[06:29] good morning [06:31] ok, gnome-shell .92 is now a valid candidate, pomodoro went in and glibc is ok [06:32] only blocked by its extensions now: * amd64: gnome-shell-extension-dash-to-panel, gnome-shell-extension-multi-monitors, gnome-shell-extension-taskbar [06:32] Morning didrocks [06:33] hey duflu [06:34] none of the 3 are updated in Debian though [06:34] Trevinho: when you merged with debian with those Breaks, were you involved in getting them updated? [07:00] good morning desktoppers [07:00] Hi oSoMoN [07:00] hey duflu [07:13] salut oSoMoN [07:13] salut didrocks [07:19] lut oSoMoN didrocks [07:19] good morning desktopers [07:19] hey duflu [07:19] Hello seb128 [07:20] salut seb128 [07:21] salut seb128 [07:31] morning guys :-) [07:34] hey Nafallo [07:41] (btw, doing gdm3 and gnome-session) [07:42] * didrocks still waits for anyone with debian upload rights to sponsor my mutter & gnome-shell branches (see scrollback from yesterday evening) so that we can sync mutter and I can push my gnome-shell merge [07:44] didrocks, speaking of those, the backlog was not clear, you mentioned salsa yesterday but you didn't push anything to it? [07:45] didrocks, https://salsa.debian.org/didrocks-guest has no recent activity [07:46] seb128: I pushed them to launchpad, that's the link I referenced [07:47] I can push them to salsa if people don't want just to pull/push [07:47] (with with the finalize changelog commit that I needed for gnome-shell, having a merge is weird IMHO) [07:48] didrocks, well, it would increase the chance to avoid someone on the Debian side not knowing you did the work and redoing it [07:48] I mentioned it to jbicha yesterday, but he was busy updating the world and started arguing that you did push to Debian since you didn't push to salsa [07:49] seb128: but if I propose a merge with the finalize commit that I needed to prepare the ubuntu branch, that's not giong to work [07:49] maybe L_aney can pick them up when he gets online [07:49] you mean? [07:49] well, I needed to prepare ubuntu/master as well [07:49] for gnome-shell [07:49] to test it [07:50] mutter is in sync, so if you did that one it should be no issue [07:50] for this, you git merge from debian [07:50] but if you do a merge instead of a pull/push, the history differs [07:50] (a merge in salsa) [07:50] and gnome-shell I would expect the update to be done in debian, including changelog/tagging [07:50] so, your "rebase" is lying [07:50] and then we merge on top? [07:50] yeah, but it means, that I have to redo it a second time [07:50] once it's merged [07:51] and I couldn't test on ubuntu without merging my WIP debian work [07:51] I see [07:51] so, basically, we are doing twice the work, and adding delays [07:52] this is really not motivating [07:52] well, hopefully Laney can review/upload/commit that when he gets online in 10 min [07:52] that's why I just asked for a pull/push [07:52] instead of a merge [07:52] yeah, let's see [07:52] you could probably ask that on salsa as well if you have a merge request [07:52] hopefully people would read and pull instead of just pressing the merge button :p [07:52] but yeah, let's wait for Laney [07:53] you think people are reading? I'm unsure ;) [07:53] but yeah, we could try next time [07:53] anyway, let's see, if we can't pull/push, that means I have to redo the work [07:53] and so, that the process is even more cumbersome [08:00] morning gang [08:01] Zoinks [08:01] Morning willcooke [08:02] yo [08:02] sup Laney? [08:02] hey willcooke, Laney [08:02] hey willcooke Laney, how are you today? [08:03] alright [08:03] what is the argument? [08:04] I need someone to pull/push my branches for mutter/gnome-shell to salsa [08:04] as I already prepared the gnome-shell branch for ubuntu, I would prefer a pull/push to not have to redo the work when remerging the debian package for ubuntu [08:05] (so, the finalize commit for g-s is already in the branch) [08:06] some complaints about whether the eventual commits in salsa end up different? [08:06] in that case, what problem would another git merge be? [08:06] it looks weird though via history [08:06] you say "based on <…>", and then, you remerge from debian with the same version [08:07] right, based on proposed state in debian, that state changes, you merge in those additional changes [08:07] I don't think that's a problem