[07:25] tsdgeos, hey, one thing I noticed on overview → go to Ubuntu Store, bottom swipe (so temp scope to overview) [07:25] tsdgeos, the overview seems to cross-fade with... the dash? [07:25] tsdgeos, feels like it should just be there at full-opacity already [07:26] let me see [07:28] tsdgeos, minor nitpick, Icon { name: "ubuntu-store-symbolic", color: "white" } would give you the app store icon (for the "Store" button in the bottom bar) colourized [07:32] ok [07:36] meh just when I thought the issue with lost input is gone, it happens again [07:38] Saviq: i don't get any scope :S [07:38] mzanetti, yeah, it's not :| [07:38] tsdgeos, ?? [07:38] tsdgeos, silo 11? [07:38] no [07:38] distro + dash_overview [07:39] uh oh [07:40] http://paste.ubuntu.com/7986661/ [07:41] good day for mhr3 not being with us anymore [07:43] ok, i manually started smart-scopes-proxy and seems better now [07:44] Saviq: does "App Store" button work for you in overview? [07:44] does nothing here [07:44] :S [07:44] tsdgeos, yes, works fine [07:44] oh man [07:45] let's get silo 11 in case it has something for that [07:45] which i don't think [07:46] tsdgeos, yeah shouldn't matter [07:46] * Saviq tries on desktop with no silo 11 [07:46] maybe my scopes are just borked [07:46] and i need a restart or something to work [07:47] i'm just doing performQuery [07:47] and expecting the scopes scope to come back to me with a "oh sure let's show this scope" [07:47] tsdgeos, did you restart scope-registry as well? [07:47] sure [07:49] Saviq, http://paste.ubuntu.com/7986711/ [07:49] card_emblems :( [07:49] Cimi, they will all be the same conflict most probably [07:50] nothing [07:50] let me try weaving [07:50] Cimi, but not sure what you want me to do, that's life [07:50] let me try to restart :S [07:50] Saviq, kill me would be nic [07:50] nice [07:52] Saviq, I work on branches for month, other branches get reviewed but not mines [07:52] so I spend ages fixing conflicts and fixing bugs that those branches introduce [07:52] Cimi, dude, I did review that before, you were away [07:52] Cimi, stop whining, all of us have to deal with conflicts [07:53] Cimi, and sorry if your branch has lower priority than other, large merges, but that's just how it is [07:53] Cimi, there's 10 of us working on the same project, there's no way to not have conflicts [07:56] Cimi, please fix those, I'll review your branch today, was planning to anyway [07:56] niet [07:57] no workie for the click in scope in overview [07:57] Cimi, and sorry for being blunt, but it's really not nice to hear you complain about conflicts for 2 weeks every day [07:57] Cimi, even if it's not true, that's how it feels :/ [07:57] Saviq, because my branch sits in the dust and I have to do every day extra work [07:57] I'd rather work on the features instead fixing conflicts [07:57] Cimi, care to see others' branches how long they've been in the queue? [07:58] Cimi, do you think someone's targeting your branches in particular? [07:58] hah no :) [07:59] tsdgeos, works here [07:59] yeah [07:59] i'll debug it [07:59] see what's happening [07:59] Cimi, imagine tsdgeos when he had to rebase dash overview on top of dash as app, think that was an easy task? did he mention it once? [08:00] tsdgeos, actually the "temp to overview" dash leaking through doesn't happen on desktop ?¿ [08:00] Saviq, I'd complain if it wasn't probably the 5th time I have to rebase emblems... [08:00] hmm or on mako [08:00] I wouldn't [08:01] Cimi, it's bad timing, is all, you came back during the sprint, when all of those things were already happening [08:01] Saviq: ?¿ [08:01] tsdgeos, ooor on my other device, any more... [08:02] :D [08:02] and back [08:02] whoa that's confusing [08:03] apps → store → overview → store... hmm, why can't I bottom-swipe any more? ah, because that is opened on top of the overview... [08:03] the navigation dots are quite ugly, were they design approved? [08:03] Cimi, the bar is a wrong colour [08:03] ok [08:03] Cimi, and they will have to change between light and dark based on the bar colour [08:04] Cimi, so yeah, not compleet [08:04] Saviq: aha, i know why it doesn't work for me [08:04] because i have the store as a "favorite" [08:05] tsdgeos, [08:05] so fixing that [08:07] tsdgeos, ok yeah, dash leaking through is happening until something fixes it... until something breaks it again [08:07] tsdgeos, like start → store → overview already shows it [08:09] booo [08:09] ok [08:09] let me see [08:09] that worked fine at some point [08:09] tsdgeos, I'm not even sure we need the opacity change there at all... [08:09] that probably broke when i did the shader thing [08:10] well, that probably is the shader thing basically [08:10] that shouldn't be showing at all :D [08:10] yeah, thought so [08:10] yeap [08:11] yeah refactoring [08:11] hehe [08:12] ok pushed [08:12] should be better now === vrruiz_ is now known as rvr [08:40] +1 [08:47] Saviq: handle is in [08:47] tsdgeos, awesomes, checking [08:47] looks a bit out of place tbh [08:48] tell me what you think and how you'd improve it [08:48] tsdgeos, well, the asset looks wrong for one [08:49] tsdgeos, is visible in preview where you can't swipe [08:49] tsdgeos, possible to fade out on bottom swipe instead of invisible? [08:50] Saviq: looks ultra bad [08:50] Saviq: comment out the enabled: overviewController.progress == 0 [08:50] in DAsh.qml [08:50] Saviq: well it's the asset i got :D [08:52] tsdgeos, sure ;) [08:52] Saviq: preview enabledness removed [08:54] ktx [08:54] Saviq, conflicts resolved [08:55] Cimi, k, will review asap [08:55] thanks :) [08:58] tsdgeos, in temp scopes it doesn't go away when you scroll up [08:58] no? [08:58] that's weird [08:58] or down [08:58] whichever way that is [08:59] yep [08:59] checking [08:59] tsdgeos, maybe it should slide down out of view instead of opacity? [09:00] Saviq: i think opacity works good enough for the vertical movement [09:00] Saviq: but maybe down works for when actually opening the