/srv/irclogs.ubuntu.com/2014/02/07/#ubuntu-unity.txt

tsdgeosSaviq: no release? :/08:06
Saviqtsdgeos, your action branch conflicted and then I couldn't get anyone to reconfigure :|08:06
Saviqwhen I resubmitted as ~unity-team08:06
Saviqso yeah, that's pretty crap08:07
tsdgeossorry :/08:08
tsdgeoswhat it conflicted with?08:08
Saviqtsdgeos, audioPlayer08:08
tsdgeosreally?08:08
tsdgeosdamn08:08
Saviqtsdgeos, yeah, /// vs. //! apparently08:09
tsdgeos:/08:09
Saviqtsdgeos, but don't be sorry, that's what this system causes08:10
Saviqtsdgeos, that would be an advantage of the staging branch08:10
tsdgeosyeah08:11
tsdgeoswhy did we decide against it?08:11
Saviqmore work than worth it08:11
Saviqtsdgeos, a solution would probably be to start pushing as ~unity-team instead08:12
Saviqtsdgeos, or well, not require a special person to be able to push a button...08:12
tsdgeosSaviq: well, pushing as ~unity-team doesn't solve much since it still needs someone to approve your branch, no?08:14
tsdgeoss/branch/change08:14
Saviqtsdgeos, no, 'cause there's no change in the CI train system08:15
Saviqtsdgeos, you just build again08:15
Saviqtsdgeos, reconfiguration is only needed when MPs change08:15
Saviqas in MP URL08:15
tsdgeosoki08:17
Saviqtsdgeos, could you review https://code.launchpad.net/~unity-team/unity8/unity8-card-overlay/+merge/204790 please?08:21
tsdgeossure08:22
tsdgeosabout the build thing08:22
tsdgeosso what do we do?08:22
tsdgeosdo i remerge my thing?08:22
tsdgeosapprove yours?08:22
tsdgeosor?08:22
tsdgeosdid you resubmit all the dependant branches on mine to depend on the new one?08:23
Saviqtsdgeos, I approved mine already, we just need to get someone to reconfigure the silo08:23
Saviqtsdgeos, let me see if any dependant branches conflict08:23
Saviqthere's only one08:24
tsdgeosok08:32
tsdgeosSaviq: do we need to resubmit it for the prerequisite branch to be updated, or that's old stuff we don't need anymore in this silo approach?08:32
Saviqtsdgeos, yeah, just did08:32
Saviqtsdgeos, prerequisite still helps to show the diff correctly08:32
tsdgeosok08:33
Saviqtsdgeos, https://code.launchpad.net/~saviq/unity8/new-preview/+merge/20529708:33
Saviqtsdgeos, so while it's not required, it's useful still08:33
tsdgeosok08:34
tsdgeosdone08:36
Saviqhow THE FUCK is that a conflict http://pastebin.ubuntu.com/6890078/08:40
tsdgeoscriss cross08:41
tsdgeosmerge with a different algorithm08:41
tsdgeosthat's what we get for moving stuff around and merging it in weird orders08:41
tsdgeosit's a bit of sad08:41
tsdgeosbut as said, without staging it's what we are doomed with :D08:42
Saviqyeah, problem is, will the train try --weave08:42
tsdgeosi guess that's a question for didrocks08:42
Saviqtsdgeos, your branch would probably merge with --weave fine, too08:42
Saviq/have merged08:43
Saviqoh maybe not08:43
Saviqthere was an actual conflict there08:43
Saviqtsdgeos, anyway, I think staging is just more work, in the normal case we'd land every day, maybe two08:44
Saviqtsdgeos, and also I'd need a `pull` instead of `merge` there, I can't live with a history of "pull from staging"08:44
tsdgeoshe he08:45
tsdgeossure08:45
tsdgeoswe just need to land every day :D08:45
Saviqtsdgeos, and we'd have to merge stuff manually into staging, keeping authors and commit messages... bleh08:45
Saviqtsdgeos, TBH what we really need is more control, and less waiting for didrocks to WAKE UP! ;P08:45
tsdgeosor that08:46
tsdgeosSaviq: readonly property var headerHeight: card.headerHeight ?08:50
tsdgeoswhy not an alias?08:50
Saviqtsdgeos, good question08:50
Saviqtsdgeos, fixed08:52
tsdgeoshmmmm08:54
tsdgeosotoh an alias is not readonly08:54
tsdgeosSaviq: can you also make it readonly? :D08:55
Saviqtsdgeos, I did08:58
Saviqtsdgeos, "upstream"08:58
tsdgeos¿?08:59
Saviqtsdgeos, it's readonly in Card.qml08:59
tsdgeosah08:59
Saviqtsdgeos, so if it's an alias in CardTool.qml, it's readonly there, too08:59
tsdgeosmakes sense08:59
tsdgeosSaviq: shoudl that "gray" be Theme.palette.selected.backgroundText ?08:59
Saviqtsdgeos, maybe it makes sense to mark it readonly there, too, for doc purposes08:59
Saviqtsdgeos, did that ↑09:01
Saviqtsdgeos, if you branched, pull --overwrite09:01
Saviqtsdgeos, fixed color09:02
tsdgeosoki09:02
MirvSaviq: hi. still row 42 reconfiguring needed I guess?09:02
SaviqMirv, indeed :|09:03
MirvSaviq: we had a team dinner so obviously no-one was looking at irc even from hotel room09:03
Mirvok, doing09:03
SaviqMirv, there's one more issue... we have a criss-cross there, do you know if it will try --weave?09:03
tsdgeosSaviq: no way to show the overlay in testXYZ or at least tryXYZ?09:03
Saviqtsdgeos, let me add09:04
MirvSaviq: we'll find soon enough http://162.213.34.102/job/prepare-silo/73/console09:05
Saviqtsdgeos, pushed09:05
SaviqMirv, thanks09:05
MirvSaviq: I don't see 'weave' in the sources at least09:05
