[05:22] <pitti> Good morning
[07:03] <didrocks> good morning
[07:09] <pitti> bonjour didrocks, ça va ?
[07:09] <duflu> pitti: Hello. And thanks again for the headache fix
[07:10] <pitti> hey duflu; no worries, thanks for pointing it out
[07:11] <pitti> didrocks: je suis déjà dans le train à België :)
[07:11] <didrocks> pitti: ça va, et toi ? :)
[07:11] <didrocks> pitti: tu arrives à quelle heure ?
[07:11] <pitti> didrocks: présque 14h30
[07:12]  * duflu would also like to ask for a translator in advance when ordering biere sans gluten 
[07:12] <duflu> :)
[07:12] <didrocks> tu as le temps d'être confortablement installé :)
[07:12] <didrocks> duflu: no problem, stay with us :)
[07:12] <duflu> Thanks didrocks. See you soon
[07:13] <pitti> duflu: ah, tu vais être à Bruxelles aussi ?
[07:13] <duflu> pitti: oui
[07:13] <duflu> I think
[07:13] <didrocks> duflu: are you coming to FOSDEM as well, or just the sprint?
[07:13] <pitti> duflu: alors, bon voyage !
[07:13] <duflu> didrocks: Just the sprint
[07:59] <mlankhorst> Hello, world!
[08:00] <didrocks> morning mlankhorst
[08:00] <mlankhorst> Morning! :-)
[08:01] <seb128> good morning desktopers!
[08:05] <didrocks> re seb128
[08:05] <seb128> lut didrocks :-)
[08:07] <Mo_1080> The following problem: I have CUPS running on Raspian as a print server. On Windows, everything works fine. Ubuntu 14.04 has also worked on the last laptop. Only on my current laptop (Ubuntu 4.14 also) I can speak to the printer, but get endless blank pages. Can someone help me?
[08:10] <seb128> Mo_1080, hey, try #ubuntu for user support
[08:11] <Mo_1080> tried that, but thanks
[08:29] <pitti> didrocks: I sent my review for the first patch FYI
[08:33] <didrocks> pitti: got it! reading, thanks :)
[08:38]  * didrocks can see that pitti has a similar nitpick on trailing \ as I do on trailing , on the Deps/Build-deps in packaging :)
[08:39] <didrocks> pitti: I agree with you, but seems that's not the style used in Makefile.am (having the last $(NULL)), would you do it anyway?
[08:41] <pitti> didrocks: I started doing that only recently
[08:41] <pitti> there are some 4 precedents, so it shold be fine
[08:42] <didrocks> pitti: ok, doing then! I'm still voting strongly in favor of the same behavior in debian/control :)
[08:43] <willcooke> mornin'
[08:43] <didrocks> hey willcooke
[08:44] <willcooke> larsu, I've got some meetings this morning, once they're done I'll see if I have time to get over to Cambridge.  I'm on school run today, so need to be home by 1500
[08:44] <pitti> didrocks: so again we talked most about the least important issue :)
[08:45] <Sweet5hark1> moin
[08:45] <larsu> willcooke: morning! We're here, let us know if you can make it ;)
[08:45] <larsu> morning didrocks, Sweet5hark1
[08:49] <didrocks> pitti: ahah, I'm continuing browsing the changes and answering some more by emails
[08:49] <didrocks> hey larsu, Sweet5hark1
[08:49] <pitti> didrocks: (just kidding)
[08:49] <pitti> hey Sweet5hark1, wie gehts?
[08:49] <pitti> hallo larsu
[08:49] <larsu> hi pitti! how goes?
[08:50] <pitti> larsu: fast! (ICE to Bruessel)
[08:50] <pitti> well, to Cologne first, then the Thalys, but same difference
[08:50] <larsu> pitti: nice! Ya, I was about to ask - hackfest doesn't start until Friday ;)
[08:50] <larsu> pitti: we'll arrive on Thursday evening
[08:50] <Sweet5hark1> pitti: excited, landed libreoffice 4.4.0~rc2 in -proposed a few hours ago ;)
[08:51] <pitti> larsu: I'll visit some friends in Leuven until then, and work from there
[08:51] <pitti> Sweet5hark1: /me feeds the buildd hamsters
[08:51] <larsu> pitti: cool, enjoy!
[08:51] <Sweet5hark1> pitti: they did hard work in cold winter ;)
[08:51] <seb128> Sweet5hark1, I promoted the packages from the other bug yesterday but didn't see that others need promoted as well, doing that now
[08:53]  * Sweet5hark1 will arrive on Friday in Bruessels with a weird Flightplan: HAM->CPH->BRU
[08:53] <Sweet5hark1> seb128: aye
[08:59] <mlankhorst> Sweet5hark1: not by car? :P
[09:01] <Sweet5hark1> mlankhorst: there was the option to travel with the LibreOffice-hackers-and-FSFE-folks-car track from Hamburg, but a/ I learned to late about that b/ It wouldnt fit with the extra days for Hackfests etc.
[09:01] <mlankhorst> and one way trips are more expensive than roundways :/
[09:02] <mlankhorst> I never understood why..
[09:03] <Laney> yo
[09:04] <larsu> yo yo
[09:05] <seb128> hey Laney, how are you?
[09:06] <seb128> Laney, happy birthday!
[09:07] <larsu> oh right, happy birthday Laney!
[09:07] <pitti> hey Laney, good morning
[09:07] <pitti> Laney: HAPPY BIRTHDAY!
