UbuntuIRC / 2012 /02 /13 /#ubuntu-desktop.txt
niansa
Initial commit
4aa5fce
[00:11] <desrt> ugh. libneon is messed up.
[00:12] <lifeless> desrt: again ? :0
[00:12] <desrt> lifeless: https://bugs.launchpad.net/ubuntu/+source/neon27/+bug/845901
[00:12] <ubot2`> Launchpad bug 845901 in neon27 "libneon can't find SSLv2_server_method" [Undecided,Confirmed]
[00:14] <desrt> it's pretty seriously messed up. i'm surprised it builds at all.
[00:14] <desrt> it's using a function that completely does not exist in any library
[00:14] <desrt> presumably it existed in libssl at some point but was removed
[00:19] <micahg> that should've been fixed many releases ago
[00:22] <broder> IIRC we killed off SSLv2 in oneiric
[00:22] <desrt> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=622140 btw
[00:22] <ubot2`> Debian bug 622140 in neon27 "neon27: symbol SSLv2_server_method used" [Serious,Open]
[00:22] <broder> because i think i ran into a boost incompatibility issue
[00:22] <broder> upgrading some of my software to oneiric
[00:23] * micahg thought it was killed long before that
=== s9iper1 is now known as bil21al
=== bil21al is now known as s9iper1
=== m_conley_away is now known as m_conley
=== m_conley is now known as m_conley_away
[05:29] <RAOF> lifeless: Why would squid be dying with a SIGABRT when I try to start it? squid3 -k parse -f /etc/squid-deb-proxy/squid-deb-proxy.conf lists a couple of warnings, complains about not being able to write to /var/log/squid-deb-proxy, but otherwise completes successfully.
[06:03] <lifeless> RAOF: so if its calling abort() it should log something first
[06:04] <RAOF> FATAL: xcalloc: Unable to allocate 4154659762 blocks of 1 bytes!
[06:04] <RAOF> Ah.
[06:05] <RAOF> Hm. It's trying to allocate and zero 4GB of memory?
[06:07] <RAOF> lifeless: http://paste.ubuntu.com/840056/
[06:12] <BigWhale> Morning.
[06:16] <GunnarHj> pitti: Good morning, Martin! Just replied on bug 930785.
[06:16] <ubot2`> Launchpad bug 930785 in language-selector "gnome-language-selector crashed with DBusException in call_blocking(): org.freedesktop.Accounts.Error.Failed: 'bg_BG.UTF-8' is not a valid locale name" [Undecided,In progress] https://launchpad.net/bugs/930785
[06:17] <pitti> hey GunnarHj
[06:17] <pitti> GunnarHj: right, but I wonder if there is a cleaner solution
[06:18] <GunnarHj> pitti: Guess we can wait a while. It's not very urgent, I suppose.
[06:23] <didrocks> good morning
[06:27] <pitti> hey didrocks, how are you?
[06:27] <didrocks> pitti: I'm fine, thanks. And you, how was your week-end?
[06:32] <pitti> didrocks: quite fine, thanks! we visited our grandma, and had a nice birthday party
[06:34] <didrocks> ah nice :)
=== Amaranthus is now known as Amaranth
[06:51] <pitti> GunnarHj: followed up to 930785
[07:13] <lifeless> RAOF: fun
[07:13] <GunnarHj> pitti: Ok, then I'll write a patch with the 'locale -a' approach.
[07:13] <lifeless> RAOF: a 27PB cache ?!
[07:14] <lifeless> RAOF: that 4G alloc is probably happening for the hash table
[07:14] <lifeless> RAOF: e.g. your config is insane
=== tkamppeter_ is now known as tkamppeter
=== rickspencer3_ is now known as rickspencer3
=== Zdra is now known as xclaesse
[09:00] <seb128> hey
[09:00] <czajkowski> seb128: morning
[09:01] <seb128> hi czajkowski
[09:05] <pitti> bonjour seb128, ca va?
[09:07] <seb128> hey, pitti, ca va bien! et toi ?
[09:07] <pitti> seb128: je suis bien, merci!
[09:07] <pitti> seb128: did anything come up at the release team meeting?
[09:08] <seb128> pitti, no
[09:08] <pitti> I saw the fun with webkit, thanks for sorting this out
[09:08] <chrisccoulson> good morning everyone
[09:09] <pitti> hey chrhi
[09:09] <pitti> hey chrisccoulson, how are you?
[09:09] <chrisccoulson> hi pitti. i'm good thanks, how are you?
[09:10] <pitti> chrisccoulson: quite fine, thanks! had a nice long weekend
[09:10] <chrisccoulson> excellent :)
[09:10] <chrisccoulson> i had quite a busy weekend moving my desk and having everything set up and ready by this morning
[09:11] <pitti> chrisccoulson: oh, you have a new "office" room now?
[09:11] <chrisccoulson> yeah, i'm downstairs now. we've freed up the old room for my daughter
[09:12] <chrisccoulson> moving my desk downstairs was not much fun
[09:26] <pitti> seb128: hm, I'm trying to build glom, but libgdamm4.0-dev doesn't exist; wasn't there some effort to pacakge it?
[09:27] <seb128> pitti, dholbach has it in a ppa he asked for review on #ubuntu-devel last week I think
[09:27] <seb128> goocanvasmm as well
[09:27] <seb128> Laney said he would have a look iirc
[09:27] <seb128> better to check with them
[09:27] <pitti> ok, thanks; ignoring for now
[10:00] <seb128> pitti, there?
[10:00] <pitti> seb128: yes, I just spoke 2 mins ago :) )
[10:00] <seb128> pitti, I want to chat about gnome-settings-daemon if you have a minute
[10:01] <pitti> seb128: what's up?
[10:01] <seb128> pitti, so, I've been looking at 3.4 compared to 3.2, I will do a second round today but with order of magnitude numbers are
[10:02] <pitti> hm, LP just went down?
