[05:14] good morning [05:37] morning [05:37] Hey didrocks [05:38] hey jibel === pavlushka is now known as Guest19993 === Guest19993 is now known as pavlushka [06:47] Morning didrocks, jibel [06:48] good morning desktopers [06:48] hey duflu, how are you? [06:49] Morning France [06:49] duflu, we finally got the ubuntu-restricted change unblocked, thanks to jbicha for the upload [06:49] hey flexiondotorg [06:49] And duflu [06:49] seb128, going well. I possibly just finished the final form of the clutter-gtk CPU fix. Upstream will have to decide what form of patch they like, but it's also still waiting for a separate smoothness fix [06:49] ah, nice [06:49] seb128, how are you? [06:50] I'm good, could have done with a bit more sleep though [06:50] I was up hacking on g-c-c until 1:30 [06:50] Fun [06:51] indeed [06:51] well at least I got things working [06:51] I've a new panel in 3.25 for the dock settings [06:51] now up to Didier to deal with the backend side [06:52] Cool. Users love settings. I vaguely recall we had a lot more to play with in Gnome 2.x [06:52] and never recovered fully [06:53] In the old days you could set refresh rates and subpixel orders ;) [06:55] Interesting thought - Any machine with <= half the power of my desktop would have been too underpowered to be able to redraw a FHD 60Hz totem window. The user experience for plenty of people should improve soon [07:04] great === maclin1 is now known as maclin [07:45] seb128, didrocks on artful just after the login the deskop icons apper behind the dock [07:45] andyrock, yes, known issue [07:45] until it refreshes [07:45] is anyone working on it? [07:46] andyrock, it you want to work on a patch that would be great [07:46] no [07:46] kk [07:46] andyrock, https://bugzilla.gnome.org/show_bug.cgi?id=760651 [07:46] Gnome bug 760651 in Desktop "Desktop icon under top panel" [Normal,New] [07:49] andyrock: FYI, I implemented this + intellihide behavior in 2011 [07:49] andyrock: you might want to check there [07:50] andyrock: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1711377 has more info [07:50] Ubuntu bug 1696621 in nautilus (Ubuntu) "duplicate for #1711377 Desktop icons slip under the top bar and dock" [Medium,Confirmed] [07:50] (even if dupped) [07:50] there is the first refresh issue [07:51] + the fact that nautilus should learn about the icon size in intellihide mode (which isn't the STRUT, has we still want apps to maximize fullscreen) [07:51] didrocks: thanks [07:57] didrocks, andyrock, the duplicate status is wrong imho, Didier's report included 2 issues, one is a dup and an upstream bug (the fact that it needs a refresh to do the correct layout) and one is a feature request (handling the dynamic behaviour) [07:57] indeed [07:57] I should perhaps have renamed/retargetted to the second issue when I commented [07:57] I would prefer to focus on the upstream bug [07:57] right,I was going to say that [07:58] it's the default config/the most noticable one [08:06] moin [08:07] * Laney has been to the dentist [08:07] morning desktoppers [08:07] Laney, was it ok? [08:08] * willcooke has the man from Lenovo coming today to replace the keyboard [08:08] got to go to the hygeinist next week /o\ === davmor2_ is now known as davmor2 [08:08] otherwise yeh [08:08] wisdom teeth man [08:08] :(( [08:08] Mine were growing well in to my 30s I think [08:09] Laney: don't have them pulled out they contain all your wisdom man ;) [08:09] hey willcooke Laney [08:09] but at least there was room for them [08:09] they're horizontal or something [08:09] morning seb128 davmor2 [08:09] Morning all [08:09] hey seb128 [08:09] hey davmor2! [08:09] can I borrow yours? [08:09] my wisdom is lacking [08:11] Morning davmor2 [08:11] hey jibel [08:12] hey Laney, willcooke [08:12] morning didrocks [08:13] good luck Laney, once this all is done, remember me to tell you about some horrible story I have with my widsom teeth :) [08:13] hey didrocks [08:13] heh [08:13] we're not at