overview? [09:00] so you have two different behaviour [09:00] becuase it's using it vs "hiding" it? [09:00] tsdgeos, yeah that might be [09:01] let me see why temp scopes doesn't work [09:01] kk [09:01] and i'll impement both [09:01] and then we can play with it [09:04] Saviq: temp scope fixed [09:07] kk [09:09] mzanetti, when you want to talk... [09:10] Cimi: need to try some more things... [09:11] ok [09:11] dandrader: hey, is your branch already in a state where I could use it as a prereq? [09:11] Saviq, https://code.launchpad.net/~fboucault/ubuntu-ui-toolkit/flickable_right_speed/+merge/229343 [09:11] mzanetti, yes! [09:11] cool [09:11] Cimi, yeah, I know, what about it? [09:11] tsdgeos, ideas? https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1348557 [09:11] Ubuntu bug 1348557 in Ubuntu UI Toolkit "Make scrolling speed resolution independent" [High,Confirmed] [09:12] Saviq, cannot be implemented that way [09:12] mzanetti, just finishing updating the vkb mock and will propose it [09:12] Saviq, florian ran out of ideas [09:12] ah, cool. I'll do my daily review then in the meantime [09:12] Cimi, ugh, attached properties don't go through? [09:12] right, damn [09:12] although weird [09:13] apparently only patching qt [09:15] Cimi, TBH I think that's really what we should do, actually fix the original bug upstream, not work around it [09:23] Saviq, should be a RTM requirementr [09:25] Saviq: shouldn't this happen automagically? https://code.launchpad.net/~pitti/unity8/update-pot/+merge/230072 [09:26] mzanetti, if we do this on every push, you'll get the comment change all the time [09:26] mzanetti, with line numbers [09:26] mzanetti, I think ideally I'd like that to happen per release [09:26] Saviq: no... I meant some launchpad bot doing this [09:26] mzanetti, no, it only does translations [09:27] ah [09:27] mzanetti, it can't do .pot file since it can't know how to do it [09:27] mzanetti, but yeah, a pre-build hook in train would probably be best for this [09:27] * Saviq files a bug [09:27] ah... right... that's why KDE has the convention of a "messages.sh" script in each repo, so the bot knows what to do [09:28] Saviq: anyways, I'll approve that branch then... looks ok [09:28] yup [09:34] tsdgeos, shall I rebuild the silo or are you doing the tab show/hide? [09:34] Saviq: ah, i already did that [09:34] tsdgeos, ok, pulling ;) [09:35] tsdgeos, sounds like you're off the hook for overview now, then [09:35] Saviq: until you find more stuff to fix ;) [09:35] tsdgeos, yeah, but I didn't ;) [09:36] ok, let's see if i can get a new image on $thephone and SIM unlocking works [09:36] tsdgeos, FWIW I wouldn't have made the tab "follow" the gesture, rather a binary shown/hidden, but that's minor [09:37] it was easir this way :D [09:37] tsdgeos, how difficult do you think it would be to hook up the activity indicator to the overview? [09:37] tsdgeos, ok, fine [09:37] Saviq: in which regard, for search? [09:37] tsdgeos, yes [09:38] hopefully not much [09:38] let me see [09:39] tsdgeos, it should be as easy as || overview.processing [09:39] yeah [09:41] yeah done [09:42] awesomes [09:45] Saviq: hey, about this: https://code.launchpad.net/~macslow/unity8/fix-1305885/+merge/227726 [09:45] Saviq: Mirco says we can't have qmltests for this kind of stuff [09:46] which seems odd... [09:46] Saviq: should we approve it nevertheless and add a TODO item for after RTM to properly mock out the stuff? [09:47] mzanetti, yeah [09:47] ack [09:47] tsdgeos, looks great, nothing else I can see broken [09:47] lol... our alarm ringtone really sounds like a lullaby [09:48] mzanetti, Saviq: also added a corresponding comment [09:49] Saviq: great! === shiznix_ is now known as shiznix [09:50] mzanetti, how do you delete an alarm? [09:50] Cimi: --wipe :D [09:50] Cimi: dunno tbh :) [09:50] ahah [09:50] indeed [09:50] what a bad UX [09:51] Cimi: swipe it to the right [09:51] undiscoverable [09:52] and many times you flick down instead right [09:52] it should be in a repeater not listview [09:53] Cimi: yeah, right [09:53] Cimi: however, the clock app is being "rebooted" atm [09:53] as in rewritten from scratch [09:56] Cimi, hmm, I removed emblem in tryCard from the data part (not template), emblem still there? [09:57] Saviq, emblems everywhere! [09:57] Saviq, I'll have a look... [09:57] Cimi, also, if emblem is defined, you always leave the space for it (wrap title where emblem would be) [09:57] Cimi, we can have a category with optional emblems, and we should only wrap title around the emblem where it's actually there, should be 0-wide otherwise [09:58] Cimi, also tryDash broken? [09:59] god [09:59] /food [09:59] all broken now [10:01] lol [10:01] i rebooted the phone [10:01] opened the greeter "too soon" [10:02] and got the "Unity Dash" booting dots animating [10:03] Saviq, testDash breaks because of the change to attributes we did [10:04] var hasAttributes = hasTitle && components["attributes"] || false; -> var hasAttributes = hasTitle && components["attributes"]["field"] || false; [10:06] Cimi: are you basing your stuff on top of dash_overview? or trunk? [10:08] tsdgeos, trunkl [10:09] MacSlow: hmm... I just realize you're reaching out of context again... [10:09] MacSlow: there is no "notification" in NotificationMenuItemFactory.qml [10:09] Cimi: you may try dash_overview, it has some substantial changes too and is next-thing-to-land afaik [10:09] mzanetti, ? [10:09] Cimi: but confirm with Saviq first [10:10] mzanetti, being on another branch atm [10:10] tsdgeos, we need to move the processing indicator in Components [10:10] have you done that already? [10:10] MacSlow: well, open NotificationMenuItemFactory.qml and search for "notification" [10:10] there isn't such a thing [10:10] yet you're accessing it [10:11] which is bad practice and makes the code a mess in the long run [10:11] why not just adding the accepted() signal to the NotificationMenuItemFactory and connecting to that? [10:11] mzanetti, well the whole