| Mirv | Saviq: ok.. | 04:01 |
|---|---|---|
| mhr3 | sil2100, didrocks, are we ready to do the unity bump now? | 07:50 |
| didrocks | mhr3: it seems like unity AP tests don't pass | 07:51 |
| didrocks | mhr3: so I think that should be fixed first? | 07:51 |
| didrocks | otherwise, we won't be able to release it :p | 07:51 |
| mhr3 | kmn | 07:51 |
| sil2100 | ;) | 07:51 |
| didrocks | mhr3: oh sil2100! | 07:52 |
| sil2100 | bregma: ^ | 07:52 |
| didrocks | sil2100: I wanted to ask you having a look at that ;) | 07:52 |
| didrocks | at least to see the current state | 07:52 |
| sil2100 | Aye aye sir ;) | 07:52 |
| sil2100 | Too bad Brandon is around later | 07:52 |
| didrocks | mhr3: btw, thorst still didn't file an entry for this | 07:52 |
| didrocks | mhr3: would be nice if he can do that | 07:52 |
| mhr3 | didrocks, i made him add it to 93 | 07:52 |
| didrocks | mhr3: oh, he added in the middle :/ | 07:53 |
| didrocks | harder to track | 07:53 |
| didrocks | ok, thanks mhr3 | 07:53 |
| mhr3 | didrocks, it's part of that feature | 07:53 |
| mhr3 | feel free to split if it makes it easier for you to track | 07:53 |
| didrocks | mhr3: no no, that's fine, let's keep it once | 07:56 |
| * sil2100 wonders why he has 2 network indicators on his desktop | 08:00 | |
| mhr3 | sil2100, for cases when one gets lost | 08:01 |
| mhr3 | you know... backup thing | 08:01 |
| mzanetti | Saviq: yesterday night unity8 crashed for me. so I looked up the crash dump and unpacked it, but it didn't contain a CoreDump. | 08:01 |
| sil2100 | Aaah, make sense ;) | 08:02 |
| mzanetti | sil2100: apport-bug said "your device doesn't have enough memory for this" | 08:02 |
| mzanetti | sorry... Saviq ^ | 08:02 |
| mzanetti | any ideas how to get some info out of it nevertheless? | 08:02 |
| mhr3 | sil2100, is it just me who can't find which unity tests are actually failing? | 08:07 |
| Saviq | mzanetti, did you try unpacking on your laptop? | 08:14 |
| mzanetti | Saviq: no. but the .crash package is only 300kb while usually they are more like 10MB | 08:15 |
| Saviq | mzanetti, yeah, looks like we won't get anything out of it | 08:15 |
| Saviq | mzanetti, it probably failed to compress the core dump when crashed (out of mem?) | 08:15 |
| mzanetti | mhm... yeah. might be. | 08:16 |
| mhr3 | how... why does mtp come up when the phone is off and it's just charging? | 08:16 |
| mhr3 | didn't even know that's possible | 08:16 |
| mzanetti | mhr3: I don't think its really off while charging | 08:17 |
| mhr3 | still, i wouldn't expect anything running in that mode | 08:17 |
| mzanetti | still, not sure if mtp should really be running | 08:17 |
| mzanetti | yeah | 08:17 |
| sil2100 | mhr3: we need to re-run the tests sadly | 08:18 |
| sil2100 | didrocks: is there a big daily tick planned now, or can I run the unity stack manually? | 08:18 |
| didrocks | sil2100: if nothing is running, please feel free | 08:19 |
| Saviq | mzanetti, why no top-approve lazy image? | 08:36 |
| mzanetti | Saviq: did I forget? | 08:37 |
| Saviq | mzanetti, dunno :D | 08:37 |
| mzanetti | Saviq: done | 08:37 |
| Saviq | mzanetti, tx | 08:37 |
| mzanetti | Saviq: regarding that mail from veebers. we still do have the upa module with mir, no? | 08:56 |
| Saviq | mzanetti, upa != qpa | 08:57 |
| Saviq | mzanetti, upa is the autopilot plugin | 08:57 |
| Saviq | mzanetti, and well, sure, we could have it still, but update to mir | 08:57 |
| Saviq | or have a new one | 08:57 |
| Saviq | sorry if the != was not something you got mixed up, but that was the only way I made sense of your question ;P | 08:57 |
| mzanetti | Saviq: I'm not sure if a unity dbus interface is way to go | 08:57 |
| mzanetti | Saviq: yeah... youre right. I mixed it up | 08:58 |
| Saviq | mzanetti, well, there isn't going to be one in mir ;) | 08:58 |
| mzanetti | Saviq: where does unity get the screen information from? | 08:58 |
| Saviq | mzanetti, from mir | 08:58 |
| mzanetti | and mir? | 08:58 |
| mzanetti | Saviq: what I mean is, it seems it would tie autopilot too closely to unity. for example running the unity8 tests we'd have a chicken egg problem, unless doing some if (unity8) hacks in autopilot etc | 09:00 |
| Saviq | mzanetti, it would just be one module in autopilot | 09:00 |
| Saviq | mzanetti, there's an X11 one, there's UPA (Unity...? Ubuntu...?), so there can be another Unity8 one | 09:00 |
| Saviq | mzanetti, it just loads all of them in sequence (I think) and uses one that works | 09:01 |
| mzanetti | Saviq: mhm... ok... yeah. makes sense. | 09:01 |
| mzanetti | Saviq: but it still might be read from the same place where Mir actually gets this information from. I don't think Mir is going to make those values up. (except for configured resolution maybe? if that's still a thing nowadays) | 09:03 |
| Saviq | mzanetti, not gonna work, autopilot shouldn't get direct access to that | 09:04 |
| didrocks | mhr3: I think sil2100 will need to revert a commit in unity8 | 09:04 |
| Saviq | mzanetti, and also there's more to that than just the screen geometry | 09:04 |
| didrocks | mhr3: an universe dep added | 09:04 |
| didrocks | so will take more time to get the results first | 09:04 |
| Saviq | mzanetti, like it should take struts (panels) into account, and that's something only Mir/Unity8 will know about | 09:04 |
| mzanetti | don't see why it would take panels into account as autopilot is not the one placing the window geometry. but anyways. I guess there are things only unity knows about | 09:06 |
| mzanetti | ok... sure, fine with me. just wanted to make sure its really what we want | 09:06 |
| sil2100 | didrocks: already did | 09:07 |
| sil2100 | didrocks: we have test results for intel from the previous run and there's 17 test failures, so hm... it's not bad, I think that's below the threshold - most of the failures are IBus failures | 09:11 |
| sil2100 | didrocks: (just out of IBus curiosity I'll check what's wrong with those) | 09:11 |
| sil2100 | didrocks: heh, ibus doesn't start | 09:12 |
| didrocks | sil2100: ok, so now, let's add mhr3's changes on top of that? | 09:14 |
| didrocks | the libunitycore ABI break | 09:14 |
| didrocks | then rebuilding unity7 | 09:14 |
| didrocks | and then unity8 | 09:14 |
| didrocks | and land that | 09:14 |
| didrocks | live in a free and happy world… | 09:14 |
| * sil2100 doesn't like the sound of that | 09:14 | |
| didrocks | sil2100: don't fear! | 09:20 |
| didrocks | sil2100: or do you want to publish unity7 now? before mhr3's merge | 09:20 |
| didrocks | mhr3: how long for you to do them? | 09:20 |
| mhr3 | didrocks, sorry, how long to do what? | 09:21 |
| didrocks | mhr3: the ABI breakage merge | 09:21 |
| didrocks | you have one branch for unity7 | 09:21 |
| didrocks | + bumping the build-dep on unity8, right? | 09:21 |
| didrocks | Saviq: how is unity8 looking btw? good to release? | 09:21 |
| Saviq | didrocks, always | 09:21 |
| didrocks | gorgiousness! | 09:21 |
| Saviq | didrocks, I completed an ask for it this morning | 09:21 |
| Saviq | didrocks, not yet in landing plan, though | 09:22 |
| mhr3 | didrocks, it's no api change, so unity8 will be really just version bump change | 09:22 |
| didrocks | Saviq: maybe your hope will get realized thanks to mhr3's ABI breakage :p | 09:22 |
| Saviq | ;) | 09:22 |
| didrocks | mhr3: yeah, just bump the build-dep so that it's easier for us please | 09:22 |
| mhr3 | didrocks, k | 09:22 |
| didrocks | mhr3: so, ack to propose them/merge them now | 09:22 |
| mhr3 | didrocks, yey! :) | 09:23 |
| mhr3 | didrocks, eh | 09:23 |
| mhr3 | didrocks, there's this weird thing in unity8 | 09:23 |
| mhr3 | libunity-core-6.0-dev (<< 7.80), | 09:23 |
| mhr3 | is there some kind of "&& > 7.1.1"? | 09:23 |
| mhr3 | >= | 09:23 |
| mhr3 | Saviq, or can we get rid of that ^ | 09:24 |
| didrocks | I think this comes from the old time | 09:24 |
| didrocks | (when they had their own copy of unity7) | 09:25 |
| mhr3 | so +1 to remove it? | 09:25 |
| didrocks | well, let's wait for Saviq maybe ;) | 09:25 |
| didrocks | mhr3: get the unity7 branch merged meanwhile | 09:25 |
| didrocks | I hope that the upstream merger is using unity7 repository as well | 09:25 |
| Saviq | mhr3, I told pstolowski yesterday that we can get rid of the << | 09:25 |
| Saviq | mhr3, https://code.launchpad.net/~stolowski/unity8/cancel-previews/+merge/187249 | 09:26 |
| Saviq | and it's happening there | 09:26 |
| mhr3 | Saviq, mind approving that one then? will be bump + needed change :) | 09:26 |
| sil2100 | didrocks: well, I don't want to publish without rebuilding... | 09:27 |
| Saviq | mhr3, you mean reviewing? ;) | 09:27 |
| didrocks | sil2100: ok, let's get mhr3's breakages in then | 09:27 |
| mhr3 | Saviq, didn't you do that already? :) | 09:27 |
| Saviq | mhr3, I'd rather you, too, I've no idea about the cancellables | 09:27 |
| mhr3 | Saviq, k, i take it as full power to top-approve :) | 09:27 |
| Saviq | mhr3, dude, you have it | 09:27 |
| Saviq | mhr3, especially for the Unity plugin, which, after all, you guys are maintaining ;) | 09:30 |
| Saviq | mhr3, you even have an INPROGRESS work item for that :D | 09:31 |
| mhr3 | Saviq, ah, thought it's doing a bit of qml as well | 09:31 |
| mhr3 | but yea... i don't need you then :P | 09:31 |
| Saviq | mhr3, nah, that happens in mzanetti's branch | 09:32 |
| Saviq | or should ;) | 09:32 |
| mzanetti | what? | 09:32 |
| Saviq | mzanetti, who cancels the previews? does it happen automagically in the Unity plugin? | 09:33 |
| mzanetti | Saviq: yeah. when I request a preview, all previous requests are cancelled automatically | 09:34 |
| Saviq | mzanetti, k | 09:34 |
| mzanetti | Saviq: there still will be a manual cancel method tho | 09:34 |
| mzanetti | Saviq: which I should call when closing the preview | 09:34 |
| Saviq | mzanetti, yup | 09:34 |
