[00:35] <josharenson> If I've pushed an amd64 snap to the store's stable channel, and later, push an armhf snap to the stable channel, it won't clobber the amd one I assume?
[03:59] <enwei> davidcalle, ping
[07:20] <om26er> Hello! How can I make available non-pypi packages to my name ? e.g. python3-gi is not available on pypi but its in Ubuntu, so how can I make that available to my snap ?
[07:36] <DanChapman> om26er: you could possibly use the "nil" plugin to get python3-gi in the stage-packages
[07:36] <DanChapman> there might be a better way. That's just what comes to mind
[07:37] <om26er> DanChapman: I just added stage-packages to my python part and it seems to have progressed
[07:56] <dholbach> good morning
[09:16] <tsdgeos> $ sudo snap remove unity8-session
[09:16] <tsdgeos> error: cannot remove "unity8-session": snap "unity8-session" has changes in progress
[09:16] <tsdgeos> what do i do now?
[09:22] <tsdgeos> ok snap changes + snap abort
[09:26] <tsdgeos> and then i had to mount something in the folder it thought there was something supposde to be there
[09:26] <tsdgeos> oh well
[09:55] <davmor2> pitti: hey dude console-conf how do I get a login setting all the ethernet connections to none?
[09:56] <davmor2> sorry network setting even
[09:58] <pitti> davmor2: sorry, wrong guy for console-conf; you want mwhudson
[10:00] <davmor2> pitti: ta
[11:02] <morphis_> sergiusens: ping
[11:42] <Mirv> yay, someone (tm) fixed the store for the content interface snaps, now ubuntu-app-platform is automatically up-to-date in the store
[11:43] <Mirv> that means there's also now first arm64 version :)
[12:03] <sergiusens> morphis pong
[12:10] <renato__> Mirv, good morning.
[12:10] <renato__> Mirv, any luck with the calculator?
[12:27] <sergiusens> jdstrand hey, with stgraber we chose to use `:` (originally `/`) as a filter per architecture, not a request to install that package.
[12:27] <sergiusens> jdstrand sidelined and fixable, this is another reason I don't want crt to fail on non matching architetures (although my case is arch all forced to arch specific)
[12:38] <davmor2> pitti, mwhudson: https://bugs.launchpad.net/ubuntu/+source/subiquity/+bug/1641110
[12:38] <mup> Bug #1641110: Setting both ipv4 and ipv6 to none on a connection still gives you an ip setup <subiquity (Ubuntu):New> <https://launchpad.net/bugs/1641110>
[12:38] <davmor2> pitti: included you incase you need to be in the discussion on what to do
[12:42] <ogra_> davmor2, i bet the connection is gone after a reboot ;)
[12:43] <ogra_> (there is a bug about "netplan apply" on actually restarting the connection with the new setup)
[12:43] <ogra_> s/on/not/
[12:43] <davmor2> ogra_: nope you can ssh into the system on reboot
[12:43] <ogra_> oh, really ?
[12:43] <ogra_> then it is different
[12:44] <davmor2> ogra_: yes set step 8
[12:44] <davmor2> see even
[12:44] <ogra_> step 8 is without reboot
[12:47] <davmor2> ogra_: d'oh let me double check it then I thought I had rebooted with it but let me double check
[12:53] <davmor2> ogra_: indeed \o/ was there another bug I can link this one too, it would explain the issue I was having with the default route too I guess right (but that is supposed to be gone in the new version)
[12:55] <ogra_> davmor2, i know there is one but i cant seem to find it :(
[12:56] <davmor2> ogra_: oh well I'll leave mine they can always link it
[12:56] <ogra_> yeah
[12:56] <ogra_> if i find my old one i'll merge them
[12:56] <ogra_> just not the findings above in yours too
[12:56] <ogra_> *note
[12:57] <davmor2> ogra_: just did on step 9
[12:57] <ogra_> :)
[12:58] <Son_Goku> sergiusens, I threw a small update into LP#1602258
[12:58] <Son_Goku> hmm
[12:58] <Son_Goku> no lp bot thingy
[12:58]  * sergiusens looks
[12:58] <sergiusens> Son_Goku it is a matter of spacing LP: #1602258
[12:58] <mup> Bug #1602258: Support other distributions as sources (Fedora, Mageia, openSUSE, Debian) instead of Ubuntu <centos> <debian> <fedora> <mageia> <opensuse> <rfe> <rpm> <Snapcraft:Confirmed> <https://launchpad.net/bugs/1602258>
[12:58] <Son_Goku> if there's anything I'm forgetting, let me know
[12:58] <Son_Goku> ah
[12:58] <sergiusens> or maybe bug  #1602258
[12:59] <sergiusens> nope, just the former
[12:59] <Son_Goku> :/
[13:01] <sergiusens> Son_Goku some other forms work too, I am just too relaxed this week to think ;-)
[13:01] <Son_Goku> I just wanted to make sure I got the highlights right
[13:01] <Son_Goku> but we're not ready to start quibbling about it yet :P
[13:03] <sergiusens> right, the general idea will change until it makes its form in code, even when playing with a potential PR it might feel wonky :-)
[13:03] <Son_Goku> yep
[13:04] <ogra_> sergiusens, hmm ... does snapcraft not take apps in the potential PATH into account at build time ? http://paste.ubuntu.com/23460983/
[13:05] <ogra_> (the binary gets copied to bin/ in the target ... the check seems to only accept it if i give the full path for "command:", even though bin/ will be in $PATH)
[13:08] <dholbach> hey... does anyone have an idea how I'd resolve something like this?
