[01:06] pitti`: help [01:06] https://bugs.launchpad.net/ubuntu/+source/share-app/+bug/1360670 [01:06] Ubuntu bug 1360670 in share-app (Ubuntu) "drop from archive" [Undecided,New] [01:06] https://bugs.launchpad.net/ubuntu/+source/libhud-qt/+bug/1360671 [01:06] Ubuntu bug 1360671 in libhud-qt (Ubuntu) "drop from archive" [Undecided,New] [02:53] Howdy [02:55] I have a question about setting up APN, specifically because I’m on an MVNO and I can get the settings, but I’m not finding the terminal commands to accomplice this. [02:55] I’ll keep googling, but so far my search terms aren’t working very well [02:56] **accomplish [02:57] if anyone is able to help I would appreciate it :) [03:10] also what is the default password? Search results say it’s “phablet” but that’s not working. === chihchun_afk is now known as chihchun [03:48] Can anyone elp me how to uninstall Ubuntu Touch from my Nexus 7 tablet [03:57] * hackersarchangel sighs [03:59] any advice on what I need to change to get SMS/Data working? === _salem` is now known as _salem [04:23] I have been editing the gsrp file under /var/lib/ofono// but that doesn’t appear to be helping. [04:37] is the GSRP file the same syntax as the android one? if so I can just drop that into place. [04:38] I know the android one is xml based, and this doesn’t appear to be but you never know [04:46] Wellark: what's up? [04:46] Wellark: the two "drop from archive" bugs? [04:46] * pitti does them [04:49] ... [04:53] pitti: I have had trouble to find someone to take care of them [04:53] so thanks! [04:54] Wellark: ah, just subscribe ~ubuntu-archive, then it usually gets taken care of [04:54] they were supposed to be dropped for 14.04 already, but got burried [04:54] Wellark: should that go from RTM, too? [04:54] pitti: they are not seeded to rtm [04:55] ack, indeed [04:58] * Wellark hopes LP would allow to set project status to "abandoned" [04:58] oh, well.. [04:59] Gentlemen, I understand you are busy working on the project, but can I snag a moment of your time? [05:02] hackersarchangel: sure [05:02] I’m having some trouble getting SMS/3G to work, and I want to make sure I’m editing the right file [05:03] also wondering what the default password is for ‘phablet’ seeing as I tried ‘phablet’ and this not working [05:03] I’m editing /var/lib/ofono//gsrp [05:03] hackersarchangel: the password is what you set the passcode to be in first run wizard [05:04] and I have a Nexus 5 on Sprint using an MVNO, and I have to change my APN in Android. [05:04] Ah! Thanks. Never even occurred to me to try that. [05:04] hackersarchangel: under /var/lib/ofono [05:05] hackersarchangel: I'm actually working on a graphical editer to modify the broadband settings ATM [05:05] Ok so I am editing the correct file. That at least helps, I now just need to figure out what the settings are supposed to be aside from just the mmsc change. [05:05] Sweet. [05:05] hopefully get that done and landed today [05:06] hackersarchangel: if you edit the files directly [05:06] you need to stop ofono while editing [05:06] I’m using the nightly but I’m not using Ubuntu One, so I’m not sure if the Multi-Rom will see that update soon or not. [05:06] I’m editing it from within Android and rebooting. [05:06] $ sudo stop ofono [05:06] [05:06] but that saves me the headache of rebooting xD [05:06] $sudo start ofono [05:07] Sir, I thank you for helping me out, this is a great start. [05:07] Now I know all my Google searching was right, I’m just not finding the right settings. [05:07] hackersarchangel: you might also have to manually tell NetworkManager to activate the new settings [05:07] even from a reboot: [05:07] ?? [05:08] so after editing /var/lib/ofono/{IMSI/gprs [05:08] do [05:08] nmcli c [05:08] and locate a line there that says something like /{IMSi}/contextX [05:09] and do $ nmcli c up id /{IMSI}/contextX [05:10] as I said, I hope to land the graphical editor today [05:10] I have a bunch of those. Tell you what, I’ll start from scratch and see what happens. [05:10] which will make all of the pain go away [05:10] Lol [05:10] Ok and should I update from within the GUI or continue using Multi-Rom to update it? [05:10] hackersarchangel: you can just do $ sudo stop ofono [05:10] $ sudo rm -rf /var/lib/ofono [05:10] sudo start ofono [05:11] and you have a "clean" setup [05:11] sweet. [05:11] Thanks a bunch man. [05:11] hackersarchangel: np. my pleasure [05:11] if it doesn’t work I’ll stick to stock Android as my daily driver but I’m interested to see how this pans out. [05:12] hackersarchangel: it should work if you enter the correct values under /var/lib/ofono/{IMSI}/gprs [05:12] you can get those values from your operator [05:12] for 3G [05:12] Right. [05:13] the only relevant value really is the AccessPointName [05:13] for MMS you need the message center and friends [05:13] The only thing they are giving me is the APN address and I think I’m fudging up too many pieces. [05:13] APN address is usually enough [05:13] unless the operator is doing something nasty [05:14] No, it’s Ting as my provider [05:14] hackersarchangel: what is your operator? [05:14] which country? [05:14] US [05:14] they are an MVNO of sprint, and the only thing I change on Android is the MMSC [05:14] to get MMS working, the 3G/LTE kicks in when doing PRL/Profile updates [05:15] been looking at some of the APN files for flashing and I am probably just mucking it up by changing too much. I’ll tinker and figure it out :) [05:16] … it just told me I had a read-only FS. [05:16] hackersarchangel: "Ting" you said? [05:16] Ting. [05:16] that's not even par tof https://git.gnome.org/browse/mobile-broadband-provider-info/tree/serviceproviders.xml [05:17] Not surprised. [05:17] hackersarchangel: hmm.. is it a CDMA phone? [05:17] as we don't currently support CDMA [05:17] They’ve only been around maybe 2 years? [05:17] OH [05:18] Wait no I’m using a Nexus 5 with a SIM [05:18] it might still be CDMA modem on it [05:18] but I think that it is CDMA [05:18] ah. [05:18] Well rats. [05:18] ok. then you are out of luck for now' [05:19] sorry ;( [05:19] So I’m guessing Sprint is a no. [05:19] Just in the general sense. [05:19] Because I can use the phone on it but no SMS. [05:20] Which if I had to live without 3G I am fine with that, but no SMS would make a lot of my friends go bonkers. [05:20] hackersarchangel: ok. normal SMS should work if you can place phone calls [05:20] but MMS are more tricky [05:20] hackersarchangel: do you have adb access? [05:21] could you provide the output of [05:21] /usr/share/ofono/scripts/list-modems [05:21] just slam it to pastebin.ubuntu.com [05:21] so what do I type to get that? I know what you mean by ADB. [05:22] hackersarchangel: plug in usb cable, type "phablet-shell" on your terminal [05:23] that should log you in to the phone via SSH connection [05:23] then you can just type /usr/share/ofono/scripts/list-modems and hit ENTER' [05:24] hackersarchangel: I found this: [05:24] http://www.4gtricks.com/2013/08/sprint-apn-settings-for-android-phone.html [05:24] we could try those values [05:24] Service org.ofono not found on DBus: org.freedesktop.DBus.Error.ServiceUnknown: The name org.ofono was not provided by any .service files [05:24] wait. [05:24] Derp'd [05:24] hang on [05:25] you probably did not log in to the phone [05:25] that just tells that your desktop/laptop is not running ofono :) [05:25] actually I forgot to restart it === _morphis is now known as morphis [05:25] ok so it’s saying I have a read-only file system [05:25] and I can’t delete the /var/lib/ofono folder [05:26] which is what I was doing then we continued onwards and I became ahead of myself [05:26] so that is another issue altogether I suspect. [05:28] http://pastebin.ubuntu.com/8156257/ [05:29] Well I’m about to lose my battery, so I’ll go ahead and see if I can get it working on my own, but if not I’ll come back tomorrow night and just hang out here while working on it. [05:30] hackersarchangel: what is the contents of /var/lib/ofono? [05:30] $ sudo ls /var/lib/ofono [05:30] drwx------ 2 root root 4096 Aug 27 00:27 310120020448886 [05:30] drwx------ 2 root root 4096 Aug 27 00:27 310120020448886-3 [05:31] ok. that -3 is weird [05:31] one second [05:31] hackersarchangel: what is the contents of /var/lib/ofono/310120020448886/gprs ? [05:32] after you have stopped ofono [05:32] just a second [05:32] you should be able to do: $ sudo rm -rf /var/lib/ofono/* [05:33] I just did that, it cleared out the folder but it did bring back the -3 [05:33] I’ll paste bin it [05:33] http://pastebin.ubuntu.com/8156283/ [05:33] hackersarchangel: please paste both /var/lib/ofono/310120020448886/gprs and /var/lib/ofono/310120020448886-3/gprs [05:34] there isn’t one in -3 [05:34] just a version file [05:34] which is blank [05:35] hackersarchangel: ack [05:35] hackersarchangel: looking at your latest paste [05:35] ack? [05:35] there are a lot of contexts pre-populated [05:35] Yeah. [05:35] hackersarchangel: ACKnowledged :) [05:35] lol [05:35] nice. [05:36] hey I’m at 1% battery life, so my MacBook is going to shut off here in a minute. [05:36] hackersarchangel: try [05:37] $nmcli c up id /310120020448886/contexts8 [05:37] *context8 [05:38] that should get you 3G [06:42] is it possible to use touch on cheap quadcore android phone? === asac` is now known as asac [06:45] good morning [06:45] o/ [06:45] hello dholbach [06:49] hi asac :) [06:50] morning [06:50] anyone had try to install touch on a cheal china quadcore phone ? [06:51] pitti, seb128: are you familiar with how .desktop files are translated? Or do you know someone who is? :-) [06:53] mardy, sort of, what's the question? [06:54] mardy, usually they have translated keys merged back in (like the source has a .desktop.in and generate the .desktop including the translations at build