[10:02] <seb128> - 350 commits, I would say: 150 are on wacom (new code, side feature), 100 on translations, 25 on xrandr (better support for docking station)
[10:02] <seb128> some 50 are bug fixes
=== rickspencer3_ is now known as rickspencer3
[10:03] <pitti> right, I had to pull libwacom from jhbuild to build it
[10:03] <seb128> and there are like 15 less easy commits like keybindings to gsettings and dropping gconf plugin
[10:03] <seb128> I'm pondering going for 3.4, reverting the keybinding and gconf stuff
[10:03] <seb128> how crazy do you think that is?
[10:03] <pitti> is the latter a problem because unity or others still expect keybindings in gconf?
[10:04] <seb128> the other option is to backport some 25 (at least) commits
[10:04] <seb128> pitti, yes, compiz still use gconf
[10:04] <pitti> I just had assumed that it would require control-center 3.4, too
[10:04] <RAOF> lifeless: How has that config gone crazy? It's unmodified from squid-deb-proxy defaults.
[10:04] <seb128> in fact the keybinding stuff is one commit
[10:05] <seb128> I think it should be easy enough to make it select gconf or gsettings at run time, which would maybe let update gnome-shell, I know they really would like the new version
[10:05] <seb128> pitti, c-c, I didn't see a reason why it would
[10:05] <seb128> they didn't change configuration keys or format
[10:05] <seb128> it's mostly bug fixing, wacom, keybinding on gsettings (to revert), xrandr better dock support
[10:06] <pitti> the other stuff sounds fine indeed
[10:06] <seb128> pitti, if you think that's not too crazy I will go ahead with a second round and do a detailled summary with diffstat, diff, etc in a bug report
[10:06] <seb128> and land it in a ppa for testing today if I can
[10:06] <pitti> seb128: I'd like to get cnd's review on the new package, how the new wacom stuff holds up with our multitouch patches
[10:06] <pitti> seb128: that sounds great
[10:07] <pitti> happy to test from u-desktop PPA
[10:07] <seb128> pitti, wacom is a separate dir, we can't just not build it if we don't want to
[10:07] <seb128> can't->can
[10:07] <seb128> like everything in gsd it's its own .so
[10:07] <seb128> we can easily drop it
[10:07] <pitti> ah, I see
[10:07] <pitti> I guess we have to for the first round
[10:08] <pitti> it's in universe, and too old
[10:08] <pitti> seb128: is it an optional build dep, or do we need to hack the makefiles?
[10:08] <pitti> seb128: if you want, I can update it to 0.2.1, if that helps
[10:08] <pitti> or tjaalton might want to
[10:09] <seb128> pitti, we need to hack the makefiles I think
[10:09] <seb128> GNOME don't want to make stuff customizable or optional nowadays, they want GNOME to be it's designed ;-)
[10:10] <tjaalton> pitti: libwacom?
[10:10] <pitti> tjaalton: yes
[10:11] <pitti> seb128: oh, it seems remmina-plugins is obsolete now, binaries are now built by remmina itself
[10:11] * pitti cleans up
[10:12] <pitti> seb128: so, let's give that a try, if it's not too hard to do; I'm happy to test with my dock, too
[10:13] <tjaalton> pitti: I'll update it, 0.3 is ok?
[10:14] <seb128> pitti, ok, I will work on that today, still wanting to summarize where we go in details before update
[10:14] <seb128> tjaalton, that would be great, thanks
[10:14] <pitti> tjaalton: sure; but as I said, we can't use it in g-c-c right now (without a MIR at least)
[10:14] <pitti> if someone could actually test this with an actual wacom device, we could do a MIR
[10:14] <tjaalton> pitti: np, it'll be available at least
[10:15] <tjaalton> oh the wacom capplet?
[10:15] <tjaalton> I have an intuos4L
[10:16] <tkamppeter> pitti, hi
[10:17] <seb128> pitti, jbicha dropped me an email where he mentioned he's wanting to file the mir for libwacom, he really wants the new g-s-d in ;-)
[10:17] <seb128> pitti, well the mir is not a blocker for ppa builds
[10:17] <pitti> hey tkamppeter
[10:17] <pitti> seb128: right
[10:17] <seb128> pitti, so I don't know need to bother for the testing round
[10:17] <pitti> seb128: you still need >= 0.2.1, though
[10:18] <seb128> pitti, right, tjaalton is fixing that for me apparently though :p
[10:19] <seb128> pitti, btw remmina 1.0, well done
[10:20] <seb128> we got it in time for ff ;-)
[10:20] <tjaalton> seb128: yeah I'd like to have the wacom gui as well :)
[10:20] <pitti> seb128: I'll seed it today once jdstrand gives the final ok for freerdp 1.0.1
[10:20] <pitti> actually, I'll update the seeds right now
[10:21] <seb128> pitti, can you drop gnome-searchtools if we didn't yet?
[10:21] <seb128> pitti, gnome-search-tool
[10:21] <pitti> what's the replacement?
[10:21] <seb128> pitti, nothing, design said nautilus search and dash are enough searches
[10:21] <pitti> * gnome-screenshot
[10:21] <pitti> * gnome-search-tool
[10:21] <pitti> * gnome-system-log
[10:21] <pitti> * gnome-font-viewer
[10:21] <pitti> seb128: ah, right, nautilus search
[10:21] <pitti> ^ those are our gnome-utils right now
[10:21] <pitti> dropping search
[10:22] <seb128> pitti, gnome-screenshot is its own source in fact
[10:22] <seb128> but yeah
[10:22] <seb128> pitti, thanks
[10:23] <pitti> - * (vinagre)
[10:23] <seb128> pitti, they made standalone components for gnome-utils this cycle, I only bothered packaging gnome-screenshot though since the other ones didn't get any activity worth getting the new version
[10:23] <pitti> + * (remmina)
[10:23] <seb128> \o/
[10:23] <pitti> remmina recommends: remmina-plugin-rdp, remmina-plugin-vnc
[10:23] <pitti> so that should suffice
[10:23] <pitti> seb128: right, I meant "what used to be -utils"
[10:24] <tkamppeter> pitti, can you upload cups-filters into Debian and Ubuntu? Thanks.
