[06:17] hey Mirv [06:17] Mirv: can you have a look at bug #1075375, seems an upgrade issue [06:17] Launchpad bug 1075375 in libunity (Ubuntu) "package gir1.2-unity-5.0 6.8.0-0ubuntu2 failed to install/upgrade: subprocess new pre-removal script returned error exit status 127" [Undecided,New] https://launchpad.net/bugs/1075375 [06:17] would be nice to investigate it a little bit [06:18] didrocks: ok, looking [06:19] thanks :) === philballew_ is now known as philballew [08:34] duflu: Hi :) Thanx 4 the reviews. I've used the default color now: https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix1070297-no-possibility-to-change-background-outline-color/+merge/132500 [08:35] MCR1: Hi. Yes, you don't need to tell me. I get notified immediately. [08:35] duflu: ok, ack :) I thought you've had problems with the launchpad mail not notifying you ;), sry... [08:37] mmrazik: Hi :) And thx 4 re-approval. [08:37] MCR1: no problem :) [08:40] duflu: Thanks a lot also for the approval. Finally resizeinfo should be tuned 8-) [08:51] MCR1: don't worry about the recent failures. I'm taking care of that. [08:51] mmrazik: thx a lot :) [10:08] * didrocks tries another way to deal with parallel and join jobs in jenkins, too hard to track in the API what is doing what === mmrazik is now known as mmrazik|lunch === _salem is now known as salem_ === mmrazik|lunch is now known as mmrazik|afk === MacSlow is now known as MacSlow|lunch === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === MacSlow|lunch is now known as MacSlow === dandrader is now known as dandrader|lunch [14:46] mmrazik|afk: hey, to be able to merge https://code.launchpad.net/~mterry/bamf/inline-debian/+merge/132944, can you please remove the packaging branch for it? (and avoid running autoreconf as well) === mmrazik|afk is now known as mmrazik [14:46] mmrazik|afk: once done, feel free to flip the swithc [14:46] switch* [14:47] didrocks: Ok. I'll do the change and approve. [14:47] mmrazik: thanks a lot :) [14:55] didrocks, will you take care of approving all the inline-debian branches? [14:56] mhr3: yeah, don't worry with that :) [14:56] mhr3: I'm doing one after another first :) [14:56] awesomeness [14:56] didrocks: it will take a moment. The job failed. The compiz inline stuff was merged manually and there is one hook which doesn't take the newly creted branch into accout [14:56] thx :) [14:56] fginther is looking into it [14:57] but I'm tempted to merge this stuff manually (or after building just the proposed branch somewhere to check it really works) [14:58] mmrazik: ok, thanks! keep me posted :) [15:16] mmrazik, hello! I've been preparing various inline-debian/ branches for unity components that will need their recipes changed to stop pulling in an external packaging branch [15:16] mmrazik, notably so far, compiz and bamf [15:17] mterry: compiz is in. fginther is checking bamf [15:17] mterry: can you work with fginther on this? He is in US timezone so it should be simpler [15:17] mmrazik, OK [15:21] mterry, hello [15:21] fginther, hi! [15:22] mterry, just ping me as you get the branches ready, we may have to do these by hand for a bit [15:22] fginther, yup. So I have several branches waiting to be approved, but I'll ping you once a human says they're OK. Only pending branch in that state is bamf === dandrader|lunch is now known as dandrader [15:33] fginther: you did see my note about dropping running autoreconf for those before building the package, isn't it? [15:34] didrocks, no sorry I missed that. Thanks for updating [15:35] fginther: yeah, so you need to: 1. remove the "external packaging bits", and 2. not installing extra rdepends and running autoreconf/creating tarballs anymore [15:35] for those packages :) [15:36] didrocks, thanks [15:36] yw, good luck! [15:38] hi, i'm trying to compile unity (after compiling nux, according to this guide: http://unity.ubuntu.com/getinvolved/development/unity/), but i'm always getting stuck at: http://paste.ubuntu.com/1337547/ can anyone please give me a hand? [15:39] mterry: ok, looking again this afternoon (done a first pass to them this morning). All looks good to me (maybe some missing lenses, like the remote video scope one for instance or the music/file lenses). I assume that you tried them in a chroot to ensure you don't miss gnome-doc-tools and so on? [15:39] didrocks, yeah I've built all in a chroot [15:40] mterry: rocking! Thanks :) [15:40] didrocks, I haven't gotten to all the lenses yet, yeah [15:40] didrocks, that's for this morning [15:40] mterry: I just conditionnaly +1 one, but didn't change the switch, I'll let you coordinate with fginther :) [15:40] * mterry gets a