SaviqMirv, do you know what to do with the branches to fix it?09:06
Saviqok let's see (/me doesn't have high hopes ;( http://162.213.34.102/job/landing-004-1-build/11/console)09:06
MirvSaviq: not offhand, but let's see and we should have didrocks woken up soon hopefully (7 minutes late! :)09:07
Saviq!!09:08
Saviqtsdgeos, added a test case, too09:11
Saviqtsdgeos, sorry for this not being really-ready ;)09:11
MirvSaviq: darn09:11
SaviqMirv, yeah :|09:12
MirvSaviq: so I guess it does bzr merge after another, in the order of branches, so that should be tweaked to succeed09:12
MirvSaviq: or, we'd need a feature to use '--weave'09:12
SaviqMirv, the last branch has the second-to-last as prerequisite, so the order can't be changed there09:12
SaviqMirv, so yeah, we need --weave (maybe with a job param, but yeah...)09:13
Saviqif only we could have rebase...09:14
Mirvdidrocks: http://pastebin.ubuntu.com/6890228/09:16
didrocksSaviq: I think we don't wait weave merge for every merge though, right?09:20
didrocks"Bazaar no longer uses weaves by default, because they were found to have poor performance and could not provide append-only guarantees."09:22
didrocksso, not sure we want that for everything09:22
didrocksand if we add an option to build, it will be for every merges in the set09:23
Saviqdidrocks, oh no, but maybe if a merge fails, it will try remerge with --weave?09:23
didrocksSaviq: do you think it will be something that everyone would wait?09:23
didrockslike first merge09:23
didrocksif fail, -> merge --weave?09:23
Saviqdidrocks, yeah, I think option would be good enough09:23
Saviqdidrocks, with a clear message in the log or so09:24
Saviqdidrocks, and maybe only if criss-cross is detected (if possible)09:24
didrockshum09:25
didrocksnot sure if I can get some easy info about criss-cross09:25
didrocksSaviq: do you have the branches so that I can play and see?09:25
didrocksSaviq: meanwhile, I think you will have to rebase, if possible :/09:26
* sil2100 looks into that in source09:26
Saviqdidrocks, row 42 in CI train09:27
Saviqdidrocks, the last two I think09:27
* Saviq checks09:27
Saviqdidrocks, rebase... not like that's something easy in bzr :/09:28
didrocksSaviq: you have a plugin09:29
Saviqdidrocks, discontinued and unsupported... never had any luck with it...09:29
didrocksoh, worked pretty good for me09:29
* Saviq tries, then09:30
* Saviq generally doesn't seem to have much luck with bzr... bisect never worked, either...09:30
* didrocks didn't try bisect for a while09:31
Saviqdidrocks, rebase → first commit, full of conflicts, files lost, MAYHEM09:33
seb128Saviq, you are doing it wrong!09:35
* seb128 hides09:35
Saviqseb128, probably09:35
seb128Saviq, sorry, friday, troll day, all that ... happy friday! how are you?09:35
Saviqseb128, sad atm, can't land since yesterday evening :(09:36
seb128:-(09:36
Saviqok, let's see, something rebased09:36
tsdgeosSaviq: cool looking overlay!09:40
Saviqtsdgeos, /me happy09:40
tsdgeosSaviq: i don't know much/anything about shaders though09:41
tsdgeosso i'll just assume the shader works because it does work09:41
Saviqtsdgeos, it's pretty simple - the vertex shader is just the default09:41
Saviqtsdgeos, i.e. no modification09:41
tsdgeosand the other multiplies09:41
Saviqtsdgeos, the other applies the images' opacity on black09:41
tsdgeossure i can see that09:41
Saviqtsdgeos, that it09:41
tsdgeosi just don't have a clue if it can be made faster/better09:41
Saviqtsdgeos, ah well09:42
Saviqtsdgeos, it's temporary anyway09:42
Saviqhum conflict in tst_Indicators.qml, that's weird09:42
tsdgeosSaviq: dednick changed lots of stuff in indicators09:42
tsdgeosmaybe it didn't play well together09:43
Saviqtsdgeos, yeah, but I'm rebasing09:43
tsdgeosah09:43
tsdgeosSaviq: so your card CI didn't succeed09:43
Saviqtsdgeos, neither branch should be touching it...09:43
tsdgeosbut it's obviously not related to your change09:43
Saviqtsdgeos, autopilot?09:43
tsdgeoschecklist-approve anyway09:43
tsdgeosyeah09:43
Saviqhttp://pastebin.ubuntu.com/6890343/09:45
SaviqWTF09:45
dednickweird. spacing?09:46
dednickon end.?09:46
Saviqdednick, no, whitespace would've complained09:48
Saviqanyway, no, that rebase isn't happening09:49
* Saviq went for manual rebase09:59
* Saviq crosses fingers http://162.213.34.102/job/landing-004-1-build/12/console09:59
* sil2100 crosses fingers as well09:59
Saviq\o/10:05
tsdgeos\o/10:05
Saviqin your FACE, bzr10:05
sil2100HA10:08
sil2100Saviq: good work!10:08
sil2100;)10:08
Saviqsil2100, Mirv, can you please reconfigure row 42 again? just added one more MP that can land10:21
sil2100Saviq: ok, will do10:24
sil2100Saviq: ok, reconfigured, re-build10:29
Saviqsil2100, thanks!10:30
tsdgeosSaviq: so looking at cardheader <-> previewheader10:40
Saviqtsdgeos, yup?10:41
tsdgeosthe json specifies emblem and attribute-x10:41
