=== Guest94403 is now known as jzheng [05:57] good morning desktoppers! [06:02] Morning oSoMoN [06:02] hey duflu, how is it going? [06:02] * duflu runs away for a while [06:02] heh, see you later then [06:26] good mornng [06:31] good mrning didrocks [06:33] hey hey oSoMoN, how is it going? [06:35] didrocks, I could have used a couple of hours more sleep, but otherwise good. you? [06:35] morning didrocks oSoMoN [06:35] morning jibel [06:38] salut jibel [06:39] oSoMoN: ah, quite similar, was awaken multiple time this night, but yeah, overall good, can still type it seems :p [06:39] not even that in my case, just a short night :) [06:39] stop partying! [06:39] :) [06:39] I wish… [06:40] heh [06:40] typing is just muscle memory, the real test is whether what you write actually makes sense [06:41] I never claimed that I was writing meaningful sentences [06:41] good, I won't pretend I do either :) [06:41] ;) [06:49] good morning desktopers [06:49] salut seb128 [06:50] lut oSoMoN, comment ça va aujourd'hui ? [06:50] seb128, nuit un peu trop courte à mon goût, mais bien sinon [06:50] et toi? [06:51] pareil [06:51] je me suis couché à 1h45 c'était pas raisonable :-/ [06:57] ouch === Guest41615 is now known as fredp === fredp is now known as Guest83014 [07:38] should we show off the trash and mount volume icon in the GNOME Shell ubuntu session? [07:38] as it's not part of the docker anymore [07:38] (any input welcomed) [07:45] didrocks: Sounds exciting. Are they new? [07:45] duflu: no, it's just an override we have [07:45] to not display them [07:45] which makes sense in unity, as they were part of the launcher [07:46] now, on the GNOME Shell session, unsure what to do (I would tend not trash, but maybe mount volumes to display?) [07:46] didrocks, I would vote to show Trash. Because it's non-obvious, and even annoying, having to open Nautilus to find it [07:47] duflu: and mount point? would you be in favor or against them on desktop? [07:47] didrocks: I like seeing mounts at a high level. Preferably in a dock or panel, but desktop is OK [07:47] Apple thinks so [07:48] thanks for your feedbacks! Let's see what other says… [07:48] Interesting. It's localized in nautilus ("Rubbish Bin") but not on nautilus-desktop ("Trash") [07:49] ahah, Bin vs Trash, reminds me fun conversation in Unity times [07:57] didrocks, showing desktop icons is fine with me [08:00] seb128: trash & mounts? [08:00] hey willcooke [08:00] morning willcooke [08:00] didrocks, yeah, I've no strong opinion [08:01] ok, that's what I did [08:01] ok, change committed [08:02] unsure about the lock screen behavior, I've reverted to upstream default for now, but we already have that one listed. [08:02] morning [08:02] the rest is way clearer for maintenance reason [08:02] if anyone wants to have a look: http://paste.ubuntu.com/25268581/ [08:03] flexiondotorg: hey, I think you are using some of those schemas file in mate or was everything forked? ^ [08:03] flexiondotorg: if so, you may want to add ubuntu to your DesktopNames list as the last element [08:07] dropping off IRC for some testing, brb [08:24] good morning desktoppers [08:24] hey chrisccoulson [08:27] didrocks, for showing mount points, I use the Removable Drives extension, which adds a drop down menu to the mostly empty panel [08:27] hi seb128 [08:31] chrisccoulson: well, I guess we are not going to add more to the panelby default, hey ;) [08:31] (once we get the indicator dicsussion done, that will be something ;) [08:32] * didrocks doesn't understand why we have the same overrides in multiple files [08:33] especially when they differ :p [08:42] duflu, https://launchpad.net/~osomon/+archive/ubuntu/cr-vaapi-test/+packages (via oSoMoN - woot!( [08:42] ) [08:43] hum, the gtk+3.0 branch is painfully outdated [08:43] * didrocks refresh it [08:44] willcooke, cool! But also not cool: I forgot about another bugfix I need to publish.... [08:44] It was always in trello but I mistakenly thought it had been resolved externally [08:44] duflu, no worries [08:44] And yet another upstream team I will need to deal with [08:44] :/ [08:48] jbicha: updated the gtk branch, please think about doing it next time :) [08:48] (even if previous tag wasn't pushed) [08:48] which was on kenvandine[m][m] ;) [08:58] * didrocks is waiting for 10 minutes on set -e; LC_ALL=C.UTF-8 /usr/bin/licensecheck --check '.