[09:15] <dpm> JamesTait, good morning! Quick question: which is/are the best project(s) in LP to report 1) issues with myapps' UI 2) mismatch between API keys used between clients and the store?
[09:28] <popey> dpm: 1) https://bugs.launchpad.net/software-center-agent  2) API keys?
[09:28] <dpm> JamesTait, popey, sorry, I meant API fields
[09:36] <JamesTait> dpm, myapps UI problems should be reported against software-canter-agent, as popey said.
[09:37] <dpm> ok cool. For the API fields, to give more context, it was about the conversation we had on e-mail recently
[09:38] <JamesTait> dpm, API fields should probably be reported against both the client and click-package-index.  Really it's a client problem, IMO, because the API predates all clients, but it might be easier to fix on the server.
[09:38] <dpm> re: mappying of api fields store <-> clients
[09:38] <JamesTait> Yeah, I assumed it was related to that. ☺
[09:40] <dpm> cool, thanks :)
[09:43] <dpm> JamesTait, also, I've got an issue whereby an app that's been just published is being seen by the client as an old revision. I guess I can send a query to the API to see which version it returns, so can I do this manually?
[09:52] <JamesTait> dpm, if you know the package name, curl -H 'X-Ubuntu-Channel: stable' https://search.apps.ubuntu.com/api/v1/package/{package_name} will give you the package metadata on the stable channel, with metadata for other channels embedded (under the _embedded["clickindex:channelpackage"] property).
[09:57] <dpm> JamesTait, yeah, the API seems to be returning the wrong (old) version: http://pastebin.ubuntu.com/15340450 - the latest published is 2.1+snap2, whereas it returns 2.1+snap0
[09:58] <dpm> JamesTait, the app has got uploads for multiple arches: rev 1 (published, 2.1+snap0, amd64), rev 2 (published, 2.1+snap2, amd64), rev 3 (*unpublished*, 2.1+snap2, i386)
[09:59] <dpm> I wonder if the latest i386 upload is obscuring the previous one for some reason, although it shouldn't do it, as it's a different architecture
[09:59] <dpm> and I'm querying from an amd64 machine
[10:05] <JamesTait> ETOOMANYCONVERSATIONS :-/
[10:08] <JamesTait> dpm, querying with -H 'X-Ubuntu-Architecture: i386' returns no results, which is what I'd expect because revisions 1 and 2 are amd64 only.
[10:09] <dpm> right
[10:11] <dpm> JamesTait, but it seems querying with 'X-Ubuntu-Architecture: amd64' returns revision 1 instead of the expected revision 2
[10:18] <JamesTait> dpm, right.  I can try a reindex from myapps to see if that refreshes it.  I'd like to understand why it's returning the old version, though.
[10:19] <dpm> JamesTait, I'm not sure if that could be related, but the automated review highlighted a checksum error, which neither I nor jdstrand could understand
[10:22] <JamesTait> dpm, gimme a few minutes to see if I can get to the bottom of what's actually in the backend, then I'll hit the reindex button.
[10:23] <dpm> awesome, thanks JamesTait!
[10:40] <JamesTait> dpm, I'm treading carefully so as not to destroy CPI again 😉  - but I do see two revisions in the backend: {u'version': [u'2.1+snap0'], u'revision': [1]} and {u'version': [u'2.1+snap2'], u'revision': [2]}.
[10:40] <dpm> is that good or bad? :)
[10:40] <dpm> does it mean the index just needs refreshing?
[10:41] <JamesTait> dpm, I'm going to take a copy of that data and index it locally to try to reproduce the issue, then reindex the package and see where things stand.
[10:41] <dpm> ok, cool
[10:41] <JamesTait> dpm, it could be an indication of a bug in selecting the "latest" revision for the package.
[10:41] <dpm> ok
[10:42] <dpm> let me know if you need me to file a bug
[10:46] <JamesTait> dpm, OK, versio 2.1+snap2 is showing for me now.
[11:10] <dpm> JamesTait, awesome. Did you figure out what it was?
[11:11] <JamesTait> dpm, still digging - currently comparing pre- and post-reindex data for obvious anomolies.
[11:12] <dpm> ok
[11:13] <JamesTait> dpm, interestingly, before the reindex, r1 of that package had a later modification timestamp than r2; after the reindex, r2 is later by ~0.03s
[11:14] <JamesTait> Not saying that's definitely the cause, but there aren't many differences.
