[00:13] Saviq, merged from trunk, still looking at keyboard [00:13] mterry, ok, kicking a rebuild [00:16] mterry, looks like we can drop indicator-sound? [00:16] Saviq, yes, and indicator-datetime [00:16] mterry, ugsb? [00:16] Saviq, need it [00:16] Saviq, at least, I don't think there was a release of that? [00:16] mterry, https://code.launchpad.net/~indicator-applet-developers/unity-greeter-session-broadcast/trunk.14.04 [00:17] Saviq, oh awesome, drop it too [00:24] mterry, ok, kicked a rebuild of unity8 and asked robru to clean it up [00:24] /sleep [00:25] Saviq, night! [00:25] let [00:25] 's see what tomorrow brings [00:25] o/ === _salem is now known as salem_ === salem_ is now known as _salem [07:10] Saviq: hey, seems that all unity8 tests are failing due to: StateNotFoundError: Object not found with name 'QQuickLoader' and properties {'scopeId': 'home.scope'}. [07:10] which relates to the scope transition [07:10] Mirv: ^ [07:12] didrocks: I believe it's what robru is explaining, that #258 needs to be built since he kicked #257 too early [07:12] Saviq: cancel, seems that it's the message from robru [07:12] Mirv: snap! [07:12] :) [07:12] :) [07:12] Mirv: building [07:13] good good [07:13] not sure why 258 didn't build (I guess because 257 was still building) [08:24] Saviq: err.. [08:24] mzanetti, whatup? [08:25] Saviq: the scopes have landed? [08:25] I was still testing :D [08:25] mzanetti, image 257 is broken [08:25] mzanetti, oh [08:25] mzanetti, you can continue testing on trunk, then, and file bugs ;) [08:25] Saviq: well, looks mostly good, but I had two questions [08:26] I flashed a clean image and installed the silo ppa. only apps scope appears [08:26] do we have all the required configs, daemons in the image? [08:26] mzanetti, there's an ubuntu-touch update, what version do you have? [08:27] mzanetti, apt-cache policy ubuntu-touch ? [08:27] let me check. flashed yesterday ~5pm [08:27] mzanetti, yeah, but then from silo did you dist-upgrade? [08:27] iirc yes. or at least upgrade and verify nothing is being held back [08:28] ah ok... yep. ubuntu-touch has been kept back, as the only package [08:28] yeah, dist-upgrade [08:28] ok. I guess that solves question nr 1 [08:29] huh? server certificate verification failed [08:29] can't remember what the other thing was right now. It'll come back once I had my coffee [08:30] Saviq: why is the new image broken? because of new scopes have landed or something else? [08:30] mzanetti, it got kicked before new unity8 got published [08:30] mzanetti, 258 is building now [08:30] mzanetti, so we lost the old scopes, but didn't get the new ones ;) [08:31] ah :) [08:33] Saviq: ah right. the perfomance issues are to be solved in a new MR, right? [08:33] mzanetti, yeah [08:34] ok. I think other than that, my testing didn't really reveal any new things, apart from the header breaking in many circumstances [08:34] which we discussed away in favor a new implementation coming up [08:35] yup [08:35] sorry I wasn't faster yesterday [08:35] no worries [08:35] QA ACK'ed it, too [08:35] but AP kills hours and hours without you even realizing [08:36] yeah it dows [08:36] does [08:46] Saviq: so, if you've cleared the silo, can I haz one for right edge? [08:46] mzanetti, ah, you could have had one regardless [08:47] reminds me. arent we in TRAINCON-0 atm? [08:49] mzanetti, on the contrary [08:50] mzanetti, we're blocking promotion on the "events queued when screen off" bug, but landings can flow freely [08:50] there is at least one in main mailbox stating we are. but I guess the amount of heat it caused got it revised [08:51] at least one mail in my mailbox... [08:51] mzanetti, read ubuntu-phone@ [08:51] mzanetti, there was a discussion about it and then it got lifted [08:51] yeah, I followed the discussion, seem I missed the official "it got lifted" mail [08:51] mzanetti, also, you can always read the current status in https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0AuDk72Lpx8U5dFlCc1VzeVZzWmdBZS11WERjdVc3dmc&usp=drive_web#gid=0 [08:56] sil2100, could