[10:25] <seb128> pitti, right ;-)
[10:26] <pitti> tkamppeter: any chance to get a simple texttops back for the release critical bug?
[10:27] <tkamppeter> pitti, will do so and release as 1.0.1 upstream.
[10:27] <chrisccoulson> hmmm, i definitely need to get a lock for this door
[10:28] <pitti> tkamppeter: should I wait for this to avoid two debian uploads in a row?
[10:28] <tkamppeter> pitti, OK, wait for that.
[10:40] <ricotz> seb128, while looking at g-c-c and g-s-d could you drop Makefile.in from patches?
[10:40] <seb128> ricotz, is that creating issues?
[10:40] <ricotz> seb128, this cleans the patch a bit and an autoreconf it done anyway
[10:40] <seb128> ricotz, ok
[10:40] <ricotz> thanks
[10:40] <seb128> yw
[10:41] <ricotz> seb128, do you know if joss has a running gdm 3.2.1.1 in debian since it is in the svn
[10:42] <seb128> ricotz, not sure, ask him on #debian-gnome I guess
[10:43] <ricotz> ah, didnt recall the nickname ;)
[11:03] <pitti> Sweetshark: hey Bjoern, wie gehts?
[11:05] <pitti> Sweetshark: the "libreoffice" metapackage depends: libreoffice-report-builder-bin, but report-builder was dropped, so these pacakges don't exist any more; can you please drop the dep?
[11:17] <seb128> hum, could people stop spamming the desktop list with weird discussions? ;-)
[11:17] <chrisccoulson> yeah, i was thinking the same thing too
[11:22] <chrisccoulson> you know when somebody references a scene from borat the discussion has lost all hope of ever being useful
[11:24] <seb128> lol
=== greyback is now known as greyback|lunch
[12:18] <pitti> rickspencer3: finally! http://people.canonical.com/~ubuntu-archive/testing/precise_probs.html
[12:25] <pitti> Sweetshark: did you get my ping from earlier about the binfilter depends?
[12:27] <Sweetshark> pitti: yes, I try to reenable building report-builder, just to see if it works.
=== MacSlow is now known as MacSlow|lunch
[12:28] <Sweetshark> pitti: I guess that would require a FFE even if it would build *confused* (the FFE definitions dont map well to the libreoffice a-whole-distro-in-one-source-package world)
[12:29] <pitti> Sweetshark: hm, we had it before, though
[12:29] <pitti> seb128: do you happen to know about packaging lightdm-gtk-greeter?
[12:29] <seb128> pitti, being done by Debian, I will sync once it's uploaded
[12:29] <seb128> pitti, I can have a look to not wait on Debian this week if you want to but not today
[12:29] <seb128> pitti, it's mostly ready in the pkg-xfce vcs in Debian
[12:30] <seb128> pitti, http://anonscm.debian.org/viewvc/pkg-xfce/goodies/trunk/lightdm-gtk-greeter/debian/
[12:34] <pitti> seb128: ah, thanks
[12:35] <seb128> pitti, ideally somebody from xubuntu or lubuntu who step up to maintain that greeter
[12:35] <seb128> since they use it and we don't...
[12:35] <pitti> right
=== greyback|lunch is now known as greyback
[13:32] <Sweetshark> pitti: seems like some on the needed deps for report-builder got demoted to universe :(
=== MacSlow|lunch is now known as MacSlow
[13:41] <pitti> Sweetshark: don't worry, we'll re-promote it once it needs it again
[13:46] <Sweetshark> bug 919659 <- oh, great. lets just kill 3.4 and backport 3.5 ...
[13:46] <ubot2`> Launchpad bug 919659 in libreoffice "[Downstream] Can't open/save spreadsheet with password" [High,Triaged] https://launchpad.net/bugs/919659
[13:46] <Sweetshark> </sarc>
[13:50] <jbicha> good morning
[13:51] <pitti> hey jbicha, how are you?
[13:53] <jbicha> I'm doing good, how was your weekend?
[13:54] <pitti> jbicha: it was really nice
[13:55] <jbicha> remmina forgot to install its desktop file, it's fixed in trunk now though
[14:00] <seb128> jbicha, hey
[14:00] <seb128> jbicha, your g-s-d work, did you start on changes for the new version there or only merging on Debian?
[14:03] <jbicha> seb128: I did both
[14:03] <seb128> jbicha, hum, ok :-(
[14:04] <seb128> jbicha, I don't want to increase diff by mixing a debian merge with the update, I will pick bits out of your work and rebase manually then
[14:12] <kenvandine> wow the archive is slow today... been downloading updates for 30m this morning
[14:12] <kenvandine> you'd think it was release day :)
[14:15] <chrisccoulson> hmmm, my neck really isn't used to looking to the left
[14:15] <chrisccoulson> perhaps i shouldn't have rearranged my desk ;)
[14:15] <kenvandine> haha
[14:15] <pitti> chrisccoulson: noone's neck is -- the monitor should be right in front of you
[14:15] <chrisccoulson> pitti - oh, my primary screen is. but the laptop screen + dock are on the left
[14:16] <chrisccoulson> they used to be on the right, but that's where i also have my mouse
[14:16] <chrisccoulson> so i thought i'd try moving the dock so that i have more room
[14:16] <chrisccoulson> it feels a bit weird though
[14:18] <didrocks> m4n1sh: hey, how are you?
=== davidcalle_ is now known as davidcalle
[14:29] <mterry> seb128, pitti: Why did we re-enable indicators for the fallback session? That will challenge some assumptions elsewhere about whether Unity==indicators (e.g. the way we swap in our date/time panel for upstream's)
[14:30] <seb128> mterry,
[14:30] <seb128> gnome-panel (1:3.3.5-0ubuntu2) precise; urgency=low
[14:30] <seb128> * debian/patches/41_classic_layout.patch:
[14:30] <seb128> - Change the defaults to be more "classic Ubuntu": Show indicators
[14:30] <seb128> & show desktop, drop GNOME clock & notification area (LP: #846378)
[14:30] <seb128> mterry, jbicha did it
[14:30] <seb128> mterry, I've no strong opinion either way
[14:30] <seb128> mterry, hey btw, had a good w.e? ;-)
[14:30] <mterry> I also thought we were trying to not modify the upstream experience
[14:31] <pitti> mterry: err, we did?