UnityMenuModel is something that's not meant to be kept around in notifications [10:14] make testDash fails on trunk for me [10:14] let me try to dist-upgrade first.. [10:26] yeah testDash is broken in trunk... [10:26] qmltestrunner::Dash::test_processing_indicator() and qmltestrunner::Dash::test_setCurrentScope() [10:27] Cimi: fix it :P [10:29] waiting saviq for a question on attributes [10:31] tsdgeos, are you exporting the processing indicator to Components? [10:32] or somewhere under qml/Dash/ ? [10:32] Cimi: which processing indicator? [10:32] the orange line? [10:32] tsdgeos, yup [10:32] there is only one [10:32] doesn't need to be exported [10:32] tsdgeos, we need it in the overview [10:32] it is in the overview [10:32] cool [10:32] there is only one [10:32] anyway [10:33] tsdgeos, thought it was outside of it [10:33] fine then [10:33] it is outside of it [10:33] there's only one [10:33] but it tells it to show [10:36] Cimi: not sure i'm explaining myself tbh [10:39] tsdgeos, when is your branch going to be merged? (really)? [10:39] tsdgeos, wondering if scope settings will be reviewed/merged before yours or not [10:40] and which needs to rebase... [10:40] Cimi: before dash_overview? [10:40] dash_overview should go in today afaik [10:40] ok [10:40] there is lot of things connected [10:40] I'll rebase on yours [10:41] Cimi, not sure what guidance do you need from me? [10:41] Saviq, testDash is broken in trunk for me [10:41] can you test? [10:41] could be due to scope v4 or dash indicator [10:41] then I have another quesitojn [10:42] Cimi, indeed two failures [10:42] Cimi, I think those get fixed with dash overview though [10:42] ehm... folks... -> http://pastebin.ubuntu.com/7987781 [10:42] we said components["attributes"] were always defined, and the way to distinguish was with components["attributes"]["field"], but in testDash this fails, where can we see how the mock processes the data? [10:42] MacSlow, yeah, merge trunk [10:43] trying to compile unity8 and really just branched [10:43] Saviq, I'm pretty sure I did just that... [10:43] mzanetti, dist-upgrade [10:43] MacSlow, trunk already asks for v4 of the scopes api, so you're not up to date [10:43] MacSlow, you may try removing builddir/CMakeCache.txt in case cmake got confused [10:43] Cimi, he is upgraded, but cmake's looking for v3 instead of v4 [10:44] Saviq, on components["attributes"]["field"] ? [10:45] Cimi, fake_categories.cpp needs to always return ["attributes"] [10:45] ok [10:45] I'll add that to card emblems [10:47] Saviq, which value for that? [10:47] Cimi, just an empty map will suffice [10:47] Cimi, see what art is doing [10:48] gatox, hey, you pung yesterday, you got along with the indicator in the end? [10:48] Cimi, yeah, testDash is fixed in dash overview [10:50] Saviq, yes, thx, i was able to find it in unity [10:52] Saviq, I've been able to compile unity8 now... but I can't start it any more... -> http://pastebin.ubuntu.com/7987829 (also deleting that file LightDM) doesn't help. What changed in the way unity8 is meant to be started? [10:52] gatox, truth be told I didn't understand your branch from a quick look, was it that the Payments button does give up a "processing" boolean that we should show the activity indicator on? [10:53] MacSlow, you don't have unity8 installed [10:53] MacSlow, you need it installed or copy the data/unity8{,-dash}.conf to ~/.config/upstart/ [10:53] MacSlow, but ./run.sh isn't working too great yet with the dash as separate process, I've a branch that needs some love [10:54] Saviq, hm... didn't touch my ~/.config/upstart in weeks... wonder how the unity8.conf was wiped [10:55] MacSlow, you don't need it there if unity8 is installed, you must've removed it somehow [10:55] Saviq, i propose the branch to have jenkins building it for me, because i was having issues building it on the phone yesterday... i wasn't able to test it yet, but i think it works... the idea is to have a signal to show the overlay indicator as when the preview is loading, when the preview is launching the pay-service or anything in the future, because right now opening applications take a while (and pay service needs to open pay-ui and [10:55] sometimes online accounts), and for some moments you have a preview where it seems nothing is going on, and it's confusing for the user [10:56] gatox, yeah, thing is that should already be happening, at least for actions that involve the scope (like pressing install) [10:56] gatox, I'll have a closer look, thanks [10:58] Saviq, ah ok, so this branch shouldn't be necessary?? let me know if we need to do something or wait for that to happen in unity, thx [11:01] gatox, this might be a special case actually (pay being a special button, opening urls taking a moment) [11:01] gatox, so it might still make sense indeed [11:03] Saviq, fixed categories mock [11:03] in emblems [11:09] Saviq, will we merge overview today? [11:09] Saviq, wondering where to rebase settings === facundobatista_ is now known as facundobatista [11:10] Cimi, rebase on overview, yeah, even if not today, that's the next thing that's going in [11:10] Saviq, yes but if albert is holiday next week [11:10] Saviq, we might have settings in before overview [11:10] Cimi, no, because we won't wait for Albert to fix any issues in there, we need this in asap [11:10] Holas [11:10] ok [11:10] \o [11:11] dednick, on that note, dash overview could use a re-review, would you be able to do it today? [11:12] Saviq: can do [11:14] gtg guys, mobo replacement [11:14] biab === MacSlow is now known as MacSlow|lunch [11:17] pete-woods: ping [11:20] greyback, Saviq, any success on the date time issue? [11:21] olli: I've tracked down root cause to be a workaround I added to qtubuntu for bug 1346633 [11:21] bug 1346633 in QtMir "[enhancement] Autopilot testing needs to know screen coordinates of widgets on screen" [Undecided,New] https://launchpad.net/bugs/1346633 [11:21] olli: so I'm working on a different workaround [11:22] heh [11:22] greyback, thx! [11:22] how bad is it... hours, days...? [11:22] olli: I'll keep you posted. In the mean time, SK guys has a workaround ready https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/pickerpanel-workaround/+merge/230065 [11:22] olli: hours hopefuly [11:23] greyback, OK [11:25] Saviq, just to be sure.... should i keep working on my branch, testing it, etc? or should wait until you tell me that is being covered or not by unity already? [11:26] tsdgeos: hi [11:27] pete-woods: your'e the usermetrics guy, right? [11:27] tsdgeos: yep [11:27] pete-woods: i see in the metrics manager you accept a domain [11:27] is it mandatory that messages go untranslated and you translate them [11:27] or can messages come in translated too? [11:27] or is domain not the translation domain? [11:27] :D [11:27] tsdgeos: nope, feel free to do whatever really [11:28] you're correct, it is the translation domain [11:28] ok, so "" and a translated mesage is also ok [11:28] tsdgeos: it's preferable to do untranslated [11:28] as if you change language then things will be translated [11:28] but it's not the end of the world if it comes in translated [11:29] pete-woods: will a reboot fix it? [11:29] tsdgeos: to fix it would require whoever is writing the data to re-run [11:29] ah [11:29] ok, so better untranslated + domain [11:29] tx === dandrader is now known as dandrader|afk [11:44] can I create a binding from the loaded component of a loader to its loader? [11:47] Saviq: could I steal 10 mins of your time? [11:47] I'd need some second opinions [11:48] Cimi: sure === lool- is now known as lool [11:49] can I do sth like readonly property bool processing: item.hasOwnProperty("processing") ? item.processing : false ? [11:49] Loader { readonly property bool processing: item.hasOwnProperty("processing") ? item.processing : false } [11:50] creating binding inside onLoaded breaks on unloaded state [11:51] Cimi: yeah, should work I guess. You might want to add "item && item.hasOwnProp..." [11:52] oh yeah correct [11:52] or maybe status === Loader.Ready instead item? [11:52] whatever you prefer [11:52] should end up being the same [11:53] Saviq, ping? [11:58] on one of my Unity 8 desktops all I get is a box that says "Hello" and has a text input ... anyone know what that's all about? === alan_g is now known as alan_g|lunch [12:10] dandrader|afk: when you get back, I need your input on https://code.launchpad.net/~gerboland/qtubuntu/fix_1351024/+merge/230094 ASAP [12:20] bregma: I think that's a password prompt [12:20] I get that when I run unity8 directly atm [12:21] not sure exactly why, though [12:21] well, it lack a certain amount of usefulness at the moment [12:22] fortunately my other test machine always brings up the online accounts app, which barfs all over and dies immediately, bypassing the greeter/lockscreen and ringing up the dash [12:22] sometimes three wrongs do make a right === MacSlow|lunch is now known as MacSlow [12:29] Saviq, pete-woods some settings seems to be undefined by default value... what shall we do in the UI? true or false? [12:30] unless model.value is expected to be undefined and we want .defaultValue or so === greyback is now known as greyback|lunch [12:33] back [12:34] mzanetti, gatox, back [12:36] Saviq, hi... i just wanted to know if i should keep working on that branch or expect unity to solve the button issue for payment too [12:36] gatox, if in current trunk it doesn't, there's nothing in queue to fix that [12:37] Saviq, ack, i'll have this working then [12:38] Cimi: you shouldn't have to handle it in the UI [12:38] Cimi: which settings, specifically are like that? [12:40] pete-woods: is there a plan to update the youtube scope in the store to the new api? [12:40] or are you waiting for a promoted image [12:40] cwayne: it's already been updated [12:41] it has been in the store a while now [12:41] huh, i didnt see an update, weird [12:41] oh well, thanks :D [12:41] np :) [12:41] 1.0.7 is what you're looking for [12:45] excellent, thanks again [12:45] Cimi: do you expect emblems to land with dash-overview? [12:45] cwayne, not up to me [12:45] cwayne, under reviews.. [12:45] pete-woods, youtube model.value [12:45] Cimi, did you fix the issues I mentioned? [12:45] in emblems? [12:46] Cimi: okay, this is in mediascanner I take it? [12:46] Saviq, which one? [12:46] pete-woods, no idea [12:47] I have this show youtube in video scope true | false [12:47] model.value is undefined for it [12:47] unless i set it [12:48] Saviq, ah the 0 padding? [12:48] Cimi: okay, reproduced it, seems to be a bug [12:48] Saviq, I try it in half hour [12:48] taking lunch break === dandrader|afk is now known as dandrader [12:50] Cimi, and the fact that it didn't "go away" when I edited the data field in tryCard [12:50] Cimi: okay, it's because there was a bug in the previous version of the settings [12:50] Cimi: leading to corrupt settings db -> rm -rf ~/.local/share/videoaggregator/ [12:51] greyback|lunch, still need my input there or does Timo's review do it/ [12:51] ? [12:54] MacSlow, like the branch much more without the regex ;) [12:54] does anybody know if there is a way to indicate that a widget should NOT have a preview? that is, ignore when the user does a long tap on it; thanks! [12:54] Saviq, I've updated the relevant branches for the text-filter... it needs another look now (https://code.launchpad.net/~macslow/unity-notifications/fix-1335787/+merge/227334, https://code.launchpad.net/~macslow/unity8/fix-1335787/+merge/230087) [12:54] Saviq, I understand... still I had to skip 7 tests which can't be covered (converted) via QTextDocument [12:55] MacSlow, on the test, please use the _data approach as we do in QML [12:56] MacSlow, an old 4.8 tutorial http://qt-project.org/doc/qt-4.8/qtestlib-tutorial2.html [12:57] facundobatista: ping [12:57] tsdgeos, pong [12:57] facundobatista: how do i translate the 7digital scope [12:57] ? [12:57] tsdgeos, lp:ubuntu-rest-scopes, see "locale" dir [12:58] tsdgeos, it uses gettext [12:58] greyback|lunch, will review it as well anyway [13:00] dandrader: yes please review anyway, since oyu know the code [13:02] facundobatista: oki [13:03] tsdgeos, any doubt there, just let me know [13:06] MacSlow, FWIW, this kind of "I may be plaintext, but I also can be html" is just broken by design [13:06] Saviq, yeah [13:06] facundobatista: https://code.launchpad.net/~aacid/ubuntu-rest-scopes/update_pot/+merge/230105 [13:06] MacSlow, either you accept HTML and all <> need to be escaped, or not, and they will be broken [13:06] MacSlow, so I'm actually happy with the results you got with QTextDocument [13:07] Saviq, yeah [13:07] tsdgeos, what is that? [13:07] Saviq, just finished adapting the test to the _data() scheme [13:07] facundobatista: updates the pot file [13:07] so we can actually translate stuff [13:07] mmm [13:08] facundobatista: basically i ran ./build/get_translatable_texts.sh [13:08] Saviq, pushed [13:08] MacSlow, kk === alan_g|lunch is now known as alan_g [13:09] MacSlow, I think you could actually use the newRow("") to store the actual string [13:10] MacSlow, this way you'll get better feedback on test output on which case failed [13:10] Saviq: how's silo 11 going? [13:11] plan to land today? [13:11] tsdgeos, hoping to [13:11] Saviq, yeah... but blows up the result-output a bit... even if all tests (of text-filter) pass [13:11] tsdgeos, thanks [13:11] Saviq, but then... hardly an issue really :) [13:11] tsdgeos, they were expecting scope favouriting to work, too, had to crush their dreams on that (shoulda kinda have mentioned it) [13:12] MacSlow, blows how? by printing the strings? [13:12] Saviq: oh [13:12] Saviq, well it spits out 26 more lines now... initially it was only one [13:13] MacSlow, well, that's the whole idea ;) [13:13] MacSlow, every case there is now treated as a separate test [13:14] Saviq, certainly looks more impressive now with 36 passed test all together :) [13:14] exactly! [13:17] facundobatista: buaaaarns, some bot unaaccepted my update [13:18] Saviq: on $thephone on first boot seems the shell is displaced like 1 pixel right [13:18] bringing back the greeter usually fixes it [13:19] wrong, blanking the screen fixes it [13:19] tsdgeos, let's check now [13:23] tsdgeos, yeah,known bug [13:23] unknown cause yet [13:23] Saviq: ok, also 5 min ago, restart was not restarting the phone but just the shell [13:23] is it known or shall i file a bug? [13:23] didn't hear that no [13:23] Saviq, I don't know why if I unsed the data you still see the emblem... [13:24] *unset [13:24] problem is that stupid me i shut down [13:24] is weird [13:24] so i can't repro anymore [13:24] if you set to an invalid data it hides, valid one it shows.,.. === greyback|lunch is now known as greyback [13:25] Saviq, might be something not updated: if you set the data to an invalid png then you remove the data, the emblem is gone [13:25] Saviq, if you don't set to an invalid image first and remove the data, it uses the previous emblem [13:26] Cimi, rather something looks at the url and only changes if it evals to true [13:26] Cimi, and "" does not eval to true [13:26] Saviq, same issue with mascot [13:26] Cimi, ok, let's leave that, unimportant as will never happen IRL [13:27] Saviq, however, changing in CardCreator.js fixes that [13:28] source: cardData["emblem"]; -> source: cardData["emblem"] || ""; [13:28] we can do the same for mascot [13:28] source: cardData["mascot"]; -> source: cardData["mascot"] || ""; [13:28] Saviq, could I get you or one of your evil minions to review https://code.launchpad.net/~bregma/unity8/merge-desktop-upstart-job/+merge/230016 so we can get the Unity 8 desktop back up and on to its next shock of the unknown ASAP? [13:29] and, possibly include the MP in the next Unity 8 ci-train landing? [13:29] Cimi, sounds good, yeah, maybe the json parser gives up null or undefined and source isn't reset with that [13:29] Cimi, so good fix [13:29] bregma, will do [13:29] bregma, ah on that note, can you drop the surfaceflinger part?\ [13:29] Saviq, in case of null image, shall we reserve space or not? [13:30] Cimi, no [13:30] I could, but I have no way to test that [13:30] bregma, there isn't a session like that any more :) [13:30] Saviq, mascot reserves space though [13:30] bregma, and I will test it [13:30] Saviq, try messing up with the url; [13:30] Cimi, I don't think it does in fact (it should not) [13:30] Cimi, if it does - bug [13:31] Saviq, width and height are static for mascot [13:31] width: units.gu(6); \n height: units.gu(5.625); \n\ [13:32] Cimi, feels like a bug [13:32] I can do sth like source status === Image.Ready ? .. : 0 [13:32] without source [13:32] I can fix that in emblems if you like [13:33] (fixed) [13:49] Cimi, yeah, invalid image should just not exist in there, if we managed to get it done with GridLayout or something, that would've happened automagically [13:49] Saviq, not fixed for mascot [13:49] thoght was easy... the calculations with sourceSize causes binding loops [13:49] cause [13:50] are there any scopes that currently utilize the alt nav support? [13:50] cwayne, it's not landed yet in unity8 at least ;) [13:51] Saviq: well we can still see the code and see if it makes sense to start integrating them into our scopes :) === _salem is now known as salem_ [13:51] cwayne, sure, dunno though [13:51] tsdgeos, did you see any real scope using dual nav? [13:52] Saviq: nope :/ [13:52] cwayne: I've seen the amazon scope doing it [13:52] Saviq, gonna commit this (and fixes tests that will break) http://paste.ubuntu.com/7989005/ [13:52] all testing is based solely on my fake plugin impl [13:52] cwayne: although obviously that support depends on stuff that hasn't landed yet [13:52] right [13:53] Cimi, from a quick skim looks fine, although are we sure cardData will always be there? [13:53] Cimi, on creation I mean [13:54] Saviq, I thought it was [13:54] Saviq, you told me we always create it or not? [13:54] might