[00:06] no one? [00:22] ShinyShoes: Well, for a start, this is not a support channel. Try #ubuntu. [05:50] hi [08:46] morning [08:46] happyaron, yo! [08:47] hey willcooke [08:48] hey seb128 - how goes? [08:49] desktoppers - there is an "all hands" meeting this morning (and again later for those in the US etc) [08:50] doing ok, woke up at 6am and the cold was annoying enough after being lying down for hours that I couldn't get back to sleep so I started the computer at 6:30, not quite p_itti early but early enough for my taste [08:50] willcooke, how are you doing? [08:50] seb128, much the same actually :) [08:51] except I didnt sit at the computer [08:51] Morning seb128 willcooke hikiko [08:51] hey flexiondotorg [08:51] willcooke, wise :-) [08:52] well, on the good sign it means I can call it a day a bit earlier this afternoon or maybe take a nap ;-) [08:52] sign->side [08:53] flexiondotorg, are you still looking at bug #1623856? or are you blocked/needing help on that? [08:53] bug 1623856 in update-manager (Ubuntu) "Updated gtk element means you need to specify a size on the details drop down or it is only 1 line high" [Low,Triaged] https://launchpad.net/bugs/1623856 [08:53] nap! [08:53] :-) [08:53] hey flexiondotorg [08:53] oh, that reminds me - seb128 we still need to work out what to do with update manager in X [08:54] seb128, Still on my list. Assistance welcome. Was going to look at it today to take a break from glaring at CSS. [08:54] good plan flexiondotorg [08:54] willcooke, ok, right, thanks for the reminder [08:55] flexiondotorg, k, well have a look today and maybe bring on the channel for discussion if you get stucked? [08:55] seb128, I'll ping you a bit later. [08:55] Currently listening to the catchy music ;-) [08:56] :-) [08:56] morning flexiondotorg willcooke seb128 [08:56] and happyaron [08:56] and desktopers [08:56] hey hikiko, had a good productive morning? ;-) [08:56] hey hikiko - congrats on fixing the laptop [08:57] thanks :) well my boyfriend did the dirty job I was just reading him the assembly instructions :P [08:57] We call that delegation [08:57] ;) [08:57] hahahah [08:57] :) [08:58] fix by proxy :p [09:04] hikiko: morning, :) [09:04] and congrats! [09:04] hey happyaron, how are you? [09:05] seb128: great, you? [09:05] I'm good thanks! [09:06] morning all [09:06] hi! [09:06] hey andyrock Laney [09:06] * Laney stabs dnsmasq [09:06] how are you? [09:07] hey seb128 [09:07] doing good! [09:07] hey seb128 [09:07] it was a bit frosty this morning [09:07] brrr [09:07] you? [09:07] hi andyrock too! [09:07] super well, yesterday was moving day [09:08] hey Laney [09:08] Laney, cold woke me up at 6am so I started the day early, doing good otherwise [09:08] andyrock, where did you move? [09:08] Laney, listening to the townhall ;-) [09:08] different flat [09:08] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=834722 [09:08] Debian bug 834722 in dnsmasq-base "dnsmasq-base: dnsmasq does not forward queries after recieving servers via DBus second time" [Normal,Open] [09:09] the townhall?????????? [09:09] the WHAT? [09:09] Barcelona city center, small but cozy [09:09] * Laney missed the memo [09:09] hi andyrock Laney [09:09] hey hikiko [09:09] oh I just didn't read the memo :S [09:09] Laney, ready allhands or warthogs lists! [09:09] read [09:09] lol [09:11] hey Laney andyrock [09:11] andyrock, back in Barcelona this year then? [09:11] yep [09:13] nice [09:14] hey happyaron [09:14] how are you? [09:14] great! [09:23] * Laney builds that dnsmasq patch [09:23] maybe my woes will end [09:29] seb128: i wrote pitti an email [09:29] andyrock, thanks [09:30] one more systemd user session regression, the recent section not working in nautilus [09:30] andyrock, that one is not for you though ;-) [09:32] i'm woderning why gnome-shell session has XDG_SESSION_ID exposed [09:33] seb128: is gnome-shell started using systemd? [09:33] because I can't find its service file [09:34] dunno [09:34] no [09:34] we were breaking new ground by doing this mostly [09:35] ah btw slow internet here http://www.speedtest.net/my-result/5764747699 [09:36] you should complain [09:37] ah ok so that's why it works for them XD [09:56] quick trip to the post office, back shortly [09:59] seb128, Trevinho https://freedesktop.org/wiki/Software/systemd/logind/ [09:59] under User Object [09:59] "Display encodes which graphical session should be used as primary UI display for the use. It is a structure encoding session ID and object path of the