[09:07]  * pitti hugs Laney and throws some candy
[09:08] <didrocks> pitti: answered and did the changes along the way! Now, looking at the DGRAM part
[09:08] <didrocks> hey, happy birthday Laney! (g+ betrayed you :))
[09:09] <Laney> thanks seb128 larsu pitti didrocks ;-)
[09:09] <pitti> didrocks: I feel a bit bad about proposing these big changes, but I honestly think it's easier, smaller, and more robust
[09:10] <didrocks> pitti: yeah, I don't mind dropping this buffer which can enlarge as requested to pass the structure right away instead
[09:11] <didrocks> pitti: I need to check about the blocking vs non blocking and do it
[09:16] <pitti> didrocks: (non)blocking shouldn't behave any different with stream vs. dgram and read() vs. recv()
[09:19] <didrocks> pitti: I'm trying to find a good example (most of them talk about getting back to TCP and doing serialization/deserialization)
[09:20] <pitti> didrocks: (non)blocking shouldn't behave any different with stream vs. dgram and read() vs. recv()
[09:22] <didrocks> pitti: I'm trying to find a good example (most of them talk about getting back to TCP and doing
[09:22] <didrocks> serialization/deserialization)
[09:22] <didrocks> (recopying back due to your lag)
[09:24] <pitti> didrocks: perhaps udev_monitor_receive_device() in src/libudev/libudev-monitor.c ?
[09:24] <pitti> didrocks: but it's really quite straightforward
[09:25] <pitti> didrocks: except for flags (where you can pass MSG_DONTWAIT for non-blocking) it has the same signature as read()
[09:25] <didrocks> ok, with recvmsg() then
[09:25] <pitti> didrocks: and correspondingly, send() and write() are pretty much the same, just that one is for dgram, the other for stream
[09:26] <didrocks> pitti: ok, giving it a try, thanks!
[09:26] <pitti> didrocks: if you have a connected socket, use recv() (easier); for an unconnected socket recvmsg()
[09:26] <didrocks> ah ok ;)
[09:27] <pitti> didrocks: the big advantage is that you don't have to worry about message sizes and boundaries
[09:28] <didrocks> pitti: yeah, I just allocate the structure and pass as a parameter, I see
[09:28] <pitti> didrocks: you send/receive one struct at a time, don't have to decode, etc.
[09:29] <didrocks> pitti: what happens as the socket is non blocking? recv() will still get the size I guess, and so, only get one struct?
[09:29] <didrocks> pitti: I'll need to adapt the send back though (using this socket to send cancel messages as a string)
[09:30] <pitti> didrocks: right, just tell it you want to get sizeof(struct progress), you'll get one at a time
[09:31] <pitti> didrocks: ah, perhaps define a "command" enum for the fsckd -> fsck direction, and send that?
[09:31] <pitti> didrocks: (if that later needs some extensions, it could become a full "command struct" again)
[09:31] <didrocks> pitti: yeah, will need to do this
[09:31] <pitti> didrocks: dgram just means you have to do exactly as many recv()s than send()s; as long as you know what kind of data you expect, you can send different lengths
[09:32] <didrocks> pitti: ok, giving it a try
[09:33] <mlankhorst> ahhh found my libreoffice on desktop bug
[09:33] <mlankhorst> how silly
[09:33] <pitti> didrocks: ply_boot_client_tell_daemon_to_display_message()... why did they have so incredibly short and hard to read identifiers? :-)
[09:34] <pitti> (could it be any longer?)
[09:34] <didrocks> pitti: I found some longer one, if you are a fan :p
[09:34] <didrocks> pitti: but yeah for ply_boot_client_* \o/
[09:34] <didrocks> ones*
[09:36] <mlankhorst> someone was abusing the api, long live asserts
[09:44] <didrocks> pitti: I need to have a fixed length then for device name
[09:44] <pitti> didrocks: hm, PATH_MAX?
[09:45] <didrocks> pitti: yeah, using that already in some other place
[09:46] <pitti> _ask_daemon() vs. _tell_daemon() ... *sheesh*
[09:47]  * pitti wants to add _if_you_please() and _nudge_daemon(), _shout_at_daemon() variants
[09:49] <didrocks> pitti: seeing the activity on plymouth upstream git, you may get maintainership role then, I would think twice, even for the sake of joke! :)
[10:27] <pitti> didrocks: ok, I'm through with the first round
[10:28] <didrocks> pitti: thanks a lot! I'm looking on how to listen on the dgram, and then, once all done, will resend you a second version with the proposed enhancement
[10:29] <didrocks> pitti: listen() is not supported on dgram socket, but I need to detect connexions, so, let's see how it goes
[10:31] <didrocks> pitti: as I can't listen, it's easy to add new ones as they come, but unsure how to detect drop connexion…
[10:36] <pitti> didrocks: oh, I wasn't aware of that; do you need this?
[10:36] <pitti> didrocks: you already have a timeout after all, and if a fsckd is at 100% and just stops sending stuff that should be ok? or am I missing something?
[10:40] <pitti> didrocks: if dgram becomes too unwieldy for some reason, you might also combine the approaches
[10:40] <pitti> didrocks: i. e. keep STREAM and use read()/write() to send structs instead of strings
[10:40] <pitti> didrocks: it's a little less robust since anyone sending garbage will break everything, but *shrug*
[10:41] <pitti> and read/write can be interrupted and only be done partially, which is ugly for structs
[10:44] <mlankhorst> compiz (core) - Info: Unity is not supported by your hardware. Enabling software rendering instead (slow).
[10:44] <mlankhorst> compiz (core) - Info: Unity is fully supported by your hardware.
[10:44] <mlankhorst> :/
[10:58] <willcooke> heh
[11:32] <brainwash> tedg: is https://code.launchpad.net/~ted/indicator-application/lp1411959-xfce-autostart/+merge/247051 a temporary build error?
[11:59] <didrocks> pitti: well, if a fsck stop sending stuff, it's blocking if I can't get the disconnect
[12:00] <didrocks> pitti: yeah, seems that sending read/write for struct is more risky than the existing approach
[12:00] <pitti> didrocks: oh, what does it block?
[12:00] <didrocks> pitti: well, I only timeout right now if number_device == 0 + timeout
[12:00] <pitti> ah
[12:00] <didrocks> pitti: I can get a timeout per device, but I need to register last message for each
[12:01] <pitti> so fsckd doesn't send a final "I'm done" message? (which could just be that curr >= max)
[12:01] <didrocks> pitti: at least, I register with the current approach exactly when a connection is drop
[12:01] <didrocks> pitti: sending a final message is fine, but the issue is more about one process dying
[12:01] <pitti> didrocks: *nod*, ok; that didn't seem crucial to me, but if it is, then we can discuss that on the ML too
[12:02] <didrocks> pitti: yeah, I think I'll keep the current code adapating to socket, and just modify to send cur, max instead of percentage
[12:02] <didrocks> adapting*
[12:02] <didrocks> pitti: in case I need to get that out again :)
[12:13] <Laney> tkamppeter: do you have a VCS for s-c-p?
