=== tvoss_ is now known as tvoss [06:28] mmrazik: not sure how it works, but maybe the publication to the public jenkins creates fake "jobs" to attach artefacts and so on [06:29] (but that's really a supposition) [06:29] didrocks: the jobs are quite real... AFAICT everything is just copied to the same directory structure as on the real jenkins [06:29] ok [07:10] mzanetti: you here? [07:33] tsdgeos: hey [07:33] paulliu, ping [07:34] mzanetti: do you know anything about ListViewWithPageHeader.clipListView ? It is only used in Dash/DashVideos.qml and somehow annotate pointed me your way [07:35] Saviq: hi [07:35] paulliu, can you please fast-track https://code.launchpad.net/~saviq/unity/8.i18n-updates/+merge/168242 :) [07:35] paulliu, should be a quick review [07:35] Saviq: yes.. [07:36] Saviq: wait for a second. [07:37] tsdgeos: let me check, one moment [07:40] tsdgeos: ok... so why exactly there is this clipping - I don't know. Maybe it improves performance a bit? The reason I used (disabled) it in DasVideo is because when the preview opens, the OpenEffect scrolls the view a bit upwards and with clipping there would be an empty space at the bottom [07:41] i did remove that line and couldn't see any difference [07:41] let me try again [07:41] mzanetti: is this on "phone" mode or "tablet" mode? [07:42] tsdgeos, AFAIR it's clipping of the internal ListView, right? otherwise there can be a case when its content is behind the PageHeader [07:43] tsdgeos: shouldn't matter I think... However, it differs a bit on where you expand the the preview. try the elements on top/bottom. Also, it might have become useless with the integration of the carousel (as that one changes the expanding position in comparison to the grid) [07:44] mzanetti, tsdgeos we need to be able to pull stuff "up" from the bottom [07:44] garg, unity-lens-mock crashing all the time, can't get the videos lens on the desktop anymore :-/ [07:44] tsdgeos, you upgraded to smart scopes? [07:44] maybe [07:44] i distugraded [07:44] so if it was there [07:44] yes [07:44] tsdgeos, yup [07:45] mzanetti, tsdgeos [...] since when you tap on some content that is only a bit on screen, it's pulled up [07:46] ok, give me a few mins until the phone is flashed and i can try to see the differences [07:46] i'll be back :D [07:46] mzanetti, tsdgeos and the clipping (and disabling of it) is there exactly for that reason [07:46] tsdgeos, and it's only used in People and Videos, because we don't have previews anywhere else [07:52] paulliu, thanks === hikiko is now known as Guest71645 === hikiko_ is now known as hikiko [08:10] mzanetti, ping [08:10] Saviq: pong [08:10] mzanetti, hey, I just thought of one thing - will the API you proposed for the launcher (sorry, didn't make it yet to review) [08:10] mzanetti, support the 'app is being installed' case? [08:11] mzanetti, when the icon shows up on the launcher with the progress bar showing installation progress? [08:11] mzanetti, I mean, it doesn't have to be built in already, just that we have it in mind [08:11] Saviq: well, there is a progress thing (which afaik is not only used while installing) [08:11] +1 [08:12] Saviq: but there is no specific is-being-installed flag [08:12] Saviq: I guess that could be easily extended through a model role tho [08:12] mzanetti, I don't think it's needed, actually, but please have a look what's needed (like the item gets highlighted after installation) [08:13] and what's there in quicklists in that case [08:13] Saviq: ok [08:18] Saviq: you ran my lvwph branch on the phone, right? [08:18] somehow it's segfaulting for me :-/ [08:18] tsdgeos, no, it failed to build [08:18] ah [08:19] tsdgeos, the qreal vs. double thing [08:19] tsdgeos, can try again [08:19] Saviq: ok, i fixed that [08:19] Saviq: btw... was hiking on Fri and suddendly I felt some cozy warm feeling in my pocked. Pulled out the phone and it was quite warm. Ran the terminal app and top and - Damn! its qml-phone-shell! [08:19] mzanetti, it's most probably related to wifi [08:19] mzanetti, or