*' --recursive --copyright --deb-fmt --ignore '^(debian/(changelog|copyright(|_hints|_newhints)))$' --lines 0 * | /usr/lib/cdbs/licensecheck2dep5 > debian/copyright_newhints [08:59] lalalala [09:00] Trevinho, on my way to the meeting, running a few mins late [09:01] didrocks: why MATE should add Ubuntu to DesktopNames? They just install their own overrides if they want different default. [09:09] muktupavels: right, but most of the override will be the same than ubuntu ones, like fonts and such [09:09] muktupavels: and if they install without the :desktop_name, it will override all sessions [09:09] we don't really want that [09:10] If it is clean MATE install then there should not be ubuntu overrides in first place, no? [09:10] they should update their overrides to add :MATE [09:11] muktupavels: I think they install ubuntu-settings bin package [09:11] Personally I don't like idea about common name in DesktopNames. Unless you also plan to provide common autostart files, ShowOnlyIn, NotShowIn... [09:12] well, that's already the case [09:12] when you have an OnlyShowIn in a .desktop file, it's common [09:12] there is only one available [09:12] I bet Mate only have overrides on their delta [09:13] but like not on which font to use, curstorsize and such [09:14] How DesktopNames will look for Unity and Ubuntu customized session? [09:14] Still I think each desktop should install their own overrides [09:20] disagree, that will mean tons of duplication [09:20] sorry, but building gtk at the same time saturate my latop in term of io [09:21] http://paste.ubuntu.com/25268851/ [09:21] muktupavels: here is the override settings to cover base, ubuntu and unity ^ [09:21] imagine if I had to duplicate "ubuntu" for the whole "unity" [09:21] and same with mate [09:21] and same with… [09:21] that would create a mess in term of experience [09:21] for few settings difference [09:21] MATE overrides should not be in that file [09:22] yeah [09:22] just cp that file [09:22] and redo it all [09:22] which sounds… silly [09:22] So Unity will have DesktopNames=Unity:Ubuntu? [09:22] but I guess that the ultimate decision will be for the Mate team [09:22] yep [09:22] Unity:ubuntu actually [09:22] Upstream GNOME DesktopNames=GNOME? [09:22] right [09:22] How will look ubuntu customized GNOME session? [09:23] and GNOME Shell with our ubuntu tweak, DesktopNames=ubuntu:GNOME [09:23] as you can see in that file, it's about different fonts and themes ^ [09:23] And you suggest that MATE adds ubuntu to their DesktopNames? [09:23] (mostly) [09:23] yeap, like for Unity: DestkopNames=Mate:ubuntu [09:23] and implement deltas in :Mate [09:23] (if any) [09:24] ofc, that's if they didn't fork the org.gnome. schemas [09:24] which I don't know [09:24] that's why I think we should left that decision to them [09:24] just being careful is they are using "global/gnome" ones [09:24] Now you decide that it good to add autostart file with ShowOnlyIn=ubuntu [09:24] so that installing Mate doesn't broke the vanilla session [09:24] It will be autostarted in MATE too and that might not be what they want [09:24] autostart are going away [09:25] once sessions are being transitioned to systemd [09:25] and they can use NotShowIn=, which will take precedence [09:25] note that it's already the case, even without "ubuntu" [09:25] I think both can not be in same file? [09:27] gtk… test failing, sigh [09:27] * didrocks needs to reboot and test, brbr [09:31] Great. I can't patch plugins-bad because artful can't build it [09:33] duflu, build issue with the new toolchain or something? [09:33] seb128, Maybe. Let me see if it's overconfiguring because I have extra packages installed [09:34] duflu, it's not listed on http://qa.ubuntuwire.org/ftbfs/rebuilds/test-rebuild-20170706-artful.html [09:38] ok, I only have one bug with those per desktop overrides [09:38] gdm is, for unknown reason, picking the :GNOME variant, even overriding global default setting [09:39] no XDG_CURRENT_DESKTOP env variable in gdm GNOME Shell though [09:39] gnome-session sets XDG_CURRENT_DESKTOP=GNOME if nothing is set [09:40] https://git.gnome.org/browse/gnome-session/tree/gnome-session/main.c#n299 [09:40] not really [09:40] XDG_CURRENT_DESKTOP=GNOME-Greeter [09:40] it seems it matches "GNOME"? [09:41] in /usr/lib/gdm3/gdm-wayland-session [09:41] same in /usr/lib/gnome-session/gnome-session-binary [09:43] chrisccoulson, how is chromium 60.0.3112.78 looking? [09:43] * didrocks tests with GNOME-Greeter, brb [09:45] muktupavels: confirmed [09:45] so GNOME-greeter matches "GNOME" [09:45] want to have a look at the patch (want some feedbacks on it?) [09:49] are you sure that XDG_CURRENT_DESKTOP == GNOME-Greeter under gdm? [09:49] yep [09:49] just look at the environ of any process running under gdm on your system [09:50] daemon/gdm-session.c: return g_strdup ("GNOME-Greeter:GNOME"); [09:50] in gdm [09:50] so yeah, that's probably why it matches GNOME aswell [09:50] just didn't see it in the binary environ dump [09:50] let me look more carefully [09:51] it will use override it finds... [09:51] yep [09:51] simply use GNOME-Greeter for gdm... [09:51] that's what I've already done :) [09:51] let me confirm from a running process [09:52] yeah, that's correct: XDG_CURRENT_DESKTOP=GNOME-Greeter:GNOMEDCONF_PROFILE=gdm [09:52] (juts the binary format didn't help) [09:53] ok, good! Will use that for "global" settings [09:53] didn't spot anything wrong with the patch for now [09:53] seb128, *shrug* it looks like it's caused by the presence of Android headers on my machine. If the normal build environment lacks those then it wouldn't try building the broken region of code. Still, I fixed it with a one line makefile fix [09:54] org.gnome.desktop.interface:GNOME-Greeter [09:54] and then you dont need "Revert with default for GNOME user session" [09:55] yep, exactly what I've done, that for ubuntu and GNOME-Greeter on cursor theme [10:18] oSoMoN, yeah, it's fine. But timed badly - I've got this, firefox and an embargoed update to do at the same time ;) [10:18] (I'm working on it) [10:18] ok, thanks [10:18] oh, and rust. Anybody, feel free to take that off my hands ;) [10:20] ISTR seb128 volunteered to maintain rustc after a few beers at GUADEC last week [10:21] the beer effect… [10:21] heh [10:22] * didrocks seconds chrisccoulson's memory [10:23] /msg chrisccoulson how much would you pay me for telling this already? [10:23] doh :p [10:23] haha :) [10:24] @all: do you have 4 workspaces showing up in your ubuntu session, even if dynamic-workspaces is set to true? [10:26] flexiondotorg: if so, you may want to add ubuntu to your DesktopNames list as the last element -> Message from flexiondotorg : he's currently on holiday but he's seen your message and he will look at that when he gets back. [10:26] ah, thanks ouroumov :) [10:26] yw [10:28] seb128, it's weird. I had to build two debdiffs. One that is ideal for archive and one which works on my machine... Please try them out: https://bugs.launchpad.net/ubuntu/+source/gst-plugins-bad1.0/+bug/1698287 [10:28] Ubuntu bug 1698287 in gst-plugins-bad1.0 (Ubuntu) "VA-API fails to initialize in a Gnome Shell Wayland session" [High,In progress] [10:28] And good night [10:37] chrisccoulson, https://code.launchpad.net/~chrassig/ubufox/multiprocessCompatible/+merge/325042 [10:38] chrisccoulson, I don't remember that... [10:38] you are saying that it's because of the beers right? ;-) [11:08] retrying to everyone running artful on this channel: do you have 4 workspaces showing up in your ubuntu session, even if dynamic-workspaces is set to true? [11:08] didrocks, checking [11:09] didrocks, I have only two, not sure what the value of dynamic-workspaces is (where do I check?) [11:10] oSoMoN: it's a gsettings key, so 2 is dynamic workspace [11:10] didrocks, I've only 2 [11:10] seb128: oSoMoN: are you sure you are on the ubuntu session? [11:10] and not the gnome one? [11:10] didrocks, the setting in Tweaks says 4. static workspaces [11:10] only 1 shows up in activities [11:10] XDG_SESSION_DESKTOP=ubuntu-wayland [11:10] didrocks, ^ [11:10] so no, not ubuntu [11:10] but ubuntu-wayland [11:10] does that make a difference? [11:11] could do [11:11] if we have a distro patch with "ubuntu" [11:11] XDG_SESSION_DESKTOP=gnome here [11:11] /!