[01:37] are there not actually any qml apps which use online accounts at all? [01:53] just updated my m10 to ota-11, with read-write mode on... is this a bad thing? [01:55] i thought updates were disabled when that's on? [02:05] swalladge: enabling read-write and using apt is a bad thing. but you probably did it manually on the device with mount -o remount,rw, instead of with phablet-config i guess [02:13] ooh [02:14] I'm pretty sure I didn't use mount commands though [02:15] shame the system can't be read-write by default - i mean what's the point of a full linux system on a device if you can't install anything apart from qml apps? [02:15] might as well have android [02:15] :\ [02:18] the point is that it's a phone or tablet, and not a PC [02:18] if you want legacy apps, then install them via libertine [02:18] that's what it is for [02:31] should look into libertine - can you install terminal programs with that as well? (ie. vim, python, mosh, syncthing, etc.) [02:39] swalladge, yes [02:42] lol that still requires read/write mode to install it === chihchun_afk is now known as chihchun [05:19] gd morning [05:20] has anybody seen or heard of webbrowser crashing and not recovering anymore after ota 11 ? [05:24] I could use some help getting it on its feet :/ [06:28] A lot of crashing sure, but at least for now I've managed to start it up again [06:28] (crashing twice in a half an our period of using) [06:33] dobey: yes, it was almost midnight :-) [06:34] dobey: no, it cannot be done without a small Qt helper, since the data in the plugin is stored using Qt's serialization [06:34] dobey: maybe we can just drop the token migration? Forcing users to re-enter their credentials is not a big deal, IMHO [07:23] Hi ! I just installed a click package from adb command line but nothing appears in the applications list === chihchun is now known as chihchun_afk [08:30] I've noticed facebook sends push notifications to chromium, which show on the desktop as long as chromium is open; even if the facebook website isn't open [08:30] That's pretty cool [08:31] does / will utouch support this sort of stuff? [08:32] ...and does anyone know if facebook does this using standards? or is this google/android specific? [08:35] If any website can push notification to the desktop without even being open this is gonna be a problem [08:36] Yes, indeed. But this can be regulated with "allow ____ to send notifications?" [08:36] ok === chihchun_afk is now known as chihchun [08:38] Just imagine some sort of web-irc. It runs on a server and when a private message is sent, it arrives to the user as a notification to the phone [08:39] Hit it, and it opens up the app/webapp. Bam! Application lifecycle drama solved [08:40] And no need for facebook messenger, just await notifications and let the user open it up when he wants to reply :) === Trevinho_ is now known as Trevinho [08:45] Or you can just use an IRC client… [08:46] This isn't currently possible in utouch, but that's not the point [08:47] There is an IRC client since a few weeks [08:47] A bit limited but it works [08:47] Hello all. There is any document about porting applications to UT? [08:47] There is no XMPP clients through :-( [08:48] Does this IRC client work when sent to the background? [08:48] brunch875: but notifications for things like github would be cool [08:48] brunch875: Oh yeah maybe not. uTouch is really bad at this [08:48] brunch875, funny, i had the same plans with the ircproxy snap http://bazaar.launchpad.net/~ogra/+junk/ircproxy/files but my time for working on it is rather limited atm ... and you need some client app as well as a way to set up the key for the notification system [08:49] ogra, always one step ahead :) [08:49] what i currently do is simply excluding the kiwi irc client from the lifecycle :) [08:49] i actually gave a talk about that irc setup at the last ubucon in berlin :) [08:49] Hey from my bq [08:49] MCMic_: test [08:50] brunch875: Nope, no background notification indeed, so sad. But it seems there not even HL on pseudo detection [08:50] As I said, a bit limited. [08:51] Yeah, that's why I mentioned this [08:51] apparently the w3c has formulated a draft for push api: https://www.w3.org/TR/push-api/ [08:51] gsettings set com.canonical.qtmir lifecycle-exempt-appids [com.ubuntu.music,com.ubuntu.developer.ogra.kiwi-irc] [08:51] thats what i use to exclude kiwi from the lifecycle [08:51] (And it’s kind of sad we can’t just port a desktop irc client and tweak the UI, developping yet another IRC client seems crazy. But maybe it uses a lib which is generic I did not check) [08:52] (just dont fotrget to close it if you dont need it, else it drains your battery) [08:52] :D [08:52] * brunch875 grabs the hacks [08:53] kiwi has builtin notification sounds if it isnt suspended and you get a ping [08:54] Is there any chance at having something based on libpurple for uTouch for instance? [08:54] MCMic: doesn't seem like there are plans for it just yet [08:55] does uTouch have a password keyring by the way? Just remembered how libpurple stores clear passwords and wondered about this [08:55] But webapps + push notifications sound like a good way to go [08:56] I don’t like/want webapps [08:56] well then, push notifications + server-heavy apps [08:56] xD [08:57] what I mean is running apps in the background isn't that necessary [08:58] I remember my last android device came bundled with a not-so-easy-to-remove facebook app which hogged RAM and drained battery [08:58] as soon as I flashed a basic rom to it, it would run much smoother [08:59] But I want app to run in the background [08:59] for what purpose? [08:59] It’s so stupid to deport something as simple as an IRC client to some server just to get notification [09:00] brunch875: it’s called multitasking :-P [09:01] Phones are real computers now, I don’t see why they should be limited in way desktop PCs are not. [09:01] MCMic: what I meant is you can have the skeleton of the IRC client installed and leave a server handle all the logic [09:01] it would make no difference to what you see [09:02] It creates a dependency to a server, a privacy problem, a useless use of bandwidth and processor [09:02] And unnecessary complicated network code [09:03] Knowing the IRC protocol, none of those would be issues :P [09:04] since you already have a dependency to server, there's no privacy, and networking/processing would remain pretty much the same [09:04] depending on one server or two is not the same thing [09:05] And this is not limited to IRC, there will be the same problem for other protocols too [09:05] I hope to have an XMPP client at some point [09:06] How does the telegram client works BTW? It has notifications but they are not synced with the app so it’s kind of weird === zsombi_ is now known as zsombi === marcusto_ is now known as marcustomlinson === \b is now known as benonsoftware === Saviq_ is now known as Saviq === K1773R_ is now known as K1773R [10:18] query for anyone who knows the answer to this - ihave the 'gmail' app/webapp thing installed, and it pops up notifications on my e-mails which is nice... but is there any reason that things liek FB/Twitter cant also do that? [10:27] Acou_Bass: api things I recall, but my twitter does give notifications [10:27] hmm fair enough, i dont think mine does but maybe thats because i dont have them setup on twitter itself [10:28] or amybe im just so unpopular that i dont get notifications on it anyway :D [10:29] I recall it wasn't supposed to work at all [10:30] I have no idea what way it does give notifications in mine [10:31] XD === chrisc_ is now known as chrisc [11:50] "openssl ciphers EXP" returns 7 export grade ciphers on OTA11. According to https://www.openssl.org/blog/blog/2015/05/20/logjam-freak-upcoming-changes/ these ciphers have been disabled in OpenSSL v1.0.1m over a year ago. What's the rationale for enabling weakest ciphers? [11:51] well, the phones are based on 15.04 ... you might want to file a bug so the change gets backported [11:52] ogra_: OK. Thanks! [11:52] phones will switch to 16.04 eventually, but that will still take a while ... [11:53] ogra_: That was my next question. :-) [11:58] Not only I cracked down the development but also solved few Libertine related problems, including an on screen keyboard. Libertine does not support phone's own OSK, so I developped my own to use inside applications running on xmir via libertine ;) [11:58] More info in the vid description, documentation in the making. [11:58] https://youtu.be/j__WaGH4aig [11:59] KrisJace, OSK support is nearly there, will be in OTA12 [12:01] great! [12:01] but I didn't want ot wait [12:01] yeah [12:02] KrisJace: u still have to click the button twice :D [12:02] and the good thing about ubuntu phones and tablets is that you can hack any aspect of them ;) [12:03] great suggestion for libertine would be to make it possible from the scpe to control xvfb and corresponding vncserver [12:03] so that you can dive into vnc remote desktop even before you dive in your phone via ssh [12:03] I have successfully installed and used this with libertine container [12:04] * ogra_ just uses a monitor :P [12:04] ;) [12:04] at least before miracast gets fully available [12:04] vnc is still nice feature [12:04] and I found it easy to use as well [12:04] plus, you get legacy x11 environment [12:04] I use it with openbox [12:04] works great [12:04] on my vid is just that [12:05] but not from libertine [12:05] from my own container [12:05] i couldn't get lazarus to work on libertine === JanC is now known as Guest27710 === JanC_ is now known as JanC === _salem is now known as salem_ [12:51] * ogra_ just opened https://docs.google.com/document/d/1yJepibh68YaQijWO3Z3dWTtTTmzXnMmEE8eswhUXzw4/edit and had to LOL ... [12:52] mzanetti, thanks for mentioning that it isnt my doc :) [12:52] lol [12:54] ogra_, I saw you mentioning that a couple of times already but people wouldn't believe it [12:54] yeah [12:54] ogra_, btw, authenticator qr code scanning now working for you? [12:55] (my "thanks" above was actually serious :) ) [12:55] dunno, when it didnt work i punched it in manually .... and now it works (i tend to not remove/add devices once they work) [13:04] mardy: yeah, i think it would be better to not migrate, than to have all the extra stuff needed to do so. only real issue there is that not having a valid account makes in-app purchases slightly weird [13:07] dobey: how so? [13:11] mardy: because QtPurchasing has no way to send errors to the application. so if the dev does things slightly wrong, things can get a little weird [13:34] dobey: so pay-ui would be one of those clients which will allow the password UI to open on top of them, right? [13:36] mardy: no [13:36] dobey: how should it handle invalidated tokens then? [13:37] mardy: well, not a different UI than we have today anyway. pay-ui uses OnlineAccounts.Client [13:37] mardy: pay-ui provides UI for logging in [13:37] dobey: I understood that you were deleting the account and recreating it, therefore showing a UI [13:37] mardy: pay-ui does Setup.exec() if there is no account [13:38] dobey: but that's the thing we decided to leave for phase 2, right? As long as the UI that we popup looks essentially the same [13:40] mardy: no. if the account is just invalidated but still there, then pay-ui will just require you to enter your password again, the same as we do today if it's been > 15 minutes since you logged in [13:41] dobey: mmm... ok, so you would block the UI, and popup your UI when signond returns you the UserInteraction error? [13:41] dobey: and then put the password+OTP you get from your UI in a new auth session? [13:43] mardy: i guess. i still don't really understand how the "correct" way for all this stuff to work is. [13:44] dobey: basically, you can choose between two ways: [13:44] mardy: but i guess no there wouldn't be a UserInteraction error. when pay-ui gets the existing credentials, they would be invalid, so the "last updated time" would be 0, and we'd show the UI to enter the password, and then would just call login() [13:45] 1) allow UI interaction to transparently happen - makes your code much easier, but you have little control over the UI (at least for now) [13:45] the goal here is to not change pay-ui [13:46] 2) prevent unwanted UI interactions, and you'll get an error; then you can collect the password/OTP yourself in your UI, and retry to authenticate with those data [13:47] dobey: both are fine [13:47] dobey: as I understand it, now pay-ui calls SSOService::login(), once it has collected the user credentials, right? [13:48] if it's necessary to show the password entry field to buy something, yes [13:48] dobey: ok, then I think that this doesn't need changes [13:48] then it calls login() and if 2fa is required, will show that field and then call login() again [13:48] dobey: perfect [13:49] dobey: what might