personally [08:07] actually it seems like the right representation of the history [08:07] sounds like a hard to read history to me [08:07] ok, sorry but I don't want another git argument [08:09] oh git, we should write songs about it ;) [08:10] Laney: still able to sponsor though? [08:10] can look [08:10] thx! [08:11] the branches are on launchpad, I can push them on salsa if you prefer and copy paste what I wrote [08:12] I am really clear in saying that there is no 'redo the work' in case I change anything or even merge the commits with a merge commit though, you simply merge an additional time and the history is correct in describing what happened [08:12] this is what I advised andyrock to do actually when preparing updates on both sides [08:12] * Laney takes a look [08:12] actually let me finish gtk3 first, I started that one yesterday [08:13] sure [08:16] seb128: in https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/1786344 what's simpliefied english? [08:16] Ubuntu bug 1786344 in gnome-settings-daemon (Ubuntu) "Default IBus input method ignored in live session" [High,In progress] [08:16] *Simplified English [08:17] maybe he meant Simplified Chinese? [08:17] yes, I think he does [08:17] I put the chinese description in google translator [08:17] "18.10 Mirror (2018.8.7) Select Simplified Chinese to boot, unable to input Chinese. " [08:17] so yeah, translations error :) [08:17] and hey andyrock! [08:18] hey :D [08:18] I was also taking a look to gnome-screenshot 3.30 [08:18] do we still need to carry out the patches for unity? [08:18] *carry [08:18] "need" [08:19] if we can it's nicer to the unity users [08:19] but it's too much work then we can argue it's not worth the effort [08:19] so depends how difficult it is to rebase those [08:19] should not be difficult [08:19] better to keep them then I would say [08:20] now I need to understand which chinese is the simpliefied in the boot menu [08:20] :D [08:22] I don't think it matter [08:22] the problem is probably true on any locale that requires ibus [08:22] or I would guess that to be the case [08:24] seb128: but I'm able to type Chinese selecting Intelligent Pinyin in a live [08:24] let me check how it works in Bionic [08:24] andyrock, well I think the bug is that you shouldn't to go select it, that should be the default [08:24] activated on login [08:26] andyrock, https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/1786344/comments/10 has a bit more details [08:26] Ubuntu bug 1786344 in gnome-settings-daemon (Ubuntu) "Default IBus input method ignored in live session" [High,In progress] [08:50] jibel, do you vnc any desktop machines? [08:50] gnome shell specifcally [08:58] * didrocks reboots with new gnome-session & gdm [09:09] willcooke, I don't [09:09] willcooke, why this question? [09:11] jibel, well... when ever I try to VNC in to my 18.04 desktop which is running as a server, I just get a black screen. It always "just worked" before 18.04. I did some google and found similar problems, but no obvious solution yet [09:12] jibel, when you get a sec, and you test? Someone said it might be caused by the video drivers [09:12] oh, I might be able to test on my other laptop [09:12] which is intel [09:12] I'll do that [09:13] hum, various imports of upstream/laste and pristine-tar then doesn't match with our launchpad branch… :/ [09:13] some fusion conflicts even [09:14] seb128: I wonder if that's not related to your merge in https://code.launchpad.net/~ubuntu-desktop/ubuntu/+source/gnome-session/+git/gnome-session [09:14] didrocks, :-/ [09:15] talk to Trevinho about it maybe? [09:15] afaik I just pushed what he had submitted [09:15] yeah, but I guess the issue is that we imported the same version than debian [09:15] but then moved forward [09:15] before reimported [09:15] so both sides have new commits (with different version), and so, different content [09:15] we did the update in Debian first [09:15] and git can't know which side to pick [09:15] then merges [09:15] hum [09:15] iirc [09:16] but maybe Trevinho had both side ready at the same time [09:16] and it got merged using the salsa web ui or something in debian and that created divergence [09:16] I don't remember now [09:16] jibel, yeah works fine on this other machine. So I think it could be nouveau [09:16] could be, anyway, this is a real pb we can get (even if it's maybe a different one this time) ^ [09:17] right [09:17] the handle of those branches is a bit tricky [09:17] I'm going to do a merge --this (or whatever the syntax is) to say to prefer the current version [09:17] yeah [09:17] unsure at this point how we could improve that though [09:17] it wasn't done for having different distros using the same branch IMHO [09:18] but we can't switch in and out [09:18] tricky [09:18] jibel, correction - it's Radeon [09:19] jibel, which rather scuppers my plans to try a different driver [09:23] willcooke, are you using gnome-shell's native remote desktop or a legacy VNC server? If the latter then it probably won't work in (X)Wayland sessions, and will show all black [09:25] Because no external VNC server can read/see the desktop in Wayland, probably [09:26] duflu, I tried both [09:26] and I'm using the X session [09:26] I now think that it's not "a black screen" [09:26] Very dark grey? [09:27] I ran xev and moved the mouse around on the whole screen and xev saw nothing at all [09:27] duflu, lol [09:27] I might have to get a screen out of the loft and plug it in, maybe there is something on the screen like "do you want to allow...." [09:29] Actually I don't even know if gnome-shell's native remote desktop support is finished/enabled/built yet [09:29] When I build mutter myself it always says "no" [09:29] It works using the built in one on my intel based laptop [09:30] Cool. I am probably just missing build-deps [09:30] does VNC send mouse movement when no buttons are pressed? It's been ages since I've looked at the protocol [09:30] jamesh, erm, yes I think so, but I will confirm [09:31] well, I can see the mouse moving around on the laptop when I move it via vnc, so I would assume yes [09:31] that'd be client related rather than server related, either way [09:32] ah right [09:32] I doubt it. The client would be able to detect any local errors. The client can't detect the difference between a server that's meant to be reporting a black screen and one that's not [09:37] didrocks: mhmh... Normally to avoid this I always made sure debian side merged through same thing then merging with Ubuntu... But it's possible that some diversion happens and it's why I wanted some time ago wanted to discuss how to handle it [09:38] Trevinho: yeah, it's an annoying thing we didn't anticipate [09:38] Trevinho: btw, did you see my questions about the 3 extensions, were there any coordination (apart from the bug filing) in debian side to update the extensions? [09:39] as right now, .92 is stuck in proposed due to this (breaks: other-extensions) [09:39] (pomodoro is synced btw, and the autopkgtest works) [09:53] didrocks: nope, I didn't do anything on extensions... Maybe I should had 😢 [09:54] And that problem... I did anticipate but I was blamed for trying to handle the problems before they were a reality [09:54] hum, I hope that those 3 upstreams we are blocked on released patches [09:54] but we need to package it, and it's in sync in debian [09:54] would be quicker for someone to do it directly in Debian and upload [09:55] rather than diverging (or waiting for Debian to review the patches, waiting with gnome-shell blocked in -proposed meanwhile…) [10:08] it would be an option to kick them out to proposed [10:35] hmm, I can't type compose characters into my terminal [10:35] this is a problem for certain of my passphrases that I need to be able to type to upload /o\ [10:36] ok, I can paste them, but still ;_; [10:37] didrocks: I think they wanted 3.30 to go to unstable, so do you want to update the branch for that or should I do it? [10:37] * Laney has done mutter already [10:38] but not gnome-shell (yet) [10:56] Laney: didier had https://code.launchpad.net/~didrocks/ubuntu/+source/gnome-shell/+git/gnome-shell/ but from what I see there was no patches refresh yet, or did I miss the commit? [10:57] probably no updates were needed [10:57] those are the best kind of updates [10:58] I'll just get to work on merging exp into master for debian [11:16] https://git.launchpad.net/~laney/ubuntu/+source/gnome-shell <- that's ubuntu/master with the new debian/master merged in [11:17] only difference is the commit I made to update references to exp branch & watch file [11:19] jibel, what are the rules around utah? whoever spots a bug, gets to fix it? [11:20] i