2014/06/27/#ubuntu-desktop.txt

=== Ursinha is now known as Ursinha-afk
=== Ursinha-afk is now known as Ursinha
=== fginther is now known as fginther|away
=== fginther|away is now known as fginther
=== Ursinha is now known as Ursinha-afk
=== Ursinha-afk is now known as Ursinha
[03:09] <eridu> Is this an appropriate place to discuss Unity issues? I've experienced a regression upgrading to 14.04
[03:10] <eridu> specifically, the behavior of alt summoning the HUD on a 500ms tap __if no other key is pressed__ seems to have changed to any tap under 500ms; so now I'm triggering it if I navigate with repeated M-f's in emacs and similar
[03:19] <darkxst> eridu, probably best to ask later, when more people are around
[03:19] <eridu> darkxst: when's a good time/time zone?
[03:23] <darkxst> eridu, maybe in 4 hours or so
[03:23] <darkxst> most of the desktop team are in europe I think
[03:23] <eridu> okay, is it mostly the london canonical people?
[03:29] <pitti> Good morning
[03:30] <pitti> Laney: reverse-build-deps already happens; reverse test-deps doesn't, as that's slightly harder to do (you need to download and unpack all sources, as test controls aren't aggregated in a Packages.gz like index)
=== thumper is now known as thumper-afk
=== thumper-afk is now known as thumper
[06:08] <didrocks> morning
[06:09] <pitti> bonjour didrocks
[06:09] <didrocks> hey pitti, how are you ?
[06:09] <pitti> didrocks: quite fine, thanks! was fun watching the game yesterday evening :)
[06:09] <pitti> didrocks: and you?
[06:10] <didrocks> pitti: I'm doing good! was fun to once again avoid watching any game ;)
[06:10] <didrocks> seb128 came late though
[06:10] <didrocks> his train was delayed by an hour
[06:10] <didrocks> so, I cheered him with beers when he arrived to recomfort him!
[06:11] <pitti> didrocks: oh, seb128 is in Lyon with you?
[06:11] <pitti> didrocks: and he didn't get up together with you? :-)
[06:11] <didrocks> yeah, he's coming for a week :)
[06:11] <didrocks> ahah
[06:11] <didrocks> you know him… :p
[06:12] <didrocks> I'm temptive to wake him up though
[06:12] <pitti> slack^Wsay good morning to him :)
[06:12] <didrocks> hehe, will tell him
[06:12] <didrocks> once he's there :)
[06:12] <pitti> 05:29:22      pitti | Good morning
[06:12] <didrocks> waow, I've been awake for 40 minutes only today :p
[06:13] <didrocks> but I went to bed late as well
[06:13] <didrocks> did you already run?
[06:13] <pitti> didrocks: no, not yet; still a bit chilly
[06:14] <didrocks> I'll try to force seb to come with us running :)
=== oCrazyLemn is now known as CrazyLemon
[07:02] <seb128> good morning desktopers
[07:02] <seb128> pitti, salut!
[07:03] <didrocks> hey seb128!
[07:03] <seb128> lut didrocks ;-)
[07:03] <pitti> bonjour seb128 ! give a hug to didrocks from me :)
[07:04] <seb128> pitti, done ;-)
[07:05] <pitti> \o/
[07:06] <pitti> seb128, jibel: so looks like it could very well be GER - FRA in the quarter finals
[07:06] <seb128> indeed, I was looking at that yesterday
[07:06] <seb128> sorry for you
[07:06] <seb128> ;-)
[07:06]  * pitti hangs his fan scarf a level higher
[07:07]  * seb128 doesn't care, he can pretend to be german if France loose
[07:07] <pitti> seb128: yeah, France is on quite a rampage, I'm not that optimistic for Germany
[07:07] <pitti> seb128: hahaa
[07:08] <pitti> although the German team did rather well yesterday; wasn't overly exciting, but nice in a technical and precision way
[07:08] <seb128> but yeah, France has been impressive
[07:08] <seb128> so let's see
[07:08] <pitti> more like proper craftsmanship than doing miracles :)
[07:08] <seb128> hehe
[07:08] <pitti> although that only goal was really a piece of art
[07:09] <seb128> I didn't see it
[07:09]  * seb128 was in the train
[07:11] <pitti> seb128: http://www.sportschau.de/fifawm2014/video/videousagegendeutschlanddiehoehepunkte102.html scroll to ~ 1 min
[07:15] <seb128> pitti, nice one
[07:17] <larsu> oh, a goal!
