/srv/irclogs.ubuntu.com/2014/06/05/#ubuntu-unity.txt

tsdgeoswhy am i still marked as away...07:29
tsdgeosoh, not aynmore07:29
tsdgeosgood07:29
tsdgeosSaviq: ping07:29
tsdgeospstolowski: ping07:34
pstolowskitsdgeos, pong in ~5 minutes (otp)07:36
tsdgeospstolowski: do you know how the spec specifies headerless categories?07:43
pstolowskitsdgeos, nope, I don't07:52
tsdgeospstolowski: because in that two_scopes you "cheat" by giving the second category the same name as the previous07:52
tsdgeosand that makes lvwph give you a headerless category07:52
tsdgeosbut not sure that's the way07:53
pstolowskitsdgeos, uh, whaat? yeah, name is the same, but id should be used to distinguish them... but I realize this is a limitation of lvwph07:53
Saviqtsdgeos, pong07:54
pstolowskitsdgeos, and the ids don't really matter at the lvwph level..07:54
tsdgeospstolowski: it's not a limitation, it's a design decision, two categories the second doesn't have a header if they share title07:54
tsdgeospstolowski: because it's basically what QML ListView does07:54
tsdgeosSaviq: headerless categories07:54
tsdgeoshow does one specify one?07:54
tsdgeosSaviq: empty category name?07:56
Saviqtsdgeos, yeah07:58
tsdgeosSaviq: ok, that's new to lvwph, i'll add that, then there's the question, in lvwph we decided that two consecutive "same category name" yields the second category with no header, do we care about that at the GenericScopeView.qml (regarding adding the top margin to the category contents) or asume it won't happen?07:59
Saviqtsdgeos, if that's a problem, we might make it explicit07:59
Saviqtsdgeos, I forgot that the sections work like this07:59
tsdgeosit's not a problem, it's just that i will need to add more code to support "empty because is the same name" headers since at the moment there's no way for a category to know if it has section or not08:00
tsdgeosso i can either have08:01
tsdgeos     topMargin: name != "" ? 0 : units.gu(2)08:01
tsdgeosor08:01
tsdgeos    topMargin: hasSectionHader ? 0 : units.gu(2)08:01
tsdgeosbut then i have to add hasSectionHader code to LVWPH08:01
tsdgeosshouldn't be hard, but it's more code :D08:02
Saviqtsdgeos, whichever way you prefer08:02
tsdgeosok, let me check how hard hasSectionHader is08:02
tsdgeosalways better to do it "properly"08:02
Saviqmzanetti, seen cwayne's email? able to reproduce?08:06
mzanettiSaviq: yep, seen it. still replying to mails. That one will be next08:06
Saviqmzanetti, kk thanks08:06
pstolowskiSaviq, any thoughts about http://pastebin.ubuntu.com/7588500/ ? ok to make it on per-category level?08:12
Saviqpstolowski, yeah missed you yesterday08:13
Saviqpstolowski, I think that's good08:13
Saviqpstolowski, you should stick to either "tap" or "click" though08:13
Saviqpstolowski, 'cause you're setting tap behaviours using ClickBehaviour ;)08:13
pstolowskiSaviq, yeah, i had a problem with naming ;)08:19
pstolowskiSaviq, in the absence of mhr3, I think I'll also take care of the plugin branch for that08:20
Saviqpstolowski, kk08:25
Saviqpstolowski, you'll need to adapt http://bazaar.launchpad.net/~unity-team/unity-api/trunk/view/head:/include/unity/shell/scopes/CategoriesInterface.h too08:26
pstolowskiSaviq, ah, ok, didn't know that, thanks08:26
Saviqpstolowski, tsdgeos, do you think we should make headerless explicit? as in explicitly set in the scope (as opposed to empty name, which is a bit of a hack)?08:27
tsdgeosSaviq: i think empty name is fine08:27
Saviqkk08:27
tsdgeosi mean if you have an empty name it will look ugly as hell08:27
tsdgeoswith the expansion pattern there08:27
tsdgeosbut nothing else08:27
