=== _salem is now known as salem_ === salem_ is now known as _salem === duflu_ is now known as duflu [09:21] Saviq: no, didn't try unity7 lately [09:32] oh man, i left yesterday thinking all autopilot tests were passing [09:32] and now i'm back to lots of them failing [09:32] * tsdgeos cries in his corner [09:37] tsdgeos, there there [09:41] tsdgeos, they almost work for me [09:42] good stuff:) [09:42] tsdgeos, after your workaround :) [09:43] Cimi: the comment of = -1, = 0 ? [09:43] tsdgeos, yes [09:43] yeah i thought i had fixed that [09:43] but no [09:46] Cimi: how did you get from all failing to some failing only? [09:46] what was wrong? [09:46] tsdgeos, that thing [09:46] tsdgeos, I'm checking now [09:46] redoing all of them [09:47] Cimi: i pushed a small change [09:47] somehow we were testing for a scope that was there when i tested [09:47] but not anymore [09:47] Hey guys. phone-right-edge was last built on Feb 21st. We're not rebuilding it? [09:47] so i went for another one that is there now :S [09:48] karni: what do you need from that one? [09:49] mzanetti: I would like up-to-date ubuntu with new-scopes face. How would I go about that? [09:49] use demo-stuff ppa instead? [09:49] nope, not that one [09:49] karni: hmm... I guess right now you'd have to build it yourself... [09:50] karni: and most likely also resolve conflicts between new scopes and right edge if you want both [09:50] that's a pity [09:50] karni: but it looks good that I'll get the right edge stuff approved into trunk really soon [09:50] I don't need right edge. I need new scopes on up to date Ubuntu [09:50] karni, unity8 from demo-stuff should be good enough [09:50] Saviq: ok! [09:50] mzanetti: Saviq: thanks guys [10:11] tsdgeos, http://paste.ubuntu.com/7072621/ [10:13] Cimi: you're outdated? [10:13] i changed com.canonical.scopes.onlinescopes [10:13] let me update [10:13] and Title.0 [10:13] maybe i didn't push [10:14] Cimi: i just pushed to r723 [10:15] make sure you have that one [10:15] I'm updating [10:18] I do unity8 -mousetouch & on desktop, and icons are not clicable. known issue? [10:18] categories expand correctly, I can swipe with mouse as well. not click icons, though. [10:21] tsdgeos: Saviq: ↑ [10:21] * karni wanted to test online accounts in unity8 on desktop [10:21] hmmmm [10:22] works here [10:22] * karni updates [10:22] wait [10:22] you mean unity8 as desktop shell? [10:22] tsdgeos: I don't know, this is highly confusing :D Just follosing desktop instructions of demo-stuff https://launchpad.net/~unity-team/+archive/demo-stuff [10:23] I launch unity8 -mousetouch & in terminal [10:24] tsdgeos: yes, I can unlock, scroll, expand categories, but click/longclick with mouse dont work [10:24] karni: i guess running ./run has the same problem, no? [10:24] um [10:25] tsdgeos: you want me to try ./run from trunk? [10:25] * karni will try [10:25] karni: ah, you're running unity8 from packages :D [10:25] karni: preferably from new scopes, but trunk works too [10:26] well, ./run from trunk won't let me unlock (because I'm suing a mouse?) [10:26] *using [10:26] hmmm [10:26] it unlocks then slides back [10:26] no, using a mouse works fine just fine in trunk + ./run [10:26] but then, I see old icons [10:26] let me try with new-scopes [10:26] right [10:27] tsdgeos: yes, same problem in new-scopes. icons don't react to mouse click. [10:28] karni: meaning you get no preview, etc? [10:29] dpm, ping? [10:29] tsdgeos: yes. tab bar works fine, swiping works, but icons are just dead. no reaction to mouse. [10:30] well, I think I'll have to stick to working on the desktop for time being. icons are not only problem, online accounts does not want to create github account, while on desktop it shows up properly. [10:31] and I can open accounts from system-settings on desktop [10:31] karni, console output when clicking / longpressing would be interesting [10:31] i'm confused now :D [10:31] are you running on desktop or not? [10:32] there's no output when I click icons [10:32] after I ./run it [10:32] is there some -v switch? [10:32] maybe didn't understood that last sentence correctly but it seems to me you say that it both works and not works in the desktop :D [10:32] karni, no, should work fine, I'll have a