[11:14] <dpm> oh, weird. r1 was uploaded last week, whereas r2 yesterday
[11:24] <aquarius> OK, confused. I have my E5 plugged in; in Ubuntu SDK, it shows in the Devices tab (as "Ubuntu Device") and says "Device Status: Ready to use". But I can't deploy to it (the deploy thing says it's not connected), and "Open SSH connection to the device" shows a terminal window for about half a second which then immediately closes. How can I go about working out what's wrong? also, it says that "Has network co
[11:24] <aquarius> nnection" is... disabled? And in the Log it says "error: device offline", although it is not offline (the browser works fine, for example)
[11:49] <t1mp> anyone knows what causes this?
[11:49] <t1mp> tim@tim-mbp:~/Downloads$ phablet-shell
[11:49] <t1mp> Host [localhost]:2222 not found in /home/tim/.ssh/known_hosts
[11:49] <t1mp> Warning: Permanently added '[localhost]:2222' (RSA) to the list of known hosts.
[11:49] <t1mp> Permission denied (publickey).
[11:53] <popey> run it again?
[11:53] <popey> For a new device I always have to run it twice
[12:01] <t1mp> popey: nope, it doesn't work. I ran it multiple times.
[12:01] <t1mp> popey: but indeed the device was recently flashed
[12:01] <popey> hm
[12:03] <t1mp> it appears to be a device issue since it works fine with my other device
[12:04] <t1mp> unless my ssh keys are wrong and they didn't need to be pushed again on the other device?
[12:45] <dpm> JamesTait, just as a heads up: we did an upload of ubuntu-clock-app.ubuntucoredev earlier on today, and for that one the index seems to be correct. Meaning clients can see the latest upload, so I guess that's good news
[12:49] <JamesTait> dpm, maybe... 😉   Things that fail all the time are usually easier to track down than things that only fail occasionally. 😃
[12:50] <dpm> indeed
[13:48] <aquarius> jhodapp, can I, from QML, give the media hub a playlist but tell it "stop playing after exactly 2 hours" even if the playlist is longer?
[13:49] <jhodapp> aquarius, sure, if you maintain the timer and your app stays in the foreground, you an just call stop on the player
[13:49] <jhodapp> or pause
[13:49] <aquarius> jhodapp, yeah, but I am assuming I'm not in the foreground -- that's why I'm handing you a playlist :)
[13:50] <aquarius> jhodapp, after two hours, the phone will have locked the screen and suspended my app.
[13:50] <jhodapp> other than that there's no way for media-hub to do that presently
[13:50] <aquarius> OK. So if I want the playlist to run for 2h, then I need to make sure it's 2h long :)
[13:50] <jhodapp> yup
[13:50] <aquarius> that's fine, no worries
[15:13] <t1mp> ahayzen: hello. Good timing :)
[15:13] <t1mp> ahayzen: I was just looking for you
[15:14] <t1mp> ahayzen: looks like I have a fix for https://bugs.launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/+bug/1554897
[15:14] <t1mp> ahayzen: is there a way you can test the fix?
[15:14]  * ahayzen ducks 
[15:14] <ahayzen> :-)
[15:14] <ahayzen> t1mp, if it is in a silo i can test on device ... or i can try the mini-app with the branch
[15:14] <ahayzen> just won't be perfectly sure, as it is super quick, but i can have a look :-)
[15:14] <t1mp> ahayzen: it is not in a silo, I committed the changes 1 minute ago
[15:14] <t1mp> ahayzen: yes, please do that
[15:15] <ahayzen> hehe :-)
[15:15] <ahayzen> t1mp, thanks for working on it so quickly :-)
[15:18] <zsombi> t1mp: one comment on the https://code.launchpad.net/~tpeeters/ubuntu-ui-toolkit/fasterWindowColor/+merge/288661
[15:19] <t1mp> zsombi: the line above your comment says: +protected Q_SLOTS:
[15:19] <t1mp> zsombi: is that what you want?
[15:20] <ahayzen> t1mp, how do you test the sdk on device? just branch on device then make and make install ? or do you only test on desktop ?
[15:20] <zsombi> t1mp: put the slot in the MainViewBasePrivate, that's what I want
[15:20] <ahayzen> oh there is a push_to_phone.sh ...
[15:21] <t1mp> ahayzen: the easiest way is to wait for CI to run the tests and generate the .deb packages that can run on the device
[15:21] <zsombi> ahayzen: hopefully works still :)
[15:21] <ahayzen> haha yeah :-)
[15:21] <t1mp> ahayzen: do you have results for the desktop already
[15:21] <ahayzen> t1mp, currently building :-)
[15:21] <zsombi> ahayzen: but that script only pushes the QML stuff
[15:21] <t1mp> zsombi, ahayzen: push_to_phone.sh will not work, because that only works for the updates qml files. This one needs to be compiled for ARM
[15:21] <ahayzen> zsombi, ah i see :-)
[15:21] <ahayzen> eh i'll test on desktop for now
[15:21] <t1mp> zsombi: ok, I got it.
