[05:02] hi [06:46] morning [07:08] Happy Monday! [07:18] ....slow morning :p [07:18] * allison1 waits for UK [07:24] allison1: morning from the UK - even if I'm not the droid you were looking for :) [07:24] flocculant: hey :D [07:45] Trevinho: hi, can you review https://code.launchpad.net/~muktupavels/compiz/lp1530277/+merge/321647? [07:45] checks [07:46] muktupavels: that won't override GDK_SCALE? [07:46] Trevinho: I think no, but I will check now [07:47] muktupavels: ack, let me know, in case you might just get the value from the env if set I guess [07:48] Trevinho, device scale must be 1 otherwise you get problems, check screenshot in bug [07:49] GDK_SCALE is now ignored, but at least decorations looks correctly, only not scaled. [07:50] Full fix will require libmetacity 3.26 where I have meta_theme_set_scale. [08:00] morning all [08:02] morgen! [08:06] Trevinho: GDK_SCALE just affects GTK+, and then gdk_window_create_similar_surface will create surface with same device scale. [08:06] muktupavels: ok, let's go with that then [08:06] hi willcooke, Laney [08:06] hey Trevinho, how's the weather :) [08:07] ah, Trevinho on IRC [08:07] what a GREAT DAY it must be [08:07] see Laney, it's *not* a holiday [08:07] AHAHAHA [08:07] how you getting settled in? [08:07] :D [08:07] hey willcooke too [08:07] willcooke: it's always pretty nice, but there are showers here and there... you know tropical weather [08:08] ~30° and very humd [08:08] humid* [08:08] but AC here works well :-) [08:09] Laney: yeah, almost... House is very near to the coworking, a little out of the center, but I've already got a scooter [08:10] good work [08:10] made some friends? [08:13] Laney: yeah, i'm with a group (http://theremotetrip.com), so... You can make friends pretty quickly. [08:19] Trevinho: hi [08:19] Trevinho: Trevinho: could you help with the possible incompatibilities between Unity's XI2 and the changes in https://code.launchpad.net/~banw/compiz/compiz.a11y-shotcuts/+merge/320091 ? andyrock tells me you might be able to at least shed some light :) [08:21] good morning, british people! [08:24] good morning desktop [08:25] hi seb [08:25] good weekend? [08:25] hey allison1 [08:25] yeah, busy with small house work mostly, but also enjoyed the sunny weather yesterday [08:25] you? [08:25] no. not really. [08:26] :( [08:26] maybe next one :) [08:29] yeah [08:32] :( [08:32] * Laney hugs allison1 [08:32] hey seb128 [08:32] hey Laney, how are you? [08:33] pretty good! [08:33] had some nice sun time down the allotment [08:33] and went to someone's 70th(!!!) birthday [08:33] * Laney 's activities are those of a 70 year old these days [08:35] re and moin [08:35] hey Sweetshark [08:36] Laney is becoming a gardening master [08:36] mostly weeding :( [08:36] oh yesterday I was lying down in the hammock and drinking tea [08:37] getting 3177 mails rn. on the first account. [08:37] hey Sweetshark [08:37] mark all as read [08:37] if it's important you'll get another one [08:37] happy to be back? [08:38] Sweetshark, has was the camping? [08:38] how* [08:38] Laney: gardening master? interesting! TBH I not having to do gardening made me go for a simple appartment instead of a house ... [08:40] it's good for mind, body and soul! [08:40] * Laney foraged some wild garlic & nettles on the way home too [08:40] soup for lunch today [08:41] seb128: camping was ... not like I remember it from childhood: . But the last two days were quite awesome as the weather improved. I visited https://en.wikipedia.org/wiki/Skamlingsbanken on the way back, its well worth the visit. [08:46] Anyway, this was just a testdrive essentially. Driving around such a big Diesel truck requires some training, esp. if you want to take it to camping sites, were roads arent always the best. Im feeling wellpreprared for a longer summer trip (would put a better matress in the van for that though). === JanC_ is now known as JanC [09:01] Morning desktopers [09:01] hey flexiondotorg [09:03] hey flexiondotorg willcooke [09:19] b4n: I can see [09:19] thanks [09:22] hey Trevinho! how are you? enjoying the cool work space? ;-) [09:22] hey seb128... Yeah, it's nice... And a thunderstorm is coming soon, so good time for working :-D [09:22] haha [09:22] b4n: well, can you give a look at unity's unity-shared/InputMonitor.cpp ? [09:23] b4n: basically that is doing a similar thing, but at unity level... [09:23] So you