look in a bit [10:32] tsdgeos: I am. I mean I can open system-settings from terminal, but I can't develop a scope - online accounts dont work on the phone the way I need, and clicking doesn't work on desktop. [10:32] Saviq: thank you [10:33] *clicking doesn't work on the desktop [10:33] hi mhr3 [10:33] karni, does the github account show up under gnome-control-center / Online Accounts? can you add it there? [10:35] dpm, hey, any chance you can update the scopes api docs on duc? michael does that usually for us, but well.. you're here now :) [10:35] Saviq: I can add it when I run system-settings -> Accounts. It doesn't show up when I use gnome-control-center -> Online Accounts [10:35] wait, let me try again with gnome-control-center [10:36] scratch that. Saviq, yes, it does show up in System Settings (gnome-control-center) -> Online Accounts [10:36] Saviq: want me to send those github account plugin files? [10:36] karni, no, unrelated [10:36] lp:githubuntu [10:37] so, it only doesn't show up on the phone [10:37] karni, in *theory* it should work in both, but e.g. U1 is more complicated, so only works on phone [10:37] right. [10:38] mhr3, for the html docs, this should happen automatically as soon as there is a new -docs package available. Do we have one containing the new API docs already? [10:39] mhr3, going forward, and this is one of the points that I want to discuss on the UDS session, we want the doxygen docs to be imported into d.u.c directly instead of publishing the raw html [10:39] dpm, no, we can't land because of the qt transition [10:39] tsdgeos: Saviq: long press does work when ./run from new-scopes branch. only click on icon (on the desktop, ./run new-scopes) does not work. [10:39] karni, activation might not work indeed, depends on the result [10:40] karni, i.e. apps won't get launched, but files, http should open [10:40] ah yes, click in ebay does work indeed [10:40] great! [10:40] Saviq: thanks :) [10:40] karni, so, fire extinguished? ;) [10:40] dpm, and yea, i know that's the plan... but i guess that will still take a while [10:41] mhr3, if you've got it in a PPA, we could still publish it. That said, given the fact that we need to update all of the docs in the next couple of weeks, I'd suggest to wait for the landing and getting the docs automatically published, rather than doing it now and doing extra work. Does that sound ok to you? [10:41] Saviq: yes. would be great to have online accounts work as expected, but I suppose this is not at all your area :) do you guys know who works online accounts? [10:41] I'll ask in u1-internal [10:41] dpm, wanted the docs to be in slightly more consumable state during uds [10:42] karni, mardy is one, kenvandine worked on them before, but think he's focused on content hub now [10:42] ah, mardy. okay! will try to get hold of him :) [10:43] mhr3, if you can point me to a PPA, I can look at getting them published from there, instead of the archive, but it will probably have to wait until tomorrow [10:43] dpm, i could build it in a ppa, but a one scopes session is already today [10:47] dpm, should appear in https://code.launchpad.net/~mhr3/+archive/staging shortly [10:58] mhr3, what's the name of the binary package providing the html docs in that ppa? [10:58] dpm, libunity-scopes-doc [10:58] ok [10:58] dpm, it didn't build yet though [10:58] yes, I saw it pending build [10:59] tsdgeos, http://paste.ubuntu.com/7072891/ [11:05] mhr3, ok, let me know when the package builds. I've now set everything up in the server, so that the docs can be published before the session today [11:05] dpm, will do, thx [11:12] Cimi: hmmm [11:12] work for me [11:13] Cimi: what do you see running the tests? [11:13] Saviq: i'm a point that all qmluitests and autopilot tests work me in new-scopes-cleanup, do you mind joining Cimi on running them? [11:24] tsdgeos, yup, will do [11:25] Saviq: i've used 5.0 for this btw [11:25] tsdgeos, yup [11:25] want me to switch to 5.2 and see what happens? [11:25] or let's do that afterwards? [11:25] tsdgeos, it looks like switching scopes is not right [11:26] Cimi: can you record a video? [11:26] maybe [11:26] either that or fix it, your call ;) [11:29] tsdgeos, fix it :) [11:29] tsdgeos, I'm testing on my second