pstolowski+1 to what tsdgeos says08:27
Saviq+108:28
=== pstolowski is now known as pstolowski|bbl
mzanettiSaviq: Re: Applications are orphaned when unity8 crashes08:38
mzanettiiirc in Oakland Mir people said it wouldn't be a problem for an app to reconnect to a new Mir server instance08:38
Saviqmzanetti, yeah, I remember that too, except that never happened ;)08:40
mzanettiSaviq: heh, true. But in any case, sounds doable and probably what we should go for.08:40
Saviqmzanetti, and then unity-mir would have to pick the running apps back up (not saying impossible)08:40
mzanettiyep. unity-mir needs fixing too for that08:41
Saviqmzanetti, for touch it's simple, later when we introduce windowing, that will get interesting when we try to restore window locations :)08:41
mzanettiSaviq: without having spend much time in thinking about it, it would seem they just need to be placed at the same place they were left08:42
Saviqmzanetti, *just* is the key word here08:42
mzanettihehe08:42
Saviqmzanetti, when the windows don't know where they were08:42
Saviqmzanetti, and the compositor just crashed ;)08:42
mzanettiright... we'd need to persist every x/y/w/h change immediately08:43
tsdgeosSaviq: pstolowski|bbl: https://code.launchpad.net/~aacid/unity8/headerless_category_margin/+merge/22213208:46
Saviqtsdgeos, looking good08:50
tsdgeoshmmmm08:50
tsdgeosSaviq: just realized we can't expand headerless categoriies anymore08:50
tsdgeosSaviq: that's fine i guess D:08:50
Saviqtsdgeos, indeed ;)08:50
Saviqtsdgeos, well TBH we'll see where this goes, the header probably won't be the expander for long08:51
tsdgeoscorrect08:51
tsdgeosSaviq: so what's happening in https://launchpadlibrarian.net/176949684/aligment_issue.png is that the cardtool is "mandating" 5 rows of summary, but since there's only 3 in that particular card, spacing looks wrong08:55
Saviqtsdgeos, right, of course08:56
Saviqtsdgeos, they should use horizontal journal...08:56
Saviqif we had integrated it...08:56
tsdgeosSaviq: i guess we should special case for when it's 1 column and then "ignore" the summary max height?08:56
tsdgeosi.e. we obey the max height of summary so that in a grid it doesn't look bad, but here it's essentially a column, not a grid08:56
tsdgeosSaviq: it'd be the same problem with the cardtool giving us the wrong height, no?08:57
Saviqtsdgeos, *or* we'd need an adaptive grid that changes row height based on each row's delegates08:57
Saviqtsdgeos, yeah, but it's not (meant to be) enforced in vjournal08:58
Saviqwhat you wrote may be the best for now (ignoring height if rows == 1)08:58
Saviqit won't break tablet and will help this use case08:59
Saviqwith a FIXME08:59
Saviqtsdgeos, basically, vJournal was exactly meant for different-height delegates08:59
Saviqtsdgeos, or well, integrating vJournal would get us closer to where we want to be, so...09:00
tsdgeosright09:00
Saviqthere's a branch from karni somewhere09:00
tsdgeosyes09:00
tsdgeoshttps://code.launchpad.net/~unity-team/unity8/new-scopes-vj-integration/+merge/21478609:00
Saviqwe could try and revive this (I know it never worked for him, though...)09:00
tsdgeosyeah should not be "that hard"09:01
tsdgeosok, let me have a look at it09:01
Saviqlet's try that09:01
tsdgeosi guess it was just hard for him because he was just dropped in the trenches of VJ, QML, Dash, GSV, etc09:01
Saviqyeah09:02
Saviqif he got RVJ working09:02
Saviqthen putting it in GSV should be mostly trivial09:02
Saviqtsdgeos, and we really need to get rid of FilterGrid, it just makes everything more complicated09:04