see this change: https://bazaar.launchpad.net/~utah/utah/dev/revision/1126 [11:20] however in https://platform-qa-jenkins.ubuntu.com/view/Desktop/job/ubuntu-cosmic-desktop-amd64-smoke-default/88/console [11:20] I see [11:20] UTAH exception: initrd file does have have gz or lz extension: /var/lib/utah/vm/utah_vm_G1kh05lzIYOIe4qu/tmpf2NIif/initrd [11:21] i wonder if UTAH is getting confused by are multi-initrd which has initrd, intel microcode, and amd microcode archives. === pstolowski is now known as pstolowski|lunch [12:08] Laney: sounds good, thanks! [12:08] Trevinho: yeah, no patch refresh needed [12:14] xnox, this one is fixed but there is another issue where the image fails to reboot after installation [12:14] xnox, heber is on it === pstolowski|lunch is now known as pstolowski [12:41] didrocks: ah, ok... as I thought something went in, but I guess it's post .30 then [12:42] didrocks: can you merge this in the mean time https://code.launchpad.net/~3v1n0/ubuntu/+source/gnome-shell/+git/gnome-shell/+merge/354321? [12:42] I don't want to do the release now though, just leave unreleased as I prefer to wait 3.28.4 [12:43] or maybe we can do an upload just with the search fixes? seb128 mentioned that, but if we go with an official point release is quite easy to verify anyway [12:46] ok [12:48] Trevinho: well, we need first to be released to cosmic [12:49] your patch* [12:49] didrocks: yeah, sure but for the VCS we can still merge, then in case we change there, I mentioned just not to do back and froth with branches around, but as you prefer [12:49] Trevinho: sure, I meant, in case you want to release it [12:50] sure sure... Although seb mentioned sometimes that is fine when a things i "on the way for sure", but I'm not rushing on releasing [12:50] I guess there is no urgency, let me just merge it [12:51] exactly no urgency... just prefer not to loose track of things [12:51] I'm preparing other components too, while for calculator I'm also updating to .2 [12:51] Trevinho: done [12:52] ta [13:17] didrocks: do you want me to upload gnome-shell? [13:17] didrocks: for gnome-calculator, the SRU can I merge on debian 3.28.2-1 first (no feature changes https://paste.ubuntu.com/p/T3wQfD7wnf/) and then apply that patch or I should just merge with updated upstream and do the versioning side alone? As since debian didn't really change anything I guess i can proceed this way, no? [13:17] Laney: to debian? Please do, I can handle the mutter sync + gnome-shell upload to cosmic [13:18] Trevinho: agreed, if the diff are minor and fix issues, I see no reason to separate your fix and the maintenance release upload [13:19] didrocks: no, to cosmic, since I re-merged, but feel free to handle [13:19] unstable is done done done [13:20] didrocks: need to do the classic merged with debian line though, right? [13:20] Laney: oh, feel free to upload if you prefer [13:20] I have no strong opinion :) [13:21] Trevinho: unsure what is the classic merged? ;) [13:21] remaining changes... [13:22] Trevinho: depends on the release team, they don't really like extra diff not fixing bugs, but if you can convince them… [13:22] should probably wait for the accepted emails from dak [13:23] Laney: still unsure, you are handling it? (I'm fine with that, you did the commits on top…) [13:23] ok yaru snap "fixed" ;) [13:23] sure! [13:23] thx ;) [13:23] next upstream release -> I'll sync a cosmic upload before UIF [13:23] yaruuuuuuuuuuuuuuuuUuuuuuuuuUUUUUUUUUUUUUUUuuuu [13:24] autopkgtest for autopilot-gtk/1.4+15.10.20150826-0ubuntu1: amd64: Regression ♻ , arm64: Regression ♻ , armhf: Regression ♻ , i386: Regression ♻ , ppc64el: Regression ♻ , s390x: Regression ♻ [13:24] :< [13:26] flaky tests for sure and you just got unlucky :p [13:26] (or should play to $$$ game) [13:27] http://autopkgtest.ubuntu.com/packages/a/autopilot-gtk/cosmic/amd64 [13:27] really unlucky! [13:28] 0, 1… it's quite close still! [13:28] still stuck behind glibc anyway [13:28] are you sure? gnome-shell doesn't show it being stuck anymore by it [13:29] oh no glib2.0 [13:29] ah ;) [13:29] but I didn't sync that so I'm not feeling too responsible for getting that to go in right now :-) [13:30] heh [13:30] wdyt we should do for gnome-shell? [13:30] (regarding extensions) [13:30] kill them to cosmic-proposed [13:31] it means that if we didn't look at