[01:15] rodrigo_, awake? [02:33] hi guys. I have what seems to be a bug on my system; my desktop icons & fonts are tiny, and I don't seem able to reset it. any ideas, please? === asac_ is now known as asac === almaisan-away is now known as al-maisan [07:09] * korben greets all [08:00] Good morning [08:20] good morning [08:22] bonjour didrocks [08:22] Guten Morgen pitti :) [08:59] didrocks: you currently claim the #1 for the remaining work to do for alpha-2 [08:59] not sure it's a nice #1 position :) [08:59] there are the oneconf WI? let's see what else [09:00] didrocks: do you want to go over them with me and see whether we can postpone/drop some stuff, and check if anything is blocked? [09:00] pitti: let me have a look first [09:00] pitti: http://people.canonical.com/~platform/workitems/natty/canonical-desktop-team-alpha-2.html is empty for me ? [09:01] http://people.canonical.com/~platform/workitems/natty/canonical-desktop-team-natty-alpha-2.html [09:01] didrocks: ah, I think that's noise from the time when we had james_w's merged project support, hang on [09:02] urgh, I think that's in the DB even [09:02] urgh :/ [09:02] please ignore them for now [09:03] pitti: other-dx-n-2d-experience-fallback (3 items): I'll be able to finish that today if I can get focus on it [09:03] right, these seem easy [09:03] didrocks: the dialog appears in the live session [09:03] packageselection-desktop-n-bringing-desktop-and-netbook-image-closer: I would say postpone. Let's discuss at the Rally, but evo express in 2.32 isn't good for default [09:03] so unless that's special magic in caper, it ought to work for installs, too [09:03] pitti: yes, I've added it without marking it as DONE [09:03] didrocks: ok, so DROP [09:04] ok [09:04] didrocks: apport hook> there's a pending branch from bryceh to make it use the X.org hook [09:04] so that should be on track [09:04] yeah + adding some small things [09:04] but it's a 2 hours work [09:04] get some plugins to be linked statically [09:05] -> wait on compiz dx expert [09:05] work on nautilus to get it not move when locking unlocking the launcher [09:05] should this be reassigned? [09:05] pitti: no, I still need to do some work and they have a WI for that too [09:05] didrocks: and is that still even an issue with 0.9? [09:05] -> not sure who put that. If it's me, I don't know what it means :) [09:05] pitti: yeah, it's still slow to load [09:05] oh yeah, I added it [09:06] (the nautilus) [09:06] didrocks: the way I understand it we have the opposite problem [09:06] I misread and saw gnome-panel [09:06] didrocks: when you show the launcher, nautilus desktop background doesn't move [09:06] i. e. the unity launcher covers your icons [09:06] pitti: it's basically https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/684649 [09:06] I manually moved them for that [09:06] Launchpad bug 684649 in nautilus (Ubuntu) (and 1 other project) "Nautilus desktop icons should never be under the launcher (affects: 7) (dups: 4) (heat: 184)" [High,Triaged] [09:06] didrocks: ah, correct [09:06] I planned to work on that in the plane [09:06] didrocks: how about we link that bug to the spec instead and drop it from the WB? it'll still be tracked as a WI [09:07] sure, will do [09:07] I'll do that [09:07] thanks :) [09:07] finish /opt work regarding the decisions above -> waiting on the DMB, in any case, that can even be done in the freeze period [09:07] (there is a WI for the DMB) [09:08] ok, moving to BLOCKED [09:08] then oneconf -> I need to work with nessita for some points. and that's mainly 2 days at the Rally [09:08] apart from "some improvements/performance tweak to oneconf USC" -> these, I have the idea how to achieve it, will do it on the plane as well [09:08] didrocks: /opt> what's still to be decided there? [09:09] I thought we exhaustively discussed it in the TB meetings [09:09] didrocks: oneconf> as we are on track in the overall cycle (http://people.canonical.com/~platform/workitems/natty/canonical-desktop-team.html) we could move it to beta [09:10] but if you want to keep it for a2, that's fine for me [09:10] I mainly want to ensure that you don't explode [09:10] pitti: on /opt, there are still some decision on Quickly to decide what the special submitubuntu command will do. Nothing in the toolchain build-dep or whatever [09:10] didrocks: so, in summary, are you comfortable with your current WIs, or do you need help with something? [09:10] pitti: also, what needs to be symlinks and such [09:11] pitti: oneconf> I would prefer doing as much work as possible during the sprint, will reconsider the WI just after [09:11] didrocks: sounds like a plan [09:11] didrocks: easier there to directly work with Naty indeed [09:11] In a nutshell, it's fine, thanks :) I'm more fixing bugs in unity than rushing on my WI those days [09:12] didrocks: yes, that was my concern; stuff like that nautilus bug could easily be moved post-alpha2 [09:12] and in fact it just did now, I think the bug isn't milestoned [09:12] oh? [09:13] right, it's not [09:13] I'll milestone it for A2 [09:13] * pitti moves to beta [09:13] ah, ok [09:13] didrocks: well, you can still fix it by beta, but it won't be an a2 blocker [09:13] erm, "by a2" [09:13] pitti: is there a way to show it as an alpha2 target still? [09:13] without being a blocker? [09:14] didrocks: conceptually, a WI should be targetted at the deadline [09:14] didrocks: you can set it as "in progress" [09:14] didrocks: but as I said, feel free to move to a2 if you prefer [09:15] pitti: no, that's fine, I've it opened in a tab as "in progress" :) [09:15] didrocks: ok, thanks for the heads-up! I'm much less scared now [09:15] pitti: thanks for the notice, I'll finish the session work today [09:16] which is basically converting the failsafe session to the new session (one hack less \o/) and check the guest session :) [09:16] niice [09:17] mvo: hey, FYI, you have now com.canonical.Unity. Just check if there is a owner to know if the panel (and not the service) is running [09:20] hello [09:21] didrocks: cool, is that in natty already? if so I change the code right away [09:21] hey seb128 [09:22] hey mvo [09:22] mvo: yes, it is :) [09:22] what is in natty? [09:22] salut seb128 [09:22] lut didrocks [09:22] seb128: a way to check if unity (the panel) is running [09:24] oh ok [09:31] pitti, did you mean to drop that nautilus locking w-i? [09:31] "- [didrocks] work on nautilus to get it not move when locking unlocking the launcher: TODO" [09:31] seb128: we discussed here with pitti before you joined. It's not dropped, just set to beta (but I still plan to work on that on the plane) [09:32] seb128: pitti added the associated bug [09:32] seb128: I linked to the existing bug instead [09:32] unfortunately that's not included in teh LP mail [09:32] seb128: bonjour [09:32] hey pitti [09:32] ok [09:33] I gues you discussed staying on classic evo as well? [09:33] could be nice to add some rational in the whiteboard for people not on IRC but following the specs [10:03] does somebody know how much longer it will take till libvala-0.10-dev has vanished from the archive (NBS)? There are still some build-dependencies left but some other packages are also in depwait on libvala-dev (currently provided by two packages and the buildds can't resolve it). Or should those package get changed to build-depend on libvala-0.12-dev? [10:03] geser: I can remove them if necessary [10:04] geser: but stuff like e. g. rhythmbox b-deps on it without an alternative [10:04] libvala-0.10-dev, valac-0.10 [10:04] pitti: do you know if packages build-depend on libvala-dev will build with libvala-0.12-dev without changes? [10:04] I don't know [10:05] depends, some do some will required patches to the source [10:05] -- natty/universe i386 deps on libvala-0.10-0: [10:05] gedit-valatoys-plugin [10:05] libafrodite0 [10:05] these are the only two binary rdepends [10:06] http://people.canonical.com/~ubuntu-archive/NBS/valac-0.10 [10:06] that's the most useful list of things to fix [10:07] ok, will try. Is it preferred to build-depend on libvala-dev or use libvala-0.12-dev (perhaps with | libvala-dev)? [10:09] if it works with both 0.10 and 0.12, my gut feeling is "libvala-dev" [10:10] last time I looked at one package it needed at least an update of the .pc file to use [10:13] ah, then it should be versioned [10:46] hum, the guest session seems broken, let's look at the gdm patch if something changed [10:57] didrocks: just curious, could you reproduce my gnome-panel issue in your vm? [10:57] geser: sorry, I didn't have the time to come to it, trying to fix other things first [10:58] np [11:08] pitti: I'm working on the gdm guest session. Is there any reason why you exec /etc/gdm/Xsession "$@" rather than setting gnome-session in /usr/share/xsessions/guest-restricted.desktop for instance? [11:09] didrocks: there was one, let me check the bzr history [11:09] pitti: I checked without any success (apart from my recent change but it can be done my wisely) :/ [11:09] Exec=/usr/share/gdm/guest-session/Xsession [11:10] didrocks: ^ you mean that? [11:10] pitti: right, and the wrapper content [11:10] didrocks: well, why wouldn't I run the gdm session? [11:10] without it, you don't get /etc/X11/Xsession.d/, locale setting, etc. [11:11] didrocks: ah, you mean gdm already does that before [11:11] pitti: /usr/share/xsessions/guest-restricted.desktop isn't ru by gdm? and this one is sourcing /etc/X11/Xsession.d/ [11:11] didrocks: right, that might simply be a thinko of mine [11:11] right [11:11] pitti: ok, I will give it a try then [11:12] didrocks: back then I might have thought "let's not hardcode gnome-session" [11:12] so that you can change the default [11:12] but I'm not stuck to it [11:13] pitti: now (in lucid) that I give the current GDMSESSION to gdm and decide which file session to execute based on that, I think we can do it adding other session desktop files [11:13] pitti: ok, I'll have a look to rewrap that then, the fridge is arriving first. Thanks! :) [11:14] hah [11:14] a bed, a fridge, what about your internet? [11:16] pitti: they started to send the new dsl boxes yesterday, but they have a backlog of three weeks. So that + building the phone line. I have little hope before the rally [11:16] didrocks: well, as long as your free connection holds up.. [11:17] more or less, today it seems to not be very kind with me (I'm disconnected a lot) [11:19] could update-manager be any slower? [11:20] ok, I've 329 updates listed [11:20] but it takes over 15 seconds to display the context menu when right clicking [11:20] with some 10 seconds of compiz marking it as not responding [11:21] that's on a recent and fast box [11:24] hey seb128! [11:24] happy new year :) [11:24] hey chrisccoulson, happy new year! [11:24] how are you? [11:25] seb128 - i'm good thanks, how are you? did you enjoy your time off? [11:25] I'm fine thanks [11:26] yes, very much, 2 weeks away from the computer [11:26] which made me realize that I really needed those ;-) [11:26] nice to be back now though [11:26] what about you? [11:26] yeah, it's good to get a break :) [11:26] i've had a fairly busy couple of weeks going to visit people [11:28] did you manage to not work? [11:28] seems you did the firefox update but otherwise? [11:28] yeah, i didn't really do any work other than that [11:28] great ;-) [11:28] i checked e-mails occasionally just to make sure there weren't any big problems ;) [11:31] seb128: hum? right click takes a long time? the initial cache calculation does, but he right click should be there instalntly [11:31] mvo, 15 seconds on a i5 ssd config [11:31] before displaying the menu [11:32] seb128: so its all there ready and you right click and it takes that long? and then you get unselect all, unselect none? [11:34] mvo, yes, it takes some 20 seconds to load [11:34] then