[07:17]  * pitti ^5s larsu
[07:18] <larsu> :) hi pitti
[07:29] <tjaalton> how to disable HUD? it opens every time I hit alt+$num in irssi..
[07:30] <tjaalton> google to the rescue
[07:32] <larsu> tjaalton: system settings / keyboard / shortcuts / launchers / key to show the hud
[07:32] <larsu> change that from alt to ... anything
[07:35] <jibel> mvo, hey, is there anything blocking the approval of update-manager and update-notifier into precise-proposed?
[07:35] <tjaalton> larsu: yep, found that.. ccsm didn't work anymore since these had migrated to dconf
[07:46] <mvo> hey jibel, I don't think anything is blocking, it just needs someone with the right permissions
[07:46] <mvo> jibel: to accept it
[07:51] <jibel> mvo, could you document bug 1333728 with impact / testing / potential risk of regression / any other info useful for the SRU team, then we'll find someone on #ubuntu-release to approve it.
[07:51] <ubot5> bug 1333728 in update-notifier (Ubuntu Precise) "update-manager should support HWE EOL transition" [Undecided,In progress] https://launchpad.net/bugs/1333728
[07:53] <jibel> mvo, http://dmz-jenkins.ubuntu-ci:8080/view/HWE%20EOL/ <- link to automated test, all red because it's testing your branch with the translation issue. Once u-m and u-n are uploaded to -proposed it should be green.
[07:53] <jibel> *tests even
[07:53] <darkxst> hey seb128, pitti, didrocks, larsu!
[07:53] <pitti> hey darkxst
[07:57] <mvo> jibel: thanks, adding this now
[08:03] <Laney> good morning!
[08:03] <Laney> happy freitag!
[08:03] <pitti> hey Laney
[08:03] <Laney> hey hey pitti, how's it going?
[08:04] <pitti> Laney: quite fine, thanks! how about you?
[08:04] <darkxst> hey Laney
[08:04] <seb128> hey darkxst
[08:04] <seb128> Laney, good morning, how are you?
[08:04] <didrocks> hey darkxst ;)
[08:04] <Laney> hey seb128
[08:04] <Laney> doing good thanks
[08:04] <pitti> pretty annoyed though that it's going to rain on Sunday -- I wanted to finally fly on a kite!
[08:04] <didrocks> morning Laney
[08:04] <Laney> we went to "middle eastern tapas" at the veggie cafe yesterday
[08:04] <Laney> was nice
[08:04] <pitti> and got an appointment on Sunday
[08:05] <Laney> fried chick peas, who knew?
[08:05] <seb128> sounds nice
[08:05] <Laney> with butternut squash
[08:05] <Laney> was very tasty
[08:05] <Laney> how's lyon? ;-)
[08:07] <seb128> nice! great weather this morning
[08:07] <didrocks> (as usual… ;))
[08:07] <Laney> haha
[08:08] <seb128> yeah, train had a 1 hour delay yesterday, "as usual"
[08:08]  * seb128 blames didrocks
[08:08] <didrocks> roh!
[08:09] <seb128> Laney, so, robert_ancell doesn't like the new service approach
[08:09] <darkxst> Laney, seb128 so the display change confirmation dialog uses a signal and method from the C-api that is not exposed on the dbus
[08:10] <darkxst> seb128, how about making it a new u-s-d plugin?
