52] <pitti> hey willcooke, how are you?
[07:52] <willcooke> so far, so good
[08:03] <Laney> moin
[08:04] <pitti> heeey Laney
[08:07] <seb128> oh, a Laney
[08:07] <seb128> morning!
[08:07] <willcooke> morning Laney
[08:09] <Laney> hi pitti seb128 willcooke!
[08:22] <Trevinho> Morongo
[08:23] <Trevinho> Morning... With right autocompletion :-)
[08:23] <willcooke> I prefer Morongo
[08:23] <willcooke> It's our new offical greeting
[08:26]  * TheMuso waves to the EU folks. :)
[08:26] <TheMuso> I like morongo. has a nice ring to it.
[08:26] <willcooke> hey TheMuso!
[08:28] <seb128> hey Trevinho, how are you?
[08:29] <seb128> Trevinho, autocompletion ... you IRC from your phone on the beach right? ;-)
[08:29] <seb128> that perfect plan just fails because of osk issues :p
[08:31] <Trevinho> seb128: hey seb128... No just less romantically moving from the kitchen to the office :-(
[08:31] <seb128> :-)
[08:32] <Trevinho> Anyway I'm good... Although autumn is arriving... Fog and rain today..
[08:33] <seb128> we had rain on sunday and it's still not great weather but forecast say we should be back to nice and sunny today for most of the week
[08:43] <Laney> 23 today apparently
[08:43]  * Laney sees some blue
[08:43] <seb128> nice
[08:43] <seb128> ramping up for the surfing w.e
[08:43] <seb128> is that place you are going to in the u.k?
[08:44] <Laney> nah portugal
[08:44] <Laney> it should be like 27-29 there
[08:44] <Laney> and not just the w.e., we are there all week ;-)
[08:46] <seb128> oh, makes more sense
[08:46] <seb128> nice
[08:46] <seb128> real holidays then!
[08:46] <seb128> enjoy :-)
[08:47] <Laney> wtf
[08:47] <willcooke> happyaron, when will the SRU be ready  for nm-applet for X?
[08:47] <Laney> still got ALL the days left to use
[08:48] <seb128> Laney, wtf?
[08:49] <Laney> ok, just 7 more :P
[08:51] <seb128> 7 remaining if you take next week and the end-of-year-week-off?
[08:51] <willcooke> happyaron, and the other updates we talked about too
[08:51] <Laney> yeah
[08:51] <Laney> seems that was automatically filled in for me
[08:52] <Laney> or I did it and forgot
[08:52] <happyaron> willcooke: nma sru is already there last week
[08:52] <Laney> haha
[08:53] <Laney> I just accidentally requested a 0 day holiday today
[08:53] <seb128> happyaron, where?
[08:53] <seb128> happyaron, did you rebase on .4?
[08:54] <seb128> happyaron, I guess I should have another look at the git?
[08:54] <seb128> Laney, seems like the canonical-is-closed week was autoregistered (or I did that earlier and forget as well)
[08:54] <willcooke> Laney, approved!
[08:54] <willcooke> Laney, did you enjoy the break?
[08:55] <willcooke> seb128, yeah I think it booked itself too
[08:55] <willcooke> Laney, should I ignore that request then?
[08:55] <Laney> willcooke: I had 10 pints and now I'm WASTED
[08:55] <Laney> GREAT HOLIDAY
[08:55] <willcooke> are you going to put another one in, or shall I edit it?
[08:55] <Laney> (yes)
[08:55] <Laney> no
[08:55] <willcooke> \o/
[08:55] <happyaron> seb128: yes in git
[08:55] <willcooke> Laney, ack
[08:55] <Laney> I think the days remaining thing starts at days - 1
[08:55] <Laney> so I was testing that
[08:56] <Laney> but it submitted the request
[08:56] <Laney> /o\
[08:56] <seb128> willcooke, happyaron, k, I'm going to have another look at the nma git for SRU sponsoring, then we need to get nm updated at some point as well I guess?
[08:56] <happyaron> yep
[08:56] <willcooke> happyaron please move quickly on that
[08:56] <willcooke> thanks happyaron seb128
[09:11] <Trevinho> ... And sun is back! ☀️
[09:15] <Trevinho> Laney: hey, if you've some time, the packages in https://requests.ci-train.ubuntu.com/#/ticket/1839 should be fine now
[09:17] <Laney> ok, i'll look in a minute
[09:34] <ochosi> jbicha: btw, final comment on the murrine issue: the same kind of workaround should work in the other direction, i.e. not using text-shadow should make the bitmap fonts work...
[10:31] <ximion> Laney: we have a meson release with D support now
[10:33] <Laney> ximion: I saw
[10:33] <Laney> been fighting with the subproject stuff
[10:33] <Laney> it's not that easy
[10:34] <ximion> Laney: I also don't think it's very well-designed - I mean, the current stuff does make sense, but it lacks a lot of flexibility
[10:36] <ximion> when I am back from Akademy, I will ask the gir-d-generator maintainer if he would merge a couple of patches to make the generator more generic and allow building the bindings on the fly as well as allowing an option to directly link too the C libs instead of loading them dynamically
[10:36] <ximion> (all of which is kind of done by my forked version already, but in a rather quick&dirty fashion (I broke the option to dynamically load the libs, and I can see why some users might want this))
[10:37] <Laney> is the maintainer active?
