=== Mirv_ is now known as Mirv [07:12] guys, the electrical company said they might cut my power today from 8 to 12, still hasn't happened but i might lose my dsl connection randomly, so don't get weirded if that happens :D [07:38] anyone here ran the shell's autopilot tests on the phone lately? [07:45] not me [08:02] Saviq: ping [08:02] tsdgeos, ping [08:02] tsdgeos, pong, rather [08:03] mzanetti, any idea why jenkins didn't pick up https://code.launchpad.net/~paulliu/unity/wrap-and-sort/+merge/165829 ? [08:04] mzanetti, is it the "just packaging changed" case where it goes fast-track? [08:07] Saviq: hmm... seems I can't reach jenkins at all [08:07] mmrazik: Anything known? ^ [08:07] Trevinho: hey, around? [08:07] mzanetti: probably related to the e-mail veebers sent? [08:07] mmrazik: likely, yes [08:08] I can't reach it too, btw [08:08] sil2100: hey! how are you? [08:08] Saviq: ok... seems the whole lab has network troubles. ticket is reported. [08:08] mzanetti, k thanks [08:09] didrocks: hello! Trying to connect to the VPN right now ;) [08:09] sil2100: no need, jenkins is down (not only jenkins, a lot of the QA lab) [08:10] :| [08:10] sil2100: one of the issue is that the admins are away for a week, so the manager will surely get phoned today :) [08:10] sil2100: anyway, meanwhile, mind pushing that to get fixed? bug #1054049. Seems a popular issue [08:10] bug 1054049 in compiz (Ubuntu) "compiz crashed with SIGSEGV in __normal_iterator() from begin() from functionSetEnabled() from glPaintOutputSetEnabled() from GridScreen::handleEvent()" [High,Confirmed] https://launchpad.net/bugs/1054049 [08:10] sil2100: also, you can work on python-evdev (get it to ubuntu today ;)) [08:13] didrocks: aye! Will do ;) [08:13] thx [08:28] didrocks: hey [08:30] hey Trevinho, see the ping to sil2100 3 lines above ^ :) [08:32] oh python-evdev.. Nice, I needed that for some hacking :P [08:32] didrocks: checking that bug, then [08:33] thanks :) [08:43] sil2100: we can probably mute that crash by checking for glScreen validity, but it doesn't seem the good fix to me... glScreen must be non-null there [08:45] Trevinho: hmm, what could cause it to be NULL then? [08:45] Any ideas? [08:46] sil2100: in theory an unproper glscreen? so maybe wrong initialization of the gl plugin [08:47] sil2100: but it should be present in other plugins also... So, or it's initialized after, or no other plugin is using it [08:47] sil2100: if unity is running there, however, it doesnt' seem the case [08:47] i,e. if it's crashing only after opening the expo [08:49] no compiz log to the bug, though [08:49] Indeed, the bug also doesn't really mention when the crash happens exactly [09:08] tsdgeos, still here? [09:15] Saviq: yep :-) [09:16] ;) [09:40] Saviq: hey, you invited me on a video call earlier this morning. What for? [09:41] greyback, mom [09:44] greyback, we had a chat with tsdgeos about the LVWPH [09:44] Saviq: ah ok [09:44] greyback, but we managed without you, slacker [09:44] Saviq: I'm sure you did. [09:45] * greyback had a looong weekend [09:45] greyback, the issue tsdgeos stumbled upon was something that we forgot to take into account when thinking of LVWPH with destroyed delegates [09:46] greyback, when you scroll back up, the previous category delegate was destroyed before [09:47] greyback, and it gets recreated when you get near it... and that's when the content you have on screen is pushed down, 'cause the contentHeight of the Flickable changes [09:48] greyback, but we were not syncing the originY from the ListView to the Flickable [09:48] greyback, and anyway with each such issue we're getting closer to say "let's just write it ourselves based on a Flickable" [09:48] greyback, why the long weekend? had fun? [09:49] Saviq: true. I struggled to get both the contentY and originY to update the Flickable's contentY correctly, It seemed in 1 frame originY changed, then the next contentY changed to offset. I never did understand why [09:49] Saviq: stag party for a good friend, in the west of Ireland [09:49] Saviq: ok, confirmed, growing items in a listview outside the viewport don't seem to influence it's positioning *but* it's hard/impossible to guarantee they'll be outside the viewport, since you can flick very fast [09:50] i.e. take this https://pastebin.canonical.com/91636/ if you scroll slowly all is cool [09:50] if you