everything is displayed, I can click on entry, read the changelog etc [11:34] but the first right click takes some 15 seconds [11:34] then "uncheck all" takes less than 1 second [11:34] checking all takes some 6 seconds [11:35] that is pretty odd [11:35] but I don't get the delay again [11:38] seb128: could you ping me again after lunch? its really a odd bug, I wonder if something outside of u-m might trigger it as I don't remember that code chaning in a while [11:39] mvo, it's probably not a new bug, it's just likely that people don't have 300 updates every day [11:39] mvo, ok, will do [11:43] seb128: I have a test box here with 1000 updates and its instant (after the initial setup). could you comment out line 476in UpdateManager.py (num_updates = self.cache.installCount )and replace wiht num_updates = 27 and see if that changes anything ? [11:43] seb128: I have a suspicion about the problem :) [11:46] mvo, now it's instant [11:46] mvo, but go for lunch we can discuss that later [11:46] seb128: thanks! [11:54] pitti: ok, it was working well with the direct call to gnome-session and different desktop file. But I just saw that you are using /usr/share/gdm/guest-session/Xsession for apparmor/gdm-guest-session. [11:54] pitti: could you please merge the last gdk-pixbuf from debian? (asking you as you did the last upload) [11:54] didrocks: right, we need the wrapper; so the wrapper should call gnome-session? [11:55] pitti: yeah, and I'll add a switch to the wrapper then [11:55] as we need different args for gnome-session [11:55] bigon: ah, too bad; in our previous pacakge we used clean-la.mk from cdbs to at least drop all the dependencies from the .la file [11:55] bigon: is it an issue for natty as well? [12:01] seb128: meh, apport chroot python bustage again, fixing [12:04] pitti: well the .la is not shipped in natty, so I guess it still is === oubiwann is now known as oubiwann_ === oubiwann is now known as oubiwann_ [12:05] bigon: we have shipped the cleaned variant for quite some time, though === oubiwann_ is now known as oubiwann [12:06] but some rdep are still shipping .la file that have references to this one [12:06] bigon: anyway, I'm happy to merge it soon; we'll watch out for build failures due to this [12:08] bigon: ah, bug 665768 [12:08] Launchpad bug 665768 in gdk-pixbuf (Debian) (and 1 other project) "libgdk_pixbuf-2.0.la missing from packages (affects: 3) (heat: 22)" [Unknown,Unknown] https://launchpad.net/bugs/665768 [12:09] bigon: btw, kenvandine added a patch to set _gir_PACKAGES in the GIR, to fix vapigen [12:09] bigon: that sounds interesting for Debian as well, did he talk to you about this? [12:11] nothing on the BTS, Ive maybe missed something on irc [12:11] any rational for this patch? [12:12] I don't know beyond that changelog entry [12:12] kenvandine: ^ did you forward it to upstream/Debian? [12:12] pitti, don't put the .la back [12:12] seb128: I don't want to [12:12] we fixed the few other .la still listing it [12:12] it's not needed in ubuntu [12:12] seb128: so we can close this bug then? thanks [12:12] yes [12:13] thanks [12:13] pitti: https://bugzilla.gnome.org/show_bug.cgi?id=635364 something here [12:13] Gnome bug 635364 in general "Add C include and exported packages information to GIR" [Normal,Unconfirmed] [12:15] looks like the patches are not the same [12:15] bigon: our patch is [12:15] +GdkPixbuf_2_0_gir_PACKAGES = gdk-pixbuf-2.0 [12:15] which looks close [12:15] +GdkPixbuf_2_0_gir_EXPORT_PACKAGES = gdk-pixbuf-2.0 [12:15] I'm not sure about the difference of these two, could be enough [12:16] kenvandine: ^ ? [12:16] seb128: I'm not happy with putting .la file back, in debian there were a crapload of other .la file referencing it [12:17] hi is there a way to block right click over the desktop and panels? [12:19] seb128: do you know who would be a good person for backporting the patch for bug 494096 to lucid-proposed? [12:19] Launchpad bug 494096 in metacity (Ubuntu Lucid) (and 3 other projects) "Clicking the title of a window is bringing a window underneath it into focus (affects: 63) (dups: 9) (heat: 294)" [High,Triaged] https://launchpad.net/bugs/494096 [12:19] seb128: I guess somewhere between you, me, and perhaps Chris? [12:20] didrocks or chrisccoulson? [12:20] chrisccoulson has a lucid install for sure [12:20] chrisccoulson: oh, you are back! happy new year! [12:20] I don't have a lucid install here (well, I have it on a computer in a box) [12:21] * pitti raises shields around didrocks, looking at his WIs [12:21] :) [12:21] ok, then chrisccoulson [12:21] chrisccoulson: would you have time to look into bug 494096? there's a patch, but it needs to be tested (it does't have a good test case) and uploaded [12:21] Launchpad bug 494096 in metacity (Ubuntu Lucid) (and 3 other projects) "Clicking the title of a window is bringing a window underneath it into focus (affects: 63) (dups: 9) (heat: 294)" [High,Triaged] https://launchpad.net/bugs/494096 [12:22] brb session restart [12:30] hi pitti, happy new year to you too :) [12:30] yeah, i can take a look at that in a bit [12:30] chrisccoulson: did you have some nice holidays? [12:30] pitti - yeah, not too bad thanks, although it was quite busy [12:30] how about you? === MacSlow is now known as MacSlow|lunch [12:31] chrisccoulson: very relaxing here; lots of family and friends, and playing Descent :) [12:31] nice :) [12:31] chrisccoulson: busy> did you work much? [12:31] pitti - i didn't work much except for doing the firefox beta 8 upload [12:31] but we spent a lot of time visiting family and friends [12:33] everything is ok now with the guest session, except that apparmor prevents the unity plugins to load… [12:33] hum, maybe it's because I have a local unity installation [12:33] hey chrisccoulson! Happy new year :) [12:33] hey didrocks, happy new year :) [12:37] pitti: do you have a script or something that display the state of .la in ubuntu (like http://release.debian.org/~aba/la/current.txt ?) [12:37] hum no, not that, it seems related to the fact it can't set a local schema to gconf [12:38] bigon: I don't think we have === njpatel is now known as njpatel|lunch === njpatel|lunch is now known as njpatel_ [13:01] re [13:01] wb seb128 [13:01] seb128: btw, using gtk2.0/gi with apport is a lot less broken than I feared [13:01] so should be easy to do if we need or want? [13:02] seb128: some methods aren't available, such as Gtk.RadioButton.new_with_label_from_widget(), and connect_signals(), but by and large it works [13:02] seb128: I'm currently fixing trunk to work with both [13:02] so I'll drop the Gtk.require_version() call, and we can control it with dependencies [13:02] seb128: we already have gir-gtk-2.0 due to the gnome-applets dependency anyway [13:03] right [13:03] seb128: apport is the easiest one to get to work with 2.0 [13:03] the question was just to know if it was working enough for apport [13:03] but I hope language-selector will work, too [13:03] it's missing all the annotation fixes that went into gtk3, so these require workarounds [13:03] we could probably revert to the pygtk version otherwise? [13:04] seb128: we could, but I'd rather avoid that if possible [13:09] ok, done with restarts for a bit [13:09] pitti, let's see what we decide to do with gtk3 then we can see what we do with the rdepends [13:10] doing grouped radio buttons with gtk2/gi is crash-land and a nuisance, though :-( [13:14] gtk_radio_button_join_group() isn't available in gtk2 :-( [13:15] could we put another langpack on the CD if we drop gtk3? [13:15] I suppose [13:15] well it's a cost benefit discussion then [13:15] a bit hard to say, as removing gtk3 removes half of the desktop due to the canberra dependency [13:15] space against work to port those backward [13:15] pitti, that was fixed 2 days ago? [13:15] oh [13:16] with a bit of salt in the middle of the cost benefit due to the theming issue [13:16] seb128: ah, just mousetweaks left [13:16] but ideally we will need a gtk3 theme anyway, would it only be for the GNOME3 ppa builds [13:16] mousetweaks depends on both 2.0 and 3.0 === oubiwann is now known as oubiwann_ === oubiwann_ is now known as oubiwann === dbarth_ is now known as dbarth [13:17] are we going to use at-spi or at-spi2 for natty? [13:18] seb128: should save us about 2.8 MB [13:18] seb128: that's dropping libcanberra-gtk3-0 gtk3-engines-murrine gir1.2-gtk-3.0 libgtk3.0-0 libgtk3.0-bin [13:18] seb128: seems the GTK3 stack isn't half as big as I feared [13:19] rodrigo_, dunno, does it impact on other things we are doing? [13:19] pitti, seems it's not worth the effort to drop it then if we get a decent theme [13:19] seb128, yes, the unity a11y work [13:19] rodrigo_, oh, you work on that now? [13:19] seb128, yes [13:20] seb128, jason assigned it to me the week before the holidays [13:20] oh ok, I didn't follow that [13:20] you should check with TheMuso [13:20] seb128: so, radio buttons are officially busted with gtk2.0/gi, there is no way to group them [13:20] seb128, right, sorry, my fault, should have told you [13:20] he said at-spi2 still had speed issues IIRC [13:20] rodrigo_, so how much time do you still have for other work? [13:21] pitti, you should just put this work on hold until we sort the GTK3 question [13:21] seb128, 50% I think [13:21] seb128: *nod* [13:21] rodrigo_, ok, thanks [13:21] seb128, just this week I'm doing almost 10% only, because I need to study the unity code [13:21] seb128, but once I'm up to speed (end of this week), should have more time [13:22] also, there's a a11y expert already working on it, so hopefully my involvement should be short [13:22] seb128, what do you want me to work on? [13:22] seb128: time for lunch then :) [13:22] pitti, enjoy! [13:23] right, lunch time here also :-) [13:23] rodrigo_, nothing specifically, I'm just trying to see where we are going, what needs to be done and who is available to do work [13:23] like how active we should be on the GNOME3 ppa, who is working on it, etc [13:24] seb128, I'm still working on it, just this week a bit slow [13:24] no worry, it's just to know how to dispatch tasks [13:24] rodrigo_, enjoy your lunch ;-) [13:24] hey mterry [13:25] seb128, feel free to send me tasks, if I can't I'll let you know, but I should have time for other tasks [13:25] but yes, lunch now, bbl [13:27] rodrigo_, ok! [13:30] seb128, hello! === dbarth__ is now known as dbarth [13:41] seb128, are there Desktop things that need doing? [13:41] mterry, nothing special, basically updating or merging things which are not green on http://people.canonical.com/~platform/desktop/versions.html [13:42] well we don't care much about the yellow ones [13:42] but everything orange or red could be updated [13:43] so feel free to pick on those [13:44] mterry, someone will need to update the preferred capplet to the GNOME3 version as well [13:44] or rather "the 2.32 code to use the GNOME3 handler system" rather [13:45] which is basically copying what they did for GNOME3 if we stay on 2.32 for natty which seems likely [13:52] seb128, I see. Is anyone working on that? [13:52] I could look into it [13:52] mterry, no === MacSlow|lunch is now known as MacSlow [13:52] mterry, if you want to that would be welcome [13:53] seb128, sure, put me down for that [13:53] mterry, ok, great, thanks [13:53] seb128, we'll decide at rally whether we stick with 2.32 for natty? Seems like we will at this rate [13:53] yes [13:54] seb128, you meant the schema handler code from GNOME3, right? [13:55] Which is affecting us because we updated glib... I see [14:00] re [14:00] mterry, sorry, compiz,unity crash [14:00] mterry, right, the new handler system for the default browser, email client, etc [14:00] seb128, yup, cool [14:01] mterry, speaking about 2.32 against 3 we will discuss it, but I'm pretty sorted on staying on what we have [14:02] mterry, there