Saviqlike the "let's sync height expansion between lvwph and lv"09:05
Saviqwhich could just be driven internally by lvwph instead09:05
tsdgeosyeah09:05
Saviqbut yeah, not right now09:06
Saviqtsdgeos, btw, ignoring height wouldn't work anyway09:07
Saviqtsdgeos, GV has static row height09:07
Saviqwe'd need a custom GV that didn't09:07
tsdgeosyeah i realized that09:07
tsdgeoswe're using a GridView09:07
tsdgeosso it'll be Grid anyway09:07
Saviqyup09:07
tsdgeosSaviq: i'll kill https://code.launchpad.net/~unity-team/unity8/new-scopes-vj-integration/+merge/214786 anyway, since it's for merging into new-scopes09:14
tsdgeosand create something based on it for trunk09:14
Saviqtsdgeos, of course, yeah09:14
Saviqtsdgeos, I know why your result on https://bugs.launchpad.net/unity8/+bug/1326397 was different09:15
ubot5Ubuntu bug 1326397 in Unity 8 "In horizontal card layout, margin between art and text is too small" [High,Triaged]09:15
Saviqtsdgeos, it looks like they didn't commit something09:15
tsdgeosmay be09:15
tsdgeosi mean the code wasn't even original compiling :D09:15
Saviqyeah I saw09:16
tsdgeoshave they updated it?09:16
Saviqtsdgeos, people seem to be abusing fill-mode09:16
Saviqtsdgeos, no, I just did what you did09:16
Saviqand checked it out in the tool09:16
Saviqand got it into a ~working state09:16
Saviqbut it's not great, but per-spec09:16
Saviqtsdgeos, they were (as people seem to do) abusing fill-mode: fitr09:17
Saviq-r09:17
Saviqor well, we're not dealing well with that in horizontal mode09:17
tsdgeosabusing API is the programmers life :D09:18
tsdgeos"you don't let me do that, eh? let's see if i can force you to"09:18
Saviq;)09:25
Saviqtsdgeos, https://launchpadlibrarian.net/176990434/Zrzut%20ekranu%20z%202014-06-05%2011%3A23%3A53.png is what I got09:25
Saviqtsdgeos, and I *think* it's our bug09:25
Saviqwell, I'm quite sure it is09:25
Saviqnot sure what it results from, though09:25
tsdgeoshmmm09:26
tsdgeossecond one looks good enough09:26
Saviqtsdgeos, yeah, sure09:26
tsdgeosi'm confused as why my result looks so wildly different from yours :S09:26
Saviqtsdgeos, because https://launchpadlibrarian.net/176990395/rsslocal.patch09:26
tsdgeosah09:27
tsdgeos:D09:27
Saviq;)09:27
tsdgeosi guess it shouldn't be too hard fixing it09:27
Saviqtsdgeos, diff between the two is 2 vs. 1 line title, and we seem to be forcing the height "the wrong way"09:27
tsdgeosbut you still have the vertical spacing problem09:27
tsdgeoswell09:27
Saviqoh yeah, that's only gonna be fixed with vj09:27
tsdgeosthe art/mascot image09:28
Saviq[...] header → art instead of art → header09:28
tsdgeosis forced to the title height09:28
Saviqart, not mascot09:28
tsdgeosno, that's correct for horizontal09:28
* Saviq not sure09:28
tsdgeosit's just that the first should remember it can have two lines09:28
tsdgeosand not one how it has now09:28
Saviqtsdgeos, ah09:28
tsdgeoswell it's correct as per what we had :D09:28
Saviqtsdgeos, sure, that might be why indeed09:28
tsdgeosso there's some playing with it09:29
tsdgeosbut should be fixable09:29
Saviqyeah, it's better already09:29
Saviqtsdgeos, in any case we'll need to look into abuses, too, at some point ;)09:29
tsdgeossure09:29
* Saviq can't see that happening in the near future...09:29
tsdgeosis it me or the canonical irc server is experiencing issues?09:29
tsdgeosi get kicked out of it a lot today09:29
Saviqtsdgeos, hmm nope09:30
Saviqseems legit for m09:30
Saviqe09:30
tsdgeos:/09:31
SaviqI'm behind a bouncer, but should be notified about disconnections in any case09:31
Saviqohkay, that's bad09:37
Saviqwth09:37