flood of didrocks reviews in his inbox [15:41] didrocks, OK [15:41] mterry: yeah, did the review this morning, but wanted to talk to you first :) [15:41] mterry: thanks for the quick changes! :) [15:41] mterry: getting the flood? you are subscribed to all merges now? :-) [15:41] (even compiz? :p) [15:42] didrocks, no, just my merges [15:42] :) [15:42] mterry: this is for this afternoon, right? ;-) [15:42] I wonder if we have a team for that: all unity stack only + compiz [15:42] mmrazik: do you know? ^ [15:45] fginther, so the following branches are approved but need their jenkins recipe fixed: bamf, dee, libunity, nux, unity-asset-pool, unity-lens-applications, unity-lens-video === mmrazik is now known as mmrazik|otp [15:45] mterry, got it. I hope to have the recipe scripts fixed shortly [15:54] didrocks: you mean for unity/compiz maintenance? [15:54] mmrazik|otp: for the whole unity stack [15:55] didrocks: it should be Stephen and his squad for 13.10 [15:55] err [15:55] 13.04 [15:55] + some more people (like duflu) [15:55] mmrazik|otp: sorry, I meant "team" as "launchpad team" where mterry can subscribe [15:55] didrocks: I don't think so [15:55] so that he receives all MR for this stack [15:55] but unity-team and compiz-team should do it [15:55] so the only option is to subscribe to every components? [15:55] or maybe not.. [15:55] mmrazik|otp: can we add him to them? [15:55] didrocks: I fear that is the case :-/ [15:55] didrocks, I can always filter out [15:56] didrocks, I already have an extensive set of filters, what's 10 more [15:56] mterry: as you prefer ;) [15:56] mterry: if you need to be part of some teams just send me a list and I'll make it happen [15:57] mmrazik|otp, I think just unity-team and compiz-team [15:57] mterry: ok [16:08] mmrazik|otp, and ~unity-lens-photos (does that really need a separate team?) [16:09] mterry, which team do you suggest, I can change it if needed. [16:09] davidcalle, the rest of the lens seem to use ~unity-team [16:10] didrocks, who can add people on ~unity-team? [16:12] davidcalle, apparently mmrazik|otp can [16:12] davidcalle: pspmteam I guess, but mmrazik|otp would no more === mmrazik|otp is now known as mmrazik [16:14] yes, it is pspmteam (or whoever is member of that team) [16:15] mterry: only davidcalle can add you to unity-lens-photos [16:16] mterry: other than that you should be member of unity-team and compiz-team [16:17] mterry, didrocks, mmrazik, thanks. So unity-team should be set as the driver or maintainer of the project? [16:17] davidcalle: I actually don't know. If you are the maintainer then the team/owner is probably correct [16:18] I mean the current one [16:18] davidcalle, looking at https://launchpad.net/unity-lens-files ... [16:18] davidcalle, pspteam is maintainer, unity-team is driver and owner of trunk [16:19] but the pspmteam indicates PS/Canonical are maintaining and fully owning it [16:19] not sure if that is the case for unity-lens-photos [16:19] mmrazik: it's a canonical project, so it should be unity-team [16:19] If we would do the same with lens-photos david would loose the right to upload to trunk [16:19] (I think) [16:19] mmrazik, it's not, I would be fine with it, as long as I'm still able to manage trunk, since I'm pretty much on my own with it [16:20] davidcalle: yeah, but we need something like this for the automagic uploading model [16:20] didrocks: what do we need? [16:21] wouldn't it be enough if some bot/user is part of ~unity-lens-photos? [16:21] I actually start to be a bit confused :) So don't worry too much about my suggestions.. [16:21] mmrazik: I don't see why unity-lens-photos should be different from any other officially supported lens [16:22] mmrazik: I'm all for consistency, so unity-team would make sense has the owner, or at least, being part of ~unity-lens-photos [16:22] to have similar right without loosing david having rights on those [16:22] didrocks, that would work indeed, I'm adding unity-team to the photos lens team, unless someone adds me or the photos lens team to unity-team. [16:23] davidcalle: let's do that for now :) I'm sure you will be soon in the unity-team, and then, we can revisit :) [16:23] the thing that concerns me is that unity-team has only canonical employees as members and I wouldn't be surprised if somebody is relying on that fact [16:24] mmrazik: non virtual ppa, yeah, it's relying on that [16:24] yes... arm builds.. [16:28] didrocks, mmrazik, ok, then I'm adding the unity-team to the photos lens team. Welcome, we have beer and cookies :) [16:29] thanks davidcalle [16:29] didrocks, np [16:30] davidcalle, :) === gatox is now known as gatox_lunch [16:38] fginther, unity-lens-photos also needs jenkins fixups [16:38] mterry, will add it to the list [16:45] Hi :) Anyone able to drag launcher icons with recent trunk version of Unity ? (staging PPA/Quantal) [16:45] Here Unity crashes immediately... [16:46] & another question: Why is Compiz not updated anymore in unity-team/staging PPA ? [16:47] Compiz is still on r3450, but latest should be r3453... [16:48] sil2100: Can you help me with those questions above ^^ ? [16:50] MCR1: let me read up [16:51] MCR1, the compiz switch to inline packaging is probably the culprit [16:51] the autolanding/dput scripts need some work to get this going again [16:51] yes, I think so also as it is r3451... [16:51] fginther, I'm working on it [16:51] fginther: Is someone working on that ? [16:52] ah cool - great thanks :) [16:52] MCR1: it seems you need to be patient ;) [16:52] sil2100: Unfortunately patience is not something I have a lot of ;) [16:53] sil2100: & I think reporting issues early is better than late... [16:54] or never... [16:54] MCR1, sil2100: the jenkins autolanding server is down at the moment, so I can't give you much at the moment [16:55] fginther: Thanks, no problem - just wanted to make sure someone's aware of it (even better if you are already fixing it...) 8-) [16:56] MCR1, no problem. sometimes these things do slip through... [17:05] larsu: ping [17:06] conscioususer, hey [17:06] larsu: have some mins? I need some extra help with gtkapplication :) [17:07] conscioususer, I found the problem to the bug you filed yesterday: wait 5 seconds before restarting your app, then it works :) [17:07] (don't ask) [17:07] (yes, a fix is coming) [17:07] sure, I've got some time :) [17:07] larsu: great! meanwhile I'll remember the 5 sec advice :) [17:07] larsu: I've been trying to use the command-line signal, but I'm missing something [17:08] http://www.pasteshare.co.uk/p/GA/ [17:08] larsu: I need this to make Desktop Actions work on primary instances [17:09] larsu: this example I pasted *almost* works, but a secondary instance does not return after pinging the primary instance, it freezes [17:09] larsu: the ping does work, though, with the correct command line being sent [17:09] conscioususer, weird, let me try it otu [17:09] *out [17:10] larsu: I'm doing what the docs said, using the HANDLES_COMMAND_LINE flag and connecting to the command-line signal [17:19] larsu: so apparently there's nothing *obviously* wrong, right? :) [17:19] conscioususer, yeah, what you're doing looks right to me [17:19] I'm trying to find the issue right now ... [17:21] larsu: seems to big to be a gtk or pygobject bug too... [17:21] *too === gatox_lunch is now known as gatox === dandrader is now known as dandrader|afk [17:28] conscioususer, yeah... some dbus method doesn't seem to be returning properly, which seems to keep the launcher instance awake [17:28] desrt, any idea what this could be ^^ ( http://www.pasteshare.co.uk/p/GA/ ) [17:30] Hey there. sorry for interruption. where in the code do I find the invokation of the dashboard of the unityshell, when cliked on the button? thanks. [17:31] devnewbee, BFBLauncherIcon.cpp:90 [17:32] unity/launcher/BFBLauncherIcon.cpp [17:34] I assume your talking about the OnOverlayShown method? [17:34] no [17:34] ubus_manager_.SendMessage(UBUS_PLACE_ENTRY_ACTIVATE_REQUEST, g_variant_new("(sus)", "home.lens", dash::NOT_HANDLED, "")); [17:34] sends a ubus message to DashController [17:34] which then shows the dash [17:34] but the Activate function is what gets called when you click on the icon in the launcher [17:38] cool. so at this point the icon is already identified and the dash is actually called by "home.lens"? [17:39] well that home.lens is the lens that gets started ie the default. The message that gets sent/received by the DashController is UBUS_PLACE_ENTRY_ACTIVATE_REQUEST [17:40] mterry, just an update, the server handling the merger testing is down, so I'm mostly stuck. Hopefully it will be up again soon [17:40] fginther, OK [17:41] bschaefer: IC. thank you so far. far more clear now. is there a code documentation somewhere? [17:42] sadly no :( [17:42] ooooh. [17:42] at lease not that im aware of [17:43] conscioususer, looks like you found a bug in GApplication, the CommandLine dbus call doesn't return a value. I wonder why this didn't turn up earlier... [17:43] * larsu might be wrong, though [17:43] ok maybe you can give me just a little hint on what Id like to do. [17:44] larsu: I'll take this as a sign of how many people are actually using GtkApplication to its full power [17:44] possibly :) [17:44] conscioususer, yeah :( Someone's gotta start.... [17:45] When an icon is