2021/04/09/#ubuntu-desktop.txt

[01:57] <KGB-2> gnome-shell ubuntu/master Daniel van Vugt * [update] merge request !47: Drop layout-Allow-updating-struts-in-Overview.patch * https://deb.li/3kKUl
[02:15] <callmepk> good morning
=== popey0 is now known as popey
[02:48] <duflu> Morning callmepk 
[02:48] <duflu> And *something*, popey?
[02:49] <duflu> Morning I guess
[02:50] <callmepk> morning duflu 
[05:48] <oSoMoN> good morning desktoppers
[05:59] <kedar_apte> like the word desktopers :)
[06:00] <duflu> Morning oSoMoN 
[06:01] <ricotz> good morning
[06:05] <oSoMoN> hey kedar_apte, duflu, ricotz 
[06:25] <luna> morning
[06:28] <seb128> goood morning desktopers
[06:29] <oSoMoN> morning luna, salut seb128 
[06:32] <ricotz> hey oSoMoN luna seb128 
[06:37] <duflu> Morning luna, seb128, ricotz 
[06:38] <seb128> hey oSoMoN, ricotz, duflu, happy friday! how are you?
[06:38] <duflu> seb128, alright. Sadly fearing the weekend and all the work it entails. How are you?
[06:39] <seb128> duflu, tired and still not over that cold but otherwise alright though sort of not ready for the weekend, those are restless atm...
[06:40] <oSoMoN> I'm tired but fine otherwise, not fearing the week-end that much :)
[06:41] <seb128> brave :)
[07:09] <didrocks> good morning
[07:47] <duflu> Morning didrocks 
[07:50] <didrocks> hey duflu 
[08:06] <seb128> hey again, sounds like I was on the empty side of the server split
[08:06] <seb128> or things just started to unsplit now?
[08:09] <Nafallo> morning seb128 :-)
[08:10] <Nafallo> morning desktopers
[08:12] <oSoMoN> good morning Nafallo 
[08:19] <seb128> hey Nafallo
[08:20] <didrocks> salut seb128, Nafallo, oSoMoN 
[08:21] <Laney> hey seb128 Nafallo oSoMoN didrocks 
[08:21] <Laney> guessing I was on the other side of a split!
[08:22] <Nafallo> damn. you're all so flexible... :-O
[08:23] <oSoMoN> hey Laney 
[08:25] <didrocks> hey Laney 
[08:25] <didrocks> it was so quiet this morning :)
[08:29] <marcustomlinson> morning callmepk duflu oSoMoN ricotz luna seb128 didrocks Nafallo Laney
[08:30] <duflu> Hi Laney
[08:33] <callmepk> hi marcustomlinson  oSoMoN ricotz Laney seb128 didrocks Nafallo Laney 
[09:17] <k_alam> Laney Hello, With glib 2.67.5 all calls for g_clear_pointer and g_object_ref causing error in Unity...I tried to workaround with add_definitions(-DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_66) to Cmakelist.txt as described here https://gitlab.gnome.org/GNOME/glib/-/merge_requests/1715...but then gtk icon tests are failing...what is known woraround
[09:17] <k_alam> here? Will it compile with 2.68 ?
[09:17] <k_alam> https://gitlab.gnome.org/GNOME/glib/-/merge_requests/1715
[09:24] <Laney> k_alam: not sure, I dont think it would be the cause of tests starting to fail to be honest
[09:29] <Laney> duflu: https://github.com/snapcore/pi-gadget/commit/5c701c62e9502999a4af3ba0f071bf3b764e1b30 looks like it already happened
[09:29] <duflu> Laney: Same time as https://launchpad.net/bugs/1918110 then
[09:29] <ubot5> Ubuntu bug 1918110 in flash-kernel (Ubuntu Groovy) "overlay_map.dtb is in wrong location" [Undecided,New]
[09:34]  * Laney claims ignorance on that one
[09:34] <seb128> Laney, I reworked my sponsoring queue collector change to avoid the web parsing
[09:35] <Laney> ah nice!
[09:35] <Laney> let me get to that today
[09:39] <seb128> thx
[09:49] <jamesh> k_alam: do you have a sample of the compile failure?
[09:51] <k_alam> jamesh: https://launchpadlibrarian.net/532623784/buildlog_ubuntu-hirsute-amd64.unity_7.5.0+20.04.20200312-0ubuntu1+202104090712~ubuntu21.04.1_BUILDING.txt.gz
[09:52] <duflu> Laney: That's the bug which meant switching from 'fkms' to 'kms' didn't work
[09:53] <duflu> Till now
[09:53] <seb128> tjaalton, did you see the xorg-server build failure and new comment on https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/1920096 ?
[09:53] <ubot5> Ubuntu bug 1920096 in OEM Priority Project focal "Cursor bleeds on the platform with Mali armsoc driver" [Critical,New]
[10:00] <Laney> aah
[10:01] <tjaalton> seb128: huh :D
[10:01] <jamesh> k_alam: It's not obvious what in the build env would cause that test to fail.  That said, it's not obvious why Unity would have a test doing that
[10:02] <seb128> Laney, did you said you had a fixed glib uploaded or did I misunderstand?
[10:02] <Laney> flatpak
[10:02] <Laney> it just got accepted
[10:03] <Laney> see if it works!