[08:10] <darkxst> the big advantage of how I have it, is that cherry-picking upstream patches is easy
[08:11] <seb128> upstream patches for what?
[08:11] <darkxst> seb128, display config and idle monitor
[08:11] <Laney> "doesn't like"?
[08:11] <seb128> Laney, https://code.launchpad.net/~robert-ancell/unity-settings-daemon/xrandr/+merge/224548/comments/539755
[08:12] <Laney> I wouldn't offer to do the work, but given that it is already done/in-progress I don't get what the cost is
[08:13] <Laney> the benefit is not having yet more forks of ancient code
[08:13] <seb128> Laney, fwd you an email
[08:13] <Laney> okay
[08:15] <Laney> did he fix up the idle monitor stuff?
[08:17] <seb128> not yet I think, he wrote on that mp that it can be done in a follow up change
[08:17] <seb128> I guess he's waiting for us to decide which road to go
[08:19] <Laney> darkxst: sounds like you're backtracking on your daemon with 'new plugin'?
[08:20] <darkxst> Laney, I meant still keep a direct copy of the code from mutter, but use a plugin rather than daemon
[08:20] <darkxst> that way, u-s-d can have direct access to the c-api
[08:21] <darkxst> it also makes the dbus activation issues, a non issue (although I haven't tested that)
[08:22] <darkxst> of course exposing an extra signal and method over dbus is not hard either
[08:23] <Laney> so you could have u-s-d call the method to show the dialog
[08:23] <Laney> ?
[08:28] <Laney> maybe we could go with Robert's approach as it's nearly working and investigate switching to yours if you want to keep working on it?
[08:29] <darkxst> Laney, u-s-d already had the code for the actual dialog
[08:29] <Laney> what's the problem then?
[08:29] <Laney> displayconfig can't call back into u-s-d to show it?
[08:30] <darkxst> just the signal that triggers the dialog now, is in displayconfig, not on dbus
[08:30] <darkxst> likewise for the function to confirms
[08:31] <darkxst> and really I am not too fussed which way you guys go, just so long as the gnome-desktop transition gets unblocked ;)
[08:32] <Laney> so you'd expose the signal and the functions and hook u-s-d's dialog up to this
[08:32] <Laney> ?
[08:33] <darkxst> Laney, yes
[08:33] <Laney> let me put it this way then
[08:33] <Laney> if you're happy to work on this in the next few days then I do still prefer this option
[08:34] <Laney> but if you don't want to or don't get it working then we should let Robert do it the other way
[08:34] <Laney> do you think that's fair?
[08:34] <Laney> I'm happy to take "mostly working" and help fix small issues, but I wouldn't say it was at that state yet from my test yesterday
[08:35] <Laney> could be small things like dbus activation making the state look worse than it is though
[08:40] <darkxst> Laney, ok, I am happy for Robert to do it his way
[08:41] <Laney> want to comment on the MP?
[08:41] <darkxst> it was "mostly working" (apart from the dialogs apparently) but I haven't really looked at it all since March
[08:42] <Laney> probably rebasing on 3.12 would shake out some issues
[08:43] <darkxst> Laney, did you use gnome-desktop 3.12?
[08:44] <Laney> yep, said that on the bug
[08:47] <darkxst> Laney, ah yes, and I even told you it was probably 3.12 related! just ignore that comment ;)
[08:47] <Laney> hehe
[08:49] <Laney> feel free to come back with a PPA or current bzr branches or something that we can work off
[08:50]  * Laney sends a little email, cc-ing darkxst 
[09:00] <Laney> there
[09:04] <Laney> TheMuso`: sorry forgot to reply--maybe that does make sense
[09:05] <Laney> TheMuso`: It surely is upstream, just reporting a bad (IMO) user experience ;-)
[09:06] <Laney> pitti: (forgot to reply to you too) maybe you could save the test-deps into a database somewhere when you see them?