[10:38] <ximion> yes, somewhat
[10:38] <Laney> cool
[10:39] <ximion> it's also a relatively low-maintenance tool, as soon as it's written you are basically fine until the GIR API changes
[10:42] <Laney> put some integration into meson for it :P
[10:43]  * Laney has just reproduced that elusive bug
[10:47] <Laney> weeeeeird
[10:50] <ximion> Laney: for compiler bugs, using dustmite is awesome to have e minimized testcase to reproduce the issue
[10:50]  * ximion needs to run now
[10:51] <ximion> Laney: LDC Debian maintainer just said he'll do an LDC upload today
[10:51] <ximion> updating to a more recent Git version
[10:52] <ximion> (packaging an alpha snapshot by accident was really not great :D )
[10:53] <Laney> nice
[10:53] <Laney> oh yeah, I have to file that ICE I saw yesterday
[10:53]  * Laney is letting doko down
[10:53] <ximion> we really shouldn't do compiler Git snapshot testing :D
[10:54] <ximion> unfortunately, multiple errors happened for this in our distribution pipeline
[10:54]  * ximion now really has to run
[10:57] <Laney> ximion: wait
[10:57] <Laney> is this https://github.com/ximion/appstream/blob/master/src/as-distro-extras.c#L200 supposed to remove broken symlinks?
[11:11] <andyrock> hey all :D
[11:12] <Laney> HUH
[11:12] <Laney> hi andyrock!
[11:15] <Laney> symlink("/var/lib/apt/lists/_srv_mirrors_ubuntu_dists_yakkety_restricted_dep11_Components-amd64.yml.gz", "/var/lib/app-info/yaml/_srv_mirrors_ubuntu_dists_yakkety_restricted_dep11_Components-amd64.yml.gz") = 0
[11:55] <ximion> Laney: jap, it's supposed to remove broken symlinks
[11:55] <ximion> and it did that, last time I checked :P
[11:55] <ximion> is it broken?
[11:59] <Laney> ximion: see PR
[11:59] <Laney> not that bit, it turns out
[12:04] <ximion> hmm, APT itself symlinks stuff?
[12:05] <ximion> in a hurry, will try to understand the issue and merge the PR later
[12:05] <Laney> shrug
[12:05] <ximion> patch itself looks good
[12:05] <Laney> makes sense if they are on the filesystem
[12:05] <ximion> more error checking is always good
[12:06] <Laney> don't know why it doesn't clean them up on an update, but I guess if you are scrabbling around in /var/lib/apt/lists/ manually then you are pretty much on your own
[12:06] <Laney> also you could call refresh-cache at any time, when they may be broken without apt having been able to clean them
[13:01] <desrt> good morning, freedom lovers!!!
[13:01] <seb128> hey desrt, how are you today?
[13:01] <willcooke> hey desrt
[13:01] <desrt> great
[13:01] <desrt> had an awesome long weekend in ottawa
[13:01] <desrt> seb128, willcooke: good morning :D
[13:02] <seb128> did you do anything fancy there?
[13:02] <desrt> yup
[13:02] <desrt> there was all kinds of fanciness, in fact
[13:02] <desrt> we had afternoon tea at the tea party, and shared a bottle of wine on the terrace at the chateau laurier overlooking the parliament and the locks
[13:03] <desrt> otherwise, we did all the typical things.. ate an obama cookie, visited the locks, visited parliament, visited the national gallery, etc.
[13:04] <desrt> was a surprisingly full weekend
[13:04] <seb128> nice
[13:04]  * seb128 looks for photos of chateau laurier, didn't know you had building like that in Canada
[13:04] <seb128> looks like a real chateau :p
[13:04] <desrt> lol.  europeans :p
[13:05] <desrt> it's a hotel :)
[13:05] <desrt> how was your weekend?
[13:05] <seb128> "designed in the French Gothic Châteauesque style to complement the adjacent Parliament buildings."
[13:05] <seb128> according to wikipedia
[13:05] <seb128> quite good but a lot less busy and fancy
[13:06] <seb128> went to Rotterdam, had lunch in the euromast tower with a nice view on the city
[13:06] <seb128> otherwise diner with friends on saturday evening and relaxing home sunday while it was raining
[13:06] <seb128> (just managed to go for a footing in between rains)
[13:07] <desrt> is 'a footing', a) a walk, b) a sporting match, or c) a group of people kicking someone on the ground  ?
[13:07] <seb128> footing->jogging
[13:07] <desrt> so closer to 'a' then :)
[13:08] <seb128> sorry, second time I make the mistake this week, I was hesitating right now pondering which was the correct one
[13:08] <seb128> the other one is french-but-english-sounding :p
[13:08] <desrt> ahh.  i thought it was some upclass-sporty term that i was unfamiliar with
[13:08] <seb128> na, it's just https://fr.wiktionary.org/wiki/footing
[13:09] <Krands> hi !
[13:09] <seb128> "    De l’anglais foot (pied) avec suffixe -ing, et pourtant pure création française (comme parking, lifting, zapping) car le terme anglais est jogging. "
[13:09] <desrt> sounds like a pretty nice weekend, and probably better for relaxation :)
[13:09] <seb128> yeah, can't complain
[13:09] <seb128> hey Krands
[13:09] <desrt> Krands: good morning :)
[13:09] <desrt> (sort of like 'handy' i guess)
[13:09] <desrt> words that logically make sense in english, used in other languages, and english people have no idea what it means
[13:10] <Krands> I know "howdy" too
[13:11] <seb128> well "handy" is proper english though?
[13:11] <desrt> not in that way
[13:11] <desrt> 'footing' is also english too
[13:11] <desrt> it's the word for where/how your feet are placed when you are climbing/hiking/etc.
[13:12] <desrt> like if you want to make your next step, but you are afraid that the ground under you may fall away, you might say "i'm unsure of my footing"
[13:12] <seb128> ah, ok
[13:12] <seb128> makes sense as well I guess
[13:12] <desrt> it can also be used as a metaphor
[13:12] <desrt> for if you are unsure about (for example) how much support you have in a given position among a group of people
[13:13] <desrt> but ya.. i guess most anglos who (a) don't travel internationally and (b) are working outside of tech, don't know "beamer" and "handy" for their 'special' german meanings
[13:52] <seb128> qengho, hey
[13:53] <qengho> seb128: Good morning!
