=== chriadam|away is now known as chriadam [01:38] cwayne: looks good! === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === vying is now known as Guest75991 === chihchun_afk is now known as chihchun === jibel_ is now known as jibel [06:51] good morning === chihchun is now known as chihchun_afk === chriadam is now known as chriadam|away === chihchun_afk is now known as chihchun [08:02] Good morning all; happy School Nurse Day! :-D === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [08:42] ogra_, following on from yesterday's discussion of disk-full problems, lp 1316978 is now open, and you may wish to close it as a dupe of the log rotation bug ( lp #1270248), which I've also commented on [08:42] Launchpad bug 1316978 in android (Ubuntu) "User copying files to ~/phablet can fill media, and prevent boot" [Undecided,New] https://launchpad.net/bugs/1316978 [08:42] Launchpad bug 1316978 in android (Ubuntu) "User copying files to ~/phablet can fill media, and prevent boot" [Undecided,New] https://launchpad.net/bugs/1316978 [08:42] Launchpad bug 1270248 in logrotate (Ubuntu) "/var/log fills up disk space on phone" [High,Confirmed] https://launchpad.net/bugs/1270248 [08:42] Launchpad bug 1270248 in logrotate (Ubuntu) "/var/log fills up disk space on phone" [High,Confirmed] https://launchpad.net/bugs/1270248 [08:55] elopio: hi! [08:55] elopio: do you have some time to debug that authentication failure? [09:03] hey mardy [09:03] I do. Just tell me know what to do. [09:04] elopio: oh, I just sent you an e-mail [09:05] elopio: just enable logging in /etc/signond.conf [09:05] elopio: then try all the flow again, and let's see the syslog [09:05] elopio: the DBus message appears to be correct [09:05] mardy: I have it enabled. Let me take a look at the syslog. [09:08] mardy: http://paste.ubuntu.com/7409314/ [09:08] maybe May 7 03:06:35 tangamandapio-desktop signonpluginprocess[5422]: oauth1plugin.cpp 254 process : Invalid provided tokens data - continuing normal process flow [09:12] elopio: indeed. Let me check the plugin code... === sephiap_ is now known as sephiap === alan_g is now known as alan_g|afk [09:13] elopio: ah, that happens if either of AccessToken or TokenSecret are empty [09:15] elopio: actually, the keys that ProvidedTokens expects are not exactly the same of the OAuth reply... [09:15] rsalveti, the SD Card bug I mentioned yesterday: https://bugs.launchpad.net/ubuntu/+source/android/+bug/1316997 [09:15] Ubuntu bug 1316997 in android (Ubuntu) "SD Card fails to auto mount in port" [Undecided,New] [09:15] Launchpad bug 1316997 in android (Ubuntu) "SD Card fails to auto mount in port" [Undecided,New] [09:15] mardy: ok! [09:15] I've put 'dummy' and now it works. [09:15] elopio: so, you can specify "AccessToken", "TokenSecret", "UserId", "ScreenName" [09:16] all the rest are ignored [09:16] weird thing is that it shows the dialog as if it had no account for some seconds. But this is moving forward [09:16] thank you very much. [09:19] mardy: what should be the userid? The same as I have in 'edam_userId'? [09:19] john-mcaleely, thanks a lot, so the 5% reserved for root wouldnt help at all [09:20] ogra_, that makes sense. I'm not clear how reservation would work for non root processes, but would that be the simple fix? [09:20] elopio: I think you don't need to set it [09:20] john-mcaleely, lol, "simple" ... i fear we need quotas or something, that smells like it can get very fiddly and awful [09:21] mardy: right, they don't seem needed. [09:21] mardy: ok, only remaining problem, the No account dialog disappears until I go to online accounts [09:21] am I missing a step? === vrruiz_ is now known as rvr [09:24] ogra_, yuk. I think I'm pleased I typed up a separate bug then :-( [09:24] yeah [09:27] elopio: mmm... I don't follow you === alan_g|afk is now known as alan_g [09:28] mardy: I add the account with my credentials.py script. [09:28] I go to reminders app [09:29] and it shows me the dialog that says: No Account, go to ... [09:29] then I opened online accounts to see what was going on [09:29] and the dialog disappeared. [09:29] as if I'm missing something on my script that gets done when I open online accounts. [09:30] elopio: are the account and its service both enabled? (check with account-console) [09:33] mardy: I think so http://paste.ubuntu.com/7409417/ [09:34] elopio: the "evernote" service is not enabled [09:34] I see. [09:35] elopio: this is done by "account-console edit --enable --service evernote", you can see how it's done there [09:36] mardy: thanks, because my first naive try was unsuccessful :) === ALan_ is now known as Guest87832 [09:51] Hi I want to use ubuntu but after downloading 14.04 the file is 930mb and my cd rw will only write to 800mb what can I do? [09:51] ok, success! [09:51] mardy: https://code.launchpad.net/~elopio/reminders-app/test_with_account/+merge/217171 [09:51] it's ready to review, please take a look when you can. [09:52] elopio: cool! OK, will do that a bit later [09:52] thanks for your help. I'm going to get some sleep now. [09:53] Guest87832, your phone has a CDRom ? [09:53] (note this is the ubuntu touch channel for the ubuntu phone OS) [09:53] I have a samsung S3 [09:54] I am not trying to put Ubuntu on my phone but on my pc [09:54] what do you plan to do with the CD ? [09:54] right, then you are wrong here ... #ubuntu isteh right channel for that [09:54] Guest87832: you have to write it on a dvdrw, or an usb stick. it doesn't fit cd anymore [09:55] I haved tried writing to USB stick but it wont write an ISO file (i Have a 16Gb stick) [09:56] https://help.ubuntu.com/community/Installation/FromUSBStick [09:56] I tried re-formatting stick as FAT 32 but and use FREEISOBURNER to write but it wouldnt work [09:56] you can check the steps for creating a stick on that page ^ [09:57] OK will try that thanks [09:57] btw I'm in Canterbury UK where are you [09:58] I was wondering if its better rying to find an older version of ubuntu then upgrading once installed. Is that a better way? [09:58] sergiusens, ricmm ... bug 1316978 ... does any of you have any idea what we could do if not quotas ? [09:58] bug 1316978 in android (Ubuntu) "User copying files to ~/phablet can fill media, and prevent boot" [Undecided,New] https://launchpad.net/bugs/1316978 [09:58] bug 1316978 in android (Ubuntu) "User copying files to ~/phablet can fill media, and prevent boot" [Undecided,New] https://launchpad.net/bugs/1316978 [09:59] hmm, even quotas might not work [09:59] since it would be per user but not per dir [09:59] Guest87832: a better way to do what? [09:59] guys, can you take that to #ubuntu ? [09:59] install Ubuntu [10:00] the easiest way is to create a bootable usb stick from 14.04 image [10:00] OK thanks will look at the page you sent and try from stick. Good luck bye [10:02] JohnLea, ping [10:16] ahayzen, pong [10:17] JohnLea, Hi I found an issue where scrolling sheets also scrolls things behind, eg a page. I wanted to confirm that no events should go to the things behind in the two cases where the mouse is inside the sheet and outside? === dbarth_ is now known as dbarth [10:18] ahayzen, yes, you are correct, no events should go to items behind the sheet in these two cases. in the second case clicking outside this sheet should close the sheet. [10:18] JohnLea, ok thanks for the confirmation :) t1mp ^^ [10:19] ahayzen: was it scrolling in a sheet or in a popover? [10:19] t1mp, yep [10:19] ahayzen: sheet or popover? [10:19] t1mp, sheet uses popover doesn't it? [10:19] JohnLea: sheets don't close when you click next to them, they have cancel/confirm buttons [10:20] ahayzen: no, but we have a "popup" as a parent of popover, sheet, dialog. easy to confuse [10:20] JohnLea: I made a note once that sheets should actually be deprecated. Do you know if that is still the case? [10:21] t1mp, hmm but the patch i did fixed the issue we were having with the DefaultSheet... oh is it because we open it with the