=== jono is now known as Guest96931 === _salem is now known as salem_ [07:37] Wellark, hey, could we get a review of https://code.launchpad.net/~mterry/indicator-network/watch-nm-dbus/+merge/219705 and https://code.launchpad.net/~mterry/indicator-network/greeter-profiles/+merge/219897 please? === zequence_ is now known as zequence [08:07] Saviq: 7.85+14.10.20140428.2-0ubuntu1 ? [08:07] it is back [08:08] tsdgeos, yeah, that one we never properly removed [08:08] tsdgeos, and we probably don't care enough [08:08] ok [08:08] tsdgeos, it got added by the train [08:08] tsdgeos, some hiccup there === salem_ is now known as _salem [09:02] mhr3: ping [09:03] tsdgeos, pong === vrruiz_ is now known as rvr [09:04] mhr3: mzanetti is complaining in the unity-api thing why Scopes::get returns a QVariant and not a ScopeInterface* while MockPreviewStack::get returns a PreviewModelInterface* and not a QVariant [09:04] mhr3: what do you think of changing the first get to return a ScopeInterface* ? [09:04] i just mimic''ed what we had in the old code [09:04] but maybe this is better? [09:07] tsdgeos, i think there's a more fundamental issue - what should the get() return [09:07] in some cases it's return a dict of all row properties [09:07] in others it's returning one specific thing [09:08] fwiw, all the other unity-api's return a non-variant-wrapped pointer to a single object [09:08] on get() that is [09:10] mhr3: well i think it should return what it is, so if it's going to be a QVariantMap, write QVariantMap instead of QVariant, if it's a ptr, the same, return the ptr instead of a QVariant [09:10] Saviq: your opinion ↑↑↑↑ ? [09:12] tsdgeos, mhr3 has a point - 'get' is not an established thing - sometimes we return an object "behind" the model, sometimes the QVariantMap [09:12] tsdgeos, yea, sure, the question remains whether those get() methods should always return a QVariantMap [09:12] tsdgeos, imo best would be get() -> QVariantMap and anything not returning a QVariantMap should be getXyz() [09:12] mhr3, I think yes, and if we want the object behind it, we should use getObject, getRaw or something (that's what's in notifications) [09:12] mhr3, violent agreement then [09:13] indeed [09:13] Saviq: mhr3: ok, i'll make get() return a QVariantMap and the ptr be getScope, getPreviewModel, etc [09:14] tsdgeos, cool, thx [09:14] tsdgeos, mhr3, do we need the get() at all? [09:15] just came to my mind :) [09:15] we should avoid it with fire [09:15] Saviq: i do not know, i'm just copying code around [09:15] i think shell indeed it in one case [09:15] * Saviq greps [09:15] but who knows maybe it doesn't anymore [09:15] this is starting to look like those Qt code reviews where people ask me to fix some other bugs unrelated to the one i'm fixing :D [09:16] tsdgeos, no no, I thought you were renaming [09:16] tsdgeos, i guess that's a good thing? :) [09:16] tsdgeos, quality++ ;) [09:16] mhr3: in general is a bit frustrating tbh [09:16] tsdgeos, *but*, unity-api is meant to sanitize after all... [09:17] sure [09:18] > i think shell indeed it in one case [09:18] wow [09:18] i use indeed to much :) [09:18] was meant to be needed :) [09:18] mhr3, we get it on activate [09:19] mhr3, to pass .result [09:19] right, so that's the one use-case [09:19] we could at least get rid of all the other get()s [09:19] Saviq: but we can get it somehow else [09:20] mhr3: you mean the getScope? [09:20] tsdgeos, i meant we could get rid of the other QVariantMap get(), if there are any [09:21] no, there's none else [09:21] otherwise ScopeInterface* get() -> ScopeInterface* getScope() [09:22] ok [09:22] * tsdgeos changes [09:24] but yea, now we'll need to land unity-api, -scopes-shell and u8 in the same go [09:24] since we're renaming [09:26] Cimi, you broke video carousel: http://imgur.com/p0oiNEo [09:28] mhr3: you can still land unity-api first [09:29] since it's "unused" [09:29] and then land scopes-shell and u8 in one go [09:29] right [09:31] tsdgeos: Saviq: so far the story sounds like that if we want to switch to 5.3 it has to be ~now and not any lingering later plan. hopefully once unity8 starts working the functionality would look good, but