=== mchro- is now known as mchro [00:38] c === ajmitch__ is now known as ajmitch === m_conley_away is now known as m_conley === m_conley is now known as m_conley_away [05:07] Good morning [06:42] good morning [06:46] morning ... [06:46] Aloha didrocks! [06:47] good morning BigWhale, RAOF! [06:54] bonjour didrocks [06:55] desrt: just sent my GNOME membership application :) [06:55] guten morgen pitti, how are you? [06:55] hey RAOF, how are you? [06:55] pitti: Pretty good. Yourself? [06:55] didrocks: quite fine, thanks! [06:57] pitti: nice :) [07:03] To the cinema! [07:05] * didrocks goes from FTBFS to FTBFS with new compiz since yesterday :/ [07:09] RAOF: what will you see? [07:14] vuntz: bonjour! [07:14] vuntz: for gnome bug 642692, I wondered how "committed" you are to the wnck_shutdown() API name [07:14] Gnome bug 642692 in general "Provide wnck_shutdown() (?)" [Normal,New] http://bugzilla.gnome.org/show_bug.cgi?id=642692 [07:15] vuntz: if the name is fixed, and we just need to discuss the implementation, I would apply that patch to our packages already for broader testing and usage in notify-osd [07:15] vuntz: but I wouldn't like to vendor-patch an API change without your consent [07:17] didrocks: I think I already asked, but I forgot, sorry: What is "Provide some user journey in manual test case form" all about? (johnlea's WI on desktop-p-unity-quality) [07:18] pitti: John as a lot of "user journey" about "how to start an application", "how to reveal the launcher", and such. He was supposed to turn them into test cases [07:18] pitti: but I covered them in my test cases at the end [07:19] didrocks: right, was just going to ask -- sounds a lot like the manual test cases you have already? [07:19] didrocks: so I guess we can drop this one? [07:19] pitti: yeah, I think we can safely drop it [07:20] done, thanks [07:20] pitti: the test cases that are not covered are the multimonitor and multitouch ones [07:20] multimonitor -> I'm waiting on the new work done upstream [07:20] multitouch -> I need time… :/ [07:23] pitti: name is going to stay, feel free to push it [07:24] pitti: I'll try to make time to look at it today/tomorrow [07:35] vuntz: ok, will upload it to Ubuntu then; merci! === tkamppeter_ is now known as tkamppeter [08:53] didrocks: I recently noticed that the launcher doesn't appear if there is a window underneath it; I just confirmed in a guest session [08:53] didrocks: do you mind a quick try on your machine, to ensure it's not something I picked up from a PPA? [08:53] go to an empty desktop [08:53] open a window (tried with terminal and devhelp, doesn't matter) [08:53] move it below the launcher [08:53] press Windows key -> launcher doesn'ta ppear [08:54] pitti: it works here [08:54] pitti: it doesn't appear at all, it's not a stacking issue? [08:54] right [08:55] hum, I don't get this here (application maximized or not) [08:55] didrocks: ok, thanks for trying [08:55] * pitti downloads today's desktop CD and tries there [08:56] pitti: but your issue remember of one I had at some point [08:56] on oneiric [08:56] so, there is maybe some screwed state around [08:56] happens for me in a freshly started guest session [08:57] start session, ctrl+alt+t, move, press Windows key [08:57] hm, wait [08:57] windows+number also doesn't work any more [08:57] didrocks: argh, silly me; ignore me, please [08:58] pitti: oh? what was it? [08:58] seems I accidentally switched my keyboard from "windows" into "Mac" mode [08:58] :) [08:58] there's a weird key combination to do that, and sometimes I hit that [08:58] *blush*, sorry [08:58] *phew* no worry ;) [08:58] * didrocks hugs pitti [08:58] * pitti hugs didrocks [09:08] hey [09:08] bonjour seb128 [09:08] hey pitti, how are you? [09:09] quite fine, thanks! how about you? [09:09] I'm good thanks [09:14] saltu seb128 [09:14] lut didrocks [09:23] good morning everyone [09:26] hey chrisccoulson [09:26] how are you? [09:27] hi seb128 [09:27] yeah, i'm good thanks. how are you? [09:27] got the tbird and ffox final builds to do today, but that should be easy :) [09:28] chrisccoulson, I'm good thanks [09:32] hey chrisccoulson [09:32] hi didrocks [10:03] seb128: remember that we talked about the segfault in gtk image