popup? [10:21] * ahayzen checks code [10:21] ahayzen: ah ok, I thought it was with a popover, not with defaultsheet. Then your question was correct [10:22] t1mp, we do PopupUtils.open() on a DefaultSheet [10:22] ahayzen: that's fine [10:22] t1mp, cool thanks :) [10:22] t1mp, yes, sheets are actually depreciated, and the sheets were only ever designed to work in the context of the phone, we never did the designs for desktop/tablet. However we don't have a replacement for sheets, in most cases where sheets were previously used we are taking the user to a new screen instead. Again we are only focused on the Phone at the moment, so there will be desktop and tablet use cases that have not been thought [10:22] through [10:22] heh [10:23] JohnLea: so we can deprecate sheets (everywhere, also for phone), and the recommendation is to use a new screen instead? [10:24] a new screen? [10:25] t1mp; that is the direction things are currently going, and is what we are doing with the designs we are currently working on. However the pattern/sdk update is not complete or published yet, and we have not had a chance yet to revisit the old designs that still use sheets, so they continue to use sheets until we can get back to them and update the designs [10:25] t1mp, more we are phasing the use of sheets out, and are not using them in new designs, or when we update designs [10:25] JohnLea: okay, we'll keep them around for now then [10:26] t1mp, cool, that sounds like the right think to do for now [10:27] JohnLea: when someone needs input from design, what should I recommend? Report a bug in ubuntu-ux, or are those bugs no longer reviewed regularly? [10:28] JohnLea: I was asked about this bug yesterday https://bugs.launchpad.net/ubuntu-ux/+bug/1295993 but there are probably more [10:28] Ubuntu bug 1295993 in Ubuntu UX "[ux]Ubuntu touch colour theme is too dark making it unusable in bright sunlight condition" [Undecided,New] [10:28] Launchpad bug 1295993 in Ubuntu UX "[ux]Ubuntu touch colour theme is too dark making it unusable in bright sunlight condition" [Undecided,New] [10:28] t1mp, yes, any issue should be reported as a bug to ubuntu-ux (also marked 'also affects' with the engineering project responsible for the component), then once this is done if the issue is urgent ping someone here to get someone to take a look at it [10:28] as we have a lot of bugs [10:29] yes I saw, 520 open, that's at least a challenge to manage [10:30] t1mp, re. that bug, we are moving to light background so that will solve that issue. However there is another bug open saying that lighter background harm battery life! Cant win them all ;-) [10:31] JohnLea: I'll copy&paste your answer in the bug and close it then [10:31] cool, thanks! [10:31] JohnLea: I think many of the design bugs are simple questions and perhaps can be closed after answering them [10:32] t1mp; we are currently reviewing all the bugs and assigning them to people, and then the next task will be for the assignees to answer and close or action. I also think most of them can be closed [10:32] JohnLea: ok that's great. Good to know. [10:32] but need to do full review (which is in progress) [10:32] perfect :) [10:39] ogra_: should mtp check avail space before starting to copy/write? [10:40] sergiusens, well, that wont help with i.e. dowloaded content [10:40] ogra_: so it's not a matter of filling up the disk, more so of leaving little avail; fwiw, the same is almost as applicable to the desktop [10:40] sergiusens: did you speak to asac about reminders? [10:40] popey: arg, forgot [10:41] asac: some peeps want reminders in the default install; are we ok with that? Test enablement should happen once it's in and can be coordinated between popey and doanac [10:42] the desktop will still boot to lightdm ... the phone wont [10:43] (due to autologin) [10:44] ogra_: nothing easy comes to mind here... :-/ [10:44] yeah ... the only thing i can imagine is to keep data and dot dirs on separate partitions ... [10:44] then we could quota it ... [11:12] sergiusens: what are reminders? [11:14] sergiusens: it's an app which lets you sync notes down from evernote and view/edit them on your phone === vying is now known as Guest19421 [11:30] popey: I know what it is ;-) [11:32] sergiusens: er, mis-hilight there, sorry ☻ [11:34] Mirv: did the train whistle? [11:35] popey: come to think of it; we agreed we would add apps into the default install if the app was of good quality [11:36] bzoltan1: yes it did === MacSlow is now known as MacSlow|lunch === jhodapp|afk is now known as jhodapp === gcollura is now known as gcollura|afk [12:10] mardy: ping! [12:10] paste.ubuntu.com/7409741/ getting that when trying to update an app on my phone - is that a signon issue? [12:10] 2014-05-07 10:50:56,524 - CRITICAL - ../../../../lib/SignOn/connection-manager.cpp 106 setupSocketConnection p2p error: QDBusError("org.freedesktop.DBus.Error.FileNotFound", "Failed to connect to socket /run/user/32011/signond/socket: No such file or directory") 1 [12:24] hm, where are XDG_CONFIG_DIRS initially defined? [12:27] popey: hi! Yes, that's a signon issue, but it's harmless [12:27] popey: in trunk I've turned that critical into a debug [12:27] * ogra_ guesses the issue is: [12:27] 2014-05-07 12:58:05,467 - WARNING - QDBusObjectPath: invalid path "https://public.apps.ubuntu.com/download/com.ubuntu.developer.bobo1993324/qmltextreader/com.ubuntu.developer.bobo1993324.qmltextreader_0.1.2-c_armhf.click" [12:28] and i think thats a mandel issue (u-d-m) [12:28] or an issue with whatever hands that url to u-d-m [12:28] ogra_: Does taking screenshots work on utopic #15? When I use phablet-screenshot, I keep getting "remote object '/tmp/mir_screencast_768x1280.rgba' does not exist". Rebooting doesnt seem to help. [12:29] nik90: filename changed [12:29] nik90, ah, thanks for the reminder ... i have a fix half done for that [12:29] nik90: look in /tmp, it's (bizzarely) got a 60Hz added to it [12:29] bug 1316806 [12:29] until that lands, just hack the filename [12:29] bug 1316806 in phablet-tools (Ubuntu) "phablet-screenshot looking for the wrong filename" [High,New] https://launchpad.net/bugs/1316806 [12:29] bug 1316806 in phablet-tools (Ubuntu) "phablet-screenshot looking for the wrong filename" [High,New] https://launchpad.net/bugs/1316806 [12:29] uh, do we need two bots? [12:29] popey, ogra_: ah thnx.. [12:30] popey, no, the second one returned after jussi added the other one it seems [12:30] * popey pokes irc people [12:33] ogra_, popey: phablet-screenshot keeps taking the screenshot of the app scope instead of the actual app that I am in. [12:34] odd [12:34] yeah [12:34] ogra_: yeah to the issue? or to it being odd? [12:34] odd :) [12:34] * nik90 tries rebooting [12:38] hello my wifi is not working after update to 12.10 [12:38] please help me any one [12:41] * ogra_ notes ChickenCutlass brought his evil twin today [12:42] ogra_, ChickenCutlass *is* the evil twin [12:42] hey. are there any examples i can go look at showing opengl's usage with mir? [12:42] ogra_, how do I kill one of them [12:43] jodh, wrangling with bug 1316978 ... is there a way to make upstart fall back gracefully to omit all logging weh the disk is full ? so that at least the session can still start [12:43] bug 1316978 in android (Ubuntu) "User copying files to ~/phablet can fill media, and prevent upstart from creating logs (which kills the session)" [Undecided,New] https://launchpad.net/bugs/1316978 [12:43] bug 1316978 in android (Ubuntu) "User copying files to ~/phablet can fill media, and prevent upstart from creating logs (which kills the session)" [Undecided,New] https://launchpad.net/bugs/1316978 [12:43] ChickenCutlass_, nickserv has a ghost command iirc [12:43] ok I will look [12:44] robjh, try asking in #ubuntu-mir [12:44] ubottu: part === MacSlow|lunch is now known as MacSlow [12:44] * ogra_ hugs jussi [12:44] there you are popey :) [12:44] master of the bots :) [12:44] * jussi