might instead port it to compiz (as it's pretty generic) [09:23] and then we could use it in unity gtoo [09:23] Trevinho: yeah I saw it, but I'm noly using raw events so IIUC it shouldn' tinterfere too much [09:23] as, the current implementation, I'm not sure weather will cause some troubles... [09:24] Like, it's quite possible that the motion detection will stop working there as unity disables that once the last client that needs it has left [09:25] as for the raw events, there shouldn't be problems.. [09:25] But in general we could just move that InputMonitor to Compiz, and expand it to use for your needs.. [09:26] Trevinho: one thing I didn't investigate yet and am not sure about, is whether that inputmonitor is using GDK a filter or X directly [09:26] b4n: X directly [09:26] that would be great I guess [09:26] k [09:26] ah, one thing that might be the root of al problems is that X's API to get the event data is very bad for multiple clients [09:26] Well, actually we get that using a Nux event monitor, but it's actually X [09:27] Yeah, not being based on callbacks, it would cause troubles indeed [09:28] b4n: as an alternative, I could just use compiz events, but not sure that some of them might be filtered and thus causing problems [09:28] some might with Compiz's actios handling I guess [09:29] I mean, compiz HandleEvent function shouldn't remove any event... And in that case we could just use that... But still it would be just better to have something like InputMonitor inside compiz at this point [09:30] how easy do you think it would be to move inputmonitor to Compiz for you or me? and also, how do you see the non-grab actions being registered/triggered wit htha scheme? [09:30] I didn't do that in the past, as no one seemed to need it for compiz, thus I just implemented in unity, although it's really generic... And a part few calls, it can just moved somewhere else and used to convert XI2 events to normal Xevents [09:31] Well, they will work indeed using this [09:31] my point is that not *all* actions should be triggered during grabs [09:31] I also wanted to use the input manager for more key actions in unity, so... it's not a bad thing [09:31] like in the lock screen, you don't want things like switcher to work [09:32] sure, but we disable that anwyay in unity [09:32] In our case, we just disable most of compiz actions... [09:32] my approach was to make actions opt-in for XI2 events [09:33] yeah, I see [09:33] and that's fine [09:33] Maybe adding even some configurability at that level would be nice, but UI would miss that [09:33] unless you don't change also some compizconfig stuff [09:34] bah, I'm not really sure it makes much sense fo rthe user to really choose [09:34] it would be nice I guess, but I don't see that as a requirement [09:34] Well, sure.. But for custom actions, for example... [09:34] anwyay, really not important [09:34] fair point [09:35] Anyway basically I was just doing something similar you did... It's I was also using some C++ templates to avoid redoundant code, so basically i'm generating events quite easily for any type of them [09:35] I saw that, pretty nice [09:35] I guess your code is far better than mine, esp. as it doesn't seem riddled with FIXMEs :) [09:36] however, at this point my concern about inputmonitor is that IIUC it listens to non-raw XI2 events, and basically forwards them all. In that case, how do you know/handle non-XI2 grabs? [09:36] It has some c++11, but still it's something we now use in compiz anyway [09:36] or do you willingly ignore all grabs? [09:36] yeah [09:37] No, we enable the input monitor only if there are callbacks wanting some kind of events... [09:37] So, in case of actions, I guess, that it could be done the same... When an action needs that, you can just add a callback and then it will just use it [09:38] yeah, but simply listening for XI2 events stops Core ones on that window [09:38] so if you forward the events, how do you know whether there was a core grab or not? [09:39] or don't you forward the Core version of the event in any way? [10:22] Sweetshark, heya [10:23] Sweetshark, I assume you will prepare the final 5.3.2 this week. so I am going to copy the 5.3.2~rc2 builds to the PPAs [10:23] note that I pushed a small fix to git [10:26] ricotz: yes. thanks. /me still wades through backlog. Do you have a link for the small fix so can put it aside for