[08:38] @unity: mterry wants opinions on whether this is the right approach or not on [08:38] https://code.launchpad.net/~mterry/unity8/turn-on-screen-when-notifying/+merge/267592 [08:38] mzanetti: ↑ i guess your opinion is particualarily interesting [10:18] cimi: did you see the other comment i made yesetdday in your ubuntushape branch? [10:20] tsdgeos, nope :/ [10:21] tsdgeos, oh no that one yes [10:21] yeah working on it [10:21] ok [10:26] tsdgeos: mzanetti off today, well he should be (i hope) [10:27] kgunn: ok, how can i help? and isn't it terribly early there? [10:27] tsdgeos: a little early :) but no help, i just say mp ping about screen on notifs [10:28] ok [10:49] greyback: what's a Rotatino? :D [10:49] damn, you saw [10:49] ;) [10:49] it's a small Rotato [10:50] a rotund latino? [10:51] that works too :D [10:51] greyback: jokes aside, is that branch for ota6? do you have a reviewer? [10:52] tsdgeos: I don't think it's critical for OTA6, as the issue is hard to reproduce. OTA7 would be ok. [10:52] oki [10:52] kgunn: opinion? [10:53] greyback: i was pushing that with bill last night, that what we had was "good enough" [10:53] altho then i saw him whinging about it failing his 100+ case [10:53] but had a hard time following [10:54] kgunn: 100+ apps? [10:54] greyback: i thot it was 100+ photos...but related to ghosting, but maybe i lost context and conflated diff bugs? [10:54] I accept the gallery app after being a picker shadow thing is worth fixing [10:55] ok, so it was "related" [10:55] yep [10:55] greyback: was bill dug in on that being fixed for ota6 ? [10:55] for Bill's issue, I mailed a test patch last night. I want Dan to have a look [10:55] kgunn: you are either in London or up really earlie [10:55] early even [10:55] kgunn: yeah, he was not budging on it, and I see his point [10:56] greyback: ok, so patched the original fix that dan had in silo 17 ? [10:56] kgunn: yep, a patch on top [10:56] it's not in silo17 yet [10:56] as it opens a application management hole, and I want Daniel's opinion on it [10:57] greyback: ok, cool, yeah the other mp albert asking about i assume is ghostApps-preemptQuitters ? [10:57] or something else? [10:58] kgunn: something else https://code.launchpad.net/~gerboland/unity8/fixGhostAppsWithRotatino/+merge/268729 [10:58] much harder to reproduce issue [10:58] which I don't think is that urgent to fix [10:58] ah...the rotatino reference (could also be very small totino's pizzas....a terrible frozen food in the U.S.) [10:59] :) [11:00] greyback: agree it's a good fix, but feels like a bridge too far [11:01] as we are well past the eve of ota6 [11:01] altho it seems pretty isolated [11:01] only in spread only for rotation (or rotatino) [11:02] rah [11:03] greyback: probably not a bad idea to prioritize it's review and testing tho, b/c overlay ppa is technically not frozen... [11:03] true [11:03] they took a snapshot this time, so things can still land [11:03] and there's a "flaw" that if they come back to trunk for a fix, they get whatever else landed ;) [11:13] and davmor2 just up early [11:15] Hi, is there anyone I can CC about dbusmenu bug? [11:15] https://bugs.launchpad.net/ubuntu/+source/libdbusmenu/+bug/1468280 [11:15] Ubuntu bug 1468280 in libdbusmenu (Ubuntu) "libdbusmenu-glib doesn't emit about-to-show signal in Vivid Vervet" [Undecided,New] [11:17] ljagielski: Trevinho should be able to lend a hand [11:19] ljagielski: that bug should have been fixed recently [11:19] ljagielski: that's when running in unity? [11:20] Trevinho: so mine is a duplicate? yes, under Unity === marcusto_ is now known as marcustomlinson [11:21] ljagielski: yes... it's fixed in 15.10, not in .04 [11:21] ljagielski: it should be a dupe of https://bugs.launchpad.net/ubuntu/+source/indicator-appmenu/+bug/1430059 right? [11:21] Ubuntu bug 1430059 in indicator-appmenu (Ubuntu Vivid) "LyX menu is not updated by indicator-appmenu in mode unity-all-menus" [Undecided,Confirmed] [11:22] ljagielski: i can prepare a SRU if needed [11:25] Trevinho: I'd be happy to have the fix in older distros, but I'veonly seen it in 15.04 [11:26] ljagielski: that's the only one affected, in fact [11:40] Trevinho: I'm glad with what I've found out, I will just use indicator-appmenu from ppa myself for now [11:41] ljagielski: I'm about to sru it now*, btw... (*after sru process is done, so subscribe to the bug if you want to help with that :)) [11:41] ljagielski: also this means that you can use a ppa in few minutes [12:56] dandrader: did gerry already catch you ? to review latest https://code.launchpad.net/~unity-team/qtmir/ghostApps-preemptQuitters/+merge/268737 [12:58] kgunn, no [12:59] dandrader: hey! You got mail :) [12:59] dandrader: ok, seems bfiller tested (see phablet) [12:59] and it's all good [13:49] tsdgeos, on that bug, is actually a bugfix [13:49] cimi: that's going to be a tough sell [13:49] :D [13:49] tsdgeos, it is correctly using PreserveAspectFit as stated, it never did before [13:50] tsdgeos, the value we had for the images inside was PreserveAspectFit, but