is still lot of work on GNOME3, we don't have a GTK3 theme and it's not clear we will have a nice one this cycle, GTK3 already got a one month delay compared to the original schedule which said decembre [14:02] mterry, it seems lot of work and risky for no real benefit, especially that they still argue about what the 2d session will be [14:04] seb128, sure. About the theme, what we'll end up using for that? We have some stuff like apport that are gtk3 now [14:04] mterry, that's an open question as well to discuss at the same time [14:04] we might just want to back out gtk3 [14:05] yar [14:05] only apport language-selector and aptdaemon are using gtk3 at the moment [14:05] seb128: do you (as an archive admin) prefer a bug to move gir1.2-soup-2.4 to main (src: libsoup2.4 (main)) or can you do it directly? (libgdata is in depwait on gir1.2-soup-2.4) [14:05] geser, I can do it now [14:05] That seems easier than not. I wish py-gi worked better with gtk2, then it would be almost a one-line change [14:06] mterry, well I still think we need a gtk3 theme [14:06] would it be only for those using the GNOME3 ppa [14:06] we need to sync with dx and design on that though [14:08] geser, done [14:08] thanks [14:09] thank you for pointing it [14:09] mterry: I got apport working 90% with gtk2/gi; we just need to fix the introspectability of gtk_radio_button_new_with_label_from_widget; if we go this direction, that shoudln't be too hard [14:11] mterry, do you want to pick that task as well? [14:11] well maybe let's wait on the decision for gtk3 on the CD [14:12] pitti, use any special way to generate the tarball for ubuntu-mono ? [14:12] if we want to keep it we need to solve at least the theme issue [14:13] cyphermox: no idea, I never touched that [14:13] bzr bd tries to do get-orig-source and fails... I created the tarball manually now, but I was wondering if there was a "magic" way ;) [14:13] oh ok [14:13] seb128, should d-conf be in the desktop-set? I can ping cjwatson if so [14:13] cyphermox: oh, just for the current one? apt-get source doesn't work? I thought it tries that first [14:13] well, I'm doing changes to it [14:13] * mterry is thinking not being in core-dev is great for finding desktop-set holes. :) [14:15] mterry, yes [14:15] hehe, indeed [14:18] ok, apparmor profile fixed \o/ [14:18] cyphermox, bzr bd should do the right thing, there is no upstream tarball, it's just built from bzr [14:18] didrocks, great, what was it? [14:19] seb128: compiz is trying to create /etc/compizconfig/config and if apparmor deny it, it won't read it [14:19] so fallback to the ini backend [14:19] temporary workaround is to allow write in apparmor, then ACL deny and libcompizconfig is happy [14:19] didrocks, why does it try to create a config in a dir which is a system one? [14:20] then, I'll have a look why libcompizconfig is behaving like that [14:20] ok [14:20] didrocks, great work ;-) [14:20] seb128: well, I had some help from jdstrand for knowing where to look :) [14:30] I'm looking at the depwait on gnome-system-tools (which was easy to fix) but got stuck when trying to build the package. The error I got is: /usr/bin/ld: /usr/lib/libgtk-x11-2.0.so: undefined reference to symbol 'atk_object_set_name' ;; /usr/bin/ld: note: 'atk_object_set_name' is defined in DSO /usr/lib/libatk-1.0.so so try adding it to the linker command line [14:31] seb128, I don't know, it would just give up.. not the biggest deal though, I created a tarball and I'm about to push a branch and merge request [14:31] pitti, bigon: I think either way works, but not sure which is more correct [14:31] I'm a little bit confused as it complains about a lib and not an object file [14:32] kenvandine: ah, thanks; so if that gnome bz patch gets applied, we can drop our's? [14:33] yeah [14:33] but [14:33] geser: looks like it's missing an -latk1.0 ? [14:33] i think if you use the --c-include we need to do something to allow uninstalled [14:33]