session to use." [09:59] we can actually use this to get the session_id [10:03] mh [10:03] good morning Trevinho! [10:03] heya seb128 [10:36] andyrock: That looks right to me [10:36] it even gives you the object path directly [11:31] Laney: I push the patch to https://launchpad.net/~fjkong/gedit/ubuntu could you help me to review it? thanks [11:34] FJKong_: Cool, thanks! I will try to soon - could you make a merge proposal please? [11:35] Laney: open a bug? [11:35] no [11:35] on that link, click "Propose for merging" and fill it out [11:35] oh i see [11:35] you should target it to lp:~ubuntu-desktop/gedit/ubuntu and not lp:gedit [11:37] I am not sure if I have permission to push there [11:37] you don't need it to make a proposal [11:42] Laney: done [11:43] FJKong_: ok, nice, will check it later on [11:43] o/ [11:52] Hello, seb128. Last night (my time), I had a discussion with attente about a global menu issue I'm facing. He (?) told me to talk to you today. [11:54] I've been running apps in containers like the ones made by Libertine for some time now. As long as the containers can access the host's DBus socket, global menu works fine for gtk2 apps, but gtk3 apps continue to have a dedicated menu bar. [11:57] To make the matter more complicated, some gtk3 apps does do the global menu fine. By the suggesstions of attente, I installed dconf-editor and nautilus. Those apps' seem to integrate perfectly with global menu. Other apps, that I want to run in containers, like Firefox and Eclipse, don't do the global menu. === hikiko is now known as hikiko|ln [12:11] Guest1796: hey, i think i figured it out. you also need libdbusmenu-gtk3-4 [12:11] (for firefox at least) [12:22] hey attente & Guest1796 [12:23] good if attente figured it out ;-) [12:23] attente: Thanks, I'll try it out right now. [12:23] How did You figure it out. [12:23] hi seb128 Guest1796 [12:24] attente knows all [12:24] Guest1796: ^ it's libdbusmenu-gtk3-4 [12:24] attente: Yea, I'm asking how did you manage to figure it out. [12:25] oh. i was in the bathroom thinking... [12:25] ... [12:25] Shower thoughts, got it. [12:25] haha [12:25] It's the colour changing lights [12:26] I was going to say [12:26] how is the tennis elbow going btw? ;-) [12:26] attente, thanks for the content-hub-glib branch, however there's a merge conflict [12:26] terrible! i don't even play tennis! [12:26] * willcooke . o O ( minds in the gutter you lot ) [12:26] kenvandine: ok, i'll refresh it [12:26] :-/ [12:26] attente, thx! [12:26] hey kenvandine, you are up early! [12:27] seb128, been working a couple hours already :) [12:27] urg [12:27] i start early on wednesday [12:27] so i can lead a coding club at the school in the afternoon :) [12:27] it's worth it :) [12:28] That's cool. What languages do you teach? [12:28] nice [12:28] we're working on python right now [12:28] 3? [12:28] moving to writing minigames for minecraft in javascript soon [12:30] Speaking of languages, I want to learn C, but any guides I find don't mention which version of C standard it's about. :/ [12:31] attente: It doesn't work for Eclipse. I'll try Firefox, but I'm connected to IRC using ChatZilla, so will leave momentarily. [12:31] Guest1796: ok. i only tried it with firefox. what version of eclipse are you running? [12:31] attente: Neon. [12:32] Guest1796: is that installed from the archive or from a download online or ...? [12:32] Download. I don't know of any archives shipping latest Eclipse. [12:33] attente: On host, global menu works, if that's your doubt. [12:33] Restarting Firefox now, BRB. [12:34] attente, how far along is your gtk branch that uses the pasteboard? [12:34] kenvandine: it works, i'm hoping to merge it as soon as the content-hub-glib is merged [12:34] woot! [12:35] woot indeed :) [12:35] :D [12:35] that's very exciting [12:35] just let me build it here and make sure the refresh works too === ubuntu is now known as Guest76787 [12:38] attente: It worked for Firefox, thank you! [12:39] Guest76787: \o/ [12:39] And about Eclipse, I remember it working fine on host, but I could be wrong. [12:39] i'll try it in a bit [12:39] I'll install Java on host as well, to try it. === hikiko|ln is now known as hikiko [12:55] kenvandine: ok, i think the content-hub-glib branch should be good now [12:59] attente, thx [13:00] attente: Global menu working perfectly fine in Eclipse on host. [13:01] attente: The UI looks quite different from running it in container, in fact. [13:02] Guest76787: ok. which jre