[09:06] <Laney> do you have a LP project for wishlist bugs? :)
[09:06] <Laney> s/a/an/
[09:06] <pitti> Laney: yeah, that "somewhere" is the hairy bit..
[09:07] <pitti> Laney: we'll have somewhat of a DB once we move to the next gen CI runner (then we can put it into swift)
[09:07] <pitti> Laney: but with jenkins, it's *eew* no
[09:07] <Laney> is jenkins the only state you have?
[09:08] <pitti> Laney: and the britney db
[09:08] <Laney> I thought part of it was driven by scripts from an account on lillypilly
[09:08] <Laney> but I guess this is exactly the bit that doesn't know test-deps yet
[09:09] <pitti> right
[09:09] <Laney> bah!
[09:09] <pitti> we thought about it, but implementing it without archive support is hilariously complicated
[09:09]  * Laney nods
[09:10] <pitti> TBH, a cron job that regularly grabs the d/t/control files from XS-Testsuite: sources would be easiest
[09:10] <pitti> a bit like Contents.gz
[09:10] <Laney> I could run something like that on codesearch
[09:11] <Laney> that has the unpacked archive
[09:11] <pitti> right
[09:33] <seb128> Trevinho, hey, is https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1170647 something you plan to look at? it's quite an active ticket
[09:33] <ubot5> Ubuntu bug 1170647 in unity (Ubuntu) "Clicking on Nautilus’ launcher icon opens new window instead of restoring the minimized one when browsing external drives/locations" [High,Triaged]
[09:34] <seb128> Trevinho, recent comments point to the fix for bug #753938 as creating the issue
[09:34] <ubot5> bug 753938 in unity (Ubuntu) "Launcher - Open Trash window prevents Nautilus being launched when a user clicks on the Nautilus Launcher icon" [Medium,Fix released] https://launchpad.net/bugs/753938
[10:18] <darkxst> Laney, replied
[10:18] <darkxst> Laney, did you ever look at fixing codesearch?
[10:18] <Laney> not yet
[10:19] <Laney> i just wrote it on my todo list
[10:27] <darkxst> Laney, ok
=== MacSlow is now known as MacSlow|lunch
=== alan_g is now known as alan_g|lunch
[12:30] <seb128> Laney, "Don't know how to respond to inline comments"
[12:30] <seb128> you select the revision in the combo
[12:30] <seb128> then you have those listed and you can replu
[12:30] <seb128> reply
[12:30] <Laney> oh yeah
[12:31] <Laney> I guess it was showing me the newest one
[12:48] <didrocks> desrt: larsu: hey, small question on the glib signals. If I send a signal on any thread, the signal will be received and handled on the mainloop thread everytime, right? (there is no magic to respawn that in a calling thread)
[12:48] <desrt> didrocks: unix signal or glib signal?
[12:49] <desrt> glib signals have no magic _at all_.  they always execute exactly in place, and then return
[12:49] <desrt> ie: always in the same thread, and the last handler is already done by the time the emission is over
[12:50] <didrocks> desrt: glib signal
[12:51] <desrt> didrocks: think of it as a function call and a foreach mixed together :)
[12:51] <didrocks> oh, I thought it was handled by the mainloop
[12:51] <desrt> so your statement about "received and handled on the mainloop thread every time" is definitely false
[12:51] <didrocks> ok, so I need to do a idle_add() and then call the signal…
[12:51] <desrt> signals and mainloops have zero integration
[12:51] <desrt> yes.  probably.
[12:51] <desrt> or g_main_context_invoke()
[12:52] <didrocks> yeah, making sense
[12:52] <didrocks> thanks desrt :)
[12:52] <desrt> note: be careful about which thread you dispatch the signal to
[12:52] <desrt> depending on what you're building, the main thread could be the wrong choice
[12:52] <desrt> (what are you building?)
[12:52] <didrocks> for UI interactions only
[12:52] <didrocks> I want that in the mainloop thread
[12:52] <desrt> maybe....
[12:52] <didrocks> sounds about right?