the lack of it [08:20] Saviq: so its already known? [08:20] mzanetti, yes, unfortunately [08:20] mzanetti, bug #1183065 [08:20] bug 1183065 in touch-preview-images "Occassional severe battery drain" [High,New] https://launchpad.net/bugs/1183065 [08:21] Saviq: cheers [08:26] didrocks: my unity-panel-service is crashing continuosly, is that a known bug? [08:27] didrocks: maybe I've still some packages from the unity-next PPA (which I disabled a few days ago)? [08:27] mardy: not really, do you have indicator-network by any chance? [08:28] didrocks: ahh!! Yes, that's from the unity-next PPA; I'll downgrade it [08:28] mardy: remove it :) [08:28] didrocks: ah, OK [08:28] mardy: see my email on ubuntu-devel (from Thursday?) [08:28] didrocks: BTW, is there a way to list all packages installed from a certain PPA? [08:29] mardy: not that I know of. At this point, you should be able to purge the next PPA though [08:29] mardy: ppa-purge does that [08:29] mardy: it does it by listing all packages with version after apt-get update [08:29] then remove the ppa [08:29] apt-get update [08:30] and making the diff :p [08:34] pstolowski: hey! seems alt + f2 is broken (again), mind fixing it? :) [08:35] didrocks: hey! dammit, i'm sure it worked last time I tried... what exactly broke? [08:35] Saviq: do you think you and kgunn will be available for a hangout later today (once kgunn is awake) for speaking about Mir and Unity 8 in saucy? [08:35] Saviq, ping [08:35] tvoss: me first! [08:35] didrocks, kgunn's on vacation [08:35] :) [08:35] didrocks, until next week [08:36] Saviq: ah, so just the 2 of us? :p [08:36] didrocks, sure [08:36] tvoss, ping-pong [08:36] didrocks, :) [08:36] Saviq: tell me when you have time, I've a flexible agenda today :) [08:37] didrocks: upgrading to see.. [08:37] pstolowski: it seems to not work only at the first alt+F2 [08:37] pstolowski: then, it's fine here [08:38] but the preview doesn't have any button [08:38] to launch them [08:38] I'm surprised seb128 stopped using alt+f2 btw :p [08:38] didrocks, it works for me [08:38] seb128: even the preview? [08:39] didrocks, https://bugs.launchpad.net/unity-lens-applications/+bug/1188705 you mean ? [08:39] Launchpad bug 1188705 in unity-lens-applications (Ubuntu) "command lens doesn't let you run a command from the preview screen" [Undecided,New] [08:39] seb128: right [08:39] didrocks, thanks for confirming my bug report ;-) [08:39] seb128: not sure I'll confirm, just to annoy you :p [08:39] seb128: thanks for opening it! [08:39] roooh [08:39] pstolowski: ^ [08:39] yw ;-) [08:39] didrocks, so for the record, I didn't stop using it :p [08:40] seb128: I can see that! :) [08:40] didrocks, seb128: right. I suspect this applies to 12.10 and 13.04 as well, let me check [08:40] pstolowski: well, 12.10 is less impacted as the preview isn't the default on single click [08:40] Saviq: is this what we want now? https://code.launchpad.net/~mterry/unity/phablet-narrowMode/+merge/167552 [08:40] pstolowski, well, left click was "run" in those version [08:40] but yeah, should impact there as well [08:41] seb128: sure. now double-click is run ;) [08:41] btw is there any shortcut to run instead of opening the preview? [08:41] oh, double click [08:41] I tried middle and right click [08:41] (pstolowski: you won an assignement) [08:41] a lot of people don't realize about the double-click [08:41] see the french forum :p [08:41] it could be nice to add middle click [08:42] seb128: +1 [08:42] didrocks: I know, I know. I've read on omg or somewhere else it's considered a bug [08:43] * sil2100 hates the double-click to run [08:43] I'll let design team handle these discussions ;) [08:43] I'm a hater, I know [08:43] pstolowski: sure, I just think we should have alt + F2 be coherent at least :) [08:43] sil2100: you really are! :) [08:43] didrocks: no disagreement on this :) [08:44] Saviq: did you get to repro the crash on the phone? [08:44] tsdgeos, didn't get to it yet [08:44] tsdgeos, will be there in 5 [08:44] oki [08:46] Saviq: ok, found