ubuntu machine [11:30] Cimi: ok, i'll test your patch :) [11:30] tsdgeos, I have no patch :D [11:30] hopefully will have until I realise what's wrong [11:30] still in testing mode [11:30] Cimi: when you have it :) === MacSlow is now known as MacSlow|lunch [11:35] Cimi: i had a problem in those tests if some notification pops up, seems to break because the mouse moving seems to happen over the notification, make sure it's not that [11:35] ok [11:36] Cimi: and i also had a problem with not enough horizontal mouse movement happening a while back, and had to increase the x move, maybe you still need more ¿? [11:37] tsdgeos, now works °_° [11:37] :D [11:37] tsdgeos, it's not reliable [11:37] it works pretty well here even looped [11:37] andnow doesn't [11:38] Cimi: which test are you running specifically? [11:38] PYTHONPATH=../tests/autopilot autopilot run unity8.shell.tests.test_emulators.DashEmulatorTestCase.test_open_scope_to_the_right [11:38] Saviq, aware of a non-virtualized ppa where i can quickly build something? [11:38] mhr3, demo-stuff [11:38] mhr3, or ppa:unity-team [11:38] Saviq, besides demo-stuff :) [11:39] mhr3, ↑↑ [11:40] Saviq, something's missing there :) [11:40] mhr3, nope [11:40] mhr3, == ppa:unity-team/ppa [11:40] https://launchpad.net/~unity-team/+archive/ppa/ [11:40] the /ppa was ;) [11:41] didn't know that's the default [11:41] Saviq, anyway, thanks [11:41] mhr3, yeah, it is [11:41] nw [11:41] tsdgeos, works on osx with vm but not with real ubuntu [11:42] tsdgeos, could it be related to mouse input? [11:42] tsdgeos, difference is that on osx I have all scopes, ebay and shopping, not on the vm [11:43] maybe I have demo ppa [11:43] it works fine here [11:43] only breaks when you talk to me :D [11:43] because of the notification thing [11:43] hah [11:44] it could be a race too [11:44] Cimi: can you try running that same test in trunk? [11:44] sure [11:51] tsdgeos, works [11:51] mhr3, get hungry! [11:51] all the time or just randomly like in new-scopes? [11:51] tsdgeos, all times [11:56] Cimi, heh, /me almost rdy to leave home [11:56] Cimi: so yeah, give it a try at fixing it, it works fine for me all the time [11:56] http://paste.ubuntu.com/7073118/ [12:05] Cimi, did you rebase https://code.launchpad.net/~unity-team/unity8/new-scopes.carousel-dinamic-fallback/+merge/207451 somewhere yet? [12:06] dpm, non-arm pkg ready @ ppa:unity-team/ppa - sorry had to move it to different ppa cause i forgot our tests don't pass in virtualized ppas [12:06] Cimi, https://code.launchpad.net/~unity-team/unity8/carouselTool-new-dash/+merge/209746 that it? [12:18] Saviq, correct [12:19] Cimi, k [12:19] tsdgeos, trunk passes for me [12:19] oh, new revision for new scopes [12:19] * Cimi pulls [12:29] Cimi: https://code.launchpad.net/~unity-team/unity8/carouselTool-new-dash/+merge/209746 === jhodapp|afk is now known as jhodapp === _salem is now known as salem_ [12:36] tsdgeos, will fix after lunch [12:37] tsdgeos, I'm sprinting with mhr3 and dednick at mines :) [12:37] cool [12:37] canonical engineers office :) [12:38] we have beers and much faster internet === MacSlow|lunch is now known as MacSlow [12:58] tsdgeos, hmm testShell from cleanup segfaults here :/ [13:02] in fake indicators... === alan_g is now known as alan_g|luch === alan_g|luch is now known as alan_g|lunch [13:21] mzanetti: ping. Can you take a look here, please? https://code.launchpad.net/~elopio/notes-app/qml_tests/+merge/210346 [13:22] elopio: ok [13:23] elopio: first of all. thanks soooo much for this [13:24] mzanetti: thanks to you. I had your pdf from the lightning talk and your dev week video all the time :) [13:25] nice :) [13:25] I just would like it better if it wasn't javascript, but it's cool. [13:25] heh [13:37] tsdgeos, yay, only one qmltest fail https://jenkins.qa.ubuntu.com/job/unity-phablet-qmluitests-trusty/1333/? === jono is now known as Guest95937 [13:40] tsdgeos, the ap tests suggest scope-registry doesn't react to the custom signal still, we might temporarily go for "initctl start scope-registry" in our pre-start (just to check the result) [13:41] tsdgeos, but mako looks good! [13:41] tsdgeos, only two failures are due to the 5.0 crasher [13:42] elopio: done. [13:42] elopio: I did some more