tsdgeosthat cardheader doesn't have ¿yet?10:41
tsdgeosSaviq: so ignore those attribs for the moment? or?10:47
Saviqtsdgeos, yeah10:48
tsdgeosokidoki10:48
Saviqkarni, hey, rebased title alignment on trunk, but I believe it's reversed: https://code.launchpad.net/~unity-team/unity8/header-alignment/+merge/20533411:00
Saviqkarni, it should always be Left-aligned in horizontal, sometimes center-aligned in vertical, no?11:01
* tsdgeos kicks launchpad11:09
tsdgeosso https://code.launchpad.net/~saviq/unity8/new-preview tells me there's 1 branch that depends on that one11:09
tsdgeosbut then the link says none11:09
tsdgeosaaaaaaaa11:09
tsdgeosSaviq: do you know what comes after that one?11:09
Saviqtsdgeos, card-overlay, no?11:11
=== anpok_ is now known as anpok|lunch
Saviqhmm no11:11
MirvSaviq: ok, looks positive you'll have a landing PPA to run tests from soonish? built for armhf now :)11:14
SaviqMirv, yeah11:16
* Saviq TestPlans11:30
sil2100jamesh: hello!11:49
sil2100jamesh: are you around? I have some problems with the new mediascanner11:49
tsdgeosSaviq: cardheader doesn't define width nor implicitwidth, that ok?12:06
=== MacSlow is now known as MacSlow|lunch
sil2100mhr3: hi!12:10
mhr3sil2100, hey12:10
sil2100mhr3: you in the office today?12:10
mhr3sil2100, on my way there12:10
mhr3sil2100, btw did you see my libunity landing request?12:10
sil2100mhr3: I'll snatch you once you appear, since I see jamesh seems to be away, and I don't see Satoris as well12:10
mhr3sil2100, satoris in on #canonical12:11
sil2100mhr3: poke robru for that today - he's doing most landings now as part of practicing ;)12:11
sil2100mhr3: oh, ok, I'll poke him there as well, but you'll be poked as well12:11
sil2100In person12:11
sil2100Since we have a big regression related to mediascanner12:12
sil2100And its scope12:12
mhr3sil2100, hmm?12:12
mhr3sil2100, i was testing that yesterday it seemed fine12:12
mhr3sil2100, anyway, will be there in ~3012:13
sil2100mhr3: yeah, so... it's something that's happening only on newest images, since after the release of unity-scopes-mediascanner and mediascanner2 the grilo plugin for mediascanner got dropped12:13
sil2100mhr3: so, basically the music-app stopped working12:13
sil2100mhr3: like, completely12:13
mhr3sil2100, it shouldn't have been dropped if music-app deps on it12:14
sil2100mhr3: also, I noticed a packaging problem I guess12:14
mhr3sil2100, but the scopes dont12:14
sil2100mhr3: I don't think it does, it's a click app and I'm not really into how click apps deal with deps12:14
mhr3they don't afaik12:14
sil2100mhr3: yeah, so basically the scopes dropped the dependency and now we don't have grilo anymore12:15
sil2100mhr3: it's not being used in either the new and old mediascanner scope?12:15
sil2100(like, at all?)12:15
mhr3sil2100, no, the scopes (both old and new) use mediascanner212:15
mhr3that doesn't do grilo12:16
sil2100mhr3: ok, but we still need the old mediascanner for music-app then, right? Those two can still run in parallel?12:16
mhr3sil2100, yes12:16
mhr3sil2100, but music-app will also transition to mediascanner2 at some point12:17
mhr3then we'll drop the whole mediascanner1 + grilo12:17
sil2100mhr3: ok then, so I guess we can either add grilo-mediascanner to the seed to fix it, or simply add a quick dependency to mediascanner to fetch grilo-mediascanner12:17
sil2100mhr3: right, but for now we need to fix the image ;)12:17
sil2100mhr3: but thanks for the info12:17
sil2100mhr3: ok, so we decided to seed the grilo plugin in the meantime, so it's being done ;)12:24
sil2100mhr3: so no action required from you, thanks for clearing things up12:24
tsdgeosCimi: Saviq: this one shall be easy https://code.launchpad.net/~aacid/unity8/preview_header/+merge/20535412:25
mhr3sil2100, good decision :)12:25
Cimitsdgeos, one extra new line on tests, after imports :)12:29
tsdgeosand probably too many imports12:29
Cimitsdgeos, tests could have something more maybe?12:30
Cimibut card header is already tested though12:30
tsdgeosCimi: yeah the only thing i am testing is my code12:31
tsdgeosCimi: fixed12:38
Saviqtsdgeos, it should implicitHeight, yes12:59
Saviqtsdgeos, the widget, that is, yeah - what you did12:59
Saviqtsdgeos, header itself has explicit height13:00
mhr3sil2100, if you still need, i'm here13:00
Saviqtsdgeos, 127+ function test_something() { a better name maybe?13:00
=== MacSlow|lunch is now known as MacSlow
Saviqtsdgeos, actually, it should be explicit height for the widget, too, no? i.e. the header doesn't really adapt when its height is changed13:09
=== dandrader is now known as dandrader|afk
karnio/13:33
=== dandrader|afk is now known as dandrader
=== dandrader is now known as dandrader|afk
karniSaviq: am I sleepy, or tryCardTool shows 11 cases in the drop down, while source contians 15 cases for that dropdown.13:52