flick fast, things animate in the middle [09:50] but since you're flicking i guess that's probably "fine" [09:51] tsdgeos, yeah, and also - it's not like our items will really change its height [09:52] tsdgeos, there's a behaviour now on them that should be disabled [09:52] Saviq: sure, but the contents may change, no? i mean it may happen that you get updated data while you scroll? [09:52] or not? [09:52] tsdgeos, so I think our case will more be one of "the delegates are of different height" [09:53] tsdgeos, sure, but that's fine [09:53] tsdgeos, and anyway 99% of the time the resulting height will be the same [09:53] tsdgeos, at least for the Dash [09:53] Trevinho: any progress on that bug? Maybe smspillaz could be of some help :) ? [09:53] (if Sam has some free time of course!) [10:04] tsdgeos, and IIUC the ListView will cache the dimensions of destroyed delegates (I would) [10:04] Saviq: it doesn't [10:04] all it has is a averageDelegateSize [10:04] which is part of the root of all evil [10:05] tsdgeos, yeah, should be updated to a list of real delegate sizes when you know them, IMO [10:05] but the averageDelegateSize is probably needed anyway [10:05] I hate ghosts... http://ubuntuone.com/5MjoQ6taFZyP85DPOreg2H [10:05] thing is, nothing guarantees the delegate sizes will be the same the next time you instantiate it [10:06] tsdgeos, of course, but it's still a better guess than average size [10:06] biab, rebooting to try and get rid of the ghosts [10:10] not good, my touchpad seems to be dying :/ [10:12] Saviq: oh no. [10:13] greyback, got 4 years door-to-door warranty, so it's gonna be ok, good thing I can disable it for the time being [10:13] actually, not d-t-d, serviceman-to-door [10:14] Saviq: fair enough, just annoying [10:15] might as well use the fancy warranty [10:26] wow... http://bazaar.launchpad.net/~unity-team/unity/phablet/revision/695 [10:27] that was quick... [10:27] Saviq: you just turn it on? [10:27] greyback, Friday [10:27] Saviq: nice [10:28] no french [10:30] or chinese or swedish, or burmese. But not bad [10:31] greyback, Friday [10:31] granted, there's like 14 translatables there ;) [10:32] but still :) [10:37] Saviq: are we going to have changes like http://bazaar.launchpad.net/~unity-team/unity/phablet/revision/695/po/pl.po all the time? [10:38] tsdgeos, no [10:38] tsdgeos, this was just on-enabled [10:38] good :-) [10:38] someone disagrees big time with paulliu's translation :D [10:39] Saviq: I thought the LP engine went through the huge list of strings and translations that were already defined in the past, to generate that on first run. [10:40] greyback, tsdgeos yeah, it might be the "share translations within the same project" effect [10:41] or not ;) [10:41] yep, no evidence either way [10:41] tsdgeos, but the translations are restricted, so only "Unity translation team" has access [10:41] tsdgeos, so that *should* be fine ;) [10:41] "Ubuntu Tranlators", rather [10:42] see https://translations.launchpad.net/unity/phablet/+translations [10:43] sure [10:43] sil2100: not really :( [10:49] Saviq, ping [10:50] tvoss, pong [10:55] mzanetti, re: syncevolution / CardDAV, did you manage to sync Harmattan? [10:55] Saviq: yes [10:55] I've had some progress with SyncML and funambol, but it was painful and flaky [10:55] Saviq: also syncevolution seems a bit flaky on harmattan [10:56] Saviq: seems to work way more stable on ubuntu phone already [10:56] mzanetti, and what do you use on your desktop? [10:56] akonadi [10:56] right :P [10:56] I wonder if there's CardDAV support for Tbird [10:56] I think there is [10:57] Saviq: http://davmail.sourceforge.net/thunderbirdcarddavsetup.html [10:57] * Saviq will have to try [11:00] mzanetti, did you build syncev yourself for harmattan or used the pkgs that are out there somewhere? === MacSlow is now known as MacSlow|lunch [11:00] Saviq: hmmm.. I think some guy posted packages... let me search the thread [11:01] Saviq: this worked for me iirc: http://talk.maemo.org/showthread.php?p=1239368 [11:02] mzanetti, cheers [11:12] Saviq_test, test [11:15] yikes there's IRC and Jabber built into Tbird now... [11:19] tsdgeos: yeah.. I know. But I think he does the translation better than I do. Actually I don't speak Chinese very well. :P [11:20] oki === tvoss is now known as tvoss|lunch === MacSlow|lunch is now known as MacSlow === mzanetti