the stage of doing anything yet :P [08:15] WAT, me ubiquity uploads failed to build [08:15] on ALL archs [08:16] but I test built it :((((((((((((((((((((((((((((((((((((( [08:16] so dbus-update-activation-environment was enough? [08:16] putting it earlier was the secret [08:16] Laney: ah ok ;) [08:16] not that I tested the actual debs [08:16] Laney: from the failure, I would say missing bzr add :p [08:17] maybe [08:17] but that computer is off now and I'm in the library so can't know today :P [08:17] cd upub [08:17] wtf [08:17] tilix ias gone mental [08:17] ;) [08:17] I could see both tabs at the same time [08:20] does anyone have a simple example of single instance checking code? [08:20] I should say non-gtk one [08:20] or non-gapplication [08:20] like command line unix style :p [08:22] depends on your language, but most of the time you write an .pid in a temp fixed location and write the pid in it [08:22] then, if the file exists, you check if the current pid is still running [08:22] if so, exit [08:22] or you overwrite it with your own pid [08:23] (and try to clean it when shutting down, ofc, the first case is to handle crashes ^) [08:23] thanks [08:23] it's C, and I was trying to avoid do all that work [08:24] what's seb128 doing? [08:24] sounds intriguing [08:24] trying to get update-notifier to not segfault under wayland [08:24] seb128: you have the run-once or so wrapper script from kirkland [08:24] I might just disable the single instance code [08:24] I can't be bothered to handle things properly [08:24] it currently uses the X clipboard for that [08:25] we have a contributor patch [08:25] https://launchpadlibrarian.net/332753117/0001-Replaced-up_get_clipboard-with-a-pidof-hack.patch [08:25] I could just apply it pretending it's fine :p [08:25] hackish hack day! [08:26] we currently have no update-notifier under wayland, which means no apport prompting for issues [08:26] which is not a good thing, we want feedback [08:26] (/me whistles) [08:27] oh, are we still using update-notifier for that [08:27] hey pitti! [08:27] hey pitti [08:27] bonjour didrocks et seb128 ! [08:27] isn't that the kind of thing that coudl be replaced with a session-systemd path unit [08:27] pitti, yeah, the apport maintainer never added a remplacement solution :p [08:27] you tell me :-) [08:27] that guy sucks! [08:28] he's probably writing javascript nowdays :) [08:28] * didrocks hugs pitti [08:28] sec [08:29] pitti, also no pkexec for system reports under wayland... [08:30] urgh, we use pkexec a lot, don't we [08:30] sort of yes :-/ [08:30] due to its efforts to grab the focus/keyboard? [08:30] (which is X only?) [08:30] I thought g-shell had a builtin polkit agent [08:31] sorry, in fact it seems it's only when .desktop use pkexec [08:31] https://bugzilla.gnome.org/show_bug.cgi?id=763531 [08:31] Gnome bug 763531 in overview "pkexec in .desktop no longer works" [Critical,Reopened] [08:32] * seb128 looks more at why system reports don't wokr [08:36] seb128: https://paste.ubuntu.com/25476960/ maybe that's useful to you [08:37] update-notifier already has the path units AFAICS [08:37] but probably not hooked up atm [08:40] Laney: but if the process crash, you don't have it cleaned up, correct? [08:40] Laney, thanks, I found that/considered it but was unsure if the predictable filename was an issue since something else could decide to hijack it [08:40] doesn't matter, it's not locked then [08:40] or take the lock [08:40] ok, as you wish [08:41] but I guess it can't do much damage [08:41] well, I'm just unsure [08:41] I'm not a security guy [08:43] what was the protection against that for the clipboard method, out of interest? [08:43] I guess none, you have a point :-) [08:44] Laney, thanks! [08:44] shrug [08:44] $ pkexec /usr/share/apport/apport-gtk /var/crash/_usr_lib_gdm3_gdm-wayland-session.1000.crash [08:44] Invalid MIT-MAGIC-COOKIE-1 keyUnable to init server: Impossible de se connecter : Connection refused [08:44] Erreur