when I for example was not able to launch system settings from indicator last week when I tried and I'm not sure why (in addition to scopes problem) [09:32] Saviq, let's revert it [09:32] Cimi, why can't it adapt to the card size? [09:34] Mirv, what scopes problem? [09:34] Saviq, I did it externally in cardCarousel [09:35] Cimi, yeah, and? it knows the card size of course? [09:35] Mirv: define "now" [09:35] Saviq, you remember my fear of the anchors fill? [09:35] mhr3: broken with Qt 5.3, since it relies on the delegate range patch that was not accepted by upstream. also the new unity8 branch somehow just crashes on 5.3 even though it works with the 5.3-way-of-doing-things-backported-to-5.2-qtdeclarative [09:35] Saviq, could be that issue [09:35] mhr3, ours [09:35] Saviq, I am not anchoring to the artShape [09:35] I am just filling the card [09:35] tsdgeos: 2-3 weeks probably [09:35] we thought filling card was enough [09:36] Mirv: even without the .0 out? [09:36] Mirv, right, thx [09:36] Saviq, see that now :) [09:36] not in case of video apparently [09:36] tsdgeos: well we can work on 5.3 rc for now and .0 should be out this week (or at latest next if it gets delayed) [09:36] Cimi, and? the carousel knows about the size of the card, there's no reason why it shouldn't work (with a borderimage on the shadow) [09:37] tsdgeos: it's just that the schedule is such that we shouldn't do major things like this anymore later in June for example [09:37] anyhow, more discussion on the subject will happen on Wed [09:37] Saviq, Cimi, i'd also say the height of the video carousel is weird... maybe fixing that would magically fix this ;) [09:38] mhr3, https://docs.google.com/a/canonical.com/file/d/0B8T-xKCA6bikUm56SG1pVEktZUk/edit [09:38] mhr3, and anyway, kinda a mediascanner scope thing ;) [09:39] Saviq, no, i mean there's loads of empty space beneath the carousel [09:39] the visual design looks ok [09:39] the real thing is empty [09:40] mhr3, in any case it's just the padding [09:44] mhr3, but agreed it seems to think it's bigger than it really is [09:46] it's scary when the tech lead says that the qml components think :) [09:48] although with all that code generation, who knows how sentient u8 is now :) [09:50] Saviq: should I be able to run unity8 on my desktop using 'start unity8' or has something changed? [09:51] veebers, should work, make sure you don't have some stale job .conf in ~/.config/upstart [09:52] veebers, anything in ~/.cache/upstart/unity8.log? [09:52] veebers, and what does `initctl status unity8` say? [09:52] Saviq: ah ok, will check and try (this is from ~last week) [09:55] Saviq: `initctl status unity8` gives (before starting): unity8 stop/waiting, `initctl start unity8` hangs for ages and I accidently just Ctrl-C it :-\ but trying to start it again states ..'job is already running..' [09:56] and now `stop unity8` is hanging for ages :-\ will ctrl-c soon [09:56] veebers, and that's current unity8? trusty or utopic? [09:57] Saviq: Hmm, trusty and should be up-to-date [09:57] veebers, right, we should SRU this [09:58] veebers, http://bazaar.launchpad.net/~unity-team/unity8/trunk/revision/866/data/unity8.conf [09:58] fyi: while stop was hanging status output is "unity8 stop/pre-start, process 13935" [09:58] * veebers looks [09:59] veebers, you need to kill any initctl still hanging, make that ↑ change to the unity8.job (you can just put current trunk's data/unity8.conf in ~/.config/upstart) and all should be good again [09:59] veebers, and you should move to utopic already ;) [10:00] Saviq: heh right, I intend to upgrade after the sprint (knowing my luck a fresh install just before a sprint would have borked my laptop :-) ) [10:00] veebers: initctl start --no-wait unity8 [10:00] * veebers tries [10:00] veebers: you are probably blocked on jobs that start on starting/started unity8, yet other conditions for them have not been satisfied and thus you are blocked. [10:00] xnox, that won't help for actually starting unity8, which didn't do --no-wait when starting indicators [10:00] xnox, nope, he's blocked on starting indicators (which unity8's pre-start does) [10:01] Saviq: horum. [10:01] xnox, yeah, we added --no-wait since then [10:01] need to SRU [10:04] Saviq: put trunk http://bazaar.launchpad.net/~unity-team/unity8/trunk/view/866/data/unity8.conf as ~/.config/upstart/unity8.conf `initctl start unity8` and it still hangs there (nothing popsup) [10:05] veebers, that's because the previous job is still hanging [10:05] veebers, make sure no initctl is still going [10:05] Saviq: ah ok, will do [10:05] veebers, last resort restart your session... [10:05] veebers: new config takes effect, only after the job is fully stopped. [10:06] xnox, Saviq: ah ha awesome that works a treat thanks guys /me finds it amusing when IRC notifications pop up in the windowed unity8 :-) [10:07] veebers, ;) [10:49] mhr3: any idea about this? [10:49] tsdgeos, "this"? [10:49] mhr3: http://paste.ubuntu.com/7487951/ [10:49] unity::scopes::ConfigException: invalid config file name: "/tmp/Runtime.ini.J10927": missing .ini extension [10:50] fix for that is being landed right now [10:50] in u8 approved mps [10:50] tsdgeos, but just run it without params [10:50] that will work [10:53] mhr3: somehow my scopes are not running :/ [10:53] this was an attempt at getting it to run [10:53] i get the [10:53] ERROR! Caught unity::scopes::TimeoutException: Request timed out after 300 milliseconds [10:53] and nothing happens [10:53] tsdgeos, stop scope-registry && stop smart-scopes-proxy && start smart-scopes-proxy [10:53] mhr3: done that like 10 times already :D [10:53] * tsdgeos does again [10:54] same [10:54] tsdgeos, hm, what's in ~/.cache/upstart/scope-registry.log? [10:54] oh wait, you had the perm issue earlier didn't you? [10:54] i did [10:54] so i guess you still do? [10:54] no [10:55] http://paste.ubuntu.com/7487972/ [10:55] that was fixed somehow [10:55] ok, it's saying the ss-proxy doesn't work [10:55] so pastebin that pls [10:56] mhr3: http://paste.ubuntu.com/7487977/ [10:56] see, still the perm issue :P [10:56] ok [10:56] at least that's what the catalan message look like :) [10:57] a chmod fixed it [10:57] where do i open a bug so this doesn't happen again? [10:57] lp:unity-scopes-api [10:58] ok, tx === zequence is now known as GrumpyStudio === GrumpyStudio is now known as zequence === MacSlow is now known as MacSlow|lunch [11:25] greyback, hey, sorry to be pushing, what do we do with the close fix branch? [11:28] mhr3: ping again [11:29] mhr3: Saviq: after all the get() changes only remaining comment's from mzanetti in the unity-api branch is http://paste.ubuntu.com/7488090/ [11:29] what do you guys think? [11:29] tsdgeos, I think we should kill it with fire... but we probably can't, yet [11:30] so just leave it be? [11:30] tsdgeos, let's see what mhr3 says, but I'm afraid so... [11:30] kill me with fire? :D [11:30] good thing there's lots of water around Malta [11:30] well, we could drop it from the shell API and deprecate it in the scope api in due course... [11:31] mhr3, ? [11:32] Cimi, so what's the verdict for the carousel shadow? [11:32] tsdgeos, Saviq, well we still need to send it to the scopes, and someone needs to provide it [11:32] mhr3, no, we don't [11:33] Saviq, I am looking at the padding first [11:33] mhr3, we really don't, not in the current meaning of form factor [11:33] Saviq, had to download and push few videos [11:33] Saviq, you're free to rename it ;) [11:33] mhr3, only I don't know what to rename it to, yet [11:33] Saviq, but scopes should know what layout they're providing results for [11:33] mhr3, what "layout"? [11:34] Saviq, tv, touch, mouse [11:34] i don't know [11:34] mhr3, yeah, sure, but we can't define those yet [11:34] but they specify the renderers, some might make more sense than others [11:34] for specific layouts [11:35] mhr3, doesn't feel convergent, tbh :P [11:35] Saviq, convergence is about proper ui, right? :) [11:35] mhr3, yeah, one that's not changing for no reason [11:35] ok, so ignore that for the moment i guess? [11:36] and since scopes define renderers, they define ui [11:36] so should be smart about it [11:36] mhr3: do you feel like adding docu for roles or shall we convince mzanetti it will just say "result returns results" [11:36] mhr3, well, then form