hovered, then it shall be checked whether there is a "group" and open tiny dashlike view with those. [17:45] larsu: so this would also happen if I wrote the example in C? [17:45] bschaefer: so what woudl I need to look at for this? [17:45] conscioususer, afaics, yes. I'm trying that right now [17:45] devnewbee, a grouping of icons in the launcher? [17:45] or the icon it self is a grouping [17:46] devnewbee, either way, you'll need to look into LaucherIcon.cpp under mouse_enter signals [17:46] mouse_move [17:46] bschaefer: Hi :) Do you have time for a short test of a Expo fix (I could even provide a binary for it, so you would not need to compile...) ? [17:46] larsu: damn, I guess I'll have to do some ugly stuff to allow different actions from secondary instances [17:47] bschaefer: Im not sure what you meant by that difference. the icon just offers on click other launchericons(the actual apps for that group, e.g. system, editors or so) [17:47] MCR1, hello, what kind of fix? plus I can just merge to a branch and compile it as easily as well [17:48] devnewbee, on which click? activate icon is on left mouse click, and quicklist is on the right mouse click... [17:48] bschaefer: Thanks :) - I fixed the broken Expo animations - here is the branch: https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix875311-expo-fadeandzoom-and-vortex-animations-black-screen/+merge/132731 [17:48] devnewbee, either way, LauncherIcon.cpp is what handles mouse events for the icons [17:49] bschaefer: well. a left click to open a quicklist with real icon sotospeak :) [17:49] devnewbee, though you might have to look at LauncherController as well because LaucherIcon only knows about individual icons [17:49] MCR1, ill take a look [17:50] larsu: I guess I'll have to register the application, check get_is_remote and activate an action if it's remote [17:50] larsu: and call run if it's not remote [17:50] bschaefer: I am currently using the .compiz-1 directory to test Compiz plug-in fixes... [17:50] conscioususer: crikey [17:50] what the heck are you doing? :) [17:51] bschaefer: ok. Ill check that. a hint how to replicate a little horizontal board for those icons? [17:51] why are you trying to do HANDLES_COMMAND_LINE? [17:51] desrt, is gapplicationimpl-dbus.c:173 missing a g_dbus_method_invocation_return_value? [17:51] MCR1, hmm Ill try to look at that later, as im working on something that I would rather not have to deal with compiz atm haha :) [17:51] larsu: no. [17:52] the reply is sent when the GApplicationCommandLine instance is finalized [17:52] devnewbee, horizontal board? [17:52] bschaefer: I just put the compiled binary from build/plugins/plugin-name compiz source dir to .compiz-1/plugins in ~ and run setsid unity to test :) [17:52] this allows the invoking instance to do things like waiting to exit until the document that was opened by that commandline to close [17:52] desrt: my app has a command line parameter that should be considered regardless if the app is already running or not [17:52] bschaefer: But no stress... [17:52] *was closed [17:52] bschaefer: ofc :) [17:52] desrt, makes sense, but then why does the pasted example not return from the launcher? [17:52] MCR1, yeah, but it takes time haha [17:53] larsu: perhaps python is not releasing the object? [17:53] * bschaefer needs coffee and to finish working on something else [17:53] bschaefer: well yes. the offerend appicons should be displayed on a board, like the launcherbar itself or the dashboard looks like. [17:53] tying the exiting of the remote to the lifecycle of the commandline object was a pretty lame thing for me to do... [17:53] it should have been an explicit call [17:53] bschaefer: a square with transparent background woudl do it. [17:54] devnewbee, weelll to do that you need to look into Nux [17:54] devnewbee, and Views [17:54] and how to generate a blur texture [17:54] desrt: should we ping pitti? [17:54] conscioususer: ideally you should break the commandline arguments out on the remote side [17:54] and send action invocations [17:54] digging around anything View, like DashView, HudView will have bits of that, along with OverlayRenderer [17:55] desrt: are actions called only on the primary instance? [17:55] yes [17:55] if you invoke an action from the remote instance it will land on the primary side [17:55] ditto activate() [17:55] and open() [17:55] desrt: oh. so that'll be easier than I thought [17:55] desrt: no get_is_remote needed [17:56] conscioususer: no. definitely not. [17:56] this is a mistake that a lot of former libunique users make [17:56] desrt, you're right about python. It works fine from C. It's quite confusing, though [17:56] larsu: i think we need some explicit exit() API [17:56] conscioususer, so this is a pygi issue, but I guess you have a different solution for now [17:56] i'll make a bug about that [17:57] desrt, agreed, thanks [17:57] larsu: it seems to, yeah [17:57] *so [17:57] bschaefer: thanks a lot . I'll come back if necessary. [17:57] oh lookie here [17:57] devnewbee, np! and good luck! [17:57] https://bugzilla.gnome.org/show_bug.cgi?id=682331 [17:57] Gnome bug 682331 in gapplication "GApplication in Garbage Collected environments would benefit from a g_application_command_line_exit() to enable remote instances to exit." [Normal,Unconfirmed] [17:57] bye [17:57] larsu: GtkApplication is not the boogeyman I thought it was after all [17:58] bschaefer: yea . bye. [17:58] conscioususer, yeah it's pretty cool once you get out of the libunique thinking :) [17:58] another bug to subscribe to... [17:58] * desrt feels like fixing this bug today [17:59] desrt, also: why is command-line called for apps that have HANDLES_COMMAND_LINE set but no arguments are given? [17:59] larsu: ok, once that gmenumodel thing is fixed, polly gtk3 will be again on the right track :) [18:00] larsu: because no arguments is just a special case of n arguments? [18:00] would be quite odd to imagine the opposite, i think [18:01] desrt, larsu: thanks a lot to you both, I can go on now and I learned a lot :) [18:01] conscioususer, I don't have time anymore to fix it today, I'll try to fix it tomorrow morning. Also the mnemonic thing, that's just a missing call to gtk_label_set_use_underline [18:01] desrt, usually you call activate when no args are given... [18:02] larsu: or open when args are given [18:02] yeah [18:02] commandline is the way of saying "i want to deal with this" [18:02] hm, okay [18:02] larsu: take your time :) [18:03] all of this magic is in the default implementation of local_command_line [18:03] you can override that and do whatever you want [18:03] HANDLES_COMMAND_LINE doesn't even have any meaning outside of that function [18:03] larsu: btw, if it's just such a harmless change, maybe it could qualify for a precise SRU? [18:04] conscioususer, I'm not on the release team, but we can certainly try. Note that this only really effects you as a developer: most users won't stop and restart polly in < 5 seconds [18:04] larsu: oh, I was referring to the mnemonic thing [18:05] larsu: the 5 seconds thing, I have no idea if the fix seems harmless or not :) [18:05] conscioususer, oh! Ya, that's a fix in gtk... that's much harder to get a SRU for [18:05] actually, I could do the fix in indicator-appmenu for now... [18:11] hmmmmmmm [18:11] this bug is difficult to fix [18:13] larsu, do you call me hard to work with there?! [18:13] seb128, never! :) [18:13] yeah yeah [18:14] but yeah, the less GTK uploads to do the happier I am ;-) [18:14] seb128, I know, I just didn't want to get conscioususer's hopes up before talking to you ;) === dandrader|afk is now known as dandrader [18:16] didrocks, argh, which one??? https://bugs.launchpad.net/ubuntu/+source/bamf/+bug/718926/comments/26 [18:16] Ubuntu bug 718926 in indicator-appmenu (Ubuntu) "Some apps don't integrate to appmenu after having their windows closed" [High,Fix released] [18:16] * larsu wonders if this is still an issue in bamf. the workaround in indicator-appmenu is buggy [18:16] mardy, poke about building unity-lens-gdocs. I'm hitting an error like "xgettext: error while opening "../unity-scope-gdocs.application.in.h" for reading: No such file or directory" Have you seen that before? [18:17] larsu: it was fixed shortly after, hence the no reference to bug # IIRC [18:17] larsu: but with every info old from 7 month, my memory can turn ugly ;) [18:17] didrocks, cool, thanks! (I'm fine with you not writing down bug numbers if you remember all of them ;) ) [18:19] seb128, is a indicator-appmenu SRU for this kind of stuff okay? [18:20] larsu, is that a real world issue or just a "if close and run it again in the next 5 seconds" [18:20] seb128, it's even better: close it and run it again in the next 5 seconds *and* have X give it the same xid [18:20] (which happens more often than one might think) [18:21] but yeah, I admit it's quite an edge case [18:21] larsu, I wouldn't bother SRUing that alone in a nonLTS [18:21] larsu, it might be good to batch with some extra fixes when we get some [18:22] seb128, right [18:29] Another fix: https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix-1075578-workspacenames-flickering-during-display/+merge/133124 === popey_ is now known as popey === salem_ is now known as _salem === mhall119_ is now known as mhall119