time) [06:54] mardy, we strip the translations and use an X-Ubuntu/GNOME-Gettext-Domain= key to load those from gettext in Ubuntu [06:54] seb128: exactly! [06:55] seb128: so, the problem with this is that g_app_info_get_display_name() doesn't translate them [06:55] hum, are you sure? [06:55] seb128: or maybe I don't have the translations, I should probably double check that... [06:56] seb128: I was just looking at this, but I didn't check yet if translations are there: https://bugs.launchpad.net/ubuntu-system-settings-online-accounts/+bug/1356939 [06:56] Ubuntu bug 1356939 in Online Accounts setup for Ubuntu Touch "List of authorized applications aren't localized" [High,New] [06:56] seb128: and we are using g_app_info_get_display_name() to get the application name [06:57] mardy, the thing is that the gettext thing doesn't work for clicks, since they don't have their .mo in the standard gettext location [06:57] so we need to merge back the inline translations for those [06:57] pitti was looking at some of those issues yesterday I think [06:58] dbarth__: my gmail webapp is kinda buggy; seems to not remember my cookies (e.g. when i start again i have to log in all over); also i cannot switch accounts anymore [06:58] seb128: these are not clicks, it's the sync daemon and account-polld; but I see that .mo files are missing, so it might be their bug after all [06:58] dbarth__: want me to file a bug or is this already known? [06:58] mardy, right [07:01] sergiusens: you don't have a bug tracker for account-polld? [07:02] mardy: ah nice, you are using GAppInfo? Our glib is patched to use gettext, but indeed that requires the .mo files to be installed in the standard paths (i. e. won't work for clicks) [07:02] mardy: but that mega-tracking but mostly applies to deb packages [07:03] mardy: there, the problem is that the .desktop files are missing the X-Ubuntu-Gettext-Domain field [07:03] pitti: yes, the information comes from libaccounts-glib, which is using GAppInfo, and then its API is wrapped in Qt and QML [07:03] mardy: ah, I meant bug 1318008, not your's (that might be differnet) [07:03] bug 1318008 in Ubuntu File Manager App "Core apps .desktop files do not include translated strings" [High,In progress] https://launchpad.net/bugs/1318008 [07:04] pitti: for account-polld, and sync-monitor, the X-Ubuntu-Gettext-Domain is properly set, but the .mo files are not there [07:17] ogra_: do you remember if there had been discussion about making our rootfs not ship any apps? [07:17] but rather move those into a "ubuntu default app pack" customization tarball or something? [07:18] ogra_: i vaguely remember that something like this was discussed at some point (3+ month ago); but can't remember what the outcome/arguments for doing/not-doing this were [07:18] asac: go ahead, i haven't noticed that myself [07:18] asac: latest -proposed? [07:19] dbarth__: no on rtm [07:19] rtm-proposed [07:19] so yeah kind of latest -proposed [07:19] ok [07:19] dbarth__: can w add gmail as a testcase to your webapp engine for landings in future? [07:20] asac: when you switch between accounts, or just plain forgets any of your authentication? [07:20] dbarth__: just simple: 1. log in; 2.log into second account; switch account; kill and start and see that you are still logged in and can switch [07:20] ok [07:20] dbarth__: kind of yeah. so i can log in and use it for one account [07:20] then i log into another account,but always end up in the first account [07:20] so the switch doesnt have effect [07:20] hmmm ok [07:21] if i kill app and start again it is logged out === dbarth__ is now known as dbarth [07:21] very weird... i am sure you will see if you try :) [07:21] dbarth: which package shall i file this against? [07:21] lp:webapps-core [gmail] [07:21] ping asac [07:21] bzoltan: dont ping, just speak :) [07:21] whats up? [07:21] need to split the project, but that's the one for now [07:21] asac: I am about to release a new UI Toolkit package and I see here http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html that the landing is blocked with "Not touching package due to block request by freeze" excuse. [07:21] dbarth: no ubuntu package? [07:21] or same? [07:22] asac: Mirv told me that it is because of the Edubuntu bet afeature freeze [07:22] err [07:22] asac: that sounds odd.. edubuntu blocking RTM [07:22] UITK is seeded in Edubuntu, so it's considered part of the freeze by default [07:23] bzoltan: not RTM, the RTM UITK release would be separate from utopic release.. it's still in a silo [07:23] bzoltan: its wrong indeed. once slangasek gets up he will fix this i am sure; until then lets see if we can get someone else in -release to pay attention [07:23] Mirv: who is the driver behind edubuntu? [07:24] Mirv: is it OK to land in RTM something what is not in Ubuntu? [07:24] asac: I'm not sure if they have a dedicated driver in the community, but the release team controls the freezes anyhow [07:25] Mirv: who is in release team anyway? [07:25] Mirv: i know the foundations folks, but they are all in portland [07:25] so looking for someone who mighit be awake right now [07:26] asac: https://launchpad.net/~ubuntu-release/+members [07:26] Laney: ^^ [07:26] help! [07:26] :) [07:27] ok shot mail to graber, steve and ian [07:27] bzoltan: but you're correct, as utopic landing is required currently before RTM landing, this makes community derivative blocking our rtm distro landing :S [07:27] not sure if he is debconf too... lets see [07:27] Mirv: can we see in the excuses that efverything else is green? [07:27] or does it not even test that? [07:28] Mirv: who says that community does not influence the Ubuntu phone project :) [07:28] asac: yes we can see, all autopkgtests have passed [07:28] if its all green we can claim that it is in and do the publish into rtm imo [07:28] to bandaid this [07:28] right [07:28] bzoltan: do you have yuour stuff in an rtm silo ready? [07:28] and QA sign off etc.? [07:28] if not, just do that and move [07:29] while you are done we might have resurrectredt aht. if not we can workaround given the argument above until they show up [07:29] asac: it is in the rtm ilo5 and ubuntu silo9 [07:29] bzoltan: and already signed off by QA for RTM? [07:29] there's RTM silo, but it's not on the spreadsheet as robru copied it manuall, and it does not have QA signoff yet [07:29] that's one thing for today, making those RTM silos not added manually to spreadsheet visible to QA team needing signoff [07:29] Mirv: ok, can you help getting that QA sign off setup and arranged? [07:30] asac: sure [07:30] Mirv: well, that manual step feels like a lesser evil [07:30] :) [07:30] and a correction, there's a manual entry on the spreadsheet [07:30] if thats what it takes let dos do it to keep the procss moving [07:30] bzoltan: please mark the line 38 tested for RTM, and let's get signoff from QA for that [07:30] right, so if bzoltan is done with testing then set it to done and get qa on it [07:30] let me know if there is any problem [07:31] yes, this should be alright now again [07:31] ok colin fixed the block [07:31] so business as usual [07:31] thanks [07:35] have you guys used ubuntu touch yet? [07:36] every day, yes [07:36] what devices have you install ubuntu touch on? [07:37] LeelooOnline: N4 is your best bet for now [07:37] asac, nope, i wasnt in that discussion [07:38] LeelooOnline: just get a used one from ebay and you will be happy :) [07:38] any larger devices? e.g. 11 inch? [07:39] LeelooOnline, if it isnt that urgent you could wait til the first ubuntu phone go on sale ... [07:39] *phones [07:39] thats tablet; you can try the N10, but we don't really put much focus on it, so it might be a bit rough :) [07:39] but should work in general. at least noone really complained to me yet [07:40] many people complain :) [07:40] just yesterday someone was here and did [07:40] not to me :P [07:40] i think the sidestage works far less than optimal :) [07:40] no, but to me [07:40] so it cant be that desastrous [07:41] would be too easy if there were no challenges [07:41] hehe [07:41] it definitely boots to a session :) [07:41] em ...I am looking for a tablet that can replace my giant laptop and running linux..... :-( [07:41] but as i heard it isnt much usable [07:41] LeelooOnline: nexus7 [07:42] yeah, the nexus7 works pretty well ... close to the quality of the nexus4 image i'd say [07:42] running nexus7 myself with touch, working smoothly [07:43] OK Thx guys. [07:43] but be careful to get the right n7 if you consider buying one [07:43] avoiding mali gpus is a good idea at least in the near future [07:44] the 2013 version or 2012 version [07:44] ?? [07:44] LeelooOnline: 2013 wifi [07:44] !devices | LeelooOnline [07:44] LeelooOnline: You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices [07:44] oh we have a bot that knows all that. nice :) [07:45] Thanx! [07:52] sil2100: wrt. bug 1357252, want me to copy the cgmanager fix from utopic to rtm? [07:52] bug 1357252 in cgmanager (Ubuntu) "systemd-shim fails to handle cgmanager being unavailable" [Undecided,Fix released] https://launchpad.net/bugs/1357252 [07:55] hey, i realllly need to run the OOBE tutorial thingy one more time [07:55] how can I do that? [07:58] pitti: yeah, that would be great :) [07:58] sil2100: done [07:58] \o/ [07:59] copy-package -b --from ubuntu -s utopic --to ubuntu-rtm --to-suite 14.09 cgmanager [07:59] Now just for tedg to re-release his u-a-l [07:59] for the record (not sure whether you can run this yourself) [08:00] I'm only a MOTU right now, so not really any power in this territory [08:07] sil2100, i think fro uinverse to universe copies that would still work for you [08:07] Probably, but not for cgmanager anyway [08:07] copy-package should be able to check permissions that fine grained [08:10] mardy: pitti