hugs ogra_ back [12:44] thanks jussi ☻ [12:44] You are most welcome [12:45] ogra_: it already does that. That bug appears to be an issue with the dbus job which is trying to write into $HOME. Maybe it could write to /run ? [12:46] hmm, i wonder why ... [12:46] * ogra_ takes a look at the job [12:47] echo "DBUS_SESSION_BUS_ADDRESS=${DBUS_SESSION_BUS_ADDRESS}" >$HOME/.cache/upstart/dbus-session || true [12:47] and [ -d "$HOME/.cache/upstart" ] || mkdir -p "$HOME/.cache/upstart" || true [12:47] hmpf [12:47] thats not easily movable to /run ... [12:48] ogra_: why not? [12:48] jodh, it is parsed by commandline shells too ... for autopilot etc [12:49] ogra_: well, the solution is to move that file, or dispence with it completely. [12:49] though if no dbus runs its probably moot to parse it [12:50] but even if we move it ... there are other session bits writing to bashrc ... that will be the next stopgap ... [12:50] sigh [12:50] i wish we could fix that on a filesystem level instead [12:51] ogra_: could those things 'initctl set-env' rather than actually modifying .bashrc I wonder? [12:52] jodh, hmm [12:54] jodh, i guess they theoretically could ... but thats a hell of a lot of stuff to re-work [12:56] jodh, heya, so upstart looks in XDG_CONFIG_DIRS/upstart for session jobs right? my problem with that is, how can we make sure the dir we need is in XDG_CONFIG_DIRS before upstart has started? === gnu is now known as Guest23724 [13:10] cwayne: you'll need to tweak the script you use to start upstart since the directory needs to exist at the point upstart starts. [13:10] jodh, any idea where that script lives? [13:11] mpt: hi! Do you have a minute (topic is "scopes and Online Accounts")? [13:12] mardy, yep [13:13] mpt: so, there is an issue: some scopes might want to use Online Accounts, but ATM the only way to grant them access to an account is if the user goes to the System Settings, opens Online Accounts, taps on an account and enables the scope from there [13:13] thank you ogra_ :) [13:13] mpt: which is not very discoverable [13:13] mardy: hello [13:13] mpt: dbarth will probably set a meeting to talk about this on Friday, would you like to join? [13:13] t1mp: hi! [13:14] mardy: can you resubmit https://code.launchpad.net/~mardy/ubuntu-ui-toolkit/lp1296728/+merge/212460 for merging to lp:~ubuntu-sdk-team/ubuntu-ui-toolkit/staging instead? [13:14] cwayne: last time I looked it was /usr/bin/ubuntu-touch-session [13:14] mardy: all our MRs go there first, and after that we land them to lp:ubuntu-ui-toolkit [13:14] t1mp: oh, you are also a victim of the citrain, I see... :-) [13:14] jodh, lovely, thank you [13:15] mardy, sure [13:15] mardy: we are still working on figuring out what is the optimal way for us to use it ;) [13:15] t1mp, btw, tested out the uitk with new headers last night, looks great! [13:15] mardy: but yeah, we have a staging now in case landings are slowed down (and it needs to land before going to trunk) [13:15] cwayne: awesome :) [13:16] cwayne: which apps are you testing? [13:16] t1mp, i tried out some of mine, and then just started adding useDeprecatedToolbar to false in some of the preinstalls to get an idea of what they'd look like [13:17] t1mp: done [13:17] cwayne: nice. We made it opt-in so the app developers can check first to see if it works for their app [13:18] mardy: thanks. We'll ask loicm to review https://code.launchpad.net/~mardy/ubuntu-ui-toolkit/lp1296728/+merge/218617 since he knows the ubuntushape best [13:19] sergiusens: what are reminders? :) [13:19] which context? [13:19] mardy, is it the same problem with System Settings “Updates” right now? It says “Credentials not found” and “Please log into your Ubuntu One account.” [13:20] (And I tap that and it animatedly switches to … itself, which I assume is a separate problem) [13:20] asac, https://wiki.ubuntu.com/Touch/CoreApps/Reminders [13:20] asac: I don't follow; reminder is just an app that needs to be added to the click seeds [13:21] t1mp, nice, good idea :) === dandrader is now known as dandrader|afk [13:26] seb128, do you know if anyone is working on the can't update apps bug? [13:27] pmcgowan, it was just being discussed on #ubuntu-ci-eng, and popey/didrocks figured out it's due to the ubuntu-download-manager update on utopic #5 [13:28] pmcgowan, mandel is not around though, we are waiting for him I guess [13:28] pmcgowan, once we find mandel someone will :) [13:28] ah very good [13:28] * ogra_ lifts the carpet to take a look ... i guess he heard doanload-manager and ran away [13:30] pmcgowan: bug #1317067 [13:30] bug 1317067 in ubuntu-download-manager (Ubuntu) "Can't update click apps since image #5 in utopic" [High,Triaged] https://launchpad.net/bugs/1317067 [13:31] didrocks, ok, there are a number of dupes for that ;) [13:31] pmcgowan: there are some infos on that one, but as long as it's targeted and assign to the right person, I don't care :) [13:31] I set to mandel for now (but he isn't around) [13:31] yep [13:32] pmcgowan: i looked at bugs you'd filed because I thought you'd mentioned it yesterday, but couldnt see one so filed a new [13:32] np [13:35] Elleo, i triggered a CI rebuild for your content-hub branch [13:37] kenvandine: okay, it's looking like things might be delayed a bit on the webbrowser, as we're needing to rename the udm qml compontent to keep it consistent with other modules (it was UbuntuDownloadManager rather than Ubuntu.DownloadManager) [13:38] kenvandine: I've submitted a branch that changes that now though, so hopefully it'll all go through fairly quickly [13:41] mpt: something to do about scopes rather; what the experience for using a specific identity would be [13:42] (still otp, but reading here) [13:48] dbarth, I did a design for that back in , and Katie did a similar design in [13:50] kenvandine: why is online-accounts a separate app from system-settings, when none of the others are? [13:50] because it can actually be run separately [13:50] I would love to know the answer to that ^ question too … It causes numerous UI bugs [13:50] for example it provides the binary to run for auth [13:50] when something needs a re-auth, etc [13:51] and apps can use it for adding online accounts [13:51] without starting all of system-settings [13:51] eventually in a modal way [13:51] it does make for an unusual experience though, when pressing "back" you actually leave one app, but never quite switch to the other [13:51] like friends-app could let you add an online account directly without taking you to system-settings [13:52] that is until we have trusted sessions :) [13:52] and actually it not switching is probably just a bug [13:52] when it exits, the previous app should be focused [13:52] It showing up as an app is a bug too. The dialog for entering a Wi-Fi password doesn’t show up as an app. [13:53] When we have printing, the printer setup won’t be an app, etc [13:53] i think that's a snap decision, right? [13:53] the plan for online accounts is to use a trusted session, to give you a modal experience [13:53] like you never leave the app [13:54] content-hub will do the same [13:54] ok, so this is a stop-gap measure [13:54] yeah [13:54] * kenvandine hopes we get trusted sessions soon [13:54] it'll improve the content-hub experience 10x === gatox is now known as gatox_lunch [14:03] right trusted session is needed (and some refactoring) to have it get back to being a normal settings plugin [14:04] the reason it's out is because this is the best way to have it deal with app requests while trrusted sessions wee missing [14:04] mardy: ^^ [14:07] Does music not work in u15? [14:07] All my songs just stay at 0:00 [14:09] Oh, an ogg works [14:10] jodh, hey, so upstart will look for jobs in XDG_CONFIG_DIRS/upstart *and* /usr/share/upstart/sessions right? [14:11] cwayne: yes - see the ordered list of search directories in init(5) under "User Session Mode" [14:13] mterry, which file format you using? [14:14] ahayzen, the ones that didn't work were mp3 [14:14] mterry, hmm ok i'll take a look later, usually i test mp3, ogg, flac, m4a [14:16] sergiusens: well, i didnt even know reminders existed, so nothing to follow, just explain :) [14:18] asac, others call it "evernote" ;) [14:19] odd name for evernote app [14:19] trademark issue? [14:19] no idea [14:19] popey might know why it is named like that [14:19] They do not allow 3rd party app devs to call it Evernote [14:19] yes, brand issue [14:19] only they can [14:20] they may change their minds later [14:21] mpt: well, that's another issue, which will be fixed with no changes to the UX design [14:21] mpt: the problem for scopes is that they don't have a UI [14:21] mardy, I don’t know what you mean by either “UX” or “UI” there… [14:21] mpt: that is, it's not clear when and how a user should grant account access to a scope [14:21] How do scopes not have a UI?? [14:22] A scope *is* a UI. [14:22] mardy, oh, do you mean they don’t have an identity like an app does? [14:23] mpt: well, AFAIK the UI is provided by unity8, the scope provides the data and style [14:23] ‘“Apps” wants access to your Ubuntu One account.’ [14:23] mpt: they do have an identity, but when and how should this ‘“Scope” wants access to your Ubuntu One account.’ appear? [14:27] mardy, what is the problem with it being presented the same way as for an app? [14:28] mpt: OK, but when? The first time one does a search? [14:28] mpt: or when the scope is first installed? [14:29] mardy, when the scope asks for access to the account. Same as for an app. === dandrader|afk is now known as dandrader [14:30] Exactly when that happens is up to the scope designer. [14:31] mpt: mmm... the difference is that an application has its own window, and will typically request access to an account in response to a user action (such as clicking on an "Add account" button) [14:32] mpt: it's not clear to me where the scope should put that "Add account" button (if anywhere) [14:35] mardy, again, that’s up to the scope designer, it’s not an Online Accounts issue. It’s possible it might be solved for default scopes by having “Dash” settings in System Settings, but that’s still up to the Dash designers, it’s not an issue with Online Accounts itself. [14:36] mardy, mikenagle is the one to ask about Dash design issues. [14:38] mpt: ah, OK, for some reason I thought that you should design the whole Ubuntu Touch :-) [14:39] mikenagle: hi! If you have some time, please read the backlog of the chat between me and mpt (last 20 minutes) [14:39] mardy, if I was doing that, we wouldn’t have “scopes” in the first place. ;-) === gatox_lunch is now known as gatox [14:52] ToyKeeper, hi, whats the bug # for the over sensitive screen rotation? [14:52] whiners ... just hold your hands steady :P [14:53] heh [14:53] :) [14:55] ogra_: parkinson bug? [14:55] lol [14:55] ++ [14:55] * ogra_ hands sergiusens the "quote of the day" award [14:56] ogra_: tbh, I had the tablet sitting on my desk and tapped on a button and it rotated :-P not sure if an earthquake was in progress and only the tablet noticed [14:56] lol [14:56] well, i dont think it is so much different from android actually [14:56] i have that too on an android phone [14:57] if i put the phone down to read with hands free and the surface isnt 100% flat it does the same on android [14:58] anyone got time for a quick mp to ubuntu-touch-session? https://code.launchpad.net/~cwayne18/ubuntu-touch-session/custom-xdg-config/+merge/218641 [14:59] what I had problem with is it tries to rotate every time the phone vibrates after I touch something [14:59] Tassadar, i proposed an MP for the heavy vibration ... make bzoltan1 approve it ;) [15:00] bzoltan1: approve it! [15:04] awe_: you tried to ping me last night [15:09] yea, was wondering what happened to the UI for SIM PIN codes? [15:10] I didn't see it in the network menu anymore and couldn't find it anywhere else... [15:10] nothing should have happened to it [15:10] the "Unlock SIM..." item is only available if A) there is a modem B) it's locked [15:10] Tassadar, https://code.launchpad.net/~ogra/ubuntu-ui-toolkit/improve-haptics/+merge/218466 FYI [15:11] awe_: you are not seeing it at all= [15:11] is this on devel-proposed? [15:11] OK, I can re-test again, but it sure wasn't there yesterday [15:11] awe_: any changes to ofono landed recently? [15:11] Wellark, yes... proposed as of yesterday [15:11] no [15:11] well, this was actually with the ofono silo [15:12] but shouldn't have had any effect [15:12] I'll re-test with the image & the existing ofono after we finish our stand-up [15:12] awe_: is that the silo that also contains the dbus control channel to phonesim plugin? [15:12] also, not sure if you saw my indicator-network bug for flight-mode [15:12] Wellark, yes [15:13] awe_: good. I'm preparing an indicator-network silo as we speak [15:13] and I can mark the ofono silo as prerequisite [15:13] https://bugs.launchpad.net/ubuntu/+source/indicator-network/+bug/1315144 [15:13] Ubuntu bug 1315144 in indicator-network (Ubuntu) "Indicator icons don't reflect modem Offline state" [Undecided,New] [15:13] awe_: ok, that should also be "fixed" with the silo i'm preparing [15:14] ok [15:14] although we are not showing anything for modem offline [15:14] but we are showing a flightmode icon [15:14] iff urfkill tells that we are in flightmode [15:14] ok [15:14] BUT [15:15] as there is no flightModeChanged signal coming from urfkill [15:15] if you go and manually activate flightmode [15:15] through dbus for example [15:15] indicator-network has no way of knowing we just went to flightmode [15:15] Tassadar: show me that MR :) [15:15] as I tried to explain in the urfkill bug [15:15] bzoltan1: https://code.launchpad.net/~ogra/ubuntu-ui-toolkit/improve-haptics/+merge/218466 [15:15] bzoltan1, https://code.launchpad.net/~ogra/ubuntu-ui-toolkit/improve-haptics/+merge/218466 [15:15] :) [15:16] makes it more feel like android ... and the phone doesnt want to jump out of your hands [15:16] ogra_: what was thelink again? [15:17] bzoltan1, less awful haptics for button presses ... [15:17] pmcgowan: https://bugs.launchpad.net/unity8/+bug/1308376 and https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1288332 [15:17] Ubuntu bug 1288332 in ubuntu-system-settings (Ubuntu) "duplicate for #1308376 Orientation lock not available (rotation/portrait lock)" [Low,Triaged] [15:17] Ubuntu bug 1288332 in ubuntu-system-settings (Ubuntu) "Orientation lock not available (rotation/portrait lock)" [Low,Triaged] [15:17] Wellark, to be clear modem == offline & flighmode == enabled are two different conditions the indicator should recognize [15:17] awe_: sure they are [15:17] awe_: I have [15:17] case Modem::SimStatus::offline: [15:17] /// @todo show something. [15:17] break; [15:17] if the lack of the flight-mode signal is truly a blocker, then cyphermox will need to fix this before we can land flight-mode [15:17] fwiw [15:18] <__lucio__> aquarius, Chipaca, hello! [15:18] it's not a blocker as long as nobody is trying to mess with flightmode [15:18] other than indicator-network [15:18] heya __lucio__ [15:18] <__lucio__> aquarius, hello mr! how are you doing? [15:18] Wellark, I don't think we can make that a limitation [15:19] Wellark, so I guess we'd need to fix [15:19] __lucio__, not too bad -- we're halfway through the week :) [15:19] awe_: yes, we can't make thst limitation in the long run [15:19] but right now it's not a blocker [15:19] it'd be broken for use to keep separate flight-mode states in the indicator and urfkill [15:19] awe_: is urfkill in a state that it can be included to the image? [15:19] once my indicator-network branch lands it pulls in urfkill [15:20] mo [15:20] s/mo/no/ [15:20] <__lucio__> aquarius, nice way to look at it, do you have some time to discuss the app you have in mind? [15:20] we're trying to get there, but it sounds like this signal needs to be added [15:21] ok. let me put that another