[05:38] <pitti> Good morning
[05:53] <didrocks> good morning
[06:06] <pitti> bonjour didrocks !
[06:08] <didrocks> hey pitti, how are the ice creams nowdays?
[06:08] <pitti> didrocks: wonderful! although I didn't have any yesterday
[06:08] <pitti> my wife is away for two days, and eating ice cream alone just isn't the same
[06:09] <pitti> but I feel my left arm and legs from Basketball yesterday (after 3 weeks)
[06:09] <didrocks> I understand you :) but a day without an icecream is it still really a day ?
[06:09] <didrocks> ahah :)
[06:09] <pitti> didrocks: it is, but for sure not a good one!
[06:17] <larsu> good morning!
[06:22] <seb128> hey larsu & desktopers
[06:26] <hikiko-lpt> hello :)
[06:27] <didrocks> hey larsu, seb128!
[06:27] <seb128> lut didrocks
[06:27] <seb128> hey hikiko-lpt
[06:27] <didrocks> morning hikiko-lpt
[06:27] <hikiko-lpt> hi seb128 didrocks :)
[06:30] <didrocks> todays is visual studio in ubuntu make day!
=== hikiko-lpt is now known as hikiko
[06:30] <didrocks> if someday someone would have preditected that some version of visual studio will be available on linux…
[06:32] <larsu> didrocks: ya. craziness
[06:32] <seb128> didrocks, yeah, I didn't know that was a thing!
[06:34] <seb128> in fact seems it's "visual studio code" and a new IDE not visual studio proper?
[06:34] <seb128> http://www.theregister.co.uk/2015/04/29/visual_studio_code/
[06:34] <seb128> nice ubuntu/unity screenshot ;-)
[06:34] <didrocks> seb128: yeah, it's a version of visual studio for the web
[06:34] <seb128> oh, web?
[06:34] <seb128> I though it was for mono
[06:34] <didrocks> supports node and javascript as well
[06:35] <seb128> the news focus on .NET
[06:35] <didrocks> well, .NET is mainly for the web as well
[06:35] <didrocks> https://code.visualstudio.com/
[06:35] <didrocks> "Build and debug modern web and cloud applications."
[06:35] <seb128> that url doesn't work here
[06:36] <didrocks> interesting, you are blacklisted!
[06:36] <seb128> seems so
[06:36] <seb128> "Firefox ne peut trouver le serveur à l'adresse code.visualstudio.com."
[06:36] <didrocks> definitively works here
[06:36] <didrocks> even on some browser like firefox :p
[06:37] <seb128> "To prove that Microsoft still loves Linux, the company demonstrated a Mono application being edited in Visual Studio Code on Ubuntu today."
[06:37] <seb128> :-)
[06:37] <didrocks> I guess they mean asp.net by mono application
[06:37] <seb128> Ubuntu demoed by Microsoft at their conf, kind of cool ;-)
[06:37] <didrocks> but yeah, really cool!
[06:38] <didrocks> it's running well, time to build support into Make!