[13:53] <seb128> qengho, what's that requirement that chromium/firefox SRUs can only go through -security and not through normal SRU process for simple non security bugfixes?
[13:53] <seb128> I'm trying to find record of it and can't, feels weird to me
[13:53] <seb128> I found the discussion on https://lists.ubuntu.com/archives/technical-board/2010-October/000520.html
[13:54] <seb128> but no bit said that we can't do normal SRUs for those, my understanding is just that usually those updates includes security fixes so they go through security for that reason
[13:54] <seb128> no?
[13:55] <qengho> seb128: it could go through SRU, I think, but doesn't have to. Being excluded from SRU requirements doesn't mean it can't go through SRU.
[13:56] <seb128> k, what I though, thanks
[13:56] <Laney> bloop
[13:56] <seb128> so it doesn't need to block on chrisccoulson
[13:57] <seb128> we could have anyone with upload rights to just do a normal SRU?
[13:57] <chrisccoulson> We don't do firefox SRUs because security updates are so frequent already that it just creates update fatigue / annoyed users
[13:57] <chrisccoulson> and there's a firefox update next week
[13:57] <qengho> I think anyone in #security (?) could copy to security pocket. If others could, I'm surprised.
[13:58] <seb128> qengho, why does it need to be in the security pocket if it's not a security update?
[13:58] <chrisccoulson> no, we can't copy updates to the security pocket. Only the other way around
[13:58] <Laney> They roll up -updates into -security ad the next update
[13:58] <Laney> IIRC
[13:59] <seb128> chrisccoulson, hey. Right, I'm not saying SRU is the way to go usually, just trying to get that "key not working since june" fixed, it's getting ridiculous that it takes us that delay to update a key and that users meanwhile are without a working service in the LTS
[13:59] <seb128> yeah
[13:59] <seb128> security updates always include the current -updates changes
[14:00] <seb128> chrisccoulson, so you are basically waiting for the update next week to include the key change?
[14:00] <seb128> chrisccoulson, same for chromium?
[14:01] <chrisccoulson> In addition, firefox built outside of https://launchpad.net/~ubuntu-mozilla-security/+archive/ubuntu/ppa will break crash reporting (we don't look anywhere else for symbol uploads)
[14:01] <chrisccoulson> seb128, it matters less for chromium - it's in universe ;)
[14:01] <seb128> so can we get it uploaded please?
[14:01] <seb128> I just want the geolocation thing to work again on Ubuntu
[14:02] <chrisccoulson> Yeah, I'll get that done
[14:02] <seb128> I don't care how/which way we do the upload
[14:02] <seb128> chrisccoulson, thanks
[14:02] <seb128> chrisccoulson, I can do a normal SRU if that spares you work, I don't see a reason why that needs go through security...
[14:02] <seb128> but if you are happy to deal with it then even better
[14:03] <seb128> you get some more free beers in octobre ;-)
[14:10] <willcooke> attente, morning.  Did you see the comment on #1420334 about CSD?
[14:11] <willcooke> bug #1420334
[14:11] <willcooke> comment #8:  https://bugs.launchpad.net/mir/+bug/1420334/comments/8
[14:14] <attente> willcooke: no, i hadn't
[14:15] <attente> yeah. it looks like those branches aren't related
[14:18] <Krands> How is the Wacom support in Xenial?
[14:19] <willcooke> attente, I was thinking more about the comment "do we really want to support CSD?".  We do, right?
[14:20] <seb128> happyaron, your nma update/SRU, looks like configure.ac bumped the libnm requirement from 1.1 to 1.2, debian/control needs to reflect that (should have been done for yakkety as well but they are likely to ask you to fix it for SRU so better to do it before we upload rather than doing a review/fix round)
[14:20] <attente> really a design decision i think. but i can't remember why i filed that bug unless we did want it
[14:21] <attente> but iirc they wanted to add a freeform surface type to mir which would be used for this
[14:21] <attente> so i'm not sure why there's backpedalling here
[14:21] <happyaron> seb128: ok, updating now
[14:23] <seb128> happyaron, also unsure what you did to patches, several got removed/added (or renamed?) but that needs to be described in the changelog, especially new added ones need to be related to launchpad bugs that are SRU compliant since they are not part of the upstream update/tested by them
[14:24] <seb128> happyaron, you might want to remove those new ones for the first SRU round and redo another one to add patches later
[14:29] <willcooke> attente, could you comment to say that yes we do want CSD.  I think I remember we need them for Gtk apps too.
[14:29] <attente> yeah. if anything to at least support header bars properly
[14:33] <happyaron> seb128: I think there aren't additions since 1.2.2? all are renames due to git-pq...
[14:34] <seb128> happyaron, current xenial version is 1.2.0-0ubuntu0.16.04.4
[14:34] <seb128> not 1.2.2
[14:36] <seb128> happyaron, http://people.canonical.com/~seb128/nma.debdiff that's the debdiff I get locally
[14:37] <seb128> happyaron, I know that the renames are due to git-pq but I'm unsure the SRU team is going to be fine with that, it makes the diff quite difficult to read/understand
[14:37] <happyaron> I'm looking at the same thing and try to find which needs to be treated
[14:37] <seb128> unsure what you can do, try asking them if they are fine with those renames
[14:38] <happyaron> ok
[14:38] <seb128> otherwise I guess you need to maintain xenial out of git-pq
[14:38] <seb128> thanks
[14:39] <happyaron> just trying to keep xenial and yakkety kinda in sync atm, that elimates a lot of wasted efforts..