later ... [10:26] Sweetshark, it is pushed to the ubuntu-zesty-5.3 branch [10:32] ricotz: what issue is that change supposed to fix? [10:42] Sweetshark, really?, it is a mismerge and the corresponding Suggests points to libreoffice-help (= 5.3) === hikiko is now known as hikiko|ln === hikiko|ln is now known as hikiko [12:49] alan_g, hello [12:49] are you around? [12:54] hikiko: ? [12:54] hey :) [12:54] alan_g, remember miral crashed inside my schroot? [12:55] I do [12:55] I did some minor changes in miral and it seems to work well now [12:55] the problem is I am not sure they are correct [12:56] would you mind to get a look in my diff? [12:56] it's very short [12:56] There were two problems. One is fixed, the other was in Mir. [12:57] I've found 2 in miral [12:57] not sure [12:57] "fixed" as in committed and in a silo [12:57] do you want to see? [12:57] Sure [12:57] https://paste.ubuntu.com/24306422/ [12:57] the first problem I have is that it couldn't find the cursor theme [12:57] had* [12:58] https://pastebin.com/31DdpmYg [12:58] and I solved it by returning a software cursor instead of throwing an exception (last change in diff) [12:59] the other one [12:59] was this: https://pastebin.com/2kACdwRr [13:01] and I am not so sure my fix is fine but I noticed that auto line was trying to convert the raw line to wide string causing exceptions for some text... also when I was printing it on gdb it seemed to have garbage inside [13:01] this change seemed to fix it [13:02] The last one looks like bug 1677550 - which is fixed. (Committed and in silo 2676) [13:02] bug 1677550 in MirAL "if you update to zesty's libstdc++-6-dev then miral-shell will crash when trying to draw its background text" [Critical,Fix committed] https://launchpad.net/bugs/1677550 [13:03] alan_g, could be [13:03] You got -r 545 [13:03] ? [13:04] On the other problem: I don't see that loading a different cursor than the one asked for is the correct behaviour. [13:06] alan_g, then, is there another option instead of exiting with exception? [13:06] i have installed default cursors etc [13:06] but miral doesnt run because it can't find them [13:07] I was going to say, I thought the depended on the cursor theme. [13:07] I was going to say, I thought the *package* depended on the cursor theme. [13:07] https://pastebin.com/31DdpmYg that's the message I get [13:08] alan_g, it did there was a bug report on that and a branch of yours that fixes it [13:08] I have dmz-cursor-theme installed [13:08] * alan_g wonders why it isn't found [13:08] but miral still crashes with that message [13:09] Could you strace and see where it looks? [13:10] alan_g, sure, I just don't know how to strace [13:10] give me a few minutes :) [13:10] strace miral-shell 2>&1 | grep open [13:14] Normally we see something like http://paste.ubuntu.com/24306579/ [13:16] alan_g, ok reverted https://paste.ubuntu.com/24306585/ [13:16] that's when I run my local build === dandrader is now known as dandrader|afk [13:17] maybe I should paste the output of the system's miral too [13:17] Here's the problem: open("/usr/X11R6/lib/X11/icons/default/cursors/", O_RDONLY|O_NONBLOCK|O_DIRECTORY|O_CLOEXEC) = -1 ENOENT (No such file or directory) [13:18] You don't have any cursors in your schroot [13:18] but I've installed the dmz-cursor-theme :S [13:19] Hmm, sorry read that wrong [13:23] hikiko: do you have XCURSOR_PATH set? [13:23] seb128: I see that langpack building for xenial has started to work in the PPA. [13:23] GunnarHj, did they? interesting [13:23] I didn't change anything [13:23] alan_g, $ echo $XCURSOR_PATH [13:24] GunnarHj, btw did you see that I uploaded zesty ones on friday? [13:24] returns nothing [13:24] seb128: Weird. But good. :) [13:24] so probably no alan_g [13:24] seb128: No, I haven't seen anything zesty yet. Where? [13:25] GunnarHj, https://launchpad.net/ubuntu/+source/language-pack-de-base/ for example [13:25] GunnarHj, see the zesty lines, there is a "proposed" version [13:25] GunnarHj, works on any locale, you just need to change the "de" by your locale [13:26] hikiko: "$ cat /usr/share/icons/default/index.theme" - do you have that set to breeze_cursors? [13:26] GunnarHj, some testing would be welcome, maybe worth emailing the translators list about, I did basic checking on the fr ones but that's about it [13:26] I have Inherits=DMZ-White [13:26] [Icon Theme] [13:26] Name=Breeze [13:27] seems