| seb128 | sil2100, I think mhr3 trolled you a bit for the duplicate indicator | 09:35 |
| seb128 | sil2100, you have nm-applet (which we use on desktop) and likely indicator-network though ubuntu-system-settings depends | 09:35 |
| seb128 | through* | 09:36 |
| sil2100 | Ah ;) | 09:36 |
| sil2100 | mhr3: could you ping me once all is prepared for the big-unity-breakage? | 09:41 |
| sil2100 | Merged in and etc. | 09:41 |
| mhr3 | sil2100, the unity7 branch got approved | 09:42 |
| mhr3 | sil2100, finishing review of the unity8 one | 09:43 |
| sil2100 | Excellent news, can't wait to see everything broken ;) | 09:43 |
| mhr3 | sil2100, then again that won't pass until 7 is merged | 09:43 |
| Saviq | Mirv, https://code.launchpad.net/~saviq/unity8/fix-qt51-ftbfs-and-unused/+merge/188003 | 09:47 |
| Saviq | mzanetti, you can review ↑, should be 30s | 09:48 |
| mzanetti | Saviq: done | 09:48 |
| Saviq | mzanetti, o/ | 09:49 |
| === seb128_ is now known as seb128 | ||
| mhr3 | Saviq, sooo... 5.1 because of the rendering lock ups? | 10:10 |
| mhr3 | or just an experiment to see whether is will fix it? | 10:10 |
| Saviq | mhr3, we've supported 5.1 for some time now | 10:11 |
| Saviq | mhr3, just it got broken yesterday | 10:11 |
| Saviq | mhr3, so got it fixed | 10:11 |
| mhr3 | ah | 10:11 |
| Saviq | mhr3, but well, https://bugs.launchpad.net/ubuntu/+source/unity8/+bugs?field.tag=qt5.1 | 10:11 |
| mhr3 | Saviq, then the question is - will it be re-integrated in the next week? :) | 10:11 |
| mhr3 | meh... tiny things :) | 10:12 |
| Saviq | mhr3, btw... we won't be able to merge the cancel-previews branch until unity goes through | 10:14 |
| Saviq | since we're not in the same stack | 10:14 |
| mhr3 | sil2100, ^ problem? | 10:15 |
| nic-doffay | Saviq, first branch: https://code.launchpad.net/~nicolas-doffay/unity8/search-indicator/+merge/188008 | 10:15 |
| Saviq | nic-doffay, thanks | 10:18 |
| mhr3 | sil2100, Saviq, then we can just merge lp:~mhr3/unity8/bump-unity-core-dep manually which only does the bump? | 10:20 |
| Saviq | mhr3, what does it matter? | 10:20 |
| mhr3 | Saviq, to ensure that unity8 gets rebuild | 10:20 |
| mhr3 | and in the correct order | 10:20 |
| Saviq | mhr3, it will get rebuilt when we merge the preview-cancel branch | 10:21 |
| Saviq | mhr3, before that we don't care | 10:21 |
| Saviq | mhr3, since you're bumping SONAME | 10:21 |
| Saviq | mhr3, so unity8 depends on the previous SONAME | 10:21 |
| mhr3 | Saviq, yea, but it conflicts with the old soname, so upgrade would remove unity8 | 10:21 |
| Saviq | mhr3, so it won't upgrade | 10:22 |
| Saviq | mhr3, or if someone does dist-upgrade, they'll know what they're doing | 10:22 |
| Saviq | or at least should | 10:22 |
| mhr3 | Saviq, ultimately this is distro thing, didrocks wanted it | 10:22 |
| Saviq | mhr3, why does it conflict, btw? | 10:22 |
| mhr3 | pkg magics | 10:23 |
| Saviq | isn't the whole idea behind SONAMEs that you can have multiple versions installed simultaneously? | 10:23 |
| mhr3 | both would provide unity-common which they can't or something | 10:23 |
| Saviq | ah | 10:23 |
| mhr3 | yea.. i know it's weird | 10:23 |
| Saviq | something's broken there, then... | 10:23 |
| Saviq | didrocks, sil2100 how do we proceed with libunity-core / unity8? we can't merge the >> 7.1.1 until it's released | 10:24 |
| mhr3 | * >= | 10:25 |
| sil2100 | Ok | 10:25 |
| sil2100 | hmmm | 10:25 |
| Saviq | or that | 10:25 |
| sil2100 | mhr3, Saviq: let me run the unity stack to rebuild unity-core | 10:26 |
| Saviq | sil2100, works for me, we'll have to wait for it to get into distro, though, right? seeing as they're not part of the same stack? | 10:26 |
| sil2100 | Saviq: it will at least help in getting the change merged into unity8, as the merger is using daily-build PPA (or at least should) | 10:28 |
| sil2100 | Saviq: for building the unity8 stack, we can force using the whole PPA | 10:28 |
| Saviq | sil2100, ok if it does then we're good | 10:28 |
| Saviq | sil2100, yup | 10:28 |
| mhr3 | so... false alarm | 10:28 |
| sil2100 | Saviq: fginther told me that all projects are using daily-build by default, if not disabled | 10:29 |
| Saviq | sil2100, k cool | 10:29 |
| Saviq | mhr3, btw, the -isactive branch is waiting for that, too? | 10:30 |
| Saviq | mhr3, seeing as it fails to build completely? should it depend on >= 7.1.1, too? | 10:31 |
| mhr3 | Saviq, yep | 10:31 |
| Saviq | nic-doffay, is the search-progress branch MP'd separately? | 10:31 |
| mhr3 | Saviq, nic's branch is built on top of my search-in-progress (prop for the unity plugin), and search-isactive is extending search-in-progress, so when reviewing that pls ignore the plugin changes, the entire thing is in search-isactive | 10:33 |
| Saviq | mhr3, yeah, but it should prereq it, then, no? | 10:34 |
| mhr3 | Saviq, well, it's an older version of the branch that doesn't need the bump | 10:34 |
| mhr3 | so that one isn't mp-ed by itself | 10:34 |
| Saviq | mhr3, won't it conflict then if we merge the plugin changes for search progress and then -isactive? | 10:35 |
| mhr3 | cause you know... why do things the simple way :P | 10:35 |
| mhr3 | Saviq, nah, should be fine | 10:35 |
| Saviq | mhr3, k | 10:35 |
| mhr3 | if it will i'll resolve it myself | 10:35 |
| mhr3 | but i trust bzr :P | 10:35 |
| Cimi | mzanetti, jenkins still hates me | 10:50 |
| mzanetti | Cimi: hmm... but its only 3 of the tests now | 10:55 |
| mzanetti | After 10.0 seconds test on HudButton.opacity failed: 1.0 != dbus.Double(0.5, variant_level=1) | 10:55 |
| mzanetti | this seems strange... do all the tests pass now on your device? | 10:56 |
| mzanetti | Cimi: ^ | 10:56 |
| Cimi | mzanetti, where do you see this? | 10:57 |
| mzanetti | Cimi: in the jenkins logs | 10:57 |
| Cimi | mzanetti, ah, last link | 10:57 |
| Cimi | mzanetti, was checking out the first | 10:57 |
| === mzanetti is now known as mzanetti|lunch | ||
| dednick | mzanetti|lunch: ping. is dbus allowed in unity8 unit tests? | 11:00 |
| dednick | Saviq: ^ ? | 11:01 |
| Saviq | dednick, ideally no | 11:01 |
| Saviq | dednick, everything should be mocked out | 11:01 |
| dednick | Saviq: hm. it's to do with unitymenumodel. | 11:02 |
| Saviq | dednick, I thought so | 11:03 |
| Saviq | dednick, and well, I imagine it'd be quite big to mock out? | 11:04 |
| dednick | Saviq: er yeah, i wouldnt even know how to go about it. Would def need changes in qmenumodel-qml | 11:04 |
| Saviq | dednick, well, we already replace qmenumodel-qml | 11:05 |
| Saviq | dednick, i.e. we have a QML-only mock | 11:05 |
| dednick | Saviq: i mean the unitymenumodel inseide qmenumodel-qml | 11:05 |
| dednick | Saviq: ah. hm.. maybe possible | 11:05 |
| Saviq | dednick, MacSlow added a minimal UnityMenuModel.qml to his snap decisions branch | 11:06 |
| Saviq | dednick, to make the qml tests pass | 11:06 |
| dednick | Saviq: ok. i'm not sure about mocking in qml though. the object i'm testing is cpp. | 11:07 |
| MacSlow | dednick, just in case... http://bazaar.launchpad.net/~macslow/unity8/extended-snap-decisions-part1/revision/311 | 11:07 |
| Saviq | dednick, well, yeah, we might need to move down to cpp | 11:07 |
| dednick | MacSlow: ta | 11:07 |
| dednick | Saviq: ok, i'll look into it | 11:08 |
| dednick | thanks | 11:08 |
| mhr3 | Saviq, could we change the renderer video uses? it's using the dvd-cover portrait thing even though pretty much all the results we'll have in 13.10 will be landscape | 11:10 |
| mhr3 | (talking about the online category there)... not sure about my videos, those will be both i suppose | 11:11 |
| Saviq | mhr3, yeah, but more landscape anyway | 11:11 |
| Saviq | mhr3, probably, yeah, pick it up with Oren? | 11:11 |
| mhr3 | Saviq, will do | 11:12 |
| didrocks | how is it going? | 11:25 |
| didrocks | mhr3: merged both? | 11:25 |
| didrocks | sil2100: ^ | 11:25 |
| sil2100 | didrocks: untiy is still building! So I guess soon unity8 pieces should be merged in | 11:26 |
| didrocks | ok ;) | 11:27 |
| === MacSlow is now known as MacSlow|lunch | ||
| mhr3 | didrocks, unity8 can't be merged before unity7 is | 11:28 |
| didrocks | mhr3: you mean, in the daily-build ppa? | 11:28 |
| mhr3 | right | 11:28 |
| mhr3 | or is it there now? | 11:28 |
| sil2100 | mhr3: it's building still | 11:29 |
| Saviq | paulliu, ping | 11:29 |
| mhr3 | yea, so waiting... | 11:29 |
| Saviq | Cimi, one for you, bug #1231937 | 11:29 |
| ubot5 | bug 1231937 in Unity 8 "Video renderer is not to spec" [High,Triaged] https://launchpad.net/bugs/1231937 | 11:29 |
| Saviq | mhr3, https://launchpad.net/~ubuntu-unity/+archive/daily-build/+sourcepub/3532745/+listing-archive-extra | 11:36 |
| mhr3 | Saviq, thx, approved cancel-previews | 11:37 |
| Saviq | mhr3, probably too soon | 11:37 |
| Saviq | mhr3, amd64 and armhf aren't there yet | 11:37 |
| mhr3 | oh well.. we'll see | 11:38 |
| mhr3 | Saviq, btw seems like we need special weather grid as well | 11:38 |
| Saviq | +will | 11:38 |
| mhr3 | Saviq, ? | 11:39 |
| Saviq | mhr3, "we will need"? | 11:40 |
| mhr3 | we already have weather results if you search for cities | 11:40 |
| Saviq | mhr3, ah, that hting | 11:40 |
| mhr3 | and the text is already overflowing a bit :) | 11:40 |
| mhr3 | Saviq, and one last renderer q - seems we can forget about dynamic renderer for 13.10, i'll replace all "dynamic" with something static if you're ok with that? | 11:42 |
| Saviq | mhr3, +1 | 11:42 |
| Saviq | mhr3, one more for dash plugins, btw | 11:43 |
| mhr3 | oh.. crap, we don't have enough renderer types for that in apps :/ | 11:44 |
| Saviq | mhr3, yeah, it's really a different content type, isn't it... | 11:44 |
| mhr3 | well.. yea | 11:44 |
| Saviq | but it's the same problem, Iknow | 11:45 |
| * Saviq hates gdocs' slides... I WANT TO ZOOM IN! | 11:45 | |