ubuntu-bug account-polld [08:10] what are you doing up already ? [08:10] ogra_: yes, but cgmanager is in main [08:11] pitti, yeah, just mentioning it :) [08:11] asac`: wrt the discussion about removing click apps by default was rejected [08:12] sergiusens: I want to add account-polld to this existing bug https://bugs.launchpad.net/ubuntu-system-settings-online-accounts/+bug/1356939 [08:12] Ubuntu bug 1356939 in Online Accounts setup for Ubuntu Touch "List of authorized applications aren't localized" [High,Incomplete] [08:12] sergiusens: but it says that account-polld has no bug tracker [08:12] mardy: done [08:12] mardy, erm, you need to use the package, not the upstream branch [08:12] mardy: "Also affects distribution" [08:12] sergiusens: ah, silly me! :-) Thanks :-) [08:13] we generally dont use the upstram branches for distro bugs [08:13] mardy: on ogra_'s recommendation we did a project wide upstream disablement of the bug trackers to force them to be ubuntu-bugs [08:13] right [08:13] so that ubuntu-bug works with them ... and you cant really know if the bug you have isnt caused by a maintainer script error [08:14] mardy: I wonder what I'm missing to fix it though; I just followed dpm's recommendations [08:14] ogra_: in an ideal world; someone would log the ubuntu bug and then a gatekeeper might link it upstream [08:14] never happens :-P [08:14] since I can't really triage ubuntu bugs [08:15] yeah, its utopic [08:15] :P [08:15] ogra_: just makes triaging harder; === ValicekB_ is now known as ValicekB [08:15] bugs are a black hole anyways [08:15] why ? [08:15] they only get noticed for real real if they are too serious or if people make you notice it :-P [08:15] weather you filter by package or by upstream project shouldnt really make a difference [08:16] you can subscribe to packages for which you want to have all bugmail [08:16] ogra_: one more step in the _process_ ;-) [08:17] in setup, yes [08:18] sergiusens: right, what was the rational for not doing it? [08:18] please remind me :P [08:18] mardy: ogra_ I wonder how the mo files got removed; they were there (I use spanish on my phone now to help with translations) [08:19] sigh .. i thought paul disabled the known broken clock test [08:19] * sergiusens never used an os in a different language than english [08:19] no image test results at all today :((( [08:19] really? [08:19] clock breaks the whole image? [08:19] asac`, yes, there is a known issue with the clock [08:19] (app) [08:19] ogra_: but how can that break the whol etesting? [08:20] it cant download the test and falls over [08:20] ogra_: and the whole infra falls over becauuse of that? [08:20] phablet-test-setup apparently runs in the install stage [08:20] can we fix that the tests become availa gain? [08:20] asac`: here's cjwatson's response for that https://lists.launchpad.net/ubuntu-phone/msg06900.html [08:20] and pulls in all tests as the first step [08:20] ogra_: clock was updated? [08:21] asac`, we'll definitely talk about that in 10min in the meeting, yes [08:21] sergiusens, yep, clock-reboot replaces it [08:21] * ogra_ gets meeting coffee [08:21] sergiusens: ok thanks. seems that didnt take into account that we could have an ubuntu-standard customization tarball that we ship as our default image [08:21] * sergiusens checks jenkins [08:21] asac`: yes; that was how my proposal started [08:21] the new unity8 is soooo lovely [08:21] ogra_: wait, can we not just make the tests available somewhere? [08:22] * ogra_ was totally surprised when the apps properly restarted [08:22] sergiusens: ok. will talk to folks next week again about this :) [08:22] thanks! [08:22] Saviq, awesome work !! [08:22] ogra_, glad, we still need some visual feedback that it restarts [08:22] but the plumbing's there [08:22] dandrader's work [08:23] * asac` wants that :P [08:23] sergiusens: thx for checking what append to clock. feels like something we could fix at least to a poinmt that the image tesing doesnt fall over [08:23] happened [08:24] asac`: that's easy; just do phablet-click-test-setup --click $click per click [08:24] thanks popey [08:25] sergiusens: how? [08:25] sergiusens: we cant taouch infra for now [08:25] asac`: I don't really mind moving some apps to such a custom tarball if it actually exists (which should happen *before* any removal); my reply was because people seemed unaware of the flexibility that click offers and thought that that approach was the only way to have some core apps not show up for carriers [08:25] asac`: well; on infra [08:25] cjwatson: all right. lets talk next week. [08:25] sergiusens: cant we fix this without infra stitches? [08:25] cjwatson: I think your reply was good; but I can attest that asac`hasn't read it yet [08:25] i meann... just put the tests in right place [08:25] asac`: however if you're doing that you should check how things work with regard to people branching the standard custom tarball, since I expect any procedures that exist now will consist of creating one from scratch [08:26] asac`: the above and the linked post are all I have to say on the subject really :) [08:26] to be clear, if at all, this is a post RTM topic [08:26] as long as people are making intelligent use of the facilities available ... [08:26] asac`, dont bother ... it is known, we discussed it in the landing meeting last night and will look at it now too [08:26] (the broken test) [08:26] ogra_: well, we dont have image results [08:27] thats not good if we can fix this without having to wait for US to wake up, i dont see why we would wait, but *shrug* :P [08:27] asac`, right, and we have psivaa in the meeting to take a look for us [08:27] * asac` stays out of micro mode [08:27] cool [08:27] seems its covered [08:27] right ;) [08:27] we'll get results somehow [08:28] ogra_: just do what I said and skip clock [08:28] sergiusens, yes, i had expected that plars had set it up that way, since we talked about it and kenw it would fail [08:30] ogra_: how does it fail; with the latest clock app from jenkins I don't see how it's failing to retrieve [08:30] ogra_: unless trunk was changed (rebased) before uploading the click [08:30] sergiusens, feelk free to come to the meeting if you like :) [08:30] ogra_: so if you back out the rebase, it should be fine [08:30] ogra_: link? [08:30] https://plus.google.com/hangouts/_/canonical.com/landing-meeting [08:43] sergiusens: I am afraid that's what happened: trunk was rebased before uploading the click. My fault sry :/ [08:43] nik90: no worries [08:43] nik90: I just approved the click [08:43] sergiusens: however the new clock click package has been uploaded [08:44] sergiusens: sweet [08:44] Hi === Mooky_ is now known as mooky [09:05] pitti, hey, regarding bug #1318008 and webbrowser-app, please see my last comment, not sure how that would be fixed? [09:05] bug 1318008 in Ubuntu File Manager App "Core apps .desktop files do not include translated strings" [High,In progress] https://launchpad.net/bugs/1318008 [09:19] jibel, hey, can you still reproduce bug #1361159 ? [09:19] bug 1361159 in webbrowser-app (Ubuntu) "webbrowser crashes when playing videos from youtube" [Undecided,New] https://launchpad.net/bugs/1361159 [09:35] oSoMoN: ah, that one has X-Ubuntu-Gettext-Domain [09:35] oSoMoN: so I guess the problem there is that unity8 doesn't read that [09:35] oSoMoN: it's certainly much better to add the domain rather than statically adding the translatiosn to the .desktop indeed [09:36] oSoMoN: but from what I heard yesterday, the apps scope is supposed to learn about that [09:45] oSoMoN: but from what I heard yesterday, the apps scope is supposed to learn about that [09:45] sorry, -EFOCUS [09:45] pitti, ok, thanks [09:54] cjwatson: ping [10:01] marcustomlinson: at conference, very spotty availability, you would be well advised to leave a message rather than just pinging! [10:02] cjwatson: alright, question: is there a way to chain nested click hooks? What I mean is, I want to execute another hook from within the first click hook. Perhaps via the Exec line, passing on the id, short-id, user and home values? [10:03] cjwatson: or is there some "correct" way to have a hook defined in a manifest trigger more than one hook? [10:03] marcustomlinson: hook Exec commands must be idempotent, so you can just execute some other program if you want [10:04] marcustomlinson: if that's not sufficient, best mail me [10:04] bit involved for 3am :) [10:04] cjwatson: oh right, sorry === Guest8994 is now known as thelionroars [10:18] cjwatson: ah I see, you can have multiple hooks sharing one hook name, then all will execute for that name [10:20] marcustomlinson: ah yes, that's what I added that for [10:21] cjwatson: nice ;) === marcustomlinson is now known as marcustomlinson| [10:21] cjwatson, what are you doing online ? you should be in a bar sipping beer at this time, shouldnt you ? [10:21] :) === marcustomlinson| is now known as marcustomlinson [10:22] (or even in a bed) === ondra- is now known as ondra [10:23] I'm in bed waiting for a transition to finish since it's big enough to require manual supervision lest the archive break [10:24] heh [10:24] cjwatson: sounds like a nice sleeping lecture! [10:25] eek, isn't it like 3 am? [10:25] let's just say that figuring out a certain Launchpad patch is rising up my to-do list [10:26] so that proposed-migration can be atomic [10:43] hi [11:11] pitti, so how would i go about your sudo askpass solution in python ? the developer mode needs to land and i somehow need a solution for the package install steps in phablet-config (so i think adding a --password option to it and if thats not suppplied ask interactively for the pw) [11:11] i assume wild subprocess calls (which is what i would do) are not really pythonish :) [11:12] ogra_: you'll need to call sudo through subprocess for sure :) [11:12] oh, indeed ... i meant more about getting the PW and dumping it into a /tmp file for ssh_askpass [11:13] ogra_: you might be able to steal from http://anonscm.debian.org/cgit/autopkgtest/autopkgtest.git/tree/virt-subproc/adt-virt-ssh#n200 ? [11:13] the sudo call is simply needing -S and the env var indeed [11:13] ogra_: although that's way more shell-ish than necessary, as that whole thing needs to be squeezed through an ssh pipeline === ecloud is now known as ecloud_wfh [11:14] ogra_: you should just need a tempfile.NamedTemporaryFile(), write the little shell script with the "echo pwd" into it, os.chmod it executable, and add it to the sudo env [11:14] k === jgdx_ is now known as jgdx === MacSlow is now known as MacSlow|lunch [11:28] mpt, what do you mean "the icons should be in color"? [11:28] on the settings wiki errata section [11:29] did the design change? I though we were design frozen for rtm? [11:29] seb128, I mean tiheum should provide some color ones :-) [11:29] mpt, "some", like for battery, or for every panel? [11:29] For every panel [11:29] urg [11:29] I don’t know whether that’s for RTM or not though [11:30] who should we ask if that's for rtm? [11:30] my understanding was that we should stop UI changes and focus on bugs/stability [11:30] I think all the important designers are in airports/airplanes at the moment [11:31] testing airplane mode :P === dandrader is now known as dandrader|afk === _salem is now known as salem_ === Elleo_ is now known as Elleo === MacSlow|lunch is now known as MacSlow === dandrader|afk is now known as dandrader === salem_ is now known as _salem [13:09] Where do I report bugs on the eBay app? I don’t see it listed at or at [13:14] dbarth: ^ mpt question [13:15] is there a list of recommended hardware? i see in that nexus status google doc that says nexus 4, its that the only supported? [13:15] GridCube: currently the nexus 4 is the main supported phone [13:16] !devices | GridCube [13:16] GridCube: You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices [13:16] others are community supported [13:21] seb128: I have a very simple but critical bugfix for ubuntu-system-settings I'd love a quick review so I can land it https://code.launchpad.net/~ralsina/ubuntu-system-settings/fix-1357506 [13:21] mpt: please file a bug against lp:webapps-core, [ebay] in the bug title [13:25] ralsina, hey, thanks, approved [13:25] seb128: awesome [13:25] ralsina, kenvandine does regular landing, I guess it's going to be in the next one [13:26] seb128: there's one in process by cyphermox [13:26] cyphermox: mind piggybacking that branch in your silo? [13:27] there are a bunch of fixes we can land [13:27] not sure if we should invalide cyphermox's one to do a proper landing including other things [13:28] or just land that and do another one [13:28] seb128: I leave it in your capable hands to decide that :-) [13:29] ;-) === _salem is now known as salem_ [13:35] mterry, since you said you have some spare cycles, are you going to look at the privacy panel bugs? ;-) [13:36] mardy: I have a question in https://code.launchpad.net/~mardy/ubuntu-system-settings-online-accounts/lp1361608/+merge/232217 [13:36] mterry, hum, settings display that I'm unsing a passphrase unlocking when I'm on swipe, weird [13:36] Thanks dbarth, reported bug 1362177 [13:37] bug 1362177 in The Webapps-core project "[eBay] app confusingly suggests that I "Download the app" or "Open in app"" [Undecided,New] https://launchpad.net/bugs/1362177 [13:37] mterry, is the canonical storage for that setting in a-s or pam or gsettings? [13:37] seb128, it's many places! ;) [13:37] mterry, :-/ [13:37] seb128, so mostly it's the shadow database [13:38] seb128, but there is an additional display hint stored in AS for choosing between passcode or passphrase [13:38] mterry, so the setting is on "passphrase", if I select "swipe" it asks for my passphrase, I enter it and click "unset" and it stays on passphrase, but the lock is actual unlocking on swipe [13:38] seb128, sounds like AS never updated its PasswordMode property (or we didn't notice it anyway) [13:39] seb128, the way we switch to swipe is to ask AS to do it for us (it calls passwd -d as root) [13:39] mterry, hum, changing passphrase and setting doing "swipe" again fixed it [13:39] -settings [13:39] mpt: ok [13:39] seb128, maybe just a UI bug... [13:39] yeah [13:39] seb128, if you can reliably reproduce, that would be hot :) [13:39] mpt: ah, those are quite annoying, but i think oSoMoN has a special trick now [13:40] dbarth: also, it links to ebay.com and not regional ebay sites ☹ [13:40] mterry, can't :/ [13:41] dbarth, no, there’s no magic trick to fix those, usually fixing that involves writing a userscript that suppresses the popup/banner/… [13:41] mterry, I'm going to keep an eye open for it though [13:41] seb128, thanks, me too [13:41] pitti: did you ever see the U1 account translated? [13:41] oSoMoN: that userscript is what i wanted to reuse [13:41] mardy: not that I can remember, but I didn't test very often with German [13:41] (in the past) [13:42] pitti: I wrote a comment to the MP; what it fixes is only something which we broke recently; it may be that the U1 plugin needs to have a new bug [13:43] mardy: ah, ok; the MP looks correct, so landing it is fine (I can file a new bug then) [13:47] chihchun, I just commented on bug #1362172 [13:47] bug 1362172 in webbrowser-app (Ubuntu) "[webapp-container] 2d context drawed to only first one canvas in one page" [Undecided,New] https://launchpad.net/bugs/1362172 [13:48] err [13:48] oSoMoN: thanks [13:48] oSoMoN: yes, it work as expected. [13:49] jdstrand, ping [13:50] chihchun, can you please mark the bug invalid then? [13:50] oSoMoN: done [13:50] thanks [13:52] liuxg: hey [13:53] jdstrand, i saw your reply on the security bug. thanks! [13:53] liuxg: oSoMoN answered the canvas issue at #1362172, it's a typo. :-) [13:53] liuxg: I'm Rex, BTW. [13:54] np [13:54] jdstrand, who will be responsible for fixing it? SDK team marked it as invalid. [13:54] chihchun, could you please send me code and I would like to have a try? [13:56] liuxg: just replied the email, the sample is on https://github.com/chihchun/ubuntuwebapp-twocanvas [13:56] liuxg: so, I said the sdk 'could' check the logs for denials. I don't think that fits in with their current priorities (it would be a wishlist bug at most). they need to decide if they want to fix it at some point [13:57] jdstrand, the developers may be confused about the errors there. [13:57] chihchun, https://pastebin.canonical.com/115967/ this is my code. it seems that you changed it. [13:57] liuxg: documentation is being written to address (it should be posted soon on developer.ubuntu.com aiui) [13:58] WHOA, I just discovered that the welcome screen slides in BOTH DIRECTIONS [13:58] liuxg: same idea, I like to make it easier to understand. It works on RTM branch [13:58] jdstrand, I think document may not be enough since the errors are misleading, and developers will guess. [13:59] Hmm, I can’t get out of the “Emergency Call” screen, that’s not so amazing [13:59] liuxg: I understand your point, and I don't disagree that it would be a nice improvement, which is why I commented in the bug the way I did. the sdk team has to decide how they want to handle it [14:00] mpt, actually emergncy call or SIM unlock ? [14:00] chihchun, I will have a try on that [14:00] ogra_, actually emergency call [14:00] liuxg: cool [14:00] mpt, i noticed that there is an X in the bottom left ... on the PIN screen that you need to tap [14:00] not really intuitive [14:01] I haven’t had the pleasure of encountering that screen [14:05] Oh, this one [14:06] mpt, right [14:07] i had a crash today where the session came up with that screen in my face (SIM was unlocked from the crashed session already) ... it took me quite a while to try the X to get out of it [14:11] reported bug 1362193 [14:11] bug 1362193 in dialer-app (Ubuntu) "Can't exit out of emergency call screen" [Undecided,New] https://launchpad.net/bugs/1362193 [14:16] asac`: sorry, I'm up but leading the conference off-site today at 8am so not in a position to help - did you find someone else to look at it? What's the actual package that's blocked and needs unblocked? [14:17] slangasek: yes, colin unblocked. [14:17] dont worry about that part [14:25] jgdx, are you cross building settings app? [14:37] is dpm away today? [14:42] renatu, salem_: is https://launchpad.net/ubuntu/+source/messaging-app/0.1+14.10.20140820-0ubuntu1 something which should be in RTM? [14:42] my next version on top of that (mark for langpacks) certainly ought to be, but I didn't want to copy it without discussing with you [14:42] asac`, sil2100: ^ FYI [14:46] popey, So I have a recoverable error for uDraw, is there some way to mark that so the next version put in the store should get checked? https://errors.ubuntu.com/problem/73de570bfb68d49176a67f61d80025dd30b3bbe9 [14:46] who do i have to convince to get https://code.launchpad.net/~aacid/dialer-app/update_pot and https://code.launchpad.net/~aacid/messaging-app/update_pot on the phone? [14:46] tedg: define "checked"? [14:47] popey, I'd say "next time this needs manual review" [14:47] popey, i.e. put it on the bad list :-) [14:47] tedg: we can leave feedback for the developer and let them know their app is broken. [14:48] from that page I don't understand what the problem with the app is? [14:48] popey, Ah, okay. Here's one for slatekit-shell as well: https://errors.ubuntu.com/problem/c039da4850108e8f1daeba1338753d1cc1961800 [14:48] seb128: ralsina: I don't mind, if you have other landings to add! [14:48] pitti, yes, definitely, the sim card stuff is required, but better talk to bfiller as he will manage the apps rtm landings I think. [14:48] popey, The icon isn't in the package, or