[00:27] greyback, ping! [00:27] cwayne: pong [00:28] greyback, hey, did you want more testing on those unity-mir fixes, or are they pushed? [00:28] cwayne: since I didn't get a mail from you, I pushed the code and have kicked off a build [00:30] should land in about 20-25 mins [00:31] greyback, awesome, thanks! [00:31] cwayne: again, thanks for making sure it was ok [00:34] greyback, np, so this is in the ppa right? not in the archive that is [00:35] cwayne: yeah the demo-stuff ppa. [00:36] perfect [05:06] is there an easy walk thru on how to create a shortcut on unity for a program compiled from source? === jono is now known as Guest99797 [05:10] anyone in here? [05:16] Lots of people. [05:17] Longfellow: You're after a .desktop file; install alacarte and it will help you create one. [05:20] Longfellow: Does your desktop file point to the right thing? Also, this is probably more an #ubuntu question (as it's support) [05:23] yeah i was in #ubuntu and no one responded to my question and saw this channel and thought i'd give it a shot === Wig is now known as Wug [07:28] o/ [07:50] Saviq_: hrm. I see the changes showed up when I dist-upgraded the image, but the branch is not merged. Have you worked on it elsewhere? https://code.launchpad.net/~unity-team/unity8/themeing-font-and-mascot/+merge/207282 [08:11] karni, nope, it's not merged anywhere?? [08:12] Saviq: hrm. I guess someone set the card colors in the Apps scope, that could have confused me [08:12] karni, probably yeah :) [08:12] Saviq: no stress! :) [08:12] Saviq: ok, I'll work on it then [08:14] karni, ok thanks, I started, but it quickly became a little more than I wanted to bite yesterday [08:14] Saviq: any diff you want to send over, or you just started? [08:15] karni, nah, not really [08:15] Saviq: yeah, took me a while yesterday (while doing other stuff in the mean time). I thought it'd be a smaller task. [08:15] okay, I got it then. [08:31] Saviq: which bug/feature/thing should i target next? [08:32] tsdgeos, have a look at https://bugs.launchpad.net/unity8/+bug/1127665 [08:32] Launchpad bug 1127665 in unity8 (Ubuntu) "[SIDESTAGE] Dash isn't visible when only sidestage app running " [Medium,Triaged] [08:32] tsdgeos, which https://bugs.launchpad.net/unity8/+bug/1213034 is related to, too [08:33] Launchpad bug 1213034 in Unity 8 "Can't dismiss keyboard by tapping outside of search entry" [High,Opinion] [08:34] Saviq: i don't see how those two are related [08:34] tsdgeos, well, if there's no dash behind sidestage-only app, dismissing it would be weird [08:34] tsdgeos, if it's there, dismissing it just means you go back to the dash [08:38] Saviq: i'm still sleepy, but i don't get it :D [08:38] one is "dash should be besible when there's not main stage app" [08:38] tsdgeos, sure, they don't really require one another [08:38] other is "i want do dismiss dash search by clicking somewhere out of the dash search" [08:39] tsdgeos, ooopf [08:39] tsdgeos, wrong bug? [08:39] maybe? [08:39] tsdgeos, https://bugs.launchpad.net/unity8/+bug/1129290 [08:39] Launchpad bug 1129290 in Unity 8 "[SIDESTAGE] can't dismiss the sidestage through grab when only sidestage app is running " [Medium,Triaged] [08:39] that's better ;D [08:39] tsdgeos, sorry ;) [08:39] ok [08:39] those are more related [08:40] i'm a bit scared that this is a bit harder than the usual bug and will at least need fixes in some non-unity8 layer [08:40] but having a look [08:41] tsdgeos, yeah, if you see it too complex, let me know [08:51] mzanetti, btw, https://bugs.launchpad.net/ubuntu/+source/ca-certificates/+bug/1258286 [08:51] Launchpad bug 1258286 in nss (Ubuntu Trusty) "CAcert should not be trusted by default" [Undecided,New] [08:51] Saviq: i don't think we can do the first one without ¿heavy? architectural changes. The Shell needs to be on top so it can catch left edge, right edge, but then if it is on top *and* we show the Dash, the