[14:39] <seb128> right
[14:39] <seb128> well see with them if they are fine with it
[14:39] <seb128> but the changelog should at least be descriptive of the renames in that case
[14:39] <seb128> like
[14:39] <seb128> * Renamed patch from git-pq
[14:39] <seb128> - thatname -> thisname
[14:39] <seb128> etc
[14:40] <happyaron> ok
[14:52] <happyaron> seb128: what about now? pushed my update just now
[14:52] <seb128> looking
[14:53] <chrisccoulson> qengho, do the packages in https://launchpad.net/~canonical-chromium-builds/+archive/ubuntu/stage have the updated key?
[14:53] <qengho> chrisccoulson: yes.
[15:13] <chrisccoulson> qengho, geolocation still fails to me with the updates from your PPA
[15:13] <chrisccoulson> (I just tried it in google maps)
[15:14] <qengho> chrisccoulson: The resources are bound to organization, not to the key. By turning off keys, we will free up resources.
[15:14] <chrisccoulson> qengho, ah, thanks
[15:30] <willcooke> BEEP BEEP BEEP It's meeting time!
[15:30] <willcooke> #startmeeting Desktop Team Weekly Meeting - 2016-09-06
[15:30] <meetingology> Meeting started Tue Sep  6 15:30:19 2016 UTC.  The chair is willcooke. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[15:30] <meetingology> Available commands: action commands idea info link nick
[15:30] <willcooke> Roll call: andyrock, attente, desrt,  dgadomski, fjkong, happyaron, hikiko, laney, qengho, seb128, sweet5hark (out), themuso (out), tkamppeter, trevinho, robert_ancell (out)
[15:30] <andyrock> o/
[15:30] <seb128> hey
[15:30] <FJKong> hi
[15:30] <dgadomski> \o
[15:31] <qengho> hi hi!
[15:31] <happyaron> \o/\o/\o/
[15:31] <desrt> word up
[15:31] <hikiko> hello
[15:32] <Laney> \o/    ¬_o/
[15:32] <willcooke> Looks like we have most folk, let's begin
[15:32] <willcooke> #topic andyrock
[15:32] <andyrock> 1. holidays
[15:33] <andyrock> 2. refactoring a little bit the quicklist code for unity8
[15:33] <andyrock> 3. started to look on how to implement tooltip on unity8
[15:33] <andyrock> (on the launcher)
[15:33] <andyrock> eof
[15:33] <willcooke> thanks andyrock
[15:33] <willcooke> #topic attente
[15:34] <attente> hi all
[15:34] <attente> short week due to holiday yesterday
[15:34] <attente> debugging a gnome-builder snap made using the custom jhbuild snapcraft plugin i've been working on the past few weeks, still not working even in devmode though...
[15:34] <attente> discussions about how mir should be dealing with popup surfaces and how clients could potentially use this to probe for monitor geometry information
[15:34] <attente> (eof)
[15:34] <willcooke> thanks attente
[15:34] <willcooke> #topic desrt
[15:34] <desrt> hey.  short week for me too
[15:35] <desrt> not super-productive, unfortunately :(  had a few technological catastrophies
[15:35] <desrt> mostly did bugs/patches/etc.
[15:35] <desrt> eof
[15:35] <willcooke> thanks desrt
[15:35] <willcooke> #topic dgadomski
[15:35] <dgadomski> hey
[15:35] <dgadomski> I haven't done much in the desktop area this week, but
[15:35] <dgadomski> * there is a LightDM issue related to password expiration & ActiveDirectory I'm examining. Already talked to robert_ancell about it.
[15:35] <dgadomski> * the snaps are still waiting to be upstreamed
[15:35] <dgadomski> plus: I'm on vacation for 2 weeks (starting Aug 12)
[15:36] <dgadomski> eof
[15:36] <willcooke> thanks dgadomski have a good hols!
[15:36] <dgadomski> thanks :)
[15:36] <willcooke> #topic FJKong
[15:36] <FJKong> * Meeting with NUDT guys at Changsha, discussing questions about kylin
[15:36] <FJKong> Sougou IM new version bug tracking.
[15:36] <FJKong> * cursor position when using AndroidStudio.
[15:36] <FJKong> * core dump debuging.
[15:36] <FJKong> * take 1 day off
[15:36] <FJKong> eof
[15:37] <willcooke> thanks FJKong
[15:37] <willcooke> #topic happyaron
[15:37] <happyaron> 1. back from Changsha on Wednesday
[15:37] <happyaron> 2. becomes part of dkms upstream
[15:37] <happyaron> 3. product documentation analysis/discussions of the sogou on phone thing
[15:37] <happyaron> 4. sponsor of some Ubuntu Kylin packages, wallpaper still pending
[15:37] <happyaron> EOF
[15:37] <willcooke> thanks happyaron
[15:37] <willcooke> #topic hikiko
[15:37] <seb128> happyaron, you also updated n-m-a ;-)
[15:38] <happyaron> :)
[15:38] <hikiko> hey
[15:38] <hikiko> sorry
[15:38] <hikiko> - finished move plugin additions (rectangle, outline mode, enable-disable blend): https://youtu.be/3NcMQujhFhs
[15:38] <hikiko> - working on unity part of other compiz plugin optimisations
[15:38] <hikiko> - fixed the compiz part of an expo plugin bug - working on the unity part ^
[15:38] <hikiko> - updated OEM documents
[15:38] <hikiko> EOF
[15:39] <willcooke> thanks hikiko
[15:39] <willcooke> #topic Laney
[15:40] <Laney> oops
[15:40] <Laney> • gstreamer 1.8.3 to yakkety and xenial
[15:40] <Laney> • fixed some appstream bugs, prepped 0.10 transition in a PPA and filed FFe
[15:40] <Laney> • many FFe reviews
[15:40] <Laney> • update gnome-calendar patches @ debian
[15:40] <Laney> • some fighting with D toolchain /o\ - ICE bug #1620681
[15:40] <Laney> • reviews and feedback on unity systemd unit, still need to test the newest version
[15:40] <Laney> • synced glib
[15:40] <Laney> • tried various builds for the webkit regression, fixed by upstream in the end
[15:40] <Laney> • fixed some autopkgtest regressions, especially ones which affected gtk
[15:40] <Laney> • organised sprint travel
[15:40] <Laney> i have no unicode
[15:40] <willcooke> :))
[15:40] <willcooke> thanks Laney
[15:40] <seb128> did we run out of unicode chars?!
