[03:44] <jamesh> robert_ancell: I've updated the two gnome-software patches I made to work against ubuntu-master.  They're available here to pull: https://code.launchpad.net/~jamesh/gnome-software/+git/gnome-software
[03:45] <robert_ancell> jamesh, you should just propose them upstream - we're not using Launchpad for G-S development
[03:46] <robert_ancell> The upstream branches are wip/ubuntu-3-xx
[03:46] <robert_ancell> where xx is 20, 22 and 24
[03:46] <robert_ancell> But first they should go into master, or be rejected as an Ubuntu specific patch
[03:48] <jamesh> robert_ancell: okay.  The pick-launch-app branch depends on unreleased snapd-glib changes, and I'm not at all sure about the UI parts of the classic-snap-install branch.
[03:50] <robert_ancell> jamesh, I want to run the UI parts against Richard upstream in case they need modifying. Once that's settled we can backport as we see fit (i.e. we don't need upstreams approval to backport into the wip/ubuntu-xx branches)
[03:51] <robert_ancell> jamesh, I'm waiting for https://github.com/snapcore/snapd/pull/3542 to be accepted before making the next snapd-glib release. I don't want to release and have the name change on me...
[03:51] <robert_ancell> I was expecting this to have been landed during the sprint, but yeah...
[03:52] <robert_ancell> Actually I think I'll just disable that change and do the release to unblock you.
[04:09] <jamesh> robert_ancell: fyi: the algorithm I used for picking a launch app is to favour apps with desktop files, and to favour the app whose name matches the snap
[04:09] <robert_ancell> jamesh, I think that's the best we can do at the moment
[04:10] <jamesh> it also uses the desktop file to launch the app if available, so we don't need to guess whether it is graphical in that case.
[04:11] <robert_ancell> Thanks!
[04:12] <robert_ancell> jamesh, let me know when you've opened bz bugs / subscribe me and I'll review there.
[04:12] <robert_ancell> jamesh, do you have GNOME git commit access?
[04:12] <jamesh> probably not any more.  I did have CVS commit access years ago ...
[04:13] <robert_ancell> ok, I'll push them if they look good
[05:55] <oSoMoN> good morning desktoppers!
[05:57] <duflu> Morning oSoMoN, seb128
[06:06] <oSoMoN> hey duflu, seb128
[06:52] <didrocks> good morning
[07:01] <seb128> hey duflu oSoMoN didrocks, how are you today?
[07:01] <duflu> seb128, good, you?
[07:02] <seb128> I'm well thanks
[07:06] <didrocks> hey seb128, duflu!
[07:31] <didrocks> seb128: so, the remaining qt components packages are still pulled on the live by fcitx
[07:32] <didrocks> (looking at binary component-mismatches to see if we missed anything)
[07:35]  * didrocks wonders why we have the syslinux theme packages seeded in supported-development-desktop, but no post xenial packag and no comment why
[07:36] <didrocks>   add syslinux-themes-ubuntu-oneiric alongside ubuntu-defaults-builder to support building images for older releases
[07:36] <didrocks> ah
[07:37] <didrocks> and as I removed yesterday ubuntu-defaults-builder… I guess we can unseed those
[07:39] <seb128> yes, makes sense
[07:40] <seb128> we have a trello card abou tfcitx
[07:40] <seb128> my gut feeling is that we should go back to ibus for all locales
[07:41] <didrocks> would +1 on this, this is what is supported upstream
[07:41] <seb128> but I don't know enough about the differences to be confident it's the right choice, we did add fcitx because some people were convinced it's better for chinese users
[07:41] <didrocks> and well, I can grab the maintainer personnally over an orange juice if needed of the chinese variant :p
[07:41] <seb128> haha
[07:41] <didrocks> (and a native speaker tester as well for mandarin)
[07:42] <didrocks> how would we take that decision?
[07:43] <seb128> good question
[07:43] <seb128> I guess we should take some input from the oem team before deciding
[07:44] <seb128> there was some push for fcitx from commercial deals at the time
[07:44] <didrocks> good point
[07:51] <flexiondotorg> Morning oSoMoN dupondje seb128 didrocks
[07:51] <didrocks> hey flexiondotorg
[07:52] <flexiondotorg> Oops, I meant duflu, not dup_ondje. Sorry for the ping.
[08:03] <willcooke> morning all
[08:05] <seb128> hey flexiondotorg willcooke, how is u.k today?
[08:06] <didrocks> hey willcooke
[08:06] <willcooke> sunny again!
[08:34] <oSoMoN> hey flexiondotorg, willcooke
[08:35] <willcooke> morning oSoMoN
[08:55] <willcooke> desktoppers: I'm verifying this on Y now:  https://bugs.launchpad.net/lightdm/+bug/1630156
[09:01] <seb128> willcooke, verifying as reproducing
[09:01] <willcooke> checking the fix works
[09:01] <seb128> ah
[09:01] <seb128> thanks
[09:03] <seb128> ah, I didn't notice it was blocked on the yakkety SRU to be verified
[09:07] <willcooke> jibel, something just crossed my mind for a test case...  we found that usb modeswitch broke a little while ago and "some" USB 3G modems stopped being able to be switched in to modem mode instead of "cdrom" mode.  Do you have such a device and could we add a test case for it?  Could be part of ISO tracker?