Dash will be on top of the sidestage app, no? [08:51] Saviq: which means, we need to make the Dash an app :D [08:51] tsdgeos, well, or clip it [08:52] tsdgeos, that'd be the only "quick" way I can think of [08:52] hmmmm, ok [08:52] can try to see if that works [08:55] Saviq: intersting... a bit sad tho too.. [08:55] mzanetti, yeah, seems they tried to bite too big a cake [08:56] mzanetti, or maybe that model is just incompatible with a trusted CA [08:56] mhr3, hey, on https://bugs.launchpad.net/ubuntu/+source/unity-scopes-shell/+bug/1282249 [08:56] Launchpad bug 1282249 in Unity 8 "After the first preview, actions gets unresponsive" [Critical,Triaged] [08:56] mhr3, what should I react to when you're replacing the preview? modelReset? [08:57] Saviq: ah, btw... I played around with the MWC image for a while yesterday: Previews have a bunch of issues. Are they known of should I report them? [08:57] mzanetti, depends on the issue ;) [08:57] dark overlay at bottom missing [08:58] app preview's screenshots not scrollable horzontally [08:58] https://bugs.launchpad.net/unity8/+bug/1281995 [08:58] Launchpad bug 1281995 in Unity 8 "Previews should have headers and "back" buttons at the top, and no "see through" at the bottom" [Medium,Triaged] [08:58] app preview's author name missing in header [08:58] https://bugs.launchpad.net/unity8/+bug/1281709 [08:58] Launchpad bug 1281709 in Unity 8 "Preview with gallery widget and multiple images: image is cut off" [Undecided,Fix committed] [08:58] Saviq, no, i do remove all + insert [08:58] mhr3, so what do I react to? [08:59] mzanetti, that I don't know about, please file against unity-scope-click [08:59] ack [08:59] Saviq, i don't see why it wouldn't work [08:59] but for the rest it looks freakin awesome. Good job on the new scopes everyone! [09:00] mhr3, it's not that [09:00] mhr3, we're triggering a "please wait" layer [09:00] mhr3, on action activation [09:00] (which btw confuses the shit out of mzanetti) [09:00] :D [09:01] mhr3, and that needs to go away when the preview gets replaced [09:03] mhr3, this was a way to avoid people tapping on the same action repeatedly [09:05] Saviq: here you go: https://bugs.launchpad.net/ubuntu/+source/unity-scope-click/+bug/1282460 [09:05] Launchpad bug 1282460 in unity-scope-click (Ubuntu) "Wrong header in app previews" [Undecided,New] [09:08] mzanetti, thanks, tweaked the title and description slightly [09:08] Saviq: doesn't this stuff use the Preview code I wrote any more? [09:08] Saviq: there should be an AppHeader.qml ready to be used [09:09] mzanetti, nope, it's completely redone [09:09] ok [09:10] Saviq, ciao [09:10] Cimi, o/ [09:12] Saviq, ah, i see [09:12] Saviq, remember the paste of yesterday? [09:12] Cimi, my log does [09:12] Saviq, what if i gave you a new prop which would be "actionBeginProcessed" [09:12] Being* [09:13] mhr3, sure, that would work [09:15] Saviq, when you have time to give me feedback [09:15] :) [09:16] Cimi, btw, when you start with "something like that is ugly, right?" don't expect me to contradict you much ;) [09:17] Saviq, I'm not expecting :) [09:17] Saviq, if I were sure I wouldn't ask [09:17] I don't know how to do it smartly, that's all [09:17] Cimi, just combine CardTool and CarouselTool, no need for two separate components [09:18] Cimi, otherwise looks fine === shiznix_ is now known as shiznix [09:18] Saviq, you told me you didn't want to have cardtool to switch category [09:19] Cimi, better than having CardTool and CarouselTool separate [09:19] Cimi, and I slept on it ;) [09:20] hah [09:23] mzanetti, did you get anywhere with the crash investigation? [09:28] Saviq: not really far... stack traces indicate its something in mir [09:28] mzanetti, don't hesitate to include them guys [09:29] Saviq: what I noticed too: when unlocking the screen, it hangs for about 2 secs [09:29] which makes swiping away the greeter quite nasty [09:29] anyone already