so [13:27] Inherits=breeze_cursors [13:27] And breeze isn't installed in the schroot? [13:27] it is [13:27] I installed cursors* [13:28] Humor me and try Inherits=DMZ-White [13:28] seb128: Asking the translators about testing in the dev release would be a new thing. Do you think it's necessary? (I can install and review Swedish if you like.) [13:29] GunnarHj, well, I don't know if a call for testing is necessary but having some people trying some locales would be nice since that machinery was buggy for a while and there could be still issues/bugs [13:30] GunnarHj, so yeah, if you want to test the ones for the locale you use that would be welcome :-) [13:31] alan_g, that works, but the system installed miral that doesn't have your fix (I think) gives now the 2nd exception [13:31] seb128: I'll do that to start with. Please note that the ubuntu-docs translations is not included in this one, so a build of a later full export will be necessary. [13:32] GunnarHj, that's a full export, do you know why it's missing and how to get it included in the next export? [13:32] hikiko: yes, the fix is in a silo awaiting QA [13:33] alan_g, I pulled latest miral and I am reverting all my changes to see if dmz-white+your fix => miral working again [13:33] seb128: It's about timing. That export was on March 28, and ubuntu-docs was built on March 31. [13:33] k [13:33] so it has the translations but outdated? [13:33] alan_g, it's back [13:33] :D [13:34] thanks [13:34] hikiko: I guess there's a cursor problem with the Breeze theme. [13:35] hikiko: "it's back" - the problem? [13:35] alan_g, I only installed breeze after I got the miral crash [13:35] seb128: So there may be a need to check the "request a full language pack export" option at https://translations.launchpad.net/ubuntu/zesty/+language-packs [13:36] GunnarHj, noted [13:36] no, I mean it works ok now, alan_g, problem solved [13:36] but it can't be breeze [13:36] * alan_g wonders why [13:36] because at the beginning I only had the dmz cursor installed [13:36] nothing else [13:36] and I was having the problem [13:36] when I saw that I can't fix it [13:36] I installed cursors* [13:37] to see if miral can detect any of them [13:37] unfortunately I don't know what my index.theme had inside by then [13:38] certainly not DMZ-White [13:38] maybe the cursor wasn't set at all [13:39] Yeah, the diagnostics from xcursor_load_theme() leave something to be desired. === dandrader|afk is now known as dandrader [13:56] seb128: There is a problem: You must build/upload the language-pack-XX and language-pack-gnome-XX packages too, not just the -base packages. Currently I can't install the Swedish base packages because of unmet dependencies. [13:58] hikiko: did reconfiguring the cursor theme also resolve bug 1677523? [13:58] bug 1677523 in Mir "mir_demo_server segfaults in schroot" [Undecided,Incomplete] https://launchpad.net/bugs/1677523 [13:59] Laney: heyo [13:59] alan_g, I forgot this report should be marked as invalid [14:00] Laney: would you be able to sponsor https://code.launchpad.net/~attente/unity-gtk-module/blacklist-anjuta/+merge/319589 [14:00] I had forgotten to mount /usr/run/1000 in the chroot [14:00] and only root could run mir [14:02] hikiko: great. Another bug bites the dust. [14:23] GunnarHj, hum, k, I don't know how to do that :-/ [14:23] that's what the langpack job generates [14:24] seb128: Basically they should be empty, but they must exist... [14:24] GunnarHj, do you know how to build those? [14:24] seb128: No, I have no idea. [14:24] :-( [14:24] me neither [14:25] seb128: It's soon an emergency. Ask pitti? [14:25] I guess you just did [14:26] seb128: Aha, didn't know he is still listening here... [14:27] he's still on the channel, dunno if he's "listening" [14:30] seb128: Looking here: [14:30] https://launchpad.net/~ubuntu-langpack/+members [14:30] Maybe sil2100 is able to help? [14:30] I doubt it [14:30] I don't think he's familiar with the langpackomatic job [14:31] I only know langpack-o-matic for the ubuntu-touch use-case, but let me read the backlog [14:32] hm, I don't know how to do that straight away, but let me take a look and maybe try to figure it out [14:33] seb128, sil2100: A workaround might be to create dummy packages manually. But there would be quite a few of those... [14:35] sil2100, GunnarHj, hum, I've an idea [14:36] let me try that [14:36] pitti, unping [15:16] GunnarHj: It won't soon be an emergency. Those things won't migrate out of proposed. [15:17] You should stop running zesty-proposed too [15:17] seb128, GunnarHj: sorry, ECONTEXT -- why were teh -gnome packages uploaded/copied but not the base packages? [15:17] pitti: Other way around [15:17] (hi!) [15:18] Laney: I know, but not so many days remain before release. [15:19] Laney: (I only enable -proposed when there is a reason.) [15:20] pitti: The -base packages were built/uploaded, but not the delta packages. [15:20] oh, that's much worse indeed; how could that happen [15:20] were they built by langpack-o-matic? [15:20] (I don't have access to the infra any more) [15:21] pitti: seb128 is the only one who can answer that. [15:22] He unpinged too, so I would say that we should hold off a bit [15:26] pitti, hey, probably because I sucked at understanding http://bazaar.launchpad.net/~ubuntu-langpack/langpack-o-matic/main/view/head:/doc/operator-guide.txt [15:27] pitti, but I've been hitting a stack of issues and "./packages upload" did nothing so I ended up doing a manual dput [15:28] pitti, I did the different steps manually as you recommended, but I think if you build without doing the upload it wipes the "updated-packages" list [15:28] then upload think it has nothing to upload [15:28] I should maybe have wipped and redone the whole dance then, rather than trying to just do the upload step [15:29] seb128: ah -- what you can do is something like ls ../zesty/sources-base/* ../zesty/sources-update/* > updated [15:29] seb128: to reconstruct a "changed package" list, if you built from scratch [15:29] (obviously not for an update) [15:30] seb128: i. e. you just dput the -base ones, but not the update ones? you can still do that then [15:30] pitti, right, thanks, I found my way around, I just ended up dputting the content of the -base and not -update [15:30] right [15:30] I didn't realize at first that both dirs were needed [15:30] I though -base was for a full export [15:30] and -update was for --update rounds [15:31] pitti, thanks for the reply! [15:31] de rien ! [15:31] nice! [15:32] pitti, how are you otherwise? still having fun with js&co? ;-) [15:32] seb128: right now I'm actually having fun with a bug in Ubuntu 16.04's pbuilder :) [15:33] but JS too of course :) [15:35] AFK for the next 2 hours or so. See you guys later or tomorrow. [15:36] See you GunnarHj [15:37] bye GunnarHj === dandrader is now known as dandrader|afk [16:38] jbicha: hey, would you be able to look at this branch when you have time? https://code.launchpad.net/~attente/gtk/gtk-mir/+merge/321739 [16:41] attente: sry got distracted, sure can do [16:41] Laney: np, thanks a lot :* [16:42] is_known_setting () is quite a thing [16:43] g_strv_contains ()? [16:44] good idea... [16:46] is there a bug for the anjuta thing? === dandrader|afk is now known as dandrader [16:50] no, i was told directly about it through email [16:50] i'll file one now [16:50] it's ok [16:51] just would have linked it to the mp if there was [16:51] no need [16:55] ok, thanks [16:55] it's here now anyways in case you want to use it: https://bugs.launchpad.net/ubuntu/+source/unity-gtk-module/+bug/1679236 [16:55] Ubuntu bug 1679236 in unity-gtk-module (Ubuntu) "Anjuta crashes when creating a window in the UI editor" [Undecided,New] [17:07] attente: https://bileto.ubuntu.com/#/ticket/2690 please test the PPA once built then someone can publish for you or I can tomorrow morning [17:07] night! [17:08] night Laney [17:10] * Laney prints the quiptic [17:10] in for a rock and roll evening :> === dandrader is now known as dandrader|afk [17:59] thanks Laney, that PPA fixes it === dandrader|afk is now known as dandrader === JanC_ is now known as JanC [19:54] willcooke, hi [19:54] morning robert_ancell [20:20] night all [20:59] mterry: why is the latest ubuntu-terminal-app not in zesty? [20:59] jbicha: I don't know... there was a silo for it at one point... [21:00] the new version looks a lot nicer :) [21:02] mterry: oh, never mind, it's in unapproved [21:02] jbicha: it is nicer, yeah! :( [21:02] :) [21:02] I mean [21:06] I assumed the :( was because it was still unapproved [21:11] Heh, then both do apply I guess :) [21:13] I've got a mp for ubuntu-terminal, who should I ping to review it? [21:18] jbicha: I'd say Kaleo (Florian) === JanC is now known as Guest6267 === JanC_ is now known as JanC