not in the path specified. [14:48] ah okay! [14:49] liuxg: could you paste the screenshot and the version you are testing? [14:49] liuxg: for #1362028 [14:50] liuxg: err, for bug pad.lv/1362172 [14:50] bfiller: is https://launchpad.net/ubuntu/+source/messaging-app/0.1+14.10.20140820-0ubuntu1 something which should be in RTM? my next version on top of that (mark for langpacks) certainly ought to be, but I didn't want to copy it without discussing that previous landing [14:50] bfiller: if it's ok, I'm happy to copy the source to RTM [14:51] chihchun, I just sent you my code. the problem is still there in my place. [14:51] liuxg: which version you are using? [14:51] pitti: messaging-app 0.1+14.10.20140822~rtm-0ubuntu1 just landed on rtm. probably best if you just push your changes on trunk and we'll pick them up on our next release to rtm [14:51] chihchun, do you mean the phone software? [14:51] tedg: i think we have a check for that now, so on update it would be picked up [14:51] liuxg: yeap [14:51] popey, Ah, okay, cool. [14:51] bfiller: uh, so you already diverged? [14:52] chihchun, your software works! my code was based on the template of the HTML 5 from the SDK. [14:52] pitti: yes, but we will be converging with trunk on the next release [14:52] tedg: but I'll let both the devs know [14:52] popey, Perhaps people have an old version of the package installed as well. We can't check that at the UAL level, only work with teh version they have installed. [14:52] ya [14:52] bfiller: the change I'm interested in has been in trunk for a few days already [14:52] chihchun, I think something could be wrong with the template and they conflict [14:52] seb128: have you done another landing for ralsina's fix or should I add it ? [14:52] chihchun, please check my code there [14:52] tedg: feel free to ping me any more of these [14:52] bfiller: so I'm not suwre what you mean by "push to trunk" [14:53] liuxg: you have a typo, you create two context from same canvas [14:53] pitti: ok, then that will get picked up on our next rtm release which will be based on trunk and not the rtm branch [14:53] popey, Sure, hopefully we'll get the ability for apport to report on click packages soon so I won't have to :-) [14:54] bfiller: ah, good; is there any reason to keep an RTM branch in the first place, OOI? just seems to make everyone's life harder? [14:54] liuxg: does the `mouseover' hook really works? your test code is not completed, I can not test it [14:54] chihchun, my software should be the latest https://pastebin.canonical.com/115978/ [14:54] chihchun, you may just copy your code there and replace them. [14:54] liuxg: ah? [14:55] liuxg: you can test my click package or let me test yours [14:55] chihchun, the two IDs are not the same. you can remove the onmouse thing there. [14:55] pitti: the rtm branch is going away, now that we have the ability to source package copy from ubuntu silos. so we'll just be using trunk. will be much easier, agreed. [14:55] chihchun, test my project. [14:55] bfiller: splendid, thanks [14:55] liuxg: I don't have it [14:55] pitti: np, it's all been a bit confusing for sure :) [14:56] chihchun, you do not have what? [14:57] chihchun, I sent my code to your emailbox already. [14:57] tedg: ok, feedback sent [14:57] chihchun, you can replace the code inside the "div" with yours. the problem then will happen. [15:01] chihchun, you are right. I might make a mistake there. === boiko__ is now known as boiko [15:02] liuxg: :-) [15:03] chihchun, the problem is that the with "drawRect" in the tag [15:03] liuxg: missing one " [15:03] pitti: you were doing phone langpacks? do you know how often do they update? [15:03] liuxg: cool. [15:04] or anyone :D [15:04] tsdgeos: at the moment I'm updating them manually with some hack; wgrant is working on producing ubuntu-rtm Launchpad exports, then we plan to update them weekly (cron'ed), or manually upon request === dandrader is now known as dandrader|lunch [15:04] pitti: ok, because again it happens that i've been tasked with "check some translation stuff" [15:05] tsdgeos: I can do another update from trunks (people still change/add strings like mad unfortunately) [15:05] and what's on the phone is outdated :/ [15:05] tsdgeos: well, outdated by one day -- I updated them yesterday :) [15:06] pitti: correct [15:06] still outdated :D [15:07] tsdgeos: note that this won't help for projects which haven't been marked as "use langpacks" yet [15:07] pitti: uss is in langpack [15:07] grep says [15:07] no? [15:08] for ubuntu, that's system-settings-online-accounts; for RTM, it's additionally unity-scope-click, system-settings, mediascanner, unity8, indicator-network, and messaging-app [15:08] tsdgeos: yes, but USS in RTM is way behind, and not yet marked :( [15:08] tsdgeos: so it'll use the old translations in the .debs [15:08] oh [15:08] there goes my testing stamina [15:18] someone please land current USS in RTM [15:18] the last 4 ubuntu landings are missing [15:18] tsdgeos: doing a langpack update now anyway [15:18] (but it won't be really visible until the above lands) [15:18] mardy: speaking of which -- any chance you can squeeze X-Ubuntu-Use-Langpack: yes into your current uss-online-accounts landing? [15:18] mardy: seems a bit excessive to do a separate landing just for that [15:18] pitti: :) [15:20] mardy: I'll include current trunk translations into the langpacks, so it won't regress [15:34] cyphermox, check with kenvandine for landings maybe? There are some other changes pending as well [15:45] Howdy, is there sound at all for the ubuntu-emulator? [15:45] hmm... my testing is done now [15:45] kenvandine: do you have pending landings now? [15:46] cyphermox, i need to do some landings today for sure [15:46] nothing pending right now === mibofra_ is now known as mibofra === Azelphur_ is now known as Azelphur [15:53] dobey: Hi! still running ubuntu-touch @ nexus 5? [15:54] dobey: I'm having problems with the sim card, sometimes it gets detected but I couldn't find a a pattern. do you have the same problem? any ideas on how to fix it? === chihchun is now known as chihchun_afk [16:14] pmcgowan, haven't been able to the last month [16:14] seb128, jgdx: i just proposed a branch for sim pin lock, but for now I just want to get a CI run [16:14] kenvandine, ack [16:14] kenvandine, hey [16:14] kenvandine, ok [16:14] kenvandine, did you see my bunch of small fixes? [16:14] not yet [16:14] been head down :) [16:14] they are mostly few liners/trival fixes [16:15] cool [16:15] i'll look at those in a few [16:15] jgdx, [16:15] while i wait for this to build [16:15] jgdx, you might want to review https://code.launchpad.net/~seb128/ubuntu-system-settings/background-header-icon/+merge/232402 [16:15] kenvandine, https://code.launchpad.net/~seb128/ubuntu-system-settings/updates-animations-click/+merge/232368 [16:15] seb128, k [16:15] https://code.launchpad.net/~seb128/ubuntu-system-settings/display-translated-names/+merge/232365 [16:16] seb128, any idea why the about panel's main qml plugin is named storageabout? === dandrader|lunch is now known as dandrader [16:17] doth not maketh sense to me [16:19] jgdx, because it combine about and a subpanel storage [16:19] jgdx, it was the first one, naming might have been suboptimal, we never bothered renaming [16:20] seb128, right. === yorick_ is now known as yorick [16:26] jgdx, kenvandine, thanks for the reviews! [16:27] seb128, thank you for the fix [16:46] pitti: since you already looked for the German translations, do you know what are the project names for the "Notifications center" indicator and the calendar/clock indicator ? === chihchun_afk is now known as chihchun [16:49] Howdy, is there sound at all for the ubuntu-emulator? [16:50] * ogra_ doesnt think so [16:52] not yet [16:55] rsalveti, ogra_, thanks [16:57] well does the physical keyboard work on the ubuntu-emulator? [16:59] rsalveti and ogra_ ^ [16:59] also not yet done, you still need to use the virtual one [16:59] ah darn [17:02] is it okay to make a temporary ubuntu account to access the ubuntu store on the emulator? [17:04] verterok: i am using it on n5, yes. but no issues with the sim card not being detected. my phone is totally unlocked straight from google though. i don't know if that matters. [17:05] dobey: yes, mine too. it happens randomly, sometimes after switching lfight mode on/off it gets detected [17:06] verterok: only problem i've been having related to the sim, is that for some reason the phone won't switch to LTE when wifi goes away [17:07] dobey: no LTE here, but I can confirm (when the sim is detected) that it switch to 3g with wifi off === Zic_ is now known as Zic [17:09] weird [17:09] Nothing_Much, sure [17:10] i don't get lte, 3g, or 2g. i just get the empty wifi icon all the time, with no wifi === chihchun is now known as chihchun_afk [17:18] i wonder why that doesn't work :-/ === dandrader is now known as dandrader|bbl [17:40] jdstrand, ping [17:40] balloons: hey [17:41] jdstrand, lol, sorry my stupidity.. I thought I found an issue.. The issue was, I'm not root [17:42] ok :) [18:01] Hey guys why implement a frontend for orientation lock without a backend? :D [18:01] lol [18:01] Yeah I noticed that too. [18:01] davmor2: backend in progress [18:02] davmor2: we heard you like locks [18:02] I didn't think frontend was visible to users though [18:02] it is. Tried it this morning to no success xD [18:02] dobey: I don't ToyKeeper likes the locks :) [18:02] greyback_: oh it is [18:03] davmor2: that should not have happened [18:04] greyback_: http://davmor2.co.uk/~davmor2/screenshots-phone/device-2014-08-27-190323.png [18:04] davmor2: I believe you :) === gatox is now known as gatox_lunch [18:35] jgdx, can you look at my sim pin branch? [18:35] https://code.launchpad.net/~ken-vandine/ubuntu-system-settings/security_panel/+merge/232440 [18:36] jgdx, you can grab the debs from the CI build, but also grab the changes to the qml files in rev 930 :) [18:36] jgdx, i'm testing your modem technologies branch now [18:51] pitti: yes, it's in there among the other things: https://code.launchpad.net/~online-accounts/ubuntu-system-settings-online-accounts/master/+merge/232414 [18:54] mterry, any thoughts on bug 1361137 ? [18:54] bug 1361137 in ubuntu-system-settings (Ubuntu) "Resetting PIN/passcode/password lock doesn't work" [Critical,Confirmed] https://launchpad.net/bugs/1361137 [18:55] mterry, i have reproduced it on utopic images [18:55] kenvandine, ugh [18:56] but not consistently [18:56] kenvandine, not sure [18:56] mterry, i think ogra_ suggested it had something to do with the passwd -d failing? [18:57] kenvandine, I think that was a misunderstanding [18:57] ok [18:57] kenvandine, what happens when we switch to swipe is that we ask AccountsService to do the work for us. It then calls passwd -d as root [18:57] kenvandine, let me finish up something here, then I can help look at that [18:58] mterry, i'd appreciate it... [18:58] i have no clue how that stuff works :) [19:04] kenvandine, OK am available [19:04] kenvandine, let me see what I can do about reproducing [19:05] kenvandine, should I be doing this with RTM image or utopic? [19:05] I guess people say it can be reproduced on both [19:05] utopic [19:05] i've not even tried it on rtm === plars_ is now known as plars [19:08] kenvandine, well... got it once [19:10] (process:7062): AccountsService-WARNING **: SetPasswordMode call failed: GDBus.Error:org.freedesktop.Accounts.Error.Failed: running '/usr/bin/gpasswd' failed: Child process exited with code 3 [19:10] mterry, ^^ [19:10] kenvandine, excellent [19:11] kenvandine, as I recall, the API call to set the password doesn't give any feedback that it failed [19:11] kenvandine, maybe we should directly call DBus to get that [19:11] kenvandine, oh no [19:12] kenvandine, that's ignorable and expected actually, now that I remember [19:12] ok, it didn't actually seem to fail when i saw that [19:12] kenvandine, calling SetPasswordMode does several things [19:12] kenvandine, it does passwd -d to delete the shadow entry [19:12] kenvandine, then it also tries to adjust the groups the user is part of [19:13] kenvandine, Touch doesn't support modifying groups yet [19:13] kenvandine, so it will fail at that and fail the call [19:13] kenvandine, but it still will have done the only thing we cared about -- emptying the password [19:13] kenvandine, so in our error case, the passwd -d call must be failing somehow, but I'm not sure how that would [19:14] ok... i finally reproduced it again [19:14] i locked with 4 digit passcode [19:14] then [19:14] switched back to swipe [19:14] and entered the wrong passcode [19:14] which gave me an error [19:14] then entered the correct one [19:15] the selector switched to passphrase [19:15] instead of swipe [19:15] kenvandine, ok, that's consistent [19:15] and now when it locks i can't unlock it [19:15] the greeter is giving me a login box [19:15] for passphrase [19:15] kenvandine, so we cleared the password (we thought) then we cleared the display hint that tells us it's a passcode not passphraes [19:15] which isn't raising the keyboard :) [19:16] kenvandine, whoa no keyboard is werid [19:16] ok... got it up :) [19:16] kenvandine, you should be able to enter your PIN code [19:16] yeah [19:16] that worked [19:16] and settings still thinks it's a passphrase [19:16] instead of swipe [19:16] kenvandine, oh, nice, I had similar issue changing auth type but couldn't reproduce, could you did ;-) [19:17] kenvandine, yup -- all this is consistent with the AS call failing early [19:17] seb128, yeah bug 1361137 [19:17] bug 1361137 in ubuntu-system-settings (Ubuntu) "Resetting PIN/passcode/password lock doesn't work" [Critical,Confirmed] https://launchpad.net/bugs/1361137 [19:17] kenvandine, I just tried to reproduce the ui bugs where on wrong password the osk would hide [19:17] seb128, seems rather unreliable [19:17] mterry, k [19:17] mterry, and trying to change it to swipe again doesn't let me input the old passphrase [19:18] i can't focus the text input [19:18] kenvandine, your OSK is weirding out [19:19] the input isn't getting a cursor when i click in it [19:19] kenvandine, ah OK. So it must be disabled somehow? [19:19] yeah [19:19] so... [19:19] the previous page still thinks it's passcode [19:19] not passphrase [19:20] but when i go into the lock security page, it's a passphrase [19:20] so some setting is out of sync [19:21] kenvandine, they both pull from AS [19:21] kenvandine, can't reproduce! :( [19:21] kenvandine, how did you see that AS error message? [19:22] tailing the upstart log === ajalkane_ is now known as ajalkane [19:24] kenvandine, oh nice, I thought AS was super-quiet [19:24] kenvandine, oh hm [19:24] kenvandine, I have a weird theory maybe [19:24] (process:3303): AccountsService-WARNING **: SetPasswordMode call failed: GDBus.Error:org.freedesktop.Accounts.Error.PermissionDenied: Not authorized [19:24] again... [19:25] so.. i rebooted it [19:25] kenvandine, when you reproduced this, is it right after boot? [19:25] and after rebooting, it said passphrase in both places [19:25] no [19:25] actually... not long after boot [19:25] kenvandine, hrm. No my theory is dumb, ignore it [19:25] mterry, but then trying to switch back to swipe it gave me that perm denied error [19:26] kenvandine, OK well that looks like the real error we're hitting then [19:26] kenvandine, we're not authorized to make the call [19:26] kenvandine, we have a crazy mechanism in place to authorize that call [19:26] so makes sense I guess [19:26] kenvandine, you're familiar with policykit, I gather? [19:27] kenvandine, we don't have a generic policykit agent to prompt the user when a policykit authorization is neede [19:27] kenvandine, ideally the shell would act as such [19:27] kenvandine, but anyway. We haven't really needed it up to now [19:27] kenvandine, and rather than make a big change like that so close to RTM, what I did was create a baby authorization process that we kick off to authorize just this one call [19:27] kenvandine, something is going wrong with it [19:27] kenvandine, some race I guess [19:28] mterry, now that i rebooted, i was able to switch from pass phrase to pass code [19:28] then from pass code to swipe :) [19:30] now i changed it 5 times without reproducing it [19:30] sigh... [19:30] must be racy [19:30] mterry, not really... i know what policykit is, that's about it :) [19:31] kenvandine, if you look SecurityPrivacy::setPasswordMode() in plugins/security-privacy/securityprivacy.cpp you can see some of the relevant code/comments [19:31] saw the comments [19:32] kenvandine, it *seems* to nicely wait for the agent to be ready before setting the password mode. The password mode call is synchronous, so we're unlikely to be killing the agent before it's used [19:32] kenvandine, maybe the agent is not responding well itself [19:34] kenvandine, do you happen to know where QProcess output goes? is it /dev/null by default or merged in with parent's output? [19:35] mterry, sorry, don't know [19:35] kenvandine, looks like we need to instrument the agent to be more verbose about what's wrong and then catch it again [19:35] kenvandine, I could totes imagine the agent is crashing or something bad before it can authorize us [19:35] it would be nice to see the output from the helper [19:35] kenvandine, exactly, that's what I want [19:38] mterry, this is interesting [19:38] i didn't reproduce the same condition again [19:38] but [19:38] i set the passcode [19:38] then switched back to swipe [19:38] and put in the wrong passcode [19:38] it didn't fail [19:38] and it did switch to swipe [19:38] kenvandine, that's expected actually... [19:39] kenvandine, so again this is because the design is whack [19:39] i installed #9 and got a weird lock screen [19:39] oh... so we can remove locking without the right pass? [19:39] even though i use PIN to unlock [19:39] kenvandine, we could fix it, just low priority [19:39] kenvandine, no you can't, really [19:39] i get a field where i have to type using full keyboard [19:39] kenvandine, but if you have authenticated with policykit recently, it remembers that [19:39] kenvandine, so it doesn't actually check your password [19:40] kenvandine, the solution is to look and ask policykit if we're authenticated before asking for the password [19:40] ok [19:40] like sudo [19:40] kenvandine, or don't show the prompt until we are asked is even better [19:40] kenvandine, but yeah it'll time out eventually or if you reboot [19:40] indeed [19:41] mterry, ok, i can't reproduce it again [19:41] over 10 times now [19:41] maybe because of the cached auth? [19:41] * kenvandine rebooted to try again [19:42] kenvandine, yes... and that would also explain why it seems to happen more on boot. And confirms the theory it's the agent's fault [19:42] kenvandine, "killall polkitd" might do same trick faster === dandrader|bbl is now known as dandrader [19:45] mterry, bingo! [19:45] after setting the passcode [19:45] i killed polkitd [19:46] and then tried switching to swipe, and put in the wrong passcode once [19:46] i got the incorrect pass error [19:46] and tried again [19:46] put in the right one [19:46] and it appeared to work, but switched to passpharse [19:46] pass phrase even :) [19:46] kenvandine, ooh yes [19:47] kenvandine, that exact sequence worked for me too [19:47] so just failing once screws it up [19:47] i guess it tries the second attempt without even using the corrected pass? [19:47] just the cached auth [19:47] kenvandine, hrm [19:48] kenvandine, just tried again and it seemed to work [19:48] haha [19:48] love it [19:51] kenvandine, after a reboot, those same steps worked [19:51] kenvandine, so some component besides AS is involved? [19:51] maybe I just need to restart USS [19:51] mterry, i can't reproduce it a second time without rebooting either [19:52] but... i can reproduce it reliably with those steps once after booting [19:52] kenvandine, looks like also if you restart USS [19:52] kenvandine, that's reliable [19:52] ok [19:52] phew [19:52] kenvandine, what a sequence :) [19:53] kenvandine, commented in bug [19:54] kenvandine, needing to restart USS seems like a big clue. I'm not sure what it means though [19:54] if (type == oldType && value == oldValue) [19:54] in setSecurity [19:54] any chance on the second try those values match? [19:55] kenvandine, shouldn't... oldType is what we're displaying to use (i.e. Passphrase mode) [19:58] tedg: do you have a minute? [19:58] we need to pass an argument to the launched app. [19:58] elopio, Sure, what's up? [19:58] Why? [19:59] tedg: in order to launch reminders using the evernote sandbox it receives a -s argument. [19:59] that works when launching the binary. But with ubuntu-app-launch I don't know what to do [20:00] So it's asking for an account "evernote" and you want that "evernote" to point to the sandbox instead of the main servers. [20:00] Can you just change the "evernote" account config? [20:01] Oh, they're different packages. So make them conflict and install it. [20:01] They can use the same name. [20:02] tedg: the problem is not on the accounts. For that we have the evernote-sandbox package. [20:02] the thing is on reminders, we need it to point to the sandbox servers [20:02] and ask to online-accounts for the sandbox credentials [20:02] that's nicely done on the app with an argument. [20:02] seb128, can you review a branch for me? [20:02] https://code.launchpad.net/~ken-vandine/ubuntu-system-settings/security_panel/+merge/232440 [20:02] kenvandine, sure [20:03] seb128, that's the sim pin lock stuff [20:03] oh, non trivial then ;-) [20:03] elopio, envvar? [20:03] unrelated to the other problem :) [20:03] kenvandine, do you need that review today? [20:03] I might defer to tomorrow if it's non trivial ;-) [20:04] seb128, yes... need it for rtm :) [20:04] pmcgowan, mind doing some testing on that? [20:04] seb128, i was trying to snag jgdx for it... but he must be afk [20:05] tedg: that would be less nice, but doable as the argument parsing is done on cpp. [20:05] tedg: but there will be no way for ubuntu-app-launch to pass arguments? There are some other nice arguments on reminders, and some other apps could use that too. [20:06] elopio, I don't see anywhere in the QML that references evernote.com or as a string in the binary. [20:06] passing envvars for everything is not as nice. [20:06] kenvandine, if I enter a wrong password twice, it seems to be fine === gatox_lunch is now known as gatox [20:07] mterry, interesting [20:07] tedg: http://bazaar.launchpad.net/~reminders-app-dev/reminders-app/trunk/view/head:/src/app/main.cpp#L86 [20:07] elopio, We're not creating command line utilities, they're GUI applications. We'll probably support actions in the future, but not arbitrary arguments. [20:08] kenvandine, k [20:08] seb128, thx :) [20:08] elopio, yes, but that's not the URL, I think it's getting the URL from the accounts plugin. [20:08] is there an ETA on when Ubuntu Phones will be available? Fall 2014 still the target? [20:08] kenvandine, yw! [20:08] kenvandine, do you plan a landing today btw? ;-) [20:08] yes [20:08] elopio, And there's no reason to have both installed, so I think they should just use the same name and reminders doesn't have to know which is installed. [20:08] seb128, but... i'd kind of like to get this branch in it :) [20:09] :-) [20:09] tedg: it's really useful to be able to switch from the sandbox to production while testing. [20:09] i can feel pmcgowan staring at me :) [20:09] if you have to remove and install, that's not so useful. [20:10] Hi all do you know if we will be able to change in the music scope to have the albums displayed instead of artist as your local music. Its kinda weird to have random people showing in there [20:10] also I'm not sure how will it work, because the evernote plugin is installed by default on the phone [20:10] seb128, disclaimer... not everything in that branch has auto pilot tests... and i even added a skip for 2 of the tests that i added in that branch [20:10] autopkgtests will install the sandbox plugin. I'm not sure if we can uninstall something with autopkg. [20:10] ran into issues getting them working reliably dealing with dual SIM [20:10] so i left the tests there with a skip... i need to go back and figure that out [20:10] kenvandine, k [20:10] but need to get it landed [20:10] yes [20:11] we can iterate then [20:11] at least there is some tests :) [20:11] and CI passes... so YAY! [20:11] balloons, or mzanetti, or rpadovani: anyone around? We need an alternate way to start the app with the sandbox because ubuntu-app-launch is not passing the -s. [20:11] elopio, I'm not sure, but it seems odd that autopkg tests are using a remote service at all. I think they need mocks. [20:11] kenvandine, do you have time to review Wellark's MP? was about to test that too [20:11] tedg: we use mocks for the majority of tests [20:11] pmcgowan, already did... [20:11] asked for a few fixes [20:12] tedg: but we also need to test the connection with the real and testing servers. [20:12] elopio, you are SURE the -s is being passed properly [20:12] ? [20:12] elopio, sure, but not in autopkg [20:12] seb128, regarding left to right and ">" arrows... Maybe I'm not thinking clearly about how the string would render in RTL, but I think we'd still want the arrows at the edges of the screen, rather than closer to the middle of the screen [20:12] asdfghjk its up to manufacturers. [20:12] pmcgowan, i didn't actually test the apn editor branch though... busy [20:13] pmcgowan, so testing would be appreciated :) [20:13] mterry, I though it was usually "icons after text" so "next >" would be "> txen" in rtl [20:13] mterry, but I'm unsure, I just know went through quite some fixing to use correct icons and order on rtl [20:13] tedg: yes, autopkg will be the runner. We can select not to run the real server tests while updating a version to the archive, and only run them nighly on the CI dashboard or something like that. [20:14] but the provisioning and configuration of the test bed will be with autopkg even for the real server tests. [20:14] mterry, they might even consider steps going the other way I think [20:14] mterry, e.g "< txen" [20:14] mterry, because the "next" is on the left on rtl [20:14] we could tell autopkg to switch the phone to read-write and uninstall the evernote plugin, so that's doable. [20:14] just not as nice as passing an argument to make the switch. [20:14] seb128, RTL blows my mind [20:15] mterry, we see going right as going further, but rtl has next on the left I think ... does that make sense? [20:15] seb128, is qml RTL-compatible? Like it swaps all the left/right anchors and such? [20:15] tedg: also there's this on the API: http://developer.ubuntu.com/api/qml/sdk-14.10/Ubuntu.Components.Argument/ [20:15] from that I got the impression that passing arguments to change the app behavoir was good. [20:16] balloons: I'm sure the -s is not being passed properly. [20:16] I'm sure it works with the binary, not with ubuntu-app-launch. [20:16] We should totally delete that from the API. [20:16] seb128, Wellark's apneditor branch includes an updated .pot file, should we drop that from the diff? [20:17] mterry, no idea about that [20:17] kenvandine, yes [20:17] kenvandine, well, one in a landing is ok [20:17] elopio, that confused, but I guess you are sure it's a problem. So, how would you like to solve it? Seems obvious to simply pass an argument. Can we fix ubuntu-app-launch? [20:17] elopio, If it's a test machine you're already going to have to install evernote-sandbox, I don't see the difficulty in removing evernote. The image has to be writable. [20:17] but usually better to just commit an update manually to trunk after the landing [20:18] Wellark, mind reverting that? [20:18] elopio, If you want you can make a separate desktop file for sandbox usage, and use that. [20:18] elopio, Make it install with the sandbox account plugin. [20:18] tedg: yes, I suppose you are right. The autopkg case would be easy having the two plugins with the same name. It can even clean up afterwards. [20:19] mterry, I don't know much about the topic, we could wait for bug reports [20:19] tedg: now, what about the other parameters? Like launching the app in portrait mode on the desktop? [20:19] mterry, but it feels like the alternate option I proposed could work as well and let the translators be in charge to do what they want [20:20] I find that nice from reminders main.cpp. And on the QML we won't be able to read an environment variable to decide, I think. [20:20] mterry, any more thoughts on that passcode issue? i'm really at a loss [20:20] elopio, On Unity7 you can use desktop file actions for stuff like that. We'll probably (assuming enough time) support that on U8 as well. [20:20] seb128, true [20:20] kenvandine, sorry, got distracted by other stuff [20:20] Been in the plans, just ENOTIME [20:20] me too... :) [20:20] kenvandine, the two-wrongs-make-a-right behavior is really frustrating [20:20] got stuff i need to land... [20:20] kenvandine, such a bizarrely narrow window for this problem [20:21] indeed [20:21] tedg: and what if we are running it on xephyr or xvfb ? [20:21] kenvandine, I think next step is just instrumenting the heck out of the helper and reproducing [20:21] mterry, just way outside of my expertise... you'd be our best bet on getting a fix for it [20:21] kenvandine, yeah