=== thumper is now known as thumper-dogwalk === thumper-dogwalk is now known as thumper === JanC_ is now known as JanC [05:24] hi [05:26] * qengho snubs hikiko. [05:29] Ummkay [05:29] Good morning [05:31] Hey hikiko! [05:37] Hey duflu, RS [05:38] RAOF :p [05:38] How are you? [05:39] Please forgive me, hikiko! [05:39] Good morning, all. [05:39] So errm you guys come here often? [05:40] I'm pretty good. [05:40] Lol, what ? qengho ? Good morning to you too!:-p [05:41] hikiko: Please don't melt today. [05:41] Oh, I didn't scroll up enough I didn't see you snubbed [05:43] qengho it's not very warm today, atm we have around 27 [05:44] Wow, that's really good. [05:44] What's the weather in Taiwan ? [05:44] Mid 30s today. [05:45] And 125% humidity? [05:45] Heh, similar to here (except today) [05:46] lol raof that describes my city [05:46] I guess applies to Taiwan too [05:46] With wind, it might be tolerable. [05:47] There is no wind. [05:47] Good morning [05:47] o/ [05:47] qengho air conditioning ? [05:47] Plenty of that. I'm hiding in a cafe. [05:47] Haha :-) [05:49] I know how to say "please, another iced coffee" better than anything else. [05:49] How ? [05:50] * hikiko tests qengho's Chinese [05:51] Something like "qing, duole yige tang bing de latte". [05:52] Oh, and large. Add a "da" in there. [05:52] Or yibei. Yeah, that one. [05:55] I'm probably wrong, but they're too polite to laugh. [05:55] :-) [05:55] lol [05:55] Yeah that's a problem with the foreign languages [05:56] People are too polite to correct you [05:56] and you can't be sure if your sentences make sense [05:57] Can't tell :p [05:58] 30 degrees already \m/ [06:00] hikiko: well, was he right or are you too polite to correct him too? :) [06:04] jbicha: You know "it's all Greek to me"? For the Greek, they say it's Chinese to them. [06:05] good morning [06:13] hi didrocks. [06:15] hey qengho! [06:20] jbicha, I don't speak chinese :) [06:20] it's chinese to me :) [06:20] hikiko: oops, sorry about that [06:21] about what? [06:26] I'm embarassed about confusing Asian countries [06:26] lol, Greece is not in asia :) well, it's too close [06:27] good morning desktopers [06:27] hi seb128 :) [06:27] hey hikiko [06:27] bonjour didrocks et seb128 ! comment allez-vous ? [06:27] hey hikiko and qengho! [06:27] salut pitti, ça va bien ! et toi ? [06:27] bonjour pitti [06:27] looks like the systemd session is getting there ? [06:28] seb128: yes, it is; currently preparing some changes to upstart, and landing the graphical-session.target unit in systemd itself [06:28] seb128: and most other bits of systemd-graphical-session are in the landing CI now [06:28] hikiko: oh I didn't realize you were in Greece either [06:28] except xeyes of coures! [06:28] course [06:28] haha [06:29] going to miss those eyes staring at me at login ;-) [06:29] seb128: I think it's just unity-settings-daemon and unity-gtk-module that's left [06:29] well, landing that silo is still some work [06:30] we should drop update-notifier from it, not sure if that's possible to do or this needs a completely new landing [06:30] robru: ^ [06:30] ça va bien pitti, et toi ? [06:31] haha jbicha, no worries :) [06:31] didrocks: je vais bien, merci ! [06:31] pitti apprends le francais? [06:31] pitti, why is update-notifier in there? [06:31] hikiko: un petit peu [06:31] seb128: it shouldn't be [06:32] tu parles bien je crois [06:33] hikiko: je crois que seb128 pense différemment :) [06:35] qu'est-ce qu'il y connait seb128 au français ? :) [06:35] lol [06:35] il est un peu francophone peut etre... peut etre il n a pas oublier son francais [06:35] didrocks: parce qu'il est demi-allemand ? :-) [06:37] je ne savais pas qu'il est demi allemand... wow, so he speaks the 3 most dominent languages in eu... he must learn spanish too and then he