[12:40] <seb128> Sweet5hark1, https://launchpadlibrarian.net/195930264/buildlog_ubuntu-vivid-arm64.libreoffice_1%3A4.4.0~rc2-0ubuntu1_FAILEDTOBUILD.txt.gz
[12:40] <seb128> Sweet5hark1, it fails to build
[12:40] <seb128> Sweet5hark1, :-(
[12:46] <mlankhorst> hooray, flipping working as intended
[12:50] <mlankhorst> oops, FTL still broken :P
[12:56] <Laney> larsu: have you looked into gnome-system-monitor black bg previously?
[12:57] <mlankhorst> had 2 different clipping bugs!
[13:04] <willcooke> http://notyetthere.org/unity8-windowed-mode/
[13:04] <willcooke> \o/
[13:10] <Sweet5hark1> seb128: yeah. well, it doesnt fail on the platforms we always had for LibreOffice: arm64 and ppc64el are new ...
[13:10] <mlankhorst> uploading +sa30, it needs some testing but I think I fixed all glitches I've had on desktop..
[13:10] <willcooke> just relashing my phone to try the windowed mode
[13:11] <mlankhorst> there are 4 different ways of getting contents to the screen :P
[13:11] <Sweet5hark1> seb128: dammit. forgot to remove those additional archs.
[13:12] <mlankhorst> but now I mastered them!
[13:13] <Sweet5hark1> seb128: bumping to a version with those removed right now.
[13:13] <Laney> Sweet5hark1: if you didn't have a build there before then proposed-migration won't complain
[13:13] <Sweet5hark1> Laney: https://launchpad.net/ubuntu/+source/libreoffice/1:4.3.3-0ubuntu3 did not have a build there ...
[13:15] <Sweet5hark1> Laney, seb128: so an upload only breaking on new archs (never build before) should work in general?
[13:15] <Laney> all good then, at least as far as proposed-migration is concerned
[13:15] <Sweet5hark1> seb128: ^^
[13:16] <mlankhorst> willcooke: I've ran into some issues with desktop after I fixed all the phone stuff, after all those fixes I noticed that compiz would work again correctly too in zero-copy mode, so I uploaded a fixed version. :P
[13:16] <willcooke> mlankhorst, oh excellent!  thanks
[13:16] <mlankhorst> FTL windowed, compiz, and glxgears windowed now work again as intended.. if only input did
[13:16] <willcooke> seb128,  are you playing with Xmir on your U8 desktop?
[13:17] <mlankhorst> it was broken, but sa30 should work (minus input)
[13:18]  * willcooke -> late lunch
[13:18] <mlankhorst> hm... Xmir might work as replacement for Xorg with the xmir driver in full desktop mode..
[13:18] <mlankhorst> haven't tried lately..
[13:22] <Sweet5hark1> seb128: FWIW, 1/ arm64 breaks on not building internal mwaw, 2/ ppc64el breaks on missing gcc-4.7 on that arch. For 1/ if we sync mwaw, we might use external mwaw (its already in main) 2/ multiple options: a/ test if any later gcc preprocs work now b/ bring gcc-4.7 to ppc64el c/ MIR and use ucpp on all platforms
[13:34] <mlankhorst> ok it seems to work
[13:36] <mlankhorst> installing ubuntu-desktop-mir, removing nvidia-prime and changing /usr/share/lightdm/lightdm.conf.d/50-xserver-command.conf from X -core to Xmir -core
[13:36] <mlankhorst> hm or not.. bah
[13:38] <mitya57> charles: JFYI: you have two indicator-datetime fixes in two different silos, while the description suggests that you wanted both fixes in one upload
[13:38] <ChrisTownsend> Anybody else missing the box to enter the password in the greeter when first logging into the system with latest Vivid updates?
[13:39] <ChrisTownsend> Password is still accepted, just no box is rendered...strange.
[13:41] <mlankhorst> no idea why xserver is ignored, but replacing /usr/bin/X with Xmir works :P
[13:51] <tkamppeter> Laney, yes, there is an upstream VCS. It is https://git.fedorahosted.org/cgit/system-config-printer.git/.
[13:51] <tkamppeter> Laney, for wha do you need it? What do you want to contribute?
[13:51] <Laney> tkamppeter: packaging change
[13:52] <tkamppeter> Laney, so you do not want to contribute to the upstream code? On the packaging level I have never used a VCS.
[13:53] <Laney> tkamppeter: indeed, OK that's fine then, thanks!
[13:53] <tkamppeter> Laney, what do you want to change on the packaging?
[13:53] <Laney> icon theme dependency (already uploaded)
[13:54] <willcooke> mirsc
[13:54] <willcooke> oops
[13:55] <tkamppeter> Laney, OK, I have seen the change.
[13:55] <mlankhorst> bleh, need hw cursor :P
[14:02] <tedg> brainwash, I would expect so, but let me see.
[14:03] <tedg> Eh, no. But known.
[14:04] <brainwash> tedg: ok, hopefully this little change can land in vivid soon :)
[14:04] <tedg> brainwash, Yeah, probably. Pushed a fix. Let's see what Jenkins thinks.
[14:04] <brainwash> tedg: thanks :)
[14:41] <seb128> Sweet5hark1, sorry was not around, reading backlog ... what Laney said, if we didn't have those archs it's fine to do without
[14:42] <seb128> let me know when you have something to sponsor
[14:42] <Sweet5hark1> seb128: hang on a sec.
[15:05] <willcooke> all - got the school run in a mo.  Should be back in time for the meeting, but if not I wont be very late
[15:05] <seb128> k
[15:18] <Sweet5hark1> ... still building source package ...