detailed explaining as usual as I imagine you want some more feedback on those first tries [13:46] mzanetti: yes, thanks. I'll take a look. [13:47] Saviq: greyback hey guys, any thoughts on what's happening with the crash on 1st boot ? do we need to take the bitter revert pill ? [13:47] kgunn: I've tracked down *a* crash problem, know how to fix, working on it now. With luck, that's it [13:48] kgunn, was QML deleting the object from under QSharedPointer, fyi [13:48] mzanetti: I will also need some help to run the qml tests as part of the make check. Can you give me a hand with that? [13:48] oh cool ! [13:48] elopio, GL is required [13:49] elopio, so make check isn't a good target, autopkgtest more so [13:49] well not cool qml deletes objs out from under QSharedPointer, but cool we found it [13:49] elopio, with xvfb-run -s "screen 0 1024x768x24" [13:49] kgunn, http://qt-project.org/wiki/SharedPointersAndQmlOwnership [13:49] it sounds good as part of autopkgtest, right. [13:49] elopio: of course [13:49] elopio, the x24 is required, otherwise gallium can't do virtual gl [13:50] Saviq: but it can be a cmake task too, right? Like make qmltests [13:50] elopio, yeah [13:51] elopio, I was playing with the idea of making a helper that would run the test one at a time (or at least suite one at a time) and record it from xvfb and put it in artifacts [13:51] elopio, but didn't get there yet [13:51] elopio, if you do, please please let us know :) [13:51] Saviq: record in video? [13:51] elopio, and we really need a common place for the QmlTest.cmake and such ;) [13:52] elopio, yeah, recordmydesktop, as autopilot does it [13:52] elopio, always helpful for debugging [13:52] just skimming that puts me in awe of you guys [13:52] Saviq: but the qmltestrunner tests are generally too fast to look at them. [13:52] I'm not sure how it would look in slow motion, maybe it could work. [13:52] elopio, it depends on the test, we (unity8) have plenty that are lookable at [13:53] elopio, also, xvfb is slower and does no transitions, so should actually show the few frames the test does [13:53] elopio, but that also means you need to take better care with the tests (especially the more complicated ones) to make them pass under xvfb [13:54] Saviq: sounds good. I'm using ubuntu-ui-toolkit as the place for all the common autopilot helpers. [13:54] for me, that's also a good place for all the qml test helpers. [13:54] elopio, on the plus side, should increase reproducability between jenkins and dev machines [13:54] elopio, could be a good place indeed, testing is after all part of our sdk [13:55] greyback: packages for you :) https://chinstrap.canonical.com/~mzanetti/right-edge-pkgs.tar.bz2 [13:56] mzanetti: nice, thanks. Hope to get to it in ~1 hour === alecu_ is now known as alecu === alan_g|lunch is now known as alan_g [14:12] greyback, so, I was talking to mzanettt. Will change qpa-mirserver code so that the QPA (SurfaceManager?) gives to QML a QMirSurfaceTextureProvider instead of a MirSurfaceItem. That's to enable multiple QQuickItems to display the same mir surface in the scene graph. E.g. a Window in the Stage and a RunningApplicationTile in the RunningApplicationsGrid showing the same mir surface texture. that removed the need and overhead of taking snapshots [14:14] greyback, think that approach would work? [14:15] dandrader: I'm not fond of passing the raw QMirSurfaceTextureProvider to QML. It won't have information like the Application it belongs to, or it's parent surface, etc. [14:16] dandrader: I'd prefer an intermediary, maybe just "MirSurface" that contains all the surface information (and it's application information), which could be used for QML to create a MirSurfaceItem [14:17] would also be able to manage the buffers, as we only want to use 1 new buffer per frame, not per instance of the MirSurfaceItem [14:19] dandrader: but while I think it's overall a good idea, I don't think we need it right now [14:23] dpm, docs looks good, thanks [14:24] mhr3, hey, scope-registry (or smart proxy, for that matter) don't start on the new upstart event yet, do they? [14:25] Saviq, trunk/distro doesn't, but once we land 0.4 it will [14:25] mhr3, k, that's what I thought [14:26] Saviq: didn't see my question