we have another can-be-flushed silo for row 43 now that we're done with scopes? [08:57] Saviq: hi! Let me check the current status and get back to you in a moment [08:57] sil2100, thanks [08:59] new scopes yayyyyyyyy [09:01] Saviq: should we retry a landing for the new icon theme? [09:01] Saviq: so, basically a branch in the toolkit + your unity8 tweaks [09:01] right? [09:01] didrocks, I wanted that today, yeah, UITK is locked, though [09:01] didrocks, in silo 009 [09:01] yeah, seeing that [09:01] didrocks, they want to land it today [09:01] thostr_: you are going to test that silo soon? ^ [09:02] then, once that's done, just one transition remains on my list before the end of cycle [09:03] didrocks, you mean week? [09:03] didrocks, we still have right edge and split greeter incoming ;) [09:05] Saviq, can i remove support for Unity 0.1? pretty pls? [09:06] mhr3, yeah, we're not rolling back ;) [09:06] \o/ [09:06] * mhr3 loves removing code [09:08] Saviq: here's a tiny one we missed when moving qml files: https://code.launchpad.net/~mzanetti/unity8/fix-commithook/+merge/212475 [09:08] Saviq: but it's only unity8? [09:08] mzanetti, saw that [09:08] didrocks, which one? [09:08] like doesn't need coordination between 10 pieces? [09:08] right edge + split greeter? [09:08] oh no [09:08] you need u-s-c [09:08] didrocks, there's a few projects [09:08] didrocks, not 10, but like 3 or 4 [09:09] didrocks, see the list in silo 004 and the new one in row 43 [09:09] didrocks, although most of that can land separately if needed [09:09] or will, anyway [09:09] Saviq: ok, I gave you 2 transitions then before the end of cycle [09:09] not more! :p [09:09] ;D [09:10] didrocks, isn't every silo a transition? :P [09:10] Saviq: yeah yeah yeah, in some way :) [09:12] Saviq, do you need designers to see split greeter? can put it on one of their devices if needed [09:12] mhr3, not sure it's ready for that, will ask mterry later today [09:13] Saviq, k, limited time offer though only today :) [09:13] :D [09:17] * didrocks will reuse the limitied time offer argument against mhr3 :) [09:18] didrocks, why do you always have to see things you're not supposed to? :P [09:18] mhr3: heh, because I have an hilight on your name to track you down! [09:18] :) [09:18] time for a nick change then :) === mhr3 is now known as mhr4 [09:19] hahaha [09:19] ah, I'm totally lost now! :) [09:19] mhr had an ABI break *again*! [09:19] now you can't talk to me [09:20] yeah, I need a rebuild :p [09:20] i didn't think that through though as you're the one who actually does have all the repo access [09:21] :) [09:23] can apport retrace pkgs from proposed? [09:24] mhr3+1, you can drop the packages from ddebs into the cache [09:25] Saviq, looks familiar? http://paste.ubuntu.com/7150345/ [09:25] Saviq, how do i do that? [09:26] mhr3+1, no, didn't see that before [09:26] mhr3+1, you want to retrace on device or host? [09:26] Saviq, device should be easier [09:26] mhr4: install the qt5 dbgsym [09:26] didrocks, no need for dbgsym [09:26] it unblocks sometimes the stack trace corruption [09:26] qtdeclarative5-dbg [09:27] I saw that with Qt [09:27] and don't ask me the rationale, I don't understand why it results in that and being unrolled once the debug symbols are there [09:27] mhr3+1, on device just grab and install the corresponding package from http://ddebs.ubuntu.com/pool/ [09:28] mhr3+1, get qtdeclarative5-dbg from distro and use apport-retrace -g on the .crash file [09:28] also if you get lots of ??? because of the JIT [09:28] force the interpreter to run [09:28] QV4_FORCE_INTERPRETER=1 [09:29] * Saviq hates it that apport-cli overwrites the .crash and you can't make it process it again [09:33] Saviq, you can, you just need to delete everything after the base64 encoded coredump [09:33] Saviq, http://paste.ubuntu.com/7150382/ [09:33] mhr3+1, I just copy it before apport-cli... [09:33] also a way :) [09:34] mhr3+1, no, nothing, any steps? [09:34] Saviq, just random crash while swiping around [09:34] mhr3+1, mhm, sounds like