Saviqkarni, there's a gazillion of branches, which one? ;)13:52
karniSaviq: sorry, the one you rebased https://code.launchpad.net/~unity-team/unity8/header-alignment/+merge/20533413:53
karniI fixed what you suggested (yeah.. that indeed was reversed) and wanted to inspect that visually as well13:53
Saviqkarni, just swipe ;)13:53
karniNACK. First one for me is "art, header, summary", last one is "art, header - portrait"13:55
karniI can see the full list from top to bottom13:55
karniFTR my network is a bit laggy13:55
karnilet me rebuild this13:57
karniI think that's my problem13:57
karniSaviq: ignore me for now.13:57
Saviqkarni, ;)13:57
karnitsdgeos: if you'd review this related branch as well, we'd have slick overlay both in trunk and new-scopes https://code.launchpad.net/~unity-team/unity8/newscopes-card-overlay/+merge/20535914:00
* karni sets description14:00
=== dandrader|afk is now known as dandrader
karnidone14:02
tsdgeosSaviq: fixed14:08
Saviqtsdgeos, cheers14:09
CimiSaviq, I'm writing the style for rating component, and I realised styleItem works even if I don't import Ubuntu.Components in the delegate file... why?14:14
SaviqCimi, it's probably the one from external scope14:15
Saviqif your whole chain of parents won't have Ubuntu.Components, it won't be there14:15
tsdgeoswooo, my hud-service went crazy14:17
* tsdgeos kills it14:17
tsdgeosbetter14:17
pete-woodstsdgeos: can you reproduce that?14:18
pete-woodstsdgeos: I am desperate to find a way14:18
tsdgeosfirst time it happens14:18
pete-woodsI've a report that sublimetext can send the new HUD into an infinite loop14:18
tsdgeosnot using that14:19
tsdgeosi was using firefox, kate, kontact14:19
tsdgeosthe same stuff i have open all the time14:19
mhr3mhall119, you forgot about me :(14:19
pete-woodsokay, thanks, I'll try running those14:20
tsdgeosbut had never had this 100% before14:20
tsdgeosso it must not be because of them14:20
tsdgeosi notice when the laptop fan spins :D14:20
pete-woodsokay, fair point, but still :)14:20
mhall119mhr3: I could never forget you14:20
tsdgeospete-woods: anything you want me to do if it happens a next time?14:21
mhall119(where/how did I forget you?)14:21
mhr3mhall119, no new scopes docs14:21
mhall119oh, right, docs14:21
mhall119mhr3: I'm going to blame the hotel wifi and claim that I didn't forget :)14:22
pete-woodstsdgeos: if you could try and get a stacktrace out of it, that would be very helpful14:22
mhr3mhall119, and i'll *almost* believe you14:22
tsdgeospete-woods: ok, will do if it happens again, sorry to have been to trigger fast killing it14:22
pete-woodstsdgeos: no worries14:22
pete-woodseven better would be a reproducable way to cause the loop14:23
pete-woods:)14:23
pete-woodsI realise that would be making my job far too easy, though14:23
mhr3mhall119, if you said that the sun in florida melted your enter, that would be more believable :)14:23
tsdgeoskarni: i'm a bit confused about that MR14:23
tsdgeosit wants to merge to lp:~unity-team/unity8/new-scopes but has a branch that wants to merge to lp:unity8 as pre-requisite ¿?14:23
* karni looks14:24
CimiSaviq, I started working on the textSummary branch, but I didn't merge there the latest changes of audioPlayer, so when I set textSummary as prerequisite I don't have the right diff on lp https://code.launchpad.net/~cimi/unity8/units8.previews_RatingStars/+merge/20537014:24
karnitsdgeos: Correct. Saviq helped me out pull as much as we could into trunk, so this change is specific to new-scopes, but does require the other branch to be merged to unity8 trunk first.14:25
karnitsdgeos: These are the new-scopes specific overlay bits.14:25
mhall119mhr3: uploading now...14:26
tsdgeoskarni: i don't understand how does it matter what is in unity8 since new-scopes is a different branch14:26
mhr3mhall119, thx14:27
karnitsdgeos: in my understanding new-scopes is unity8 with only set of changes required to enable new scopes UI/toolkit.14:27
tsdgeosyes14:27
karnitsdgeos: when we work on dash toolkit, we do as much as we can in trunk, and as little as we can in new scopes14:27
tsdgeosok, maybe it's different understanding of "prerequiste" branches14:28
karnibecause trunk does not contain CardFilterGrid and CardCarousel14:28
tsdgeosfor me a prerequisite is "something that you need landed before you land this one"14:28
karnitsdgeos: I still believe the other merge to trunk is required for this merge to go smooth.14:28
* karni looks14:28
tsdgeosok14:28
karnitsdgeos: give me a sec please14:29
tsdgeossure14:29
mhall119mhr3: check the docs and let me know if your new stuff is there14:30
mhall119mhr3: it would be nice if you could write a parser for your docs to push them to the new API website on your own14:30
mhr3mhall119, yep, it is, thx14:30
mhr3mhall119, i might get to that... in 201514:31
mhall119:-P14:31