need to be changed (dunno, I didn't check) is error handling [13:49] dobey: basically the OnlineAccounts.Client stuff should be called only on the AccountNotFound error, and not in other cases [13:49] it shouldn't because libu1auth should send the error signals the same [13:50] dobey: it might be already like this, I didn't check [13:50] yes, we only call Setup.exec() when there is no existing account [13:51] dobey: and you call login() when you either don't get a token, or the token is too old, right? Is this the logic? [13:51] we require the password to be entered when the token is too old, yes [13:53] i guess we need to make sure "invalidated (thus empty credentials) token" doesn't result in credentialsNotFound() but gives an empty token [13:56] dobey: ah, right, then my branch was not doing the right thing there: I was indeed emitting credentialsNotFound() [13:58] mardy: btw, why are there no docs for OnlineAccounts.Client on the developer site? [14:03] dobey: I think I pinged someone about it a couple of times, then forgot [14:03] dobey: let me double check... [14:07] dobey: or maybe it was about bug 1512375, but I don't have any reason to believe that the outcome would be much different :-) [14:07] bug 1512375 in Ubuntu Developer Portal "Add API documentation for Online Accounts v2 " [Undecided,New] https://launchpad.net/bugs/1512375 [14:07] dobey: I'll file a bug [14:11] ok [14:12] was trying to figure out how to actually use an online account from a qml app last night, and couldn't figure it out [14:12] for a personal project [14:12] davidcalle: hi! who should I ping about bug 1588836 and bug 1512375? [14:12] bug 1588836 in Ubuntu Developer Portal "Add API documentation for Ubuntu.OnlineAccounts.Client" [Undecided,New] https://launchpad.net/bugs/1588836 [14:12] bug 1512375 in Ubuntu Developer Portal "Add API documentation for Online Accounts v2 " [Undecided,New] https://launchpad.net/bugs/1512375 [14:12] mhall119: ^^ [14:13] mardy: ping whomever is developing those APIs, the docs are imported from source [14:13] mhall119: it's me :-) [14:14] is this a new package? [14:14] mhall119: what should I do? It's not clear to me [14:15] mhall119: one is relatively new (1 year?) the other is much older [14:15] mardy: we currently use ./get_package.py accounts-qml-module-doc [14:15] mhall119: can we add more packages? [14:16] mhall119: i think you need qml-module-ubuntu-onlineaccounts2-doc and qml-module-ubuntu-onlineaccounts-client-doc [14:16] instead of accounts-qml-module-doc [14:17] dobey, mhall119: I wouldn't completely remove accounts-qml doc, but it would be good if there was a way to make it less prominent [14:17] mardy: yes, but it'll require me pushing changes to the devportal code [14:17] mardy: sorry, docs are either there or not there [14:17] mardy: i guess it will remain for the older framework versions [14:17] mardy: if you update the old docs with a warning about them being deprcated, that will get imported [14:17] mardy: but we could drop it from the new one [14:17] dobey: right [14:18] ah, yes, any released framework docs will not be changed, those are frozen forever [14:18] so we can drop it from newer ones [14:18] mhall119: ok, then let them be published, I'll update them to specify their scope [14:19] mardy: I've assigned the bugs to me, I'll let you know when they go live [14:19] mhall119: excellent, thanks [14:21] mhall119: is it possible to customize the name shown in https://developer.ubuntu.com/api/qml/current/ (for example to add a version number?) [14:22] mhall119: I'm asking because accounts-qml-module-doc and qml-module-ubuntu-onlineaccounts2-doc have the same import string, but a different version [14:22] mhall119: if it's not possible, then you should remove accounts-qml-module-doc [14:26] Hi, I am not sure whether I am at the right place to ask this question, I am having trouble when trying to connect my Meizu Pro5 to my lenovo WD100 wireless display adaptor [14:27] mardy: the docs should have the import string in the, like https://developer.ubuntu.com/api/apps/qml/sdk-15.04.4/Ubuntu.OnlineAccounts.Account/ [14:27] Keerthi_: didn't you just ask on askubuntu? [14:27] Keerthi_: I suppose this is as good a place as any, morphis and kgunn might be able to help [14:27] Keerthi_, is that a plain miracast