[12:52] <desrt> what are you building?
[12:53] <didrocks> well, the developer tools, I have long standing threaded calls
[12:53] <desrt> if it's UI, you might already be in the main thread always
[12:53] <didrocks> and I want the response to the user to be sent back to the UI thread
=== alan_g|lunch is now known as alan_g
[12:53] <desrt> if it's non-UI then it stands to reason that maybe someone wants to use your stuff without UI
[12:53] <didrocks> so that the UI can display what it needs to display
[12:54] <desrt> like maybe one day you write a dbus/network service to handle these requests and it uses other threads...
[12:54] <desrt> anyway... you should probably use the async convention here
[12:54] <didrocks> actually, I just want all the responses send back to one thread, then, there is multiple backends installed
[12:54] <desrt> and GTask
[12:54] <didrocks> (only one run at a time)
[12:54] <didrocks> can be a CLI backend, a ncurses, a QML one…
[12:55] <desrt> take a look at GTask
[12:55]  * didrocks opens tab
[12:55] <desrt> it basically gives you a very easy way to have some code run in a thread
[12:55] <desrt> and return the result to the "user's thread" when it is done
[12:55] <desrt> it manages all the boring details like threadpools, etc. for you
[12:57] <didrocks> oh, excellent
[12:57] <didrocks> will definitively have a look, thanks!
[12:58] <desrt> there's a lot of examples of its use all over the place in GIO
[12:58] <desrt> for example, in all of the standard classes like GInputStream, we "emulate" the async version (if the underlying stream does not support it) by creating a GTask and using it to run the "sync" verison on a thread
[13:00] <didrocks> interesting
[13:04] <darkxst> didrocks, did you ever take a look at the tracker MIR?
[13:05] <didrocks> darkxst: not yet, I was discussing with it with seb128, I'll give it a look in the incoming days, but tracker will need time to review
[13:05] <didrocks> however, once I'll jump on it, I'll go to the end in one shot :)
[13:06] <darkxst> didrocks, ok, I added a nautilus patch to MIR, that only loads the tracker engine for GNOME
[13:06] <darkxst> ^to the MIR bug
[13:06] <didrocks> darkxst: yeah, I saw that, I think we should have a gsettings option rather
[13:06] <didrocks> wdyt?
[13:07] <darkxst> didrocks, not sure it makes a lot of sense, only GNOME uses tracker
[13:08] <darkxst> and if people want to disable indexing there is the optional UI for that
[13:09] <didrocks> darkxst: yeah, but if they want to enable it in Unity?
[13:13] <darkxst> didrocks, that wouldnt do much unless Unity grows tracker integration
[13:13] <didrocks> darkxst: nautilus itself isn't enough?
[13:14] <darkxst> didrocks, nope, nautilus provides the "search provider" that hooks into tracker index
[13:14] <darkxst> gnome-shell calls nautilus' "search provider"
[13:15] <didrocks> ok in that sense, I think it still would be better as a gsettings in case another desktop environment grows its integration, but that's not a worry until this is the case
[13:17] <darkxst> didrocks, happy to add that later, and upstream would mostly likely take patches for that
[13:19] <didrocks> sure ;)
[13:53] <seb128> Laney, what was the status of https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1324618 ?
[13:53] <ubot5> Ubuntu bug 1324618 in ubuntu-system-settings (Ubuntu) "Storage app list does not always show icons when they are available" [High,New]
[13:55] <Laney> ummmm
=== MacSlow|lunch is now known as MacSlow
[14:01] <Laney> haha oh here it is
[14:01] <Laney> uhh, yeah, thanks for pointing that out
[14:01]  * Laney runs
[14:02] <seb128> yw ;-)
=== alan_g is now known as alan_g|tea
=== m_conley_away is now known as m_conley
=== alan_g|tea is now known as alan_g
[15:53] <Laney> seb128 is on a triaging spree!