can communicate everywhere... [06:37] no, he must first learn Dutch :) [06:37] * pitti hugs seb128 [06:37] pitti: exactement ! :-) [06:37] :s/everywhere/everywhere in eu [06:37] il habite aux Pays bas (I am sure this is not written like that but ok :P) [06:37] ? [06:38] hikiko: non non, très proche : "Pays-Bas" [06:38] :D [06:39] il y a 16 ans ou je n ai pas eu l occasion d ecrire une phrase en francais... :D [06:43] * seb128 hugs pitti & didrocks [06:47] c'est l'heure du petit déjeuner, brb [06:47] * didrocks hugs seb128 back, bon app! [06:47] 'ci :-) === davidcalle|afk is now known as davidcalle [07:47] morning all! [07:49] willcooke, hey! you are back ;-) [07:49] hi seb128, yeah afraid so [07:49] lol [07:49] had a nice break though, will share some pics later [07:49] we started wondering [07:50] we remember that you said you had thu/fri off [07:50] nice! [07:50] yeah, I forgot I'd booked Monday off [07:50] ah, k [07:50] logged in to my calendar on Monday morning, thought "oh" and promptly turned the computer off again and went out :) [07:51] lol, k [07:51] I had plans to drop you a Telegram, and then forgot, sorry about that [07:51] it was not in the shared calendar [07:51] figured you'd work it out [07:51] no worry [07:51] oh, you're right I thought I'd added it there. [07:51] naughty [07:52] hey willcooke! [07:52] morning didrocks [07:53] 'sup, willcooke. [07:53] afternoon qengho - how goes over there? Still good? [07:56] willcooke: yep. got to get back home soon. I leave it alone for just a few weeks and *the whole country* goes to poop. [07:56] * qengho sighs. [07:56] :D [08:03] morning! [08:04] hey Laney! [08:05] * seb128 is disturbed by the fluctuations in Laney's schedule recently, saying hi one minute earlier today than yesterday, what's going on! [08:05] Laney, how are you? had fun at climbing? [08:05] I hope you track my exit time as much as my entrance ¬_¬ [08:06] ya, I'm good [08:06] lol [08:06] we had a mosquito in the bedroom last night so there was some late night fun catching that [08:06] climbing was okay but it gets hot in there [08:06] sweaaaaaaaatyyyyyyyy [08:06] what about you? [08:06] hey Laney [08:07] I played tennis from 5pm to 6:30pm, that was good [08:07] hey pitti [08:07] how's it going? [08:07] otherwise a bit of work still after that, dinner and some tv, nothing fancy [08:07] got annoyed by a mosquito as well in the living room, but we didn't get it [08:08] at least it was not in the sleeping room so we didn't get bitten during the night! [08:08] morning Laney! [08:08] evolution works though [08:08] the sound of it moving around was VERY irritating [08:08] I hate when they pass by the ears [08:09] ola didrocks [08:09] ¿cómo estas? [08:09] (or something) [08:11] https://code.launchpad.net/~kaihengfeng/ubiquity/fix-usd-scaling/+merge/298284 [08:12] shrug [08:12] really? [08:13] lalala [08:13] /unsee [08:13] but *really* [08:13] * seb128 stares at cyphermox for approving [08:14] Laney: going well, thanks! landing more and more user session bits [08:14] nice work! [08:15] will you land -pre in another PR? [08:15] Laney: yes, I will; I added it to the Debian package for nwo [08:15] what happened with that PropagateDependencies thing? [08:15] so with upstart ubuntu28 and systemd 231-1 we have all the necessary infra in place [08:15] Laney: no reply yet [08:16] Laney: I'll bring that up at systemd.conf at the latest, I'll do a presentation about graphical systemd sessions there [08:16] Morning! [08:16] bongiorno Trevinho! [08:16] righto [08:16] hey Trevinho [08:17] Laney: so ATM all the hacks are in http://bazaar.launchpad.net/~ubuntu-desktop/gnome-session/ubuntu/view/head:/debian/data/run-systemd-session [08:17] Laney: so over time we can reduce this to become tasteful hopefully :) and then extend this to more