delegate creation/destruction gone wrong [09:35] on that note, I suspect we're keeping too many delegates around [09:37] didrocks: Saviq: testing silo 9... [09:39] sil2100, bump about prep silo for right edge, sorry for nagging :) [09:41] Saviq: I think I already assigned it some minutes ago ;) [09:41] Just in a meeting now still! [09:41] sil2100, ah... [09:42] * Saviq should check the spreadsheet... [09:43] mzanetti, right edge going into ppa:~ci-train-ppa-service/landing-015/ [09:43] thanks Saviq & didrocks :) [09:43] & sil2100! [09:43] ^ [09:43] yw ;p Sorry for not mentioning [09:44] mzanetti, btw, looks like we need some more version bumps, though [09:44] mzanetti, we need debian/changelog bump in https://code.launchpad.net/~mzanetti/unity-api/new-screenshot-and-focusing-api/+merge/199810 [09:44] mzanetti, so that others can depend on the -dev package at the right version [09:45] unity-mir in that case [09:45] Saviq: so which one do I need to bump? [09:45] ah. changelog [09:45] mzanetti, yeah, in unity-api [09:45] I thought that happened automagically when landing [09:45] mzanetti, and then >= that in unity-mir [09:45] yeah, sure [09:45] mzanetti, it does [09:46] mzanetti, but if you don't have that ↑↑ in unity-mir [09:46] mzanetti, it will build against any unity-api there is [09:46] mzanetti, if API changes, you need to manage the deps yourself [09:46] ah, sure, makes sense [09:47] mzanetti, we will probably see it with unity-mir failing to build in the silo now [09:47] mzanetti, ah, and yeah, right-edge conflicts with trunk... [09:48] no surprise there :D [09:48] ok, on that stuff to fix [10:02] how can I debug a qml module not loading? [10:02] or which are the requirements for it to load? [10:05] Cimi, what do you mean "not loading"? [10:05] Saviq, cannot find a module [10:05] Cimi, means you don't have it on your QML2_IMPORT_PATH or its qmldir is missing [10:06] Cimi, qmlplugindump is helpful in that === mhr4 is now known as mhr3 [10:39] JohnLea: hey, who should I asked to have a bug related to phone and new scopes for the wallpaper support in term of design strategy? [10:39] JohnLea: so that we can point people to one [10:39] JohnLea: I know you're not the one working on that, but you are my connection with the design team :) [10:41] mpt, ^ (seems like touch/unity8 dropped support to customizing the user session background) [10:42] which means at least the settings need to be updated [10:44] lol [10:44] /food [10:44] mpt didn't like that [10:48] dednick, I have this code http://paste.ubuntu.com/7150619/ [10:48] dednick, complains about [10:48] Error: Cannot assign QObject* to UnityMenuModel* [10:48] for menuStack.head = menuModel; [10:48] Cimi: dont think you need a stack [10:48] dednick, I used the code from the system settings [10:49] Cimi: Stack is for when the model is a tree. [10:49] the settings menumodel is flat. [10:51] Cimi: but in this specific case, you're assigning the stacks head to a sort model, not the qmenumodel [10:51] Cimi: s/qmenumodel/unitymenumodel [10:51] dednick, ok [10:51] now the filter filters everything, though [10:51] if should be menuStack.head = unitymenumodel (where the UnityMenuMode { id: unitymenumodel } [10:57] Saviq: one small issue in new scopes: in the scopes-scope, if there is no icon, the text lacks leftMargins [10:57] should I report a bug or is already some fix in progress? [11:00] dednick, how do I use filterRole and FilterRegExp? [11:00] Cimi: hang on, i'm looking at it now [11:01] dednick, this is my file http://paste.ubuntu.com/7150676/ [11:06] Cimi: use filterRole: 5 [11:07] not 4 [11:07] mzanetti, indeed [11:12] dednick, mmm [11:12] menuitemfactory expects unitymenumodel, though [11:14] sil2100: you should ping someone that knows about packaging for https://code.launchpad.net/~sil2100/dee-qt/add_cpp_symbols/+merge/202679 no? like didrocks maybe? [11:16] tsdgeos: ah, this one [11:17] tsdgeos: ok, will re-visit it, since I need to double check something there now, and will poke someone from the core-devs ;) [11:17] tsdgeos: thanks for reminding me about that one [11:18] yw [11:19] Cimi: hm. all of this code is pretty old compared to what's in unity8 not. [11:20] dednick, shall I grab your code? [11:20] Cimi: but basically, you need to be setting the unitymenumodel on the factory rather than the filtered model, and then do index conversion between the filter model and the unity menu model [11:20] yeah I thought about it [11:20] havinmg two models [11:21] basically [11:21] new crash https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1297223 [11:21] Error: ubuntu bug 1297223 not found [11:26] Cimi: try http://pastebin.ubuntu.com/7150757/ [11:28] Cimi: but yeah, the MenuItemFactory code should probably be updated at some point. [11:28] dednick, just menuitemfactory? === _salem is now known as salem_ [11:28] dednick, I'm rewriting my own code here [11:28] dednick, so I can copy that [11:28] and update [11:29] Cimi: are you using the code in ubuntu-system-settings/wifi ? [11:29] or adding another copy? [11:29] ubuntu-system-settings/plugins/wifi [11:30] dednick, adding another copy [11:30] dednick, since i have different design [11:30] Cimi: PageComponent & MenuItemFactory should be the same though.. [11:30] dednick, I will have to intewgrate password entry in the list [11:30] i see [11:31] dednick, no more snap decisions [11:31] Cimi: well that's the same for everywhere [11:31] MacSlow: ^ ? [11:32] ? [11:32] MacSlow: above discussion about password entry being part of the menu item. Is that still being done? [11:32] for wifi password [11:32] Cimi, could you elaborate on that? [11:33] MacSlow, when you tap an access point, the password entry appears in the listview [11:34] MacSlow: you were doing something about changing the wifi password from being a snap desicion into being integrated in the indicator menu items [11:34] dednick, last status of that is... what cimi described is the utimate goal... but for that to work much infrastructure is needed on the ofono and/or telephony backend-side... so for the time being we keep the snap-decisions we have atm [11:35] dednick, initially that was the plan... but there's this blocking on infrastructure atm [11:36] MacSlow, ouch, why is so difficult? [11:37] om26er, make public? [11:37] Cimi: because the backend password entry is a request based mechanism rather that a "conncet with this password" [11:37] ahhh [11:37] Cimi, ask Wellark [11:37] I'll skip it for now then [11:37] Cimi, or dednick :) [11:37] ciao dednick :P [11:37] dednick, wanna come here thursday? :P [11:38] Cimi: I'm off to Scotland on Thursday i think. [11:39] Cimi: but anyway, I think it'll be the same design everywhere for the list. So i think all the menuItem stuff in the ubuntu-system-settings should be the same. [11:40] Saviq, was a crash, should I wait for retracer [11:41] Saviq, made it public now https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1297223 [11:41] Ubuntu bug 1297223 in unity8 (Ubuntu) "unity8 crashed with SIGSEGV" [Medium,New] [11:43] om26er, oh interesting, that we didn't see before, wonder what retracer will come up with [11:43] mzanetti, ↑↑ crash in launcher [11:43] ah it retraced already [11:43] hmm [11:44] mzanetti, ah, yea i hit that yesterday [11:44] but was running ppa, so no nice stacktrace [11:44] any steps to reproduce? [11:44] anyway, you do have one now [11:49] Saviq: this one updated: https://code.launchpad.net/~mzanetti/unity8/make-xvfbtestSomething/+merge/211573 [11:51] Saviq, bug 1297239 [11:51] bug 1297239 in unity8 (Ubuntu) "Screen flickers with right edge switcher" [Undecided,New] https://launchpad.net/bugs/1297239 [11:51] om26er, will be fixed with right edge [11:52] great! [11:52] om26er, it happened before, too, but probably because background was darker it wasn't as noticable [11:52] mzanetti, assign yourself to ↑? [11:53] mzanetti, and thanks [11:53] om26er: you don't happen to know how to reproduce that? [11:54] mzanetti, the crash? no it happened random [11:56] * mzanetti flashing latest proposed image and trying to reproduce