way.. if I now land indicator-network that pulls in the urfkill we have in the archive will that break everyones devices and desktops where unity8 is installed? [15:21] please don't do that [15:21] so, I need to throw out flightmode code then [15:21] there are still bugs in urfkill that we're trying to fix [15:21] ok. I can do that [15:22] I will schedule a hangout tomorrow for the usual network/telephony sync time so that we can discuss [15:22] it sounds like we're not all on the same page [15:22] __lucio__, sure thing. What I believe would be useful for push notifications are two things: the basics, which I'm sure you're already working on, and secondly a way for an app to ask for a list of push notifications that arrived since it last asked, along with their payloads. [15:23] mardy, I just realized: Having a scope ask for access to an online account, rather than the Dash as a whole, is like having a Web site ask for access to your location, rather than the Browser as a whole. [15:23] mpt, that's what websites do, no? [15:24] aquarius, that’s my point [15:24] __lucio__, the use case here is: other platforms send a push notification to my, for example, messaging app. The OS shows the notification and I can read it in the OS. Tapping the notification (or the messaging menu) launches my app, but my app doesn't know what the notifications were! [15:24] awe_: I see only one critical bug against urfkill [15:24] mpt, oh, good, OK then, I'll shut up again ;)] [15:24] <__lucio__> aquarius, thats the mailbox model, as opossed to the mailmain model, where we start the app and deliver notifications to it [15:24] and that's fix committed [15:24] aquarius: that's one of the bits of the api, yes [15:24] __lucio__, oh, a notification will always start my app? [15:24] mardy, so if it’s possible for the browser, it should be possible for the Dash. [15:24] __lucio__, presumably it starts some sort of invisible daemon subset of my app, right? [15:24] <__lucio__> aquarius, no, the user interacting with the notification will do it [15:25] <__lucio__> actually, we deliver "notified users", not notifications :) [15:25] __lucio__, ah, no. Because if I get ten notifications from, say, Twitter while I'm asleep, I'd like the Twitter app to be able to get the content of those ten notifications when it starts up even if I only interact with one [15:25] mpt, will you be in malta ? [15:25] awe_: could we have a hangout later today on the urfkill situation? I need to prepare a silo and I need to know what are the limitations [15:26] ogra_, yes [15:26] aquarius: __lucio__: a general answer, and some considerations [15:26] <__lucio__> go ahead [15:26] mpt, good ... i need to discuss some error mesage stuff with you we will need to show to the user [15:26] __lucio__, it is immensely annoying that the OS has the content of those notifications but the app doesn't -- I start up my messaging app with no network, and it won't show me the messages even though the phone already has them [15:26] aquarius: __lucio__: first, there will be an api endpoint to get a backlog of notifications [15:27] aquarius: __lucio__: these are notifications that have been delivered to the phone, no server-side work [15:27] local api endpoint? or server api endpoint? [15:27] aquarius: __lucio__: the app still needs to deal with there being more notifications than whatever arbitrary limit we set [15:27] aquarius: local [15:27] Wellark, sure...let me check with cyphermox. maybe +2h from now? [15:27] aquarius: as far as the server is concerned those notifications were delivered [15:28] aquarius: (at this stage, where we are not implementing at-most-once) [15:28] Sure; if my app gets a zillion notifications then we'll only cache the last fifty or ten or whatever; I would expect whatever shows me recent notifications as part of the OS would have that same cache. [15:28] ogra_, ok. In the meantime, may answer your question. [15:28] awe_: 2h from now is 20:30 to me.. so the sooner the better [15:28] (the pull-down top menu in Android or iOS, for example) [15:28] awe_: but 2h works for me, too [15:28] aquarius: caches are hard enough without having multiple of them and making promises about keeping them in sync :) but yes [15:28] awe_: or later if absolutely no other chance [15:29] awe_: just ping me when you have a time to propose [15:29] thanks! === dandrader is now known as dandrader|afk [15:30] mpt, i think we have an error message for that on the desktop already ... but we will need to adapt that for the phone (the issue is that if the disk is full you will not be able to run apps etc, we need to tell teh user to free diskspace somehow ) [15:30] aquarius: there will also be a bit of the api whereby if an app is running it'll get the notifications directly, if it wants to [15:30] Chipaca, ya, I'd expect that, indeed -- that's part of the "basic" stuff I mentioned : [15:30] Wellark, I just need to hear back from cyphermox [15:30] Chipaca: "if it wants to" ? [15:31] is there a fix to phablet-screenshot yet? [15:31] lucio_: was thinking of making that part optional [15:31] mhall119, i was planning to work on that but a big big bug cam in ma [15:31] cam in my way now [15:31] mhall119: yeah, or just fudge it locally to add _60Hz to the url [15:31] *came [15:31] s/url/filename/ [15:31] right, you can just hack it for the moment [15:31] Chipaca: why optional? [15:32] lucio_: because it makes the logic a little bit more complex; there's yet another thing the appdev needs to do [15:32] lucio_: it enables them to do better stuff, but it's more work [15:32] popey, mhall119 i would just add that bit but the script should stay backwards compatible so its a little more than just adding the missing bits [15:32] right [15:33] Chipaca, fire an event inside the app. The "optional" part is whether my app *does* anything with that event; I shoudl always *get* the event. [15:33] or the mir screenshot tool shouldn't add _60Hz to still image shots [15:33] seems like a bizarre change to me [15:33] aquarius: but if it doesn't do anything, then nobody did anything with that notification [15:33] it shouldnt add a frequency at all [15:33] i still dont see what that is for [15:34] Chipaca, right. Which is the app's fault. Apps that are currently the open foreground app need to deal with notifications themselves; the OS should not show them. That's what I think. Others may differ. :) [15:34] aquarius: lucio_: if that is the consensus, I'm fine with it being not-optional. Mo' YAGNI mo' fun. [15:35] Chipaca, "I think it" is not really a consensus :) But you have my opinion now :) [15:35] yeah, "i think" [15:35] * Chipaca nods [15:35] dont know how to know [15:36] lucio_: one way is to implement the minimal one and wait for the feature requests [15:36] aquarius: is the intermediated delivery model good enough for your needs? [15:36] Chipaca: +1 [15:36] nah. This is new stuff for Ubuntu. You need, in my opinion, to set a strong consistent guideline on how you want people to work with and deal with notifications, so that everyone copies it and people who do it wrong look like they're falling behind. [15:37] lucio_, explain "the intermediated delivery model"? [15:37] aquarius: your app wont know about it until the user clicks on it [15:38] mterry, poke ... [15:38] lucio_, hrm. Then I think no. :) [15:38] aquarius: .more() [15:39] ogra_, hi [15:39] aquarius: isnt this the iphone model? [15:40] mterry, i'm just fighting with a nasty bug (session doesnmt start when home is full) ... while i will be able to fix the session startup i was wondering how we can allow the user to make emergency calls still ... and i assume you have some way to do that in your split greeter world where the user isnt even logged in ? [15:40] lucio_, Chipaca, I can think of three entry points here. (1) my app is running and foreground, and a notification arrives. I believe in this situation that an event should fire in my app, somehow; it is my