[15:29] <willcooke> back
[15:30] <willcooke> #startmeeting Weekly Desktop Team Meeting - 2015-01-27
[15:30] <meetingology> Meeting started Tue Jan 27 15:30:53 2015 UTC.  The chair is willcooke. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[15:30] <meetingology> Available commands: action commands idea info link nick
[15:31] <willcooke> Roll call:  attente_, desrt, dgadomski_, didrocks, FJKong, Laney, larsu, mlankhorst, qengho, seb128, Sweet5ark1
[15:31] <didrocks> hey
[15:31] <dgadomski_> hey o/
[15:32] <willcooke> I have updates for Themuso, robert_ancell, tkamppeter
[15:32] <willcooke> s/for/from
[15:33] <willcooke> Let's get this party started:  attente_
[15:33] <willcooke> #topic attente_
[15:33] <attente_> hi
[15:33] <attente_> implementing the menu api on the mir server side, small menu fixes to the gdk backend to use it
[15:33] <attente_> going to kernel bisect for https://bugs.launchpad.net/mir/+bug/1409133
[15:33] <attente_> that's all from me
[15:33] <willcooke> good stuff, thanks attente_
[15:33] <willcooke> good luck with the bisect
[15:33] <attente_> thanks
[15:34] <willcooke> Let me know if you need anyone else pulling in to help
[15:34] <willcooke> #topic desrt
[15:34] <desrt> hi!
[15:34] <willcooke> How's CAM?
[15:34] <desrt> i have been sitting here for hours waiting for my turn
[15:34] <larsu> willcooke: under a bridge
[15:34] <desrt> it's lovely.
[15:34] <willcooke> :D
[15:34] <willcooke> Sorry I couldnt make it today - hopefully tomorrow will be easier
[15:34] <desrt> been getting a lot of good work done here.  a lot of long-term items are getting reviews and merged
[15:34] <willcooke> Awesome!
[15:34] <desrt> the unix mount monitor stuff will land before i leave, it seems
[15:35] <willcooke> great news
[15:35] <desrt> also we landed the G_DECLARE_TYPE header macro stuff
[15:35] <desrt> and now we are doing g_autoptr(GObject) which is a local variable that unrefs itself on scope ending
[15:35] <desrt> which is ... nice
[15:35] <desrt> GListModel probably also lands today or tomorrow
[15:35] <willcooke> neat
[15:35] <desrt> also chatting about some other things.... it's a good hackfest, in other words
[15:35] <willcooke> great news
[15:35] <desrt> that's all i really have to say
[15:36] <willcooke> thanks desrt
[15:36] <willcooke> #topic dgadomski_
[15:36] <dgadomski_> hey! I have some questions today and I appreciate your help in getting more info on those:
[15:36] <dgadomski_> * one of the users is having problems with getting external mics working on his Thinkpad X240 (bug #1414706). It does not work neither directly nor via a docking station. I believe it has something to do with either CTIA<->OMTP jack incompatibility or it may require some quirks passed to the snd-hda-intel model=. Do you know who could have some more info in that area?
[15:37] <seb128> dgadomski_, try asking diwic or themuso
[15:37] <willcooke> Off the top of my head, I'd say TheMuso is probably a good starting point.  He's in UTC-12 so email would be easiest
[15:37] <willcooke> what seb128 said :)
[15:37] <dgadomski_> cool, thanks!
[15:37] <dgadomski_> * Somehow related to bug #1104230 - if you have some windows on an extra display, and then you disconnect the display, the window is moved to the display that is left. If you then reconnect that extra display - the window does not move back. Is a feature to restore the window missing or is it a bug? If it is missing - is it triaged to be fixed at some point? It's super annoying for a group of users using lapt
[15:37] <dgadomski_> ops with docking stations.
[15:37] <seb128> diwic is on #ubuntu-devel atm
[15:38] <willcooke> dgadomski_, yeah I've heard people complaining about that one as well
[15:38] <willcooke> dgadomski_, bregma_ & co should be able to help there
[15:38] <seb128> feature or bug doesn't make much a difference, does it?
[15:38] <seb128> but yeah, it's for bregma_'s team and a compiz thing
[15:39] <dgadomski_> yeah, if it's a bug there is a chance that the implementation is already there, just not working ;)
[15:39] <dgadomski_> * bug #445333 - I have talked to seb128 about it a while ago, are there any plans of including the fix in gtk2?
[15:39] <bregma_> well, on 14.04 when I reconnect my external display ALL windows move to it, so it's just a colleciton of bugs
[15:40] <seb128> dgadomski_, did upstream decide on what they want to do with this one?
[15:40] <seb128> I guess we could SRU it, it's just a non trivial change and would be good to have it landing upstream first
[15:41] <dgadomski_> seb128: they still haven't made decision yet, they are being super cautious with adding anything to gtk2
[15:41] <seb128> rightly so :-)
[15:42] <dgadomski_> there are some users already using the hotfix I provided and they haven't reported any regressions yet, but we don't want to break gtk2 :)
[15:42] <dgadomski_> I would appreciate considering it though
[15:43] <seb128> yeah, we should probably start by uploading it to vivid
[15:43] <dgadomski_> I mean consider backporting not breaking :)
[15:43] <dgadomski_> I have posted a debdiff for vivid so we could give it a try
[15:43] <seb128> right
[15:43] <seb128> I'm going to have a look at sponsoring it
[15:44] <dgadomski_> thank you
[15:44] <dgadomski_> that's all from me today, thanks
[15:44] <willcooke> thanks dgadomski_
[15:44] <willcooke> #topic didrocks
[15:44] <didrocks> Ubuntu Make:
[15:44] <didrocks> * Releasing Ubuntu Make 0.4.1 installing by default the java appmenu support and an intellij fix due to their web page change (http://blog.didrocks.fr/post/Bringing-appmenu-support-for-java-application-and-Ubuntu-Make-0.4.1-with-an-Intellij-IDEA-fix)
[15:44] <didrocks> * Finishing up review and merging IDEA (non community) edition support
[15:44] <didrocks> * Start helping a new contributor for other IDEs support
[15:44] <didrocks> Systemd:
[15:45] <didrocks> * continue on fsckd communication, bringing: final plymouth communication, being able to cancel current fsck at boot time, i18n support for themes not supporting it.