with the new iconhelper stuff? I have a tiny testcase now http://paste.ubuntu.com/815206/ - obviously what the testcase is doing is not correct and yet it would be nice if gtk would not segfault here, especially not when using gtk :/ - essentially the problem is that icon_helper is NULL on destory and that causes the segv [10:03] seb128: fwiw, oneiric did not segv in the testcase [10:04] mvo, hey, right the old code was doing less in destroy [10:20] mvo, can you open a bug with the testcase? [10:20] mvo, btw is there any chance you look at s-c with the new webkit? [10:20] mvo, nothing else is having an issue, I wonder if that's a s-c bug or if you could come with a small testcase [10:20] seb128: where can I download it? desktop ppa? [10:20] mvo, yes [10:21] mvo, ppa:ubuntu-desktop/ppa [10:21] seb128: sure I can file a bug, I don't have much hopes other than "WONTFIX" though, this is why I did not do it already [10:21] mvo, well you seem to care enough to write a testcase, so we can as well have the bug for the record [10:21] even if it's wontfix it still show a problem we might hit again [10:22] * mvo nods [11:45] RAOF: FYI, milestone "precise-beta-1" does not exist (it's ubuntu-12.04-*); but as b1 is the default milestone for that blueprint I just merged it with "work items:" === greyback is now known as greyback|lunch === MacSlow is now known as MacSlow|lunch === Omega- is now known as Omega [12:30] A question... Silly but important one! Is ~/Documents in Slovenia (or Germany or whatever) also ~/Documents or it gets translated? [12:32] BigWhale: it does get translated, but not quite at runtime [12:32] BigWhale: see ~/.config/user-dirs.{dirs,locale} [12:33] BigWhale: /etc/X11/Xsession.d/60xdg-user-dirs-update checks for locale changes, or "first login" creation of those [12:34] pitti, ah, thanks! [12:41] didrocks: would it be okay for me to upload an unity-2d to drop the invalid "use-strut" from /usr/share/GConf/gsettings/unity-2d.convert ? [12:42] pitti: sure, you need to push the same change back to lp:unity-2d as well though [12:42] didrocks: hm, can't push to the branches [12:42] Vcs-Bzr: http://bazaar.launchpad.net/~unity-2d-team/unity-2d/oneiric [12:42] pitti: is it invalid? I didn't follow the story [12:42] that seems out of date, and a /precise doesn't exist [12:42] pitti: yeah, it's just lp:unity-2d [12:43] didrocks: yes, it makes gsettings-data-convert crash and thus breaks the settings migrations for every upgrader [12:43] didrocks: ah, no separate upstream/packaging branches? [12:43] pitti: upstream really wanted the packaging to be in upstream trunk [12:43] didrocks: normal MP, or what is the procedure these days? [12:43] and that was on the armel time (when the armel team maintained it) [12:43] pitti: yeah, normal MP [12:44] will also fix Vcs-Bzr: while I'm at it [12:44] pitti: thanks :) [12:44] IIRC, use-strut was for the autohide mode [12:44] * didrocks checks the schema [12:46] hum, that's weird [12:46] pitti: wait, pinging greyback [12:46] ah, not online right now :/ [12:46] ah, ok; committed locally for now [12:47] those keys should still be there, hence my surprise [12:47] grepping for "strut" there is not much [12:47] yeah [12:47] no querying from gsettings, anyway, so it doesn't look like an omission in the schema [12:47] pitti: it's not, ther is stil super-key-enable [12:47] didrocks: r804 [12:47] pitti: but hide-mode is not available as well [12:48] This branch removes the use-struts dconf option and enables struts automatically when the user selects hide-mode 0 (never hide). [12:48] pitti: do you have "hide-mode" in the schema? [12:48] yes [12:48] urgh, dconf-editor doesn't show it here [12:48] * didrocks pokes gsettings directly [12:49] pitti: ok, got it there :) [12:49] r804 doesn't look like an accident, anyway [12:49] yeah, makes sense, I was afraid about not seeing "hide-mode" anymore as well [12:49] but seems a dconf-editor issue [12:50] so yeah, just MP please :) [12:51] https://code.launchpad.net/~pitti/unity-2d/fix-gsettings-migration/+merge/89882 [12:53] didrocks: I suppose I shouldn't upload before it is merged [12:53] pitti: if this can wait, yes :) [12:54] if not, please go ahead, I'll erase in next upload [12:54] a day or two is certainly okay, it's just a major regression since december [12:54] fortunately we have test cases for this since yesterday :) [12:55] hello everyone! I saw Chris email re: X updates... and I see all the updates in apt. Is it safe to upgrade? [12:56] nessita: I've been running with those packages for a week or so, no trouble here [12:56] pitti: heh, excellent! will be no thursday 2nd february for next release :) [12:56] Thinkpad X201/intel arrandale [12:56] didrocks: ugh, 2nd Febryary? [12:57] that's still quite some time [12:57] it's also the day of soft-freeze for alpha-2 [12:58] well, /me bows to the gods of workflows :) [12:59] pitti: yeah, post alpha2 [12:59] didrocks: can we at least get this into alpha-2? [12:59] to avoid breaking the migration for every upgrader? [12:59] pitti: the 2 february is the alpha2, isn't it? [12:59] pitti: so please go ahead and upload [13:00] pitti: next release will only be post alpha2 (freeze on monday) [13:00] ok, doing that then; I don't care much about preserving the changelog [13:00] didrocks: will that cause you lots of trouble? [13:00] * pitti is torn back and forth between pissing off you or upgraders [13:01] pitti: no no, don't worry, maybe I won't have the changelog in the next upload but that's it :) [13:02] yeah, I don't care much [13:03] seb128: ^ once I locally fix the unity-2d conversion schema on the live system, it stops crashing; so shotwell seems to be a corner case only, or already fixed? === MacSlow|lunch is now known as MacSlow [13:08] didrocks: hm, how do you build a source package from these branches these days? use --commit? [13:09] pitti: with source 3 arguing at you for having distro-patch, right? [13:09] pitti, not sure I understand the shotwell issue, the key they list in the .convert is in one of the .schemas, that case might be fixed by the gconf commit I showed earlier [13:09] didrocks: yes [13:09] seb128: oh, is that already 8in? [13:09] pitti: remove source 3, I didn't for unity-2d yet but as it doesn't play well with merge-upstream… [13:09] pitti, no, still on my list for today [13:09] pitti: I did for others, especially thinking about you with SRU :) [13:09] seb128: so it's not that [13:09] didrocks: ok, will do that [13:10] pitti, well, it doesn't happen every time for sure, I don't know what conditions trigger it, it's not as obvious as the unity-2d case where an invalid key was listed === greyback|lunch is now known as greyback [13:27] good morning [13:27] seb128: could we get librsvg updated to 2.35.1 to fix bug 920796? [13:27] Launchpad bug 920796 in librsvg "gcompris FTBFS: 'RsvgSizeFunc' is deprecated [-Werror=deprecated-declarations]" [Medium,Fix released] https://launchpad.net/bugs/920796 [13:28] hey jbicha [13:28] hey jbicha [13:29] jbicha, guess we could, evfool did the update to .0 since .1 has an abi break which got revert in git after .1 but they didn't roll .2 yet [13:29] jbicha, so if we do we need to backport extra fixes as well, I was going to wait for .2 [13:29] see bug #918016 [13:29] Launchpad bug 918016 in librsvg "Request to update to new upstream version 2.35.0" [Wishlist,Fix released] https://launchpad.net/bugs/918016 [13:31] jbicha, do you want to do the update? or should I add it to my list? [13:31] seb128: oh ok, I don't have upload permissions for it [13:32] jbicha, I can sponsor it if you want ;-) [13:34] seb128: ok I'll work on it today :) [13:34] jbicha, thanks === m_conley_away is now known as m_conley [14:26] Which resolutions for icons does Unity use by default? Is there any benefit to installing x64 and x128 versions if they are just resized versions of a x256 image? [14:27] arand: for most icons, it shouldn't matter; our standard icons are SVG anyway (and thus scale well to any size), and the launcher uses someting in the range of 48 to 64 [14:28] arand, you can try asking #ubuntu-unity if you want details on what unity uses in the launcher and dash [14:28] I already did so and got no responase :/ [14:28] desrt, hey [14:29] hey [14:29] arand, when? maybe it was a bad time and you should try again? [14:29] desrt, some bugs,questions for you about the new appmenu show,hide xsettings based [14:29] About an hour