responsibility to *listen* to that event. [15:40] aquarius: with you so far [15:41] yup [15:41] lucio_, Chipaca, (2) my app is not running and a notification has arrived. I tap that notification (either at the moment it arrives in a bubble or in some sort of "previous unacknowledged notifications" list. The OS then starts my app. My app should, on startup, have a way to request all notifications that it hasn't already requested. [15:41] (and obviously I should *do* that on startup) [15:41] ogra_, well split greeter mode is still the simple case of no lock -- so emergency dialing isn't a problem yet. But the greeter has its own home dir. Can we allocate some space to it via mount config? [15:42] aquarius: yup. [15:42] mterry, well, i assume the homedir doesnt live under /home, right ? [15:42] lucio_, Chipaca, (3) my app is not running, and a notification (or many) have arrived. I start my app from the Dash. This is the same as case 2 -- I should have a way to request all notifications that I haven't yet requested, and I should do that on startup. [15:42] ogra_, right. /var/lib/lightdm [15:42] aquarius: correct [15:42] either we deliver notifications when they arrive and on app start, or we signal "more notifications available" and have the app fetch them [15:43] mterry, yeah that wont help much if the autologin ran ... [15:43] (on signa and startup) [15:43] ogra_, right now temporary writable space [15:43] lucio_, Chipaca, the reason (2) and (3) are separate is that one potential solution here is that in case (2) when I start my app from a notification, the startup should pass the notification *to* the app startup process. I think that that's wrong, because it will discourage peolpe from just checking for notifications on startup/resume in the proper way. [15:43] mterry, sad, i was hoping you have some way that makes dialer-app run without needing to log or write to disk otherwise [15:43] thanks though [15:43] ogra_, there was talk of making the dialer-app a qml plugin [15:44] would that help in case where you have no writable diskspace ? [15:44] aquarius: the app should know if it was started from the user interaction with the notification or from the dash [15:44] why? [15:44] ogra_, I don't know. I'm not sure what the problem is with launching it right now (like, I don't know what it writes to). But dialer-app has a special emergency mode -- presumable we can just make it not write in that case [15:44] actually, so you can know *what* they pressed on. Yes. [15:44] Chipaca: also, when starting the app, the app should be able to clear notifications that are displayed [15:45] lucio_: agreed on "should". [15:45] lucio_: not promising that yet :) [15:45] Chipaca: then we need a clear story so that we dont have those around and have the user hand cleaning notifications [15:46] but it may be worth passing, maybe, just the notification ID or similar, and only providing a "fetchAllUnrequestedNotifications()" API function, *not* one to just get one notification by ID. Then you have to fetch them all and find the one that they clicked on in the list, which is a good encouragement to deal with all notifications, not just the clicked one. [15:46] fetchAllUnrequestedNotifications() would clear them from the notification cache, surely? [15:46] mterry, yeah, that would help ... the issue atm is that the session doesnt start at all when /home is full, i cant really restrict the user from filling it up ... i can change the session mgr in a way that it only writes to tmpfs dirs but that wont help with apps ... thats why i ask about the dialer since that still needs to function even with full home === mzanetti is now known as mzanetti|otp [15:47] aquarius: depends, clicking on one notification to solve one issue might not mean the other issues are solved [15:47] aquarius: lucio wants us to remove them from e.g. the messaging menu once the apps have them [15:47] once the apps decides to do so [15:47] W/E [15:47] :) === dandrader|afk is now known as dandrader [15:47] w/e? [15:48] lucio_: shorthand for "yes sir, anything you say sir" [15:48] lucio_, no, but it does mean that the app *knows* that that notification occurred. At that point, the OS doesn't need to keep the list of them around any longer; it's the app's responsibility now [15:48] ell [15:48] aquarius: so the app should re notify the user? [15:48] lucio_, and the OS can be sure that the app knows about that notification because the app is only allowed to request all outstanding notifications, not just one specific one. [15:48] the app can crash between getting them and storing them [15:49] lucio_, the app can do what it wants. You can't know what an app should best do with a notification. [15:49] but we can ignore that for now i guess [15:49] bzoltan, btw, that haptic MP should also help with ToyKeeper's pet bug 1290201 [15:49] bug 1290201 in dialer-app "haptic feedback / vibration audible during calls" [Undecided,New] https://launchpad.net/bugs/1290201 [15:49] at least a little :) === mzanetti|otp is now known as mzanetti === chihchun is now known as chihchun_afk [15:56] mmh, i like the model of start with ID of clicked notification, require get all, clean outstanding notifications on get_all [15:57] Chipaca ^ what do you think? doable? [15:57] lucio_: very [15:58] lucio_: if we want to guard against the app crashing we need to do more work, but i think we don't [15:58] yeah [15:58] lucio_: (because the app should be able to, at a bit more cost, go get whatever info it was being sent in notifications if it lost them) [15:59] makes sense [15:59] aquarius: ^ would that work for you? notice we dont offer the "put notification in lock screen" [15:59] feature [15:59] but that's more a failure of the notification framework we have on the phone, than of *push* notifications itself [16:00] whoops [16:00] Chipaca: more or less, its also a feature that we dont let you do everything [16:00] lucio_, Chipaca, agreed. There may be, currently, no way to see a list of notifications you've got but not acknowledged yet if you failed to tap them when they appeared, but I bet we'll grow one at some point. [16:01] lucio_: I'm assuming that when that feature exists, we'll be exposing it to push clients [16:01] Chipaca: what i mean is, if you replace our current lock screen with your own, you still need us to allow putting notification in there [16:02] and that wont change [16:02] lucio_: ok === jhodapp is now known as jhodapp|lunch [16:03] lucio_: aquarius: anything else? otherwise I'll go push kids through homework [16:03] i am happy [16:04] Chipaca, I don't have anything else -- my primary goal was to push (!) the idea that apps can ask the phone for notifications that the phone knows about rather than being forced to hit its own internet server for them [16:04] aquarius: well, we're already required to keep the id around for the server, so keeping the whole hting is very little more work :) [16:06] Chipaca, I have one little extra thing, which is that it would be nice if I could include more in a notification payload than just the text which is displayed to the user [16:06] aquarius: it's a json document :) [16:07] Chipaca, so I can send a notification to a messaging app like { display_to_user: "Chipaca says: hi!", payload: { from: "user101204", user_display_name: "Chipaca", sent_at: "20140705T170000" }} or similar [16:07] aquarius: quite [16:07] Chipaca, ah, cool, so it already *is* that? and there's a specific field which is "the thing which gets displayed in the notifiy bubble"? [16:08] Kaleo: do you know of any good tutorials explaining how to properly use WorkerScript and Loader to optimize an app? [16:08] aquarius: some fields we interpret as to how to notify the user, the rest , up to size limits, is up to you [16:08] aquarius: well, it isn't implemented, but yes. There's even a first pass of a draft at a table of the fields that are relevant [16:08] lucio_, cool, so what I wanted is already