adapter (only standards compliant ones work) [14:28] i dont have a pro meself [14:28] * ogra_ only has MS adapters [14:29] Keerthi_: can you explain a bit more what kind of problems you have when connecting? [14:30] seems thats a lenovo specific device that only works with lenovo hardware [14:30] i guess they add something on top of miracast (like chromecast does) [14:30] mhall119: ok, then I guess you should remove the old one; I'll add a comment on the bug [14:31] I am able to see, that is the connection is established, but when finalising, it shows as disconnected in the device [14:31] that is phone [14:31] and i get some strange sounds on TV and video flickering [14:31] ogra_: that is possible [14:32] Keerthi_: can you paste me the content of /var/log/upstart/aethercast.log? [14:32] ok, let me grab that [14:32] ogra_: it claims it is compatible with "android 4.2, Windows 8 and above versions" [14:33] hi, I read ota-11 brings firefox to m10. Is this also true for smartphones? [14:33] morphis, well, in the comments on amazon i read that people return it because it only works with lenovo devices [14:33] ogra_: ah, should have read to the bottom of that page :-) [14:33] not generically with any miracast or WiDi [14:36] ogra_, morphis: i already asked keerthi to file a bug: https://askubuntu.com/questions/781370/ota-11-not-able-to-connect-meizu-pro-5-to-lenovo-wd-100-adaptor [14:36] dobey: sounds good [14:37] ah, you use it with a HTC phone ... [14:37] so it is at least compatible with something non lenovo === dandrader is now known as dandrader|afk [14:43] Keerthi_: please file the bug as i suggested in your question on askubuntu === chihchun is now known as chihchun_afk [14:45] dobey_ :Thank you, sorry got lost in between [14:46] Keerthi_: and attach the aethercast.log to the bug please :) [14:49] dobey_:sure, can you please let me know the path again [14:49] Keerthi_: /var/log/upstart/aethercast.log [14:49] might be worth to add that to askubuntu [14:50] * ogra_ guesses we'll have more people coming with such issues [14:50] dobey_: thank you [14:51] ogra_: yes [14:51] ogra_: that is what we have to waltk through :-) [14:52] yep [14:52] the briar patch of non-compliant things [14:53] dobey: yes [14:58] morphis, kgunn: btw, i've noticed lately that unity8 is constantly using between 3-8% CPU while the phone is doing nothing. is it perhaps related to aethercast integration? [14:59] dobey: is that being with aethercast connected or disconnected? [15:00] morphis: i don't have any miracast devices to connect it to, so disconnected. and even if i "sudo service aethercast stop" it's still doing it [15:01] hi there! [15:02] does anybody know if is possible to install ubuntu touch in the Aquaris e5 hd 4g? [15:04] dobey: that's unity8 not unity8-dash? [15:05] kgunn: yes [15:05] oh wait dobey so you still see aethercast running? [15:05] * ogra_ noticed that his turbo doesnt seem to actually sleep :/ [15:06] kgunn: yes, but even without aethercast running, unity8 still uses the cpu [15:06] the battery graph is constantly going down ... no matter if the device was on or off [15:07] dobey: so "doing it"==unity8 using cpu, not aethercast still running [15:07] are we not putting it into deep sleep state ? [15:07] ogra_: how does that work? i would presume if unity8 keeps running....it won't let the kernel trigger that? [15:07] could be fault of shell... [15:08] kgunn, well, to my knowledge we had issues on all devices where the wlan driver kept it out of deep-sleep [15:08] i'm not talking about unity8 specifically here [15:08] dobey: what would be interesting, does unity8 show the same issue after fresh reboot (w/o having previously connected to aethercast) [15:08] sorry, coincidence that dobey mentioned that :) [15:09] kgunn: i don't have any devices to connect aethercast to [15:09] kgunn: so yes [15:09] ah, k [15:10] Saviq: ^ anyone already looking after such a thing? [15:10] unity8 running 3-8% constantly [15:11] kgunn, dobey, bug #1579031 maybe [15:11] bug 1579031 in Canonical System Image "High CPU usage (~10%) on Pro 5 when idle (with screen on)" [Undecided,Confirmed] https://launchpad.net/bugs/1579031 [15:12] I didn't find a relation between that and aethercast, but that doesn't mean there isn't one [15:12] s/find/notice/ [15:13] my mx4 