have misunderstood [13:54] Cimi, yeah, we do, but it might not be bound in time [13:54] Cimi, like if that binding evaluates before cardData is bound from the result [13:54] Saviq, because I removed it from attributes [13:55] I can add them everywhere [13:55] Cimi, yeah, now I think about it again, we need at least a check for cardData, because that might be null on creation [13:55] ok [13:55] adding all of them [13:55] Cimi, if it's defined then ["attributes"] will be there too [13:55] I mean always [13:55] so we don't need to check for that, but the cardData object itself might be null on creation [13:59] Saviq, all pushed - apart the fix for not allocating 0 size in case of broken mascot image [14:00] Cimi, yeah, ok, unrelated bug [14:00] (which requires lil more thought) [14:00] jeez that grew [14:00] Cimi, whitespaces again :| [14:01] oh fuck [14:01] Cimi, can you add a prerequisite that removes all the spaces and ; from cardCreator [14:01] damn qtcreATOR [14:01] ok [14:01] (; where it doesn't make sense, so outside of JS) [14:02] nope [14:02] let me revert that [14:03] Saviq, but why we added whitespace at end of line in first place? [14:03] Cimi, readability in cardCreator I expect [14:03] Cimi, so that you don't get lines that all end with \n\ [14:03] ok [14:03] or actually end with\n\ [14:04] Cimi, the card creator tests leave the result as a file on failure, shouldn't you just copy those onto the expected ones [14:04] Cimi, and check the diff that it makes sense [14:09] Saviq, spacing issues fixed I think [14:09] diff seems sane now on LP [14:10] Cimi, still 1200 lines for a small emblem ;P [14:10] pffff [14:10] :D [14:11] Cimi, '' vs. "" changes?? [14:11] 200 - var mascotShapeCode = ""; [14:11] 201 - var mascotCode = ""; [14:11] 202 + var mascotShapeCode = ''; [14:11] 203 + var mascotCode = ''; [14:11] Saviq, code consistency [14:11] Saviq, we always used '' apart here [14:11] Cimi, you really don't want this to be reviewed do you [14:11] ? [14:12] what's the problem? [14:12] Cimi, having an MP with a bunch of unrelated changes doesn't help review speed [14:12] bunch = 2 lines? [14:12] Cimi, there's a whole slew of renames too [14:12] and there's more "" to '' elsewhere [14:13] Saviq, you're right about the '' [14:13] Saviq, but it's a easy review [14:13] Cimi, well, it's not easy any more :P [14:17] tsdgeos, you stuck in SIM land or can help with hopefully a small thing? [14:17] Saviq: i am trying to rebuild gallery not to crash [14:17] but if it's quick, sure [14:18] tsdgeos, we need to forward processing from GSVs open on top of overview [14:18] tsdgeos, if you open a temp scope in overview, no processing indicator [14:18] tsdgeos, should be another || gsv.processing [14:18] or so [14:18] correct [14:18] let me see [14:21] Saviq: pushed [14:21] tsdgeos, awesomes, thanks [14:22] mzanetti, could I volunteer you for https://code.launchpad.net/~mterry/unity8/wrong-password-handling/+merge/229184 please? [14:26] another one for evaluation: keyboard persists after you open a scope from search [14:26] tsdgeos, ↑ [14:26] tsdgeos, if you think easy, please fix, otherwise we'll do it [14:26] i fixed that [14:27] someone broke it :/ [14:27] bastards [14:27] was probably me :D [14:28] with other page header changes [14:28] let me see if it's easy to fix [14:36] Cimi, nice timing ;) [14:36] Saviq, hah [14:37] Saviq, bloody xchat without highlights for words with @ [14:37] missed the ping [14:38] Saviq: which silo is dash overview? [14:38] or tsdgeos: ^ [14:39] dednick: 11 [14:39] tsdgeos: ta [14:40] dednick, there's a few fixes on top of the silo now [14:40] smaller ones [14:42] Saviq: if you want any extra help testing that silo, just give me a shout [14:43] pete-woods, we've plenty of hands, it's going into user testing on Monday ;) [14:43] pete-woods, but thanks, you can test it out anyway [14:45] Cimi, were you not rebasing emblems on dash_overview? [14:45] Saviq, no [14:45] okay, just wanted to help (this silo is particularly exciting) [14:45] Saviq, scope settings [14:45] Cimi, ah [14:46] Cimi, looks like you want to rebase emblems, too, 3 conflicts [14:46] Saviq, or dash overview... [14:46] :D [14:46] Cimi, or... not [14:46] Saviq, which one gets first? [14:46] Cimi, overview [14:46] ok [14:47] Cimi, small, easy conflicts though [14:48] Saviq, dash overview has whitespace changes too [14:49] Cimi, are we really playing the blame game now? === gatox is now known as gatox_lunch [14:49] Saviq, yeah! [14:49] Saviq, anyway pushing now [14:49] Cimi, well, I'm not [14:50] Cimi, did you have to rename anchors to artAnchors etc? were they clashing? [14:51] Saviq, some of them were [14:51] Saviq, so I changed name for readability [14:51] Cimi, and btw , ' is only used when there's quotes inside the string already, to avoid escaping it [14:51] dednick, any qmltest for the prompt session stuff? [14:51] Cimi, everywhere else we actually use " [14:52] Cimi, so TBH I'd rather ' be an exception than a rule in that particular file [14:52] Saviq, I asked that question at the sprint [14:52] difference / what we prefer [14:52] Cimi, it doesn't make a difference, so *changing* it doesn't make sense to me [14:52] dandrader: not at the moment. [14:52] just for the sake of changing it [14:53] dandrader: on my todo list [14:53] especially in a branch that doesn't deal with that wholesome, but actually implements things [14:58] Saviq, https://code.launchpad.net/~cimi/unity8/card_emblems_2/+merge/230126 [14:58] rebased [14:59] Cimi, ok, I'm reviewing the original one first anyway [14:59] Saviq, i can change '' nack to "" [14:59] back [14:59] Cimi, don't, just don't do such changes unnecessarily, they fit in a separate MP that only deals with such changes [14:59] ok [15:00] Cimi, maybe remove the superseding MP for now, I'm halfway through the review in the non-superseded one [15:00] Cimi, this way comments should go over [15:00] Cimi, or wait, maybe they'll go over anyway, let's see [15:05] opinion