[06:38] <seb128> great
[06:40] <didrocks> ok, the "direct" eurostar is taking more than changing in Lille
[06:40] <didrocks> (and it's arriving at 22:12)
[06:54] <pitti> didrocks, seb128: I think https://www.brainyquote.com/quotes/quotes/l/linustorva381582.html was from around 1991; took a while, but I think we can safely say it's true now :)
[06:56] <didrocks> pitti: ahah, indeed :)
[07:02] <didrocks> seb128: still didn't get a link? I wonder if it's just because you have an x68 machine (they don't seem to have a download available for you)
[07:23] <willcooke> morning
=== mitya57 changed the topic of #ubuntu-desktop to: Welcome to the Desktop Team | https://wiki.ubuntu.com/DesktopTeam | For support please join #ubuntu | Test gtk 3.16 from ppa:ubuntu-desktop/ww & file bugs with the tag "gtk316"
[07:27] <didrocks> hey willcooke
[07:31] <didrocks> seb128: the website is down for me as well now FYI
[07:31] <larsu> same here (worked earlier)
[07:32] <didrocks> larsu: I was blaming my code first :p
[07:32] <didrocks> (support ready, then wanted to have a run before implementing mock tests)
[07:32] <larsu> wow you're fast ;)
[07:33] <didrocks> I wonder who did this framework to add new support so flexible while being nicely tested… OH WAIT! :)
[07:33] <larsu> hehe
[07:34] <lng> Hi! I have two monitors, but both show the same output and it's detected as one in Displays section. I use Gnome. Here is more details: http://ubuntuforums.org/showthread.php?t=2273883&p=13265949#post13265949
[07:34] <didrocks> let me add the large tests in between (even if I can't really test it for now)
[07:38] <didrocks> and large test implemented ;)
[07:40] <seb128> didrocks, sorry was afaik for a bit, glad to see it's not only me ;-)
[07:41] <didrocks> seb128: not glad for me, it's blocking me now :p
[07:41] <seb128> hehe
[07:41] <didrocks> seb128: we could had it with full tests in a couple of hours! Now, we won't :p
[07:42] <seb128> you should have downloaded the site while it was up to be able to mock it/work offline :p
[07:42]  * didrocks creates some mock ssl certificates on code.visualstudio.com for medium tests meanwhile
=== ara is now known as Guest6469
[07:47] <didrocks> some people are reporting it's crashing on trusty though
[07:59] <willcooke> lng, this is a developer channel rather than a support channel - you would be best to ask in #ubuntu where people who know about this sort of thing will be
[08:04] <Laney> hullo
[08:04] <willcooke> what ho Laney
[08:05] <willcooke> damn it window focus
[08:05] <larsu> willcooke: welcome to the wonderful world of compiz
[08:05] <willcooke> time to explore "focus follows mouse again"
[08:05] <willcooke> or can't I do that in Compiz?
[08:05] <Laney> thanks mitya57!
[08:06] <larsu> willcooke: probably there's a plugin for that :P
[08:06] <didrocks> hey Laney
[08:07] <Laney> hey didrocks et willcooke
[08:07] <Laney> what up?
[08:07] <didrocks> willcooke: yeah, think about enabling locally integrated menu though (if you care about reaching your menus)
[08:07] <Laney> hud!
[08:07] <larsu> hi Laney!
[08:07] <larsu> didrocks: oh shit good point :)
[08:08] <larsu> didrocks: I guess you could still get at them with Alt-Something
[08:08] <didrocks> larsu: I know some QA guy who had to stop using ffm when Unity came around because of that precise reason :p
[08:08] <didrocks> or rather "a QA guy"…
[08:09] <Laney> hey larsu!
[08:09] <larsu> didrocks: you mean *the* qa guy ;)
[08:10] <didrocks> right, *the* one ;)
[08:11] <larsu> seb128: I don't know about bug #1445540
[08:11] <ubot5> bug 1445540 in gtk+3.0 (Ubuntu) "GTK draws its own (double) window decorations under Mir" [High,Triaged] https://launchpad.net/bugs/1445540
[08:11] <seb128> hey willcooke & Laney
[08:12] <larsu> seb128: we won't have default gtk apps on unity8, right? I suggest just going with upstream headerbars
[08:12] <seb128> larsu, I think it's an unresolved question, until unity8 displays decoration (if it does) we should keep the csd
[08:12] <seb128> larsu, how does that work with unity8 if they display decoration? same issues than we currently have in u7?
[08:14] <larsu> seb128: johnlea said unity8 will allow csd and ubuntu apps will use csd
[08:14] <larsu> seb128: this might be a development/design not in sync issue actually
[08:15] <seb128> larsu, k, so close it as invalid?
[08:15] <seb128> or just let it triaged and ignore it
[08:15] <larsu> seb128: I'll comment and leave it to wishlist
[08:15] <seb128> larsu, thanks
[08:39] <willcooke> qengho, running Version 42.0.2311.135 Ubuntu 14.04 (64-bit)
[08:39] <willcooke> qengho, still get the crash at startup
[08:45] <willcooke> seb128, I've got a "The application Chromium Web Browser has closed unexpectedly." window up.  There's a load of debugging info in there
[08:45] <willcooke> is there a text file equivilent somewhere?