factor doesn't give enough info anyway [11:37] tsdgeos, i'd say add a FIXME, and it will be just a doc fixme [11:37] mhr3, because 7" tablet would be the same as a 20" tablet [11:37] mhr3: fixme for what? [11:37] tsdgeos, "document what values are supported" [11:38] mhr3, tsdgeos asked about model roles [11:38] mhr3: can you do it since you know what you want to do? :D [11:38] Saviq, tsdgeos, eh sorry, i was still talking about the formFactor [11:39] if the roles need documenting, we're doing it wrong [11:40] I agree [11:40] ie the enum value is not good :) [11:41] mhr3: so document the fixme for the formfactor and done? [11:42] yep [11:42] Saviq: it needs a little updating against the short app Id change, and trying to fix as best as possible the issue you reported. Hopefully have something in a couple of hours [11:42] greyback, ok, just let me know if this is taking too much of your time (MAE higher prio) [11:42] Saviq: sure [11:42] mhr3: ok, i'll do it [11:43] tsdgeos: just saw this one in the MR: 9 + -- Albert Astals Cid Thu, 15 May 2014 17:20:44 +0200 [11:43] I guess should be your canonical mail [11:43] probably [11:45] alright, i have half day off today, so cya tomorrow === mpt_ is now known as mpt [11:46] mzanetti: fixed [11:59] Saviq, ok so the delegate of the carousel video has empty space at the bottom [11:59] Cimi, fix it! [11:59] Saviq, I am :D [11:59] Cimi, tx [11:59] Saviq, looking to see if it comes from card === alan_g is now known as alan_g|lunch [12:01] Saviq, cardCarousel.components && cardCarousel.components["art"]["aspect-ratio"] || 1.0 [12:01] Saviq, this seems 1.0 in video carousel [12:02] Cimi, unlikely [12:02] Cimi, at least the scope sends 1.5 [12:02] Cimi, you can see in the scope tool [12:02] Saviq, TypeError: Cannot read property 'art' of undefine [12:02] cardCarousel.components is undefined [12:02] Cimi, can we please not do IRC debugging again? [12:02] I'm live digging [12:02] ah ok [12:03] I like it [12:03] Cimi, get your own channel then ;P [12:03] Saviq, yeah but nobody will join >D [12:04] Cimi, yeah, that's 'cause it's really unproductive for the other party [12:05] Saviq, yes [12:05] Saviq, btw bug is fixed [12:05] Saviq, s/cardCarousel/cardTool/g [12:05] good [12:05] pushing soon [12:05] Cimi, testable? [12:05] Saviq, looks more like someone mistake [12:06] maybe yeah,,, [12:06] Cimi, that's what tests are for ;) [12:06] we should test if the carousel tile width and height have same aspect ratio of the backend data [12:07] can be done with some objectName magic maybe, othewise aliasing properties [12:10] tsdgeos, will you have a look at what Mirv reported about Qt 5.3 and your patch? [12:11] /latefood [12:13] Saviq, ping me when you back [12:20] mhr3, you in the office one of these days? [12:20] Cimi, wed probably [12:21] ok [12:21] mhr3, want to try my wifi card? [12:22] or flashing your bios together? [12:23] Cimi, oh yea, do you want mine? [12:23] Cimi, i can give you mine and try to do nasty things to yours :) [12:23] mhr3, not that I want yours [12:23] mhr3, mine works fine [12:23] Cimi, but is it ac? [12:23] mhr3, nope [12:24] Cimi, see, so you can have an upgrade [12:24] mhr3, but I don't have an AC ap anyway [12:24] mhr3, neither I know anyone with one [12:24] give it few months :) [12:25] but yea, i was trying to flash my bios back and forth and all the way around, but couldn't make it work [12:26] the latest bioses have some stupid security features [12:26] and i have a feeling that if i play a bit more with it, my laptop will just stop booting [12:26] ok >D [12:26] let's try mine, otherwise send it back [12:27] it won't just work out of box === MacSlow|lunch is now known as MacSlow [12:27] there's only one wifi/wimax intel card that this laptop supports [12:27] it could work [12:27] but i could flash the epprom on the card ;) [12:27] how do you know if a card is supportd or not? [12:28] there's a list [12:28] link [12:30] mhr3, my 5100 is bundled with some thinkpads, so it might be that it works [12:33] Cimi, http://download.lenovo.com/parts/ThinkPad/ThinkPadEdgeE430_2012-04-10.pdf [12:34] mhr3, nothing basically :D [12:34] pretty much :) [12:35] mhr3, https://www.diggloving.com/product_info.php?language=en¤cy=GBP&products_id=2233 ? [12:35] yea, that's the one intel it supports [12:37] Cimi, but you still need to buy it from lenovo for about 25 pounds... [12:37] so... yea [12:37] Saviq: close fix branch updated, please lemme know if I succeeded, or if I broke something else [12:37] Saviq: heya, is the whole removing "Preview" from the preview titlebar going to be discussed in Malta? [12:53] Cimi, back [12:53] greyback, ok, looking [12:53] sil2100, ping [12:53] pstolowski: pong :) [12:55] cwayne1, not sure, it might come up in the dash design session [12:58] Saviq: would you mind making sure it's discussed? it looks not great when we send screenshots to partners and the preview title is cut off... [12:59] cwayne1, hmm with the planned design it might get even worse https://sites.google.com/a/canonical.com/unity8dash/toolkit/14-previews [13:00] cwayne1, since it's meant to hold content title after all [13:00] cwayne1, that's not signed-off design, yet, though === alan_g|lunch is now known as alan_g [13:02] cwayne1, in any case, I'll make sure to mention it, yes [13:04] Saviq: thank you === pete-woods is now known as pete-woods-lunch [13:05] Saviq: so next thing i have to bother you about :) any update on shell rotation? [13:06] cwayne1, you can check the status out in https://launchpad.net/~unity-team/+archive/phone-right-edge [13:06] cwayne1, there's rough edges, but just upgrading from that PPA should give you a rotating shell [13:08] Saviq: sure i will [13:09] Saviq: do we plan on having it landed before MAE? or are we going to need to hack it again like MWC [13:10] cwayne1, we'll need a custom image for sure [13:10] cwayne1, but only one [13:11] daw [13:18] Saviq, sorry I am busy with the test [13:18] Cimi, nw, it's you who's asked me to ping when back ;) [13:18] Saviq, anyway I noticed that there is more padding above the carousel then at the bottom (in music) [13:19] Saviq, this is due to the sum of internal padding of carousel centering the items [13:19] Cimi, mhm [13:19] Saviq, plus the empty space of the header of the category [13:19] in Music scope [13:20] "Songs " [13:20] no idea how to best fix this [13:20] design wise... [13:20] it doesn't look nice [13:21] Saviq, carousel has even padding top and bottom of tiles [13:21] Saviq, but at top you add the empty space next to the label [13:22] we should probably remove the carousel top padding [13:22] because we know we have the padding due to the label [13:23] greyback, only thing I didn't expect: upstart-app-stop does not remove the app from the list, but that's probably correct - until you remove it from the dash, it's meant to be restartable, even though the job goes away? [13:24] Saviq: nope, it should go away. Darn [13:24] greyback, orly? [13:24] so you can distinguish "expected" stop vs. "unexpected" stop? [13:24] Saviq: how does a user stop an app from the CLI? [13:25] Saviq: upstart can distinguish those events [13:25] greyback, ok, so yeah, that'd be a problem still [13:25] so I broke something... [13:26] weird, I've a test for that and all [13:27] greyback, to make sure we're on the same page: https://code.launchpad.net/~gerboland/unity-mir/fix-upstart-closed-apps2/+merge/218721/comments/525521 [13:29] greyback, that'd be the only thing, all the other issues are fixed! [13:29] Saviq: ack, on it === _salem is now known as salem_ [13:37] * greyback sees his test in wrong.... [13:37] Saviq: what do you think it should do? Second opinion wanted :) [13:42] mzanetti: phone right edge ppa looks likes its got all new goodies...i was gonna try, just checking if there was a reason i shouldn't ? [13:42] greyback, I *think* you're right, it probably should close [13:42] kgunn: hit it! [13:43] Saviq: yep, I agree [13:43] greyback, it should only restart when upstart didn't expect it [13:43] kgunn: it'll crash after screen blanking right now, and can't start apps currently. [13:43] greyback, basically, retcode !