[14:31] <mterry> seb128, yeah, it was good :)
[14:31] <pitti> mterry: I think we should not have any indicators in gnome-shell or gnome-panel
[14:31] <mterry> pitti, I tend to agree
[14:31] <seb128> pitti, cf the changelog I just copied
[14:31] <pitti> right, I think we should revert that
[14:31] <seb128> I've no strong opinion, jbicha and ricotz are in charge of GNOME upstream experience ;-)
[14:31] <pitti> jbicha: ^ opinion?
[14:32] <seb128> pitti, well he's the one who did the change so I guess he's in favor of it
[14:32] <pitti> jbicha: I'd rather ship a real upstream experience now
[14:32] <mterry> seb128, yar but he may not be aware of the consequences
[14:32] <seb128> I'm unsure, people updating for lucid would probably like to still get their gnome-panel session similar to what they are used to
[14:32] <pitti> jbicha: we don't really test indicators and other ubuntu-isms in gnome upstream, so we better not modify it too much?
[14:32] <mterry> jbicha, just FYI, other parts of the desktop now detect whether they are in Unity or not to expose indicator preferences
[14:33] <seb128> mterry, well what you say that is that we shouldn't have added indicator-applet back
[14:33] <pitti> correct
[14:33] <seb128> in the archive
[14:33] <pitti> we even removed it, didn't we?
[14:33] <mterry> Oh, I guess
[14:33] <seb128> pitti, because it was not ported to GNOME3
[14:33] <mterry> That's my biggest concern, yeah
[14:33] <pitti> no, because we (well, seems "only some of us") didn't really want it back
[14:33] <seb128> well, on the other side quite some people like gnome-panel with indicators
[14:33] <mterry> seb128, have them email GNOME to add it. ;)
[14:34] <seb128> lol
[14:34] <seb128> well we have indicator-applet and it's working
[14:34] <mterry> Ah, but it's not quite. Users have no way to modify some of the settings
[14:34] <pitti> and now we don't have a real gnome any more
[14:34] <seb128> it's about time we really fork gnome-control-center
[14:34] <seb128> we stack hacks on hacks between sessions
[14:35] <seb128> we should have an ubuntu-control-center and a gnome-control-center
[14:35] <seb128> but that's a discussion for next cycle ;-)
[14:35] <seb128> pitti, well gnome-shell is real GNOME
[14:35] <mterry> seb128, perhaps. But then to fix this, we'd be talking about swapping out the control center for the fallback session. Which gets us further from the GNOME experience.
[14:35] <seb128> ok, I'm out of this discussion
[14:36] <mterry> :)
[14:36] <seb128> I don't think gnome-panel is worth the time we are loosing discussing it
[14:36] <mterry> seb128, care to talk about keyboard indicators instead?
[14:36] <seb128> mterry, I'm happy to talk keyboard ;-)
[14:36] <mterry> I don't get the no-keyboard-indicator bug you see. So I'm not sure if I'm doing the same thing as you
[14:36] <seb128> mterry, sorry for the list spam, I should draft emails rather than just commit after each step of testing :p
[14:36] <mterry> heh
[14:36] <seb128> mterry, the user switching one? or the "in session indicator"?
[14:36] <mterry> The in-session indicator one
[14:37] <seb128> mterry, my switch to workspace <n> keybindings are broken as well with your update
[14:37] <seb128> so something is broken with g-s-d
[14:37] <seb128> I tried to --debug run it but I didn't see anything obvious
[14:37] <mterry> seb128, ??
[14:38] <mterry> seb128, hmm. sounds like maybe it's blocking on something? Never gets to the point where it can show keyboard indicator or handle events
[14:38] <seb128> mterry, well I've alt-1 for switch to ws1 for example, with the g-s-d from the ppa that stopped working
[14:38] <mterry> seb128, like if it couldn't get a response from accountsservice when it tried to tell it about your layouts?
[14:38] <seb128> so I'm wondering if there is an issue leading to some plugins not loading
[14:39] <seb128> hum
[14:39] <seb128> mterry, could be yes
[14:39] <mterry> My patch is real small. It's just calling accountsservice calls (but doing so synchronously)
[14:39] <mterry> seb128, can you double-confirm you have the accountsservice from the PPA?
[14:40] <seb128> mterry, do you use the gir?
[14:40] <seb128> ii accountsservice 0.6.15-2ubuntu4+keyboard1 query and manipulate user account information
[14:40] <seb128> ii gir1.2-accountsservice-1.0 0.6.15-2ubuntu4 GObject introspection data for AccountService
[14:40] <seb128> ii libaccountsservice0 0.6.15-2ubuntu4+keyboard1 query and manipulate user account information - shared libraries
[14:40] <seb128>
[14:40] <seb128> seems I didn't update the gir
[14:40] <seb128> could it break stuff?
[14:40] <mterry> No, shouldn't... it doesn't use the gir
[14:40] * kenvandine wishes robert_ancell was online for a hug, he cleaned up all the compiler warnings in gwibber :)
[14:41] <seb128> hehe
[14:41] <kenvandine> and he started a branch with bake :)
[14:42] <seb128> mterry, gdb on it seems to indicate it's not blocked
[14:42] <seb128> #1 0x00b05020 in __GI___poll (fds=0x9c5f590, nfds=10, timeout=3999)
[14:42] <seb128> at ../sysdeps/unix/sysv/linux/poll.c:87
[14:42] <seb128> #2 0x007d280b in g_poll (fds=0x9c5f590, nfds=10, timeout=3999)
[14:42] <seb128> at /build/buildd/glib2.0-2.31.16/./glib/gpoll.c:132
[14:42] <seb128> #3 0x007c4dde in g_main_context_poll (n_fds=10, fds=0x9c5f590,
[14:42] <seb128>
[14:43] <mterry> seb128, curious... my patch seems so harmless besides the possibility of blocking: https://launchpadlibrarian.net/92613030/gnome-settings-daemon_3.2.2-0ubuntu15_3.2.2-0ubuntu15%2Bkeyboard1.diff.gz
[14:43] <mterry> And it's not just crashing for you it sounds like
[14:44] <seb128> mterry, ok, scrap the "keybindings are not working"
[14:44] <seb128> mterry, they seem to work, maybe I had the wrong layout and didn't notice since the indicator is not showing
[14:44] <mterry> ah, heh
[14:44] <seb128> mterry, but you got the indicator to show?