the plan. I like it when that happens ;) [16:08] :) [16:08] aquarius: lucio_: um. toplevel fields are limited, but there's a "payload" one [16:09] Chipaca: even better [16:09] doesn't matter how the json is organised -- pick whatever's best for you. Just as long as I can add data which is meaningful to the app but not displayed to the user :) [16:09] aquarius: lucio_: but that might just be a way to say "nuh, you were using unsupported fields, and that field you were using now means stuff so that's why your app broke", although tvoss would rather we enforce it :) [16:10] aquarius: I even support the idea of notifications that aren't displayed at all (but am unsure whether that is actually useful :-) )0 [16:10] Chipaca: thats why payload works, right? [16:10] Chipaca, fair enough -- define a "payload" field and promise I can put what I like in it, and filter all top-level unknown fields on the server [16:10] lucio_: yep [16:11] I personally think that notifications delivered while my app is foreground should not get a notify bubble, but I can see the other side of that argument. === bfiller is now known as bfiller_afk [16:12] I'm not sure that I can think of a reason to deliver a notification which doesn't get a bubble even if my app is not foreground, but if it turns out to be important then you just document a dont_show_a_bubble:true top-level field and have notify-osd pay attention to it ;) [16:13] ok, i'm off. bbl! [16:13] as am I, actually -- cheers for the time, lucio_ and Chipaca! [16:13] cheers [16:13] thanks [16:13] sergiusens: hi, I've just had my attention drawn to the nuntium package in NEW. Why is this using golang-go? The plan as of last October was that go packages would be built using gccgo [16:16] beuno: the store is only returning 100 clicks when queried is this intentional? [16:16] sergiusens: also, what are you expecting this upstart job to do? because it appears to have a race condition [16:16] on 5th may it returned 281 for me [16:18] sergiusens: finally, debian/nuntium*.lintian-overrides are absolutely wrong. It's not appropriate to suppress lintian messages about things that should be fixed in the package [16:22] mzanetti: ping [16:22] hi mhall119 [16:22] mzanetti: hey, are you aware that Tagger no longer works? [16:22] mhall119: no, I wasn't [16:22] mhall119: thanks for the report [16:23] will check it out [16:23] mzanetti: http://paste.ubuntu.com/7411305/ is the log [16:23] mhall119: i replied to your store upload of tagger a while back btw [16:23] hmm... I think I've fixed this already [16:23] when I select to scan a code, it'll open the camera and give me a preview for almost a second before crashing [16:23] yeah. I know this issue [16:23] and I've fixed [16:23] popey: my upload? [16:24] I mean, mzanetti [16:24] oh, ok [16:24] ah there it is [16:24] yep... didn't pass the store review and I forgot about it [16:24] sorry popey [16:24] np ☻ [16:24] your app ☻ [16:25] * popey goes to harras children into eating vegetables. [16:25] I gotta have more QR code [16:34] Hi, Anything new about MOTO G port? [16:38] Anybody working on ubuntu-touch on moto-G handset? [16:38] check the devices wikipage [16:38] !devices | grepped [16:38] grepped: You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices [16:38] if someone works on it he should have added it there (and a way to contact him/her) [16:44] ogra_, ping [16:45] cwayne_, i'm here [16:46] mhall119, sorry, not from the top of my head [16:46] and from the bottom ? === chihchun_afk is now known as chihchun [16:48] ogra_, you have a script to create webapps right? [16:48] cwayne_, well, kind of ... its buggy and all but i can give it to you ... [16:49] ogra_, works for me :) [16:50] cwayne_, people.canonical.com/~ogra/webapp-script.tgz [16:50] just a bunch of templates ... you still need to create/find an icon [16:50] (there is a README, tell me if something is missing from it) [16:54] ubot5, thanks. Which device are you working on? I have background of QT/C++ under linux, is it sufficient to start my own port? [16:54] grepped: I am only a bot, please don't think I'm intelligent :) [16:55] ubot5, god! you are a bot, what IRC client have hacked? [16:55] grepped: I am only a bot, please don't think I'm intelligent :) === dandrader is now known as dandrader|lunch [17:05] slangasek: wrt to overrides; can I just leave them be? afaik, there are no changelogs for any of these packages [17:06] sergiusens: I don't really care about the override for the changelogs, but the other two overrides are hiding reports of real bugs - one a policy violation that we don't care too much about (manpage), and one that's a security issue (lack of hardening due to use of golang-gc) [17:06] slangasek: wrt to gcc-go and golang-go all I know of is a current internal thread going on; but I can try and switch; I guess more packages would need this too (usensord, ubuntu-push-notifications) [17:07] slangasek: what's the race condition in the job? [17:07] slangasek: I'll drop the overrides then [17:07] the internal thread is that lots of people have lots of opinions; but the agreed approach TTBOMK is to use gccgo here [17:08] I'll check what I have to export to get it to build with gcc-go and see how it works out [17:08] sergiusens: the race condition is that it starts on x or y, and the pre-start script stops the job if it's started in response to x and y hasn't happened yet; if y happens while the pre-start script is running, the job will stop and never be started [17:08] sergiusens: thanks. In the meantime, I think I'm going to reject this package from the queue; I'm around today if you want a second set of eyeballs on the gccgo stuff [17:09] slangasek: ack, thanks === jhodapp|lunch is now known as jhodapp === bfiller_afk is now known as bfiller [17:41] bzoltan, hey, didn't the webapp template in ubuntu-sdk used to automatically create the manifest.json? [17:48] kenvandine: if you still want suggestions for pathwind, pass --fullscreen to qmlscene [17:48] dobey, i need to upload a new build, it doesn't use qmlscene anymore :) [17:49] but it does fullscreen [17:49] ah ok [17:49] made some changes for building on android and ubuntu [17:49] dobey, i also fixed the icon :) [17:49] yay :) [17:49] i'll publish a new version soon [17:49] haven't had anytime this week [17:50] heh [17:51] * ogra_ wonders what happens with --fullscreen on idle [17:51] the "jetpack" behavior also makes it difficult to play [17:51] ogra_: idle? as in screen is locked? [17:51] does it properly lock the screen ? [17:51] yes [17:52] dobey, define difficult? [17:52] idle as in idle [17:52] :) [17:52] (not activly locking, just leaving it alone) [17:54] kenvandine: complete lack of control. i can't keep the character in the middle of the screen to dodge things for example. so sometimes it is impossible to avoid getting hit by something [17:55] ogra_: anyway, yes it locks properly :) [17:55] awesome [17:55] though i don't really understand why fullscreen isn't the default [17:56] popey, hey, are there plans for the core apps to switch to the new headers? [17:56] also, online accounts is making my life miserable right now :( [17:56] dobey, because everything is fullscreen on the device :) [17:56] kenvandine: no it isn't [17:56] what? === dandrader|lunch is now known as dandrader [17:57] kenvandine: the phone top panel is there in pathwind for example. --fullscreen puts the app on top of that panel [17:57] oh that [17:57] everything is only "maximized" on device [17:58] some things are full screen, but not everything [17:58] yeah, it'll be full screen [17:58] and edges don't rotate [17:59] nor does the top panel [18:00] which gets really awkward if you hold the phone upside down :P [18:00] yeah [18:01] what the heck process is the one that is loading the qml plug-in when i try to add an account? [18:01] ubuntu-system-settings-online-accounts [18:01] i think [18:02] i don't think that's actually a process is