on that? [09:30] mzanetti, no, didn't see that before [09:31] Saviq: I only see that with the complete mwc image (including sample data etc). didn't notice that before when I manually installed the ppas and branches [09:31] mzanetti, might be that something gets refreshed [09:31] Saviq: I have a question. How do you work on stuff that you want to test with new-scopes, but are intended to be merged to trunk? do you force new scopes and just work on the trunk branch? or do some rebasing-shenanigans? [09:36] Saviq: actually its worse: drag the greeter half away and release it so that it snaps back. on the last pixel it hangs already, printing this to the logs: http://paste.ubuntu.com/6964506 [09:37] so yeah. it seems to recreate all the scopes stuff when the greeter is locked. and because we only lock it after waking up (instead of before going to sleep) we notice the hanging when turning it on [09:44] mzanetti, that would mean that if you wait for a few seconds after powering up, it'd be fine again, which it isn't [09:44] mzanetti, I think the scopes refresh their data when unlocking [09:44] mhr3, comment ↑? [09:45] mzanetti, or actually yeah, both happens, really [09:45] Saviq: same if you have one app open and drag it a bit from left to right and back [09:45] it stutters, printing the same scope messages [09:46] Saviq, no, they don't refresh [09:49] Saviq, i see that i'm setting loaded to false as soon as the preview realizes that it will be replaced [09:49] should i be doing that? [09:49] mhr3, think it's fine [09:50] Saviq, there will be a delay between that and it removing and adding new widgets though [09:50] mhr3, that's fine, we can put the activity indicator on then [09:50] comes back to the activity indicator [09:51] Saviq, your tiny patch was broken btw [09:51] the one that added it back [09:51] mhr3, it was incomplete, yeah [09:51] it was never displayed when the preview was loading [09:51] because of the fadeIn thing [09:52] mhr3, does click scope report loaded=true yet? === dandrader is now known as dandrader|afk [09:53] Saviq, no.. c++ lambda madness in there [09:53] tvoss will check :) [09:53] mhr3, so no activity indicator yet anyway [09:53] right [09:54] Saviq, columns? :) [09:54] mhr3, yeah, tweaking one other thing first [09:54] Saviq, maybe we can rename cardtool into something else [09:54] Cimi, no, leave it [09:54] Cimi, it's all about cards [09:57] karni, I usually merge new-scopes, then shelve the new changes, uncommit, unshelve, commit [09:58] * karni parses [09:58] gotcha [10:00] Saviq, https://code.launchpad.net/~mhr3/unity-scopes-shell/processing-actions/+merge/207384 [10:01] mhr3, see, newlines in init list good! [10:01] Saviq, btw did rosie ping you yesterday? she said that pretty much everything card-related is using too large font [10:02] mhr3, yeah, but implementing the whole ruleset is not trivial, still hope to get it in, though [10:02] ok [10:04] mhr3, top-ack, or do we want to check that it works first? ;) [10:05] Saviq, tests say it does :P [10:09] mhr3, I like it how we got away from libunity(-core) and things can actually happen now ;D [10:10] all hail my awesome unit tests! :) [10:10] not even scopes-api has so good end-to-end testing [10:10] which is sad really :P [10:12] ;) [10:16] Saviq: the clipping seems to work [10:16] i only need to make sure i don't break everything [10:16] tsdgeos, yup, awesome [10:17] Saviq: i'm taking we still want disabled (and a bit opaque) when the sidesstage is shown? [10:17] s/opaque/transparent [10:17] tsdgeos, yeah, darkened and disabled, tapping on it should bring the dash in I think [10:18] tsdgeos, as should dragging the sidestage away [10:18] so tapping on dash "minimizes" the sidestage? [10:18] ok, can try [10:18] tsdgeos, FYI small addition bazaar.launchpad.net/~unity-team/unity8/new-scopes/revision/660 [10:19] ah [10:19] there's one thing i wanted to do yesterday [10:19] mhr3, so where do I set the number of columns on? [10:19] and i forgot [10:19] the pageheader mousearea should be an abstractbutton [10:19] so we get the haptic ZZZZZZ when clicking on it [10:19] Saviq, the preview stack, has widgetColumnCount prop [10:20] tsdgeos, ah so we only get it on AbstractButton do we?! [10:20] or the preview model itself, since you don't care about the stack these days yet [10:20] Saviq: think so yes [10:20] mzanetti, ^^ there, reason for using AB instead of MouseArea ;P [10:20] mhr3, I wonder if it should be higher, though... [10:20] or we need to add the HapticsEffect [10:20] ourselves === dandrader|afk is now known as dandrader [10:21] mhr3, like it's fairly static, and it'd be better if the previewModel knew it already before it's created? [10:21] will try to get that done in a sec (AB thing) [10:21] Saviq, the model is created empty, so doesn't matter much [10:21] Saviq: heh [10:22] speaking of the haptics effect. Am I the only one that thinks it is waaaay to strong? [10:22] it is ;) [10:22] Saviq, also, shouldn't orientation affect it? doesn't seem that static to me [10:22] mhr3, I mean it spans all preview stacks/models [10:23] right [10:23] mhr3, so either I store it somewhere and bind to all the preview stacks/models, or you do [10:23] then again previewstack is owned by qml and is therefore detached from everything else [10:23] mhr3, mhm [10:23] mhr3, ok [10:24] do you know a search string that returns me 4-5 elements? [10:24] always getting 8 :) [10:27] Saviq: https://code.launchpad.net/~aacid/unity8/new-scopes-haptic-back/+merge/207390 [10:27] tsdgeos, just push to new-scopes [10:27] (next time) [10:27] ok [10:28] Saviq: how can I force new-scopes when I run from unity8 trunk? UNITY_FORCE_NEW_SCOPES=1 ./builddir/tools/unity-scope-tool doesn't work. [10:28] Saviq, http://paste.ubuntu.com/6964688/ [10:29] karni, you can't "really" force new scopes (i.e. you can't have Cards) [10:29] karni, so only try* in trunk makes sense [10:29] test*/try* [10:29] oh ok.. [10:29] thank you [10:29] Cimi, MP please [10:30] Cimi: in CardCarousel.qml, you remove CardTool { id: cardTool ..}, and in line 25,26 reference cardTool.* [10:31] hmm, should remove the FORCE flag now [10:31] karni, the item cardTool is passed to DashRenderers [10:32] karni, line 208 of that diff [10:32] Cimi: I don't know the detail, but I guess what I said is not a problem then. [10:32] karni, it is instantiated in genericscopeview [10:32] karni, and sent in a different way [10:32] works [10:32] thanks though [10:32] I see [10:40] Saviq, quickie https://code.launchpad.net/~mhr3/unity-scopes-shell/no-force-new-scopes/+merge/207398 [10:41] karni, fyi ^ [10:41] * karni looks [10:41] mhr3: ack [10:45] mhr3, wanna check it out http://paste.ubuntu.com/6964770/ ? [10:47] Saviq, should work on desktop, right? [10:47] mhr3, yup [10:47] if i stretch the window [10:47] mhr3, works with scopes here [10:47] Saviq, in that case +1 :) [10:50] Saviq, i'm seeing odd crashes with scope-tool, not sure if it's just me [10:50] Saviq, maybe cause i don't have demo ppa's uitk? [10:50] Saviq, http://paste.ubuntu.com/6964780/ shortly after startup and things showing up [10:51] mhr3, or thumbnailer [10:51] mhr3, actually album art fetcher [10:51] mhr3, the threaded thing didn't get merged yet [10:51] Saviq, no, shouldn't be, i didn't enable those scopes [10:51] got just scopes in my envvar [10:52] mhr3, I only saw that trace with the music scope before [10:52] but it doesn't have the actual albumart code [10:52] those traces did [10:53] Saviq, i've actually seen it yesterday when playing with the activity indicator in previews [10:53] and your code is doing something similar [10:53] cause the previewModel can be null [10:54] Saviq: I have to admit it's a bit of pain to work on trunk for Card background features. there's only tryCardTool in trunk, and all tryCard*stuff is in