mhall119but it's easy14:31
mhall119I even made a nice python library for you14:32
mhall119s/nice/functional/14:32
mhr3can't it be both? :)14:32
karnitsdgeos: unity8-card-overlay (being merged to unity8 trunk) defines property bool showHeader in qml/Dash/Card.qml, which is then used in new-scopes. My bad, please don't review the other branch now. We'll first get the trunk one merged, we'll merge changes from trunk to new-scopes, and only then we'll approve the MR I sent over into new-scopes.14:34
karnitsdgeos: Now I understand the flow Saviq usually executes when there are related changes to trunk and new-scopes.14:34
tsdgeosoki14:35
karnifirst merge to trunk, then merge trunk to new-scopes, only then approve and merge related branches into new-scopes14:35
karnitsdgeos: unless, you want to review it, but not top-approve it yet. that'd be the same thing, basically.14:35
mhall119mhr3: it can be, if it was written by somebody with more talent than me :)14:35
karniYou're fresh on the subject, as you just reviewed the trunk one14:35
mhall119mhr3: I just have a bad habit of re-inventing Django's model objects for any given python project14:36
mhall119but in this case it makes sense, because it's talking directly with Django models on the server-side14:37
tvossmzanetti, ping14:40
=== rachelliu_ is now known as rachelliu
mzanettitvoss: hi14:40
elopioping mterry: I have the test passing, without depending on the calendar14:42
mterryelopio, calendar?14:43
elopiosorry, still early. s/calendar/camera14:43
elopioI'm wondering if we want this test in unity8, because to make it work we need the real application manager.14:43
karniSaviq: pushed fix to https://code.launchpad.net/~unity-team/unity8/header-alignment/+merge/205334 . All tests pass, but the problem appears when you switch card layouts more than once in testCardTool. "Title - horizontal" > "Title - vertical" > "Title - horizontal" boom Card height undefine., no card displayed.14:48
tsdgeosoh god14:54
tsdgeossomeone did add the automagically creating setters and getters for QPROPERTIES14:54
tsdgeos\o/14:54
tsdgeosat last :D14:54
karniwohoo14:55
tsdgeoshttps://plus.google.com/117221897452321521192/posts/hZQDDnwduhs14:56
Cimitsdgeos, help with rating widget? from the json, I don't understand which are the data I get14:57
tsdgeoslet me see14:58
mzanettitsdgeos: nice" that is awesome!14:59
tsdgeosCimi: which one are you doing? "rating-input" ?14:59
Cimitsdgeos, I currently just worked on the rating widget under COmponents15:00
MacSlowgreyback, hey... can you make it to the hangout?15:00
Saviqelopio, hey, do you have a maguro?15:01
tsdgeosCimi: ok, so i didn't understand your question :D15:01
elopioSaviq: I don't.15:01
elopiobut I know people. What do you need?15:02
tsdgeosmhr3: ping?15:02
mhr3tsdgeos, sup15:02
Saviqelopio, http://ci.ubuntu.com/smokeng/ we're getting some apparently random ap failures on maguro15:02
tsdgeosmhr3: so how do we do progress reporting for "type": "progress" ?15:02
Cimitsdgeos, but I was expecting a rating bar15:02
Cimitsdgeos, not input and review15:02
mhr3tsdgeos, there's some progress provider thing15:03
Cimitsdgeos, it looks like we're missing that widget in the spec15:03
tsdgeosCimi: the rating bar is part of the header i think15:03
Cimitsdgeos, scope like IMDB won't feature single reviews15:03
mhr3tsdgeos, it's used by the app preview15:03
mhr3tsdgeos, the old one15:03
elopioSaviq: robotfuel and davmor2 have maguros. Any of you have some time to try to reproduce them?15:03
Saviqelopio, one of them was in the emulators tests, another in hud, will have to check it out then15:03
Saviqelopio, davmor2 was already looking into it15:03
davmor2elopio: tests are rerunning now :)15:04
robotfuelSaviq: elopio can try to reproduce on a maguro, if you need me to15:04
tsdgeosmhr3: com.canonical.applications.Downloader ?15:05
elopioSaviq: yeah, I saw the error they mentioned on the open scope test I added. Without a maguro or a video of the run, it's really hard to understand what's going on. The log says autopilot tried the drag correctly.15:05
Saviqelopio, indeed15:05
tsdgeosmhr3: but that's too specific, no?15:05
mhr3tsdgeos, yea, that thing15:05
elopiorobotfuel: thanks, but it seems davmor2 is way ahead.15:05
robotfuelack :D15:06
mhr3tsdgeos, we didn't really define much what's the data for that widget15:06
mhr3tsdgeos, see....15:06
tsdgeosmhr3: i know, but if we need to code it ... :D15:06
mhr3tsdgeos, http://people.canonical.com/~mhr3/unity-scopes-api/classunity_1_1scopes_1_1_preview_widget.html#progress15:06
tsdgeosmhr3: maybe that name should come as part of the json too?15:06
tsdgeosmhr3: i can't edit the json document, so it would be "source": ["dbus-name" : "somename", "dbus-object": "somestring" ]; instead of "source": null , yes?15:08
Cimitsdgeos, ?15:11