ago. [14:29] desrt, if you have a few minutes [14:30] seb128: for you, i have _hours_ [14:30] :) [14:30] ;-) [14:30] desrt, ok, so first some people uninstall appmenu-gtk to get ride of the appmenu in oneiric [14:31] desrt, interesting side effect is that they get the xsettings set because unity-panel-service is running but the menus not exported... [14:31] desrt, in precise [14:31] where should that be fixed? (trying to figure where to send the bug) [14:31] in the gtk patch? [14:31] it should be fixed by telling them to stop running the appmenu indicator [14:32] ok, so uninstalling indicator-appmenu as well I guess [14:32] well [14:32] or installing back appmenu-gtk rather [14:32] pitti: So there's no benefit to installing smaller icons unless they are actually made for that format? the gtk-icon-cache will generate and store appropriate resizes anyways (so there's no performance benefit)? [14:32] isn't the UBUNTU_MENUPROXDY env var only set by the xsession scripts in the case that appmenu-gtk is installed? [14:32] 80appmenu: export UBUNTU_MENUPROXY="libappmenu.so" [14:32] 80appmenu-gtk3: export UBUNTU_MENUPROXY="libappmenu.so" [14:33] desrt, right, thing is that if you uninstall the gtk3 version only you still get the variable [14:33] nice house of cards here [14:33] which is sort of a mixmatch broken config [14:33] seb128, have you also seen the bug where people are running nautilus with sudo and not getting a menu at all? ;) [14:33] FML. [14:33] chrisccoulson, shush, that's my second one [14:33] lol [14:33] chrisccoulson, you are spoling my question! [14:33] chrisccoulson, that makes sense, no env variable set [14:33] ok [14:34] :) [14:34] kenvandine, right :) [14:34] annoying though [14:34] desrt, so second is what chrisccoulson mentioned [14:34] not specific to nautilus [14:34] sudo gedit is the same [14:34] can we just kill our stupid users? [14:34] YES! [14:34] * kenvandine thinks we need a better way than setting env [14:34] let's have the same conversation about what happens if you do X forwarding [14:34] gsettings key? [14:34] i bet the answer is sort of the same! [14:34] desrt, if it wasn't for users, our job would be AWESOME :) [14:35] seb128, that would be better i think [14:35] arand: I don't think gtk-icon-cache caches pre-rendered icons [14:35] desrt, we need a "if (getuid() == 0) exit(-EDUMBASS)" in gtk [14:36] kenvandine, I don't think the issue is the environment, sudo doesn't strip it, does it? [14:36] $ sudo sh -c "echo $UBUNTU_MENUPROXY" [14:36] libappmenu.so [14:36] but we would need to disable them if libappmenu isn't loaded or the setting was disabled [14:36] seb128, how about gtksu ? [14:36] i think it might [14:36] desrt has a point that Xorg forwarding will hit similar issues [14:37] sudo from a terminal i think is ok [14:37] kenvandine, well first, should stuff running with sudo have their menus exported or not? [14:37] i wonder if xsettings was the wrong approach [14:37] kenvandine, no it's not [14:37] but "Run as" kind of thing is broken [14:37] ok [14:37] kenvandine, sudo gedit leads to the bug since the xsettings is set for the session [14:37] without consideration for the user [14:38] kenvandine, the menu showing in precise is done based on the xsettings [14:38] ok [14:39] seb128: for the first complaint my answer is that you need to educate your users (which will probably happen automatically on various forums) [14:39] seb128: for the second complaint my answer is... uhm... i dunno [14:39] desrt, thrid one, is that gnotravex stopped having is gmenu showing in unity appmenu where that was working during the rally ... do you know why? [14:39] no. i find that most curious. [14:39] :-( [14:40] oh. yes. [14:40] it makes sense, actually [14:40] oh? [14:40] so gtk checks the xsettings property to determine if it should show the menu locally [14:40] it sees that shell-shows-menu is set and decides the answer is "no" [14:40] which means that the gtk module has nothing to rip out [14:41] ok, makes sense [14:41] i think it may have been a mistake to use the same xsettings for two very similar but not-quite-the-same meanings [14:41] desrt, you still plan to make appmenu consume gmenus this cycle? i.e to fix that? [14:42] desrt, I