tsdgeoswe broke tryOrganicGrid, tryVerticalJournal and tryHorizontalJournal when going to displayMargin :/09:59
* tsdgeos fixes09:59
=== pstolowski|bbl is now known as pstolowski
Saviqtsdgeos, oups10:08
Saviqtsdgeos, TBH, when you're at it... maybe just integrate all of them at once? it should be copy/paste for the most part?10:09
tsdgeosyeah10:10
Saviqgreyback, there a reason why ubuntu-touch-meta is still older in the PPA than in distro?11:09
Saviqand if not, should we just drop it from there?11:09
greybackSaviq: it's a custom one I pushed. Then distro bumped theirs. We're still on image 58, so it's not impacting us11:09
Saviqgreyback, oh ok11:10
Saviqgreyback, it *might* impact us still if cwayne wants to build an image with the PPA... not sure what his plan is there11:10
greybackSaviq: will check with him11:11
Saviqgreyback, kthxbai11:11
=== MacSlow is now known as MacSlow|lunch
Saviq/fodo11:33
Saviqhm?11:33
tsdgeosSaviq: :/11:41
tsdgeosSaviq: there's a not so good think about VJ11:41
tsdgeosheight is randomly limited at 35gu of height per design11:41
tsdgeosresulting in cut text11:41
tsdgeoswhich is awful11:41
tsdgeosi.e. the end of http://imgur.com/Ae68nDR11:42
tsdgeosnot sure how we can fix that tbh11:43
=== jhodapp|afk is now known as jhodapp
tsdgeosor even worse11:51
tsdgeosthis one11:51
tsdgeoshttp://i.imgur.com/crhA6Hs.png11:51
tsdgeosfood11:52
Saviqtsdgeos, that's fine, we're talking with Mike today about this, we'll fix it11:58
Saviqtsdgeos, plan was to add a shadow or something11:58
=== alan_g is now known as alan_g|lunch
=== _salem is now known as salem_
=== MacSlow|lunch is now known as MacSlow
=== pete-woods is now known as pete-woods-lunch
tsdgeosSaviq: ok12:54
tsdgeosMirv: have you seen the patch that is supposed to fix the crash in javascript 5.3 arm crash?12:57
=== salem_ is now known as _salem
=== _salem is now known as salem_
=== alan_g|lunch is now known as alan_g
Mirvtsdgeos: seen yes, got e-mail 45min ago. tested/done not yet.13:06
tsdgeosMirv: cool :)13:06
Mirv(plus I need to leave now for today)13:06
Mirvbut hopefully it'll solve it, there's no end of possible places it could be in use even if webbrowser and UITK are now fixed13:06
tsdgeosSaviq: i can't find in vertical-journal where the number of columns comes in13:43
tsdgeosdon't we have a variable for that?13:43
Saviqtsdgeos, you mean *into* vj?13:49
tsdgeosSaviq: yeah13:49
tsdgeoshow many columns vj has to have13:49
Saviqtsdgeos, that's dynamic based on card size13:50
Saviqtsdgeos, that's what ResponsiveFoo does13:50
tsdgeosSaviq: hmmmm13:51
tsdgeosbut the card size is based on the column width :D13:51
Saviqtsdgeos, no, the other way around13:51
tsdgeosit's a bit recursive, no?13:51
Saviqtsdgeos, card size is small, medium, large13:51
tsdgeosah right13:51
Saviqthat directly corresponds to GU13:51
tsdgeosgot confused13:52
tsdgeosso if you get https://code.launchpad.net/~aacid/unity8/dashviews_integration/+merge/222192 and change victor's code to be a vertical-journal13:52
tsdgeosshould work13:52
tsdgeosit's working for me13:52
tsdgeoswith the said cutting problems13:52
tsdgeosthere's an interesting bootstrapping problem in the height of the thing13:53
tsdgeoswonder if that's were karni got stuck13:53
tsdgeosthe rest of the code was mostly fine13:53
tedgci-eng is busy13:53
karnitsdgeos: ^ ^13:53
tedgSaviq, So do you think that the autopilot-legacy stuff is the issue? The whole thing seems weird.13:53
karniglad to hear it. tsdgeos, I don't recall what stopped me there. but I recall the codebase was new to me at the time as well.13:54
=== pete-woods-lunch is now known as pete-woods