[15:45] <didrocks> * review in progress, doing some polishing
[15:45] <didrocks> Misc:
[15:45] <didrocks> * patch pilot
[15:45] <didrocks> * AA duties
[15:45] <didrocks> .
[15:45] <willcooke> thanks didrocks
[15:45] <didrocks> yw
[15:46] <willcooke> #topic FJKong
[15:46] <FJKong>  hey
[15:46] <FJKong> * Fix bug: after preview skin from right click menu status bar icon error
[15:46] <FJKong> * Fix bug: crash when switch skin from menu.
[15:46] <FJKong> * bug review: status bar icon error from some skin, find reason it is caused by size of picture and trasparent, won't fix this type
[15:46] <FJKong> * bug review: font size won't changed in setting menu, can't be reproduced
[15:46] <FJKong> in progress: skin error after first installed package
[15:46] <FJKong>  here is mine
[15:46] <FJKong> .
[15:46] <willcooke> thanks FJKong
[15:47] <willcooke> #topic Laney

[15:47] <Laney> HELLO
[15:47] <Laney> • Mainly worked on splitting adwaita-icon-theme
[15:47] <Laney> ∘ getting the icons in the right package
[15:47] <Laney> ∘ removing all main deps on gnome-icon-theme to try and get g-i-t off the cd and ideally demoted to universe
[15:47] <Laney> ‣ need to fix a couple of remaining BDs now, including gnome-system-monitor, see later on
[15:47] <Laney> ∘ updating our icon theme to inherit from this
[15:47] <Laney> ∘ found some gtk reftests don't work with our icon theme, added a patch to force these to adwaita for now (at the very least we need rtl icons to fix this, or to skip that test ...)
[15:47] <Laney> • Poke at gnome-system-monitor transparent background, don't know what happen, larsu please halp
[15:47] <Laney> • Merge webkitgtk, need to SRU, will do this week
[15:47] <Laney> • Test eog without libgrip patches, using GTK's new GtkGesture support. It works! So those are now gone
[15:47] <Laney> • poppler transition, couple of packages needed porting, nobody seemed to notice (good)
[15:47] <larsu> Laney: drop overlay-scrollbars!
[15:47] <Laney> • got pung to clean up the cdimage building machine as it was running low on space
[15:47] <Laney> • Dig into a couple of proposed-migration issues for people
[15:47] <Laney> 🎂
[15:47] <seb128> lol
[15:48] <Laney> you get one more cycle of this shit, then we can :p
[15:48] <seb128> I could have bet on the o-s comment from larsu!
[15:48] <Sweet5hark1> ;)
[15:48] <larsu> Laney: I looked into it but couldn't find a solution either
[15:48] <seb128> can't you just override the bg color for g-s-m?
[15:48] <Laney> it's something to do with the border
[15:48] <seb128> you did it for deja-dup iirc
[15:49] <Laney> I can get the colour on the GtkBox but there's still a transparent border
[15:49] <Laney> which is even still there with * { background-color: red }
[15:49] <Laney> and border-color
[15:49] <Laney> ho hum
[15:49] <larsu> ya, that's what I saw as well
[15:49] <larsu> maybe we've reached the end of that hack
[15:50] <Laney> how can there be an unstyleable transparent area?
[15:51] <Laney> anyway, post-meeting ;-)
[15:51] <larsu> that's a good question

[15:51] <willcooke> thanks Laney
[15:51] <willcooke> #topic larsu
[15:51] <larsu> * finsihed up some gtk 3.14 theme fixes
[15:51] <larsu> * fixed the volume slider (wasn't moving properly when dragging)
[15:52] <larsu> * put gedit menubar back (upstream, almost ready to merge)
[15:52] <larsu> I have a package built for that, but it still has a header bar. Adding a titlebar makes it look weird
[15:52] <larsu> we decided to talk about what to do there when we see each other next week
[15:53] <larsu> * fixed an indicator-messages problem that phone people were hitting (unable to reply to messages)
[15:53] <larsu> * at a hackfest since sunday! Lots of discussions etc.

[15:54] <willcooke> thanks larsu
[15:54] <willcooke> #topic mlankhorst
[15:54] <mlankhorst> noon
[15:54] <mlankhorst> oh just in time
[15:55] <mlankhorst> working on xmir, getting lts-utopic packages accepted
[15:55] <mlankhorst> most of the desktop works now on es2 and mesa
[15:55] <mlankhorst> fixed the mesa dri2 acceleration too
[15:55] <mlankhorst> not much else going on right now :)
[15:55] <willcooke> thanks mlankhorst
[15:56] <willcooke> hopefully we can make a start on input stuff next week with the Mir guys
[15:56] <mlankhorst> cursor support would be nice too
[15:56] <willcooke> #topic qengho
[15:56] <qengho> Heya.
[15:56] <qengho> * More Mir work. Trying to figure out how to get around a dlopen'd protobuf problem. I may need advice here. More later.
[15:56] <qengho> * New chromium release 40.0.2214.93. Testing and fixing now.
[15:56] <qengho> EOF.
[15:56] <willcooke> !!  fast
[15:56] <willcooke> :)
[15:56] <willcooke> thanks qengho
[15:56] <seb128> new chromium \o/ :-)
[15:57] <qengho> seb128: You have DRI3, yes? I'll have something for you to test in 4h.
[15:57] <willcooke> mlankhorst /might/ be able to give you some pointers with Mir?
[15:57] <seb128> qengho, I don't know, I've an i5 intel video, how do I tell?
[15:57] <seb128> @mir, there is #ubuntu-mir as well, bunch of helpful people there :-)
[15:57] <meetingology> seb128: Error: "mir," is not a valid command.
[15:57] <qengho> seb128: if you had the crasher you asked about, you do. The intel driver is a good indicator you do.