de segmentation (core dumped) [08:44] $ pkexec sleep 1; /usr/share/apport/apport-gtk /var/crash/_usr_lib_gdm3_gdm-wayland-session.1000.crash [08:44] works [08:44] fun [08:44] indeed :-/ [08:46] you mean the "connection refused" thing was from pkexec, not apport? the second command runs apport-gtk as user, so no surprise if that works [08:47] ah, crap [08:47] thanks for pointing that out [08:48] I'm trying to figure out how to get a backtrace [08:49] ah, apport has the dump [08:49] even if unpack hits [08:49] ValueError: ['UserGroups'] has no binary content [08:52] #0 0x00007f9f8a172c99 in _gtk_style_provider_private_get_settings ( [08:52] provider=0x0) at ././gtk/gtkstyleproviderprivate.c:123 [08:57] duflu, koza seb128 - otp still, will be a bit late / won't make it to BT meeting. Is there much to discuss today? [09:00] willcooke, running late too. Nothing I know of other than Pulse 11 is interesting [09:01] willcooke, duflu, koza, oh sorry, I'm in the middle of hacking and forgot about that one again [09:01] let's skip? [09:01] +1 [09:01] * duflu wonders if koza is in there alone [09:01] jibel, ? ^ [09:03] duflu, ah sorry, owm [09:04] duflu, or just skip :) [09:04] koza is not even in the room [09:05] let's skip then [09:05] duflu, pulseaudio 11 is for next cycle imho, I agree with what you wrote on the trello board [09:05] jibel sorry [09:06] And good news... bluez has not released recently [09:06] jibel seb128 duflu terribly sorry, I have been reading BT spec and did not realize it is 11 already [09:06] koza, nw duflu was wondering if you were in the meeting alone, and you were not :) [09:07] Best excuse ever for missing the BT meeting [09:07] :) [09:07] koza, we decided to skip, there is not much to share and everybody is busy [09:07] exactly, same situation on my side - not much to share, busy with other BT stuff [09:08] (everyone now wants to use Bluetooth, this is crazy) [09:08] I should try this Bluetooth thing some time [09:08] seb128, still should PPA sooner [09:34] Laney, your suggestion works fine, https://code.launchpad.net/~seb128/update-notifier/work-under-wayland/+merge/330271 [09:35] Laney, thanks again ;-) [09:35] now to debug the under pkexec segfault [09:37] seb128: didn't want to check errno? [09:37] I guess you just exit anyway so maybe you don't care [09:38] good stuff [09:43] Laney, right, I don't want to enter into the business of dealing with situations were things in a weird state [09:43] just exit for now [09:52] so it's basically "can't use gtk under sudo/pkexec" [09:52] if you do that on eog or gedit it tells you it can't open the display and bails out [09:52] it's just that python code/apport is handling errors less well [09:58] k, I'm offline for a bit for debuggign&lunch, bbl [11:10] man from Lenovo will be here in a mo to replace my keyboard \o/ [11:19] oki, bbl [12:09] yay, keyboard fixed! [12:10] willcooke, it was still the space bar not working on the right issue? [12:10] seb128, yeah [12:10] good that they fixed it :-) [12:11] The space bar is much more positive now, you can tell that you're pressing it [12:11] The engineer runs Ubuntu too :) [12:12] nice [12:13] did you get the spacebar heating option? ;) [12:13] willcooke: oh wow, they sent someone? I also had the extra 24/7 worldwide plan, but they just fedexed me a replacement [12:13] not that I'm complaining, it's easy enough to replace; I'm just surprised [12:13] jbicha: indeed - winter's coming! :-) [12:14] pitti, yeah, bloke came out and did it for me. It was quite involved on this X270, he had to take the main board out, and the battery, and the hdd [12:15] willcooke: oh, wow; took me 5 mins to replace on my x240, keyboard was pretty much the first thing that came off [12:15] maybe the design changed quite a bit [12:17] Ha, I offered him a cup of tea when he got here (natch) and the first thing he asked was "which model is it?" - and suggested that if it was a 240 then there wouldnt have been time to boil