is now known as mzanetti|lunch === tvoss|lunch is now known as tvoss === _salem is now known as salem_ [12:24] sil2100: mind having a look? https://code.launchpad.net/~didrocks/unity/add-autopilot-dep/+merge/165871 [12:36] sil2100: and https://code.launchpad.net/~didrocks/autopilot/remove-circular-dep/+merge/165873 [12:36] thanks :) === mzanetti|lunch is now known as mzanetti [12:58] mzanetti, still having problems with input events? [12:58] dandrader: I didn't really continue on friday... So yes, nothing changed [12:59] did you read you IRC backlog? [13:00] dandrader: yes [13:00] dandrader: but didn't really understand it [13:00] mzanetti, short version: sync with latest trunk again as you're using an outdated/problematic MouseTouchAdaptor [13:02] dandrader: oh... [13:02] * mzanetti tries [13:12] greyback, are you making heavy modifications on Stage.qml? [13:13] Saviq: i've been testing the ListView alone and it also has it's fair share of problems with items that animate it's size, as far as i understood you said that this animating is not really expected on our componenets either, i am thinking it'd make sense to fix that first since it's one of the things that makes my life harder [13:13] dandrader: not heavy, just moving the file. I do more damage to Shell.qml though :) [13:13] tsdgeos, of course, disable the behaviour [13:14] dandrader, there's an issue with the launcher [13:14] dandrader, once you drag it down [13:14] dandrader, to scroll to the top items [13:15] dandrader, it will then scroll up there by itself every time you open the launcher === mmrazik is now known as mmrazik|afk [13:17] Saviq: arggg, running ./build gives me a non-empty diff, is someone working on that? [13:17] tsdgeos, not sure what you mean? [13:17] Saviq: all the .po files get scrambled [13:18] http://paste.ubuntu.com/5706763/ [13:18] tsdgeos, weird, it should only happen on make translations :/ [13:19] can you repro that? [13:19] tsdgeos, I get the same [13:19] that's a pain :-/ [13:19] hmm.. seems that we should avoid make translation on build? [13:20] well, we should avoid things moving inside the file for no reason :D [13:20] if you look the contents of the file are the same [13:20] almost the same [13:20] they just moved around [13:21] tsdgeos, http://pastebin.ubuntu.com/5706780/ [13:21] paulliu, ^ [13:21] Saviq, reproduced the issue [13:21] dandrader, will you try and fix? [13:22] mzanetti, should I fix that Launcher issue or is it your turf now? [13:22] Saviq: but the only thing this makes is not adding it to the regular make, no? [13:22] is that what we want? [13:22] tsdgeos, yes [13:22] oki [13:22] tsdgeos, we should only update the .pot file from now on [13:22] hmm wait [13:23] BTW, in DashBar.qml:98 we are using lens.name to get the icon. But now the lens.name is got translated. So it cannot find the icon. There's a comment there that we should use lens.iconHint.. [13:23] tsdgeos, that would mean they're not built on install [13:23] paulliu: that's not new, been hapeening here for a long time :D [13:23] What's it exactly? Should I just add a iconHint there? [13:23] Saviq: what we need is "compile but not merge" [13:24] not sure if that can be done :D [13:24] tsdgeos, indeed [13:24] with that command [13:24] i mean [13:25] Let me see. There are 3 macros in FindGettext. The other 2 should do the trick. I'll test it. [13:25] paulliu, thanks === qengho_ is now known as qengho [13:30] dandrader: depends how critical it is... my branch might need a bit to land still [13:31] mzanetti, paulliu standup? [13:31] mzanetti, I think it's not critical as it doesn't block any use case. [13:32] d'oh [13:32] Saviq: I was doing it! [13:32] dandrader: a first try after merging trunk: it _seems_ to be fixed [13:32] greyback, :P [13:32] greyback, here you go [13:32] mzanetti, good [13:33] dandrader: will report long term results [13:33] greyback, can you hear us [13:33] ? [13:34] Saviq: I can barely hear you [13:34] Saviq: we can't hear you [13:34] but I hear everyone else fine [13:37] tsdgeos, I think you're a good candidate to review this: https://code.launchpad.net/~dandrader/unity/phablet_autopilotTouchOnly/+merge/165666 [13:39] paulliu: I couldn't quite hear you, can you give me a summary? [13:39] greyback: yeah.. I'll add that on notes. [13:39] paulliu: thanks [13:43] mzanetti, Saviq, so, should I start working