[09:10] <jibel> willcooke, I don't have such device but we can add it to the test cases as an optional test
[09:11] <willcooke> jibel, thanks.  I used to have one, it's probably in the bottom of my "box of things".  I'll go looking for it later
[09:14] <jibel> I stopped using this kind of device long time ago and use my phone instead
[09:14] <jibel> as access point
[09:21] <didrocks> ahah, qtdeclarative5-examples was what is making 20+ Qt packages in main
[09:21]  * didrocks continues his demotion journey
[09:27] <andyrock> morning!
[09:28] <seb128> hey andyrock, how are you?
[09:28] <oSoMoN> good morning andyrock
[09:29] <andyrock> hey oSoMoN
[09:29] <andyrock> hey seb128 good you?
[09:29] <seb128> andyrock, good as well!
[09:33] <seb128> k, going for an early lunch, bbiab
[09:33] <oSoMoN> bon appétit seb128
[09:34] <seb128> 'ci
[09:44] <jbicha> btw, this was the trick I used to get fcitx on the Ubuntu GNOME iso but without Qt: https://bazaar.launchpad.net/~ubuntu-gnome-dev/ubuntu-seeds/ubuntu-gnome.artful/view/head:/live#L62
[09:45] <didrocks> oh, interesting, thanks jbicha!
[09:46] <didrocks> I don't know if there are many Qt extensions that people want though
[09:52] <jbicha> in Artful, we now have an Addons→ Input Sources category in GNOME Software so someone could add appstream metadata for the different Fcitx frontends/components
[09:55] <didrocks> but what happens if they add a Qt module without the corresponding package?
[09:55] <didrocks> is it installed?
[09:55] <jbicha> but then we also get the issue we were discussing with GunnarHJ yesterday where language-selector pulls in fcitx (with Qt) for simplified Chinese
[09:56] <jbicha> we could have Qt recommend fcitx support, but I'm not sure whether that's a good idea
[09:58] <jbicha> part of my proposal for LP: #1585903 was to have gtk2 recommend ibus and the Ubuntu gtk2 theme…
[10:01] <didrocks> yeah, tricky
[10:06] <jbicha> the gtk2 part makes sense to me, having Qt recommend fcitx is a bit more questionable to me
[10:07] <jbicha> Fedora installs Qt by default because they wanted to ensure that Qt theming (with Adwaita) would be great when installing a Qt app
[10:08] <didrocks> hum, just to get the qt gtk engine installed by default I guess
[10:09] <jbicha> yes
[10:10] <jbicha> we might have done the same thing in Ubuntu GNOME (which defaulted to Adwaita) once the 2nd required package for that to work was packaged
[15:20] <seb128> Trevinho, I verified that u-c-c yakkery SRU so it should unblock your landing for tomorrow
[15:20] <Trevinho> seb128: thanks
[15:21] <seb128> Trevinho, yw!
[15:28] <seb128> bah, every cycle firefox gets stucked in devel-proposed for most of the cycle due to build issues on "side" archs :-/
[15:29] <seb128> chrisccoulson, I guess those armhf and ppc64el build issues are not something you are currently looking at?
[15:40] <flexiondotorg> Hi Desktopers
[15:40] <flexiondotorg> I was just asked what will happen to user running Ubuntu GNOME 17.04 when they upgrade to 17.10?
[15:41] <flexiondotorg> Will they get GNOME with the Ubuntu bits or stick with vanilla GNOME?
[15:42] <seb128> hey flexiondotorg
[15:43] <seb128> flexiondotorg, I think Ubuntu GNOME said they would prefer their users to get the Ubuntu session
[15:43] <seb128> but jbicha can probably give you a better answer
[15:43] <flexiondotorg> Thanks
[15:45] <seb128> yw
[15:48] <chrisccoulson> seb128, I'm not looking at those issues
[15:49] <seb128> chrisccoulson, what I though, I start wondering if we should just stop building firefox on those archs, unsure it has any users there and it's creating work and issues every cycle ... do you see any value in keeping them? do they allow to find bugs that might impact other archs for example?
[15:50] <chrisccoulson> I've never really seen value in building firefox on those architectures
[15:52] <seb128> k, thanks
[15:52] <seb128> on that note I'm going off IRC for a bit, moving back from the coffee place I was working from this afternoon
[15:52] <seb128> bbiab
[16:05] <seb128> k, back but on&off from IRC, I'm still going to look at the backlog and do a few things in between non computer work
[16:26] <oSoMoN> seb128, FYI: some slow progress on the LO x86 crash, it seems it might be a JDK bug in some x86 specific code, the kernel folks are looking into it
[16:41] <chrisccoulson> seb128, I should do what you do and find random places to work from :)
[17:15]  * kenvandine just freed up 457G of space with a rm -rf phablet 
[17:15]  * kenvandine hopes anything interesting in there has been pushed to bzr somewhere
[18:27] <willcooke> night all