the kettle :) [12:18] hah - so, this is much more expensive for them, teaches them right to not make the design worse :) [12:20] seb128, didrocks: I got a fix [12:20] mutter's fault [12:20] andyrock, woot [12:20] well done [12:20] I'll propose a fix upstream [12:21] andyrock, send...excellent [12:21] it's 1 line [12:21] don't forget to reassign the bug in bugzilla [12:45] Ooh, suddenly a dropbox icon appears at the top of my 17.10 panel [12:45] but it's not clickable :( [12:46] (known bug?) [12:47] Trevinho: ^ [12:47] andyrock: that's nice! the env var trick worked I guess for debugging? [12:47] jbicha: btw, did you get the FFe lined up for g-c-c? [12:47] didrocks: yep [12:47] great ;) [12:50] didrocks: yes, use LP: #1712797 [12:50] Launchpad bug 1712797 in gnome-control-center (Ubuntu) "FFe: gnome-control-center 3.26" [Undecided,Confirmed] https://launchpad.net/bugs/1712797 [12:51] jbicha: excellent! I just have some debug stack to get in the UI, and we'll be good to go! [12:51] did you guys happen to fix LP: #1712104 with this patch set? [12:51] Launchpad bug 1712104 in gnome-control-center (Ubuntu) "Add Hot Corner switch to Dock Settings" [Wishlist,New] https://launchpad.net/bugs/1712104 [12:51] I don't know if we want to add this, do we? [12:51] right now, ubuntu session is "no hot corner", vanilla is "hot corner" [12:51] it's a popular feature for some people [12:52] but it's a good point, seb128, willcooke, Laney, what do you think about surfacing this option? ^ [12:52] (with off by default in the ubuntu session, not selectable in the vanilla one) [12:52] I think most people that have already been using GNOME have gotten used to the hot corner and it's frustrating if it's not there [12:52] why not selectable? [12:52] didrocks, I think off by default is the right way to go, I dont have any strong preference for making it an optional thing [12:52] jbicha: well, it will be upstream vanilla experience in the GNOMEsession, so on by default there [12:52] why is that a dock setting? [12:53] Laney: where did we say it's a dock settings? [12:53] oh [12:53] that's the bug [12:53] no, it would be in a different frame [12:53] same panel [12:53] or I could add it to the Tweaks app > Top Bar if you don't want it in g-c-c right now [12:53] (which shouldn't be entitled "Dock") [12:54] I woudn't do it as a distro patch, but that might just be me [12:54] I think a settings with off by default isn't that crazy, in the same panel [12:54] I think there are people who don't mind the Ubuntu session but like the hot corner and telling them to use dconf-editor or gsettings directly isn't great [12:54] tweaks would be ok [12:54] Laney: not a distro patch, but as we disable it by default… [12:54] and it's a separate panel, where we have our dock settings also [12:54] I think the argument applies to people who don't like it enabled by default and would want to disable it [12:54] the GSetting is upstream too [12:55] nope [12:55] the gsettings is not upstream yet :( [12:55] it's a patch made by upstream [12:55] but not in tree [12:55] sure [12:55] but if they take it then it makes sense to expose it [12:56] seb128: should I open a new bug for mutter [12:56] or can I just send the patch file in the nautilus bag [12:56] I cannot re-target it [12:57] however, it's some place where we reverted the upstream behavior due to our dock, I guess it's sensible to expose a way to put that gesture back in, but that's just IMHO [12:57] let's see what seb128 thinks about it as well once he's back [12:58] if upstream could commit that patch to gnome-shell, then it makes sense in Tweaks>Top Bar upstream too [13:13] andyrock, reassign the bug to the right component as I said earlier [13:14] andyrock, why not? [13:14] seb128: how can I reassign [13:14] didrocks, jbicha,no strong opinion, unsure how the hot corner fits with a dock, you are likely to trigger it by error when going over the most used/top icon [13:14] or the ui is really bad [13:15] or