Saviqkarni, mhr3, tsdgeos, merged everything into lp:unity-team/unity8/new-scopes15:11
Saviqkarni, tsdgeos, so reviews of https://code.launchpad.net/~unity-team/unity8/new-scopes/+activereviews can be done again15:11
tsdgeosCimi: which ratings are you speaking about?15:11
tsdgeosCimi: i.e. do you have a design so we can we sure we're speaking about the same thing :D15:12
karniack15:12
* karni tries to reproduce with a test the problem in tryCardTool in header-alignment branch15:13
karniIronically, when using tryCardTool, problem is obvious, but from the test (by setting the drop down selector) test passes, when it should fail.15:14
=== anpok|lunch is now known as anpok
Cimitsdgeos, get my branch15:20
Cimitsdgeos, well, I have design in mind15:20
Cimididn't check specs15:20
Cimiah I see15:21
Cimibasically rate this15:21
Cimitsdgeos, Saviq putting just an icon of a star with a label (4.9) means nothing15:21
Cimiin the header15:22
Cimiyou need a visual clue of the range15:22
Cimiotherwise you put %15:22
karniSaviq: FTR whenever you have a moment - can't make this use case fail from code, left a comment on branch. clearly it fails when run manually on tryCardTool https://code.launchpad.net/~unity-team/unity8/header-alignment/+merge/20533415:22
tsdgeosCimi: talk to design :)15:22
Cimiso the range is per cent15:22
Cimitsdgeos, we need to add 5 if we want to turn 4.9 meaningful15:22
Cimiso 4.9 / 515:22
Cimi4.9 could be super crap if it's out of 10 or great if itos out of 515:23
mhr3tsdgeos, well, it really depends on the progress indicator, i guess the only one ew have now is dbus-based, but basically we were saying in the spec that we don't care, whatever the progress provider requires should be there15:23
mhr3tsdgeos, s/indicator/provider/15:24
tsdgeosmhr3: not sure i understand you, you mean that you don't want to spec the dbus names?15:25
mhr3tsdgeos, we should support multiple types of progress indication, the dbus-based is just one of them15:26
mhr3tsdgeos, although the only one atm15:27
tsdgeosmhr3: ok, but still we should spec them all, no?15:27
mhr3tsdgeos, right, the thing is that what they need can change independently of scopes or the shell15:28
tsdgeosmhr3: so how is the shell going to automagically adapt?15:28
mhr3tsdgeos, what's why the spec says "whatever the provider needs"15:29
tsdgeosi mean eventually15:29
tsdgeoswe need to do code to do stuff15:30
tsdgeosif that code has unknown input15:30
tsdgeosthat's kind of hard to achieve15:30
Cimitsdgeos, but even then15:34
Cimitsdgeos, the json how is supposed to set the value?15:35
tsdgeosCimi: can you show me the design we're talking about?15:35
Cimitsdgeos, https://docs.google.com/a/canonical.com/document/d/1n880Fih5KyGPcoP5chidnHDG_8TxXUgSuij7f4rHpuk/edit#heading=h.c86ldo7kh1u15:36
Cimihttps://docs.google.com/a/canonical.com/document/d/1NmiM4UCnJgf6IEawmfyTOHRNAA5ZGrqpyrPqPOibwc8/edit15:36
tsdgeosCimi: so the header?15:36
Cimitsdgeos, no, rating input15:36
tsdgeosit's input15:37
tsdgeosyou don't get anything15:37
tsdgeosyou set it15:37
tsdgeosno?15:37
Cimitsdgeos, ok, but then this widget needs a way to set the new input, no?15:37
tsdgeosCimi: the triggered signal with some stuff in the data variable15:37
tsdgeosi'd say15:37
Cimitsdgeos, so I have to add a signal?15:40
Cimitsdgeos, currently, rating has no signal, just you use the property value15:41
tsdgeosCimi: previewidget has a signal15:46
karniSaviq: If you have something I could bite today, feel free to hit me. Previews are the last bit we care about, and it's my last day at the sprint, so whatever you have, happy to join effort.16:04
Saviqkarni, lp:~saviq/unity8/newscopes-new-dash-look16:05
Saviqkarni, run it in actual unity (not the dev tool)16:05
* Saviq pushes to ~unity-team16:05
elopiomterry, Saviq, mzanetti: so what do you think about https://code.launchpad.net/~elopio/unity8/url-dispatcher_test/+merge/205037 ?16:06
elopioshould we merge it even if it uses the real unity-mir ?16:06
Saviqelopio, IMO ap tests are integration tests, so they should use the real thing, so I'm good - but obviously you need to disable it for desktop scenarios16:07
Cimitsdgeos, which signal?16:08
tsdgeosCimi: see the actionpreviewwidget branch16:08
tsdgeoswe need to land all that crap :D16:08
=== alan_g is now known as alan_g|vt
mhr3Saviq, we seem to have lost the new-scopes preview integration branch16:09
mhr3any idea where did it go?16:09
karniSaviq: ack16:09
tsdgeosCimi: this signal https://code.launchpad.net/~unity-team/unity8/action_preview_widget/+merge/20523216:10
tsdgeosmhr3: https://code.launchpad.net/~unity-team/unity8/newscopes-preview/+merge/205029 ?16:10
Cimitsdgeos, just land those!16:10
elopioSaviq: ok. It's disabled for desktop.16:10
Cimiwe have deps of deps16:10
tsdgeosCimi: well if our tests weren't that unstable, that'd be done ages ago16:11