\ gdm is buggy [11:11] but I'm pretty sure I had the same in an x11 session [11:11] so if you select another user [11:11] though gdm doesn't respect the session you pick [11:11] and never set a default session to it [11:11] it can go to the gnome one [11:12] ah, interesting, gnome-tweaks says that I have static workspaces [11:13] didrocks, 2 as well in "ubuntu" [11:13] ok [11:13] so, settings mutter to dynamic workspaces works [11:13] here [11:13] unsure the difference between the gnome shell and mutter settings [11:13] seb128: $ gsettings get org.gnome.mutter dynamic-workspaces [11:14] false [11:14] * didrocks puzzled [11:14] it's a recent artful daily install [11:14] like just before GUADEC [11:14] not an upgrade ? [11:14] I didn't change much [11:14] no [11:15] I had false [11:15] which is the default [11:15] same here [11:15] but true on the g-s settings [11:15] [11:15] This key overrides the key in org.gnome.mutter when running GNOME Shell. [11:15] [11:16] org.gnome.shell.overrides dynamic-workspaces true [11:16] org.gnome.mutter dynamic-workspaces false [11:16] can you try to reset both keys? [11:16] and I've 2 workspaces which behave in dynamic way [11:17] same values [11:17] yeah, but same shell behavior? [11:17] yes [11:17] 2 workspaces and dynamically changing [11:17] if I reset them both, if have: [11:17] true/false as you did [11:17] but 4 workspaces :/ [11:17] you have a ghost :p [11:18] the weirdest part [11:18] what does gnome-tweak-tool tells you? [11:18] is if I go the gnome session [11:18] I have dynamic workspaces [11:18] I wonder if we don't have a distro patch with something that I have installed after upgrade doing this [11:18] gnome-tweak-tool says it's static [11:18] and 4 workspaces [11:18] look at the tweaks code to see what it reads? [11:18] just what I downloaded :p [11:19] tweaks is telling it's dynamic for you? [11:19] self.settings = Gio.Settings("org.gnome.mutter") [11:19] self.key_name = "dynamic-workspaces" [11:21] didrocks, https://pasteboard.co/GEGTPsP.png [11:21] hum [11:21] so it disagrees with the mutter setting [11:21] not for me [11:21] ups [11:21] well, you mutter settings is false [11:22] yes you are right [11:22] and I don't see the code looking for the shell override [11:22] weird [11:22] isn't it? :p [11:22] restart of the shell doesn't change anything [11:22] nope [11:23] and I have that on multiple run [11:23] the weirdest part is having that only in the ubuntu* sessions for me [11:23] and not on gnome one [11:23] if that was dconf related, it would be global [11:23] that's why I'm thinking about a distro patch somewhere, as this is an upgrade [11:23] and an additional package I have you don't [11:25] yeah, still weird [11:25] I can try to look a bit more in a bit [11:25] but first moving location, be back in 10 min [11:25] if you have time, will be great :) [11:25] yeah [11:25] brb [11:58] seb128: just did a diff with a new user between the ubuntu and gnome session [11:58] on the gsettings database [11:58] nothing in the diff relevant [11:58] I'm pretty sure we have a distro-patch for num-workspace somewhere if "ubuntu" is in session name [11:59] (something we don't install anymore by default maybe, which will explain why you are not seeing it [11:59] ) [11:59] distro patch to what? [11:59] unsure, this is what we would need to find out [11:59] or I would have fixed it :p [11:59] do you get the same behaviour with a new user on your machine? [11:59] yep [11:59] ubuntu -> 4 workspaces [11:59] gnome -> dynamic [12:00] unsure if anyone else have a desktop with unity they upgraded [12:01] I bet Laney does [12:02] I wiped my old upgraded machine [12:07] seb128: are you on a vm or bare install? [12:07] bare install [12:07] ok, so I won't screw you up :) [12:08] what did you want to try? [12:08] installing unity [12:08] (well, unity-session) [12:08] and see the behavior after a logout/login [12:08] I've a difficult time believing that one package installed by unity could change that [12:08] what do you think it is thus? [12:08] we don't have shell extensions or such [12:09] good question [12:12] in G-S code it's quite clear that it's the g-s override settings which get on top of mutter one [12:12] so even unsure why it "works" when I use the mutter one [12:12] wonder if there is something with the mode [12:12] let me try to hack the desktop file to remove the mode [12:13] ok, that's it [12:14] (but tweaks is broken anyway) [12:14] didrocks: the gtk test failure is LP: #1709225 [12:14] Launchpad bug 1709225 in pango1.0 (Ubuntu) "gtk test regressions with pango 1.40.8" [High,Confirmed] https://launchpad.net/bugs/1709225 [12:14] thanks jbicha [12:16] didrocks, what mode did you remove from where? [12:16] I didn't remove any mode [12:17] I did add a mode [12:17] let me try to hack the desktop file to remove the mode [12:17] the net effect is that the keys in org.gnome.shell.overrides from the main path: /org/gnome/shell/overrides/ aren't read anymore [12:17] or I read that wrong [12:17] ah, for testing, yeah [12:17] I did remove the "ubuntu" mode I added [12:17] so the issue is due to local changes you didn't upload yet? [12:17] and those keys are changing GNOME Shell behavior compared to mutter [12:18] yep, the thing is that I have that experimental mode for quite a while [12:18] this is why you didn't get it yet [12:18] k [12:18] that explains it [12:18] yep :) [12:18] so, basically GNOME Shell disagrees with mutter defaults [12:18] and use those keys ^ [12:18] which can be changed by mode [12:18] but even the default "no mode" read it [12:19] however, there is no fallback mecanism [12:19] like if the mode overrides keys aren't found [12:19] it will fallback to mutter defaults [12:19] (I'm sure Tweaks don't take that into accounts) [12:19] the only way I can think of is either shipping our own overrides, or override mutter default based on desktop name [12:20] as I think the new glib feature will enable us to remove those GNOME Shell overrides [12:20] (for classic and other modes) [12:20] I would rather go with mutter overrides [12:20] wfm [12:20] we are talking about 5 keys FYI [12:21] ok, at least this canbe explained :) [12:22] https://git.gnome.org/browse/gnome-shell/tree/src/shell-global.c#n1489 [12:22] yeah [12:22] I was expecting the default would be "no mode path -> revert to .overrides" [12:22] which is G-S without modes [12:22] as our session mode is inheriting from "user" [12:23] but global->session_mode seems to be the exact string your pass to the mode [12:23] (also, having the list hardcoded is… hem…) [12:24] so, let's use the per desktop overrides in mutter [12:24] and bring back this sanity to GNOME Shell as well with this [12:24] muktupavels: making sense to you as well? ^ [12:26] didrocks? removing override schemas? yes [12:26] muktupavels: right! [12:26] and only using the :desktop_name override to implement the samebehavior [12:43] Salutations! please #see [12:43] https://bugs.launchpad.net/ubuntu/+source/grub-installer/+bug/1709306 [12:43] Ubuntu bug 1709306 in grub-installer (Ubuntu) "Xubuntu installer with usb in UEFI BIOS mode ubiquity GRUB Installer crashed with "grub-efi-amd64-signed" package." [Undecided,New] [12:43] and [12:43] https://bugs.launchpad.net/ubuntu/+source/grub-installer/+bug/1629348 [12:43] Ubuntu bug 1629348 in grub-installer (Ubuntu) "installer crash (grub2) during installation in UEFI mode" [Critical,Invalid] [12:44] hey sary, unsure you are on the right channel or going the right way about explaining what you want? [12:45] seb128: unsure! care to point me to the right team IRC channel ..! [12:48] any other additional informations i should add to the bug report before leaving the live-session! [12:50] sary, try speaking to cyphermox on #ubuntu-devel [12:51] seb128: am on it , thanks! [12:51] yw [13:31] moin [13:34] hey Laney [13:35] bon matin Laney [13:35] how are you? enjoying debconf? [13:35] hey Laney [13:41] hey seb128 oSoMoN [13:41] seems ken left [13:41] yeah it's good, getting tired and slow though [13:41] me and ximion have been fixing stuff [13:41] he's just rebooting after removing/reinstalling half his desktop [13:41] Laney, ah nice :-) [13:41] hmm [13:41] side chats! [13:41] * seb128 likes the sound of things fixed [13:41] Laney, do you know if "g-s doesn't find printer drivers" is a known issue? [13:42] work in fedora [13:42] I tried to install a printer yesterday and it gave me a "can't find driver" when on a fedora livecd it works on the same printer [13:43] no, don't know how that works [13:43] do they have metainfo files or something? [13:48] no clue [13:48] look inside the pkg [13:48] i know fedora added a lot of those things downstream [13:48] the origin was fedora [13:48] in the g-s details page [13:49] right, so get the source / rpm and look if it has a metainfo file [13:49] but it was on a live iso on my laptop (usb printer detection doesn't work in a vm) [13:49]