| mhr3 | also, do i see a weather preview there? :) | 11:46 |
| mhr3 | wooo | 11:46 |
| Saviq | yeah | 11:47 |
| mhr3 | adding to "not-to-spec list" | 11:47 |
| mhr3 | paulliu, you had a branch that rendered the info hints? | 11:47 |
| === hikiko is now known as hikiko|lunch | ||
| mhr3 | was that only for movies or for any preview? | 11:48 |
| Saviq | bug #1231951 bug #1231948 | 11:50 |
| ubot5 | bug 1231951 in Unity 8 "Need a renderer for dash plugins" [High,Triaged] https://launchpad.net/bugs/1231951 | 11:50 |
| ubot5 | bug 1231948 in Unity 8 "Need a renderer for weather" [High,Triaged] https://launchpad.net/bugs/1231948 | 11:50 |
| Saviq | Cimi, those are for you atm, but we might take them away from you at some point ;) | 11:51 |
| Saviq | mhr3, so what do we do with the dash plugins? | 11:51 |
| mhr3 | Saviq, i'll think of something | 11:51 |
| Saviq | mhr3, can we ride the wave on unity's ABI bump again today? | 11:51 |
| mhr3 | new content type or whatever | 11:51 |
| * Saviq remembers saying that we should use strings for that :D | 11:52 | |
| mhr3 | Saviq, i knew you'd mention that :P | 11:52 |
| mhr3 | Saviq, anyway, it'll be just new api, no need to rebuild anything | 11:52 |
| mhr3 | besides libunity itself | 11:52 |
| Saviq | mhr3, k | 11:53 |
| Saviq | mhr3, affected libunity on bug #1231951 | 11:55 |
| ubot5 | bug 1231951 in Unity 8 "Need a renderer for dash plugins" [High,Triaged] https://launchpad.net/bugs/1231951 | 11:55 |
| mhr3 | thx | 11:55 |
| * Saviq food | 11:55 | |
| paulliu | mhr3: not yet. I'm about to implement it. | 12:00 |
| paulliu | mhr3: Currently fixing the design-branch currently | 12:00 |
| mhr3 | sil2100, didrocks http://10.97.2.10:8080/job/unity8-saucy-i386-autolanding/509/console | 12:06 |
| mhr3 | too soon, or daily ppa not enabled there? ^ | 12:06 |
| === alan_g is now known as alan_g|lunch | ||
| sil2100 | fginther: ping | 12:10 |
| paulliu | mzanetti|lunch: https://code.launchpad.net/~paulliu/unity8/unity8-fixgenericpreview/+merge/187762 | 12:11 |
| paulliu | mhr3: hi | 12:15 |
| paulliu | mhr3: so is there any design docs for the hints to be displayed? | 12:15 |
| Cimi | guys I have issues with network | 12:20 |
| Cimi | no more network card | 12:20 |
| Cimi | what happened with indicator-network? | 12:20 |
| Cimi | on my laptop | 12:20 |
| Cimi | I'm considering reinstalling | 12:20 |
| sil2100 | fginther: ping | 12:27 |
| sil2100 | mhr3: we need fginther around to access the merger bits ;/ | 12:28 |
| === mzanetti|lunch is now known as mzanetti | ||
| mzanetti | paulliu: cheers | 12:36 |
| mzanetti | greyback: hey | 12:37 |
| mzanetti | greyback: it's still not possible to run unity with mir and some command line args. | 12:37 |
| mzanetti | as mir aborts for some reason | 12:38 |
| greyback | mzanetti: yep https://bugs.launchpad.net/mir/+bug/1226227 | 12:38 |
| ubot5 | Ubuntu bug 1226227 in Mir "libmirserver parses arguments and fails if it's not something it understands" [Medium,Triaged] | 12:38 |
| greyback | mzanetti: it's pretty high up on my fix list, can you wait a day? | 12:39 |
| mzanetti | greyback: sure... I just need to start debugging stuff with mir enabled. and this makes it quite uncomfortable | 12:39 |
| paulliu | mzanetti: thanks. I'll now work on infoHints and the movie-preview. | 12:40 |
| mzanetti | paulliu: ok. cool. FYI: https://code.launchpad.net/~mzanetti/unity8/switching-previews/+merge/187327 | 12:41 |
| greyback | mzanetti: yep, understood. | 12:41 |
| mhr3 | paulliu, Saviq, i'd say let's make the info hints look like what's designed for weather, objections? | 12:42 |
| mhr3 | of course specific preview renderers might know better | 12:42 |
| mhr3 | paulliu, that's last slide on https://docs.google.com/a/canonical.com/presentation/d/1X3_G6reCpcFJxzxn1Xtv9wK5n3-wOW-6z2l0qSZ8-sg/edit#slide=id.g1026aae2e_225 | 12:42 |
| mzanetti | cyphermox: what is the correct component to report bugs related to Networking on the phone? | 12:43 |
| mzanetti | cyphermox: to be precise: cellular networking doesn't really work at all. Except you completely disable wifi and reboot. | 12:44 |
| mzanetti | and even then it only comes up every second time | 12:44 |
| paulliu | mhr3: ok.. I'll puit that after the descriptions in GenericPreview. | 12:44 |
| === alan_g|lunch is now known as alan_g | ||
| === hikiko|lunch is now known as hikiko | ||
| === MacSlow|lunch is now known as MacSlow | ||
| Saviq | paulliu, mhr3 yeah, sounds about right, although that takes quite a bit of space | 12:51 |
| mhr3 | Saviq, that's good, some previews often have only that and no other data | 12:51 |
| Saviq | mhr3, right, good for me then | 12:51 |
| cyphermox | mzanetti: make sure you are running the very latest image, and you can file bugs against network-manager with a full copy of syslog | 12:52 |
| didrocks | mhr3: not sure TBH… | 12:52 |
| didrocks | mhr3: can something try to build it locally, like Saviq? | 12:52 |
| didrocks | and we push to trunk directly | 12:52 |
| didrocks | (as dailies is just around the corner, we'll know if it's broken quickly and can back out) | 12:52 |
| mhr3 | didrocks, sil suggested waiting for fginther | 12:52 |
| didrocks | mhr3: I'm afraid about the timing | 12:52 |
| didrocks | time to get that in | 12:52 |
| didrocks | time to test | 12:52 |
| mzanetti | cyphermox: ok. will do so... thanks | 12:52 |
| didrocks | time to push that to distro | 12:52 |
| didrocks | time to have someone in the release team to review it… | 12:53 |
| didrocks | let's see what Saviq thinks about it | 12:53 |
| * didrocks takes a shower meanwhile | 12:53 | |
| Saviq | didrocks, mhr3, I'm here, but not sure what's happening? | 12:54 |
| mhr3 | Saviq, just a check that it builds afaiu | 12:54 |
| * Saviq reads backlog | 12:54 | |
| didrocks | and push to trunk if it builds | 12:54 |
| didrocks | then, we kick dailies | 12:54 |
| didrocks | and we know/test if it's good or not | 12:54 |
| didrocks | will be easy to revert if not | 12:55 |
| Saviq | didrocks, mhr3 do we really need this in? can't it just wait its turn? what're we gaining? | 12:56 |
| didrocks | Saviq: maybe 2 hours | 12:56 |
| Saviq | didrocks, mhr3 is it only about rebuilding unity8 after libunity merged? | 12:56 |
| didrocks | Saviq: and avoiding having sil2100 going to bed at 10PM | 12:56 |
| mhr3 | Saviq, i'm fine if we push just the bump | 12:57 |
| mhr3 | lp:~mhr3/unity8/bump-unity-core-dep | 12:57 |
| mhr3 | Saviq, yep | 12:57 |
| didrocks | Saviq: right | 12:57 |
| Saviq | ok, /me tries with daily-build enabled | 12:57 |
| Saviq | sil2100, right, so upstream merger doesn't have daily-build in the end, does it | 12:57 |
| mhr3 | right | 12:57 |
| Saviq | though so, and thought it was on purpose - as we only wanted per-stack archives... just that unity8 should be in the same stack unity is... | 12:59 |
| sil2100 | Saviq: I think the merger doesn't use the notation of stacks, but probably they removed the dependency indeed | 13:02 |
| Saviq | sil2100, it does | 13:03 |
| Saviq | sil2100, we have a per-stack local repo | 13:03 |
| Saviq | sil2100, so anything that gets merged into trunk | 13:04 |
| Saviq | sil2100, goes into that per-stack local repo | 13:04 |
| Saviq | sil2100, so say, if unity-api merges something in trunk, we can build against it even though it's not there yet in distro - but it's not using daily-build PPA | 13:04 |
| sil2100 | That's something ;) | 13:04 |
| Saviq | we should have PPAs for that - /me loves that daily release uses PPAs for building and not local pbuilders | 13:05 |
| Cimi | Saviq I'm feeling a lil sick I need to rest for a bit… bad food maybe :\ I am working on autopilot tests (again, they fail on jenkins) and plugin for sim card for welcome wizard | 13:15 |
| Cimi | using ofono the latter | 13:15 |
| Saviq | Cimi, k, feel better | 13:15 |
| Saviq | you need to stop eating but food, people! | 13:16 |
| Saviq | we're in berserker mode! ;) | 13:16 |
| Cimi | Saviq, heh, london | 13:16 |
| Saviq | s/but/bad/ | 13:16 |
| Cimi | heh | 13:16 |
| Saviq | that was bad spelling | 13:16 |
| Cimi | I'll be fine in essn than an hour, just time to digest | 13:16 |
| Cimi | *les | 13:16 |
| Cimi | *less | 13:16 |
| sil2100 | Saviq: does it build correctly with the bump of mhr3's? | 13:17 |
| fginther | sil2100, pong | 13:17 |
| Saviq | sil2100, checking now | 13:17 |
| kgunn | mterry: ping | 13:19 |
| mterry | kgunn, hello | 13:19 |
| kgunn | mterry: hey...sorry to distract you from wizard work :) | 13:19 |
| kgunn | mterry: but i gotta hot bug | 13:20 |
| mterry | kgunn, :) | 13:20 |
| mterry | kgunn, ok, good, throw it over | 13:20 |
| === pete-woods is now known as pete-woods-foodl | ||
| kgunn | mterry: ok...it might require a little chat with gerry...but | 13:20 |
| kgunn | mterry: https://bugs.launchpad.net/qtubuntu/+bug/1226234 | 13:21 |
| ubot5 | Ubuntu bug 1226234 in Unity 8 "QT_LOAD_TESTABILITY=1 does not work for loading the testability driver under mir" [Undecided,Incomplete] | 13:21 |
| mterry | kgunn, huh, OK | 13:22 |
| kgunn | mterry: just sharing from earlier chat with gerry...its probably right we mir should do a better job with unrecognized command line params...but the thot is to "do something quick" in unity-mir | 13:28 |
| kgunn | mterry: in order to get the AP tests working again...its pretty much the last thing stoping mir from being on by default | 13:28 |
| mterry | kgunn, that's bug 1226227 | 13:29 |
| ubot5 | bug 1226227 in Mir "libmirserver parses arguments and fails if it's not something it understands" [Medium,Triaged] https://launchpad.net/bugs/1226227 | 13:29 |
| * kgunn imagines mterry taking his wizard hat off for the day | 13:29 | |
| kgunn | mterry: right... | 13:29 |
| mterry | OK, I'm guessing these are really dups? | 13:30 |
