| === Jucato_ is now known as Jucato | ||
| falcon | hello people, I'm new here and want to port ubuntu touch for Zuk Z1. Can anyone get me started? | 04:18 |
|---|---|---|
| falcon | Anyone? | 04:54 |
| lotuspsychje | falcon: repeat your issue plz? | 04:54 |
| falcon | hello people, I'm new here and want to port ubuntu touch for Zuk Z1. Can anyone get me started? | 04:55 |
| falcon | the wiki is a little bit blunt | 04:55 |
| lotuspsychje | falcon: porting ubuntu touch isnt very easy mate | 04:55 |
| lotuspsychje | falcon: perhaps try the XDA forums and try to find an existing project? | 04:55 |
| lotuspsychje | falcon: another way around is too sell your device and buy a supported ubuntu touch device | 04:56 |
| falcon | I have ported Sailfish OS. Same with libhybris. Their build wiki was good. | 04:56 |
| lotuspsychje | falcon: we alway looking for porters, makes it good for the community | 04:57 |
| falcon | So I guessed this will be worth the try. | 04:57 |
| lotuspsychje | falcon: try the XDA forums and mariogrip website | 04:57 |
| falcon | Not to sound very aggressive here. But mariogrip's ubports community has some serious psychos out there. I'll try though | 04:58 |
| falcon_ | just a question is ubuntu's source compatible with cm-12.1 abi? (Lollipop)?? | 05:01 |
| lotuspsychje | falcon_: dont know that one mate | 05:01 |
| lotuspsychje | but latest OTA should be more android friendly so i read | 05:02 |
| falcon_ | But I'll say porting wiki is really inadequate. | 05:03 |
| lotuspsychje | falcon_: unfortunately porting isnt very easy, we need some breakthrough worldwide still in my opinion | 05:05 |
| lotuspsychje | falcon_: im waiting for ubuntu-touch could be installed on ANY device, would rock my world totaly | 05:05 |
| falcon_ | even me. | 05:05 |
| falcon_ | I have tried almost all other open oses (ffos,sfos,all android variants, maemo/meego,iOS) etc., Only ubuntu is pending. | 05:06 |
| lotuspsychje | falcon_: perhaps your the one that will lead us further :p | 05:07 |
| falcon_ | I ported only one of em.... But then who knows. :P | 05:07 |
| Dylan | Trying to install on a OnePlus X...got to the point to install and select a channel but I get "Device MSM8974 not found on server". Any help? | 08:43 |
| Dylan | Trying to install on a OnePlus X...got to the point to install and select a channel but I get "Device MSM8974 not found on server". Any help? | 08:56 |
| === JamesTait is now known as Guest95749 | ||
| === Guest95749 is now known as JamesTait | ||
| === JanC is now known as Guest86538 | ||
| === JanC_ is now known as JanC | ||
| === tosky_ is now known as tosky | ||
| Silicium | hi there | 11:25 |
| taiebot | Hey LoquiIm on the store :) | 13:30 |
| kenvandine | mterry, is there anyway i can get apps launching in the unity8-session with strict confinement? | 13:48 |
| mterry | kenvandine: yeah... should be working? I had it working when I was testing the unity8 interface, maybe you need that snapd branch to land first... But you tried it with the mir and unity7 interfaces and it didn't work? | 13:48 |
| kenvandine | right | 13:49 |
| kenvandine | mterry, it wasn't working for me last week and tedg suggested you might have something working | 13:49 |
| mterry | kenvandine: ok I guess you have to wait for a unity8 PR to land | 13:49 |
| mterry | Which is stuck in review issues right now | 13:49 |
| kenvandine | or i can merge that into my branch :) | 13:49 |
| mterry | kenvandine: oh it's not a unity8 PR but a unity8-interface PR for snapd -- which is... here: https://github.com/snapcore/snapd/pull/2787 | 13:50 |
| kenvandine | yeah, i have a snapd branch for the content-transfer interface | 13:50 |
| kenvandine | which is what i'm trying to test | 13:50 |
| mterry | kenvandine: if it doens't work after that, let me know | 13:50 |
| kenvandine | mterry, thx | 13:50 |
| dobey | whoot | 13:57 |
| mterry | tedg: is there a way for UAL to be able to tell an app its own AppID? | 14:14 |
| mterry | I think it sets APP_ID in Touch? But I don't think that's true in snappy | 14:15 |
| mterry | Because it doesn't really control that launch | 14:15 |
| mterry | Content Hub is relying on APP_ID right now | 14:15 |
| mterry | But maybe if there were an api it could call to get its appid from UAL? | 14:15 |
| mterry | tedg: I guess an app could do it now by iterating the running apps and checking if its own pid matches any of them | 14:20 |
| javier4 | I got this error trying to cross-compile wpa for armhf on a x86_64 host. | 14:27 |
| javier4 | "/usr/include/qt4/QtCore/qatomic_armv5.h:131: Error: no such instruction: `swpb %al,%bpl,[%rbx]'" | 14:27 |
| dobey | mterry: why would the app even need to know it's own full app id? | 14:27 |
| javier4 | Reading online it seems to be due to an incorrect setup of qmake. In particular it seems to not find the arm toolchain. | 14:27 |