mterryelopio, your test got fancy!16:11
mterryelopio, real unity-mir is better, yeah16:11
Saviqmhr3, huh https://code.launchpad.net/~unity-team/unity8/newscopes-preview/+merge/205029 ?16:11
Saviqmhr3, but it'll be in new-scopes itself soon16:12
Saviqmhr3, and in demo-stuff, then16:12
mhr3Saviq, ah, wip, that's why i didn't see it16:16
mhr3how come https://code.launchpad.net/unity8 doesn't list it?16:16
mhr3right... because i'm blind16:17
elopiomterry: and just wait for the python-fixtures to start landing in ubuntu-ui-toolkit, it's going to shine :)16:17
mterry:)16:17
karnibrb reboot16:19
=== dandrader is now known as dandrader|lunch
karniSaviq: run that branch  with ./run ?16:28
karniIs that what you meant?16:28
Saviqkarni, yes16:28
Saviqkarni, we need to change text colours16:29
karniSaviq: I did. fonts are white on light background16:29
Saviqkarni, yeah, that's what needs fixing16:29
karniwait, I'm sure it was fixed somewhere already16:29
karniI did see it look right before, with the light background.16:29
karnion it16:29
tsdgeoselopio: is _open_scope_scrolling yours?16:30
tsdgeosin dash.py?16:30
elopiotsdgeos: and yours ;)16:31
tsdgeossure16:31
tsdgeoselopio: there's a bug16:31
tsdgeosmoving doesn't do what you think it does16:31
tsdgeosand that is why https://jenkins.qa.ubuntu.com/job/autopilot-testrunner-otto-trusty/2656/testReport/junit/unity8.shell.tests.test_emulators/DashAppsEmulatorTestCase/test_get_details_Desktop_Nexus_10_/ fails sometimes16:31
elopiointeresting.16:31
tsdgeosmoving is "vertically moving", not horizontally moving16:31
karniwtf.. I'm having unity7 rendering issues ://16:31
karnimakes it hard to work16:32
elopiotsdgeos: is there a horizontally moving?16:32
tsdgeoselopio: thinking about it16:32
tsdgeoselopio: we way need to resort to testing the x as we did somewhere16:32
tsdgeosor let me check16:33
* elopio fires up the vis.16:33
tsdgeosactually yes16:33
Saviqtsdgeos, karni, mhr3, newscopes-preview merged into new-scopes, too16:33
karni\o/16:34
Saviqso quality of new-scopes just went drastically down16:34
Saviqbut at least everything's in one place16:34
karni:D16:34
tsdgeoselopio: what we want is moving but on dashContentList not on the scope16:34
tsdgeoselopio: since the scope is verticall and dashContentList is the horizontal one16:34
tsdgeoselopio: am i making sense?16:34
elopiotsdgeos: you are, yes.16:34
elopiolet me see how can I get that element.16:34
tsdgeoselopio: do you think you can code that? i think it'll be faster if you do it than if i do, my autopilot+python foo is not very food16:35
tsdgeosgoof16:35
tsdgeosgood16:35
mhr3Saviq, it enables further dev, so +116:35
elopiotsdgeos: yes, I'm trying.16:35
Saviqmhr3, yeah, why I did it exactly16:35
tsdgeoselopio: cool16:36
mhr3Saviq, re-building demo ppa16:36
elopiotsdgeos: and just for future reference, how can I know if the moving is vertical or horizontal? It's not clear to me as I can move the scope in both directions.16:36
mhr3thostr_, ^^ previews hooked up soon in demo ppa's unity8 (+scope-tool)16:37
tsdgeoselopio: let me rephrase16:37
tsdgeosmoving is moving16:37
tsdgeosamazing skills i have here :D16:37
tsdgeoselopio: basically the dash is listviews inside listvides16:37
tsdgeosso there is an external one that is horizontal16:37
tsdgeosand that external one, contains one listview per scope that is vertical16:37
tsdgeosso it's not that moving is vertical16:38
mhr3Saviq, also, i have a branch with addSpecialCategory, wanna try it out?16:38
tsdgeosis that the moving for the listview you are testing is the vertical listview16:38
tsdgeosand we have to test the moving for the horizontal one16:38
tsdgeoselopio: ↑↑↑ making sense?16:38
elopiotsdgeos: yes yes, I also made a poor question :)16:40
elopioso, the thing is that I need to wait for the list I'm moving.16:40
elopioa container might have moving = False even if one of its elements have moving=True.16:40
elopiothat's good to know.16:40
tsdgeosit could yes16:41
tsdgeosnote that in this case is the other way around16:41
tsdgeosyou're testing for the inner thing and that one is not moving16:41
tsdgeoswhat is moving is the outer thing16:41
tsdgeoswell the inner thing is being moved16:41
tsdgeosbut not moving itself16:41
elopiooh, right.16:41
tsdgeosthe naming is not amazing16:41
tsdgeosmoving == scrolling || overshooting16:42
tsdgeosthat makes it easier to understand i guess16:42
tsdgeosmoving != "travelling around" on your parent16:42
elopiotsdgeos: it's clear, thanks :)16:42
elopioit's really bad that we don't have a good way to actually test this. I would need to slow down the animations.16:43
tsdgeosyou could do that16:44
tsdgeosi guess16:44
tsdgeosor maybe not16:44
tsdgeoselopio: yeah it's a problem with these kind of tests16:44
tsdgeosbut if you look at the video of the failure closely16:44