| greyback | mterry: not really. One is that Mir's command line parser quits if a argument it doesn't recognise is passed. Other is that somewhere the QT_LOAD_TESTABILITY is being ignored/lost | 13:31 |
| mterry | Or not and you'd like both fixed :) | 13:31 |
| Saviq | Cimi, mterry standup? | 13:32 |
| mterry | greyback, OK. I just got confused by the conflation of the two, but I guess kgunn was just saying that those were the two blockers | 13:32 |
| greyback | mterry: I suspect QT_LOAD_TESTABILITY is the one AP actually uses, not appending "-testability" to the command line | 13:32 |
| Saviq | greyback, mterry QT_LOAD... is the new thing that should be used | 13:32 |
| Saviq | -testability is legacy | 13:32 |
| mterry | Saviq, thanks for standup warning | 13:33 |
| didrocks | Saviq: is the build finished? | 13:36 |
| didrocks | Saviq: sorry to be pushing but we really need to have this unity8 building for testing in daily-release ppa ;) | 13:37 |
| Saviq | didrocks, 40%, sorry | 13:37 |
| sil2100 | Saviq: does it build? ;) | 13:37 |
| Saviq | didrocks, will know in 10 mins after having tested | 13:37 |
| didrocks | sil2100: I have an idea if you don't mind messing up with the system | 13:37 |
| sil2100 | didrocks: wazzup? | 13:37 |
| * Saviq does the right thing and builds it in sbuild and installs on the device to test | 13:38 | |
| didrocks | sil2100: so taking trunk, merging mhr3's branch | 13:38 |
| didrocks | sil2100: pushing to ~sil2100/unity8/test | 13:38 |
| didrocks | setting cu2d to it | 13:38 |
| didrocks | (deploying) | 13:38 |
| didrocks | and starting the stack | 13:38 |
| didrocks | so that if Saviq +1, we can just bzr push ~sil2100/unity8/test to trunk | 13:38 |
| sil2100 | didrocks: just to see if it builds? Or in preparation for releasing? | 13:38 |
| didrocks | sil2100: prep for releasing | 13:39 |
| sil2100 | didrocks: makes sense! | 13:39 |
| sil2100 | Doing! | 13:39 |
| didrocks | doesn't sound insane? | 13:39 |
| didrocks | thanks sil2100 ;) | 13:39 |
| Saviq | what's ~sil2100/unity8/test? :D | 13:39 |
| didrocks | Saviq: that's the future! | 13:39 |
| didrocks | lp:unity8's future more exactly ;) | 13:39 |
| mhr3 | Saviq, did i trained you to do these things? :) | 13:40 |
| mhr3 | train* | 13:40 |
| greyback | mzanetti: something like this: http://pastebin.ubuntu.com/6162798/ | 13:42 |
| Saviq | nic-doffay, adb shell "find /home/phablet -name unity8-build-deps*deb -exec rm {} +" | 13:42 |
| mterry | Saviq, I got disconnected while you were talking, so probably missed a fair chunk. You might want to update your section in the notes | 13:42 |
| Saviq | nic-doffay, should fix you | 13:42 |
| Saviq | mterry, will do | 13:42 |
| mzanetti | greyback: yay! that's it... probably we could add adding the ddebs repo to apt, but then it's mostly what I thought of | 13:43 |
| greyback | mzanetti: true. Wonder if ppl would like a shared repo where we stick handy scripts | 13:43 |
| mzanetti | greyback: yeah, definitely... we need that | 13:43 |
| Saviq | greyback, btw, ssh-copy-id is probably the correct way of pushing the ssh key over | 13:44 |
| Saviq | greyback, also, s/service/initctl/ | 13:44 |
| greyback | Saviq: I know, but adb is easier sometimes | 13:44 |
| mzanetti | Saviq: it is, but it requires you to enter the passwd | 13:44 |
| mzanetti | adb push doesn't | 13:44 |
| Saviq | mzanetti, which is a good thing ;) | 13:44 |
| mzanetti | Saviq: not in such a script, no | 13:44 |
| nic-doffay | Saviq, cheers giving it a go. | 13:45 |
| Saviq | mzanetti, thing is that script assumes you only have a single ssh key | 13:45 |
| Saviq | ssh-copy-id does not | 13:45 |
| Saviq | anyway | 13:45 |
| mzanetti | thats a valid point | 13:45 |
| Saviq | and in a default location, for that matter | 13:45 |
| greyback | works for me (R) | 13:45 |
| Saviq | ® | 13:46 |
| mzanetti | that guy and his special chars | 13:46 |
| mzanetti | ® | 13:46 |
| mzanetti | oh... my kbd can that too :D | 13:46 |
| mzanetti | still can't do the arrows tho | 13:46 |
| Saviq | ↓↑←→ | 13:47 |
| mzanetti | … | 13:47 |
| Saviq | ;P | 13:47 |
| Saviq | that's like GameBoy Combo | 13:48 |
| Saviq | ↓→→→↑↑←→AB | 13:48 |
| mzanetti | lol | 13:48 |
| mzanetti | Killer Combo! | 13:48 |
| === _salem is now known as salem_ | ||
| Saviq | didrocks, sil2100, I'll just force the branch to go through upstream merger if it's fine with you | 13:54 |
| Saviq | no manual pushing | 13:54 |
| didrocks | Saviq: hum, that will still take an hour | 13:55 |
| Saviq | didrocks, no, 3 mins | 13:55 |
| didrocks | ah, if it's 3, fine | 13:55 |
| didrocks | sil2100: remove your ack then ;) | 13:55 |
| didrocks | sil2100: we will have the branch now ^ | 13:55 |
| sil2100 | :| | 13:55 |
| sil2100 | I ran the stack already :| | 13:55 |
| Saviq | someone's trigger-happy | 13:55 |
| Saviq | sil2100, it's not there in trunk yeyt | 13:56 |
| Saviq | yet | 13:56 |
| sil2100 | didrocks: what to do in this case ;p? | 13:56 |
| Saviq | I'm almost there | 13:56 |
| sil2100 | didrocks: should I kill the stack? Or just let it go? | 13:57 |
| didrocks | sil2100: kill kill kill | 13:57 |
| sil2100 | didrocks: can I kill the prepare job without consequences? | 13:57 |
| Saviq | there will always be consequences! | 13:57 |
| sil2100 | I killed stuff | 13:58 |
| Saviq | sil2100, I'll ping you when ready | 13:58 |
| sil2100 | Let's wait for the merge to go in then, I'll redeploy the stack back... | 13:58 |
| * Saviq wants cross building... | 13:58 | |
| didrocks | sil2100: yes you can! | 13:59 |
| Saviq | mhr3, renderer for dash plugins affects home scope? | 14:00 |
| mterry | greyback, what library listens for QT_LOAD_TESTABILITY? | 14:00 |
| mhr3 | Saviq, yes, it defines them all | 14:00 |
| mhr3 | (cause it has all the master scope definitions) | 14:00 |
| greyback | mterry: Qt does. probably qtbase | 14:01 |
| Saviq | mhr3, so we have a triplet? [ renderer, content_type, renderer_hint ] now? | 14:01 |
| mterry | Hmm, didn't see it there | 14:01 |
| mhr3 | Saviq, we do | 14:02 |
| Saviq | mhr3, and the last one is a freeform string? :D | 14:02 |
| mterry | Ah... in a recent ubuntuX patch, we added it | 14:02 |
| mhr3 | Saviq, yes, hope you're happy :P | 14:03 |
| Saviq | mhr3, no, in fact I'm not - instead of one freeform string we have 2 enums and a freeform string :D | 14:03 |
| Saviq | crazy! ;P | 14:03 |
| mhr3 | Saviq, we have two well defined enums, those should be enough for anything | 14:03 |
| mhr3 | Saviq, but you always want crazy things... like tinting a category white... therefore hint | 14:04 |
| Saviq | but in case they're not ;) lol | 14:04 |
| mhr3 | Saviq, if you want too much we can concatenate the triplet in the plugin and you'll have your one string :P | 14:05 |
| Saviq | rotfl | 14:06 |
| mhr3 | Saviq, but for real - we will want things like "grid in landscape mode" vs portrait, yet it's still grid, i didn't want to extend the types themselves | 14:07 |
| Saviq | mhr3, I understand | 14:07 |
| mhr3 | Saviq, and i also didn't want to extend content into music_from_your_gradma | 14:08 |
| mhr3 | vs music_from_the_store | 14:08 |
| nic-doffay | Saviq, getting this on run_on_device now: http://pastebin.ubuntu.com/6162898/ | 14:10 |
| Saviq | nic-doffay, it didn't build | 14:10 |
| Saviq | nic-doffay, run_on_device -s; run_on_device again | 14:10 |
| === iahmad is now known as iahmad|afk | ||
| Saviq | nic-doffay, and please paste more context next time | 14:10 |
| Saviq | sil2100, DONE | 14:12 |
| sil2100 | \o/ | 14:12 |
| sil2100 | Saviq, didrocks: ran unity8 stack o/ | 14:13 |
| mzanetti | greyback: how do you test the stuff on the device with mir? | 14:15 |
| Saviq | didrocks, sil2100 so... I'm removing the landing ask then... | 14:16 |
| mzanetti | greyback: I can run unity8 only once, then I have to reboot | 14:16 |
| greyback | mzanetti: reboot? Shouldn't be necessary. | 14:16 |
| greyback | what's the error? | 14:16 |
| Saviq | sil2100, actually right, you're tackling that | 14:16 |
| mzanetti | greyback: the old instance doesn't go away from the screen any more and when starting the next one I get resource busy or something like that | 14:17 |
| mterry | greyback, how's https://code.launchpad.net/~mterry/unity8/load-testability/+merge/188064 ? seems hackish, but consistent with existing hack | 14:17 |
| Saviq | sil2100, didrocks, btw what's the difference between "Saucy" and "Head" in http://10.97.0.1:8080/view/cu2d/ ? feature branches? | 14:18 |
| greyback | mzanetti: that means an application is still running. Unfortunately that blocks mir starting. Workaround is to kill all apps | 14:18 |
| kgunn | greyback: sorry to pester - are these instructions still good ? | 14:18 |
| kgunn | https://pastebin.canonical.com/98152/ | 14:18 |
| mzanetti | greyback: oh... I see. yeah, I had an app running. cheers | 14:19 |
| kgunn | i get unknown job unity | 14:19 |
| kgunn | when i try to stop | 14:19 |
| Saviq | kgunn stop unity8 | 14:19 |
| mhr3 | you need UPSTART_SESSION set | 14:19 |
| mhr3 | initctl list-sessions | 14:20 |
| Saviq | kgunn, yeah, and use "sudo -u phablet -i" instead of "su phablet" | 14:20 |
| Saviq | kgunn, or at least "su - phablet" | 14:20 |
| kgunn | thanks guys.... | 14:20 |
| greyback | mterry: well I suppose that'll work, but inside Qt itself, I thought there was code listening for the same? | 14:20 |
| olli | Saviq, kgunn what is the issue with AP not running on u8/mir? | 14:20 |
| mterry | greyback, it's in QApplication | 14:20 |
| Saviq | olli, a few, I'm afraid | 14:20 |
| kgunn | btw...checking is mhr3 was talking to me...looked at scroll back, nearly peed pants laughing at "extend content into music_from_your_gradma" | 14:21 |
| Saviq | olli, bug #1226234 and bug #1226227 that's two | 14:21 |
| ubot5 | bug 1226234 in Unity 8 "QT_LOAD_TESTABILITY=1 does not work for loading the testability driver under mir" [Undecided,Incomplete] https://launchpad.net/bugs/1226234 | 14:21 |