the launcher crash while waiting for the right edge silo to be built [11:59] mzanetti, did you update the branches already? [11:59] mzanetti, it doesn't auto-build, I need to push a button === MacSlow is now known as MacSlow|lunch [12:02] does the tiles in apps lens no longer have "pressed" state ? seems there is no difference [12:04] Saviq: I updated them an pushed the button already [12:09] mzanetti, ah [12:09] mzanetti, you have the button pushing powers, /me forgot :) [12:09] yeah, I use that power carefully only :D [12:10] mzanetti, FYI, when bumping changelog, you don't need to add the +foo [12:10] I don't need or I must not? [12:10] mzanetti, don't need to [12:10] mzanetti: glad QSignalSpy works. Have you new-scopes merged yet? [12:10] mzanetti, best to put "7.80.6-0ubuntu1" or so [12:10] greyback: yeah, silo will be buit soon [12:10] mzanetti: nice [12:10] mzanetti, so that we know what you added and what the system added [12:11] ah ok. Saviq, should I fix or its ok for this time? [12:11] mzanetti, no no, it's fine [12:11] .. [12:12] wrong window, sry [12:13] didrocks, (when) should we use ~ in Depends, i.e. (>= 1.2.3~)? === hggdh_ is now known as hggdh [12:16] Saviq, hm, interesting side-effect of https://code.launchpad.net/~mhr3/unity8/fix-1297246/+merge/212593 surfacing is now lazy, shell no longer fires up queries to all the scopes on startup [12:24] mhr3, ha, interesting indeed :) [12:25] mhr3, since we need to be able to supply them from the backend, we need the backend to kick the initial one, too, I'd say === jhodapp is now known as jhodapp|bbiab [12:26] Saviq, aren't we all in for being more lazy? [12:26] or perhaps i misunderstood [12:27] you're saying you *do* want something to fire up the queries asap [12:27] ? [12:27] mhr3, that's what we were doing until now [12:28] mhr3, would you delay that until it's shown or something? [12:28] Saviq, yes, that is what happens with that branch ^ [12:28] mhr3, ah [12:28] mhr3, I thought you meant they're not fired at all on startup [12:29] i.e. you ended up with empty scope until you did something [12:29] at startup they aren't, once you reveal the scope they are [12:29] mhr3, that sounds fine I think [12:29] agreed [12:32] didrocks, btw, please please apport-cli the .crash files first, or whoopsie-upload-all, otherwise I have to do it locally (and actually fail to now) === jhodapp|bbiab is now known as jhodapp [12:54] Saviq: hmm... another issue in new scopes (invisible tho): I messed up when merging the right edge, accidentally dropped the paper background image [12:54] Saviq: turns out, there's still the normal background behind it [12:55] mzanetti, prep an MP please, not a big issue [13:02] Saviq, https://plus.google.com/101899139221851198594/posts/EQYGqtG6Cpo [13:03] mhr3, I saw that in the tool, too, never on device [13:04] mhr3, we'll have to check out what's happening [13:04] Saviq, i do see it with ./run though [13:05] which would mean it'll be most likely visible in the preview session [13:05] mhr3, depends, it might be xcb-specific [13:05] mhr3, but we still need to check it out for sure [13:05] true [13:06] Saviq, opening bug then [13:06] mhr3, I've a feeling it's 5.2-related, might be caused by all the warnings we get in the console, which we need to shut up === alan_g is now known as alan_g|lunch [13:06] yea, didn't see it pre-5.2 [13:06] i think [13:08] Saviq, https://bugs.launchpad.net/unity8/+bug/1297270 [13:08] Ubuntu bug 1297270 in Unity 8 "Shell flickers when flicking through carousels" [Undecided,New] [13:20] greyback, have you ever seen this kind of crash (first item in the "Issues" session)? https://docs.google.com/a/canonical.com/document/d/1IiHBDIW_e0qnGt-po1D2z5HJKrNhBwh6pdILeEN2sgA/edit# [13:21] dandrader: QRasterPlatformPixmap <- looks wrong, we shouldn't be using htat [13:21] I don't get how the QPA works fine in a device and crashes in another... [13:21] greyback, yeah, I found it weird... is that software rendering? [13:22] dandrader: for some reason Qt/QPA is deciding not to use a texture