[13:08] <dholbach> go get -t -d ./github.com/containous/traefik/...
[13:08] <dholbach> github.com/containous/traefik/web.go:13:2: no buildable Go source files in /home/daniel/traefik/parts/traefik/go/src/github.com/containous/traefik/autogen
[13:12] <dholbach> ok, looks like something running another command is required before running the build
[13:22] <sergiusens> ogra_ can you add more details to your statement?
[13:23] <ogra_> sergiusens, dunno, what more details do you need ? the dump plugin copies the script to bin/ ... then something seems to check the "command:" content if the script exists and is executable
[13:24] <ogra_> (and fails because it does not seem to look in bin/ by default)
[13:24] <sergiusens> ogra_ this should work, give me a snapcraft.yaml and I'll check
[13:24] <ogra_> sergiusens, https://code.launchpad.net/~ogra/+junk/kodi-pi
[13:25] <ogra_> (beware, the tree is huge)
[13:25] <ogra_> if you want a full build log (huge as well :) ) https://code.launchpad.net/~ogra/+snap/kodi-pi
[13:26] <ogra_> i'm probably totally wrong about that message though ... the above is what i interpret from the message it gave me
[13:27] <ogra_> (i just changed "command: runner_kodi" to "command: bin/runner_kodi" in the last commit in that bzr tree)
[13:28] <ogra_> (the build for that change is still running)
[13:28] <sergiusens> ogra_ huge trees should go to git! ;-)
[13:29] <ogra_> yeah yeah ...
[13:30] <sergiusens> ogra_ if you dump `.` you get everything
[13:31] <sergiusens> ogra_ do something like this instead https://github.com/morphis/plexmediaserver-snap/pull/3/files
[13:31] <mup> PR morphis/plexmediaserver-snap#3: Update to use latest snapcraft features <Created by sergiusens> <https://github.com/morphis/plexmediaserver-snap/pull/3>
[13:31] <ogra_> sergiusens, but that means the script needs to live in bin/ in the source, right =
[13:31] <ogra_> ?
[13:32] <sergiusens> ogra_ yeah, is that a problem?
[13:32] <ogra_> nah, just wanted to make sure ... a test build takes 1.5h :)
[13:32] <ogra_> i still dont get why we couldnt keep the copy plugin around ...
[13:33] <ogra_> it was easy to understand ... for dump i need to ask someone nearly every time i use it ... way to complex
[13:35] <sergiusens> ogra_ soon these wrapper scrips will be a thing of the past ;-)
[13:35] <sergiusens> oh that's a big one though :-P
[13:35] <ogra_> i think thats an upstream one
[13:36] <ogra_> that isnt originally my tree ... i'm just trying to make it build in LP currently
[13:36] <ogra_> (with plenty of awful hacks (as usual :P ))
[13:37]  * ogra_ fires off another build
[13:37] <dholbach> I was just chatting to somebody who has build artifacts for four architectures which they would probably just dump into a snap - what would be the best way to snap and publish them?
[13:37] <dholbach> Or rather... dump into four different snaps.