| ubot5 | bug 1226227 in Mir "libmirserver parses arguments and fails if it's not something it understands" [Medium,Triaged] https://launchpad.net/bugs/1226227 | 14:21 |
| mterry | greyback, which sits on top of QGuiApplication | 14:21 |
| Saviq | olli, and one other that we just got informed of this morning | 14:21 |
| mterry | greyback, and unity8 only uses QGuiApplication | 14:21 |
| kgunn | olli: its being worked | 14:21 |
| Saviq | olli, is that we need to expose geometry info from unity8 | 14:21 |
| greyback | mterry: got it | 14:21 |
| sil2100 | Saviq: we've using Saucy for now, once T development opens up it will be in Head | 14:21 |
| sil2100 | Saviq: right now Head doesn't have too many projects in it | 14:22 |
| Saviq | sil2100, yup, thought that's it | 14:22 |
| olli | kgunn, Saviq, that's the current holdup for going default with mir/u8 | 14:22 |
| mzanetti | mterry: greyback: FYI: I patched this into Qt and it seems to get accepted for 5.2 | 14:22 |
| kgunn | olli: just watch mterry & greyback discuss... | 14:22 |
| mzanetti | greyback: mterry: https://codereview.qt-project.org/#change,66513 | 14:22 |
| olli | kgunn, Saviq do you guys have an ETA | 14:22 |
| kgunn | olli: we're trying for today...but honestly it might be monday | 14:22 |
| mterry | mzanetti, oh good. so that's the long-term fix | 14:22 |
| kgunn | olli: no matter what...the best we could do is get it on the ASK sheet | 14:23 |
| kgunn | so no way its in an image before Mon | 14:23 |
| olli | kgunn, ok | 14:23 |
| mterry | kgunn, what's the plan with unity-system-compositor? I'm assuming our first cut at Mir support is without it? Do we still plan it for 13.10? | 14:23 |
| olli | so, there is some frustration with someone unnamed that it comes as a surprise now that we are switching to u8/mir and can't run our tests... | 14:24 |
| kgunn | mterry: so u-s-c would only be needed for split greeter yeah ? | 14:24 |
| olli | kgunn trying to manage that atm ;) | 14:24 |
| kgunn | olli: ack :-// | 14:24 |
| mterry | kgunn, uh, I guess so, if everything works in the user session Mir... I thought there was some concern about wanting root to have access to the devices rather than phablet | 14:26 |
| kgunn | mterry: you got me there...dunno....i do recall that topic, but don't know the priority | 14:26 |
| kgunn | mterry: i'll try to follow up on that | 14:27 |
| mterry | Saviq, mzanetti: heyo! Could I get a review on https://code.launchpad.net/~mterry/unity8/statsWelcomeScreen/+merge/184153 ? | 14:27 |
| kgunn | mterry: for the moment...i'd assume split greeter is ok for deferal (per jamie discussion we had) | 14:27 |
| mterry | If that doesn't land soonish, I should probably file a bug against system-settings to hide that setting | 14:28 |
| greyback | mterry: comment added | 14:28 |
| kgunn | mterry: which would mean u-s-c can lag as well | 14:28 |
| mzanetti | mterry: right... forgot it again :/ | 14:28 |
| mzanetti | sorry | 14:28 |
| mterry | greyback, that's the same code QApplication does | 14:28 |
| mterry | greyback, they don't check value either | 14:28 |
| greyback | mterry: really? Huh, ok | 14:28 |
| greyback | mterry: consider it approved then :) | 14:28 |
| mterry | greyback, (it's our patch on top of Qt, not real Qt code) | 14:29 |
| mterry | kgunn, yeah fair | 14:29 |
| mterry | kgunn, so much code that almost made it ::sad trombone:: | 14:30 |
| mterry | greyback, are you doing the argument parsing bug? | 14:30 |
| kgunn | mterry: ...sure, i just keep saying...day 1 post phone v1...we're back to bid-ness | 14:30 |
| mterry | kgunn, fair enough | 14:30 |
| greyback | mterry: I've not tried it yet, but I can. Fancy trying? You need to override the default argument parser that Mir supplies. so it's a unity-mir thing | 14:34 |
| nic-doffay | Saviq, here's the output for both run_on_device -s and run_on_device | 14:35 |
| nic-doffay | http://pastebin.ubuntu.com/6162984/ | 14:35 |
| nic-doffay | Still having the same issue as before. | 14:35 |
| Saviq | nic-doffay, your compiler died | 14:36 |
| Saviq | nic-doffay, "c++: internal compiler error: Killed (program cc1plus)" | 14:36 |
| Saviq | nic-doffay, if you can, I'd flash the device from scratch - using phablet-flash ubuntu-system --channel devel-proposed | 14:37 |
| Saviq | nic-doffay, and start over | 14:37 |
| === pete-woods-foodl is now known as pete-woods | ||
| Saviq | nic-doffay, be aware you'll lose data this way, if you have any on the phone | 14:37 |
| mterry | greyback, sure, I'm looking at the parser bug | 14:42 |
| Saviq | kgunn, olli, greyback filed https://bugs.launchpad.net/unity-mir/+bug/1232054 for tracking | 14:43 |
| ubot5 | Ubuntu bug 1232054 in unity-mir "Need to expose geometry for autopilot consumption" [Critical,Triaged] | 14:43 |
| olli | thx saviq | 14:44 |
| === alan_g is now known as alan_g|tea | ||
| greyback | Saviq: can you do a quick review pass on: https://code.launchpad.net/~gerboland/unity-mir/use-upstart-app-launch2/+merge/187769 | 14:48 |
| === alan_g|tea is now known as alan_g | ||
| Saviq | greyback, can you add //TODO to screenshot provider that we should use a {previous screenshot, app-provided splash screen, default splash screen} | 14:58 |
| Saviq | greyback, bug #1124265 | 14:58 |
| ubot5 | bug 1124265 in unity8 (Ubuntu) "[Shell, UI Tollkit] white page is displayed before app opens" [Medium,Triaged] https://launchpad.net/bugs/1124265 | 14:58 |
| greyback | Saviq: in that MR? | 14:59 |
| Saviq | greyback, also, isn't it customary to name the method ::instance() and not ::singleton() ? | 14:59 |
| Saviq | greyback, yes | 14:59 |
| Saviq | greyback, you added some comment there | 14:59 |
| Saviq | or at least a LOG() | 14:59 |
| Saviq | but a TODO there would be useful probably | 15:00 |
| Saviq | greyback, does SIGSTOP; upstart_app_launch_stop_application work? i.e. does upstart send SIGKILL, or just SIGTERM | 15:02 |
| Saviq | greyback, also, since we don't support re-connecting to apps yet, should we maybe stop all of them on exit? | 15:04 |
| Saviq | s/stop/kill/ | 15:04 |
| greyback | Saviq: I don't get your question on SIGSTOP | 15:05 |
| Saviq | greyback, we suspend an app | 15:05 |
| Saviq | greyback, then we call upstart_stop_app | 15:05 |
| Saviq | greyback, SIGTERM won't reach the app | 15:05 |
| Saviq | greyback, since it's stopped | 15:05 |
| Saviq | greyback, so does upstart kill it then? | 15:05 |
| greyback | Saviq: ah I see. Yes upstart kills it | 15:06 |
| Saviq | greyback, as in SIGKILL? | 15:06 |
| Saviq | greyback, now that I think of it... should we really remove the app from our stack if it died? | 15:06 |
| Saviq | /crashed? | 15:07 |
| greyback | I don't know how, I just know the app is gone and removed | 15:07 |
| didrocks | Saviq: we got quite some regression on desktop: http://10.97.0.1:8080/job/autopilot-saucy-daily_release/label=qa-nvidia-gtx660/2185/ | 15:07 |
| Saviq | greyback, even if it's SIGSTOP'd first? ;) | 15:07 |
| didrocks | Saviq: but it's possible that it didn't pick the unity new version TBH | 15:07 |
| didrocks | (not sure how sil2100 picked it) | 15:07 |
| Saviq | didrocks, uh oh | 15:07 |
| didrocks | Saviq: you did run the unity8 AP tests on phone, right? | 15:07 |
| sil2100 | What's up? | 15:07 |
| Saviq | didrocks, I just used it for a bit, will run tests now | 15:08 |
| didrocks | Saviq: ok, thanks! | 15:08 |
| greyback | Saviq: a later MR will inform if app crashed or not (upstart-app-lib update will give me that info) | 15:08 |
| sil2100 | uh oh | 15:08 |
| sil2100 | Saviq: 14 AP failures... | 15:08 |
| didrocks | sil2100: I asked paul to run some AP tests, can you look at those issues? ^ | 15:08 |
| sil2100 | Saviq: 14 AP failures for unity8 | 15:08 |
| didrocks | I didn't look at the unity desktop tests | 15:08 |
| greyback | Saviq: but if app crashes, it hasn't saved it's state, so won't restore state. | 15:09 |
| didrocks | sil2100: the unity desktop tests didn't run | 15:09 |
| didrocks | sil2100: can you fix that? | 15:09 |
| didrocks | (because of the ABI change) | 15:09 |
| greyback | Saviq: in which case, I think best to remove, notify user of fail, and allow user to relaunch if they want | 15:09 |
| didrocks | sil2100: I think for unity8, we'll just rely on paul and Saviq's feedback | 15:09 |
| Saviq | greyback, k | 15:09 |
| mterry | kgunn, do we still need to fix the argument parsing if we are paying attention to QT_LOAD_TESTABILITY? | 15:10 |
| kgunn | mterry: hmmm....guess not if you consume it right | 15:11 |
| sil2100 | Ah... eh | 15:11 |
| kgunn | mterry: altho a test with it to ensure nothing else is being sent | 15:11 |
| Saviq | sil2100, did we do "full PPA" btw? | 15:11 |
| Saviq | gobject/gsignal.c:2475: signal 'changeset-started' is invalid for instance '0x8fcb090' of type 'DeeSharedModel' | 15:12 |
| Saviq | suggests we didn't | 15:12 |
| Saviq | although that shouldn't matter | 15:12 |
| mterry | kgunn, didn't quite parse you | 15:12 |
| Saviq | sil2100, didrocks they're all ~autopilot failures | 15:12 |
| didrocks | Saviq: so autopilot is broken? | 15:13 |
| Saviq | sil2100, didrocks i.e. the tests didn't really start even | 15:13 |
| sil2100 | Saviq: I didn't do a full PPA one, but unity, libunity and such were upgraded | 15:13 |
| Saviq | "'NoneType' object has no attribute 'introspection_iface'" | 15:13 |
| kgunn | mterry: just saying, to test the full ap run with the QT_LOAD_TESTABILITY in place....and make sure no other niggly little flag is being flowed thru to crash mir | 15:13 |
| sil2100 | Saviq: how come? Since only 7 per platform failed, while there are around 30 right? | 15:13 |
| didrocks | interested | 15:13 |
| kgunn | mterry: altho...you might be certain there is none | 15:13 |
| Saviq | sil2100, yeah, I know it's weird | 15:13 |