sessions than just ubuntu and xubuntu [08:18] Laney: the rest looks fairly good to me now [08:18] wow, look at that broken indentation -- /me fixes [08:19] hey Trevinho, how are you? [08:20] seb128: hey seb, well thanks... You? [08:20] pitti: there's nothing in this world a bit of sleeping can't solve [08:21] :) [08:21] naps are good! [08:21] pitti, why do you set GNOME_DESKTOP_SESSION_ID= a second time there? [08:21] or unset it [08:21] Trevinho, I'm good thanks :-) [08:22] seb128: I want to unset it after the session ends [08:22] seb128: ... but I'm doing that wrong indeed, this needs to move down; thanks for spotting! [08:22] pitti, yw! [08:22] seb128: pushed [08:23] k, makes more sense now [08:23] thanks! [08:23] I get the --quiet is-active call is cheap but that's happening while the session is open right? [08:23] wouldn't it make more sense to have at least 10s there? [08:23] or is that going to delay closing/shutdown by as much? [08:24] seb128: yes, it will [08:24] logout, not shutdown [08:24] I see [08:25] seb128: that's the bit which I want to fix first, it's really ugly [08:25] we should probably have a bug targetted to yakkety to fix that one [08:25] right [08:25] seb128: I'll add a WI for, say, 16.09 [08:25] thanks [08:25] (holidays and too much to do in August) [08:25] I think it's fine for the unstable cycle [08:25] that one could be something like watching the logind session on dbus like for g-k [08:25] probably? [08:25] but I wouldn't like the idea to have a such hack in a release [08:26] writing a small downstream helper which listens to the signal is fairly simple, if we don't get it into systemctl proper in time [08:26] the logind session can get you stuck though no? [08:26] like if there are processes in its session not properly closing/exiting [08:27] the state changes to 'closing' [08:27] I sometime end up with indicator-session showing an user still logged in [08:27] k [08:27] seb128: we just need to listen to systemd's signal that the thingy unit we start goes to "stopped" [08:27] that's how we know when to kill gnome-keyring-daemon [08:27] or you can watch some unit on systemd's api [08:28] conceptually we just want systemctl start --wait-until-stopped ${DESKTOP_SESSION}-session.target [08:30] morning all [08:31] hi andrea [08:31] hikiko: hey, did you see https://code.launchpad.net/~3v1n0/unity/shadows-on-existing-pixmaps-cleanup/+merge/301099 ? [08:31] hey andyrock, how are you today? [08:33] could not sleep that well [08:33] you seb128 ? [08:33] I'm good thanks [08:33] andyrock: too much excitement in yesteday's ride? [08:33] oh, too warm in Italy? [08:34] he lives in the top of an hill, it can't be hot :-D [08:34] haha [08:34] ahaha too many thoughts [08:34] (very nice place, though... with sea view) [08:34] yes Trevinho I think there are corner cases though if we move the func to the DecorationsManager [08:35] what you mean? [08:35] Trevinho, I want to check something - reviewing it [08:38] vabbè guardo i singoli commit [08:39] hikiko: ok... I've just moved that to the window, since it builds textures for a window. Plus i've removed some duplicated code and usde better C++ features [08:40] yeah, I saw you renamed some variables etc, thanks [08:45] Trevinho: Am I imagining things or has widget scaling got better in 16.10? 