the problem [08:47] tsdgeos, ah good [08:47] Saviq: run_on_device rsync is not removing the old ListViewWithPageHeader.qml and confusing the hell of everything [08:48] now i just need to found why everything is empty! :S [08:48] tsdgeos, yeah, we should add --delete [08:51] mzanetti, dunno, do we? :) [08:51] ? [08:51] mzanetti, narrowMode [08:51] mzanetti, it's you who proposed that, I'm good either way [08:51] Saviq: --delete doesn't seem the best thing, need to compile everything again :D [08:51] mzanetti, --exclude builddir [08:51] tsdgeos, ↑ [08:52] -mzanetti [08:57] tsdgeos, seems to work fine here [08:57] Saviq: actually I proposed something different. I proposed to fix the issue properly but neither design nor Michael seem to want to address this issue right now so he came up with this workaround. Hence the question if you're fine with that [08:58] tsdgeos, the swiping velocity is limited, right? [08:58] hmmm [08:58] shouldn't [08:58] brrr [08:58] go to the music one [08:58] it'll break crazy :-/ [08:59] Saviq: anyways... I'll approve it for now [08:59] tsdgeos, works fine... [08:59] mzanetti, if it solves a problem, do [09:00] Saviq: scroll to the end, then drag a lot to the top and release [09:00] tsdgeos, not seeing anything wrong [09:01] tsdgeos, and if you want to see why the clipListView - go to videos, drag so that Les Miserables is barely visible, and tap on them [09:02] it jumps to the previous "category" here [09:02] wonder why you don't see it [09:03] tsdgeos, ah yeah, just got the jump [09:05] tsdgeos, it also seems to stick to the top when I show the carousel just a bit [09:06] Saviq: meaning you can't scroll down? i think that has to do with the jump and the list getting confused at how height it is [09:07] tsdgeos, I can, it just jumps to the top [09:07] i see [09:07] well, let me fix the one i can repro first [09:07] and then fix yours :D [09:07] :) [09:10] and yes [09:10] somehow the speed seems "weird" [09:11] tsdgeos, http://qt-project.org/doc/qt-5.0/qtquick/qml-qtquick2-flickable.html#maximumFlickVelocity-prop [09:12] Saviq: sure, but we weren't setting before nor now [09:12] so should not have changed [09:12] tsdgeos, I believe we were [09:12] /are [09:12] tsdgeos, LVWPH.qml:56 [09:12] 156 [09:14] oh [09:14] :D [09:34] Mirv: hum, so, did you look at the platform stack? is it good for me to look at the packaging change? [09:44] Saviq: ok, i fixed the jumpiness at the bottom, can you tell me how to reproduce the problem you mentioned with the carousel? [09:45] tsdgeos, drag all the way down, drag back up so that just a bit of the carousel is visible, release [09:45] tsdgeos, for me it made it jump to the top [09:47] Saviq: can you try with the new fix, maybe it also makes a difference? [09:47] can't repro [09:47] tsdgeos, yeah, will do [09:47] MacSlow, hey [09:48] tsdgeos, dude, you broke git-bzr-ng with the ++ in your branch name ;) [09:48] :D [09:48] hah! [09:48] could be worse [09:48] yesterday there was a cve on subversion on repos being destroyed because of newlines in the filenames :D [09:52] nice === mzanetti is now known as mzanetti|lunch [09:58] tsdgeos, yeah, can't reproduce [09:59] cool :-) [09:59] seb128, hey seb [09:59] MacSlow, hey, how are you? [09:59] seb128, busy as usual :) [10:00] MacSlow, quick notify-osd question, can Bubble objects have a refcount > 1 ? [10:00] seb128, they should not iirc [10:01] MacSlow, stack_allocate_slot() has a [10:01] if (stack_is_slot_vacant (self, slot)) [10:01] self->slots[slot] = BUBBLE (g_object_ref ((gpointer) bubble)); [10:01] so I was wondering [10:02] seb128, oh... that slot-allocation stuff... [10:02] didrocks: hmm, I reviewed sil2100's packaging, but I understood he'd be doing the configuration changes if you now mean adding the new packages to the stack? [10:03] Mirv: no, I mean about the platform stack :) [10:03] the platform stack itself was fine, pending on the QA stack only, although now I can't seem to be able to connect to jenins [10:03] Mirv: hum, I can connect here [10:04] Mirv: I'm seeing a packaging change due to powerd new package [10:06] MacSlow, that code seems complex... I was just trying the segfault which is happening for quite some users since saucy, the issue is that IS_OBJECT on an unrefed object segfaults ... and stack_layout() unref an object without setting the pointer to NULL [10:06] didrocks: ok, I can't for some reason or another. but you asked if it's good for you to look at, yes feel free to look at :) but if something needs to be done to powerd, just tell me what [10:07] tsdgeos, there's some slight hiccups when the delegates are created, otherwise it's working great [10:07] Mirv: no, it seems good to me, the packaging change as well. If you can't access to it, mind if I publish for you? [10:07] Saviq: yeah i noticed that too :-/ [10:07] Saviq: wonder where it's coming from since i'm supposed to be generating them async [10:08] tsdgeos, not sure we'll be able to do anything [10:08] tsdgeos, it might be IO, even [10:08] didrocks: yes, please, and thanks! [10:08] Mirv: doing :) [10:08] Saviq: i'm going to work on getting the section headers up and running first [10:08] Mirv, didrocks: https://code.launchpad.net/~sil2100/python-upa/packaging_review/+merge/168393 [10:09] and then we can have a look at that [10:09] Mirv, didrocks: https://code.launchpad.net/~sil2100/autopilot/get_manual_version_to_trunk/+merge/168385 [10:09] Reviews please! [10:09] tsdgeos, yeah, of course [10:09] * didrocks lets Mirv doing that, I'll rereview when newing python-upa :) [10:09] sil2100: ok, will do [10:09] sil2100: though, don't downgrade the dep on the second on [10:10] sil2100: just backport the changelod I guess to unblock the stack [10:10] didrocks: ok, so I'll revert your switch then [10:10] Leaving the changelog entry intact [10:10] right :) [10:10] it's just a nasty workaround to not have autopilot stuck in -proposed [10:11] tsdgeos, btw, merge with trunk, there's a conflict [10:12] didrocks, Mirv: updated [10:13] thx [10:14] seb128, I'd need to dig myself in there again to remember all of it... [10:14] Saviq: done [10:15] tsdgeos, cheers [10:15] need to restart the session, everything is borked, can't even alt+f4 [10:15] MacSlow, hum, k === om26er is now known as om26er|away [10:24] MacSlow, Trevinho: I've added details to https://bugs.launchpad.net/notify-osd/+bug/1189281 ... is there a chance any of you could look at it? the issue is "simply" that we need to set bubble = NULL after the g_object_unref when the refcount reachs 0 ... not sure how to do the "if ref reachs 0" though, would be easier if bubbles were not refcounted over 1... [10:24] Launchpad bug 1189281 in notify-osd (Ubuntu) "notify-osd sefaults in bubble_get_id() trying to access an unrefed object" [High,Confirmed] [10:25] seb128, one can't easily query the ref-count... [10:25] MacSlow, right, which is why I'm dropping the ball, I don't understand the notify-osd memory handling well enough and I don't want to spend a day on it === mzanetti|lunch is now known as mzanetti [10:37] paulliu, https://code.launchpad.net/~saviq/unity/phablet.raring.i18n-updates/+merge/168397 please :) [10:40] Top-edge swipe on lock-screen is supposed to be able to open the indicator menus, right? [10:40] Saviq: ok..wait [10:40] greyback, yes [10:40] Saviq: ok === MacSlow is now known as MacSlow|lunch === alan_g is now known as alan_g|vt === alan_g|vt is now known as alan_g [11:57] who can test my branch with alltests not running unity? [11:58] this thing keeps crashing in saucy [12:03] didrocks: the 2 HUD failures that we have are regressions, but I think we're being hit by that since a longer time [12:04] greyback: ping [12:04] mzanetti: pong [12:04] greyback: I'm testing your branch [12:04] greyback: have a weird issue [12:05] mzanetti: go on [12:05] sil2100: ok, so pinging ted once he's around? [12:05] greyback: open the phone-app, then open the notes-app. Edit a note. [12:05] greyback: finish editing a note (OSK hides again) [12:05] didrocks: yes, I'm looking for the bug for that