[08:45] <willcooke> oh
[08:45] <willcooke> /var/crash
[08:46] <didrocks> yeah, just open it and read if you want to ensure we didn't see how which tabs you were :p
[08:48] <willcooke> lol
[08:48] <willcooke> I'll upload it somewhere and send a private link to qengho
[08:49] <seb128> willcooke, right
[08:49] <seb128> willcooke, you can also apport-retrace it locally to provide a backtrace
[08:50] <willcooke> oki - can I do that later on, or do I have to do it now?
[08:50] <willcooke> like
[08:50] <willcooke> if qengho wants me to do it in a few hours, will the data still be there?
[08:51] <didrocks> as long as you keep the .crash and you don't upgrade libraries, you will be able to retrace
[08:52] <willcooke> cool, thx
[08:55] <didrocks> yw
[09:04] <didrocks> nice, I receive different kind of error page on visual studio website everytime I try a download from it
[09:04] <didrocks> they are clearly working on it: http://pastebin.com/LXbdtset
[09:05] <didrocks> (a mono stacktrace, asp verbose mode enabled)
[09:07] <willcooke> ha
[09:12] <flexiondotorg> I'm the lead for Ubuntu MATE. If I wanted to make LIM an option what libraries/settings do I need to incorporate?
[09:13] <didrocks> I guess it's a question for Trevinho ^
[10:21] <didrocks> ok, the site is back up, the remaining crash was due to user agent (which put the server in error)
[10:21] <didrocks> so using chromium user agent for now in make
[10:49] <willcooke> didrocks, o_O
[10:49] <willcooke> didrocks, so Ubuntu support isn't quite what it should be?
[10:49] <didrocks> willcooke: well, it's more their server-side doesn't handle no user agent, so not a biggie
[10:49] <willcooke> ah, I see
[10:49] <didrocks> willcooke: I would be interested in someone trying on trusty though
[10:50] <didrocks> I read on the net that it's crashing
[10:50] <willcooke> sure, what do I do?
[10:50] <didrocks> ok, large and medium tests work \o/
[10:50] <willcooke> woo
[10:50] <didrocks> let me wrap that in nice commit messages and pushing it
[10:51] <didrocks> then, you can have a try
[10:51] <willcooke> :D
[10:55] <didrocks> willcooke: ok, so:
[10:55] <didrocks> 1. git clone https://github.com/ubuntu/ubuntu-make
[10:55] <didrocks> 2. cd ubuntu-make
[10:55] <didrocks> 3. bin/umake web visual-studio-code
[10:55] <didrocks> path, accept licence, and so on…
[10:56] <willcooke> ImportError: No module named 'bs4'
[10:56] <willcooke> installing bs4...
[10:57] <willcooke> Now installing the Python 3 version...
[10:57] <willcooke> ;)
[10:57] <willcooke> fixed
[10:57] <didrocks> willcooke: oh, I thought you had ubuntu-make installed (and so all deps :p)
[10:57] <willcooke> I guess it's on my vm
[10:57] <didrocks> makes sense
[10:58] <willcooke> WORKS!!!!
[10:58] <willcooke> well
[10:58] <willcooke> LOADS!!!
[10:59] <didrocks> hehe :)
[10:59] <didrocks> let's see once installed
[11:00] <willcooke> I quite like it
[11:01] <didrocks> willcooke: so, it's basically atom under the wood
[11:01] <didrocks> (yeah you have another installation of chromium)
[11:01] <willcooke> :)
[11:02] <didrocks> ok, so let's release it and write a blog post I guess
[11:04] <willcooke> \o/
[11:04] <willcooke> Now *thats* snappy
[11:04] <jcastro__> didrocks, man awesome, I'm in malta and came in and wanted to see if someone wanted to do the new ide
[11:04] <jcastro__> you rock
[11:04] <willcooke> damn straight
[11:04] <didrocks> heh, thanks guys!