[15:41] <willcooke> #topic qengho
[15:41] <qengho> * Yesterday, national holiday.
[15:41] <qengho> * Tracking down crasher in Yakkety chromium. Possibly freed-pointer sanity test and new gcc optimizaion. Tricky.
[15:41] <qengho> * Experiment to avoid SUID-sandbox warning in Chrome snap.
[15:41] <qengho> * Preparing Chromium new upstream release, v53.
[15:41] <qengho> * In a few days, turn off old Google-API keys.
[15:41] <qengho> EOF←  compose keys are better
[15:41] <willcooke> :) thanks qengho
[15:41] <willcooke> #topic seb128
[15:42] <Laney> I was just reading a bug instead of finding one :(
[15:42] <seb128> • several rounds of yakkety daily iso/install testing, reported issues and debugged some
[15:42] <seb128> • some desktop updates/syncs (libmtp, hicolor-icon-theme)
[15:42] <seb128> • some packaging changes review for the touch team
[15:42] <seb128> • sponsoring (libreoffice, libreoffice-l10n, n-m-a)
[15:42] <seb128> • asked around about yakkety translations to find out who could get those started, thanks w_grant!
[15:42] <seb128> • joined some snappy discussions
[15:42] <seb128> • daily review of recent launchpad bugs reports&triaging

[15:42] <willcooke> thanks seb128
[15:42] <willcooke> #topic TheMuso
[15:42] <willcooke> * Continued working with reporter of bug 1574324 to track down problematic Canonical authored bluetooth patches in pulse, making good progress so far.
[15:42] <willcooke> * Took a look at bug 1580882. I am unable to reproduce locally accross some intel HDMI and AMD HDMI hardware. I do have other hardware I could test, but I am not sure if it would make a difference. Have asked for some logs and am awaiting a reply.
[15:42] <willcooke> * Further testing/debugging of a11y-profile-manager in ubiquity oem-config mode, and trying to get the oem config setup wizard to be accessible, still trying to work out what runs as what user, and whether a11y infrastructure starts too late for the window to be visible via assistive technologies.
[15:42] <willcooke> #topic tkamppeter
[15:43] <tkamppeter> - cups-filters: Several cups-browsed fixes, partially relatd to snappability but also general:
[15:43] <tkamppeter>    o Cleaned up HTTP access to the local CUPS daemon, to assure that only one method (domain socket or localhost:631) is used, to avoid that cups-browsed suddenly hangs (esp. on shutdown).
[15:43] <tkamppeter>    o Close http connections after use to fix another memory leak.
[15:43] <tkamppeter>    o treat remote printers correctly also if they are on a non-standard port.
[15:43] <tkamppeter>    o In load-balanced print queue clusters make sure that the options of the remote printers get listed in print dialogs.
[15:43] <tkamppeter> - Bugs.
[15:43] <willcooke> thanks tkamppeter
[15:43] <willcooke> #topic Trevinho
[15:43] <Trevinho>  - Fixed systemd branches
[15:44] <Trevinho> - Prepared and published new SRU for xenial
[15:44] <Trevinho> - Some more work in lockscreen grabbing
[15:44] <Trevinho> - various reviews/fixes/triaging
[15:44] <Trevinho> eof
[15:44] <willcooke> thanks Trevinho
[15:44] <willcooke> #topic robert_ancell
[15:45] <willcooke> No update from Robert, but spoke to him yesterday.  He's been implementing the new snap store auth in libsnapd
[15:45] <willcooke> #topic AOB
[15:45] <willcooke> Townhall meeting tomorrow
[15:45] <willcooke> Thanks for getting travel reqs in.  If you haven't already , please do so
[15:45] <willcooke> Anyone got anything else?  Going in 10...
[15:46] <willcooke> (ish)
[15:46] <willcooke> #endmeeting
[15:46] <meetingology> Meeting ended Tue Sep  6 15:46:45 2016 UTC.
[15:46] <meetingology> Minutes:        http://ubottu.com/meetingology/logs/ubuntu-desktop/2016/ubuntu-desktop.2016-09-06-15.30.moin.txt
[15:46] <willcooke> thanks all
[15:46] <seb128> thanks
[15:47] <Laney> merci!
[15:47] <seb128> attente, what sort of issue do you have with gnome-builder snap? that might not be the easiest example, you should perhaps try with something easier like the calculator ;-)
[15:47]  * Laney tries Trevinho's new unity now
[15:48]  * Trevinho goodbyes Laney :_D
[15:48] <attente> seb128: right now it's looking for libjpeg.so.62. i've tried staging libjpeg62 as a stage-package directly, but snapcraft still seems to omit it entirely
[15:48] <Laney> NOPE
[15:48] <Laney> in a vm
[15:48]  * Laney isn't that crazy :)
[15:48] <Trevinho> ohhhh... not brave as I thought you were...
[15:48] <Laney> you want seb128 for that kind of crack
[15:48] <Trevinho> what is it a wall to climb, compared to that?!