it, there are some chances that dashtopanel and other aren't installable [13:31] they won't be in the release, that's right [13:31] same as what'll happen in debian [13:31] (I don't know if people are using the extension websites or packages) [13:31] oh, debian may remove them? [13:31] they'll be auto removed from testing [13:31] https://tracker.debian.org/pkg/gnome-shell-extension-dash-to-panel [13:31] see the note there [13:32] interesting [13:32] ok, so removing them in the release pocket (and not -proposed), yeah mean? [13:32] there's a script demote-to-proposed or something [13:32] not in ~ubuntu-archive [13:32] which copies to cosmic-proposed and delete from release [13:33] it's in lp:ubuntu-archive-tools [13:33] oh, I didn't add it to my $PATH one this laptop… [13:33] * didrocks fixes [13:33] should block them in proposed-migration too otherwise they might get copied straight back [13:33] ok, I'll do that, makes sense [13:33] * Laney does [13:33] oh right ;) [13:36] there [13:36] now it should pick copying gnome-shell rather than the extensions [13:36] ok, I'm running demote-to-proposed now [13:37] winning [13:37] I'll push gnome-shell to git, will upload once the new one moves over [13:37] and done! yeah, sounds good to wait for the new one moving now [13:37] thx Laney [13:39] didrocks: spotted another small thing https://code.launchpad.net/~3v1n0/ubuntu/+source/gnome-calculator/+git/gnome-calculator/+merge/354328 [13:40] hmm there's already an ubuntu/3.30.0-1ubuntu1 tag there but that's a bit different to what I'm going to upload [13:40] shall I la la la and delete it? [13:41] Laney: yeah, that was my thing about "need to be redone or pending review in debian", but that's fine, just delete and replace it [13:41] alright [13:42] Trevinho: merged [13:43] interesting, I didn't work on calculator, but I see that git push is failing on pristine-tar and upstream/latest and there are conflicts :p [13:43] weird though, because it seems people used -f rather than merging, it's puzzling [13:44] O_O [13:50] didrocks: thanks, also this branch needs to be pulled to ~ubuntu-desktop as there's not a an ubuntu/bionic yet: [13:50] https://code.launchpad.net/~3v1n0/ubuntu/+source/gnome-calculator/+git/gnome-calculator/+ref/ubuntu/bionic [13:52] Trevinho: can't we create one so that I can see the diff? ;) [13:52] didrocks: yep, just do it with `git checkout -b ubuntu/bionic ubuntu/1%3.28.1-1ubuntu1` [13:52] didrocks: so I can prepare the MP [13:53] Trevinho is on the path to upload rights, I like it [13:54] not sure I want them :D [13:54] Trevinho: ubuntu/bionic branch create [13:54] created* [13:54] cool [13:55] I could have put the gbp and control commits before the merge... but meeh, I'm lazy now to rebase :) [13:56] lots of diff, but actually debian/ diff is small [13:57] Trevinho: did you have a script for generating the Yaru changelog? [13:57] didrocks: mhmhmh, wait [13:57] didrocks: `gbp dch` just that [13:58] interesting ;) [13:58] you have to re-order the entries because it doesn't group by person [13:58] works well indeed [13:58] bit annoying [13:58] ah ok Laney [13:58] thx for mentionning! [13:58] like if it's Laney didrocks Laney, you get two [ Laney ] bits [13:58] yep.. [13:58] yep, got you [13:59] I mentioned the same last day... need to fix that upstream, laney I saw you already sent a patch to Guido, so you're the good one to ask for fixing that too :) [14:00] nah, bus factor is bad 😌 [14:00] I think I screwed things in calculator... let me redo it [14:01] gbp dch --multimaint-merge you can probably set that in debian/gbp.conf to [14:01] too [14:02] --multimaint-merge Merge commits by maintainer, default is 'False' [14:02] yes [14:02] I just found that :> [14:03] ALSO https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=881467 [14:03] Debian bug 881467 in git-buildpackage "git-buildpackage: gbp dch should filter out commits behind upstream-vcs-tag" [Wishlist,Open] [14:03] the tip in there is going to change my life [14:04] oh, lovely... `dch dch debian` for life [14:04] s/dch/gbp/ [14:04] I've been manually deleting those entries [14:22] didrocks: mhmh so https://code.launchpad.net/~3v1n0/ubuntu/+source/gnome-calculator/+git/gnome-calculator/+merge/354330 is the merge, but... there's lot of "garbage" because i think debian introduced to the VCS pre-compiled files, so