[15:23] <ahayzen> t1mp, hmm i got this when building http://pastebin.ubuntu.com/15341641/ .. and then "module "Ubuntu.Components" plugin "UbuntuComponents" not found" when i try to run :-)
[15:24] <t1mp> oh shit
[15:24] <t1mp> zsombi: ^no isTabFence
[15:24] <t1mp> ahayzen: which ubuntu release are you running?
[15:24] <ahayzen> t1mp, Wily
[15:25] <t1mp> ahayzen: we have a patch for Qt that is only available for vivid (in the overlay ppa) and for xenial.
[15:25] <zsombi> ahayzen: brilliant, go to Xenial :D
[15:25] <ahayzen> should i somehow build under the sdk-dev thing ...
[15:25] <ahayzen> zsombi, i've got todo uni work on here rather not have Beta builds ;-)
[15:26] <zsombi> ahayzen: I know, was just sayin... :P
[15:26] <ahayzen> hehe ;-)
[15:26] <ahayzen> 'soon' :-)
[15:26] <ahayzen> maybe i can just comment that out for now? as i'm not testing that part anyway
[15:30] <ahayzen> yeah commenting that line out worked :-)
[15:31] <ahayzen> t1mp, from what i can see with the mini-app on the desktop it looks fixed, but i'd like to be perfectly sure with music on a device once there is a deb/silo
[15:31] <t1mp> ahayzen: cool. How did you test it?
[15:31] <t1mp> ahayzen: okay, I will let you know when the deb packages are done for that MR
[15:31] <ahayzen> launching the mini-app like 20 times and staring at the screen to see if the app goes white before dark grey
[15:32] <ahayzen> (on the desktop, and to get around the build failure i just commented out that line as it was todo with listitems)
[15:32] <ahayzen> thanks :-)
[15:32] <t1mp> ahayzen: ah, nice. :) I was wondering how you got around the build failure without upgrading your system :)
[15:32] <ahayzen> :-)
[15:33] <t1mp> ahayzen: in a month or so, xenial won't be beta any more. So then you can safely upgrade.
[15:33] <ahayzen> yeah, although it'll probably be Mayish before i do, as i have exams and presentations todo from this laptop
[15:33] <t1mp> and in case you are interested, I installed xenial last week and so far it has been totally stable for me.
[15:33] <t1mp> ahayzen: uhh, yeah. That sounds like something important :)
[15:34] <ahayzen> from what i've heard it is stable :-) but Uni work so :-/ maybe if i get a spare weekend or something but unlikely
[15:53] <ahayzen> t1mp, do you think this will also fix the issue of the scopes/indicators initially being white on boot? or are they done differently ?
[16:39] <t1mp> ahayzen: I haven't checked. But I assume that they don't use MainView.
[16:39] <ahayzen> yeah same :-/
[16:39] <t1mp> ahayzen: the way to fix them may be the same as the fix for MainView though.
[16:39] <t1mp> I never noticed that they are white on boot
[16:40] <ahayzen> wonder why they are white at boot then, maybe the mako is slow
[16:40] <t1mp> if other devices are faster, that only means the issue is less visible, not that it doesn't exist
[16:40] <t1mp> let me reboot my mako to see
[16:41] <ahayzen> :-)
[16:42] <ahayzen> when booting i see the indicator bar go white then black ... then if i unlock the device the whole section is white then the bg and "Today" appears where the scope goes
[16:42] <ahayzen> don't remember those happening pre-silo050 but maybe i'm just looking for it after the music issues :-)
[16:43] <t1mp> I don't see it here. Not sure which version I have installed though
[16:43] <ahayzen> i just flashed the latest rc-proposed and saw it with the silo
[18:03] <bregma> is there any sort of file chooser API for apps under Ubuntu Touch?
[18:05] <ahayzen> bregma, do you mean something like content-hub ?
[18:06] <ahayzen> https://developer.ubuntu.com/api/apps/qml/sdk-15.04.1/Ubuntu.Content.index/
[18:57] <balloons> popey, so looking at https://core-apps-jenkins.ubuntu.com/job/filemanager-app-ci/53/console, it seems it suffers from the same issue as docviewer, but succeeds. Should this also be wiping the download cache? It looks like it's using the cache, and thus is perhaps why it's working a-ok
[19:00] <balloons> I'm going to assume yes, and wipe the cache unless I hear otherwise
[19:26] <popey> balloons: yeah