Saviqtedg, it definitely is weird, and I'm not sure what's happening, I only really know it's ual + url-dispatcher that, upgraded, 'cause this to happen13:55
SaviqI didn't have the time to dig more13:55
tedgSaviq, What's a reasonable way to test what autopilot is doing to setup unity8? Can I just start it from the command line?13:57
Saviqtedg, that's even more interesting... it runs it via upstart13:57
Saviqtedg, tweaking a bit of env13:57
tedgSo I don't see an XDG_DATA_DIRS in my global environment just normally.13:59
tedgSo it seems Unity8 can handle that case.13:59
Saviqlet me try and get a diff between a working and a broken env14:01
tsdgeosSaviq: the current ResponsiveVerticalJournal has a maximumNumberOfColumns property14:01
tsdgeosbut we don't really need it, do we?14:02
tsdgeosmaximumNumberOfColumns should just be infinite14:02
Saviqtsdgeos, not *really*, we have it in RGV14:02
Saviqtsdgeos, but that's more general purpose14:02
Saviqtsdgeos, so I'm fine without it14:02
tsdgeosi mean you can just set more spacing and stuff if you want fewer columns et al14:03
Saviqtsdgeos, yes, I agree, RGV just lets you set up more to your liking, but it was a more gen-pop type of thing14:04
Saviqtsdgeos, and karni (rightly so) just copied the API14:04
tsdgeossure, not "blaming" anyone14:04
tsdgeosjust thinking if we need it or not14:04
Saviqtedg, http://paste.ubuntu.com/7595031/14:05
Saviqtedg, might it be that us overriding XDG_DATA_DIRS causes havoc? (it worked fine a long time)14:08
tedgSaviq, I bet it's assuming /usr/share/ for XDG_DATA_DIRS unless set. When set, it needs to have the base.14:08
Saviqtedg, yeah, we had that for a long time now...14:08
=== dandrader is now known as dandrader|afk
tedgAlso a bit odd about the LD_LIBRARY_PATH, do you guys use any Android compat?14:08
Saviqtedg, nah, those are liblightdm mocks14:08
tedgOr is that all through Mir?14:08
Saviqtedg, or you mean the original ones?14:09
Saviqtedg, not sure where those come from14:09
tedgSaviq, Seems like that it should include the original ones, plus the others.14:09
tedgBut that's probably not our issue right now.14:09
tedgGrep'ing autopilot there's no XDG_DATA_DIRS14:09
tedgWhere should I be looking for the setter there?14:09
Saviqtedg, it's in unity8's suite14:10
tedgSaviq, I think the issue is probably with the split greeter fixes, I bet the Xsession stuff was setting the var.14:10
tedgSo by removing that we ended up in a different place.14:10
Saviqtedg, but was that in url-dispatcher / UAL?14:11
tedgSaviq, No, guessing overlapping landings.14:11
Saviqtedg, but hear me out again... I can upgrade everything, and everything's fine until I upgrade url-d / UAL14:11
Saviqand yes, I know it doesn't make sense14:11
Saviqtedg, but you are right14:12
Saviqtedg, if I add /usr/share at the end14:12
Saviqtedg, the problem is gone14:13
tedgWoot! :-)14:13
Saviq*HOW* url-d / UAL causes it, you tell me14:13
tedgNot sure why that'd be the case.14:13
tedgDo you want me to prepare a patch?14:13
Saviqtedg, no, I'm on it14:13
tedgSaviq, Cool, thanks!14:14
Saviqoh yay, and eclipse doesn't start :|14:14
tedgeclipse --force-start14:19
mterrySaviq, I was looking at memory usage of unity8 when apps are open, and it was consuming enormous amounts of memory per app (like 10-20MB).  That's surely not right?14:21
Saviqmterry, screenshots14:21
Saviqmterry, and sure, that's not right, but explainable atm14:22
mterrySaviq, because we take full screen images?  Hrm14:22
Saviqmterry, and we're probably even more wasteful (like keeping a copy or so)14:23
Saviqmterry, so, yes, we need to fix that14:23