[13:40] <ogra_> dholbach, one way is to wrap it into a Makefile that checks the build arch ... but i guess sergiusens would call that cheating ;)
[13:40] <ogra_> (checks and acts arch specific then)
[13:41] <sergiusens> I was working on an idea of per architecture parts but we are scaling down on features as people are starting to get lost in a sea of options
[13:41] <sergiusens> dholbach you can just create for parts and dump it into the snap
[13:41] <ogra_> dholbach, https://git.launchpad.net/~ubuntu-kernel/ubuntu/+source/linux-snap/+git/xenial/tree/Makefile?id=ed92e3a0272c85c2640ca42322e799fcf69a9034 is an example
[13:41] <sergiusens> set architectures in snapcraft.yaml to the corresponding ones
[13:42] <ogra_> yeah, that would be helpful
[13:42] <sergiusens> ogra_ well not going to happen for the time being, Makefiles or build systems is the way
[13:42] <abeato> sergiusens, hey, noticed this: bug #1641123
[13:43] <mup> Bug #1641123: snapcraft does not properly detect the architecture when kernel is 64 bits but userspace is 32 <Snapcraft:New> <https://launchpad.net/bugs/1641123>
[13:43] <ogra_> but a good thing to have on the TODO
[13:43] <sergiusens> ogra_ it is labeled NO in the inbox of tasks ;-)
[13:43] <ogra_> :(
[13:43] <sergiusens> NOs can be YES in the future, albeit distant
[13:45] <sergiusens> abeato force the architecture as I mentioned just above
[13:46] <abeato> sergiusens, hmm do not see that
[13:46] <ogra_> just "architecture: i386" in your snapcraft.yaml
[13:46] <abeato> oh, I see, thanks
[13:53] <sergiusens> ogra_ abeato architectures: [armhf]
[13:53] <Mirv> renato__: no, getting qml: [LOG]: Unable to calculate formula : "98+66", math.js: SyntaxError: Parse error
[13:53] <renato__> Mirv, ok I am trying to discovery which file is missing
[13:54] <abeato> sergiusens, ack
[13:56] <ogra_> oops, indeed
[13:57] <Mirv> renato__: actually, I think it might be that you're testing it (when not using platform snap) without xenial overlay, would I be right?
[13:57] <dholbach> sergiusens, do we have an example of that?
[13:57] <Mirv> renato__: in that case it wouldn't be platform snap issue, but the fact that calculator doesn't work properly with Qt 5.6. I've existing bug about it.
[13:57] <Mirv> renato__: if you're using stock xenial, you have Qt 5.5
[13:58] <renato__> Mirv, could be, let me try to pack it with overlay
[13:58] <renato__> I have a vm with that
[14:01] <dholbach> sergiusens, ogra_, just so I understand it correctly: one architectures: [armhf, ...] entry and then a specific  architecture: armhf  entry in the relevant part?
[14:01] <renato__> Mirv, yes you are right the problem is related with qt version
[14:01] <ogra_> dholbach, no, that was what sergiusens mentioned as possible future plans
[14:01] <sergiusens> dholbach yeah, you are just dumping binaries, right?
[14:02] <dholbach> sergiusens, that's how I understood them, yes
[14:02] <renato__> popey, ^^
[14:02] <popey> ah
[14:02] <renato__> popey, who can help us on that? Looks like the mathjs does not work with qt5.7
[14:02] <popey> is it an upstream bug with math.js? if so, I'd file a bug on their github page
[14:02] <popey> they're pretty responsive
[14:03] <renato__> popey, https://bugs.launchpad.net/ubuntu-calculator-app/+bug/1620333
[14:03] <mup> Bug #1620333: calculator doesn't calculate: parse error in math.js <Canonical System Image:Confirmed for popey> <Ubuntu Calculator App:Incomplete> <https://launchpad.net/bugs/1620333>
[14:03] <popey> I mean, is it a bug in the upstream math.js library we're using?
[14:03] <renato__> popey, I can try create a small test if that helps
[14:03] <popey> or is it a bug in qt?
[14:04] <popey> it probably would, yes
[14:04] <renato__> popey, I think this is a qt bug. But honestly I do not even know where to start :D
[14:04] <popey> would file upstream at https://github.com/josdejong/mathjs/issues
[14:05] <popey> heh, me neither!
[14:05] <popey> need a qt person
[14:05] <renato__> popey, we can try report a bug on mathjs project to see if they give us any hint
[14:07] <Mirv> renato__: ok, so practically calculator "works", it's just it has problems with Qt version. great!
[14:07] <renato__> Mirv, looks like that
[14:07] <popey> yeah
[14:24] <ogra_> YAY !
[14:24] <ogra_> Priming kodi
[14:24] <ogra_> Snapping 'kodi-mir' ...
[14:36] <mup> Bug #1641132 opened: no way to include devmode snaps in snap prepare-image? <Snappy:New> <https://launchpad.net/bugs/1641132>
[14:51] <mup> PR snapcraft#898 opened: Revert "Allow for architecture-specific packages (#876)" <Created by sergiusens> <https://github.com/snapcore/snapcraft/pull/898>
[14:53] <Suraj311> Plz help me
[14:53] <Suraj311> my pi 3 keep asking about connection
[14:54] <Suraj311> How to configure wlan at first time boot with ubuntu core
[14:54] <ogra_> you can not (as noted in the release notes)
[14:54] <Suraj311> is there any windowX environment for configuring the network
[14:54] <Suraj311> ?
[14:55] <ogra_> configure it with cable attached ... then log in and run "sudo console-conf" to re-configure it for wifi (adn turn off ethernet)
[14:55] <Suraj311> i don't have LAN
[14:55] <Suraj311> i only have mobile internet
[14:55] <Suraj311> i am using internet by hotspot
[14:55] <ogra_> well, there is currently no way to configure wlan on first boot on a pi3 ...