| mterry | dobey: content hub uses it to talk to peers over dbus | 14:27 |
| javier4 | I'm totally ignorant in QT developing. The only file that I think could be a config for qmake is the wpa_gui.pro that qmake should use to generate its makefile. | 14:27 |
| javier4 | https://www.irccloud.com/pastebin/ni0ymJpB/ | 14:28 |
| dobey | mterry: ah so you mean the content-hub api that the app is using, needs to know the app id? | 14:31 |
| mterry | dobey: yeah, libcontenthub or whatever | 14:31 |
| dobey | mterry: i think kenvandine added some API in the last UAL landing specifically for that | 14:34 |
| dobey | or at least, to find the application by pid, and focus it | 14:35 |
| mterry | I see Application::findInstance(pid_t) | 14:35 |
| mterry | Which yeah I can call for each running application | 14:35 |
| dobey | yeah, and then Application::appId() gives you that | 14:36 |
| dobey | javier4: not sure what you're trying to do exactly there, but it looks like that is using qt4, and we don't use qt4 on the phone | 14:37 |
| tedg | mterry: for snaps they get everything but the app name as variables from snapd | 14:38 |
| tedg | mterry: apps don't generally have permission to use UAL, requires a bunch of access to systemd. | 14:38 |
| mterry | tedg: ah including the ability to list running apps? | 14:39 |
| mterry | makes sense if confined | 14:39 |
| tedg | mterry: I think we could still set the envvar, but would love to avoid it. | 14:40 |
| javier4 | dobey: I'm trying to build a customized version of wpa package. Even the standard one includes a qt gui. I don't care if it doesn't get used. | 14:40 |
| mterry | tedg: well I would propose that (A) we get snappy to expose the current command in env and (B) add an api to UAL to get the current appid (and then it composes those pieces together, so that apps don't need to know format of appid parts) | 14:40 |
| tedg | mterry: running app list is an information leak... could find out you have the Fox News app running and tell all your friends 😉 | 14:41 |
| dobey | javier4: disable building the qt gui then? | 14:41 |
| mterry | Yeah makes sense | 14:41 |
| javier4 | dobey, commenting out this target from main wpa_supplicant Makefile should be enough, right? | 14:45 |
| javier4 | wpa_gui-qt4: wpa_gui-qt4/Makefile wpa_gui-qt4/lang/wpa_gui_de.qm | 14:45 |
| javier4 | $(MAKE) -C wpa_gui-qt4 | 14:45 |
| kenvandine | tedg, lol... i swear i don't have a fox news app :) | 14:45 |
| kenvandine | mterry, yeah, the content-hub service gets the pid of the connecting app from dbus and calls findInstance(pid_t) | 14:46 |
| kenvandine | at least for where we need it in the service now | 14:46 |
| dobey | javier4: i guess so, yes | 14:47 |
| javier4 | dobey: I'll try. Thanks. | 14:47 |
| mterry | kenvandine: yeah I'm thinking app-side though, where it exposes its own appid path on dbus | 14:47 |
| kenvandine | mterry, yeah, i think I might need that too | 14:48 |
| dobey | hrmm | 14:49 |
| kenvandine | for libcontent-hub used on the app side | 14:49 |
| kenvandine | i think right now it checks APP_ID in the env | 14:49 |
| dobey | probably some issues there with having multiple versions of an app, too | 14:49 |
| mterry | kenvandine: right that's exactly where I'm looking actually :P (working on a patch to contenthub to use UAL for parsing appids instead of embedding format knowledge there) | 14:49 |
| kenvandine | mterry, excellent... that will be a welcome patch :) | 14:49 |
| kenvandine | mterry, i'm probably close to running into that issue :) | 14:50 |
| dobey | mterry: well you can just assemble the appid using ual API, if you have the three individual pieces | 14:51 |
| dobey | mterry: and it sounds like snapd already sets env vars that provide two of the pieces? | 14:51 |
| mterry | yeah you can call discover() and give the three pieces -- but you would have to know you're on snappy and what the env vars are | 14:52 |
| mterry | and do something else if you're not on snappy | 14:52 |
| dobey | well, on deb there is only one piece | 14:53 |
| mterry | Point is, it's a complicated landscape that apps shouldn't have to care about. UAL knows about whether an app is in libertine or whatever. apps shouldn't have to piece that together themsevles | 14:54 |
| dobey | sure, so i think you need to call findInstance() and work back from that then, and we need to solve the confinement problems there | 14:57 |
| taiebot | Did not realise but with the updated linux support i can make videocalls on skype now with the Skype (unofficial) app http://www.omgubuntu.co.uk/2017/03/skype-linux-beta-video-calling | 19:44 |
| === iahmad_ is now known as iahmad | ||
| === Elleo_ is now known as Elleo | ||
| === Jucato_ is now known as Jucato | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!