[15:48] <seb128> lol
[15:48] <Laney> sudo dpkg --extract unity.deb /
[15:49] <seb128> close enough, I usually dpkg-deb -x :p
[15:49] <attente> seb128: what is the difference between libjpeg62 and libjpeg-turbo8 anyhow?
[15:49] <Laney> hardcore
[15:49] <seb128> attente, -62+turbo8? ;-)
[15:50] <attente> oh. so one is just faster than the other?
[15:51] <Laney> it's an ABI-compatible alternative implementation
[15:51] <seb128> attente, joke aside I don't know, but I think the later one used to support formats that libjpeg didn't (jpeg2k maybe?)
[15:51] <qengho> a fork, iirc
[15:51] <seb128> but dunno which one to prefer
[15:51] <seb128> attente, what is your snapcraft.yaml?
[15:51] <seb128> for the missing lib
[15:52] <Laney> Trevinho: I got a desktop at least :-)
[15:52] <attente> https://www.irccloud.com/pastebin/8jp0EbvI/
[15:52] <Laney> what else should I check
[15:52] <Laney> killing unity while it's locked?
[15:52] <attente> it's not going to mean much without the plugin though
[15:52] <seb128> attente, there is no stage-package in there though
[15:52] <attente> the plugin is adding libjpeg62 manually to the stage packages
[15:52] <seb128> oh
[15:53] <seb128> hum
[15:53] <attente> (as a temporary hack) but that doesn't seem to be working
[15:53] <Trevinho> Laney: having the panel working, lockscreen with indicators
[15:53] <Trevinho> and all the service on set when locked
[15:53] <seb128> do you have a link to part/code?
[15:53] <Laney> Trevinho: what does 'on set' mean?
[15:53] <Laney> :(
[15:53] <attente> i cleared the cache, but i don't really want to rebuild it since it takes forever to build webkit
[15:53] <seb128> k
[15:54] <seb128> attente, anyway, install libjpeg62 on the host should work no?
[15:54] <Trevinho> Laney: I mean the services are running as expected :-)
[15:54] <seb128> since the snap mount /usr/lib from the host
[15:54] <Laney> ok!
[15:54] <seb128> attente, not a solution but to see if you have other issues
[15:54] <attente> seb128: https://github.com/attente/snapcraft/tree/wip/attente/jhbuild-v4
[15:54] <attente> seb128: yeah, i installed it on the host hoping it would work, but it didn't
[15:55] <seb128> what's the error exactly?
[15:55] <attente> gnome-builder: error while loading shared libraries: libjpeg.so.62: cannot open shared object file: No such file or directory
[15:55] <attente> i'm not sure how it decides to use libjpeg.so.62 over libjpeg.so.8 though
[15:56] <attente> libjpeg.so.8 is in the snap
[15:56] <seb128> well that's what you built with
[15:56] <seb128> ldd on the binary should tell you what it wants
[15:57] <attente> [0] william@adele:/snap/gnome-builder/current$ ldd /snap/gnome-builder/current/bin/gnome-builder | grep jpeg
[15:57] <attente> 96:	libjpeg.so.8 => /usr/lib/x86_64-linux-gnu/libjpeg.so.8 (0x00007f5ffd4af000)
[15:57] <Trevinho> Laney: maybe testing if it still works with upstart too might be nice... I've tested that and seems to go well though
[15:57] <Laney> Trevinho: wtf....
[15:58] <attente> wait, i guess that's outside of the snap launching environment
[15:58] <seb128> attente, well the binary is not going to change, so it looks for libjpeg.so.8
[15:58] <Laney> Trevinho: http://people.canonical.com/~laney/weird-things/gtkmodule.png
[15:59] <Trevinho> Laney: oh...
[15:59] <Laney> what's happening there?
[15:59] <Trevinho> Laney: is that only onterminal?
[15:59] <Laney> no, everything
[16:00] <Trevinho> Laney: as I suspect might be the old thing of the dbus activation... although.... we update the env var before the terminal should run...b
[16:00] <attente> seb128: so i don't really understand why it's looking for libjpeg.so.62 directly. shouldn't it be looking at whatever libjpeg.so is symlinking too usually?
[16:00] <seb128> attente, no, you link to a soname
[16:00] <Laney> eeeeeeeeeeeeverything
[16:00] <seb128> attente, it uses libname.so.<soversion>
[16:00] <Trevinho> Laney: is that in strings /proc/$(pidof gnome-terminal-server)/environ|grep GTK_MODULES too?
[16:00] <seb128> attente, otherwise it wouldn't know how to decide to load libbar1 or libbbar7
[16:01] <attente> seb128: does that mean the problem is i built with the wrong libjpeg-dev package?
[16:01] <seb128> attente, the .so is only used for linking at build time
[16:01] <Laney> Trevinho: yeah
[16:01] <Trevinho> Laney: I've no clue then... it's something gtk related btw
[16:01] <seb128> attente, yeah, it means in your build tree you have libjpeg.so -> libjpeg.so.8
[16:01] <Laney> Hhaha
[16:01] <Laney> suuuuuuuuuuure
[16:01] <seb128> attente, so it register to load libjpeg.so.8
[16:02] <Trevinho> Laney: global menus are there too?
[16:02] <Laney> Trevinho: wait
[16:02] <Laney> I have no indicators
[16:02] <Trevinho> ah... so no ups running
[16:02] <Trevinho> ?!
[16:02] <attente> seb128: ok, how do you register it?
