=== JanC_ is now known as JanC [07:02] good morning [07:04] Morning didrocks [07:06] hey duflu [07:29] good morning desktoppers [07:29] Hi oSoMoN [07:32] good morning everyone [07:33] salut oSoMoN, hey ricotz [07:35] Hi ricotz [08:08] gooood morning desktoppers and happy friday! [08:10] hey seb128, happy Friday [08:12] hey didrocks :-) [08:25] marcustomlinson, ping, please see pm [08:25] hey seb128 didrocks duflu [08:26] hey ricotz, how are you? [08:26] ricotz, do you need Marcus for a libreoffice issue? can someone else maybe help if he's not around? [08:26] seb128, hi, I am fine, how are you? [08:27] I'm good, it's fosdem's friday! [08:27] seb128, he pushed a build to deal for diskspace issue, which seems insufficient [08:27] Morning seb128. Happy... Brussels? [08:27] seb128, I have another solution, hopefully better [08:28] seb128, in progress here https://launchpad.net/~libreoffice/+archive/ubuntu/libreoffice-prereleases/+packages?field.name_filter=&field.status_filter=published&field.series_filter=focal [08:28] hey duflu, how is your friday going? getting some success of your gjs improvements? ready for the w.e? [08:28] seb128, oh fosdem :) [08:28] seb128, Friday is going well but today upstream has been silent. So that's at least not bad news. [08:28] ricotz, ah, good ... I've not followed the details, but why are builds hitting enospace now? toolchain changes? [08:29] seb128, likely, that is a downside of merging the l10n build [08:29] https://git.launchpad.net/~libreoffice/ubuntu/+source/libreoffice/log/?h=wip/ricotz-focal-6.4 [08:31] ricotz, Bjoern is still around or did you resurect some old patch? ;) [08:31] seb128, cherry-picked it [08:31] marcustomlinson, ping [08:31] I see, cool [08:31] ricotz, it's a bit early for him I think [08:32] or rather I think he's around earlier but then step out and come back a bit later [08:32] seb128, https://launchpad.net/~marcustomlinson/+archive/ubuntu/libreoffice/+packages?field.name_filter=&field.status_filter=&field.series_filter=focal [08:36] duflu, I wonder if we regressed grub, it's not supposed to apply the purple screen/do anything if it's not displaying the menu [08:37] duflu, by 'ACPI BGRT changes in the kernel' you mean FRAMEBUFFER_CONSOLE_DEFERRED_TAKEOVER=y ? [08:37] seb128, I've never seen it not apply the purple screen. Assuming it's the only thing that ever displays a blank purple screen [08:37] Yes I mentioned that in the same bug :) [08:37] it's not applying the purple bg on my xps [08:37] well, assuming you use the spinner theme? [08:38] seb128, I mean without the flickerfreeboot package installed. Sorry I used the same tag just because it is related [08:38] So no, I do not mean the spinner theme [08:38] ah ok [08:38] that makes more sense now [08:39] Just a plain focal install, and one reboot [08:39] btw we decided to switch to spinner by default for focal, at least to evaluate it [08:39] That sounds good and would at least solve that new bug :) [08:39] it's not fully flicker free but we still believe it's an improvement [08:39] :) [08:40] seb128, however it is telling grub to not display a splash screen should probably be done independently in grub though. So the bug does not return in the absence of the plymouth theme [08:41] yes, well that's the intend, unsure what's the issue there [08:41] seb128, completely separately I think Ubuntu needs a little logo on top of the BGRT [08:42] otherwise it looks like a hung BIOS [08:42] yes [08:42] x_nox was suggesting added the 'ubuntu' logo at the same position than in gdm [08:42] adding [08:43] bottom of the screen centered [08:43] I can't remember but we can debate the perfect position once something is visible [08:43] right [08:44] It's a slow graphics path though. So we either want zero animation or a small one [08:44] Actually, zero. Because it's during kernel boot [08:44] Animation is probably not possible at that time [08:45] duflu, the grub patch we are using that I though should avoid the colored background is the most recent commit on https://github.com/jwrdegoede/grub2/commits/for-upstream [08:45] anyway, bugs that we should probalby polish before the LTS [08:46] 👍 [08:49] duflu, I created https://trello.com/c/aa3X88QY/118-plymouth-boot-improvements [08:50] seb128, maybe deduplicate with https://trello.com/c/FiGczlI0 [08:51] duflu, fair point, let me add the reference/checklist to that one [08:52] duflu, k, updated [09:11] morning desktoppers [09:11] Morning marcustomlinson [09:12] question (to whoever sees it first): can a package in main recommend a package in universe? Or is that still too strict a dependancy? [09:12] hey marcustomlinson, how are you? [09:13] marcustomlinson, you should lower to a suggest iirc [09:14] Yeah I guess, because recommendations get automatically installed [09:14] ? [09:14] I'm doing ok seb128 thanks. Still recovering, but at least it's friday [09:15] duflu, right, they don't fail installation if they are missing but that would give inconsistent behaviour depending of your activated sources [09:17] hmm, so if the package doesn't have to be installed but does help significantly, it can be from universe? [09:18] to be clear here, I'm wondering if LibreOffice can recommend fonts from universe [09:19] marcustomlinson, no [09:19] main package need to have their recommends in main [09:19] so either you need to MIR those or to lower the recommends to a suggests [09:20] ok that's a clearer answer :) [09:20] thanks seb128 [09:20] np [09:21] if the fonts are specifics to a locale you might be able to make the languages-selector support pull them in when enabling the said language [09:31] ricotz: I was working outside of IRC hours ;) [15:21] good morning desktopers [15:22] hey hellsworth [15:22] o/ marcustomlinson ! [17:07] who knows mesa? I'm onto a new crash when using Indirect GLX, which seems specific to the client application I'm running - glxgears runs fine without causing this issue, as does supertuxkart (albeit with graphical abberations in STK's case because it expects newer opengl): https://paste.ubuntu.com/p/cqrqq4FZWD/ [17:09] the application code I'm experimenting with can be gathered from https://github.com/irixxxx/igl.git [17:09] specifically any of the examples [17:11] make sure you don't have anything important open/unsaved if you attempt this, because it WILL kill your Xorg :-p (you also need Xorg compiled from the debdiffs in LP #1776447 [17:11] Launchpad bug 1776447 in xorg-server (Ubuntu) "Xorg's Indirect GLX broken from upstream regression" [Medium,Triaged] https://launchpad.net/bugs/1776447 [17:11] this doesn't appear to be a regression from those debdiffs, just an unmasking [18:01] try #dri-devel [18:05] thanks :-) [20:01] diddledan: good news, soon we'll have gtk2-common-themes for arm64 and armhf :) [20:02] diddledan: and i'm submitting a PR that fixes the gtk theme if you set it to use the system theme [20:02] gimp that is [20:02] \o/ [20:18] tjaalton, it might be a stack overflow somehow runaway recursion [20:49] diddledan: getting a weird issue trying to build gimp [20:50] Issues while validating properties: Specified icon 'gimp.png' does not exist [20:50] that line in the yaml hasn't changed in nearly 2 years [20:50] is the icon there? [20:51] it's in the top level dir [20:51] should be in the root of the repo, yeah [20:52] hmm [20:52] a second run and it is working [20:55] odd [21:09] diddledan: https://github.com/snapcrafters/gimp/pull/97 [21:09] snapcrafters issue (Pull request) 97 in gimp "Set necessary env for system gtk themes to work and clean up content interfaces" [Open] [21:09] diddledan: this does revert one of your commits, but we should have gtk2-common-themes for all supported arches by early next week [21:10] we can wait for that before merging this, but comments would be appreciated [21:39] diddledan: once this PR is merged, we'll get all the arches https://gitlab.gnome.org/Community/Ubuntu/gtk2-common-themes/merge_requests/1 [21:39] Ubuntu issue (Merge request) 1 in gtk2-common-themes "Updated to build with bionic" [Opened] [21:57] diddledan: what's gtk3 in gimp? [21:58] nothing directly, but there might be plugins that want to use GTK3. also if I were to add darktable that definitely is GTK3 [21:59] gtk2 and gtk3 use different env variables [21:59] so shouldn't clobber each other [21:59] I could only find GTK_PATH applied to both [21:59] how can i verify that i didn't break gtk3 in gimp? [21:59] GTK_EXE_PREFIX allows both to coexist without setting GTK_PATH [22:01] GTK_EXE_PREFIX simply requires that gtk-2.0 and gtk-3.0 exist at /usr/lib instead of /usr/lib/$SNAPCRAFT_ARCH_TRIPLET [22:01] ok, i'll look at this some more later on [22:02] at least glimpse-editor looks much better now :) [22:02] https://usercontent.irccloud-cdn.com/file/bvJt1cPG/Screenshot%20from%202020-01-31%2013-23-37.png [22:02] lines 167 thru 171 do the moving about - as we're now using gtk-engines we might need a bit more than a simple symlink - maybe a dir with symlinks inside [22:03] nice :-) [22:03] i look more carefully at gimp and make sure my fix doesn't break gtk3 [22:03] none of the gtk3 stuff is in the glimpse snap, so simpler case [22:03] aye [22:04] gimp does looks great with my branch though :) [22:04] but i probably busted something else [22:04] I'm not currently using gtk3 but I was mulling adding darktable into it because people keep moaning they can't use darktable and gimp snap together :-) [22:10] diddledan: actually, if we move the GTK_PATH to environment under the app it wouldn't effect darktable [22:10] right now it snap global [22:11] but if we move it to the apps section, gimp and darktable could have separate env [22:11] good point [22:12] although, I don't know if they can launch each other in addition to talking to eachother?