16.04 and below is annoying to configure at 1080p but 16.10 looks better proportioned [08:46] I mean 16.10 deals better with non-integer scaling values [08:48] andyrock: You're back? [08:49] hey duflu [08:49] andyrock: You were away from Ubuntu, no? [08:49] duflu: jejej nope :D [08:50] Oh. I don't pay enough attention [08:50] pitti: what was the problem with talking to the user systemd on dbus? [08:50] Laney: "the problem"? [08:50] the one tedg and I had [08:50] duflu: btw gtk scaling is still 1 or 2 [08:50] laney@nightingale> busctl --user introspect org.freedesktop.systemd1 /org/freedesktop/systemd1 ~ [08:51] Failed to introspect object /org/freedesktop/systemd1 of service org.freedesktop.systemd1: Process org.freedesktop.systemd1 exited with status 1 [08:51] andyrock: Yeah, but something looks better in yakkety. Maybe it's the Ubuntu titlebars only [08:51] Laney: ah, so it wasn't systemd-shim and it wasn't cgmanager; it seems I don't remember any more, or we never actually demystified it [08:51] I forgot if we did [08:51] I actually made some experiments last w.e. to try to make it floating [08:53] andyrock, don't waste effort on that, I think upstream made an explicit choice by making it an int and wouldn't take a patch to change that [08:54] i'm actually using windows 10 on my spare time because of that XD [08:54] duflu: hey, well... in gtk? [08:54] otp.. [08:54] dell xps 15 just sucks with scaling 2 [08:55] andyrock, :-( [08:55] 1.75 would be perfect [08:56] * Sweet5hark1 find awesome work by ochosi on g+. [08:56] ochosi: awesome, thanks that helps a lot! [08:56] good morning Sweet5hark1 [08:57] good morn'. [09:03] Sorry was on the phone Trevinho [09:03] andyrock: I've found GTK's integer staling > 1 only really works on 3K or 4K displays. 1080p is really hard but it's better in yakkety [09:04] I guess because the widgets only have integer scaling factors, everything only looks right on integers [09:04] * Trevinho won't ever forgive gtk for not having supported floating scaling! [09:05] Yeah GTK is now a little too old-fashioned [09:05] duflu: i've a 4k display and scaling 2 does not feel ok [09:06] :( [09:06] andyrock: Yeah same problem. I guess at 15" a scale of 2.0 is too big [09:06] yeah so would actually be not that difficult to support floating scaling [09:07] I believe GTK still has some bitmap-based widgets that need integer scaling..? [09:07] Trevinho told me that is because of pixel alignment [09:07] That's what I mean [09:08] but on high-dpi 1px less 1 px more is not visible at all [09:08] yeah... rounding wouldn't have been a problem.... Really not a problem. [09:08] nobody ever noticed the rounding we do in unity7... [09:08] And indeed we still have integer-sized windows (of course) [09:08] I think you're right... we'd be better off forcing GTK to use floats than dealing with just ints [09:08] which... we might try to do [09:09] The loss of quality isn't as bad as the gain in readability [09:09] Maybe [09:13] pitti: I think it probably only works under dbus user sessions [09:13] Laney: oh! is that what you didn't have installed? very plausibly [09:14] Laney: otherwise systemd --user uses its internal dbus socket (/run/user/1000/systemd/private) [09:14] pitti: not on my outside system [10:55] pitti: https://paste.debian.net/785326 <- rough first draft [10:59] Laney: nice! [11:00] I'd still like to get this into systemctl proper to avoid the python overhead and the dependencies, but this is better than the polling loop already [11:01] right, it should be, but this is better for now imho [11:01] so do what you will with it :-) [11:01] * Laney goes away into juju land for a while [11:02] https://yy2.staticflickr.com/5307/5629329064_28902fc8e3_z.jpg&f=1 [11:05] jibel, FYI "initctl start indicator-network" in the terminal app gets me a properly working indicator in u8 [11:07] oh ! and it even shows my existing vpn setup to choose from [11:31] Laney, your python dbus foo is strong, you should rewrite that ubiquity hack with proper code :-) [11:40] I should add a review comment at the minimum [11:42] yes please [11:46] seb128: the dbus-monitor polling one? omg yes :) [11:46] pitti, yeah :-/ [11:47] also, it's "fu" :) [11:47] oh, right ... thanks! [11:53] done [11:54] I used *DISAPPROVE* [11:54] felt better than