I'm not allowed [13:15] :D [13:15] andyrock, the product should be a combo [13:15] just select another value [13:15] on the top of the page, second item [13:15] or first rather [13:15] ok, looks like more for Tweaks thus [13:16] +1 [13:16] (can be conditional if the key is there) [13:16] let me record that on the bug report [13:16] we can always add it next cycle if the user feedback tells us it's a wanted option [13:17] I'm not allowed to do that https://usercontent.irccloud-cdn.com/file/bWbgePvT/Screenshot%20from%202017-09-06%2015-17-05.png [13:17] sounds good to me [13:17] andyrock, ok, let me reassign for you [13:19] seb128: it's mutter not gnome-shell [13:19] andyrock, I got an error with mutter so picked the other one [13:19] kk thanks [13:19] those are the same people looking at both [13:19] so should be fine [13:19] https://git.gnome.org/browse/gnome-tweak-tool/commit/?id=29073dd8a70e5 [13:19] jbicha, nice [13:20] that's on git master, I need to ask the translators if I can commit to the 3.26 branch [13:26] seb128, jbicha - I realised I never got round to testing the power button stuff, do I need to add some PPAs etc to see it? [13:26] I don't know, power/suspend handling is crap for me atm [13:27] I need to debug that but I'm waiting for 3.26 to land [13:27] also I had the suspend button extension which stopped working [13:27] willcooke: if you don't want to use dconf-editor or gsettings, you need gnome-control-center 3.25.92 which is in the GNOME3 Staging PPA right now [13:27] quite annoying :-/ [13:27] jbicha, I can dconf-editor, that's fine [13:29] gsettings set org.gnome.settings-daemon.plugins.power power-button-action 'interactive' [13:30] jbicha, works great! ship it! [13:30] lol [13:31] seb128, are you having issues with that setting then? ^ [13:31] Hey all, where does unity store the display settings? I can't get it to start proper, only clue is a dialog saying it can't restore display config [13:31] urgh, sounds familiar ^ ? [13:31] is this the monitors.xml thing again? [13:32] yeah, ditch (or move) the .config/monitors.xml [13:32] Also gnome-shell only starts when I have my ext monitor connected... [13:32] Saviq, you reckon it might be this? https://bugzilla.gnome.org/show_bug.cgi?id=786660 [13:32] Gnome bug 786660 in general "3.25.91: Unable to log in: invalid monitor configuration, Logical monitors not adjecent" [Critical,New] [13:33] Ok ditching monitors.xml helped with the dialog, but unity still doesn't start :'/ [13:33] No window decor, runs a few secs and exits [13:34] willcooke: jbicha: the settings "interactive" doesnt work for me [13:34] ah, you have to press for more than 0.5s [13:34] (or 1s) [13:35] "Fatal: Couldn't open display :0" [13:35] didrocks, that's hardware dependent I think. When does it generate the ACPI event [13:35] I don't remember to have that delay on unity [13:35] but I might be wrong :) [13:35] didrocks, on X220 - yeah, I does [13:35] jbicha: shouldn't that be our default for the ubuntu session? [13:35] lemme check [13:35] willcooke: if you have it on, yeah :) I normally press that button only when I'm upset, so not often (hem ;)) [13:36] But yeah, gnome-shell now starts, so monitors.xml helped here [13:36] didrocks: there are a few issues with it being default but since it's in Settings>Power, it's easy for users to opt in to [13:36] jbicha: do we have logs of those issues? [13:37] didrocks, confirmed, same on u7 [13:37] willcooke: ok, "good" then :) (and thanks for checking!) [13:37] willcooke: and you know what, my screen is still really readable and clean! :) [13:37] I don't like having the behavior of the Power Button be different if you log in to GNOME or log in to Ubuntu [13:37] Saviq, I think Trevinho is your best bet for help there [13:38] jbicha: that's not really an "issue" as per deficiency [13:38] what should happen if no one is logged in? [13:38] like, in gdm? I don't think the settings impacts gdm here, does it? (as it's per user) [13:38] it's