[09:46] cimi: what's the status of card-social, want me to review? [09:46] tsdgeos, debugging why the scope does not work for me :) [09:46] ok [09:46] tsdgeos, https://code.launchpad.net/~stolowski/+junk/scope-social_actions [10:21] cimi: want me to try to help you debug what's wrong or you have it? [10:22] tsdgeos, for now let me see [10:24] oki [10:26] did a quick review of the code and looks reasonablre === dpm_ is now known as dpm [10:46] tsdgeos, I tried putting an onCardDataChanged jsonstringify cardData inside cardCreator (when !isCardTool) but it is not printing anything [10:46] cimi: phone? desktop? [10:47] desktop [10:47] did you run make? [10:47] cardcreator needs make [10:49] tsdgeos, ok I changed a bit into onTitleChanged: console.log("UCD inside card", JSON.stringify(cardData["socialActions"])); [10:49] tsdgeos, but is undefined [10:49] tsdgeos, same is social-actions [10:49] print cardDAta [10:50] tsdgeos, it doesnt print [10:50] what code do you have exactly? [10:51] tsdgeos, but I can print cardData["attributes"] fine [10:51] tsdgeos, http://paste.ubuntu.com/15471189/ [10:52] tsdgeos, print http://paste.ubuntu.com/15471192/ [10:53] the other UCD print is from cardtool fake template [10:53] components["social-actions"] is indeed defined for both [10:53] I guess we have a bug somewhere in pawel branches no? [10:57] cimi: http://paste.ubuntu.com/15471216/ works fine to print cardData for me [10:58] runnign make tryCard i get [10:58] http://paste.ubuntu.com/15471219/ [10:59] tsdgeos, for cardTool? [10:59] i don't understand why cardData would be undefined for you [10:59] tsdgeos, I run with unity scope tool [10:59] cimi: what do you mean "for cardTool"? [10:59] tsdgeos, the template card with fake data from cardtool? [11:00] tsdgeos, when isCardTool is true [11:00] yes, what about it? are you asking if it prints there? [11:00] or the one with false [11:00] well, you can see it is there too [11:00] second like [11:00] line [11:01] tsdgeos, works fine too here for tryCard [11:01] tsdgeos, not scope tool [11:02] mmm maybe I need to run smart proxy something [11:03] nah always not working with that [11:06] yeah [11:06] actually it seems json.stringify doesn't know how to stringify the real cardData since it's a QQmlDMAbstractItemModelData [11:06] wonder how i did print that [11:09] tsdgeos, you told me to use stringify :D [11:09] i did [11:09] but yeah I understand [11:09] i've been known to lie :D [11:09] tsdgeos, it worked for the fake ones because we were treating them as json [11:16] cimi: ok, this is what you want [11:16] + onCardDataChanged: { console.log("root", root); for (var k in cardData) console.log(k, cardData[k]); } \n\ [11:17] gives me something like http://paste.ubuntu.com/15471434/ [11:18] that betteR? [12:32] tsdgeos, http://paste.ubuntu.com/15471733/ [12:33] cimi: ask pstolowski [12:34] pstolowski, I updated my card social branch, but the model doesnt have socialActions defined === maclin1 is now known as maclin === alan_g is now known as alan_g|lunch [13:36] Saviq: http://people.canonical.com/~gerboland/perf/ [13:36] perf needs symbols files to backtrace properly [13:37] it'll have same limitations as gdb probably [13:37] greyback, yeah got those [13:37] Saviq: what are you seeing? Can you paste me a report [13:37] Saviq: in general, everything I know, I got from http://www.brendangregg.com/perf.html [13:38] # Sample CPU stack traces for the PID, using dwarf to unwind stacks, at 99 Hertz, for 10 seconds: [13:38] perf record -F 99 -p PID -g dwarf sleep 10 [13:38] is probably the one I use most [13:39] greyback, http://pastebin.ubuntu.com/15472057/ [13:39] Saviq: huh it saved nothing [13:40] try use the -F flag to increase the frequency of the perf snapshots [13:40] does "perf top" give you much? [13:41] greyback, http://pastebin.ubuntu.com/15472067/ [13:42] Saviq: wtf [13:42] as if that function is stuck in infinite loop [13:43] greyback, what's "dwarf"? [13:44] Saviq: attempts to unwind the stack using libdwarf, instead of trying to follow the frame pointer [13:44] dwarf: No such file or directory [13:45] hmm, maybe older perf doesn't support