Needs Fixing [11:55] ogra_, on yakkety? [11:55] ogra_, how do you install the terminal app? [11:55] jibel, xenial with overlay ... but i guess it works on yakkety too [11:55] jibel, there is a link in mhalls blog post [11:56] sadly AltGr still doesnt work in unity8 apps for me :( [11:57] (it works in libertine apps ... but there i have no keyboard repeat at all ... which is super annoying ) [11:58] ogra_, the fix for that is in the next Xmir upload [11:58] yay [11:58] key repeat, I mean [11:58] yeah, thats what i understood [11:58] AltGr might be in Mir 0.24 maybe [11:59] maybe [11:59] bregma, btw, my high load seems to have been eds running in the libertine container ... battling with the natively running one ... [11:59] my fan is silent when i dont run evo [11:59] ogra_, thanks the indicator starts with initctl [12:00] seems to be some race on session startup [12:00] ogra_, I can now upload crashes without switching to unity7 ;) [12:00] lol [12:00] well, switching back to u7 i usually spend 10mins clicking away apport windows [12:00] * ogra_ blames pitti [12:00] :P [12:01] they're like cockroaches [12:01] yeah [12:01] and every time i think i'm done a new one pops up [12:01] *just* like cockroaches [12:01] :) [12:02] so now having the music app behave would be awesome [12:02] it starts but freaks out when you actually want to play someting [12:03] and even more funny ... *everything* i click in the indicator that opens a remote app (system settings, clock etc) actually opens the music app by default if i have it installed [12:04] aha ... uninstalling and re-installing it seems to have fixed that last bit [12:06] hmm, there must be a way to make libertine use a proper cursor [12:07] ogra_, yes, but it's still a work in progress [12:08] ha, got it ... [12:09] just installing dmz-cursor-theme along with light-themes and ubuntu-settings gets me the full default theme support [12:09] that looks a lot better :) [12:17] hmm, funny, installing lautfm-player gets me proper playback ... seems to be an issue in the music app itself [12:19] bregma, I cannot start libertine on yakkety, it crahes with bug 1606556 [12:19] bug 1606556 in libertine (Ubuntu) "/usr/bin/libertine:11:QListData::size:QList:ContainerAppsList::size:QQmlAdaptorModel::count:QQmlDelegateModel::_q_modelReset" [Undecided,New] https://launchpad.net/bugs/1606556 [12:21] seb128: hey, ping for n-m/1.2.2, ;-) [12:22] happyaron, hey, right I know it's on my list, just got busy and previous week was .1 etc, trying to get to it this afternoon [12:23] ty [13:10] tedg: hey, could you please merge again https://code.launchpad.net/~ted/unity/systemd-unit with andyrock's branch? [13:10] (not sure they cleanly merge right now) [13:10] Trevinho: Ah, yeah, I can. [13:12] tedg: ta [13:15] tedg: also, I was thinking... using that code in the current yakkety causes the Start/Stop calls to not happen since the proxy is not connected... They get queued though... Maybe you could avoid to call start/stop if the systemd_proxy_ is not connected? [13:16] tedg: at the beginning I had the idea of just provinding one of the two wrappers, depending on what is running in the session, but... It's a little to annoying, and I didn't want you to create an abstract class just for that, but... Maybe... [13:16] Trevinho: Oh, because it's connecting to the session bus instead of the user bus right now? [13:17] tedg: yes [13:17] Trevinho: We won't have the Upstart stuff for long hopefully, so I don't think it's worth abstracting. [13:18] tedg: yeah, same I thought... [13:18] tedg: but currently guess we'd get some "Timed out waiting for proxy" erros in unity logs [13:18] so... [13:19] tedg: is an env var available when systemd is there? [13:19] tedg: I think checking if it's connected is fine. We can drop that too once