[15:58] <seb128> k
[15:58] <seb128> I'm happy to do testing then
[15:59] <willcooke> qengho I saw that missing url bubble again the other day, but I still haven't worked out the cause of events which makes it happen, so I think it might be a JS thing, perhaps the URLs are not really URLs at all but JS functions. Going to look more next time
[15:59] <willcooke> #topic seb128
[15:59] <seb128> • tested touch rtm updates, reported some bugs
[15:59] <seb128> • some small ubuntu-system-settings reviews
[15:59] <seb128> • updated unity8 bugfix branch after review comments, got it landed
[15:59] <seb128> • updated casper to have the desktop-next iso starts the live session without the configuration wizard, users want to try the desktop not configure their user account/computer
[15:59] <seb128> • tried qtmir fix for gtk rendering issues on unity8
[15:59] <seb128> • worked on some ubuntu-system-settings privacy panel bugs
[15:59] <seb128> • helped debugging unity8-desktop not starting after yesterday updates (platform-api issue, being addressed)
[15:59] <seb128> • usual share of desktop related bugs triages and discussions

[16:00] <willcooke> thx seb128
[16:00] <willcooke> #topic Sweet5hark1
[16:00] <Sweet5hark1> heya
[16:00] <Sweet5hark1> - ran autopkgtests for 4.4.0 locally, tweaked around some failures
[16:00] <Sweet5hark1> - uploaded libreoffice 4.4.0~rc2/vivid
[16:00] <Sweet5hark1> -- includes these MIRs: libfreehand, libodfgen, libabw, libe-book, libeot
[16:00] <Sweet5hark1> -- other MIRs/system libs punted for an later upload
[16:00] <Sweet5hark1> -- (new) arm64 doesnt build yet because of internal mwaw (to investigate: can we simply build with external mwaw -> needs sync)
[16:00] <Sweet5hark1> -- (new) ppc64el doesnt build yet because of gcc-4.7 missing, options: a/ check if newer gcc preprocessors work for LibreOffice by now b/ port gcc 4.7 on ppc64el too c/ MIR and use and test using ucpp instead on all platforms
[16:00] <Sweet5hark1> -- however, forgot to remove ppc64el, arm64 from platform, done now with the just-created http://people.canonical.com/~bjoern/vivid/4.4.0/libreoffice_4.4.0~rc2-0ubuntu2_source.changes
[16:00] <Sweet5hark1> - finished libreoffice 4.2.8/trusty SRU
[16:00] <Sweet5hark1> - some investigation on libreoffice android viewer, tiled rendering, gtk+tiled rendering on MIR options
[16:00] <Sweet5hark1> - LibreOffice migrated from freedesktop to http://bugs.documentfoundation.org
[16:00] <Sweet5hark1> -- thus freedesktop bugzie should have lost 25% of bugs, users and traffic over the weekend. To other freedesktop projects: enjoy your speedup.
[16:00] <Sweet5hark1> - this week: FOSDEM preparations (talks, slides, BoD and employee meetings)
[16:01] <willcooke> thanks Sweet5hark1
[16:01] <willcooke> #topic robert_ancell
[16:01] <willcooke> Worked on:
[16:01] <willcooke> - Released unity-greeter 14.04.11 (SRU)
[16:01] <willcooke> - Released lightdm 1.13.1
[16:01] <willcooke> - Released simple-scan 3.15.4
[16:01] <willcooke> - Sponsored GNOME updates
[16:01] <willcooke> - Updated Sudoku and Mahjongg to 3.14
[16:01] <willcooke> Currently working on:
[16:01] <willcooke> - Unity Greeter update for vivid (tests are failing on vivid, not on 14.04)
[16:01] <willcooke> - TPM support
[16:01] <willcooke> - Updating Indic font packages
[16:01] <willcooke> #topic themuso
[16:01] <willcooke> * Made a breakthrough WRT bug #1066157. Just got to make the dash work properly with orca when open and closed more than once. Additionally a small bit of work on the Orca side will be needed for Vivid.
[16:01] <willcooke> * Updated pulseaudio to version 5.99.3 in the transitions PPA, ppa:ubuntu-desktop/transitions.
[16:01] <willcooke> #topic tkamppeter
[16:02] <willcooke> - system-config-printer: Uploaded and verified fixes of second SRU for Utopic. Now automatic driver download should work again in all situations. Also uploaded a second SRU for Trusty with the same fixes and uploaded the fixes to Vivid and submitted them upstream.
[16:02] <willcooke> - Started discussion on project ideas for Google Summer of Code 2015.
[16:02] <willcooke> - Bugs
[16:02] <willcooke> #topic Any other business
[16:02] <willcooke> It's the start of the annual review process
[16:02] <willcooke> YAY
[16:02] <willcooke> Expect emails and stuff
[16:02]  * didrocks got emails, even spam, waiting for stuff
[16:03] <willcooke> I hope you all have a fun time @ FOSDEM
[16:03] <willcooke> See some of you on Sunday/Monday
[16:03] <willcooke> #endmeeting
[16:03] <meetingology> Meeting ended Tue Jan 27 16:03:46 2015 UTC.
[16:03] <meetingology> Minutes:        http://ubottu.com/meetingology/logs/ubuntu-desktop/2015/ubuntu-desktop.2015-01-27-15.30.moin.txt
[16:03] <willcooke> thanks all
[16:03] <seb128> thanks
[16:03] <didrocks> thanks willcooke
[16:04]  * willcooke -> Back to shouting at mirscreencast
[16:04] <dgadomski_> thanks
[16:04] <seb128> Sweet5hark1, so http://people.canonical.com/~bjoern/vivid/4.4.0/libreoffice_4.4.0~rc2-0ubuntu2_source.changes is ready for sponsoring?
[16:04] <Sweet5hark1> seb128: yes
[16:04] <Laney> don't you want to wait for the current version to migrate?
[16:04] <didrocks> willcooke: hopefully it's not mirsoundrecorder :)
[16:05] <seb128> Sweet5hark1, done
[16:05] <seb128> Laney, ups, sorry, read your comment after dputting
[16:05] <Laney> guess not
[16:05] <Sweet5hark1> seb128: libreoffice-l10n should not need a rebuild
[16:05] <Sweet5hark1> seb128: thanks. sorry for the extra hassle
[16:06] <seb128> Sweet5hark1, no worry
[16:08] <attente_> qengho: i think i might've had a similar problem because i accidentally had the mir android packages installed
[16:15] <qengho> attente_: which problem? DRI3 crasher?