for sure, I can take this on [20:21] ;) [20:21] elopio, You have a confusing setup. Please upgrade to a modern display server :-) [20:21] i'd appreciate it... [20:21] elopio, Not sure what you're saying :-) [20:21] mterry, but i'm happy to help repro/test [20:22] elopio, If you're running on those you don't need UAL because there's no access control at all. It's the wild west. [20:22] balloons: I don't know how I would like to solve it. That's why I'm attacking tedg with questions to figure out the best way. [20:23] tedg: we use virtual buffers so the tests don't steal the input from your machine, and to run on test beds without unity. [20:23] tedg: so if we are not running unity, we shouldn't launch the apps with ual? [20:24] elopio, You can, but if you don't have Unity there's not a reason you have to. I'd argue that using the X11 backend isn't really testing them very well. [20:24] We don't ever expect them to actually run in production with that backend. [20:25] tedg: we catch most of the errors on X11, because we are not testing the display server, we are testing the functionality of the qml in integration with other services. [20:26] the reason we have to repeat the tests on the phone is just because we don't have full coverage of everything, so many problems surface only on the dashboard. [20:26] I think there's confirmation bias there. "we run most of our tests on X11" == "most of the issues we find with our tests are on X11" :-) [20:26] but that's a separate problem. [20:26] tedg: no, we run the same tests in X11 and then on the phones. [20:27] X11 gives us faster feedback. [20:27] So you find the errors on Mir, but then verify them on X11. [20:28] sometimes. [20:28] sometimes we can simulate on X11 the conditions that caused the error on the phone. [20:29] it's generally not a display issue. It's generally either an issue of the size of the window, or on one of the services the app depends on. [20:30] Okay, but in general, I'd say you want the apps to run on Mir/Unity, and if that's the case you want UAL. [20:30] The X11 is just a sideshow in that regard. [20:30] tedg: I agree. We need to run on X11 because it's what we currently have in the development machines. [20:31] maybe an alternative would be to install MIR on the testbeds, but not unity. === beldar is now known as Guest47982 [20:31] tedg: but so the bottomline of your opinion is that if we need to change the behavior of the app, we should do it on the desktop file? [20:32] elopio, Yes, for non-production cases. Don't really want apps modifying their own desktop file (not that they could, but in theory). [20:33] tedg: yes, we can just do it on the fly on the tests, and remove the test desktop when they finish. I think I can get something working on reminders for you to see if you like it. [20:34] assuming that if I change the exec to evernote -s it will pass the argument. [20:34] I'll give it a try. Thanks tedg. [20:34] :'( I gave my evening to translate some core apps, but lp is having to much errors ;'( [20:34] timeouts [20:34] cm-t: yes, it's bad. I have to retry almost every time I open or submit something. [20:35] the good part is that it doesn't lose the translations. === wilee-nilee is now known as Guest90398 [20:37] balloons: do you know if we can pass arguments to the autopkgtests ? [20:38] elopio: yes, at least nothing is lost === Fyodorovna is now known as Guest75308 [20:39] elopio, sure, but that's an odd question.. you mean pass from the runner to the test or ? [20:39] what's your language cm-t? [20:39] balloons: yes. I would like to select which tests to run. === Guest75308 is now known as friendly-fascism [20:40] elopio, you mean autopilot tests yes? if so then yes [20:40] balloons: lets say that we get a fake evernote server. On MPs and for getting the package updated in the archive, I would like to run the tests with the fake server. [20:40] but on the dashboard, I would like to run them against sandbox. [20:41] it would be nice to use adt-run for every case. [20:41] sure, but I'm not sure if that logic belongs in the runner.. perhaps jenkins, perhaps the test [20:46] elopio: [FR-fr] [20:47] balloons: the one that choses the flags to pass would be jenkins or the tester. [20:47] elopio: but i'm not in the ~ubuntu-l10n-fr so it will need a validation (I will send them an email to ask tell them) [20:48] cm-t: I've just asked to be added to the ~es team so I can review suggestions. [20:48] elopio, so you would configure jenkins to invoke adt different ways depending on what you want, and the tests should support it [20:48] now I have to wait 2 weeks while they vote ×_× [20:49] balloons: I think so, yes. [20:49] like daily against sandbox, weekly against production, things like that. [20:49] elopio, so one way to do it is via ADT_AUTOPILOT_MODULE env var.. [20:50] you can pass a full arg list for AP that way to be whatever you want [20:52] balloons: I like that. [20:52] hum, exec: evernote -s doesn't work. [20:53] * balloons notes he's not sure how passing args to autopilot gets you what you want, but :-) [20:55] balloons: something like adt-run suite ADT_AUTOPILOT_MODULE='u1-server-url=staging.one.ubuntu.com' [20:55] yes, but 'u1-server-url=staging.one.ubuntu.com' is passed to autopilot, not the test [21:00] balloons: ah that's the other part. We have an open bug for autopilot to accept arguments. [21:00] elopio, :-) [21:02] elopio: Didn't thomi implement that feature so you could pass arbitary command line args through to the test? [21:02] elopio: or are you wanting something different? [21:03] veebers: it's still open: https://bugs.launchpad.net/autopilot/+bug/1266589 [21:03] Ubuntu bug 1266589 in Autopilot "Pass argument values from autopilot run to the tests" [Undecided,New] [21:03] * veebers looks [21:04] elopio: does this help? http://developer.ubuntu.com/api/devel/ubuntu-14.10/python/autopilot/api/autopilot.html#autopilot.get_test_configuration [21:05] * balloons waves to veebers [21:05] veebers: yes! [21:05] whoa, when did this happen? [21:05] * balloons facepalms [21:06] veebers coded it, landed it and updated the docs 30 seconds ago. [21:07] elopio: heh, sorry we should have been more on the ball with updating that bug. thomi implemented it in Malta I believe [21:07] hey balloons o/ [21:08] ok, so not so old.. just malta [21:08] balloons: where is the binary for the reminders app that's preinstalled? [21:08] hey veebers I'm looking at our favorite mp again while I can sneak some time [21:09] elopio, click contents is your friend [21:09] balloons: nice, I've been meaning to take a look myself but haven't :-\ [21:09] elopio, /lib/arm-linux-gnueabihf/bin/reminders for the phone [21:09] balloons, elopio: on that note. What would the best method of exposing these types of features to yourself and other developers? Would hate to have other features missed [21:10] * balloons notes he has a backlog of 2 posts to make about recent AP additions [21:10] namely screenshots and subunit [21:11] veebers: if you started to blog and linked the posts on g+ I would always notice. [21:11] that's how I know what balloons and pitti are doing. [21:12] elopio: very good point, I should do that [21:12] veebers, the mailing list posts the other day made a difference for me [21:12] the mails to the list like what you did for the subunit file with screenshot will make me read them too. [21:12] that's a least something [21:12] I intended to more or less reblog and talk about it [21:12] balloons: aye, those were features we really wanted to hightlight. Using trv with a subunit result should be nice to use [21:12] but I already have too many mails, and they almost never have cool new stuff. [21:13] elopio: right, I wouldn't want to spam the list with minor features/fixes etc. === dandrader is now known as dandrader|afk [21:21] balloons: $ less /lib/arm-linux-gnueabihf/bin/reminders [21:21] /lib/arm-linux-gnueabihf/bin/reminders: No such file or directory [21:24] elopio, sorry, it's a relative dir [21:24] /usr/share/click/preinstalled/com.ubuntu.reminders/ [21:24] or under opt/click [21:25] for instance, I see /usr/share/click/preinstalled/com.ubuntu.reminders/0.5.226/lib/arm-linux-gnueabihf/bin/reminders [21:25] balloons: ah, thanks. [21:43] any click dev to see what cause this http://paste.ubuntu.com/8163075/ ? [21:44] cjwatson: any idea ? === salem_ is now known as _salem === asac` is now known as asac [22:05] daker: whats the issue? [22:05] popey: ctrl+F : com.canonical.payui [22:05] i still fail to see the problem [22:05] popey: 7 entries [22:06] you specified --all [22:06] dont do that and you wont see 7 entries [22:06] that's what u-s-s does [22:06] why is that a problem? [22:07] over time various versions of that app have been on your device [22:08] and they are not deleted on upgrade ? [22:08] the apps are, yes [22:08] look in /opt/click.ubuntu.com/ [22:08] specifically /opt/click.ubuntu.com/com.canonical.payui/ [22:08] probably only one in there [22:08] 0.3.14 [22:09] no :( [22:09] http://paste.ubuntu.com/8163237/ [22:10] interesting, i only see one here [22:10] so click isnt cleaning up for you? [22:11] calculator, calendar, music, camera, clock, dropping-letters, filemanager, gallery, reminders etc... all have more than one entry :( [22:11] popey: yes [22:12] https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1265250 [22:12] i have more than one entry for some too [22:12] Ubuntu bug 1265250 in ubuntu-system-settings (Ubuntu) "Storage shows same app multiple times" [Low,Confirmed] [22:12] ok, i see. [22:12] popey: comments see attachements [22:14] ah, thought this was familiar [22:14] i filed a dupe [22:14] 315MiB used by apps [23:22] are there any QML ubuntu SDK programmers around? [23:46] howdy everyone [23:49] who else in here is running Ubuntu Touch on a Nexus 5?