are you using? [13:02] attente: 8. [13:03] I'll upload some screenshots, give me a moment. Is imgur OK? [13:04] Guest76787: sure [13:07] attente: http://imgur.com/a/PXRIi [13:12] * flexiondotorg lunches [13:16] that's pretty disconcerting... [13:20] attente: I'm thinking there's some specific package for integrating Java SWT into GTK. [13:20] attente: Or maybe not, why'd it be installed in a vanilla desktop install. [13:26] there used to be one but that stopped working iirc [13:27] http://www.webupd8.org/2015/10/ubuntu-1510-and-1504-update-disables.html [13:28] https://launchpad.net/ubuntu/+source/jayatana/2.7-0ubuntu4 [13:28] seb128: It now works without it. [13:28] weird [13:28] Look at the screenshot. It's working on host, but not on container. [13:29] i'm pretty stumped tbh [13:31] ... The pleasures of developing an extensive and modular system... [13:42] Host Eclipse logs: http://paste.ubuntu.com/23416155/ [13:43] Container Eclipse logs: http://paste.ubuntu.com/23416160/ [13:48] * flexiondotorg returns [13:53] attente, seb128, look at the logs if it helps. Personally, I can't find anything of note there. [14:18] seb128, I've looking at LP: #1623856 [14:19] Launchpad bug 1623856 in update-manager (Ubuntu) "Updated gtk element means you need to specify a size on the details drop down or it is only 1 line high" [Low,Triaged] https://launchpad.net/bugs/1623856 [14:19] If you're up for sharing some insight, I'm game :-) [14:20] did you find an easy way to get the widget/situation displayed to test/iterate? [14:20] or are you basically going through an "download some upgrade" game every time? [14:21] I've got a VM snapshot I can revert to. [14:21] I've got a reproducible setup. [14:21] And error logs and use gtk inspector to identify the selectors. [14:22] https://usercontent.irccloud-cdn.com/file/hbuiAfJV/Screenshot%20from%202016-11-02%2010-20-21.png [14:22] Guest76787: sorry, i'm not having much luck here. fwiw, that is what it looks like in my container ^ [14:24] flexiondotorg, let me have a look [14:28] seb128, See the GTK warnings here - https://bugs.launchpad.net/ubuntu/+source/deja-dup/+bug/1623835/comments/2 [14:28] Ubuntu bug 1623835 in deja-dup (Ubuntu) "Updated gtk element means you need to specify a size on the details drop down or it is only 1 line high" [Low,Triaged] [14:30] seb128, And here is the GTK inspector info - http://imgur.com/a/05gRH [14:31] flexiondotorg, in what file is the ui described? [14:31] I was just about to start looking at that. [14:31] seb128, You'll remember I said the Deja Dup issue is the same? [14:32] yes, which still seems weird ot me [14:32] The reason is, the first action Deja Dup performs is to install some plugins using aptdaemon. [14:33] ah right [14:33] well deja-dup has the same issue on its backup view [14:33] Ah, OK. [14:33] attente: I see, thank you. [14:35] flexiondotorg, step 1 is to find where the UI is described [14:35] Got the three .ui files open. [14:35] they are from aptdaemon ? [14:36] Yes. [14:40] flexiondotorg, k, found an easy way to reproduce, the aptdaemon source has a gtk3-demo.py small ui which does it [14:41] Thanks. [14:45] flexiondotorg, that seems to fix it [14:46] flexiondotorg, http://paste.ubuntu.com/23416391/ [14:47] seb128, Thanks. [14:48] What was your process of get to that? [14:49] looked at bugzilla.gnome.org bugs since march which mention size or height issues in their title/descriptions [14:50] also the error your shared [14:50] to try to poke around to see if others had similar issues and how they solved it [14:50] there were some suggestion to tweak min content size on scrolled windows [14:50] so I grepped aptdaemon's source for scroll [14:51] Right, OK. [14:51] that gave me 1 source/section [14:51] Understood. [14:51] then I tried the tweak [14:51] I'm not sure it's the right/correct/only solution but that should go [14:51] So I've notice the aptdaemon gtk3 widgets are using deprecated GtkVBox and GtkHBox. [14:51] gtk geometry allocation and why things behaves differently between versions is always weird to me [14:52] Is that worth addressing? [14:52] seb128, Thanks for the time. [14:52] yw! [14:52] I can test futher. [14:52] I wouldn't bother porting aptdaemon to modern gtk [14:53] it works with current gtk3 which isn't going to change [14:53] And also use what I learned to work on Deja Dup backup. [14:53] if we invest work is likely going to drop aptdaemon and just use packagekit [14:53] flexiondotorg, thanks [14:53] seb128, Yep, I