on fix that Launcher issue? [13:43] dandrader: ok, i'll give it a shot [13:43] dandrader, yes please [13:43] ok [13:43] dandrader: its most likely just a wrong height anyways... should be easy [13:43] mzanetti, yes, I hope so [13:43] :) [13:44] tsdgeos, thanks! [13:45] tsdgeos, btw, remember that issue that sometimes the launcher gets displace some dozens of pixels to the right? [13:45] yep [13:45] tsdgeos, it happened in my machine if I do a drag using a single move event (straight to its final position) [13:45] ouch [13:46] tsdgeos, if I do it with a couple of intermediate moves, that issues doesn't happen [13:46] interesting [13:46] well at least we know it''s not the machine going crazy D: [13:46] tsdgeos, so that's reflected in some of my changes in this merge proposal [13:47] ok [13:48] Saviq: https://code.launchpad.net/~aacid/unity/doNotAnimateFilterGrid/+merge/165886 [13:48] dandrader: CI not liking the change? [13:48] sil2100: hum, still not around? needing help for anything? :) [13:49] cyphermox: ping [13:50] mzanetti: pong [13:50] cyphermox: hey. you still maintaining the syncevolution package? [13:50] tsdgeos, it needs some modifications (like a "chmod 666 /dev/autopilot-uinput"). [13:51] didrocks: I'm here ;) [13:51] mzanetti: I never really did, it was a drive-by fix [13:51] didrocks: ah shit! Missed some pings ;)! [13:51] mzanetti: anything I can help with? sponsoring? [13:51] cyphermox: ah... that one with the syncevo-local-sync binary missing, right? [13:51] * sil2100 scrolls back [13:51] mzanetti:I can't remember tbg [13:51] *tbh [13:51] cyphermox: that issue seems back again... http://notyetthere.org/ [13:51] I'm quite busy today too :/ [13:52] didrocks: that's the problem when I leave my irssi opened on this window, I don't see the notifications when someone pings me ;) Those get lost when there's a lot of talking on the channel [13:52] didrocks: looking! [13:52] cyphermox: no worries... I just thought you would be the package owner... [13:52] thanks [13:52] mzanetti: I'll look at it now, I can start NM long enough ;) [13:53] cyphermox: basically the fix is to add "usr/lib/syncevolution/syncevo-local-sync" to debian/syncevolution-libs.install [13:54] dandrader: oh, so the launcher thing, what you did is workaround it instead of fixing it in the launcher itself? [13:54] tsdgeos, what launcher thing? [13:55] dandrader: "the launcher gets displace some dozens of pixels to the right?" [13:55] tsdgeos, yes. I added a better emulation of a real drag [13:56] but this means that the problem could theoretically still happen, no? [13:56] tsdgeos, I've no idea how this single-move-event drag causes it [13:56] didrocks: are the mergers still down as well? [13:57] sil2100: seems so, as it's part of the lexington lab [13:57] tsdgeos, yes, if you do a drag using a single event it will [13:57] dandrader: should we write that as a blueprint item for furhter investigation in the future? [13:58] tsdgeos, like a light-weight bug report. yeah... [13:59] dandrader: oki, please add taht somewhere then :D [13:59] i guess we have a launcher blueprint [14:01] tsdgeos, why me! ?:) [14:07] dandrader: because you found it and workarounded it [14:07] and you know what has to be written :D [14:07] dandrader: i'm not saying that you assign it to you, just that you write the item === mmrazik|afk is now known as mmrazik === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader [14:43] sil2100: the branch you approved didn't land: https://code.launchpad.net/~elopio/unity/fix1181677-execute_action_by_id/+merge/164608 [14:43] but the errors don't seem related to my change. Should I just approve again? [14:44] elopio: let me check that [14:45] elopio: heh, yes indeed, those are some random failures - I'll re-approve it [14:45] thanks sil2100! [14:45] elopio: it'll take some time until it gets merged again as the merger is down currently [14:45] that [14:46] 's fine, I can wait. [14:46] tsdgeos, jenkins down [14:46] tsdgeos, so no CI [14:46] Saviq: dandrader's one went through CI (if you are refering to that) === mmrazik is now known as mmrazik|afk [14:47] tsdgeos, no, lp:~aacid/unity/doNotAnimateFilterGrid [14:47] ah, ok [14:51] bregma: Trevinho: sil2100: do you see some weird labels on alt-tab with Unity 7? [14:52] didrocks: no, why? [14:52] didrocks: or better, what you mean by weird exactly? [14:52] look