it [13:46] greyback, same thing in perf report :/ [13:52] Saviq: :( We need to learn what that symbol corresponds to. I don't really know what else can be done [13:53] greyback, at least for now we seem to have found why the freeze - we were never changing timezone so early before [13:53] and with new OOBE tz change is part of it [13:54] Saviq: but you were thinking rotation might also have something to do with it? [13:54] greyback, yeah, tz change + rotation triggers it [13:54] it's real dumb [13:56] bizarre [13:56] if you want me to look into it, say the word [13:56] greyback, I think we'll let Mir folk look into it, assuming we get clear steps to repro [13:56] ok [13:58] cimi, will take a look [14:18] dandrader_: will you review https://code.launchpad.net/~aacid/unity8/rootObjectCanBeNull/+merge/289800 ? It's on some code that seems you made (or copied from some other file) [14:19] tsdgeos, sure === dandrader_ is now known as dandrader [14:46] cimi, are you sure you're using shell plugin from the silo? [14:50] pstolowski, which branch? [14:53] pstolowski, yes [14:53] I built locally [14:53] pstolowski, does it work for you? [14:58] cimi, i need to flash my phone to check. only looked at my code (which is exactly the same as for inline playback) [14:58] cimi, have you pushed all your changes? i'll rebuild the silo [14:58] pstolowski, I did [14:58] k [15:07] mmm looks like lp is annoyed [15:08] pstolowski, no pawel we need this merge https://code.launchpad.net/~cimi/unity8/card-social/+merge/289663 [15:08] Saviq, greyback I reproduce the freeze once by following the steps from the bug report. and i've just got another freeze after activating app-switcher [15:09] pstolowski, I think we found reliable steps to repro - and it's a preexisting issue triggered by the silo [15:09] pstolowski, I just updated the bug report a moment ago [15:09] Saviq, ack [15:10] related to timezone change, which we didn't do in OOBE so far [15:10] and triggered reliably by device rotation [15:10] Saviq, the freeze I just got with app switcher is probably something else. unity8 eats very little cpu [15:11] mhm [15:11] Saviq, in the original case i couldn't even adb [15:11] pstolowski, yeah, that's when adb asks unity8 whether it's unlocked [15:11] and it times out [15:14] Saviq, is this only on 1st timezone setup? i couldn't reproduce it when changing timezones later [15:15] pstolowski, yeah, needs a wipe === dandrader is now known as dandrader|lunch [16:09] dandrader|lunch: ping when you're back === dpm is now known as dpm-afk [16:32] dandrader|lunch: commented on the homekeypage [16:50] tsdgeos, pstolowski, josharenson seems like the rebuild of silo 71 is going well [16:51] cimi: I'll make sure to dist-upgrade when its done === dandrader|lunch is now known as dandrader [16:54] tsdgeos, ping [16:54] dandrader: the question i made on the MR [17:00] cimi: How do you install that test scope? I built/installed it w/ click/pkcon but its not showing up in the scope list [17:04] cimi, oh, i need all the filters stuff in silo 71 :( [17:04] cimi, and yes, i remember you mentioned rebasing our stuff on top of filters.. [17:27] josharenson, you installed the silo? [17:28] cimi: yes but haven't updated it in ~ an hour [17:29] josharenson, we need to wait for unity-api stuff [17:29] ok [17:29] josharenson, build didnt go well I think [18:36] Saviq, ping [18:47] dandrader, pong [18:51] Saviq, for animated cursors, I wanna use the AnimatedImage item. But for that I've to write a QImageIOPlugin to read xcursor files. It's a Qt GUI plugin. Ideally it should live in a separate package and project. But that's probably a lot of bureaucracy, so at last at first I guess we should just put int in unity8/plugins (along with those qml plugins)? [18:52] s/last/least [18:57] dandrader, maybe it makes sense upstream, simply? [18:58] dandrader, we can carry it as a distro patch until it lands upstream in Qt and we have their version [19:07] Saviq, using xcursos is quite a niche. Only compositors written in qt would be interested in that [19:08] dandrader, yeah, maybe - no opposition on including it in unity8 from me [19:08] ok