mterrySaviq, do we need the full screen image for anything but the thumbnail?  Seems like we could optimize that image a bunch if not14:24
Saviqmterry, right edge14:25
Saviqmterry, thumbnail is going away real soon14:26
Saviqmterry, but then14:26
Saviqmterry, with QtComp we'll only load screenshots if there's no surface from the app14:26
Saviqmterry, so it should get better already14:26
=== dandrader|afk is now known as dandrader
tsdgeosSaviq: so the question now is, do i do HJournal and oGrid in the same branch or in a dependant branch so that in case we want to merge VJournal first it's easier to review?14:27
Saviqtsdgeos, make it sep MPs then14:33
=== dandrader is now known as dandrader|afk
Saviqhuh... 14" Chromebook... quite (oxy)moronic if you ask me...14:47
tsdgeos\o/ sane branch naming in Qt14:57
tedgSaviq, So is dash going to become its own process?15:03
Saviqtedg, yes15:03
Saviqtedg, soon15:03
tedgSaviq, Okay, so a trusted session for pay should be over the dash PID then?15:03
Saviqtsdgeos, where where?15:03
Saviqtedg, yup15:04
tsdgeos5.3.1 => refs/for/5.3 (for now, refs/for/5.3.1 when it's available)15:04
tsdgeos5.3.2 => refs/for/5.3 (after the 5.3.1 branch is created and 5.3 bumped)15:04
tsdgeos5.4.0 => refs/for/dev15:04
tsdgeosSaviq: ↑↑15:04
tedgSaviq, Will it be it's own upstart job? Looking for how to get the PID.15:04
=== dandrader|afk is now known as dandrader
Saviqtedg, yes15:04
tsdgeosSaviq: so now it's either dev for "whatever is next" or branch name with version number15:04
tsdgeosinstead of release, stable, dev15:04
Saviqtsdgeos, yay15:04
Saviqtedg, do you have another question to which I could reply with https://www.youtube.com/watch?v=8ZToP48Mrp8&feature=youtu.be&t=9s15:06
tedgSaviq, Is the weather in London cloudy all the time?15:10
tedg:-)15:10
Saviqtedg, https://www.youtube.com/watch?v=8ZToP48Mrp8&feature=youtu.be&t=13s15:11
tedgheh15:12
tsdgeosSaviq: Horizontal Journal: "Only over 60 GU device, on smaller devices this category layout becomes a grid."15:13
tsdgeosweird?¿15:13
Saviqtsdgeos, not really... HJ doesn't make sense on smaller screens I think15:15
Saviqtsdgeos, truth be told I have not seen *any* design using HJ15:15
tsdgeos:D15:17
Saviqtsdgeos, you can check with mikenagle, but I doubt he'd come up with something *right now*15:19
Saviqtsdgeos, maybe we should ignore HJ for now15:19
tsdgeosworks for me :D15:19
tsdgeosSaviq: so OrganicGrid? or just leave the VerticalJournal and try to fix the problem with have with art when the header line is only 1 line instead of two?15:20
Saviqtsdgeos, the latter15:20
tsdgeosmakes sense15:20
=== gatox is now known as gatox_lunch
tsdgeosSaviq: do you have a sec?15:51
Saviqtsdgeos, yes16:00
tsdgeosSaviq: i'm slightly confused by all the height/width sizing code of artShape and artShape.image16:01
tsdgeosi'd say that code is origianlly yours, do you remember it?16:01
Saviqtsdgeos, you and me both ;)16:01
Saviqtsdgeos, but yes16:01
* Saviq looks16:01
tsdgeosi mean why do we have different width/height sizes for artShape and its image?16:01
Saviqlol16:02
SaviqI tell you guys, Albert has a timer that plugs his computer out ;D16:02
Saviqplugs out...16:03
Saviqyes, that's an expression16:03
Saviqtsdgeos, was just LMAO16:03
Saviqtsdgeos, at you getting disconnected at 6pm sharp ;)16:03
Saviqtsdgeos, the thing with image and the art is... there's so many combinations...16:03
Saviqtsdgeos, but in general, art is foo-wide, and high based on aspect-ratio, and the image is cropped16:04
Saviqtsdgeos, but when it's fill-mode: fit16:04
tsdgeosSaviq: router being funny16:04