[14:45] <mterry> seb128, yeah. I have 4 layouts configured, and I see the indicator
[14:45] <seb128> hum
[14:45] <seb128> d-feet returns what seems the right keymaps list
[14:46] * mterry wonders how widespread this is; needs another tester
[14:50] <mterry> didrocks, heyo!
[14:50] <jbicha> have y'all tried default gnome-panel without the indicators? it's pretty horrible, the battery icon does nothing, it doesn't scroll well, and so on
[14:51] * mterry waves to jbicha
[14:51] <jbicha> mterry: hi!
[14:51] <kenvandine> jbicha, nope, only with the indicators :)
[14:51] <seb128> mterry,
[14:51] <seb128> $ gsettings get org.gnome.libgnomekbd.keyboard layouts
[14:51] <seb128> ['de', 'fr\toss']
[14:51] <seb128> mterry, so that seems correct...
[14:52] <mterry> jbicha, I just wanted to make sure that it was understood that the preferences for the indicators don't show outside of Unity
[14:53] <jbicha> mterry: :( well at least it looks prettier
[14:53] <mterry> seb128, and you proved that g-s-d is not blocked with gdb... If you add a layout, does accountsservice reflect that over DBus?
[14:53] <didrocks> hey mterry :)
[14:54] <mterry> seb128, (the XKeyboardLayouts attribute)
[14:54] <jbicha> I think we really need a separate OnlyShowIn: value for GNOME Classic than for Shell
[14:55] <seb128> mterry, it does
[14:55] <mterry> seb128, so that means my patch in g-s-d is doing its job. Just somewhere along the line the indicator got lost
[14:56] <mterry> didrocks, can you do me a favor and try the keyboard packages from the desktop PPA?
[14:56] <seb128> mterry, I can easily trigger the issue if you need help debugging, like I "sudo apt-get install gnome-settings-daemon=3.2.2-0ubuntu15" and restart g-s-d and I get an indicator
[14:56] <mterry> didrocks, I want to see if this is a French thing or not :)
[14:56] <seb128> I use the ppa and restart g-s-d and I get none
[14:57] <didrocks> mterry: sure, you meant, the g-s-d here?
[14:57] <mterry> didrocks, there are four packages, accountsservice, g-s-d, lightdm, unity-greeter
[14:57] <mterry> didrocks, and then you have to log out and back in twice. :)
[14:57] <mterry> didrocks, might just be easier to dist-upgrade rather than pick them out
[14:58] <didrocks> mterry: ok, that can wait for ~ half an hour?
[14:59] <mterry> didrocks, sure
[14:59] <mterry> didrocks, no rush, thanks!
[14:59] <didrocks> will do :)
[14:59] <mterry> tedg, did you get a chance to look at my indicator-power patch? I wanted to distro-patch it in for FF, but would feel better if I could say "backported from trunk" in the changelog :)
[15:00] <seb128> mterry, I'm pretty puzzled at why your diff would break anything...
[15:00] <tedg> mterry, Yeah, I saw it, I'm not sure if it conflicts with charles' patch though.
[15:00] <tedg> mterry, I haven't looked at the two together yet.
[15:01] <mterry> seb128, me too :-/
[15:01] <mterry> charles' patch, eh? /me looks
[15:01] <seb128> mterry, can you confirm one of the other issues I reported at least? ;-)
[15:01] <mterry> seb128, heh, sure. let me try the switch user one
[15:02] <mterry> seb128, also, I tried switching layouts in the greeter and I wasn't able to log in anymore. So for me, it seems layout switching is working
[15:03] <mterry> seb128, wish there was a place to type that let you see what you were typing
[15:03] <seb128> mterry, hum ok, anyway none of those seem blocker for an upload so maybe get that in precise and let's see what bugs come
[15:03] <mterry> seb128, the indicator one is kinda bad
[15:03] <seb128> mterry, can't you enable back the "other" in some config way?
[15:03] <seb128> for nfs,ldap logins
[15:03] <mterry> seb128, other?
[15:03] <seb128> mterry, the entry that let you type an username
[15:04] <seb128> mterry, isn't that required for i.e ldap logins?
[15:04] <mterry> seb128, oh right. no? not sure what happened to that, but haven't seen it in a while. let me check cde
[15:04] <mterry> code
[15:04] <seb128> I used to type in that field to see what I typed
[15:04] <seb128> mterry, it was disabled by default, but I'm pretty sure that respect a config, like show user list
[15:05] <seb128> mterry, http://bazaar.launchpad.net/~unity-greeter-team/unity-greeter/trunk/revision/236
[15:05] <seb128> mterry, the check is easy to ack away at least for testing
[15:05] <seb128> -if (greeter.hide_users_hint)
[15:05] <seb128> +if(1)
[15:05] <seb128> I guess
[15:07] <mterry> seb128, or greeter-hide-users=true in /etc/lightdm/lightdm.conf
[15:07] <seb128> mterry, well then I guess you loose your users to test what config reading
[15:08] <seb128> it would only list the username,password prompt in that case?
[15:08] <mterry> seb128, I think so? Let me try. Not sure I have the absolute correct syntax yet
[15:08] <seb128> mterry, well that's what I would think the option do, hide your users
[15:09] <seb128> that might not be good for testing
[15:09] <seb128> I guess easier is to do a local build and show the chars in the password entry
[15:10] <ronoc> mvo, ping
[15:10] <mterry> seb128, hrm. I also don't get the switch-user behavior you describe. When you switch users, does the user list have to scroll to get to your name?