never worked [13:50] now it dows [13:50] does [13:52] cimi: right, but which one does look better? [13:52] tsdgeos, none [13:52] tsdgeos, stretching is not nice too [13:52] tsdgeos, maybe we should have crop? [13:53] honestly if this is a "don't use depreacted mebmers" MR [13:53] i'd prefer to not change the behaviour [13:53] tsdgeos, but the behaviour was wrong [13:53] sure [13:54] but keeping MRs focused really helps [13:55] I see your point [13:55] tsdgeos, I will not use the new ubuntushape then here [13:55] cimi: why not? [13:55] i see that you did move something out of the image [13:56] to the ubuntu shape [13:56] if you don't do that, do we keep the old behaviour? [13:56] tsdgeos, it uses the old ubuntushape if I don't move [13:56] tsdgeos, if I use "source" property instead "image" it switches to the new ubuntushape [13:57] tsdgeos, when using the new ubuntushape, the properties of the Image, like alignment and fillmode MUST be set from the ubuntushape and not inside the source [13:57] tsdgeos, they are ignored [13:57] what is inside the Image { ... } is ignored I mean [13:57] but you say they were ignored before too? [13:58] so maybe just leave them and they will behave as before? [13:58] maybe even add a // TODO investigate proper way to do this since moving this out changes behaviour [13:58] or something? [13:58] tsdgeos, I believe ubuntushape takes the texture of source property and applies horizontal/vertical alignment and fillmode set with sourceFillMode / sourceHAlignment [13:59] tsdgeos, before, there was simply a bug somewhere... [13:59] ok :/ [13:59] i don't know [13:59] tsdgeos, inside the image: Image { ... } we had preserveFit [13:59] tsdgeos, but clearly was not working [14:00] tsdgeos, I can use the old ubuntushape as said, and keep as it is [14:00] but that's also not so great [14:00] as we're doing this branch to use the new one :D [14:01] is there no way we can use the new one and keep the behaviour? [14:01] or you think that's not right either? [14:02] tsdgeos, I think it was using crop before [14:03] i mean we don't know if whoever coded the code actually wanted preserveFit or just had it there from a copy &paste and what actually liked was how it behaved [14:03] o stretch maybe [14:04] * cimi looks [14:13] there is a "ProportionalShape", maybe that helps? [14:15] greyback, it's a different thing [14:15] greyback, that is an ubuntushape which radius is proportional to the dimension [14:15] width/height [14:16] cimi: ok, just thought I'd suggest in case it wasn't known about [14:16] carry on :) [14:22] Out of curiousity, does testDirectionalDragArea work for anyone else? I get a fatal error from ubuntu-ui-toolkit about not enough GL surfaces to go around. I've been looking into it, but want to confirm I'm not the only one [14:23] mterry: i have a fix :D [14:23] tsdgeos, oh good! [14:23] tsdgeos, why was it not seeing the openglcontext aboutToBeDestroyed signal? [14:23] mterry: https://code.launchpad.net/~aacid/ubuntu-ui-toolkit/betterConnectForAboutToBeDestroyed/+merge/268454 [14:24] mterry: "The old code assumes that the UbuntuShape will still be around when the opengl context is destroyed, but that seldom happens so connect the signal to a lambda that will be there and does the cleanup properly" [14:24] tsdgeos, ah makes sense [14:24] tsdgeos, thanks [14:24] yeah now we need the thing to land into a release [14:25] tsdgeos, we can also work around it ourselves by not using Button in that test [14:25] tsdgeos, if we wanted to not be on their schedule [14:25] mterry: but i'd actually leave it [14:25] in case it comes back :D [14:25] it's good we caught this [14:25] a bit too high on the stack [14:25] but at least it's caught somewhere [14:25] tsdgeos, for sure. once they landed the fix, I'd want to go back to Button [14:26] Just saying it was an option if we liked to get passing tests again :) [14:26] Since our unit test framework is not the ideal place for catching UITK bugs :) [14:29] sure [14:29] i tried adding a test there [14:29] but they don't have xvfb nor opengl on their tests [14:29] so wasn't an option [14:29] unless i wanted to spend weeks [14:30] i can also pester the sdk guys to see when/if they'll release a new sdk with the fix [14:31] Yeah, if it's not soon, we should work around it temporarily. Seems silly to have our unit tests hostage to another project [14:39] mterry: ah i reviewer your branch for saving to disk the delay stuff [14:39] it ain't working [14:39] josharenson, looks ok on a first, quick, look [14:40] tsdgeos, I noticed, was about to investigate why... [14:40] It worked for me... [14:41] mterry: maybe you forgot to commit? [14:41] the variable that makes the thing work is never set [14:41] it's either undefined or 0 [14:44] tsdgeos, I used crop, which looks better to me, and removed unused properties [14:44] oki, did you push? [14:48] dandrader: cool, seems more elegant than the original one === dandrader is now known as dandrader|afk === alan_g is now known as alan_g|EOW === dandrader|afk is now known as dandrader