about getting an update on the unity8 crash (or maybe I missed your answer)? [14:26] om26er, hey, I saw we jumped the gun on the unity8 unlock changes? is there a plan? [14:26] didrocks, greyback found the issue, working on a fix now [14:26] oh excellent! unity-mir? [14:26] didrocks, missed your q indeed [14:26] no worry :) [14:26] Saviq, now using process_helpers from unity8. and plan ? [14:27] mhr3, cool :) [14:27] om26er, ah, I saw on Friday there was a problem (i.e. unity8-autopilot not installed)? [14:27] om26er, if that's resolved, we're good [14:27] didrocks, yes, QML deleted an object from under unity-mir's QSharedPointer [14:28] not nice! thanks guys and keep us posted, we'll get that landed ASAP [14:28] Saviq, on a CI job that ran yesterday it seems to be using the new unlocker [14:28] Saviq, not sure about daily smoke testing though [14:28] DONE: finished system-settings test addition branches, one is approved, the other is ready for review. [14:28] TODO: Join Apps team standup [14:28] BLOCKED: None [14:28] NEXT: null [14:28] didrocks, well, it's a known thing about object ownership in Qt, we just missed it [14:29] oop [14:29] https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-mako/5754/console [14:29] Saviq: I wonder if latest AP didn't regress 2 of your tests due to unicode error btw [14:29] tsdgeos, you're green in otto with the hack to start scopes [14:29] Saviq: awesome [14:31] Saviq, so it seems the daily smoke testing is using a different script [14:31] the old one === dandrader is now known as dandrader|afk [14:43] mzanetti: what's the style I should follow to name tests? Something like test_ClickNoteMustExpandIt ? [14:44] dednick, http://pastebin.ubuntu.com/7073881/ [14:44] Saviq: is it with 5.2? [14:45] dednick, that's 5.0, and I just started wondering... [14:45] elopio: in unity we do this. if the qml file is called NoteList.qml the test would be named tst_NoteList.qml [14:45] tsdgeos, UnityEmulatorException: No scope found with id scopes [14:45] elopio: sdk however, seems to have renamed their tests lately to only lowercase letters [14:45] tsdgeos, it's my remaining issue [14:45] mzanetti: I'm doing that for the qml file. But I'm wondering about the name of the test function. [14:45] ah. [14:46] tsdgeos, does that look to you like something related to the 5.2 fix? i.e. looks like stuff's executed on deleted objects (see pastebin above ↑↑↑) [14:46] * Saviq upgrades to 5.2 [14:46] elopio: yeah, mostly ok, CamelCase, but starting lowercase is what we do I think [14:46] elopio: e.g. test_clickNoteMustExpandIt() [14:46] Saviq: ah, that's with the changes I made to fix 5.2 issues :( [14:46] dednick, yeah... [14:46] Cimi: ahhhhhhhh, can you see if reverting this helps? http://bazaar.launchpad.net/~unity-team/unity8/new-scopes-cleanup/revision/722 [14:47] Cimi: do you have the scopes of scopes in there? [14:47] Saviq: wonder if it's this one: https://bugreports.qt-project.org/browse/QTBUG-34351 [14:47] tsdgeos, how do I install them? [14:48] dednick, could be [14:48] Saviq: this one still good? https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta2 [14:48] * Saviq wonders if that's fixed, btw [14:48] Saviq: not familiar with the backtrace [14:48] mzanetti, landing-006 instead [14:48] Saviq: ack, thanks [14:48] should have everything in it === MacSlow is now known as MacSlow|afk [14:49] mzanetti: ok, I've made some changes, when you have some time please take a look. I do appreciate lots of feedback, even for small details :) [14:49] mzanetti, yeah, I'm just upgrading from it and looks fine [14:49] elopio: ok, cool. Really happy to help with this. [14:49] mzanetti, and the tests are https://code.launchpad.net/~unity-team/unity8/fix-5.2-tests/+merge/209058 [14:49] Saviq: puurfect! thanks [14:49] purrfect. even :D [14:50] mzanetti, you might wanna review that branch in the mean time :) [14:50] Saviq: the fix-5.2-tests or elopio's? [14:50] I guess both [14:50] mzanetti, both ;) [14:50] ok [14:51] mzanetti: mine is not urgent, as the notes app will be deprecated. I'm just learning. [14:51] elopio: you can start doing some for the reminders-app then :P [14:51] * mzanetti has been lazy with testing on that [14:52] mzanetti: I will, but not until