pretty steady at 2.9% for unity8 [15:13] and battery drained overniht first time in a while [15:13] thats screen off [15:14] eeewww [15:15] dobey: then it doesn't seem to be related [15:15] its a feature: "you can now leave your screen on all the time and will not use more battery than with screen off" [15:16] ;) [15:16] * brunch875 applauds [15:16] kreillin seems fine running stable [15:16] krillin [15:22] Saviq, shall I file a separate bug for screen off case [15:22] seems it regressed within last 3 weeks [15:24] pmcgowan, yeah, that sounds like it didn't go to sleep, not even about unity8 [15:24] Saviq, but its at 3% constant [15:24] whereas the other mx4 doesnt do that with old image [15:24] separate from the battery drain issue [15:25] and sus[end_blocker tells me it is sleeping [15:25] pmcgowan, any case, yes, file a bug :) [15:26] I can confirm on krillin [15:28] Saviq, https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1588873 [15:28] Launchpad bug 1588873 in unity8 (Ubuntu) "CPU usage with screen off" [Undecided,New] [15:32] mardy: Sorry I was away. I just followed up on bug 1587829. [15:32] bug 1587829 in ubuntuone-credentials (Ubuntu) "Password error message is not localized" [Undecided,Confirmed] https://launchpad.net/bugs/1587829 [15:36] mpt: thanks [15:37] mpt, mardy: that error is provided by the server; i moved the bug [15:37] dobey: do you think that the server can deliver the translated message? [15:38] mardy: not sure [15:39] dobey: I'm almost sure it cannot :-) [15:40] dobey: the bug must be fixed in ubuntuone-credentials, similarly to how the bug on the wrong email address was fixed [15:40] huh? [15:41] what bug? [15:43] dobey: http://bazaar.launchpad.net/~ubuntuone-control-tower/ubuntuone-credentials/trunk/view/head:/online-accounts-provider/NewAccount.qml#L130 [15:43] dobey: not sure if there was a bug, actually [15:44] mardy: that line is probably a bug in fact [15:44] So either the server provides an error code, and the client localizes it [15:45] Or the server provides an English error message and promises never ever to change it, and the client matches it and localizes it [15:45] dobey: I have the branch where I think I'm handling the errors in a better way, but I'm not putting it for review since you asked not to put too many things in the plate [15:46] mpt: 3) the client tells the locate to the server, and the server returns localized errors [15:46] mpt: but in any case, the client needs to be changed too [15:46] s/locate/locale/ [15:46] well, frankly, u1-credentials needs a major rewrite and redesign, but trying not to just dive in to do that, and fix things incrementally instead [15:47] dobey: yep, but just FYI I have a branch where I'm rewriting the U1 plugin not to use libubuntuoneauth at all, so when you want to see it, just tell me [15:49] first things first [15:49] mardy, yeah, I guess (3) allows changing the server errors later without requiring people to update the client too [15:50] exactly [15:52] anyway, back to more immediate concerns [15:56] the first of which is lunch. :) === tyhicks` is now known as tyhicks === dandrader|afk is now known as dandrader [17:26] Hi is it normal than vimeo scope got removed from the image today on rc-proposed ? Also i cannot find it anymore on the store. [17:26] i think some old already-deprecated framework versions were removed in latest image [17:29] dobey: so Bye bye vimeo scope :'( it was making the video aggregator scope quite nice now there is only youtube and my videos as aggregators. For privacy i do not like showing my videos and what's left of the video scope is youtube results.. [17:32] taiebot: seems like a bug that just needs to get fixed [17:35] taiebot: just filed bug #1588920 for it [17:35] bug 1588920 in Vimeo Scope "Framework version out of date" [Undecided,New] https://launchpad.net/bugs/1588920 [17:42] dobey: thanks [18:18] mardy: not still around are you? === pindonga` is now known as pindonga === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === aaron is now known as Guest86371 === salem_ is now known as _salem === Guest86371 is now known as ahoneybun [21:19] Hello, is there anyone experienced on Samsung Galaxy Note 4 with Ubuntu touch?_ === yorick_ is now known as yorick [23:20] mm anyone with a Nexus 7 on rc-proposed?