the issue... however `git diff debian/3.28.2-1` is what you might look at or `git diff ubuntu/1%3.28.1-1ubuntu1 debian` as the rest is just garbage happened on merging with upstream that is not anymore in [14:22] debian/master but was there in debian/3.28.2-1 tree [14:23] ah, interesting, autopilot was relying on some defaults in ambiance [14:23] ambiance won't ever change! :) [14:23] Laney: so, failing for quite a while? But I guess nothing triggered it [14:23] Trevinho: ok, I'll have a look (maybe later though, testing Yaru for now) [14:23] it fails now that Yaru is default I guess [14:24] didrocks: sure, waiting new fresh look :) [14:24] but really it should be setting properties if it wants to test them, not assuming the theme is static [14:24] waow, the over-amplification is interesting [14:24] * didrocks prefers it to default upstream one [14:24] Laney: yeah, on the other side, they want to test on the default [14:24] Laney: what pkgs? [14:25] what are they relying on? [14:25] assuming some border width is 5 [14:25] ah, yeah, so that's more debatable :) [14:26] it's like "give me all the boxes with this width" to test if selecting works [14:26] but the only reason they had that width was because of the theme [14:26] test's .ui file should set it [14:27] right [14:28] 👍 [14:38] SHELLLLLLLLLLLLLLLlllllllllllllll [14:44] tjaalton, thanks for the xorg minor update, I'm still going to annoy you a bit by mentioning that libxss has a minor update available (https://gitlab.freedesktop.org/xorg/lib/libxscrnsaver/tags/libXScrnSaver-1.2.3) and libinput/wayland would be nice to update :) [15:27] taking gnome-getting-started-docs, brain is fried ;) [15:35] was waiting ages for the autopilot-gtk build to finish, turned out my ssh connection to home had frozen /o\ [15:38] "waow, that touch command was taking ageesssssss" [15:39] it was just after starting the testsuite [15:39] which sometimes doesn't output stuff for ages [15:39] ok, easy to get missed then [15:39] terrible timing :P [15:39] yep [15:44] 😸 (ok, can paste emoji, new gnome-characters works enough) [15:52] :) [15:59] seb128: do you remember you should redo the upload of https://code.launchpad.net/~3v1n0/ubuntu/+source/nautilus/+git/nautilus/+merge/350174 ? :) [16:08] Trevinho, yes, I asked you the other day if you could amend the vcs to have the right version if you remember? [16:09] seb128: ah, I forgot that bit... But release was fine with such version eventually, no? [16:12] no since they rejected asking to change and reupload [16:14] seb128: yes, but then sil corrected: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1782681/comments/3 [16:14] Ubuntu bug 1782681 in nautilus (Ubuntu Bionic) "Upgrade to 3.26.4 and SRU it" [High,In progress] [16:24] seb128: the problem with libxss for instance is that it's lacking six years of janitorial pkg maintenance, and the upstream changes are worthless [16:25] one manpage typo fix, five autogen cleanup commits, one warning fix [16:25] libinput 1.12rc is in a staging ppa, final expected next week === pstolowski is now known as pstolowski|afk [16:37] hi, i noticed there is http://cdimage.ubuntu.com/bionic/daily-live/current/ - but the name seems to be misleading, it was last built in july. [16:38] was this meant to be just a temporary build, is it going to go away soon, or will there likley be dailies, and the build scripts just dont work out, yet? [16:38] They are promoted to that directory when they pass automated testing - I think the QA guys are looking into why it's failing [16:38] the raw daily images end up in http://cdimage.ubuntu.com/bionic/daily-live/pending/ [16:39] thanks Laney! [16:39] no problemo 🤠 [16:46] tjaalton, k, thanks, fair enough for libxss I'll mark to ignore for our report (and good to know for libinput) [16:50] Trevinho, reuploaded nautilus bionic [16:56] willcooke: kenvandine Wimpress I added a spreadsheet link to our desktop doc we were in. It has some screenshots for some of the apps [16:56] popey, thx [17:07] thanks popey [17:09] popey, can we get edit rights? [17:13] on that note, quittin time. Night all [17:56] seb128: thanks [17:57] seb128: wayland uploaded [18:38] looks like my autopilot didn't work properly [18:38] despite having run the autopkgtests locally [18:38] WAH [18:39] oh yeah, it'll need the new gtk :/ [19:20] tjaalton, great :)