[16:02] <Laney> indeed
[16:02] <seb128> attente, the -dev include the .so symlink, you install the -dev corresponding to the version you want to link to :p
[16:03] <Trevinho> Laney: weird, I mean ups has an After= stanza
[16:03] <attente> seb128: i could've sworn i linked against libjpeg-dev which depends on libjpeg8-dev which depends on libjpeg-turbo8-dev
[16:03] <Laney> Trevinho: It should be in unity7.service.wants/, no?
[16:03] <Trevinho> (i changed it to "Before" for a while, but then it would have caused unity to restart on ups restart)
[16:04] <seb128> attente, right
[16:04] <seb128> attente,
[16:04] <seb128> $ dpkg -L libjpeg-turbo8-dev | grep .so
[16:04] <seb128> /usr/lib/i386-linux-gnu/libjpeg.so
[16:04] <seb128> $ ls -l /usr/lib/i386-linux-gnu/libjpeg.so
[16:04] <seb128> lrwxrwxrwx 1 root root 16 févr. 22  2016 /usr/lib/i386-linux-gnu/libjpeg.so -> libjpeg.so.8.0.2
[16:04] <seb128> attente, that gives you libjpeg.so.8
[16:04] <seb128> well libjpeg.so.8.0.2
[16:04] <seb128> on xenial
[16:04] <seb128> attente, if you want libjpeg62 you need to build with libjpeg62-dev
[16:05] <Trevinho> Laney: isn't the after thing enough? It always worked here
[16:05] <qengho> attente: perhaps you linked against the system libjpeg instead of the one mentioned in your yaml?
[16:05] <Laney> Trevinho: no, that just sets the order if they get started at the same time
[16:06] <Laney> it doesn't actually make the thing get started
[16:06] <qengho> "you" == the build tool of whatever you're packaging and its possibly-dumb LDFLAGS handling.
[16:06] <attente> seb128: sorry! i think i didn't really explain this well. i want it to link against libjpeg.so.8, but i'm getting the error. so i was just trying to add libjpeg62 as a hack to see if that would work, but that didn't work either
[16:06] <attente> qengho: that can't be possible because the plugin builds in a lxc container
[16:06] <Trevinho> Laney: oh, why does it start always here then?
[16:06] <seb128> attente, you might have another binary in your stack that link to libjpeg62? like libgtk or gdk-pixbuf or something?
[16:06] <Laney> Trevinho: Search for "After=" in here https://www.freedesktop.org/software/systemd/man/systemd.unit.html
[16:07] <Laney> don't know
[16:07] <Trevinho> Laney: yeah, I noticed that...
[16:07] <Laney> systemcl --user list-dependencies --reverse unity-panel-service.service might give you a clue
[16:08] <seb128> attente, can you put your .snap in a location where I can download it?
[16:08] <Trevinho> https://www.irccloud.com/pastebin/e3iqhBte/
[16:08] <Trevinho> Laney: ^
[16:08] <attente> seb128: also not possible afaics because those are built as part of the jhbuild process. i bootstrap the lxc container with libjpeg-dev
[16:08] <seb128> attente, well, it could be some lib from the host
[16:08] <seb128> that are bindmounted
[16:08] <seb128> you use /usr/lib from your host
[16:08] <attente> seb128: ok, i'll see where i can upload it. it's 180 Mb
[16:09] <Trevinho> Laney: /usr/lib/systemd/user/unity7.service:Requires=unity-settings-daemon.service unity-panel-service.service bamfdaemon.service
[16:09] <seb128> attente, people.canonical.com?
[16:09] <seb128> if you can ssh to there
[16:10] <Laney> Trevinho: hmmmmmmmmmmmmmmmmmmmmmm indeed that is there
[16:10] <seb128> attente, can you ldd -r /snap/gnome-builder/current/usr/bin/gnome-builder | grep jpeg ?
[16:11] <attente> seb128: http://attente.ca/gnome-builder_3.21.90_amd64.snap
[16:12] <Laney> Trevinho: and it works this time :|
[16:12] <Trevinho> Laney: anyway that might have been a 'Wants' thought, isn't it? I mean, I guess that unity shouldn't stop if I stop the ups or bamf... It just don't works correctly, but... it would work
[16:12] <Trevinho> well,bamf maybe is more important, but still people can work without them too
[16:13] <Trevinho> Laney: so... I don't like the idea that stopping ups would stop unity7 and then the whole session
[16:13] <Trevinho> isn't this too much?
[16:14] <attente> seb128: 96:	libjpeg.so.8 => /usr/lib/x86_64-linux-gnu/libjpeg.so.8 (0x00007fdba7513000)
[16:14] <attente> seb128: but this is doing an ldd without the default wrapper script
[16:15] <Laney> ah
[16:15] <Laney> I bet it was a missing daemon-reload when I updated the package
[16:15] <attente> or the jhbuild environment for that matter... so i wouldn't trust that ldd output actually
[16:15] <Trevinho> Laney: i wanted to suggest that... but, I thought you did it :-)
[16:15] <Laney> would have expected the package to do that for me :|
[16:15] <Laney> user units aren't as well supported as system ones
[16:16] <Trevinho> Laney: no since it's session side
[16:16] <Trevinho> well, it might do that, but it's not done as every user
[16:16] <seb128> attente, right, I guess you should add a part that starts bash in your snap (but needs a rebuild...)
[16:16] <seb128> attente, so you can log into a shell and do the poking around from inside the env
[16:16] <attente> nooooo
[16:16] <Trevinho> Laney: there should be something like an apt trigger that does that for the available users, isn't it?
[16:16] <attente> lol. ok
[16:17] <Laney> something would be nice
[16:17] <Laney> not sure how you would do it...