wanted: do we consider the orange line under the indicators bar to be part of the panel? Or is it separate? [15:06] greyback, I'd say panel [15:07] Cimi: me too [15:10] Cimi, reviewied [15:10] -i [15:11] greyback, panel for sure [15:11] Saviq: ok [15:11] Saviq, When emblem is declared, but invalid, there's no right text margin. ? [15:11] Cimi, text reaches right edge [15:11] Saviq: commited the unfocus [15:11] has 1 gu here [15:11] Cimi, should have 1gu margin on the right [15:11] maybe in particular cases [15:11] Cimi, when emblem invalid or not there? [15:12] panelHeight is not including the orange line, and so we're actually drawing over the top 2DP of app surfaces [15:12] Cimi, it's fine when it's not declared [15:12] greyback, ouch [15:12] tsdgeos, thanks [15:12] Saviq, you right [15:12] I right! [15:12] Saviq, font hinting got me [15:13] thought was padding [15:13] alecu: ping [15:13] Cimi, we have a bug there btw, there should be no margins if there's no background and touchdown only covers art [15:13] tsdgeos: hi [15:13] alecu: i'm pretty sure that this is wrong [15:14] alecu: http://paste.ubuntu.com/7989543/ [15:14] you can't have "static" + translations [15:14] since the translation will get run on "start" time [15:14] before anything is set [15:14] and you end up with no translation [15:14] which is what i'm seeing [15:14] am i making sense? [15:15] tsdgeos: yes, I think I understand [15:15] want a patch? [15:15] tsdgeos: sure! or a bug should be enough [15:15] i have 45 min without much to do [15:15] so i'll give you a patch :D [15:16] Saviq, no bg? [15:16] greyback, So I'm looking at app startup, and I have a case where it's taking Unity 640ms to execute the focus observers. Do you know why that could be? [15:16] Cimi, say art + title [15:16] tsdgeos: thanks! [15:17] Cimi, title is unnecessarily squeezed 2gus from the sides [15:17] Cimi, when there's no background and touchdown will only affect art [15:19] Saviq, paste me components? [15:19] tedg: focus observers? Can you elaborate please. Some data would be nice [15:19] alecu: hmm, i'll open a bug :D [15:19] greyback, The observer for the focus and starting events with UAL. [15:19] tedg: I always like questions along the lines "your code is slow, do you know why?" :D [15:19] alecu: not as easy as i thought [15:19] greyback, Let me put the data somewhere [15:19] ok got it [15:20] Cimi, but that bug's in trunk anyway [15:20] how should is bt? [15:20] tedg: thanks. Did you acquire it with that method your blogged about? [15:20] http://gould.cx/ted/blog/Measuring_Upstart_App_Launch_using_LTTng [15:20] greyback, Yes, I've got a tar of it. [15:20] Saviq, let me see if I can quickly fix it [15:21] alecu: https://bugs.launchpad.net/ubuntu/+source/unity-scope-click/+bug/1354501 [15:22] Ubuntu bug 1354501 in unity-scope-click (Ubuntu) "Do not use class static translatable messages" [Undecided,New] [15:22] greyback, http://chinstrap.canonical.com/~ted/dialer-start-20140808-095853.tar [15:22] tsdgeos: I'm just testing it on my device in spanish, and you are right, it looks very weird. [15:22] greyback, You can just run the untar'd dir: babeltrace dialer-start-20140808-095853 [15:22] tsdgeos: thanks for the bug [15:22] alecu: no prob [15:23] greyback, Looking at the "observer_start" and "observer_finish" events [15:23] screw you SSO === dandrader is now known as dandrader|lunch [15:24] Heh, yes. [15:25] greyback, So this is just clicking on the dialer in the dash === om26er is now known as om26er|dinner [15:28] um. woops. just managed to delete my entire working folder. goodbye code. [15:28] ouch [15:28] tedg: ok interesting. I'll have to dig to see what's so slow there [15:28] good news is that i've got lots of space now. [15:29] greyback, Cool, thanks! [15:32] Saviq, fixed but doesn't look that nice [15:32] Cimi, that is the design everywhere, though, that they line up on the sides [15:33] ok [15:33] I will push [15:38] greyback, So I did a run of the clock app and it only took 65ms. [15:38] greyback, Not sure what the difference is. [15:38] greyback, The only thing I can think of is that clock is a confined/click where dialer is unconfined/legacy. [15:40] tedg: that's an astute observation, can you see if other unconfined apps have the same slowdown? [15:40] greyback, Just a sec, grabbing a couple more dialer runs to see if it was just a fluke. [15:44] greyback, Did two dialer runs, one after another. The first took 600ms, the second less than 60ms. Not sure it's a confinement issue. === gatox_lunch is now known as gatox [15:47] greyback, Hmm, browser took ~150ms. Feel like I'm not getting consistent results, any idea of what I should stop/start/do per run? [15:49] tedg: tbh what you're doing is the most realistic, so we need to figure out why the lack of consistency [15:49] tedg: I think it's probably a good idea to add lttng tracepoints to qtmir [15:51] greyback, Ah, that's a good idea [15:57] Saviq, pushed... required more time to deal with emblems, subtitles ect [16:01] Saviq, Indent please. ? [16:01] your diff comment [16:03] Cimi, there's no indentation in { } [16:03] Saviq, gotcha - fixed [16:04] Saviq, are we sure of emblem requirement? [16:05] Cimi, not sure what you mean === dandrader|lunch is now known as dandrader [16:05] Saviq, using statusicon [16:05] and apparently don't care... stupid ^W [16:05] Cimi, yes, it needs to be colourized [16:05] Cimi, well, you can actually use Icon [16:05] Cimi, passing source: to it [16:05] Cimi, since it's meant to be square [16:06] Saviq, trying [16:09] Saviq, icon does not have source [16:10] Saviq, only set of named images from the iconset [16:12] Saviq, I can try with qt graphical effects [16:16] but we will have to convert the palette colors into HSL then getting just H [16:21] mzanetti a bit of a last minute thing... dash-as-app, dash is foreground, long-left edge swipe, can we ~easily make launcher stay on screen? [16:21] Saviq: hmm... depends a bit on the ~easily [16:22] mzanetti, as