| didrocks | sil2100: while this is running, your phone is free right? can you try latest autopilot and run all tests on it? | 15:13 |
| didrocks | sil2100: the AP regresssion is fixed | 15:14 |
| mhr3 | the dee error is #79 in the landing plan, i have no idea why is it striked through | 15:14 |
| mhr3 | clearly only half of it landed | 15:14 |
| mzanetti | mterry: are you sure that QApplication listens for QT_LOAD_TESTABILITY? | 15:14 |
| greyback | Saviq: if I do s/singleton/instance/ I'll have to touch a few more files. Want it? | 15:14 |
| mzanetti | mhr3: I don't think it does | 15:14 |
| mzanetti | sorry... mterry | 15:14 |
| mterry | mzanetti, we distro-patched it | 15:14 |
| Saviq | greyback, not really, was just asking | 15:14 |
| sil2100 | ACK | 15:15 |
| mzanetti | ah ok | 15:15 |
| sil2100 | Let me upgrade my phone... | 15:15 |
| Saviq | greyback, since I remember we were using ::instance() in other places, but not hung up on it | 15:15 |
| mterry | kgunn, I'm not certain, no | 15:15 |
| greyback | Saviq: ok. Yep I usually use "instance" but someone added singleton in unity-mir, so I did the same | 15:15 |
| greyback | Saviq: todo pushed. The screenshot logic needs more intelligence in shell, as right now it just asks for all to update in one go | 15:16 |
| greyback | I've idea how to solve that | 15:16 |
| Saviq | greyback, sure, I just wanted a TODO, not a fix :) | 15:17 |
| Saviq | greyback, looks good | 15:17 |
| greyback | Saviq: thanks | 15:18 |
| mterry | kgunn, to test the AP tests, we just need to run my branch through phablet-test-run -n unity8, right? | 15:19 |
| sil2100 | Saviq: hmm, so what can we do with unity8 ? | 15:19 |
| mterry | kgunn, I assume the other app tests don't care about our testability? | 15:19 |
| Saviq | sil2100, I'm running ap on the device | 15:20 |
| Saviq | sil2100, but looking at the videos... unity8 never started there... | 15:20 |
| Saviq | or wait, it took long to start | 15:20 |
| Saviq | and timed out maybe | 15:20 |
| Saviq | AHJ | 15:20 |
| Saviq | F*CKING HUD | 15:20 |
| Saviq | sil2100, didrocks, I say hud-service is crashing like crazy, unity8 is waiting for it | 15:21 |
| nic-doffay | Saviq, having issues with setup now after the flash: http://pastebin.ubuntu.com/6163138/ | 15:21 |
| Saviq | sil2100, didrocks, and autopilot times out waiting for it | 15:21 |
| didrocks | Saviq: but hud-services crashes are not new? | 15:21 |
| didrocks | (not due to those changes, right?) | 15:21 |
| Saviq | didrocks, no | 15:21 |
| didrocks | ok | 15:21 |
| Saviq | didrocks, well, they're unrelated simply | 15:22 |
| didrocks | Saviq: just tell me the result by your local run :) | 15:22 |
| Saviq | nic-doffay, where did you get -updates and -security from? | 15:22 |
| Saviq | didrocks, yup, it's coming | 15:22 |
| * mhr3 wonders why hud wasn't disabled completely for now | 15:22 | |
| didrocks | you are using packages from the daily-build ppa, right? | 15:22 |
| greyback | Saviq: and sorry, but need one more code review: https://code.launchpad.net/~gerboland/unity-mir/add-fail-and-resume-focus-support/+merge/188040 | 15:22 |
| sil2100 | Saviq: you think that re-running the tests would help? | 15:22 |
| Saviq | sil2100, doubt it, fixing hud would | 15:23 |
| Saviq | sil2100, or well, fixing our stuff to not wait for hud | 15:23 |
| didrocks | sil2100: we are already running the unity7 ones, right? let's get those and trust on the manual testing for unity8 phone | 15:23 |
| didrocks | we don't really care of unity8 desktop results ;) | 15:23 |
| Saviq | sil2100, didrocks | 15:23 |
| Saviq | Ran 24 tests in 403.408s | 15:23 |
| Saviq | OK | 15:23 |
| didrocks | \o/ | 15:23 |
| didrocks | this is with the daily-build ppa? | 15:23 |
| Saviq | nic-doffay, try again, not sure why you got 404 there | 15:23 |
| Saviq | didrocks, yes, not with the full, though, just unity8 and deps | 15:24 |
| Saviq | didrocks, can do full if you need? | 15:24 |
| didrocks | Saviq: good enough (you got the scopes?) | 15:24 |
| Cimi | Saviq, I'm fine since more than an hour but I'm losing time cause ubuntu doesn't work anymore for me after the upgrade | 15:24 |
| Saviq | prolly not | 15:24 |
| nic-doffay | Saviq, I've even reflashed. | 15:24 |
| Cimi | I'm reinstalling | 15:24 |
| didrocks | one sec, giving you the source list | 15:24 |
| Saviq | nic-doffay, somehow you're getting 404s on the repositories... | 15:24 |
| sil2100 | didrocks: right ;) | 15:24 |
| didrocks | Saviq: http://people.canonical.com/~platform/cu2d/results | 15:24 |
| nic-doffay | Saviq, yeah and before I didn't have them. | 15:24 |
| nic-doffay | No idea why not. | 15:24 |
| didrocks | ==== unity ==== | 15:24 |
| Saviq | nic-doffay, yeah, try again | 15:25 |
| didrocks | Saviq: please install what's under it (which is already on the phone) | 15:25 |
| didrocks | so no compiz, no nux I guess :p | 15:25 |
| Saviq | didrocks, I went for dist-upgrade :P | 15:25 |
| nic-doffay | Saviq, I've reflashed already. | 15:25 |
| sil2100 | didrocks: ouch.. but I see that unity7 tests for intel have problems: | 15:25 |
| Saviq | nic-doffay, just apt-get update on the phone | 15:25 |
| didrocks | Saviq: you are so brave | 15:25 |
| Saviq | nic-doffay, let me try too | 15:25 |
| sil2100 | didrocks: [273121.435980] [drm:i915_hangcheck_elapsed] *ERROR* stuck on render ring | 15:25 |
| didrocks | Saviq: there is a new autopilot, so maybe side-effects | 15:25 |
| nic-doffay | Saviq, I've tried still get 404 | 15:26 |
| didrocks | sil2100: so, nvidia will be our only hope? ;) | 15:26 |
| didrocks | sil2100: nvidia is running? | 15:26 |
| Saviq | nic-doffay, well, it's not *your* problem I imagine... some mirror is tricking you | 15:27 |
| Saviq | nic-doffay, you in the office maybe? | 15:27 |
| sil2100 | didrocks: yea... ;p | 15:27 |
| * didrocks crosses fingers then | 15:27 | |
| didrocks | sil2100: I hope that running autopilot on the phone with latest image will be fine as well | 15:27 |
| didrocks | so that we can get that one off | 15:27 |
| didrocks | autopilot + unity7 + unity8 | 15:27 |
| didrocks | will be a nice week-end! | 15:28 |
| * sil2100 hopes so as well | 15:28 | |
| nic-doffay | Saviq, that I am. | 15:30 |
| Saviq | nic-doffay, I expect there's a mirror that's playing tricks with you ;) | 15:30 |
| Saviq | nic-doffay, it's going to work for you at home | 15:30 |
| Saviq | nic-doffay, -security and -updates are pretty new things still | 15:31 |
| Saviq | nic-doffay, BUT | 15:31 |
| Saviq | nic-doffay, you still didn't make your device writable | 15:31 |
| Saviq | nic-doffay, and that's your bigger problem | 15:31 |
| Saviq | nic-doffay, adb shell touch /userdata/.writable_image; adb shell reboot | 15:32 |
| nic-doffay | Saviq, I doubt it since I don't have a connection there yet! | 15:34 |
| Saviq | nic-doffay, lol | 15:34 |
| Saviq | nic-doffay, so yeah, ↑↑ will help you regardless of the 404s | 15:35 |
| Saviq | greyback, https://code.launchpad.net/~gerboland/unity-mir/add-fail-and-resume-focus-support/+merge/188040/comments/429736 | 15:37 |
| mzanetti | Saviq: do we anything special in mir to enable the edges while there is an app focused? | 15:38 |
| Saviq | didrocks, sil2100 ok so... got failures on phone after dist-upgrade from daily-build :/ | 15:38 |
| Saviq | mzanetti, InputArea | 15:38 |
| Saviq | didrocks, sil2100 exact same failures | 15:38 |
| didrocks | Saviq: argh… | 15:38 |
| didrocks | mhr3: help ^ | 15:39 |
| didrocks | Saviq: downgrade autopilot first | 15:39 |
| didrocks | and please rerun ;) | 15:39 |
| Saviq | didrocks, stuff that got upgraded: http://pastebin.ubuntu.com/6163219/ | 15:39 |
| sil2100 | Saviq: :< | 15:39 |
| didrocks | sil2100: please rollback the 2 autopilot packages | 15:39 |
| sil2100 | Right, might be AP's fault | 15:39 |
| sil2100 | didrocks: should I remove them from daily-build? | 15:40 |
| mhr3 | Saviq, what failures? | 15:40 |
| Saviq | mhr3, http://10.97.0.1:8080/job/autopilot-saucy-daily_release/2185/ | 15:40 |
| didrocks | Saviq: please do | 15:40 |
| didrocks | sil2100: ^ | 15:40 |
| didrocks | sil2100: and remove autopilot from the ppa | 15:41 |
| sil2100 | ;) | 15:41 |
| Saviq | didrocks, sil2100 will know pretty soon | 15:41 |
| didrocks | Saviq: I cross fingers *very hard* | 15:41 |
| mhr3 | Saviq, is that supposed to be completely clean normally? | 15:42 |
| Saviq | mhr3, as in 0 failures? yes | 15:42 |
| mhr3 | uh | 15:42 |
| mhr3 | sounds like something is very much broken | 15:42 |
| sil2100 | didrocks: btw. we not releasing autopilot then today? | 15:42 |
| mhr3 | and mostly coming for ap itself | 15:43 |
| mhr3 | missing position attrs | 15:43 |
| * Saviq thought we've failed to switch to home scope on startup, but in fact we've just started showing videos and music above apps | 15:43 | |
| mhr3 | yea, which is a bug... but not critical :P | 15:44 |
| didrocks | sil2100: no, please poke thomi with the same bug | 15:44 |
| Saviq | mhr3, depends | 15:44 |
| Saviq | mhr3, since our ap tests assume apps are on top in home ;) | 15:44 |
| mhr3 | eh :/ | 15:44 |
| mhr3 | fwiw pstolowski has a branch that fixes it | 15:45 |
| Saviq | mhr3, but not the cause here | 15:45 |
| mhr3 | i just wanted tests for it | 15:45 |
| mhr3 | and it's taking a while to write some | 15:45 |
| sil2100 | Man, exciting Friday releases \o/ | 15:45 |
| mhr3 | anyway guys, i have a train to catch | 15:45 |
| didrocks | sil2100: it is! | 15:46 |
| didrocks | Saviq: give us good news ;) | 15:46 |
| didrocks | it's all on you ;) | 15:46 |
| sil2100 | :< | 15:47 |
| didrocks | sil2100: can you just install unity7 on your desktop and dogfood a little bit? | 15:47 |
| didrocks | sil2100: as we won't be able to trust I guess AP test results | 15:47 |
| sil2100 | I already upgraded before, just need to switch to guest | 15:48 |