one, as I think I filled one before [12:05] greyback: now swipe from the right [12:05] seb128, I still have bug 1181717 [12:05] bug 1181717 in unity (Ubuntu Raring) "compiz crashed with SIGSEGV in g_object_unref() from unity::bamf::Application::~Application()" [Undecided,New] https://launchpad.net/bugs/1181717 [12:05] mzanetti: ok gimme a minute to set up phone [12:05] in saucy [12:06] Cimi, talk to Trevinho [12:06] sil2100: thanks! [12:07] andyrock, hey andrea [12:07] I still have bug 1181717 [12:07] bug 1181717 in unity (Ubuntu Raring) "compiz crashed with SIGSEGV in g_object_unref() from unity::bamf::Application::~Application()" [Undecided,New] https://launchpad.net/bugs/1181717 [12:07] do you have the bt? [12:07] can be a different crash [12:08] 7.0.0daily13.06.07-0ubuntu1 [12:08] I will try to give you a bt [12:08] Cimi, thx [12:09] Cimi, seb128: really? are you on trunk? [12:09] Trevinho, I'm not having the issue, don't look at me ;-) [12:09] seb128: hehe, ok [12:09] Trevinho, andyrock: is there any chance one of you could look at the notify-osd unref issue I pointed earlier? [12:09] mzanetti, reviewed launcher API [12:10] seb128: I've prepared again the MR for gdbus https://code.launchpad.net/~3v1n0/bamf/libbamf-gdbus-round2 [12:10] seb128, maybe i was offline [12:10] Trevinho, great [12:10] andyrock, [12:10] Saviq: thanks... [12:10] mzanetti, can you drop ~mzanetti/unity-api/launcher or at least mark it abandoned? [12:10] seb128, can you post the link again? [12:10] seb128: I've reverted the change that caused the crash as well, do I need distro ok or upstream? [12:10] mzanetti, since we've moved it to ~unity-team [12:10] Saviq: ok [12:10] Trevinho, upstream is fine [12:10] andyrock, http://paste.ubuntu.com/5751417/ [12:10] seb128: ok [12:11] Trevinho, ^ [12:11] 7.0.0daily13.06.07-0ubuntu1 [12:11] Cimi: on trunk? [12:11] 7.0.0daily13.06.07-0ubuntu1 [12:13] it happens running make alltests in my project [12:13] like, everytime [12:14] Cimi, I can give a look but it can takes a couple of hours/days/weeks ;) [12:14] mhr3: is really that expensive to use g_object_add_weak_pointer? Not that is a solution, at all... But looking at gobject code it doesn't look so much... === MacSlow|lunch is now known as MacSlow [12:15] dednick, is it right we cut the diff from over 6k down to 4k for the indicators? [12:15] * Cimi installs KDE [12:15] * Trevinho slaps Cimi [12:16] Trevinho, it's weeks I'm losing time because unity and bamf crash, I have work to do [12:16] Saviq: ? as far as i see, it's 8k [12:16] Cimi: indeed, but it's quite weird,... [12:17] dednick, right [12:17] Trevinho, andyrock: the notify-osd issue is bug #1189281, I did a summary of the problem on the bug, it's basically a variable that is g_object_unref() then code which does IS_OBJECT() on it, which segfaults with the new glib ... not sure we can set it to NULL after unref though since the ref count seem to go > 1 so it only needs to be = NULL if it goes to 0 [12:17] bug 1189281 in notify-osd (Ubuntu) "notify-osd sefaults in bubble_get_id() trying to access an unrefed object" [High,Confirmed] https://launchpad.net/bugs/1189281 [12:17] seb128: yes, I read that... [12:17] Trevinho, if you want to reproduce it, pick up lp:~cimi/indicators-client/system-components [12:17] Trevinho, mkdir build [12:17] cmake .. [12:17] Cimi: ok, thanks.. .I was about to ask that [12:17] make alltests [12:17] after a couple of tests it segfaults [12:18] (couple can be 6-7) [12:18] it's a race, we know it, so it's not like 100% happening [12:18] here happens 100% of the time though, but randomly [12:18] dednick, ah, from 9k lines - still, almost 10% down is good :) [12:19] dednick, /me tries to review [12:19] Saviq: have fun :) I added some comments for a bit of help [12:19] dednick, yeah, saw that [12:21] Trevinho, maybe this time the problem is in PanelMenuView... [12:21] andyrock: mhmh... it could be, but I think it's managing things correctly... [12:21] andyrock: that code should be ported to BamfApplicationManager though [12:21] Trevinho, can you check better? :D [12:22] k [12:22] Trevinho, I'll give a look too [12:23] Anyone else suspect ninja is causing a slower build on the device? === alan_g is now known as alan_g|lunch [12:29] greyback, there was a missing -j$NUM_JOBS in build [12:29] mzanetti: what is the bug? I don't see anything wrong [12:29] greyback, but should be fine now [12:29] Saviq: ok, will pull [12:29] greyback: it just switches between notes-app and notes-app [12:29] greyback: the phone-app gets lost in the app stack somehow === om26er|away is now known as om26er [12:29] mzanetti, notes-app is special [12:30] mzanetti: not reproing here :( [12:30] mzanetti, I think it does something weird [12:30] hmm... ok... can't reproduce this with todays stock image here [12:30] mzanetti, it shows up twice in running apps, for example [12:30] mzanetti: this on phone or desktop> [12:30] greyback: phone [12:31] mzanetti, it feels like Notes opens two surfaces initially, this might make the wm confused [12:31] mzanetti: I'll try a clean reflash and install my build and check it out. [12:31] Saviq: build failed on jenkins https://code.launchpad.net/~saviq/unity/phablet.raring.i18n-updates/+merge/168397 [12:32] paulliu, yes, the job shouldn't eve run [12:32] even run [12:32] paulliu, that branch is in manual merge mode now [12:32] Saviq: ok. [12:32] paulliu, I'll take care of that when sergiusens comes online [12:32] Saviq: ok. [12:34] Cimi, i get a qt5 error running cmake [12:34] Cimi, http://paste.ubuntu.com/5751489/ [12:38] andyrock: you're probably missing "qtdeclarative5-dev" === _salem is now known as salem_ [12:40] andyrock: if you're building unity8, use first the "build" and then the "run" script [12:40] greyback, ok thank but I'm not building unity8 [12:40] andyrock: ah ok [12:46] my desktop is completely broken after upgrading to saucy (typing from my laptop). even the greeter if f**d up [12:46] s/if/is [12:47] what's the easiest way to get some kind of working desktop? [12:48] dandrader: install another desktop environment. And switch to it when login? [12:49] dandrader, greeter? [12:49] dandrader, mine is mostly usable [12:50] my greeter is missing icons, etc. but yeah, I could say "mostly usable" [12:50] as I can log in from it [12:51] I was going for a saucy chroot but g_settings_new was crashing when running unity8 inside it [12:51] So I just gave up and upgraded [13:02] Trevinho, it is expensive because each gobject wrapper instance will be calling both add_weak_ref and remove_weak_ref === alan_g|lunch is now known as alan_g [13:21] Saviq, thanks for porting my narrowMode branch to unity/8.0 [13:21] mterry, :P [13:22] mterry, you owe me a beer for that! [13:22] the diff was so big [13:22] I'll buy you a tiny beer [13:22] Little thimblefull [13:23] ;D [13:24] Saviq, what's the best way to check from another Qml file if a transition is being performed? [13:24] Not just on startup... [13:25] nic-doffay, just bind to it [13:25] nic-doffay, or, slightly better [13:25] nic-doffay, property alias rotating: helper.rotating in Shell.qml [13:25] nic-doffay, and then shell.rotating [13:29] ok Saviq how do I perform a state change on a variable change then? [13:30] nic-doffay, either use states: [ State { when: shell.rotating }, State { when: !shell.rotating } ] [13:30] nic-doffay, or state: shell.rotating ? "rotating" : "static"; states: [ State { name: "rotating" }, State { name: "static" } ] [13:30] nic-doffay, note, it's pseudo-code, again [13:30] Saviq, cool, the first seems more plausible. [13:31] nic-doffay, make sure to read through http://qt-project.org/doc/qt-5.0/qtquick/qtquick-statesanimations-animations.html [13:33] Saviq: greyback:n ic-doffay [13:33] standup? [13:33] tsdgeos: in other meeting, have to miss it [13:33] tsdgeos, launching mumble now [13:34] mzanetti: yep reproduced, will fix [13:34] greyback: \o/ [13:34] greyback: standup first [13:34] oh.. soryy [13:37] aaargh why don't I get pings about the standup anymore [13:38] Saviq: i did ping you :D [13:38]