in, Monday morning for user testing ;) [16:22] yes please [16:22] mzanetti, they start 9am in London [16:22] I hate it disappears [16:23] line 246 Launcher.qml [16:24] if we have a way to check from launcher if we are in the dash... doable easily [16:24] Saviq, do we? [16:24] Cimi: ApplicationManager.get(0).appId == "unity8-dash" [16:24] Cimi, sure we do, launcher is shell [16:25] Cimi: however, you need to update it manually as "0" doesn't change === om26er|dinner is now known as om26er [16:25] mzanetti, don't we have like focusedApp though? [16:25] mzanetti, being a shortcut to 0? [16:25] err. right [16:25] ApplicationManager.focusedApplicationId [16:26] Saviq: trying to get away from that though [16:26] but I guess ok for the hotfix [16:26] mzanetti, yeah right, we have two focused apps at any given time.. [16:26] (on tablet) [16:26] Saviq: yes, for once. [16:26] Saviq: also, ApplicationManager shouldn't care in the long run [16:27] trueth [16:27] Saviq: we just tell it which apps to suspend [16:28] Cimi: are you on it or should I fix it? [16:28] mzanetti, I am not on it, but that's the line to edit... [16:28] ack [16:28] that you knoew [16:28] knew [16:29] Cimi: wrong line :P [16:29] even wrong file :D [16:29] lol [16:29] ah no... I guess we could just add it here [16:30] :P [16:30] nah... Shell.qml is better [16:31] mzanetti, think so, shouldn't the onDraggingChanged have a different behavior if dash is foreground? [16:31] mzanetti, sure, Launcher shouldn't know directly, but Shell can tell it [16:31] Saviq: which onDraggingChanged [16:31] ? [16:32] Saviq: found it... [16:32] mzanetti, Launcher:244 [16:32] Saviq: why should that have a different behavior? [16:32] mzanetti, maybe I'm just reading it wrong ;) [16:32] Saviq: actually I'm not entirely happy with that code any more... [16:32] mzanetti, anyway, you know your way around it [16:32] Saviq: its a bit of a mix between the launcher knowing about stuff and the shell controlling the launcher [16:32] mzanetti, yup indeed [16:33] mzanetti, we need a hotfix, not a proper solution, for now ;) [16:33] that's the thing... one writes it to not know about anything and the next one hotfixes something into it... [16:33] mzanetti, that won't even land in trunk [16:33] mzanetti, we just need it in the silo [16:33] http://paste.ubuntu.com/7990053 [16:34] Saviq, btw seen I fixed the rest of the things? [16:34] apart the Icon which doesn't support source [16:34] still need to compile and test... [16:34] Cimi, yes it does (ugh, maybe not in trunk yet?) [16:35] Saviq, not on my pc [16:35] it fails [16:35] mzanetti, I still think we want to change the status to visible [16:35] Cimi, https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1284233 [16:35] mzanetti, thus I said line 246 [16:35] Ubuntu bug 1284233 in Ubuntu UI Toolkit "Can't use uris for Icon source" [Undecided,Fix released] [16:35] Cimi: ? [16:35] Cimi, Icon has a source property since Monday [16:36] Saviq, in staging maybe [16:36] :D [16:36] Cimi, RELEASED [16:36] Cimi, 1.1.1179+14.10.20140804-0ubuntu1 [16:36] OKM [16:36] OK [16:36] OKOK [16:36] OK [16:36] :D [16:37] Cimi, https://launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/1.1.1179+14.10.20140804-0ubuntu1 [16:37] it's friday late afternoon, I started at 8:30am and I am well cooked... [16:37] mzanetti, mzanetti a bit of a last minute thing... dash-as-app, dash is foreground, long-left edge swipe, can we ~easily make launcher stay on screen? [16:38] mzanetti, does your code keep the launcher on screen? [16:38] Cimi: yes [16:38] but just figured it keeps it on the screen a bit too often [16:38] cool then [16:40] mzanetti, yeah, it needs to switch to visible state indeed [16:40] mzanetti, it has to be the same exactly like a short swipe [16:40] yep... but there's still an issue [16:41] mzanetti, aanyway [16:41] mzanetti, we have an hour on Monday morning [16:41] before UK wakes up [16:41] not so sure any more if I can do this in an hour :D [16:41] :| === dpm is now known as dpm-afk [16:42] mzanetti, that's fine, if we can't, we can't, they had could've let us know earlier [16:42] well, I'll try [16:42] Saviq, mistery solved - CardCreator.js was importing 0.1 not 1.1 [16:43] muahahaha [16:43] scnr [16:46] ok, I gotta unglue... [16:51] Saviq: Cimi: hotfix... not to be landed for real: http://paste.ubuntu.com/7990208 [16:52] Saviq, Using icon now, added colorisation and fixed png asset for test [16:52] (pushed) [16:53] mzanetti, hah, line 246 :P /troll [16:53] Cimi: yeah... but we really don't want to land this as is [16:54] of course [16:54] mzanetti, also in spread? [16:54] but enjoy weekend guys [16:55] o/ [17:11] * greyback EOW === alan_g is now known as alan_g|EOW [17:24] dednick: hey, still around? === boiko_ is now known as boiko [17:41] mzanetti, aaaghhg /build/buildd/unity8-8.00+14.10.20140808.1/qml/Launcher/Launcher.qml: bad whitespace in line 249 ;P [17:41] we all love that test :) [17:42] mzanetti, I'm fixing [17:42] I'm on it Saviq [17:42] too late [17:42] meh [17:43] mzanetti, aren't you using QtCreator? it strips any trailing whitespace here [17:43] Saviq: I am for most stuff. But for such hotfixes its too much effort to configure qtcreator [17:43] still not using colocated branches [17:44] Saviq: I need to change the builddir, add -j9 to makeopts and if I want run to work add -mousetouch and whatnot [17:44] hence I often just ./build.sh, nano and run.sh [17:46] oh jeez [17:46] mzanetti, well, yeah, I'm not using QtC to build/run anyway, but yeah colo branches FTW [17:47] but nano? ;) [17:47] Saviq: yeah, why not? [17:47] gets the job done [17:48] and I mostly just have to press one key to do something instead of 3 per command :P [17:49] the only nasty thing is on *wrt, where it mostly autocompletes to nandwrite 'cause nano isn't installed :D === om26er_ is now known as om26er [19:44] mzanetti, sudo nano ftw! [20:06] Cimi: ? [20:49] mzanetti, I ran it many times :) === salem_ is now known as _salem