[11:10] <didrocks> uploaded to the ppa (vivid, utopic, trusty), waiting for publication before getting the blog post out
[11:16] <willcooke> \m/
[11:16] <Laney> \m/ >_< \m/
[11:16] <willcooke> excellent
=== MacSlow is now known as MacSlow|lunch
[11:42] <didrocks> published in the ppa, blog post and g+ post done :)
[11:42] <didrocks> davidcalle: waow!
[11:43] <didrocks> I guess it's a new record, sharing in <5s :p
[11:43] <willcooke> didrocks, congrats
[11:43]  * davidcalle is actually a bot
[11:44] <didrocks> :)
[12:29] <willcooke> qengho, anecdotal information about new Cr from the ppa:  there seems to be a lot more screen tearing
[12:29] <willcooke> when scrolling up and down pages esp.
[12:30] <qengho> willcooke: huh. How many "GPU crashes" did you see per day, previously?
[12:31] <willcooke> crashes, only at start up
[12:31] <willcooke> once it was running, it was solid
=== MacSlow|lunch is now known as MacSlow
[12:54] <flexiondotorg> Has someone some pointer on how I might be LIM working in Ubuntu MATE? I've installed all the *appmenu* packages. But nothing. Guessing I might need to export some environment variables?
[12:56] <didrocks> flexiondotorg: as told earlier this morning, let's wait for Trevinho to be around, I don't know if the code is in unity or if the decorator with LIM support is in compiz itself, he would know
[12:56] <Trevinho> didrocks: oh, I'm around... I just didn't see any ping :o
[12:56] <didrocks> Trevinho: there were at least one :p
[12:56]  * Trevinho now does
[12:56] <flexiondotorg> Trevinho, See above :)
[12:57] <Trevinho> flexiondotorg: I've seen it...
[12:57] <Trevinho> flexiondotorg: so... for ubuntu mate you need to implement it by scratch mostly
[12:58] <Trevinho> flexiondotorg: the only thing you can get from unity-panel-service is the full list of menus around
[12:58] <Trevinho> flexiondotorg: but then you've to make gtk-window-decorator to support them, and it's not easy, considering how it's done (that's why I wrote a new decorator from scratch for unity)
[12:59] <Trevinho> flexiondotorg: the unity decorator is not a separate plugin from unity..
[13:00] <flexiondotorg> Trevinho, All understood. So, I don't need of of the appmenu stuff?
[13:00] <Trevinho> flexiondotorg: but....... in theory you might create a new compiz plugin from scratch using most of of its code (as it doesn't depend much on unity code).. But you could use Libunitycore
[13:00] <flexiondotorg> Trevinho, I need unity-panel-service running and gtk-window-decorator adapted to display the menus?
[13:00] <Trevinho> flexiondotorg: indicator-appmenu is used, but via unity-panel-service
[13:01] <flexiondotorg> Trevinho, Is indicator-appmenu required if I am only interested in lIM?
[13:01] <Trevinho> flexiondotorg: well... not sure you're already using unity-panel-service for indicators... In case you do, you can just re uset it.
[13:01] <flexiondotorg> *LIM?
[13:01] <Trevinho> flexiondotorg: otherwise no, you can use directly indicator-appemnu
[13:01] <Trevinho> appmenu*
[13:01] <Trevinho> flexiondotorg: yes...
[13:02] <Trevinho> flexiondotorg: it exposes you the menus  of the focused app or, of every app (depending on the environment variables you set)
[13:02] <flexiondotorg> Trevinho, OK, so this menu introspection is used elsewhere in Unity too?
[13:04] <Trevinho> flexiondotorg: mhmh, no... In unity we have unity-panel-service which loads libappmenu (with the others) and talks to LibUnityCore wich abstracts the indicators...
[13:04] <Trevinho> flexiondotorg: this was needed at the time we also had unity2d/unity3d... to allow code reusage
[13:05] <Trevinho> flexiondotorg: then, we use these things in two sides: PanelMenuView which handles the menus when an app is maximized, and in DecoratedWindow wich is the new decorator code
[13:05] <Trevinho> for (restored windows)
[13:07] <flexiondotorg> Trevinho, Thanks for such a detailed replay. Very useful. Thanks you.