as expected on all my machines [14:53] sometimes, can't be sure about the reproducer, but I have the (1) new email label on thunderbird [14:53] let's say thunberbird is the 3rd app in alt tab [14:53] I alt tab to another app [14:53] thunderbird is not anymore the 3rd app [14:53] but the label stays (and behind the icons?) [14:54] then, I just alt-tab to other apps [14:54] and the label always stays on this 3rd app (and behind) [14:54] never got that? [14:54] didrocks: hm, didn't see that before [14:55] * bregma upgrades his Saucy machine to today's latest to check there [14:55] bregma: I got it twice on raring [14:55] so doesn't seem to be saucy specific [14:55] and now, I changed my ws [14:56] so it resetted, can't recordmydesktop [15:00] Saviq, mzanetti: there you go https://code.launchpad.net/~dandrader/unity/phablet_fixLauncherTopScroll/+merge/165896 [15:00] dandrader, cheers [15:01] thanks dandrader [15:04] Saviq, mzanetti: ops, gotta fix one small detail still [15:05] it should scroll back to have the home button visible again on the next appearance === greyback is now known as greyback|food [15:08] mzanetti: I uploaded syncevolution to saucy btw [15:09] cyphermox: the same version as in raring or the fixed one? [15:09] well, fixing the missing binary [15:09] cyphermox: awesome! thanks a bunch [15:10] we should SRU this too, but I can't do it now [15:10] as soon as it looks more like my code is working, I'll do the SRU too ;) === tvoss is now known as tvoss|test === tvoss|test is now known as tvoss [15:13] ok, now it's really good-to-go [15:14] dandrader, no worries, we son't be merging until jenkins is back anyway :/ [15:14] s/son't/won't [15:15] Saviq, speaking of Jenkins. Will it be simple to have it running our autopilot tests using touches? [15:15] that modification I mentioned on the e-mail [15:16] dandrader, yeah, that should be easy [15:16] dandrader, it's just about adding the testing user to the right group [15:29] Saviq: https://code.launchpad.net/~paulliu/unity/i18n-po-msgmerge/+merge/165899 [15:29] Saviq: I found that FindGettext sucks. [15:29] paulliu, ;) [15:29] Saviq: So just copy the rules from it and use it. Now we don't need to make install. [15:30] paulliu, you should remove the if(isRunningInstalled()) from main.cpp, then, right? [15:30] Saviq: There are 3 macros in FindGettext, the first and second do msgmerge on build which we don't want it. The 3rd one works, but it install the mo files base on LANG, not project name. [15:31] Saviq: ok.. So we want to call bindtextdomain on all cases? [15:33] Saviq: pushed. [15:33] paulliu, yeah I think it's fine, at least for now [15:33] paulliu, cheers [15:36] Saviq, I was thinking: when the user presses close to the right edge of an application, should we really hint that you can switch apps? [15:36] Saviq, ie.: should we really slide in the next app a bit? [15:36] dandrader, we probably should (after a delay) [15:36] dandrader, but no, not by sliding the next app in [15:36] dandrader, rather via a glow [15:37] Saviq, won't that interfere with a long tap for instance? [15:37] dandrader, yeah, but the edge drag should ignore a long tap like this [15:37] dandrader, that's just me thinking, mind you, [15:37] dandrader, IANAD (IANADesigner) [15:38] Saviq, like its adding the distractions that we want to avoid by properly detecting a directional drag before doing the thing [15:38] dandrader, sure, I think that's the place for "let's learn with our users" [15:38] dandrader, i.e. hint it until we recognize a correct edge swipe for 10 times or something [15:39] dandrader, potentially reducing the strength of the hint over time [15:39] dandrader, but again, IANAD [15:39] dandrader, just repeating some things I've heard proposed [15:40] Saviq, ok. well, I'll leave that slide-in hint for now then === greyback|food is now known as greyback [16:17] dandrader: can you point me to your branch on your Stage changes, so I can see what you're doing and minimize conflicts [16:18] greyback, lp:~dandrader/unity/phablet_edgeDragInStage [16:19] dandrader: thanks === dandrader is now known as dandrader|lunch === dandrader|lunch is now known as dandrader === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === popey_ is now known as popey === salem_ is now known as _salem === ajmitch_ is now known as ajmitch === _salem is now known as salem_