agree about leaving aptdaemon widget on gtk3. [14:54] I did remove the HBox and VBox deprecations from some MATE bits, but ultimately didn't gain anything other than soothing my OCD. [14:54] flexiondotorg, just for extra infa, https://bugzilla.gnome.org/show_bug.cgi?id=771743 was the bug that hinted me to try to min-content-height tweak [14:54] Gnome bug 771743 in Mouse "Mouse panel very small when touchpad is present, on Wayland" [Normal,Resolved: fixed] [14:54] seb128, I've also looked at the zebra striping bug. [14:55] what was that one again? [14:55] There is an patch available, that could be refactored. [14:55] oh the list one [14:55] https://bugs.launchpad.net/ubuntu/+source/ubuntu-themes/+bug/945430 [14:55] Ubuntu bug 945430 in ubuntu-themes (Ubuntu) "Lists lack zebra-striping" [Low,Triaged] [14:55] right [14:55] would be nice to fix but that's probably not a priority [14:55] The patch has been reviewed and updated upstream a few times, but not accepted yet. [14:55] I wouldn't invest too much effort into it [14:56] OK, I'll have a stab at it one evening. [14:56] thanks [14:57] annoying but typical example of GTK upstream :-/ [14:57] "I didn't much time thinking about the removal at the time because none of the themes seemed to use it..." [14:57] seb128, The patch add new selectors. [14:57] So themes would have to change as well. [14:57] right [14:58] The existing theme syntax is no longer valid. [14:58] well, would it change look on an unmodified theme? [14:58] oh, that's more annoying [14:58] It does require a patch and theme change to reinstate stripes. [14:59] Which may be good enough reason to not prioritise it? [14:59] it's a regression in 16.10 only right? [15:01] What the striping? [15:01] the bug is a bit confusing [15:01] it says gtk 3.18 allowed to do that but 3.19 removed the option [15:02] unsure our theme was using the feature in 3.18/16.04 though [15:02] the launchpad bug suggests it wasn't [15:02] I think striping is missing from 16.04 too. Just double checking some themes... [15:03] https://bugzilla.gnome.org/show_bug.cgi?id=757495 first comment suggests that 3.18 made it possible at least [15:03] Gnome bug 757495 in Widget: GtkTreeView "GtkTreeView odd/even row styling no longer works" [Normal,New] [15:03] seb128, Laney, Trevinho this should fix the issue (works on X, i'm checking it on Y) [15:03] andyrock, "this"? [15:03] Stripes missing on my 16.04. [15:03] i assume using /org/freedesktop/login1/user/self is safe [15:03] https://code.launchpad.net/~azzar1/unity/fix-double-lock/+merge/309859 [15:04] andyrock, using user/self makes sense to me [15:04] Never heard of that [15:04] is it documented? [15:04] dunno [15:04] but I'll check if it works properly [15:05] I mean is it guaranteed to not be randomly removed? [15:05] seb128, Sould the update-manager fix be SRUd to Yakkety also? [15:05] there's no mention of 'self' in https://freedesktop.org/wiki/Software/systemd/logind/ [15:06] flexiondotorg, my understanding is that gtk 3.18 was making it possible and 3.19 removed the support for it, so basically we could probably add it back to the lts by tweaking the theme [15:06] where > 16.10 would be more work [15:07] flexiondotorg, I wouldn't bother looking at it for > 3.18, not high priority enough, might be a nice SRU for the LTS if it's an easy tweak but again a low priority/friday afternoon thing [15:08] seb128, I'll double check the theme but I'm sure I've seen the appropriate styling that should enable it. [15:08] For 16.04. [15:08] I don't know that you can do it for treeview rows with 3.20 [15:09] Someone's phone is ringing in the library [15:09] * Laney does a Very British Hard Stare [15:09] you probably can't [15:09] He's withered to nothing [15:09] Success [15:09] that's what the bugzilla bug I pointed before says [15:09] Yeah I think you'd need to add some style classes or something [15:09] Company removed the option because he though noone was using it [15:09] ah yes [15:09] there's a patch but it's not finished / committed [15:09] right [15:10] so we could restore it once that is done [15:10] * Laney subscribez [15:10] yeah [15:10] I doubt they are going to do much work/changes on gtk3 now though [15:10] shrug [15:10] find out at the time [15:11] Laney, does http://paste.ubuntu.com/23416391/ looks like right or a hack to you? (that fixes the update-manager details scrollview being not having enough height) [15:11] seb128: I think you need something like that now [15:11] ok I double