it? [18:02] it's provided by that package [18:02] nope [18:03] /usr/bin/online-accounts-ui ? [18:04] i think that's just the list of accounts ui [18:04] no [18:04] that is also what handles auth [18:04] oh, actually signon-ui might still handle the auth [18:05] how is it being started? [18:05] signon-ui? [18:05] or online-accounts-ui? [18:05] also the fact that all this is separate processes is really annoying [18:05] online-accounts-ui is started when it's needed by system-settings or when there is an api call using the appaccess api [18:06] kenvandine: well, whichever one is actually loading the qml plug-in [18:06] because i'm trying to run with an env var which is woefully not getting passed to whatever is loading the qml plug-in that needs the env var :( [18:06] system-settings might load the plugin itself, if you are going in that way [18:06] i don't recall [18:07] and i think if you use appaccess it uses the other [18:07] apparently it's not [18:07] i think it's started by dbus [18:07] it might get spawned when it's needed by system-settings too [18:09] sigh, this is a horrible mess [18:10] maybe i should just reboot again [18:10] brb [18:14] Saviq, hey my phone UI is frozen and unity8 is running at 100%, anything I can do to debug? [18:15] kgunn, ^^ === chihchun is now known as chihchun_afk === dandrader is now known as dandrader|afk [18:47] ogra_: https://bugs.launchpad.net/click-package-index/+bug/1317233 [18:47] Ubuntu bug 1317233 in Click Package Index "Store only returning 100 results" [Undecided,New] [18:47] confirmed === Tassadar_ is now known as Tassadar [18:51] cwayne_: it does create when you create the click package ... the manifest files are editable on the Publish page [18:53] bzoltan, ah, ok === dandrader|afk is now known as dandrader [19:01] pmcgowan: do you still have it in that state ? [19:03] kgunn, no but I entered a bug https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1317235 [19:03] Ubuntu bug 1317235 in unity8 (Ubuntu) "UI frozen and running at 100% CPU" [Undecided,New] [19:03] not sure the bt is useful [19:06] pmcgowan, yeah, not gonna help much... that's basically the mir main loop, and we can't be sure that's what was looping, either [19:07] drat [19:07] pmcgowan, I'll try and come up with a plan on what to do in a case like this - we basically need something to attach to the process and profile it for a few seconds to see what is spinning [19:07] pmcgowan, I'll have to ask around [19:07] ok [19:19] ogra_: https://code.launchpad.net/~rsalveti/project-rootstock-ng/adding_option_distro_series/+merge/218502 [19:19] in case you didn't see it yesterday [19:19] minor fix [19:38] Saviq, hey so what module controls the rotation? sorry I thought it was the shell? [19:38] maybe ricmm can tell me ^^ [19:54] hi guys, its late but i got an issue during boot. While moving the virtual dirs proc and sys to root, im not further able to connect through adb. Maybe someone can give me some hint to fix it?? [20:02] pmcgowan, it's via sensors directly [20:02] pmcgowan, there's a sensor of type "orientation" [20:03] Saviq, wheres the logic that detects a rotation change and changes the window [20:03] pmcgowan, which takes values from different sensors into account and comes up with a {0,90,180,270} reading [20:03] pmcgowan, SDK [20:03] pmcgowan, in MainView [20:03] so its my bug, damn [20:03] :D [20:03] yeah was just looking for it there [20:03] thanks [20:03] pmcgowan: that it's too sensitive? [20:04] pmcgowan, but the fact that it's too sensitive is a sensors bug I'd say [20:04] popey, yes [20:04] I dont know [20:04] Saviq, so where is the logic that says "45 degrees is a rotation change" [20:04] thats where the bug is [20:04] also we relayout when the phone gets tapped or josstled [20:05] not sur ehow that happens [20:05] indeed [20:05] pmcgowan, accelerometer [20:05] nothing to do with a rotate [20:05] i have mentioned this in the past too.. pretty sure there's a bug [20:05] Saviq, so which package? [20:05] qtubuntu-sensors? [20:05] bug 1286150 [20:05] pmcgowan, I meant acc is why it thinks it rotated when it's jostled [20:05] bug 1286150 in qtubuntu-sensors (Ubuntu) "Rotating detection is very sensitive mako #212" [Undecided,New] https://launchpad.net/bugs/1286150 [20:05] pmcgowan, and yes, there [20:06] Saviq, I cant find the logic there yet [20:06] or I dont understand it [20:06] pmcgowan, altho I'm not entirely sure we can amend the logic at that level (i.e. it might come from hardware and/or android parts) [20:07] pmcgowan, ricmm knows best about those [20:07] Saviq, ok will ask him manana === dandrader is now known as dandrader|afk [20:16] dobey, pathwind with the better icon and fullscreen is in the review queue, i'll call this the "dobey release" :) [20:16] dobey, thanks for the feedback! [20:17] sure [20:25] slangasek: can you please take a look at https://code.launchpad.net/~sergiusens/nuntium/packaging_fixes/+merge/218693 ? [20:32] kenvandine: approved [20:33] rock on popey! [21:07] sergiusens: looking [22:01] tedg, can you help test indicator-messages in split mode with me? [22:01] tedg, I'm wondering if I'm seeing odd behavior [22:01] mterry, For a bit, I need to run in a few. [22:01] tedg, we can pick up tomorrow [22:01] mterry, I usually grab a webirc session and log into #tedtest [22:02] tedg, say what? [22:02] mterry, For testing the messaging menu, so that I can ping myself [22:02] tedg, oh to get incoming messages [22:02] tedg, that's clever [22:03] tedg, that will help decide if I'm seeing a problem in telephony or indicator-messages [22:03] Having issues flashing nex4.. Hangs on the "google" boot screen. I can adb shell (into busy box) and I can get to the ubuntu recovery console... but I cannot boot :-/ tips? [22:03] tedg, how do I set that up on the phone? [22:03] mterry, Hmm, not sure. [22:04] mterry, But reality is that indicator-messages has no real knowledge of greeter or not. [22:04] mterry, So it's not getting info from the session. [22:04] tedg, fair... so likely telephony-service [22:05] mterry, That would be my guess, but it also conveniently results in no bug for me :-) [22:05] josharenson: boot to recovery and flash again with ubuntu-device-flash? [22:06] dobey, tried 3 times [22:06] tedg, I just get weirdly inconsistent results is all. Anyway. Tomorrow [22:06] mterry, Sounds good. [22:06] dobey, any use in reflashing android and trying everything again? [22:06] sergiusens: commented on the MP [22:07] josharenson: no idea. reboots and/or reflashing have always worked for me when i hit a similar issue [22:08] doeby, guess I'll keep trying the same things if this isn't a known issue. Thanks though. [22:10] sergiusens: fwiw I think it's a gccgo-go bug that this doesn't work: http://paste.ubuntu.com/7412873/ [22:11] sergiusens: (and we should file that bug, but not be blocked by it) === dandrader|afk is now known as dandrader [22:28] sergiusens: so your changes look fine to me in principle, but as I still can't get the package to build... [23:27] slangasek: hmm, let me give it another spin, it did build fine for me in a schroot; I'll bootstrap a new one to reverify [23:30] sergiusens: trusty or utopic? [23:30] slangasek: yeah, my bad was I did trusty... [23:30] ok, please let me know if you see the same issue then [23:31] slangasek: so building twice works :-/ http://paste.ubuntu.com/7413178/ seems gccgo isn't building it's deps before the target [23:32] oh, interesting [23:32] I guess you were building in place which is why it worked for you [23:32] so yeah, fun [23:33] slangasek: nope, I build multiple ways :-) bzr bd; bzr bd -S && sbuild *dsc; go build; go build -compiler gccgo [23:33] I'll try a utopic build right now [23:34] ok [23:36] fwiw I usually don't go install that much and just go build [23:52] slangasek: yeah, fails on utopic for me too :-/ [23:53] I'll give this a triple check from your comments [23:54] blast; I know what it is :-/ [23:54] will fix in a bit