Saviqtsdgeos, funny, sure, you just have a "disconnect at 6pm" timer don't you ;)16:05
Saviqtsdgeos, in fit mode it is fitted in those same bounds (foo-wide, aspect-ratio-high)16:05
Saviqtsdgeos, and they should always have the same dimensions, why do you say they don't?16:06
tsdgeoswell16:06
tsdgeoswe have16:06
tsdgeosheight = Qt.binding(function() { return !visible ? 0 : image.fillMode === Image.PreserveAspectCrop ? image.height : width / image.aspect }); \n\16:06
tsdgeosfor artShape height16:07
tsdgeosso sometimes is the image height16:07
tsdgeossometimes not16:07
tsdgeosbut i guess it has to do with the cropping or not you mention16:07
tsdgeosthing is16:08
tsdgeoswe're calculating that fixedArtShapeSize16:08
tsdgeosbut not applying it correctly i'd say16:08
tsdgeosmy current fix is16:08
tsdgeoshttp://paste.ubuntu.com/7595728/16:08
tsdgeosbut i'm thinking that is going to break stuff16:09
tsdgeosand maybe i need to apply it to the shape and not the image16:09
tsdgeosanyway, i really need to disconnect or i'll be late to tapas with friends16:09
tsdgeoswill try to think/find all combinations tomorrow16:09
* tsdgeos waves16:09
Saviqtsdgeos, p/16:09
Saviqfook, missed him16:10
anpok_Saviq: we heard that trusted-sessions are already in use in some branches... could you point us to some?16:25
Saviqanpok_, _in use_?16:25
Saviqanpok_, let me fwd you the mail from Nick / Kevni16:26
SaviqKevin16:26
anpok_in use - as in - somebody tries to implement something with it :)16:26
Saviqanpok_, ah yes, mardy is16:26
Saviqanpok_, with the branches mentioned in the email, though, u-s-c crashes for me on startup16:27
Saviqanpok_, here are the packages I built on Tue16:28
Saviqhttp://people.canonical.com/~msawicz/unity8/trusted-sessions.tar.xz16:28
Saviqanpok_, but I didn't have time to investigate16:28
anpok_ok thank you16:29
=== gatox_lunch is now known as gatox
mhall119Trevinho: ping16:54
Trevinhomhall119: pong16:54
Trevinhomhall119: hi16:54
mhall119Trevinho: hey, could you run a session in UOS next week that goes over the features of Unity 7 and how to use them to maximize productivity? it's something that we've been requested to discuss16:55
Trevinhomhall119: yeah... But not sure how much there's to talk about them :) I mean, is there much to explain from an user prospective?16:57
mhall119Trevinho: I think there is, the features that you (and to a lesser extent I) take for granted may not be well known16:57
mhall119even things like explaining what the HUD is, where it's data comes from, and how the fuzzy matching works with it, would be something a lot of people probably don't fully grasp16:58
Trevinhomh, ok... I try to write down something16:58
mhall119the Launcher super+# shortcuts, window snapping, alt+~ instead of alt+tab, etc16:59
mhall119thanks16:59
Saviqmhall119, I won't be able to do UE Live next week, am in Berlin for QtCS, you'll have to make kgunn do it for you ;)17:02
mhall119Saviq: thanks, I know lots of good presenters will be at QtCS, bad timing for us but there weren't many other options17:03
mhall119kgunn: ping17:03
Saviqmhall119, oh yeah, and there's ODS, too...17:05
=== alan_g is now known as alan_g|EOD
=== dandrader is now known as dandrader|lunch
=== om26er_ is now known as om26er
=== dandrader|lunch is now known as dandrader
cwayneSaviq: seeing a lot of crashes with shell rotation on the n418:17
cwaynemay need to drop it and only apply those changes to n7 to lock into landscape18:17
dandradercwayne, tell me the steps to reproduce it18:17
dandradercwayne, btw, it's me, greyback and mzanetti working on it. so you can bug us regarding that ppa18:18