[15:10] <mterry> seb128, why not good for testing?
[15:10] <seb128> mterry, no, I think the greeter is spawned with that user selected
[15:11] <mterry> seb128, yar, but for me, it animates scrolling to the user. Does it do that for you?
[15:11] <seb128> mterry, vt switching sucks too much there, by the time I see the greeter it's already rendered, if it had to do an animation I don't see it
[15:11] <mterry> k
[15:11] <seb128> mterry, what user is selected by default?
[15:11] <seb128> mterry, the user I tested happens to be the first in the list
[15:11] <seb128> if that makes a difference
[15:11] <seb128> so no scrolling needed
[15:12] <seb128> mterry, that's maybe a corner case if the user selected for switching is the default one
[15:12] <ronoc> mvo, the dependency problem with apt-get install libpackagekit-glib2-dev
[15:12] <mterry> seb128, yup, if it scrolls, no bug. But I do get it for first user
[15:13] <seb128> mterry, I'm building a g-s-d locally with your patch and some printfs, let's see how that goes
[15:13] <mterry> seb128, it's not the default one, but the first alphabetically
[15:13] <ronoc> mvo, is preventing it from being released into precise
[15:13] <seb128> mterry, right
[15:13] <seb128> mterry, ok, good, so you can at least confirm this one ;-)
[15:13] <mterry> now let me try disabling chars
[15:13] <mterry> seb128, oh, I bet it looks horrible too for you! I didn't put the light-themes changes in the PPA
[15:14] <seb128> mterry, it looks weird but I didn't stop at cosmetic issues, it's not uif yet ;-)
[15:14] <mterry> still, gotta have all the bling one can get
[15:14] <seb128> mterry, hum, in fact I had only one glitch
[15:14] <mvo> ronoc: the dependency problem that it pulls in packagekit itself?
[15:14] <seb128> mterry, we got a light-themes snapshot recently, I think it includes the theming work Cimi did
[15:15] <seb128> mterry, or was there extra tweaks since?
[15:15] <ronoc> mvo, that's the one
[15:16] <mterry> seb128, no, don't think that includes his lightdm tweaks
[15:17] <mterry> oh it does, just in a different place...
[15:17] <seb128> mterry, wtf
[15:17] <mterry> so now i have to wonder how to get it to apply...
[15:17] <seb128> mterry, so your ppa package rebuild locally works...is there any chance the i386 ppa build built with the old accountsservice and that's creating issues or something
[15:18] <mterry> seb128, I can't see how... doesn't need its headers or dbus api or anything beforehand
[15:18] <mvo> ronoc: if the python-packagekit recommends get lowered to suggests, that should be all that is needed to fix it
[15:19] <mterry> seb128, does it also fix the switching layouts issue? :) /me hopes
[15:19] <seb128> mterry, on the greeter? I can't tell, I don't want to reboot now
[15:19] <ronoc> mvo, ok probably doesn't make a difference but for the record I'm not using python
[15:19] <didrocks> seiflotfy: do you kow if m4n1sh will get around soon? I think we need an activity-log-manager release with the COPYING issue fixed if we want to push it to main (and preferably the gtk warning/treeview issues fixed as well)
[15:20] <seb128> mterry, I lost your lightdm changes when I fixed lightdm yesterday and didn't restart since I got the new update from the ppa so I'm running the archive version
[15:20] <ronoc> mvo, it would be nice to get this today, any chance you could fix that and kenvandine can try the release again
[15:20] <mterry> seb128, no problem. all that's needed is liblightdm-gobject, so just having unity-greeter restart is enough
[15:21] <seb128> mterry, ok, that I can try
[15:23] <m4n1sh> didrocks: back. so it is not pushed in main yet?
[15:23] <m4n1sh> didrocks: since it is uploaded to repos, so is FF applicable?
[15:23] <seb128> mterry, yeah, that fixed it
[15:23] <m4n1sh> since this release is bug fix
[15:23] <seb128> mterry, so dunno wth with the ppa build
[15:24] <didrocks> m4n1sh: no, but putting in main requires to meet the FF
[15:24] <m4n1sh> oh noes
[15:24] <didrocks> m4n1sh: and there is no official release with the fixed copyright file, so I can't ask for a MIR
[15:25] <didrocks> m4n1sh: as you have that in trunk, can you cut a release with it?
[15:25] <m4n1sh> I fixed COPYING, the binary name issue
[15:25] <m4n1sh> yes
[15:25] <didrocks> m4n1sh: I'll push and harass mterry so that he can review the MIR
[15:25] <m4n1sh> but there are two open bugs
[15:25] <didrocks> great :)
[15:25] <didrocks> hum, which ones?
[15:25] <didrocks> (the desktop files are fixed)
[15:25] <didrocks> ?
[15:25] <m4n1sh> yes
[15:25] <m4n1sh> i checked it in a VM
[15:25] <m4n1sh> works
[15:25] <m4n1sh> two bug fixes are about application tab
[15:26] <m4n1sh> i can fix it now
[15:26] <m4n1sh> which leads to crash
[15:26] <m4n1sh> and other in history tab, which is mostly about wrong error messages
[15:26] <didrocks> m4n1sh: ok, tell me when you will have processed the release then :)
[15:26] <m4n1sh> trying in 2 hrs
[15:26] <didrocks> thanks :)
[15:26] <mterry> seb128, OK, with your manual build then, the only bug is the user-switching one? That's not bad
[15:27] <seb128> mterry, yes
[15:27] <m4n1sh> parallel clothes washing and bug fixing.. oh boy
[15:27] <seb128> mterry, I feel uneasy about the ppa build stuff though
[15:27] <seb128> mterry, did you try g-s-d from the ppa or your local build?