[16:17] <Laney> worth a bug if you want to file one
[16:17] <Laney> anyway, wants/requires> probably wants is nicer in case ups gets a bug that makes it never start or something
[16:19] <attente> seb128: oh. sorry, i forgot you're on i386 still
[16:22] <Trevinho> Laney: yeah, I agree... we can land that later anyway probably
[16:22] <Trevinho> Laney: as I would like to make anyway the session to stop when in locked state... but I'll do that later
[16:25] <Laney> stop?
[16:31] <Trevinho> Laney: I mean stop graphic-session.. thus killed
[16:39] <seb128> attente, does it make any difference?
[16:40] <seb128> attente, and I've an amd64 laptop as well ;-)
[16:40] <attente> seb128: pretty embarrassing... i think i'm running my host jhbuild...
[16:40] <seb128> lol
[16:40] <seb128> sorry ;-)
[16:40] <attente> seb128: no, it's good. thanks for the tip!
[16:40] <seb128> yw!
[16:41] <attente> seb128: and welcome to the future :P
[16:41] <seb128> :-)
[16:44] <chrisccoulson> qengho, ok, chromium is copied
[16:44] <Trevinho> Laney: anyway, if you think is fine, just approve the MP and publish the stuff... As I'd love to finish this systemd saga
[16:44] <seb128> chrisccoulson, thanks!
[16:44]  * qengho hugs chrisccoulson.
[16:44] <seb128> chrisccoulson, firefox is going to wait next week update I guess?
[16:46] <chrisccoulson> seb128, yeah. I'll upload firefox builds to the security PPA later
[16:46] <seb128> chrisccoulson, which means we can't invalidate the old key until next week?
[16:47] <chrisccoulson> seb128, it doesn't seem worth starting the SRU process at this stage (the next release will be published before the SRU moves from -proposed to -updates)
[16:47] <seb128> or I guess we could even if firefox uses it
[16:47] <seb128> we are over the quota and it's not working anyway
[16:47] <chrisccoulson> yeah, I guess it doesn't matter if it doesn't work already
[16:49] <jbicha> xnox: since you did the last software-properties upload, are you interested in reviewing https://code.launchpad.net/~jbicha/software-properties/use-gi-require_version/+merge/304193
[16:49] <seb128> attente, in your snap ldd says that libwebkit2gtk-4.0.so.37 links to libjpeg.so.62
[16:49] <qengho> seb128: other google services are working. Only geolocation is not. I don't think we want to break suggesions, spelling, sync, ....
[16:49] <seb128> qengho, oh ok, indeed then
[16:49] <seb128> shrug
[16:50] <attente> seb128: oh :/
[16:50] <chrisccoulson> ah, good point - although I think the only other one firefox uses is safe browsing
[16:51] <attente> seb128: so why does libwebkit2gtk-4.0-37 depend on libjpeg8?
[16:52] <attente> seb128: when you did the ldd, how did you enter the shell? is it possibly using the host environment there too?
[16:53] <seb128> oh, could be ...
[16:53] <seb128> $ strings squashfs-root/lib/libwebkit2gtk-4.0.so.37.14.4 | grep libjpeg
[16:53] <seb128> libjpeg.so.62
[16:53] <seb128> attente, ^
[16:53] <seb128> so, no
[16:53] <seb128> it's in the binary
[16:53] <seb128> dunno about the depends
[16:53] <attente> interesting
[16:54] <attente> in the plugin i just stage that package
[16:54] <seb128> the libwebkit2gtk-4.0-37 depends seems buggy
[16:54] <seb128> or those are swappable
[16:54] <seb128> Laney said they are abi compatible
[16:55] <seb128> so there might be some magic behind
[16:56] <attente> seb128: actually, that might've been produced by the jhbuild
[16:56] <seb128> well if you stage package the binary is not rebuilt?
[16:57] <attente> i'm not sure
[16:57] <attente> apt-file says that so doesn't exist
[16:57] <attente> so i assumed it was built in jhbuild, but let me confirm
[16:59] <attente> yeah, it was jhbuilt
[17:00] <attente> maybe i need to do this from scratch again to be sure. it's just that it takes so long to restart everything
[17:01] <seb128> :-/
[17:02] <seb128> but yeah, /lib is not the deb
[17:02] <seb128> it would be in /usr/lib/<arch>
[17:02] <seb128> anyway that's where your libjpeg62 is coming from
[17:03] <seb128> staging the libjpeg62 deb should have worked though
[17:03] <attente> ok, thanks. yeah, i'm not sure why the staging hack didn't work either
[17:03] <seb128> though where do you stage it from?
[17:03] <seb128> https://github.com/attente/snapcraft/commit/00253022d8ab66335d60a3fee746d76d7fe34577 doesn't have it
[17:04] <seb128> I guess it was a local hack?
[17:04] <attente> yup
[17:04] <seb128> stage packages don't work in some context
[17:05] <seb128> like I had cases where I used an organize rules and ended up with having some content missing
[17:05] <seb128> to be safe you can move the stage-packages to its own part
[17:05] <seb128> e.g
[17:05] <seb128>     deb:
[17:05] <seb128>         plugin: nil
[17:05] <seb128>         stage-packages:
[17:05] <attente> oh. i didn't know about nil
[17:07] <xnox> jbicha, not really =/
[17:10] <jbicha> xnox: np
[17:10] <seb128> those warnings are annoying, we should have reverted that change in the deb
[17:18] <Laney> night!
[17:18] <seb128> night Laney
[17:19] <seb128> jbicha, next time you can probably drop the  "gedit replaces gedit-common (<< 3.18.1-1ubuntu1)" change, that version was before xenial
[17:39] <seb128> 'night
[17:52] <marlinc> Is there a particular channel where I can questions about Polkit?
[18:03] <willcooke> night all
[20:27] <Krands> Would using gdm instead of lightdm  be a better choice than lightdm?
[20:27] <Krands> I have so many instabilities with lightdm