we get migrated over if it's taking too long. [13:20] ok... [13:20] tedg: I mean, it might be not connected at the very start of the session... Like if starting with unity already locked (because of a crash) though [13:20] but... Yeah, as you say, this is something we can get rid of [13:21] or, juts check for an env var presency for both upstart and systemd wrappers [13:23] tedg: also /usr/share/unity/unity-compiz-profile I'd prefer it to be in /usr/lib/unity [13:24] tedg: and probably renamed unity-compiz-profile-set or -selector or... anything that explains better :-) [13:25] K, I was avoiding changing the codebase too much, but since andyrock moved the .conf file into the build system we can move it to lib. [13:33] tedg: yeah, that's nicer [13:43] Laney: seb128: do you want to file a merge proposal for ubiquity, since you know how to "do it the right way" [13:44] cyphermox, hey, going to let Laney comment on that, I don't know offhand but shouldn't be hard to write, I was just reacting to calling cmds and parsin std output this way [13:44] Not sure why the scare quotes, but I don't particularly have time right now, sorry [13:45] I would ask the contributor, but maybe they won't do it now it got merged [13:45] cyphermox, I guess we can still revert and ask the submitter to do it properly [13:46] seb128: I think we'll want to do the change and do another upload [13:47] need to find a "we" then [13:54] how can git rebase fail on a commit in the upstream branch? [13:55] huh, weird [13:55] Laney: wait, gbp? [13:56] real git [13:56] Laney: not sure what you mean, but last time I fell into trouble was when I rebased master and scribbled over/deleted the merge commit from upstream [13:56] no merge commits here [13:56] I think there was a force push though [13:56] * Laney spanks ximion [13:57] ah, or that [13:57] pitti: it was ; ; merge failure in one of those [13:57] I don't see how that can happen in normal circumstances [13:57] git log master..HEAD revealed the problem though [13:58] -> it showed one of the changes that should have been on master [13:59] now to try and repair this without brute force [13:59] * Laney tries git rebase -i; delete force pushed commit [14:01] pitti: if you're on debian-devel, would be nice to reply to http://article.gmane.org/gmane.linux.debian.devel.general/210066 [14:02] Laney: I'm not, but thanks for the pointer; I'll read it RSN [14:03] Laney: https://lists.debian.org/debian-devel/2016/07/msg00484.html has the Message-Id: for replying [14:04] Laney: what? where? [14:04] pitti: nod (I used http://mid.gmane.org/ as a lazy way to get a link) :) [14:07] ximion: my branch has "e50f836d47ecf09ffbd3fa672542685a39461d0e Use appender more often and apply some style fixes" but master has "c638e73255030b047db317799187a15698d0dad0 Use appender more often and apply some style fixes" [14:07] ximion: it could be that I messed it up by mistake though [14:09] doesn't look like your fault, actually... [14:10] something went wrong there [14:11] without looking I would guess commit --amend and push --force [14:13] probably, but I don't remember conciously force-pushing anything during Debconf [14:13] anyway, the branch wasn't protected yet (= force-pushing was possible) [14:13] * ximion fixed that a few minutes ago [14:13] those IPAs were quite strong [14:15] Indian Pale Ales? [14:17] yummy [14:18] the easiest way to fix this would propapbly be to export the patches and import them on a clean branch [14:18] no way, git is powerful enough [14:18] looks like I messed something up there in the past, sorry for that [14:18] neat :) [14:18] I did an interactive rebase and told it to delete the conflicting patch [14:19] then there is a real conflict in getIndexFile [14:19] "Successfully rebased and updated refs/heads/online." [14:23]