=0 [13:43] kgunn: but it respawns if it crashes and it rotates [13:43] mzanetti: any idea why apps don't launch? [13:43] mterry, first of all thanks for taking time out of your weekend to fix the conflicts [13:44] greyback: no. dandrader has the suspicion because of newer mir and we need to update something [13:44] mterry, unfortunately it was a nail to your own coffin - I was able to test stuff ;) [13:44] mterry, and found a bunch of issues that still need attention [13:44] Saviq, oh did you find a problem? (Good ish?) [13:44] mterry, https://code.launchpad.net/~mterry/unity8/split/+merge/213149 [13:44] mzanetti: hmm ok [13:44] greyback, has the launching scheme changed in the meantime (the way appids are handled, I don't know) [13:44] ? [13:44] or the appid format [13:44] dandrader: actually... it works here [13:45] dandrader: nothing that should break anything [13:45] any changes were backward compatible [13:46] Saviq, in the mock of fake_categories.cpp [13:46] Saviq, can you explain me if (index.row() < m_specialCategories.count()) at line 97 [13:46] Saviq, OK, will comment in MP, thanks for the testing! [13:46] basically why we have this if/else [13:47] yeah. it's working. was able to launch gallery, system settings and phone app [13:47] mterry, yeah, some things I'm sure are expected [13:47] mterry, just wanted to write all down [13:47] mterry, but there's a few blockers still [13:47] Cimi, IIUC it checks if it's not the last category [13:48] but not calculator [13:49] Saviq, dunno, I will copy to both if/else [13:49] Saviq, I want to play with different aspect ratio here [13:49] and check in test DashContent === salem_ is now known as _salem [13:49] the aspect ratios [13:51] Cimi, m_specialCategories is a QList that you can override in tests [13:51] Cimi, altho looks like it's not used any more [13:51] Cimi, ah no, it is [13:51] addSpecialCategory [13:51] Cimi, Apps is a special category [13:52] Cimi, I mean running apps in the Apps scope [13:52] Cimi, we need to inject that category to display the running apps (not for long, though! so we'll be able to get rid of all that) [13:54] * mzanetti likes how launchpad estimates a build completed a minute ago while its still running === _salem is now known as salem_ [13:58] kgunn: hmm... I'd need to delete a package in the ppa (cause I messed up with versioning). What's your eta? [13:58] mzanetti: i just flashed....so whenever, no rush [13:59] kgunn: ok... then do the upgrade and let me know onces you fetched it === dandrader is now known as dandrader|afk [14:00] Saviq, in make tryDashContent, I have few empty categories [14:00] Saviq, do you have the same? [14:01] Cimi, doesn't look like it [14:01] Cimi, sounds like one of your renderers fails to parse [14:03] kgunn, fyi, I cleaned up the old recipes we had in https://code.launchpad.net/~unity-team/+recipes [14:04] cool, thanks, it was a mess [14:04] Saviq, thanks so much for the testing! I should have tested manta more recently. Got complacent with mako [14:05] Saviq, but SIM unlocking is a pain, since I don't have a SIM that unlocks to test [14:06] mterry, nw, I'm only just flashing now properly, will test there again [14:06] mterry, you should be able to set a PIN in Android, or do you mean there are SIMs which don't support it at all? [14:06] Saviq, I'm sure there will still be some problems, since I haven't given malta the attention it deserves [14:06] * mterry busts out his manta [14:13] Hi! a little help please. If you have user accounts authenticated with LDAP (say), and use nss tables other than passwd to store user information, then you find that the system automatically creates an entry in the user account database on login. This is handy: it means the user gets listed by lightdm in the greeter, for example. However, these users cannot be deleted. If you use the accounts-daemon from the control center to remove them, you get that th [14:13] e user isn't in /etc/passwd [14:13] tsdgeos, builddir/plugins/Dash/createCardComponent:40:51: Unable to assign [undefined] to QString where is that? [14:15] Cimi: the card creator [14:15] tsdgeos, yeah, but how do I get the line? [14:16] it's there [14:16] line 40 [14:16] obviously line 40 of the generated code [14:16] you'll have to print the generated code [14:16] tsdgeos, and where is the generated code? [14:16]