for the image, which is wrong. Wonder why however [13:23] dandrader: this is nexus4 I assume [13:23] greyback, yes, it's written in a line above the stack trace [13:23] ah sorry, missed that [13:24] dandrader: any output from the binary as it's running that might be handy? [13:24] Saviq: it's if you have a version which isn't released, what's your case? [13:24] greyback, not really. but the qml-demo-shell works fine.. [13:25] didrocks, not applicable, then [13:26] dandrader: interesting, can you get more frames in that stacktrace? What's loading the pixmap? [13:26] greyback, now flasing my Nexus 7 to try out the stuff there. I'm curious to know if it works there :) [13:26] Saviq: I did that, right? [13:26] greyback, no. below that the stack is garbage [13:26] for the apport-cli [13:26] or corrupted [13:26] didrocks, no, the .crash I got was naked [13:27] Saviq: https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1297240 [13:27] Error: ubuntu bug 1297240 not found [13:27] dandrader: could you try installing the UITK debug packages, just in case that's the source? [13:27] Saviq: seems the retrace worked? [13:27] might give us a few more frames [13:27] didrocks, yeah, but the one you attached was not processed AFAICT [13:28] didrocks, I grabbed it from the email somewhere.. [13:28] didrocks, anyway, you know what to do, so let's not drag that any more :) [13:28] greyback, hmm, that's an idea. also a virtually fully commented out unity8 works fine. so another approach is to slowly uncomment things until the crash pops up [13:28] Saviq: what should I do? you have the bug report retraced? [13:29] didrocks, nothing, the retrace is as good as it gets [13:29] so everything's fine? (and I can close the other .crash) [13:29] great :) [13:29] Saviq: let me look if I can make this bug public [13:29] didrocks, I already did, core is stripped already [13:29] and nothing else sensitive I don't think [13:29] yeah, just quickly checked, nothing sensitive [13:30] Saviq: I'm duplicating the other bug [13:30] didrocks, kies [13:30] I can also build qt myself to get all the gory details and debug outputs imaginable. just hope I don't lose a week on that like that white flash bug. Mustering the courage to jump on it :) [13:30] greyback, ah, btw, virtual keyboard is working! [13:30] got it on Friday [13:31] dandrader: ah cool [13:31] dandrader: I'm working on mir+qpa stuff now, so please pull regularly [13:31] I just removed one hack from Mir, and fixed a hang at shutdown [13:32] greyback, we will still have to keep the UbuntuKeyboardInfo hack (to tell us the vkb opaque rect) until the touch cancelling work lands (which is a big diff, better to land separately) [13:32] dandrader: perfect. Good to keep changes reasonably sized [13:32] greyback, on the for-qpa-mir-server branch? [13:32] dandrader: mir-for-qpamirserver, and qpa-mirserver [13:33] dandrader: am trying to get it running on desktop again - I need to re-enable a PointerController [13:33] crashing for some reason, but almost there [13:35] ok [13:39] tsdgeos, so... of course I need to pull you... bug #1297240 - I wasn't able to repro and can't really discern anything from the (otherwise quite good) trace... [13:39] bug 1297240 in unity8 (Ubuntu) "unity8 crashed with SIGSEGV in QQuickWindowPrivate::polishItems()" [High,New] https://launchpad.net/bugs/1297240 [13:42] mterry, hey, looks like lightdm got released, let's drop it from silo? [13:42] mterry, also, think it's time to put the split greeter in design hands for testing? [13:43] Saviq, yes to first [13:44] Saviq, design has given video on what boot animation should look like. The silo doesn't represent that yet, and that's really the only visual change (except for indicators having different menus I guess). So I want to land the animation tweaks before asking design to feedback [13:45] mterry, got it [13:45] Saviq, but I think I should send out a functionality call for testing. Despite the keyboard Mir bug [13:45] sil2100, can you please drop lightdm package from silo 004 ppa? [13:46]