[15:54] <seb128> Laney, yeah, reviewing the settings bugslists, lot of small things in there :/
[15:54] <seb128> Laney, like your click sorting stopped working for some reason!
[15:54] <Laney> yeah
[15:54] <Laney> that definitely worked
[15:54] <seb128> nice to see people doing work though
[15:55] <Laney> i looked at that but didn't solve it, can try again
[15:55] <seb128> like cyphermox adding some bluetooth new feature, Wellark and Satoris doing work for networks, Jonas updating the background screen
[15:55] <seb128> oh, gatox working on the updates as well ;-)
[15:56] <seb128> that one is quite tricky, we still have quite some UI issues
[15:56] <seb128> (like the top anchors not being correct when the install button is not listed, or like the individual items that should hide the install bar when paused)
[15:58] <gatox> seb128, i'll propose that branch that checks for the battery befor installing a system update as soon as i can test it... having some problems with my phone
[15:58] <seb128> gatox, thanks
[15:58] <seb128> Laney, do you have your icon one almost ready? I was pondering doing another landing
[15:59] <Laney> yeah noticed it didn't fix everything, just testing another one
[15:59] <seb128> k
[16:04] <Laney> ooh, close that time
[16:04]  * Laney makes one tweak
[16:27] <Laney> there
[16:27] <Laney> not quite there but it's better
=== gatox is now known as gatox_lunch
[16:41] <seb128> Laney, do you still want to work on it or should I review/land it like that?
[16:42] <Laney> seb128: I think it's okay to review it now and I'll look at the other issue next week
[16:43] <Laney> it only impacts one thing I have installed
[16:43] <seb128> k
[16:57] <seb128> Laney, that code is unhappy
[16:57] <seb128> 2014-06-27 18:57:46,500 - WARNING - QFSFileEngine::open: No file name specified
[16:58] <seb128> it dumps like 15 like that
[16:58] <Laney> wasn't that there before?
[16:58] <seb128> let me check
[16:58] <seb128> it was ;-)
[16:59] <Laney> not sure where it's from, mind
[16:59] <seb128> yeah, I haz icons!
[16:59] <Laney> woohoo!
[17:01] <Laney> right I gotta go meet my dad for a beer (he's in town and I'm catching a train away for the weekend in 90 minutes)
[17:01] <Laney> happy weekend desktop!
[17:01] <seb128> Laney, thanks, have a good w.e as well!
[17:02] <Laney> enjoy lyon ;-)
[17:02] <Laney> I heard a rumour that didrocks has some sake with your name on it
[17:03] <seb128> \o/
[17:04] <didrocks> yeah, but my wife doesn't want us to mix sake and beers :p
[17:04] <seb128> we are going to need it
[17:04] <didrocks> what a dilemna!
[17:04] <seb128> not sure we get food any time soon
[17:04]  * didrocks is wondering between blackmailing seb128 or just tell that directly to Julie
[17:04] <seb128> they decided to engage in japanese cooking, seems non trivial
[17:04] <seb128> roh
[17:05]  * sarnold . . o o O O ("how long could it take to serve uncooked fish?")
=== alan_g is now known as alan_g|EOW
[17:07] <didrocks> it's real japanese food :)
[17:07] <didrocks> the one you cook, not the fancyness of sushi and mai
[17:07] <didrocks> maki*
[17:09] <sarnold> mmmm. katsudon. tonkatsu. udon. mochi. <3
=== gatox_lunch is now known as gatox
=== linuxturtle is now known as jbrett
=== rsalveti_ is now known as rsalveti
=== sarnold_ is now known as sarnold
=== bdrung_ is now known as bdrung
=== m_conley is now known as m_conley_away
[22:22] <Noskcaj> Would the desktop team be willing to maintain uhttpmock in ubuntu. It's a small package that libgdata needs to be able to build. bug 1327458
[22:22] <ubot5> bug 1327458 in uhttpmock (Ubuntu) "[MIR] uhttpmock" [Undecided,New] https://launchpad.net/bugs/1327458