cwaynedandrader: just general use for me, opening an app or two and rotating the shell18:18
cwaynei get hangs and unity restarts, but no crash files18:18
kgunnmhall119: what's up ?18:25
Saviqcwayne, huh, no crash files? that's real weird18:28
Saviqcwayne, like nothing in /var/crash/ ?18:28
cwayneSaviq: nothing related to unity at least18:30
greybackcwayne: how about unity-system-compositor? or lightdm?18:31
Saviqcwayne, please clear it whole and next time you get it crash take a look what changed18:31
cwaynei took the update out of the mae-flash script though so I haven't used the shell-rotation bits in a few hours, let me add it back in and see if i can get some debug18:31
cwaynegreyback: all i had was a mediascanner crashfile18:31
greybackcwayne: that won't affect unity818:31
cwaynei know, that's why i said nothing related to unity :)18:32
greyback:)18:32
mhall119kgunn: we need a UOS session about Mir/Unity 8, either planning or presentation (with Q&A), can you run that next week?18:32
cwaynelet me reinstall the ppa and see if i can get some repro steps18:32
kgunnmhall119: yeah, i got side tracked on adding a session...let me do that18:33
kgunnmhall119: my only request is to have it on Thurs...i'm out of office tomorrow until Wed morning (having a day to get oriented would be nice :)18:33
mhall119kgunn: that's fine, if you set your availability when you registered for the summit it'll warn the scheduler if they try to put it any other day18:38
cwayneoh, there was a unity8 update 10 minutes ago from that ppa?18:40
cwaynegreyback: mzanetti^ what changes were in there?18:40
kgunnmhall119: sil2100 ...fyi, my session http://summit.ubuntu.com/uos-1406/meeting/22290/utopic-ue-unity8-mir-roadmap/18:41
* kgunn needs to go update availability18:41
greybackcwayne: some tablet fixes mainly - for spread. The too-tall window problem too (i.e. panel on top)18:41
cwaynehm, it does seem a bit more stable than this morning, even on the n4 though18:44
mhall119Trevinho: can you register for uos-1406 so I can get this session in for you?18:44
mhall119Trevinho: http://summit.ubuntu.com/uos-1406/registration/18:44
sil2100kgunn: thanks!18:50
cwaynegreyback: rotation does seem a bit too sensitive still though18:54
greybackcwayne: noted, will see what we can do18:55
kgunncwayne: greyback ...isn't rotation sensitivity what ricmm addressed? (malta demo)18:57
greybackkgunn: yes but he's not shared the code yet18:58
* cwayne wasn't in malta18:58
mhall119kgunn: approved it, thanks18:58
mhall119kgunn: anytime on thursday good for you?18:59
kgunnmhall119: sure19:00
mhall119kgunn: ok, got you scheduled for 1400 UTC19:00
kgunnta19:01
anpok_hm in the unity8 changes that landed overnight.. was there a change that might cause surfaces to be resizes temporarly when the osk is shown19:09
anpok_(or was it always like that)?19:09
Saviqanpok_, no, there's no surface resizing in current u819:13
Saviqanpok_, anything that could happen would happen in qtubuntu19:13
Trevinhomhall119: done19:27
cwaynegreyback: mzanetti dandrader: camera seems to be rotated oddly in n7 pinned to landscape20:18
dandradercwayne, does it rotate at all?20:19
dandradercwayne, when you rotate the device20:20
cwaynedandrader: no20:36
cwaynedandrader: also, dragging from the right on n7 with no apps open drags a black box from the right20:40
dandradercwayne, yeah, that's know20:40
cwayneok cool, sorry :)20:40
dandradercwayne, np. keep it coming :)20:40
cwaynehmm, shorts app seems broken20:48
=== salem_ is now known as _salem

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