next week probably. It's just that I started this notes tests before they told me I should just ignore the error I was trying to understand. [14:53] no problem :) but good to know its for learning and not production code that needs to get in. I'll be extra picky then :D [14:54] tsdgeos, works now I think [14:54] tsdgeos, I didn't have the package installed [14:54] well [14:54] we need to make sure it's installed [14:54] but i guess it is [14:54] because CI seems to be happy [14:56] tsdgeos, Cimi, what package? [14:56] Saviq: the one that provides the scopes scope, not sure what's the name [14:56] we should probably tweak ./build [14:56] tsdgeos, unity-scope-scopes [14:59] Saviq: so we should depend on that? [14:59] Saviq, that one [14:59] Saviq: https://code.launchpad.net/~gerboland/unity-mir/fix-shutdown-crash2/+merge/210427 - could you please nominate someone to test & review it for me please? === dandrader|afk is now known as dandrader [14:59] Saviq, I have mhr3 and he told me :P [15:00] tsdgeos, we do already - but ./build -s doesn't install it yet, but it should [15:00] Saviq: also the clickscope is not there, which the tests need too [15:00] ah, ok [15:01] tsdgeos, indeed, the changes we did to debian/control need to be "converted" into ./build, too [15:01] greyback: i don't mind testing...but how to confirm ?...realizing first boot after install is the issue...? [15:01] as we can't automagically install runtime deps, since they're dynamic on package build [15:02] tsdgeos, Saviq so tests pass for me now [15:02] on new scopes and trunk [15:02] greyback, kgunn, I'll do packages and let's verify we don't get crashes on `stop unity8` or `restart unity8` [15:02] Saviq: ack...(..and duh) [15:02] Cimi: good stuff [15:02] let's see CI runs with latest changes i made [15:03] tsdgeos: do you know why waitForRendering() sometimes stalls for ~3 secs? [15:03] Saviq: so you're just gonna stick it in silo ? [15:03] kgunn, nah, x-build locally [15:03] kgunn, you take care of silo? [15:03] Saviq: yeah..i'll do silo [15:03] mzanetti: no clue, if i had to guess it may be because of racyness, but no idea really [15:04] Cimi: can you fix https://code.launchpad.net/~unity-team/unity8/carouselTool-new-dash/+merge/209746 should be quick [15:06] tsdgeos, sure [15:07] tsdgeos, which target branch? [15:07] tsdgeos, new-dash-cleanup? [15:08] Saviq, ? [15:08] Cimi: no, the same [15:08] ok [15:09] kgunn, http://people.canonical.com/~msawicz/libunity-mir1_0.3+14.04.20140311.1-0ubuntu1_armhf.deb [15:11] greyback, think we should add FIXME/TODO? setting the ownership should really work (did you try on .data(), or just the QSP itself?) [15:11] ooops [15:11] I pushed in new-scopes-clean-to-trunk [15:11] let me revert [15:11] Cimi, I can push if you want [15:11] overwrite, I mean [15:12] Saviq, ok thx [15:12] Saviq, I did bzr push :parent on the wrong folder [15:12] Saviq: I tried setting ownership on hte pointer returned by .data() [15:13] since QML engine can't deal with the QSP [15:13] greyback, mhm, wonder why that didn't work [15:13] greyback, worth investigating [15:13] tsdgeos, fixed [15:13] Saviq: agreed I am suspicious why [15:13] will add a fixme in the short term [15:13] Cimi: thanks [15:13] Cimi, overwrote [15:16] tsdgeos, shall we move to qt 5.2 now? [15:17] Cimi: you mean testing that the new-scopes-cleanup tests pass with 5.2? [15:17] greyback, +1 [15:17] tsdgeos, gotcha [15:17] yep [15:17] Cimi: if you have time for that it'd be great yes [15:17] didrocks, kgunn is preparing a CI Train row for the unity-mir fix [15:17] Cimi: also add the fix 5.2 branch to your tests [15:17] didrocks, verified it works already [15:17] didrocks: line 35 [15:18] tsdgeos, I do, ok [15:18] Saviq: kgunn: \o/ assigning the wagon [15:18] lol [15:20] Saviq: um, how would you feel about a quick fix? [15:21] elopio: some comments [15:21] kgunn: done and building [15:21] Saviq: done [15:21] thanks greyback btw! [15:21] Saviq: as in, this crash is caused by unloading the indicators in the model destructor. So a bit pointless really. [15:23] Mirv: due to this unity-mir urgent landing, you will have to rebuild (if I can't that done before my EOD) unity-mir in your silo === MacSlow|afk is now known as MacSlow [15:29]