[13:07] <Trevinho> flexiondotorg: let me know if you need further explainations...
[13:08] <Trevinho> flexiondotorg: in case you'd like to create a new decorator, and share the code with the unity one we might create a library maybe...
[13:08] <flexiondotorg> Trevinho, I will do. I've filed those notes and will do some research to better understand how this all integrates.
[13:18] <Trevinho> flexiondotorg: the way we do things is not the "most optimized" so far because of this service around, but it still helps to solve some issues (such has handling the lockscreen better), but in general code should be quite self-explainatory
[13:29]  * Laney mews
[14:36] <xclaesse> seb128, was thinking about the inconsistent GtkHeaderBar in unity when the window is maximized. Here is what I currently get: http://people.collabora.com/~xclaesse/tmp/window-decoration-maximized.png
[14:36] <xclaesse> seb128, on the left it is nautilus maximized (no GtkHeaderBar afaik), on the right it is devhelp maximized
[14:36] <seb128> xclaesse, right, I know
[14:37] <xclaesse> seb128, the theme could just hide close/max/min buttons
[14:37] <seb128> we could remove the decoration from the headerbar in those cases I think
[14:37] <xclaesse> seb128, and theme it as a toolbar
[14:37] <seb128> that would display a standard toolbar
[14:37] <xclaesse> and that would solve the problem, no?
[14:37] <seb128> yeah, I guess so
[14:37] <seb128> larsu, Laney, ^ would that be easily doable?
[14:38] <xclaesse> there are little size/padding differences as well, even in the windowed state, but that's cosmetical, probably just a few values to tweak in the css
[14:39] <Laney> we already want to remove the buttons
[14:39] <Laney> don't know about re theming
[14:39] <seb128> xclaesse, I think it's not that easy, larsu looked at it and gtk doesn't let you use lower padding
[14:41] <xclaesse> probably because they are in the same GtkHBox than other widgets in the headerbar which force it to have bigger height
[14:42] <xclaesse> or something like that... anyway that's cosmetical, if we can already hide decoration buttons when maximised that would improve it a lot already :)
[14:43] <seb128> right
[14:43] <xclaesse> atm only devhelp seems to have that issue, in the apps I'm using, but more will come in upcoming GNOME releases
[14:47] <seb128> gnome-system-log in vivid has the issue
[15:01] <Sweet5hark> seb128: I had a look at that libreoffice 4.4 crash on close ...
[15:02] <seb128> Sweet5hark, did that result in anything useful?
[15:04] <Sweet5hark> seb128: in a cruel twist of fate, it was caused by the fix for another crash on close. Will look into this with the author of the commit. https://bugs.documentfoundation.org/show_bug.cgi?id=84935
[15:04] <ubot5> bugs.documentfoundation.org bug 84935 in BASIC "Intermittent crash on exit" [Normal,Resolved: fixed]
[15:04] <seb128> Sweet5hark, how did you manage to nail it down to that commit without being able to reproduce?
[15:05] <Sweet5hark> seb128: looked at the stacktrace and the git log.
[15:05] <seb128> good :-)
[15:06] <Sweet5hark> seb128: Often that is not helpful, but in this case it was. ;)
[15:20] <jcastro> didrocks, hey so the wiki and github descriptions of umake need to be updated
[15:21] <jcastro> it doesn't list like the categories you can use, etc. makes it seem like it's only for android sdk
[15:21] <larsu> xclaesse: we can't hide those buttons
[15:22] <larsu> xclaesse: and even if we could, moving the rest of the headerbar content around is a bit weird imo
[15:22] <xclaesse> larsu, cannot be worse than current state ;-)
[15:22] <larsu> xclaesse: which current state?
[15:22] <larsu> ?
[15:22] <Laney> less weird than showing the buttons twice
[15:22] <Laney> buuuuuuuuuut </headerbars>
[15:23] <larsu> yeah...