| Saviq | didrocks, sil2100, the videos are so weird, unity8 either just started or just finished... | 15:48 |
| didrocks | Saviq: TBH, I just trust your manual testing at this point :) | 15:49 |
| didrocks | sil2100: see you? (or not :p) | 15:49 |
| Saviq | didrocks, sil2100 PASS | 15:51 |
| didrocks | \o/ | 15:52 |
| * didrocks hugs Saviq | 15:52 | |
| * didrocks hugs sil2100 (only if unity7 is fine in his guest) | 15:52 | |
| didrocks | so autopilot is really broken | 15:52 |
| didrocks | thomi: we need you running autopilot on all apps and unity8, it seems trunk is really broken atm ^ | 15:53 |
| sil2100 | :D | 15:54 |
| didrocks | sil2100: you're back, it looks good then? ;) | 15:55 |
| sil2100 | Looks fine, but just give me 5 more minutes to test it on my main session | 15:55 |
| didrocks | sure | 15:55 |
| didrocks | sil2100: meeting in 5 minutes btw | 15:55 |
| sil2100 | brb | 15:55 |
| sil2100 | Ok, stupid me ;/ | 15:55 |
| sil2100 | What the f**k am I using screen for if I wanted to shut down IRC when restarting sessions | 15:56 |
| didrocks | ahah | 15:56 |
| kgunn | mterry: i got totally distracted with mir team...did you get your answer on running AP test setup? | 15:59 |
| kgunn | mterry: i got totally distracted with mir team...did you get your answer on running AP test setup? | 15:59 |
| mterry | kgunn, I know how to run AP tests, just wanted to confirm which tests are failing on Mir. Is it just unity8 or unity8 + apps? | 16:00 |
| kgunn | mterry: yes...its unity8 on mir | 16:00 |
| Saviq | mterry, everything | 16:03 |
| Saviq | mterry, see "Unity, autopilot, UPA and Mir" in unitynextuiteam@lists | 16:04 |
| Saviq | sil2100, didrocks sorry it took so long | 16:05 |
| seb128 | does anyone know how to make an app using the dark theme with our toolkit? | 16:07 |
| seb128 | e.g like https://wiki.ubuntu.com/Appearance?action=AttachFile&do=get&target=phone-background.mockup.png | 16:07 |
| sil2100 | Ok, all seems to look alright | 16:09 |
| didrocks | Saviq: no worry, thanks for the feedback? | 16:09 |
| didrocks | sil2100: coming to the hangout? | 16:09 |
| didrocks | sil2100: or can you publish? | 16:09 |
| didrocks | sil2100: maybe reviewing the packaging change | 16:09 |
| didrocks | but please push | 16:09 |
| sil2100 | Dogfooding done, ok... so... | 16:10 |
| sil2100 | didrocks: let me check what components in the unity stack to release | 16:10 |
| sil2100 | didrocks: should I also publish mediascaner etc. as in landing asks? | 16:10 |
| Saviq | seb128, http://bazaar.launchpad.net/~unity-team/unity8/trunk/view/head:/Shell.qml#L75 | 16:11 |
| Saviq | seb128, only you want "SuruDark" | 16:11 |
| seb128 | Saviq: thanks | 16:11 |
| Saviq | seb128, any update on being able to run ubuntu-system-settings without having to build a .deb / install? :) | 16:12 |
| seb128 | Saviq: Satoris is looking at it as part of https://code.launchpad.net/~jpakkane/ubuntu-system-settings/cmake/+merge/184123 | 16:12 |
| Saviq | ah good, transitioning to cmake :D | 16:12 |
| seb128 | "good" | 16:12 |
| seb128 | I kind of like qmake :/ | 16:13 |
| seb128 | but at the end if somebody wants to do the work and maintain the build system, good for them | 16:13 |
| Saviq | it's simple... too simple... | 16:13 |
| seb128 | yeah, the usual | 16:13 |
| seb128 | it makes simple things really easy | 16:13 |
| * Saviq just found unity-lens-people-build-debs... | 16:13 | |
| nic-doffay | Saviq, more issues I can't identify now: http://pastebin.ubuntu.com/6163355/ | 16:15 |
| Saviq | nic-doffay, jeez dude ;) | 16:16 |
| Saviq | nic-doffay, you got it bad | 16:16 |
| Saviq | nic-doffay, go into the device, remove builddir in ~phablet/shell | 16:16 |
| Saviq | nic-doffay, try again (no -s needed) | 16:16 |
| sil2100 | didrocks: quick ACK's http://10.97.0.1:8080/view/cu2d/view/Saucy/view/Unity8/job/cu2d-unity8-saucy-3.0publish/lastSuccessfulBuild/artifact/packaging_changes_unity8_7.81.3+13.10.20130927.3-0ubuntu1.diff and http://10.97.0.1:8080/job/cu2d-unity-saucy-3.0publish/lastSuccessfulBuild/artifact/packaging_changes_unity_7.1.1+13.10.20130927.1-0ubuntu1.diff | 16:16 |
| nic-doffay | Saviq, haha apparently. | 16:17 |
| sil2100 | didrocks: both look ok, there are all the known changes inside | 16:17 |
| didrocks | sil2100: no .diff? | 16:17 |
| didrocks | (on the machine?) | 16:17 |
| sil2100 | didrocks: what do you mean? | 16:18 |
| didrocks | sil2100: as publish didn't run | 16:18 |
| didrocks | because -check didn't pass | 16:18 |
| sil2100 | didrocks: I did them ;) | 16:18 |
| didrocks | we don't know which diff | 16:18 |
| sil2100 | didrocks: the foo + auto_publish | 16:18 |
| sil2100 | Trick | 16:18 |
| sil2100 | ;) | 16:18 |
| sil2100 | didrocks: so these are the latest diffs for things that will be released | 16:19 |
| didrocks | oh good! | 16:19 |
| didrocks | :) | 16:19 |
| mterry | Saviq, I can't see who sets QT_LOAD_TESTABILITY or passes -testability to unity8's tests; autopilot-qt doesn't seem to? | 16:23 |
| Saviq | mterry, there's nothing generic that does that | 16:23 |
| Saviq | mterry, for unity8 it's our tests that do | 16:24 |
| Saviq | mterry, for others, probably too | 16:24 |
| mterry | Saviq, I didn't see it in the code nor phablet-test-run | 16:24 |
| Saviq | mterry, hmm then autopilot-qt must, /me looks | 16:25 |
| Saviq | hmm hmm hmm | 16:26 |
| Saviq | mterry, ah! | 16:27 |
| Saviq | mterry, autopilot itself | 16:27 |
| Saviq | mterry, autopilot/introspection/qt | 16:27 |
| Saviq | .py | 16:27 |
| mterry | oh weird, ok | 16:27 |
| mterry | Saviq, I assumed the existence of autopilot-qt meant there wasn't qt code in autopilot :) | 16:27 |
| Saviq | mterry, autopilot-qt is Qt's support for autopilot, not the other way around :) | 16:28 |
| Saviq | mterry, the qt driver is built into autopilot | 16:28 |
| === jhodapp|afk is now known as jhodapp|lunch | ||
| mterry | Saviq, gotcha | 16:28 |
| mterry | greyback, well, looks like the autopilot tests pass other non-Mir arguments to unity8, so we need to fix this unrecognized-arguments bug. But as my string of comments in that bug indicate, looks like its easiest to just change Mir itself, though alan_g didn't like that idea | 16:34 |
| alan_g | mterry: I'm all for changing Mir as my comment in that thread indicates | 16:34 |
| kgunn | mterry: i think it was to ignore it that he didn't like | 16:35 |
| mterry | alan_g, your proposal was to let the app strip arguments from the list? | 16:35 |
| mterry | alan_g, that would be easier if Mir prepended --mir- to its options | 16:36 |
| alan_g | mterry: My proposal was to separate the construction of the options from the construction of DefaultServerConfiguration | 16:36 |
| kgunn | alan_g as you're about to eod, and this one is kinda hot...could racarr tackle this, this afternoon ? | 16:36 |
| Saviq | <facepalm> | 16:37 |
| * Saviq just started fighting a bad pixel on the tv... just to discover that it was a charger LED reflection when he started moving... | 16:38 | |
| kgunn | :)) where physics and sw meet | 16:38 |
| alan_g | kgunn: if racarr is happy with that. | 16:39 |
| kgunn | alan_g: i have no concept of scope on effort...there...is it even doable in a couple of hours | 16:39 |
| alan_g | kgunn: I've not looked closely - but between hours and a day(ish) | 16:41 |
| nic-doffay | Saviq, finally have results on the phone! | 16:41 |
| kgunn | alan_g: ug...could we do a quick and dirty to ignore unknown args , with full intention of properly fixing (...or is that really really bad...willing to take advice) | 16:41 |
| mterry | greyback, what if we had unity-mir create QGuiApplication before it created the ShellServerConfiguration? Wouldn't that strip the argv of "confusing" arguments? | 16:42 |
| kgunn | mterry: to make sure i understand...this is truly needed even with the QT_LOAD_TESTABILITY in place ?? | 16:43 |
| kgunn | e.g. it is still hot? | 16:43 |
| mterry | kgunn, well, we don't just use QT_LOAD_TESTABILITY. We still pass -testability (which could be fixed, but other Qt arguments like -frameless are also passed) | 16:44 |
| kgunn | Saviq: from earlier, mhr3 said i needed UPSTART_SESSION set to initctl list-sessions, am i correct in thinking that means "export UPSTART_SESSION=list-sessions"? | 16:44 |
| greyback | mterry: that would require a bit of restructuring. Instead of mir being main thread and qapp child thread, you'd switch it to the other way around. It's possible, but not trivial | 16:46 |
| mterry | greyback, other option is restructuring Mir around a bit, which I'm worried would mean another ABI bump or similar. alan_g, I'm still not entirely clear on your proposal. You say decouple options and config? | 16:48 |
| alan_g | kgunn: I think it is possible - would also be possible to add argc & argv to the virtual parse_options() call which would give the client a chance to control things | 16:48 |
| alan_g | mterry: yes it would mean an ABI bump | 16:48 |
| mterry | alan_g, what about ignoring unknown args just as a short term solution? | 16:49 |
| greyback | mterry: other option is in unity-mir:/src/unity-mir/shellserverconfiguration.h - note it inherits DefaultServerConfiguration, which has a "parse_arguments" method. Try overriding that | 16:50 |
| alan_g | mterry: I think it is possible | 16:50 |
| alan_g | greyback: but there's no access to argc and argv - it is misguided | 16:50 |
| mterry | greyback, yeah, I left rambling comments on the bug about the various ways we can't do it via Mir as is | 16:51 |
| greyback | ah ok, that was my first idea | 16:51 |
| mterry | greyback, there are a couple options, but each of them ultimately needs access to some private variable :) | 16:52 |
| === jhodapp|lunch is now known as jhodapp | ||
| alan_g | mterry: It is ugly, but you could save argc and argv for yourself. | 16:58 |