[15:27] <mterry> seb128, I'm currently on the PPA
[15:27] <mterry> seb128, no break that I can see
[15:27] <seb128> mterry, though I don't understand how the ppa could make any difference
[15:28] <mvo> ronoc: let me talk to ximion (in #Packagekit) about the best course of action then I can do aupload
[15:28] <ronoc> mvo, thanks
[15:28] <seb128> mterry, there are some warnings like /build/buildd/gnome-settings-daemon-3.2.2/plugins/keyboard/gsd-keyboard-manager.c:130: warning: undefined reference to `XkbKeysymToModifiers'
[15:28] <mterry> seb128, oh wait! I just killed gnome-settings-daemon and got your bug
[15:28] <seb128> but that's nothing that could create that...
[15:29] <mterry> seb128, so maybe I was on a local build after all and just now switched to the PPA one
[15:29] <seb128> mterry, ok, so ppa is b0rked, next question: "why"?
[15:29] <mvo> ronoc: did you see the reply from glatzor? that sounded encouraging, didn't it?
[15:29] <mterry> seb128, I will start a rebuild and at least see if it's a Heisenberg
[15:30] <ronoc> mvo, yep just responding now. I have already done everything he suggested, will do the renaming/refactoring object name after FF
[15:32] <mterry> brb
[15:32] <mvo> ronoc: cool - so that leaves some bugs, right? i.e. you don't always see signals?
[15:33] <ronoc> mvo, no I think it's fine
[15:33] <ronoc> mvo, afaict from my testing last week it seemed good to go
[15:33] <ronoc> mvo, it's the packaging
[15:33] <ronoc> *just
[15:33] <ronoc> deps
[15:35] <seb128> mterry, ok, got it
[15:35] <ronoc> mvo, mpt agreed to drop the requirement to represent 'updates installing'
[15:35] <ronoc> that made my life a hell of alot easier
[15:35] <ronoc> i.e. didn't have to bother with transactions
[15:35] <seb128> mterry, easy one ;-)
[15:35] <mterry> seb128, oh good!
[15:35] <seb128> mterry, the ppa build has " App indicator support: no"
[15:36] <mterry> heh!
[15:36] <seb128> mterry, I think something with the recent transitions
[15:36] <mterry> why.... h,
[15:36] <mterry> OK, so on that theory, the rebuild will be better
[15:36] <seb128> mterry, I wonder if it works locally because I still have an old lib or something
[15:36] <mterry> Oh, you think it needs a configure.ac change
[15:36] <seb128> yes
[15:38] <mvo> ronoc: ok, new PK is uploaded
[15:38] <pitti> didrocks: impressive compiz changelog!
[15:38] <mvo> ronoc: aha, nice to hear :)
[15:38] <didrocks> pitti: 4 months of upstream work :)
[15:38] <mterry> seb128, part of the configure script is checking for the gtk2 version, part is checking for the gtk3 one
[15:38] <ronoc> mvo, sweet thanks a mill :)
[15:38] <ronoc> kenvandine, ^^
[15:39] <ronoc> session restart brb
[15:39] <kenvandine> mvo, ronoc: thx
[15:40] <mvo> ta
[15:40] <mvo> eh
[15:40] <mvo> yw
[15:40] <mvo> you know what I mean :)
[15:41] <seb128> mterry, I don't get why it worked before
[15:41] <mterry> seb128, OK, I have a keyboard3 packaging building in PPA that I will ping you to try later. Should fix the indicator thing
[15:41] <seb128> mterry, in auto case it checks for the gtk2 version
[15:41] <mterry> seb128, a dep probably pulled in libappindicator-dev
[15:42] <seb128> mterry, yeah, must be
[15:42] <seb128> mterry, thanks!
[15:42] <seb128> mterry, I'm glad we sorted it ;-)
[15:42] <mterry> me too
[15:42] * mterry now looks at theming problem
[15:42] * mterry wishes things would stay fixed
[15:43] <seb128> mterry, oh, I know what changed
[15:43] <seb128> mterry, the gtk3 indicator-dev used to depends on the gtk2 one since they had the .h common for both version
[15:44] <mterry> ah...
[15:44] <seb128> mterry, that got fixed last week, they each have their .h now
[15:44] <seb128> ok, great, mystery solved ;-)
[15:44] <mterry> That doesn't seem right (sharing a .h). But I probably did that patch originally... My bad :)
[15:44] * seb128 back to updating g-s-d to 3.3
[15:46] <seb128> stupid question
[15:46] <pitti> mvo: hm, I don't understand your packagekit change -- shouldn't packages which need the API depend on PK | aptdaemon-pkcompat?
[15:46] <seb128> but how do I bzr bd-do to refresh my patches nowadays?
[15:46] <pitti> seb128: you mean in an UDD branch?
[15:46] <seb128> it refuses to log me into the shell because one patch fails to apply ... which is exactly why I want to log in, to refresh it
[15:47] <seb128> pitti, no, ubuntu-desktop debian only types
[15:47] <seb128> bzr bd-do fails because one of the patches doesn't apply
[15:47] <pitti> eww
[15:47] <seb128> how do I refresh patches then?
[15:47] <seb128> I guess that's rather an #ubuntu-devel question
[15:48] <pitti> hm, I don't see a --dont-apply-patches-dammit option
[15:48] <pitti> seb128: you could manually unpack the tarballs with tar xf ../foo.tar.bz --str=1
[15:48] <pitti> refresh the patches, and then "bzr clean-tree --ignored --unknown" again
[15:48] <pitti> but again, eww
[15:48] <seb128> pitti, well I'm rm debian/patches/series; bzr bd-do; bzr revert debian/patches/series
[15:48] <seb128> but that sucks
[15:49] <pitti> yes
[15:49] <seb128> well rather comment the patches from the serie and uncomment
[15:49] <seb128> but that's ridiculously stupid to do ;-)
[15:50] <pitti> mvo: oh, it's for the library, I see
[15:53] <mterry> tedg, ah, I see his branch. I think I like mine slightly better, but I'm probably just being biased. I don't care which gets implemented, but that decision should probably happen soon, because I don't want to distro-patch the wrong one and end up with a dead-end gsettings value
[15:58] * tedg sets up the octagon ... mterry, charles_, fight!
[15:59] <tedg> Of course I mean the octagon of love. This is a family IRC channel.