checked [15:11] where did 300 come from? [15:12] two different users logged at same time [15:12] just me trying a random value to see if the call makes a difference [15:12] which is does [15:12] user/self points to differents objects [15:12] should be fine [15:12] I don't like to add fixed values though [15:12] would be nice if GTK was clever enough to fit to the space available as it used to do :-/ [15:13] andyrock: but is it part of the API? [15:13] how did you find out about it? [15:13] andyrock, I think Laney's point was that it might go away in a systemd update if it's not properly documented [15:13] so we shouldn't rely on it [15:13] or get upstream to document it as officially supported [15:16] ah... [15:16] is there a freenode channel for systemd [15:17] andyrock: Yea, it's #systemd. [15:17] if you use get at least a test [15:17] so the day systemd change we get something flagged red === didrocks1 is now known as didrocks [15:20] seb128: adding such a test in our framework is hard [15:21] we don't rely on real-world dbus [15:21] talk to #systemd upstream and see what they say [15:21] maybe inside cmake we can do something [15:23] andyrock, but at the same time getting the uid is easy and that's probably supported/stable [15:23] i know but self is nicer :P [15:26] andyrock: Also, ew [15:26] use the second part of the tuple instead of string concatenation on the session id [15:27] mmm [15:27] dunno in this case is going to be a bit harder to still support XDG_SESSION_ID [15:27] maybe not [15:27] i'll update the branch [15:28] in that case you should be using GetSession IMHO [15:29] I can use the second part still supporting XDG_SESSION_ID [15:29] i'm pretty sure Trevinho will not accept deprecating XDG_SESSION_ID [15:30] nah, I'm saying that in this case you should be calling org.freedesktop.login1.Manager GetSession (XDG_SESSION_ID) to get the object path [15:32] ah... we likely did like that because our dbus api is async and it's not always trivial to get dbus properties [15:33] the risk is that the implementation changes one day and breaks you [15:33] same as the other case [15:33] :P [15:34] let's see if Trevinho still wants the XDG_SESSION_ID thing [15:34] maybe he's using it in some scripts [15:34] otherwise it's easier to remove that code path [15:35] * andyrock 's day is almost over and he wants to propose it before eod [15:37] seb128, The default scrolled window sized in 16.04 was 200px. So I'll use that. [15:38] flexiondotorg, wfm, the 300 value was a random test, I played around a bit and was leaning toward 180px [15:38] so you are close [15:38] I measured it ;-) [15:38] apt install screenruler [15:38] :-) [15:39] I didn't look at deja-dup but I expect it's going to be a similar issue [15:39] There's a case in update-manager too if you weren't already aware [15:39] the "technical description" bit [15:40] Laney, Yep. Saw that. [15:40] ok [15:40] Thanks for the prompt though. [15:40] Deja Dup is the same issue. [15:41] that was expected ;-) [15:41] unsure if we should bother SRUing those to yakkety [15:42] I would, shouldn't be hard [15:42] I'm happy to. [15:42] Just a case of merge proposals for Zesty and debdiffs for Yakkety, right? [15:45] If you want, or you can do MPs for both (I can push the branches to ubuntu-yakkety) [15:46] I've not done a MP for a prior release before. [15:46] What branch to I merge to? [15:46] Just change Vcs-Bzr [15:46] s/to/do/ [15:46] I think it needs to exist to make a proposal to it - that's what I just offered to do [15:47] bzr branch ...; bzr push ...-yakkety [15:47] OK [15:48] your call for the SRUs, it has a small work cost between upload/review/test and it's a non LTS and only cosmetic issues [15:48] but polish is always nice [15:53] flexiondotorg: created lp:~ubuntu-desktop/deja-dup/ubuntu-yakkety and lp:~ubuntu-core-dev/update-manager/yakkety - doesn't look like aptdaemon has one (no Vcs-Bzr at least) [15:54] Laney, thanks. [15:55] * Laney spends his last £1 [15:56] Laney: perhaps relevant :) http://hyperallergic.com/299912/the-photographers-of-1870s-london-who-documented-their-disappearing-city/ [16:02] sarnold: you calling me an old relic? [16:03] Laney: hehe, no :) it just triggered the "wow london is old" feeling.. [16:06] Ah, my quaint old currency [16:07] here we're suffering the consequences of 1970s architectural vandalism [16:08] * Laney stabs the Broadmarsh centre [16:13] Laney, did the Snenton market revamp ever get finished? [16:13] *Sneinton [16:14]