| mterry | alan_g, you mean as a subclasser of DefaultConfiguration? Well, but then we don't have access to program_options | 16:59 |
| mterry | So we can't emulate the parsing that Mir does | 17:00 |
| didrocks | Saviq: sil2100: thanks for everything :) | 17:00 |
| didrocks | (all in and waiting) | 17:00 |
| sil2100 | ;p | 17:00 |
| alan_g | mterry: but you can override parse_options() | 17:00 |
| sil2100 | didrocks: np! Sorry it was so bumpy! Today wasn't my best day | 17:00 |
| didrocks | heh, don't worry! | 17:00 |
| mterry | alan_g, but that takes a ProgramOption object, which we can't control | 17:01 |
| alan_g | mterry: but you *can* do the parsing any way you like. | 17:02 |
| alan_g | Including allowing unrecognised options | 17:02 |
| mterry | alan_g, but rest of Mir looks at that ProgramOption object, and we can't fill it with the results of our parse | 17:02 |
| mterry | alan_g, because its variables_map is private | 17:02 |
| alan_g | Sorry, have to go | 17:03 |
| mterry | alan_g, bye | 17:03 |
| alan_g | mterry: but that's what the default does, so why can't you? | 17:03 |
| === alan_g is now known as alan_g|EOD | ||
| mterry | alan_g|EOD, I'll discuss with racarr I guess. Have a good day | 17:04 |
| mterry | racarr, poke! | 17:04 |
| mterry | alan_g|EOD, (though if you look back at this log later, the answer to your question is because we only have access to program_options in parse_options which assumes we use ProgramOption. If we roll our own ProgramOption, we need to build program_options ourselves to get access to it. But we can't do that because add_option() isn't virtual | 17:06 |
| mterry | alan_g|EOD, though... maybe we could let the default build its ProgramOption and translate that into our own... Super gross though | 17:06 |
| mterry | racarr, you around? | 17:38 |
| Saviq | kgunn, no, you need to "su - phablet" or "sudo -u phablet -i" for the environment to be set up | 17:43 |
| Saviq | kgunn just "su phablet" won't do | 17:43 |
| mterry | kgunn, Saviq: where might phablet-test-run logs go to? I'm testing my allow-unrecognized-options change, and I'm still not seeing unity8 come up | 18:10 |
| Saviq | mterry, -v will give you verbose logs | 18:13 |
| Saviq | mterry, otherwise it (autopilot) doesn't store them anywhere | 18:13 |
| mterry | yeah, I don't get any output though. Like no output from unity8 crashing or whatever | 18:13 |
| mterry | hm | 18:13 |
| Saviq | mterry, can't you just try running unity8 -testability? | 18:13 |
| mterry | Saviq, that's fixed, yeah | 18:14 |
| Saviq | or -whatever, for that matter? | 18:14 |
| mterry | Saviq, I tested my change itself. But it doesn't seem to let autopilot get further | 18:14 |
| mterry | or at least, autopilot chokes on somethingn else now | 18:14 |
| Saviq | mterry, ok, so now you might be hitting bug #1201116 | 18:14 |
| ubot5 | bug 1201116 in Unity 8 "[dash] constant increase in memory consumption when navigating through lenses" [Critical,Incomplete] https://launchpad.net/bugs/1201116 | 18:14 |
| Saviq | wrong | 18:14 |
| Saviq | bug #1232054 | 18:14 |
| ubot5 | bug 1232054 in unity-mir "[mir] Need to expose geometry for autopilot consumption" [Critical,Triaged] https://launchpad.net/bugs/1232054 | 18:15 |
| mterry | that makes more sense | 18:15 |
| Saviq | mterry, again, see the thread in unitynextuiteam@lists.c.c | 18:15 |
| mterry | Saviq, ah, at c.c! I looked for that list earlier and couldn't find it. Figured I was being dumb | 18:15 |
| Saviq | mterry, you should be subscribed to it | 18:16 |
| Saviq | mterry, if you're not, we need to fix that | 18:16 |
| mterry | Saviq, I don't think I am | 18:16 |
| Saviq | mterry, I'll forward the thread to you, as I don't think it's possible to access the archives... | 18:16 |
| mterry | Saviq, ok, thanks | 18:17 |
| Saviq | mterry, and subscribe to https://lists.canonical.com/mailman/private/unitynextuiteam/ in the mean time | 18:17 |
| Saviq | mterry, someone will probably have to accept you | 18:17 |
| Saviq | mterry, ok, you've got mail | 18:18 |
| Saviq | mterry, probably with some duplicates, too... sorry for that | 18:18 |
| mterry | Saviq, ick, gmail doesn't handle that gracefully | 18:20 |
| mterry | kgunn, can you add me to that list? There isn't a subscribe button there | 18:20 |
| mterry | racarr, https://code.launchpad.net/~mterry/mir/unregistered-options/+merge/188125 when you get back | 18:20 |
| Saviq | mterry, if you go to https://lists.canonical.com/mailman/listinfo/unitynextuiteam | 18:21 |
| mterry | kgunn, ^ that fixes the unregistered option thing, but doesn't unblock autopilot by itself | 18:21 |
| Saviq | mterry, there's no "Subscribe" button in the middle !? | 18:21 |
| mterry | Saviq, no, it's just a username/password "let me in" button to see the archive | 18:22 |
| mterry | Saviq, listinfo | 18:22 |
| mterry | Saviq, ah. I didn't edit the url you gave me before | 18:23 |
| Saviq | mterry, sorry, wrong link | 18:23 |
| Saviq | ah, I got the archive! | 18:23 |
| Saviq | for some reason the password I had there was weird | 18:23 |
| mterry | Saviq, yeah, it's odd they don't use openid for that ye | 18:23 |
| mterry | t | 18:23 |
| Saviq | mterry, indeed | 18:23 |
| Saviq | mterry, it's worse - password reminder sends the password to you in plaintext | 18:24 |
| Saviq | which is wrong on oh so many levels | 18:25 |
| mterry | Saviq, :) a good reminder to change it though! | 18:25 |
| * Saviq just put in a random 16-character pass | 18:26 | |
| Saviq | yay password hasher | 18:26 |
| mterry | Saviq, kgunn: seems like racarr isn't around either. Is there another Mir person who would be qualified to approve the unrecognized-options branch? | 18:30 |
| mterry | I suppose any of them *could*, but someone who cares about that bit of code | 18:31 |
| Saviq | mterry, the kiwis and other kangaroos will be around over our weekend | 18:32 |
| Saviq | mterry, so we should be fine Monday morning | 18:32 |
| mterry | Saviq, fair enough | 18:32 |
| * mterry goes back to greeter bugs | 18:33 | |
| kgunn | mterry: racarr is prob at lunch....or there are the kiwis/kangaroos | 18:36 |
| kgunn | Saviq: greyback ...just fyi, i was able to run the QML_RENDERER_TIMING, but at least on SF the osk goes all wonky (sometimes renders but most of the time not...and flickers) | 18:40 |
| kgunn | just sharing.... | 18:40 |
| greyback | kgunn: just with QML_RENDERER_TIMING? | 18:41 |
| * greyback EOD | 18:43 | |
| === jhodapp is now known as jhodapp|afk | ||
| Saviq | has unity7 stopped switching workspaces for anyone else, when clicking on the launcher? | 19:44 |
| Saviq | it changes focus, but doesn't switch to the workspace the app is in | 19:44 |
| Saviq | bregma, you'd know about this ↑ | 19:45 |
| Saviq | and /me reboots in the mean time to see if it helps | 19:45 |
| bschaefer | Saviq, its a know problem and fixed in trunk :) | 19:46 |
| bschaefer | known* | 19:46 |
| Saviq | bschaefer, oh cool, thanks | 19:50 |
| bschaefer | np, hope a release comes out soon for compiz ... | 19:50 |
| Saviq | bschaefer, you were next on my queue of people to ping about this :) | 19:50 |
| ChrisTownsend | Saviq: Do an upgrade now and it'll be fixed with the new compiz package. | 19:50 |
| bschaefer | Saviq, :) | 19:50 |
| bschaefer | ChrisTownsend, o it was released? | 19:50 |
| ChrisTownsend | bschaefer: Yep, just a bit ago. | 19:51 |
| bschaefer | sweeet | 19:51 |
| Saviq | ChrisTownsend, yeah, upgraded just before reboot | 19:51 |
| Saviq | so have the fixed one now :) | 19:51 |
| bschaefer | well then, that solves that problem | 19:51 |
| ChrisTownsend | Saviq: Cool | 19:51 |
| * ChrisTownsend Looks forward to not getting any more of those bugs | 19:52 | |
| * bschaefer agrees! | 19:53 | |
| === salem_ is now known as _salem | ||
| Saviq | ChrisTownsend, bschaefer how about bug #969039 ? ;D | 19:56 |
| ubot5 | bug 969039 in compiz (Ubuntu) "HUD appears when tapping Alt+Left very quickly" [High,Confirmed] https://launchpad.net/bugs/969039 | 19:57 |
| bschaefer | Saviq, yeah, its a very annoying problem :(, been around for a while | 19:57 |
| bschaefer | Saviq, its been a while since i've looked into it though | 19:57 |
| kgunn | mterry: ping | 20:04 |
| kgunn | mterry: you there ? | 20:47 |
| === jhodapp|afk is now known as jhodapp | ||
| mterry | kgunn, am back | 20:56 |
| mterry | kgunn, sorry, ran an errand | 20:56 |
| kgunn | mterry: no problemo | 20:58 |
| kgunn | i hate to ask this...at your 5oclock | 20:58 |
| kgunn | could you retarget your mp ? | 20:58 |
| mterry | kgunn, sure. We have a 13.10 trunk? | 20:58 |
| kgunn | https://code.launchpad.net/~mir-team/mir/development-branch | 20:58 |
| kgunn | mterry: we're using ^ that as our staging | 20:58 |
| kgunn | so we don't break stuff with our api shennanigans | 20:59 |
| kgunn | but...if you retarget | 20:59 |
| kgunn | maybe racarr or kdub could review quickly (cough cough) | 20:59 |
| kgunn | then we could get that merged for the aussie/kiwi crew... | 20:59 |
| mterry | kgunn, will do | 20:59 |
| kdub | review what? | 20:59 |
| kgunn | then robert a can merge to trunk, update ask sheet...yadda yadda | 20:59 |
| mterry | kgunn, kdub: https://code.launchpad.net/~mterry/mir/dev-unregistered-options/+merge/188168 | 21:03 |
| mterry | This is not very user friendly (no warning or errors about unused options). But it fixes the problem for now | 21:03 |
| kgunn | kdub: just so you know...its so the server doeesn't freak out when it gets args it doesn't understand | 21:03 |
| kgunn | basically...its blocking mir from being turned on by default | 21:04 |
| mterry | one of the things :) | 21:09 |
| mterry | bbiab | 21:15 |
| kgunn | mterry: dang you ...yes...1 of the things blocking mir | 21:28 |
| kgunn | :) | 21:28 |
| === jhodapp is now known as jhodapp|afk | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