don't care much about 1, but I'm a bit nervous about the sudo, fwd and gmenu on unity stuff for the lts ;-) [14:42] i wasn't planning on it, no [14:42] but it seems like i may have to [14:43] desrt, so you think the current xsettings are not a good fit after all? [14:43] the trouble with any dbus-based scheme for knowing when to have the menus is that we get into that silly dancing-menu situation [14:43] xsettings are solid and there right from the start. nothing to wait for. [14:44] epiphany also uses the new appmenu but I was leaving that in the gnome3 ppa until webkit migrates to precise [14:44] I suspect several of the games will be using the new menu soon [14:44] jbicha: several already are [14:44] jbicha, right, gnotravex is my testcase in precise [14:44] seb128: for the record, i consider the gnotravex case to be the only truly worrying one [14:45] desrt, doing sudo gedit and having no menu is a bit annoying as well, people do need to edit system files ;-) [14:45] seb128: i'm with chrisccoulson on this point [14:45] g_assert (getuid() != 0); [14:45] what do you recommend? [14:45] how do you tell users to edit system files? [14:45] nano? [14:45] vim! [14:45] ... [14:45] :-) [14:46] not very constructive guys :p [14:46] I agree the nautilus case is on crack [14:46] but I disagree the "edit files in etc using gedit" is [14:46] it's equally bad [14:46] that said, i don't think that this particular way of failing is really acceptable [14:47] again -- i'd prefer if it exploded in your face immediately [14:47] but i doubt you'll want to do that [14:47] we should just crash when we fail! ;) [14:48] jbicha: GSettings-style software engineering practices :) [14:48] grrr [14:49] i hate that :) [14:49] i know [14:49] what we need is a reliable way to determine if we are in the proper session [14:49] ie: detect these weird sudo/x11-fwd cases [14:49] interesting story [14:49] yeah [14:49] i was talking to christian hammond (chipx86) over the weekend [14:49] and he said that your gtk patching makes his life absolutely miserable [14:50] it interacts quite badly with vmware's native-window thing [14:50] because it rips the menu out of the window [14:50] but then vmware puts the window on the windows machine [14:50] desrt, I'm not even sure why the sudo case is not working, would it make sense to use the current user session bus for sudo as well? [14:50] and the menu is nowhere to be found [14:50] so this is another related issue... [14:50] seb128: that's a really complicated question that you're not likely to get a satisfactory answer to [14:51] ok, not the way to go then [14:51] seb128: mostly because any person who is qualified to answer that you ask it to will tell you "don't do that" [14:52] desrt, I guess the sudo issue will happen the way under gnome-shell right? like if you run gnotravex it will get no menu right? [14:52] quite odd [14:52] DISPLAY makes it into sudo [14:52] but DBUS_SESSION_BUS_ADDRESS does not [14:52] seb128: sudo gnotravex? :) [14:52] desrt, yes, as a testcase :p [14:53] ya. i bet it's broken. [14:53] "don't do that" [14:53] desrt: DISPLAY is one of the few whitelisted sudo env vars which are kept [14:53] pitti: can we turn that off? :) [14:53] ./plugins/sudoers/env.c, initial_keepenv_table[] [14:53] desrt: err, no? [14:53] pitti: of all people i know that i don't need to have the "sudo [x application] is bad" conversation with you :) [14:54] desrt: well, you can turn it off for a particular call, or in /etc/sudoers, but we use it for $deity knows many things, and so do users [14:54] desrt: well, it's not that I like them, but people do use them.. [14:54] right. fair enough. [14:54] oh wow, this hud things brings back memories of Office XP's personalized menus [14:54] so it's pretty simple [14:55] we need a robust mechanism for determining if the connecting client is in the "native context" of the x server [14:55] desrt: there was an entire project and library (libgksu/gksu/gksudo) written around it, so apparently it's useful [14:55] which in this case means "can connect to the same session bus" [14:55] desrt: although in the past it was a lot more important [14:55] desrt: like, "gksu synaptics" in a .desktop file [14:55] these days