new-scopes. [10:55] wait.. I did merge, I should rebuild [10:55] karni, huh? [10:55] karni, tryCard is there in trunk [10:55] karni, I tested your changes on trunk yesterday [10:55] oh, it is [10:55] Saviq: that's all I needed ;) thanks.. [10:56] Saviq, eeeh, nevermind, it was the albumart thing, my env wasn't what i expected :/ [10:56] mhr3, oof [10:58] mhr3, columns pushed [10:58] yey! [10:58] Saviq, so, can we merge the albumart fix? [10:59] tsdgeos, ↑ [10:59] weren't you reviewing that? [10:59] don't remember :D [10:59] url? [10:59] tsdgeos, https://code.launchpad.net/~jpakkane/unity8/threadfix/+merge/206948 [11:00] ah [11:00] i only complained about it [11:00] that probably counts as reviweing [11:00] heh [11:00] it does :) [11:00] tsdgeos, indeed! ;D [11:00] mhr3, so, activity.visible: !loaded && !processingAction ? [11:00] or no [11:01] well, yes [11:01] !(loaded || processingAction) [11:01] no! [11:01] loaded && !processingAction [11:01] Saviq: ok, i'll review [11:02] Saviq, well... you can't interact with the preview when processingAction == true, and you want to know that it's being reloaded, so i'd just base it off loaded [11:02] mhr3, "can't interact" - that's the thing [11:02] mhr3, we need to prevent the interaction [11:02] mhr3, and ideally show the user that it's busy [11:03] performing the action should be fast enough, getting the new preview is what will take longer [11:03] Saviq, so i'd say don't overcomplicate it [11:04] there's something that's breaking the logging [11:04] mhr3, "should be" [11:04] and i get all the stuff in a single line :-S [11:04] tsdgeos, scopes ↑↑ [11:04] mhr3, scopes log errors with no \n [11:04] and/or warnings [11:04] Saviq: but why would that mess with console.log ? [11:04] i.e. i'm getting [11:04] tsdgeos, because console.log prints at the end of a line with no \n [11:04] stagesOuterContainer.x 2129.573974609375stages.x 915.4135131835938stagesOuterContainer.x 2129.141357421875 [11:05] tsdgeos, ah [11:05] tsdgeos, not sure then [11:05] qWarning needs \n? [11:05] tsdgeos, maybe your terminal got confused [11:05] mhr3, no [11:05] if so then whoops :) [11:05] might be [11:05] when no whoops [11:05] s/when/then/ [11:05] mhr3, so, action means a road trip to the scope [11:06] mhr3, don't you think that might take long enough that the activity indicator should kick in? [11:06] mhr3, maybe after 50ms or so? [11:06] Saviq, i'd maybe start desaturing the buttons, wouldn't show the indicator [11:06] Saviq: that change we're doing in the art thing [11:07] mhr3, and something we can do *now*? ;P [11:07] we're blocking the thread [11:07] Saviq, heh :) [11:07] Saviq, wasn't qml making that thing easy? :P [11:07] but that's ok [11:07] mhr3, weren't we supposed to work off of designs? :P [11:07] since we're doing QQmlImageProviderBase::ForceAsynchronousImageLoading so it should be a different thread than the main one [11:08] Saviq, no dash designers, we do the design now ;) [11:08] tsdgeos, yeah [11:08] i honestly don't see why this would change anything [11:08] since we change from blocking in non main thread [11:08] to blocking in a thread that is not the not the main thread D: [11:08] mhr3, ↑ explain [11:08] but if you guys says it fixes it [11:08] tsdgeos, it's not about blocking [11:08] tsdgeos, it's about crashing [11:08] tsdgeos, it was spinning the loop of that thread [11:09] Saviq: sure i know [11:09] * Saviq shuts up [11:09] tsdgeos, which meant is started processing other events in that thread etc, and that could obviously horribly break something [11:09] mhr3: i don't see why, since that thread is ad-hoc for image loading, i don't see why it would go wrong [11:10] but if it fixes it, it fixxes it [11:10] Saviq: are we still eventually going to rip that up of unity8 or? [11:10] tsdgeos, yes, definitely [11:10] tsdgeos, well, you know it's just a bad idea to keep respinning the mainloop [11:10] greyback, hey, alive