possible to lose work if something accidentally hits the Power button and you don't notice for 60 seconds [13:39] shouldn't the behavior be the same though? [13:39] wasn't that the case in unity and installing ubuntu gnome? [13:39] you had different behaviors [13:39] and I don't think the multiple session thing is something people will switch back and forth [13:41] are you sure that Unity defaults to power off when power button is pressed? [13:41] willcooke, sorry, GNOME fail I don't get pings notifications :-/ no, no issue with that setting, just GNOME behaves crap regarding to power manager here, like it suspend when it should, can't configure the lid close behaviour, sometime doesn't suspend when it should [13:41] TBH, I think it will be more common to have people installing Tweaks and trying to enable/disable mod extensions like crazy, sounds something more important [13:41] jbicha: yes [13:41] ooh! I had that today. My laptop didn't suspend on lid close, just sat there getting warm [13:42] didrocks, jbicha , I didn't follow the whole discussion but unity default to display the dialog with all the action on power button event [13:42] didrocks: ok, I'd like the same settings for power button action for both GNOME and "Ubuntu" [13:43] jbicha: I don't think we want to change any upstream default in the vanilla session [13:43] that's the point of this session [13:45] hmm, maybe [13:51] Trevinho, you should maybe add X-Ubuntu-Use-Langpack: yes to debian/control for unity so it keeps getting its translations included in langpacks [13:52] seb128: oh, yeah... if I could just get it built in 17.10 after gcc change.... I'll be back on this soon, but having no failures in my archs makes things a bit more complicated [13:53] I might try with some qemu though [14:20] willcooke, those weird visual glitches don't happen only on unlock for me but also when starting e.g firefox [14:20] which opens maximized here [14:20] unsure but could be xwaylands apps only [14:20] I'm seeing it under X as well [14:21] it's pretty recent here I think [14:21] it was not bothering me before [14:21] I wonder if that's a gnome-shell 3.25 regression [14:21] I thought it might be to do with the launcher [14:21] or a combination [14:21] it could [14:21] so, that rather neatly raises this... [14:21] but it was not doing it with 3.24 I think [14:22] andyrock, While you're poking around inside mutter, could you see if you can work out what's going on there ^ [14:22] you can recreate the issue in question by: [14:22] maximizing a window [14:22] lock [14:22] unlock [14:22] see the window do a little wiggle [14:23] willcooke: i'll take a look [14:23] thanks! [14:33] kenvandine, hey, I tested eog snap yesterday and it shows the thumbnails for me [14:39] Trevinho: any idea about snapped remmina complaining that it can't open :0 display? I'm on wayland atm, so maybe something not working with XWayland? the app isn't shown in the gnome-shell launcher, so meh... [14:39] seb128, awesome, thanks [14:40] Saviq: mh, no I've not tried that... [14:40] Saviq: I mean, I did try that some time ago [14:40] seb128, one thing i'm noticing though, we don't get window icons for snaps [14:40] but... Not recently [14:40] if you uninstall the eog deb [14:40] then run the snap, you won't get the window icon [14:40] Saviq: it's using the snap preload though, so maybe I should fix it to use new technologies [14:40] i think we need snapd to copy out the icon like it does the desktop file [14:41] and we need to add whatever directory it's copied to for theme icons? [14:41] Saviq, Trevinho, https://github.com/snapcore/snapd/pull/3398 for the "snaps not listed under wayland" (which is different of your display issue) [14:41] seb128: ack, thanks [14:47] Trevinho: can confirm it's a wayland issue, the snap works fine under Xorg [14:49] Saviq: mh, i see... I'll give it a look in next days, but there shouldn't be much different between real x and actual xwayland [14:49]