[10:03] <seb128> ah
[10:03] <seb128> thx
[10:04] <KGB-0> mutter ubuntu/master Daniel van Vugt * [open] merge request !72: Add 61-mutter.rules-Enable-KMS-modifiers-for-Raspberry-P.patch * https://deb.li/lDai
[10:59] <k_alam> jamesh: My unity branch is here: https://code.launchpad.net/~khurshid-alam/unity/+git/unity/+ref/build-hirsuit, and the recipe is https://code.launchpad.net/~khurshid-alam/+recipe/unity-build-hirsuit....on local machine it compiles for groovy...may be some regression with gtk
[11:00] <jamesh> k_alam: the test is making assertions about what it believes the current default theme is, which would depend on the build environment
[11:00] <jamesh> k_alam: are you building with e.g. sbuild?
[11:01] <k_alam> the current default theme is yaru, it is on build-depends, no I am using debuild
[11:02] <jamesh> k_alam: if you want to try and reproduce the failure locally, you'll probably need to use sbuild.  That should use an environment closer to what is used by the LP builders
[11:02] <jamesh> It basically constructs a minimal chroot containing only the packages you claim to depend on and then runs the build in that env
[11:26] <jamesh> didrocks999: This is the polkit change I was thinking of: https://gitlab.freedesktop.org/polkit/polkit/-/merge_requests/75 -- however it only seems to update the search path for JS rules rather than the XML .policy files
[11:31] <didrocks999> jamesh: waow, this is indeed really recent, but yeah, seems to be only JS rules… Happy that we found a way so that we can still run tests even on focal
[11:50] <GunnarHj> Hello seb128! I just filed bug #1923187. Would you be able to run the commands to promote libmarisa0 to main?
[11:50] <ubot5> bug 1923187 in ibus-libpinyin (Ubuntu) "[FFe] Enable opencc support" [Wishlist,New] https://launchpad.net/bugs/1923187
[11:53] <seb128> GunnarHj, hey. it should go the other way around, you upload first which makes it show on component mismatch then we promote, especially that if that ffe doesn't get accepted then we have no need to promote the package in hirsute no?
[11:54] <GunnarHj> seb128: I see. So if the FFe is accepted I just upload and the promoting is fixed afterwards.
[11:56] <seb128> GunnarHj, correct
[11:57] <GunnarHj> seb128: Thanks for clarifying.
[12:01] <seb128> np!
[12:03] <GunnarHj> seb128: Another thing: Do you have an idea offhand why lightdm fails to build in hirsute?
[12:03] <GunnarHj> https://launchpad.net/ubuntu/+archive/test-rebuild-20210325-hirsute/+build/21263293
[12:05] <seb128> GunnarHj, libsystem.c:333:1: error: no previous prototype for ‘__xstat’ [-Werror=missing-prototypes]
[12:05] <seb128> it seems?
[12:06] <seb128> GunnarHj, it was fixed upstream with https://github.com/canonical/lightdm/pull/168/files, let me cherry pick that fix
[12:08] <GunnarHj> seb128: I can do that if you like. (I'm about to make a tiny guest session fix but got stuck.)
[12:09] <seb128> GunnarHj, if you want sure
[12:09] <seb128> thanks
[12:41] <GunnarHj> seb128: That seemed to be it as regards lightdm. However, there is some symbols noise in the buildlog:
[12:41] <GunnarHj> https://launchpadlibrarian.net/532666335/buildlog_ubuntu-hirsute-amd64.lightdm_1.30.0-0ubuntu4~ppa2_BUILDING.txt.gz
[12:41] <GunnarHj> Is that important enough to require a refresh of the symbols file?
[12:42] <seb128> GunnarHj, those are c++ symbols marked as optional, I would let them as they are
[12:42] <GunnarHj> seb128: Ok, thanks. Then I'll do some tests and upload later.
[12:43] <seb128> 👍
[13:16] <popey> didrocks999 there's an interesting thread on the GNOME discourse. I mentioned ubuntu-report. You might want to take a look :). Could be an interesting way to work with upstream. https://discourse.gnome.org/t/help-creating-a-simple-reporting-app/6076
=== didrocks999 is now known as didrocks
[13:23] <didrocks> popey: good answer! I will follow it and see what Allan answers to it, keep me posted in case I missed anything, but indeed, ubuntu-report can completely be repurposed to it. In addition, it provides a C library to integrate with any graphical application (this is how gnome-initial-setup interacts with it)
[13:24] <popey> Nice!
[14:20] <apw> seb128, i am using the default 'LP: #12345678' highlight (I think this is default) when the item is near the bottom of gnome-terminal (less that about 6 lines) the menu fails to open; the cursor changes but the menu does not appear.  at about 7 lines it starts appearing and is even shifted so the cursor is bottom left still.
[14:20] <ubot5> Error: Launchpad bug 12345678 could not be found
[14:50] <hellsworth> good morning desktopers
[14:53] <didrocks> Laney: compared to this morning: more integrations tests and they are now running on Github Action: https://github.com/ubuntu/adsys/pull/118 ! :) (still need the one mocking AD though)
[14:54] <didrocks> hey hellsworth 
[14:56] <hellsworth> hey
[14:56] <hellsworth> hey there didrocks :)
[14:56] <Laney> didrocks: NIIIIIIIIIIIICCCCCCCCCEEEEEEEEEEEEEEEEEEEEEEEEEE
[14:57] <didrocks> and the most crazy commit of the day: https://github.com/ubuntu/adsys/pull/118/commits/d190d211a2ed5b4350a4d720b0c9b0a768d04b0e (running the same tests twice, but alternating which side is in process for coverage and which side is a subprocess)
[15:00] <didrocks> and just reached 650 tests!
=== ijohnson is now known as ijohnson|lunch
[18:20] <RikMills> glib2.0 migrated :)
[18:21] <RikMills> flatpak fix by L@ney obviously worked
[18:29] <oSoMoN> nice
=== ijohnson|lunch is now known as ijohnson