[14:56] <ogra_> (there is a bug with the driver)
[14:56] <Suraj311> is there any way by tethering?
[14:56] <Suraj311> usb tethring?
[14:56] <ogra_> nope
[14:56] <Suraj311> :(
[14:57] <ogra_> you need ethernet ... the next image releasse will hopefully fix this
[14:57] <Suraj311> then is ubuntu mate will be good for IOT development?
[14:57] <ogra_> why not :)
[14:58] <Suraj311> Ok thanx Ogra_
[14:58] <Suraj311> will go for mate
[14:58] <ogra_> (i dont think it matters what you use on your desktop for IoT development ...)
[14:59] <mup> PR snapd#2270 opened: store: check payment method before TOS for a better UX <Created by pete-woods> <https://github.com/snapcore/snapd/pull/2270>
[16:06] <mup> Bug #1641150 opened: snap hooks are not run with environment similar to apps: PATH, LD_LIBRARY_PATH <Snappy:New> <https://launchpad.net/bugs/1641150>
[16:09] <Bits4Bytes> Can you package closed source programs into snaps?
[16:09] <ogra_> why woldnt you
[16:10] <ogra_> snaps are binary packages, they dont really care ...
[16:10] <qengho> Bits4Bytes: Yes! That's one of the goals!
[16:10] <ogra_> if you create an account in the store you need to sign that you have the distribution rights for the software you upload though ...
[16:11] <qengho> Bits4Bytes: It used to be that to be safe, someone had to review code put into Ubuntu and others. The isolation of snaps lets Ubuntu+ get out of the way and let people publish without hurting users.
[16:11] <ogra_> (so just grabbing random closed source binaries off the internet will likely not work if they are not distributable legally)
[16:12] <Bits4Bytes> Not going to publish it. I have a java program that I'd like to run on a Snappy Ubuntu Core on a Rasberry Pi.
[16:59] <dholbach> have a great weekend everyone!
[17:12] <popey> https://www.gl-inet.com/mt300a/ is a neat thing (which would probably be nice with snappy)
[17:13] <popey> ooh, they sell them on amazon
[17:13] <popey> https://www.amazon.co.uk/GL-MT300A-Ext-MicroSD-pre-installed-Repeater-Tethering/dp/B01EWCO7CS
[19:24] <mup> PR snapcraft#898 closed: Revert "Allow for architecture-specific packages (#876)" <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/898>
[19:33] <mup> PR snapcraft#899 opened: Release changelog for 2.22.1 <Created by sergiusens> <https://github.com/snapcore/snapcraft/pull/899>
[20:42] <mup> PR snapcraft#899 closed: Release changelog for 2.22.1 <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/899>
[21:00] <mhall119> zyga: can you tell me what I'm doing wrong with my configure hook? http://paste.ubuntu.com/23462794/
[21:36] <qengho> mhall119: I just fought this. I think your snapd is not new enough.
[21:36] <qengho> mhall119: and that you're on snappy on classic, amirite?
[21:39] <mhall119> right, on 16.04
[21:39] <qengho> mhall119: You just gotta wait. :(
[21:39] <qengho> ...I think.
[21:40] <mhall119> that's not my forte :)
[21:40] <qengho> mhall119: You just gotta compile it on your own. :)
[21:40] <mhall119> that's also not my forte :)
[21:41] <qengho> mhall119: You just gotta complain to zyga until he cracks.
[21:42] <qengho> mhall119: I have two packages that use it. Fine on a snappy core box.
[21:42] <qengho> mhall119: "sshesame" and "tor-middle-relay" in candidate channels.
[21:44] <qengho> mhall119: http://bazaar.launchpad.net/~privacy-squad/+junk/sshesame-snap/view/head:/snapcraft.yaml#L18
[21:47] <mhall119> not complaining to zyga, that's something I'm good at
[21:47] <mhall119> s/not/now/
[22:20] <CountryfiedLinux> howdy
[22:21] <CountryfiedLinux> Will there be a time when an Ubuntu install is all snappy packages by default?
[22:21] <CountryfiedLinux> Right now it's up to the user to figure out which packages have snaps. It would be nice for there to at least be a snappy edition.
[22:22] <qengho> CountryfiedLinux: Not in foreseeable future. Some things may transition to snapps altogether, like maybe libreoffice. But there will be deb for a long long time.
[22:23] <qengho> Some deb could be transitional empty packages that install snaps, but others will be deb "forever".
[22:24] <CountryfiedLinux> qengho, Perhaps someone will take the initiative of making a Snappy edition with a graphical Snappy Store. Maybe :)