[16:18] <seb128> hum
[16:18] <seb128> is anybody else having lot of spacing in their sound menus "app entries"?
[16:18] <didrocks> not more than usual, but I didn't restart after updating today
[16:18] <seb128> didrocks, thanks
[16:18] <seb128> didrocks, want to try in a guest session maybe? ;-)
[16:18] <didrocks> seb128: confirmed in a guest, after starting rhythmbox -> rhythmbox registered, but no controls showing on
[16:19] <seb128> didrocks, thanks!
[16:19] <didrocks> yw :)
[16:19] <seb128> larsu, Laney, ^ any idea?
[16:20] <seb128> ja
[16:20] <seb128> it's today's light-themes update
[16:21] <seb128> I guess it's it as well that makes the password entry on the greeter not visible
[16:21] <seb128> yep, confirmed
[16:21] <larsu> head -> desk
[16:21] <seb128> mitya57, ^
[16:22] <seb128> larsu, sorry :-(
[16:22] <larsu> seb128: which updates are in there? The gnome-panel one?
[16:22] <seb128> larsu, https://launchpad.net/ubuntu/+source/ubuntu-themes/14.04+15.04.20150126.1-0ubuntu1
[16:23] <larsu> could be any of them :/
[16:26] <Sweet5hark1> seb128: oh right: could you please sync libmwaw_0.3.4-1 from debian experimental to vivid? for the next libreoffice upload ...
[16:26] <Sweet5hark1> (so not terribly urgent)
[16:27] <seb128> larsu, let me know if I can help testing/finding the issue
[16:27] <larsu> seb128: thanks
[16:28] <seb128> Sweet5hark1, synced
[16:28] <Sweet5hark1> seb128: awesome, thx
[16:28] <seb128> yw!
[16:31] <attente_> qengho: mir protobuf issue
[16:32] <qengho> attente_: Ah. Thanks.
[16:32] <attente_> think it had to do with having both the mesa and android packages installed simultaneously. could be a different problem though
[16:35] <qengho> attente_: I think this is different.
[16:36] <seb128> shrug
[16:36] <seb128> I again have polkit auth screwed after some updates
[16:36] <attente_> ah, ok
[16:37] <seb128> pitti, did you have hints on how to debug that? logind restarted some 15 minutes ago, I guess that was part of the upgrades, since I get polkit prompt for every action (like things wanting to write user datas to accountsservice, update-manager to check for updates, etc)
[16:38] <pitti> seb128: ah, you couldn't reproduce last time with merely restarting logind, right?
[16:38] <pitti> seb128: what's loginctl show-session $XDG_SESSION_ID
[16:38] <seb128> pitti, indeed not
[16:38]  * mitya57 reads the scrollback
[16:39] <seb128> pitti, http://paste.ubuntu.com/9900729/
[16:39] <seb128> pitti, I guess active=no is wrong?
[16:40] <mitya57> larsu: "could be any of them" — no, could be any of *your* changes ;)
[16:40] <seb128> pitti, that session is c5 but I've a c2 which is active
[16:40] <seb128> pitti, http://paste.ubuntu.com/9900750/ is c2
[16:43] <pitti> seb128: right, that looks wrong, if c5 is your active session
[16:43] <pitti> Timestamp=lun. 2015-01-26 08:13:13 CET
[16:43] <pitti> seb128: is that plausible? i. e. is your computer running since (at least) yesterday morning, when you logged in first?
[16:44] <seb128> pitti, yes, that seems about right for my uptime/last boot
[16:44] <seb128> pitti, I did close my session and logged back in since though
[16:44] <pitti> seb128: i. e. did you log out and back in, and perhaps there was still some leftover process from the old c2 session which kept the session alive?
[16:44] <larsu> mitya57: dude...
[16:45] <larsu> seb128: found the issue - the menu item sets the spinner class on itself and the way spinners are drawn got changed in gtk
[16:45] <seb128> larsu, ah, great :-)
[16:45] <seb128> larsu, that was efficient, how did you find it?
[16:45] <seb128> pitti, http://paste.ubuntu.com/9900820/
[16:45] <larsu> seb128: gtkinspector!
[16:45] <larsu> seb128: navigated to the menu item to see what style is set on it
[16:46] <didrocks> seb128: don't start blaming chrome!
[16:46] <didrocks> I see what you are doing here :)
[16:46] <seb128> pitti, so maybe it has nothing to do with the update, I also restarted my session which could be the issue
[16:46] <seb128> didrocks, :-)
[16:46] <pitti> seb128: ah, that makes sense; they are both on vt7 and display :0, so I guess logind becomes confused
[16:46] <seb128> pitti, but why is that session still marked as active?
[16:46] <pitti> seb128: well, I think the old logind might still have known which one is "really" active, but as that's due to watching VT changes it doesn't get written to /run/systemd/
[16:46] <seb128> shouldn't the logout have flagged it inactive?
[16:47] <pitti> seb128: I suppose the restarted login just iterated over them and marked the first one active that "fitted"
[16:47] <seb128> oh, ok
[16:47] <pitti> seb128: yeah, I guess that's roughly the direction of the bug
[16:47] <seb128> that would make sense
[16:47] <mitya57> larsu is super fast
[16:47] <seb128> want me to report that?
[16:47] <pitti> seb128: would you mind filing one with all the output from above and the summary of how you (re)started sessions?
[16:48] <pitti> seb128: I think it should be possible to reproduce that somehow; i. e. log out with some leftover processes (so that c2 doesn't close), log back in, restart logind
[16:48] <seb128> right
[16:49] <pitti> seb128: hm, actually /run/systemd/sessions/c2 does have the ACTIVE flag
[16:49] <pitti> seb128: so it might be more complicated, like restarting logind while the session gets changed or whatnot
[16:49] <seb128> I updated from update-manager on desktop
[16:49] <seb128> so I for sure didn't log out during the update
[16:50] <pitti> ok, good data point
[16:50] <pitti> seb128: well, it could still be that logind re-evaluates ACTIVE at startup (which would make sense)
[16:50] <pitti> seb128: blindly trusting what's in /run doesn't sound like a good idea
[16:51] <pitti> seb128: playing with that in a VM
[16:54] <seb128> pitti, https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1415104
[16:55] <pitti> seb128: merci
[16:55] <seb128> pitti, merci à toi !