we have polkit and root-running d-bus backends and all that [14:55] we aren't quite at zero yet, but very close [14:55] pitti: indeed... that's what i was getting at [14:56] all of this will soon be an unpleasant memory [14:56] not that pkexec would be any better, though :) [14:56] (if you use it to launch X-ish stuff) [14:56] so i have a thought [14:56] unix:abstract=/tmp/dbus-JZ8XIKEi1E,guid=80949262084dbaf430eea21d00000011 [14:56] instead of storing the shell-shows-app-menu property as a bool [14:57] we could store the dbus guid [14:57] and we only consider it valid if we have the same guid on our own session bus [14:57] desrt: in fact, I think I (well, apport) is one of the last bad guys, aside from checkbox-gtk and update-notifier [14:57] otherwise we ignore it, because obviously it won't be working anyway [14:57] that will catch every broken case [14:58] this may well break in a systemd future, though [14:59] jbicha: do you know whether freerdp will be uploaded to Debian soon? [14:59] jbicha: (you committed 1.0 to git, but wasn't uploaded yet) [14:59] desrt, well let's do it in a way which works in the present and adapt it to what is available in the futur when that happens [14:59] seb128: i want to make sure that it continues to be possible to do this without synchronous dbus calls [15:00] desrt, well I guess at this point it's better to switch to a bug or list discussion [15:01] interesting [15:01] gdbusconnection has a "guid" property [15:03] pitti: the issue is that someone else proposed different packaging for freerdp and I'm not sure what of that needs to be added to git [15:03] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=647498 [15:03] Debian bug 647498 in freerdp "freerdp: Consider packaging git snapshot?" [Wishlist,Open] [15:04] jbicha: ah, thanks; I'll have a look ASAP [15:04] pitti: thanks [15:04] jbicha: probably tomorrow, though, need to run in about 20 mins [15:05] pitti, will you be back for the meeting if there is one? [15:05] pitti, or do you want me to handle it? [15:06] pitti, well "it" the "ping people and see if we do a meeting" ;-) [15:06] pitti, ignore the question I just read your reminder email after the first line :p [15:06] seb128: what, there is content in the email reminder? [15:07] it's just not a link to click on it? :p [15:07] didrocks, :-p [15:07] seb128: would be nice, thanks [15:07] didrocks, I learnt my lesson ;-) [15:07] pitti, will do [15:07] seb128: but I don't think we'll have one, no items until now [15:08] pitti, right, it's likely we will not, still I can do a round of ping and see if any topic comes [15:08] pitti, I know nessita want to make a call for endorsements for her per package upload application wikipage [15:08] yes I do! [15:08] ;-) [15:08] shall I drop the request now then? [15:09] nessita, you can mention it yes ;-) [15:09] doesn't hurt [15:09] we can mention it again at meeting time as well [15:10] hello everybody! If you can/want to invest some minutes, I would appreciate some endorsement for my per package upload application: https://wiki.ubuntu.com/NataliaBidart/PerPackageUploadApplication [15:10] and I will also welcome any suggestion/improvement that you consider necessary [15:19] ok, I'm out for half an hour or so, will be back for the meeting time for sure ;-) [15:21] hey cyphermox [15:21] cyphermox: seems I can't activate the bluetooth anymore [15:21] hey didrocks [15:22] are you aware about such issues? [15:22] how's it look? [15:22] there's tons of issues with bluetooth ;_; [15:22] like, the menu just shows "bluetooth active/disable settings" [15:22] not the long menu like before [15:22] didrocks: are you using a usb bluetooth dongle or integrated? [15:22] and if I start gnome-control-center [15:22] cyphermox: it's integrated [15:22] didrocks: problem would be that bluetoothd isn't running for you [15:23] I can't enabled it [15:23] ah [15:23] should I launch it? [15:23] you can, and it's going to solve the issue for you [15:24] I'm still working on figuring out why the hell it won't start on boot like it was so clearly working so well before [15:24] good night everyone! need to run now [15:24] have a good night pitti [15:25] cyphermox: hum, it exits immediatly [15:25]