yet? [11:10] mhr3: but that is *not* the mainloop in there [11:10] it's a separate thread for image loading [11:10] so you're not respinning the mainloop [11:11] tsdgeos, but there is *a* mainloop processing events for that thread [11:11] which might be requests for another image [11:11] maybe it could break some implicit ordering of things, and qt didn't handle that [11:11] Saviq: yes yes [11:12] mhr3: ok [11:12] greyback, could you mark the bugs as Fix Committed for those that we have fixed for mwc? like https://bugs.launchpad.net/unity8/+bug/1281199 I believe? [11:12] Launchpad bug 1281199 in unity8 (Ubuntu) "Can only pull launcher out in landscape from top left on flo mwc image" [Critical,In progress] [11:13] Saviq: yep, just am pulling updates to confirm [11:13] greyback, ok great [11:13] thanks [11:13] Saviq: I suppose that's ready for new-scopes, still needs tests for trunk https://code.launchpad.net/~unity-team/unity8/themeing-font-and-mascot/+merge/207282 [11:14] karni, awesome, thaks [11:14] Addressed your review comments [11:14] :) [11:14] karni, don't put internal properties on root objects in a component [11:15] hrm [11:15] karni, remember that root component's props are basically its API [11:15] right.. [11:16] karni, we usually put those somewhere lower down the hierarchy [11:16] karni, or sometimes in a separate QtObject { id: priv } [11:16] Saviq: you mean these should have been aliases? not sure how can I bind from Card without putting properties at CardHeader root [11:16] karni, CardHeader they need to be there [11:16] karni, but not in Card.qml [11:16] oh [11:17] now I understand. [11:18] karni, one more thing: no mascot shape when summary and art is mapped, as we're putting those on a default, white background [11:18] Saviq: for instance, inOverlay - making readonly is good, or should I hide it anyway? [11:18] karni, if no user of Card needs access to it, hide it [11:18] ack [11:20] mzanetti, awesome, can you include Mir folks in the crash investigation? [11:20] karni, so basically property bool useMascotShape: background.visible [11:20] Saviq: I'm just merging the right-edge unity-mir branch with trunk, seeing that ricmm fixed something in there related to crashing apps. will make sure it still happens with that, and then forward things to the Mir people [11:21] Saviq: ack [11:21] mzanetti, ok awesome, [11:21] karni, so you probably don't even need the useMascotShape prop anywhere in card.qml, just bind CardHeader.useMascotShape directly to background.visible [11:22] Saviq: just did that :) [11:22] karni, and maybe s/useMascotShape/shapeMascot/ [11:22] Saviq: fontcolor, I think I need. for the summary. make it readonly would suffice? [11:22] or hide in QtObject ? [11:23] karni, just put it on the Label [11:23] karni, same with the functions [11:23] ok :D [11:23] karni, you put them on the root object, which means they're accessible to Card users, which they shouldn't be [11:23] karni, just hide them in Label { id: summary } [11:23] yep. thanks Saviq [11:23] karni, and then bind for CardHeader [11:24] karni, we're not breaking overlay are we:? [11:24] yes, gotcha [11:24] karni, we don't want summary to be white if overlay is enabled [11:24] good catch :| [11:25] I'll make sure that works. [11:31] Saviq: seems its fixed by now [11:31] it behaves somewhat odd still, but doesn't crash any more [11:32] * mzanetti pushes the merged branch to the mwc image [11:38] Saviq: I'm confident I'm not breaking overlay (summary is now grey even when overlay enabled), but I'm having trouble testing card-background: "color:///#112233" -- has JSON spec changed? [11:39] do I need to make it card-background: { "type": "color", sth: "#aarrggbb" }? [11:40] mhr3: ↑ [11:40] karni, do you have unity-plugin-scopes from yesterdya? [11:41] I am up to date. I'm working on trunk, using tryCard [11:41] oh.. trycard.. that isn't going through the plugin [11:41]