[16:56] <pitti> seb128: hm, what's weird is that c2 is in state "opening"
[16:56] <seb128> yeah
[16:56] <pitti> seb128: I tried that and the old one is in state "closing" after logging out and back in
[16:57] <seb128> pitti, is the timestamp supposed to be the session start?
[16:57] <pitti> seb128: ah, reproduced!
[16:57] <seb128> oh, great!
[16:57] <seb128> how?
[17:01] <pitti> seb128: sent to the bug report; pretty much as I suspected above
[17:03] <seb128> pitti, makes sense, thanks for looking at it!
[17:03] <pitti> seb128: I'll look at it tomorrow, I'm too tired today I'm afraid
[17:04] <seb128> pitti, yeah, no hurry, it's not like it was happening often or not easy to workaround
[17:04] <seb128> would be nice to fix for vivid release though
[17:04] <pitti> seb128: yes, absolutely; I haven't looked into the code yet, but my gut feeling is that it's rather shallow
[17:05] <seb128> good
[17:12] <mitya57> larsu: please ping me if you need approve/landing
[17:14] <larsu> mitya57: will do, thanks (at a hackfest, might take a while)
[17:16] <mitya57> larsu: btw, seb128 also talked about an issue with unity-greeter, that's probably because you dropped an unico rule from unity-greeter.css, right?
[17:16] <seb128> mitya57, larsu, I tried to add that line back it doesn't fix it
[17:17] <larsu> right, unico rules didn't do anything anymore
[17:18] <seb128> larsu, btw I expect users are going to be highly confused by the greeter one, the password entry is missing
[17:18] <seb128> you can actually type but I expect we start receiving soon "I can't log in anymore" bugs
[17:19] <mitya57> I can revert both larsu's changes temporarily if needed
[17:20] <seb128> mitya57, thanks, let's see if larsu comes with a fix today, if not that might be a good idea indeed
[17:21] <larsu> seb128: can I test the greeter in my session?
[17:21] <seb128> larsu, unity-greeter --test-mode
[17:22] <seb128> larsu, be careful, it might disable some g-s-d/u-s-d plugins from your user config
[17:22] <larsu> awesome! Thanks
[17:22] <seb128> e.g changing the active gsettings key
[17:22] <seb128> just in case you notice that e.g multimedia keys stop working
[17:22] <larsu> I can't log in anymore!?!!!!1!
[17:23] <seb128> lol
[17:25] <larsu> same issue, it sets spinner
[17:45] <paulltk_> hello, i've got a problem with dual booting my hp envy 4-1100ed. My pc was delivered with Windows 8.1, and now is want to install ubuntu besides it. I've made a bootable USB stick with ubuntu 14.01 on it, my BIOS settings are ok, my internal hdd has unallocated space and i did everything i had to do to install ubuntu according to several 'how tos', but when i boot ubuntu fron the USB, i don't get the choice how to install ubu
[17:45] <paulltk_> the screen where i’m supposed to choose a partition on which i want to install ubuntu, except the fact that it shows no storage at all. The section is completely empty. Does someone know what’s going on?
[17:49] <willcooke> hey paulltk_ - you would be best of asking in the #ubuntu channel.  Those guys are better placed to help I think
[17:50] <willcooke> oh, looks like you are already there
[17:53] <paulltk_> willcooke, thanks, yeah i didn't know where i could ask best
[17:53] <willcooke> paulltk_, as a side note - could it be a UEFI BIOS thing?  Secure boot?
[17:56] <paulltk_> willcooke, secure boot is disabled, i think not so, besides i thought that ubuntu had a signature?
[17:58] <paulltk_> willcooke, anyway thanks for your help!
[17:58] <willcooke> paulltk_, nw
[19:10]  * willcooke -> EOD
[19:32] <larsu> Laney: still around?
[19:35] <mitya57> my IRC says he's away
[19:36]  * larsu is trying to find out why we're not shipping all icons in adwaita-icon-theme
[19:36] <larsu> but it can wait till tomorrow - it's his bday after all
[19:37] <mitya57> some of the icons are in adwaita-icon-theme-full
[19:37] <larsu> interesting...
[19:38] <mitya57> that's what we had in gnome-icon-theme previously
[19:43] <mitya57> larsu: btw, any update on the theme issues? should I do the temporary revert as agreed with seb128?
[19:52]  * mitya57 → EOD, too late here
[20:09] <larsu> mitya57: no, found both issues already. Just have to come up with patches
[21:09] <robert_ancell> larsu, what's the other MP you refer to in https://code.launchpad.net/~larsu/unity-greeter/dont-set-spinner/+merge/247770?
[21:10] <larsu> robert_ancell: https://code.launchpad.net/~larsu/ubuntu-themes/unity-greeter-pw-entry/+merge/247771
[21:10] <larsu> it sets a -gtk-icon-source on the entry
[21:10] <larsu> problem is that we don't draw spinners anymore unless they're spinning
[21:10] <larsu> so setting the .spinner class results in getting opacity=0 for your widget when it's inactive
[21:11] <larsu> robert_ancell: I have the same problem in i-sound, please wait with merging that second one
[21:11] <robert_ancell> larsu, do we need a dependency on unity-themes?
[21:11] <larsu> no
[21:11] <robert_ancell> ok
[21:11] <larsu> whichever theme you're using should support this
[21:11] <larsu> Adwaita does, for example
[21:19] <larsu> robert_ancell: done
[21:20] <larsu> thanks for the review
[21:20] <robert_ancell> np
[21:21] <larsu> note: this won't work until our adwaita-icon-theme gets process-working-symbolic
[21:21] <larsu> or out theme gets a custom one...
[21:21] <larsu> no clue why it's not in the package, will ask Laney tomorrow
[21:21] <robert_ancell> larsu, yeah, that's why I was wondering if we should put a dependency in debian/control. But it's probably just easier to wait
[21:22]  * larsu nods