tsdgeoshttps://jenkins.qa.ubuntu.com/job/autopilot-testrunner-otto-trusty/2656/artifact/results/autopilot/artifacts/unity8.shell.tests.test_emulators.DashAppsEmulatorTestCase.test_get_details%20%28Desktop%20Nexus%2010%29.ogv16:45
elopiotsdgeos: where did you get a video from?16:45
tsdgeosyou'll see it's clearly that isue16:45
tsdgeosissue16:45
elopiooh, it's failing on trusty too, I thought it was only maguro.16:45
tsdgeosthe mouse moves and tries to click before the moving is done16:45
tsdgeosand ends up in the wrong place16:45
karnisaldkfjlsadkj why doesn't this list colors http://design.ubuntu.com/apps/style/typography16:46
elopiodavmor2, Saviq ^16:46
karniI lost the link to Ubuntu Theme colors on d.u.c16:47
elopiothanks to tsdgeos I now know what to fix.16:47
tsdgeoselopio: not really sure that one is the same than http://ci.ubuntu.com/smokeng/trusty/touch/maguro/167:20140206.1:20140115.1/6480/unity8-autopilot/741839/16:47
tsdgeosbut may be16:47
tsdgeosthe hud one, sadly, no clue16:48
Saviqelopio, tsdgeos, thanks, that makes some sense16:49
Saviqand on that note...16:49
Saviqor well, bombshell16:49
Saviqo/16:49
Saviqbe back on the 17th16:49
tsdgeosenjoy!16:49
tsdgeoselopio: i'm going to EOD soon, do you have anyone to review that potential patch in your timezone (you're americas based, right?)16:50
elopiotsdgeos: yes. Lets see... mterry?16:50
mterryelopio, sure, I can do that today16:50
tsdgeosthat should work16:50
davmor2elopio: ah nice does that mean you can test fro a fix there and then I can confirm on maguro?16:55
Cimitsdgeos, in oreviewactions you imported qtquick 2.116:55
Cimi*previewactions16:55
Cimitsdgeos, why that?16:56
tsdgeosCimi: for no reason really16:56
tsdgeosi think we should be importing 2.1 everywhere16:57
tsdgeosCimi: i could change it to 2.016:57
tsdgeosbut let's just leave it16:57
elopiodavmor2: more or less, I can't make a proper test atm as I don't know how to keep the list moving.16:57
tsdgeosor it'll create a massive cris cros again in the merges16:57
karnitsdgeos: I used "grey" in CardHeader.qml as text color. Saviq changed it to theme color, Theme.palette.selected.backgroundText, which is white. And we have light background in new-scopes. That means, the Theme does not contain a color we can use in stead of "grey".16:58
karnitsdgeos: Your take on this?16:58
elopiobad news, this can't be the same maguro problem.16:58
elopiotsdgeos, davmor2.16:58
karnitsdgeos: basically, Ubuntu.Components.Themes.Palette would need to be updated for new-scopes.16:59
tsdgeoselopio: oh :/17:00
tsdgeoselopio: why not?17:00
tsdgeoskarni: yes, Saviq was hinting that on our mumble chat17:00
elopiotsdgeos: on the scroll we have self.dash_content_list.currentIndex.wait_for(original_index + 1), and that's what's failing on maguro.17:00
tsdgeoskarni: we may want to choose or tweak the theme17:00
tsdgeoselopio: ah17:01
tsdgeoskarni: or hardcode some stuff17:01
karnitsdgeos: shall I make it grey temporarily? fonts are now invisible if you ./run newscopes-new-dash-look17:01
tsdgeoskarni: sure, and add a todo17:01
karnitsdgeos: I say I put "grey" in there so we see anything, and leave a todo to update the theme17:01
karniok17:01
tsdgeoskarni: we'll have a look at it next week need to run now17:01
karniok!17:01
tsdgeosdavmor2: can you reproduce the issue? i guess not, right?17:01
elopioit doesn't even get to the point where we wait for the animation to finish and click something else.17:02
elopiothe problem there seems to be that the drag is actually failing to move to the next one.17:02
tsdgeosor that there is no next one17:02
tsdgeosi.e. hasn't loaded yet?17:02
tsdgeosdavmor2: if you can reproduce it'd be good if you can run the test exporting QML_BAD_GUI_RENDER_LOOP=1 just to make sure it's not the awful 5.0 qml scene graph playing tricks on us17:03
* tsdgeos really has to run now17:04
tsdgeostil morning!17:04
davmor2tsdgeos: I have run two sets of test let me grab the ouput from that first run got  http://paste.ubuntu.com/6891542/ and  the second got http://paste.ubuntu.com/6891973/17:08
elopiomterry: https://code.launchpad.net/~elopio/unity8/fix1277591-open_scope_scrolling_waiting_for_wrong_property/+merge/20542317:12
elopiodavmor2: if you could capture a video of what unity is doing when you get that failures, it'd be awesome.17:13
davmor2elopio: HA!   that will definitely have to wait till Monday17:14
mterryelopio, grabbing lunch but will review your branches after17:18
elopiomterry: that's just fine, thanks.17:20
=== alan_g|vt is now known as alan_g
=== karni is now known as karni-lunch
=== dandrader|lunch is now known as dandrader
=== karni-lunch is now known as karni
* greyback eow o/18:24
=== dandrader is now known as dandrader|bbl
cwaynehm, my theming is all messed up after updating20:43

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!