[15:23] <xclaesse> larsu, currently buttons are duplicated
[15:23] <xclaesse> larsu, maybe GtkHeaderBar should have a "toolbar" mode where it only show custom widgets
[15:24] <xclaesse> because that's really what it is, once you remove all info that unity panel already display, GtkHeaderBar becomes a toolbar
[15:24] <larsu> xclaesse: I don't know. Lots of applications use the center of the header bar as a title
[15:24] <larsu> and having 2 titles looks weird
[15:24] <didrocks> jcastro: I don't think we should update the category, but rather point to --help (and tab completion) which does all the thing), but good point, I'll reword to not make it sounds like it's for android sdk only
[15:25] <xclaesse> larsu, title can/should be removed from the headerbar is well
[15:25] <larsu> xclaesse: then it looks empty (see current evince)
[15:25] <xclaesse> larsu, should have a way to tell GtkHeaderBar "my WM already show title and close button, please hide them"
[15:25] <larsu> xclaesse: how about epiphany then?
[15:26] <larsu> where title and locationbar are the same?
[15:26] <didrocks> larsu: evince> not empty, there is this awesome "open file" icon
[15:26] <didrocks> </friday> :)
[15:26] <xclaesse> larsu, it's custom widget => keep it
[15:26] <larsu> xclaesse: how do you know if it's a custom widget?
[15:26] <larsu> didrocks: :P
[15:26] <xclaesse> larsu, GtkHeaderBar API knows it
[15:26] <larsu> xclaesse: not saying this can't be done, but it's hacky
[15:27] <xclaesse> widget you add with gtk_header_bar_set_custom_title() or gtk_header_bar_pack_start() must be kept in the hypotetical "toolbar" mode, the rest hidden
[15:28] <larsu> ugh, magic
[15:28] <larsu> but then, gtkheaderbar is already a mess
[15:28] <larsu> anyway, not sure we want this in unity at all
[15:28] <larsu> still have the argb window problem, for example
[15:29] <xclaesse> unless ubuntu wants to rewrite all GNOME apps, you'll have to deal with it, no?
[15:29] <larsu> we are
[15:29] <Laney> we change default apps to not use them
[15:30] <xclaesse> doesn't macosx has the same issue ?
[15:30] <larsu> xclaesse: I've been advocating to add proper support for csd windows in compiz for years
[15:30] <larsu> xclaesse: nothing happened yet
[15:30] <xclaesse> I think gtk upstream should have some support for the unity case
[15:30] <Laney> personally would vote for trying the hiding buttons thing (see if it really is weird) and leave the other stuff
[15:30] <larsu> it does
[15:30] <larsu> and my patch adds it to gedit as well
[15:30] <larsu> not sure why it's still not applied
[15:30]  * Laney goes out, bbl
[15:30] <larsu> I guess because there's a toolbar missing
[15:31] <larsu> but that's what it looks like on osx today
[15:32] <xclaesse> IMO patching every single app isn't a brillant idea, should try to get GtkHeaderBar widget handle the unity/maxos case
[15:32] <larsu> I agree
[15:33] <xclaesse> IMO a "toolbar-mode" property where all non-custom widgets are hidden makes sense
[15:33] <larsu> no
[15:33] <larsu> that's just stupid
[15:33] <xclaesse> then unity could set that property when maximized, and theme it as a toolbar
[15:34] <larsu> it should be automatic
[15:34] <xclaesse> yeah, or have it know what the WM wants
[15:34] <larsu> right
[15:34] <larsu> so another xsetting?
[15:35] <xclaesse> something that tells "I already dispaly close/max/min buttons and the title if it's not custom widget, please hide them"
[15:36] <larsu> "but only when maximized"
[15:36] <larsu> this is weird
[15:36] <larsu> I'd rather just keep the traditional title bar for now
[15:37] <xclaesse> I guess the theme css can know if it's a GtkHeaderBar inside a maximized window, then make the background like toolbar's
[15:37] <larsu> until we see how (and *if*) unity design deals with csd in the future
[15:38] <larsu> xclaesse: you can't properly hide widgets from css (only make them invisible, which means you'll still be able to click)
[15:38] <larsu> also, the theme doesn't know which desktop it runs on
[15:38] <larsu> unless we finally give in and tie desktop env and theme together
[15:39] <xclaesse> larsu, yep that's why hiding widget must be done by GtkHeaderBar widget itself (via a property, or a xprop, or whatever). The ubuntu theme should just deal with the background of the headerbar that looks weird when maximized and make it lool like toolbar
[15:40] <larsu> xclaesse: ah, right
[15:40] <larsu> I've been planning on doing that anyway
[15:40] <xnox> I'm done with O_CLOEXEC -> i aggree
[15:40] <larsu> not sure if people will like that, since it will change the look of many apps
[15:40] <larsu> for example, nautilus
[15:41] <xnox> desrt: ^
[15:41] <xclaesse> larsu, nautilus' toolbar maximized looks correct to me, it continues nicely the unity panel, no?
[15:42] <larsu> xclaesse: it's a technically a header bar, so it would get restyled if we implement your proposal
[15:43] <larsu> xclaesse: see how this is a shitty problem? We've been talking about this for a while now. Don't even get me started on supporting LIM on windows with csds
[15:43] <larsu> which is more or less impossible
[15:43] <larsu> but people like it, so there goes
[15:44] <didrocks> xnox: hum, what happened with O_CLOEXEC?
[15:44]  * didrocks does use it
[15:45] <xnox> didrocks: i've hit a race with it -> http://gtk.10911.n7.nabble.com/I-m-done-with-O-CLOEXEC-td86719.html
[15:46] <xclaesse> larsu, it's already a GtkHeaderBar in nautilus? the background looks different to devhelp's: http://people.collabora.com/~xclaesse/tmp/maximized.png
[15:47] <larsu> xclaesse: ambiance styles it differently if it's not a titlebar
[15:49] <didrocks> xnox: interesting… we should definitively forbide multithreading and be back to one processor as well :)
[15:50] <xclaesse> xnox, IIRC glib itself is done with CLOEXEC
[15:50] <xclaesse> wasn't desrt taking about it recently?
[15:50] <didrocks> xclaesse: I guess that's the link he posted above
[15:51] <xnox> xclaesse: see thread i linked. yes it's about glib, and that desrt will stop doing it =)
[15:52] <xclaesse> ok :)
[15:59] <xclaesse> larsu, checking GtkHeaderBar API, there is already "show-close-button" property. So one step is to make that vie a style property, right?
[16:01]  * xclaesse is not familiar with GTK's CSS, dunno what it can do, like having a different class for when widget's window is maximized
[16:01] <larsu> xclaesse: please don't add this API yet. Unity is *not* ready for this yet
[16:03] <xclaesse> larsu, unity's UI is going to change that much ?
[16:04] <larsu> xclaesse: wtf?
[16:04] <larsu> xclaesse: it won't
[16:04] <larsu> it's not ready for the reasons I stated above: LIM and argb windows not getting shadows
[16:04] <xclaesse> so why shouldn't we care yet about making GtkHeaderBar look good ?
[16:05] <xclaesse> not getting background is another orthogonal issue, no?
[16:05] <larsu> not getting a shadow is a no go
[16:05] <larsu> we will always patch in a title bar if that is the case
[16:05] <larsu> srsly. Don't add API in gtk for unity if unity doesn't use the features yet
[16:05] <larsu> step one should be to fix unity
[16:07] <xclaesse> if you say so... just though they are 2 unrelated things that can be fixed in parallel
[16:08] <larsu> nah, sadly not
[16:08] <larsu> gotta run now, sry
[16:08] <xclaesse> larsu, ok :-)
=== alan_g is now known as alan_g|EOD
[17:41] <desrt> xnox: fight the system!
=== qengho_ is now known as CardinalFang
=== davmor2_ is now known as davmor2
[18:58] <CardinalFang> So quiet in here today. I guess we fixed all the bugs.  :(
[18:58] <ogra_> snappy ate them
=== rsalveti_ is now known as rsalveti
=== \b is now known as benonsoftware