[15:59] <mterry> tedg, yeah, I would think instead of an octagon, we hug it out in the circle of friends
[16:01] <didrocks> seb128: you can bzr bd-do
[16:01] <didrocks> wait it to fail
[16:01] <didrocks> cd ../build-area/<your_project>
[16:01] <didrocks> quilt push/pop whatever you need
[16:01] <didrocks> and cp the patch…
[16:02] <seb128> didrocks, I like my edit series still better :p
[16:02] <didrocks> :p
[16:02] <seb128> at least it copies all the diff over for me
[16:02] <didrocks> yeah
[16:02] <seb128> but thanks ;-)
[16:02] <didrocks> yw ;)
[16:02] <didrocks> mterry: ok, ready now, too longer than expected
[16:02] <didrocks> mterry: so, you want me to dist-upgrade from the ubuntu-desktop and try lightdm/keybindings?
[16:03] <tedg> jono, I think mterry has found a new UDS activity... we set up a circle of friends, get those blow up sumo suits, and we use it to decide "hard conversations"
[16:03] <mterry> didrocks, hold off on that for a sec
[16:03] <mterry> didrocks, new package building
[16:04] <mterry> tedg, heh
[16:04] <didrocks> |o|
[16:05] <mterry> didrocks, what's that mean? :)
[16:06] <jono> tedg, lol
[16:06] <mterry> Is amd64 in bad shape in the archive? getting various unmet dependencies
[16:06] <didrocks> mterry: I hold up my arms, don't touch the keyboard anymore :)
[16:06] <seb128> didrocks, we figured the issue so there is less need for another tester
[16:06] <didrocks> ah great :)
[16:06] <tedg> mterry, Woodpecker has put a whole in his tree, away from keyboard to get shotgun.
[16:06] <mvo> pitti: yeah, just the libs to avoid that they pull in pk
[16:06] <mterry> tedg, heh
[16:06] <tedg> hole
[16:07] <mterry> seb128, you on i386? that build succeeded
[16:07] <seb128> mterry, yes
=== charles_ is now known as charles
[16:07] <mterry> try keyboard3
[16:07] <seb128> mterry, but I'm pretty sure that was the indicator stuff being disabled
[16:07] <seb128> mterry, you would feel better if I try the update from the ppa still? ;-)
[16:07] <charles> mterry, tedg: I don't mind which branch gets used, I kind of like my branch o/c :)
[16:08] <mterry> seb128, yeah :) This one says "App indicator support: yes" so just want to confirm it works
[16:08] <charles> I thought I marked the ticket as "in progress" though -- I'm not sure why the duplication of effort happened?
[16:08] <seb128> mterry, ok, I will test it, but only because it's you! ;-)
[16:08] <mterry> charles, this is why we need sumo suits! Yeah, sorry, I had the work item from UDS, but I never saw the bug
[16:08] <charles> is there something else I needed to do to flag the ticket to avoid duplication? :)
[16:08] <charles> mterry: ah, ok
[16:09] <mterry> charles, just that the bug and blueprint never got connected
[16:10] <mterry> tedg, check his branch first, since I'm pretty sure this was my mistake and it looks like his branch fixes some other stuff too? :)
[16:13] <seb128> mterry, hum, I wonder why the amd64 builders are unhappy
[16:14] <pitti> seb128: failing on build depends?
[16:14] <pitti> http://people.canonical.com/~ubuntu-archive/testing/precise_probs.html
[16:14] <pitti> might be the fontconfig upload
[16:14] <charles> I think I like my branch a little better since it binds directly to the gsettings. mterry's is a smaller patch though, and should work, so they're both good
[16:14] <pitti> fontconfig is built on all arches, so I hope it'll clear up in some 20 mins
[16:15] <pitti> if it's not that, I need to dig deeper
[16:15] <seb128> pitti, seems to be
[16:15] <seb128> pitti, danke
[16:15] <seb128> mterry, I'm an happy man again, thanks ;-)
[16:16] <seb128> (the ppa build works fine)
[16:16] <mterry> seb128, yay!
[16:16] <mterry> so I'll push once I figure the theme. Can fix the scrolling thing later, I think I know what that is anyway
[16:23] <Sweetshark> pitti: how come that even though: https://launchpad.net/ubuntu/+source/libxml-openoffice.org there is no libxml-java-openoffice.org to be found in my pbuilder?
[16:25] <Sweetshark> (and yes, I added a universe line the sources.list)
[16:27] <pitti> seb128: precise_probs back to normal, now mostly compiz-y stuff
[16:27] <seb128> pitti, great, thanks
[16:27] <pitti> Sweetshark: it existed until oneiric, not in precise any more
[16:28] <Sweetshark> pitti: ugh, doh!
[16:28] <Sweetshark> pitti: I keep mixing up release names again and again.
[16:29] <pitti> https://launchpad.net/ubuntu/+source/libxml-openoffice.org/+changelog
[16:29] <pitti> Sweetshark: ^ It was removed from Debian and we followed suit
[16:29] <pitti> it had no reverse dependencies any more
[16:29] <pitti> Sweetshark: if we need it back, we can reintroduce it
[16:30] <pitti> but it does smell pretty stale
[16:31] <Sweetshark> no, actually we need the "libxml-java" source package from debian ...
[16:31] <pitti> compiz failing on many arches doesn't help installability, of course
[16:31] <pitti> Sweetshark: ah, fun; that was also removed a while ago, so it's back now?
[16:32] <pitti> Sweetshark: syncing
[16:32] <Sweetshark> pitti: yes, that and a bunchload of other java-related stuff
[16:33] <ricotz> Sweetshark, hey, this doesnt sound like fun having it backported to lucid :\
[16:33] <pitti> Sweetshark: ah, needed unblacklisting (done now), that needs some time to propagate
[16:34] <pitti> Sweetshark: I'll sync it tomorrow
[16:35] <Sweetshark> ricotz: disable reportbuilder then?
[16:36] <Sweetshark> pitti: same for the flute source package, I guess?
[16:36] <pitti> Sweetshark: https://launchpad.net/ubuntu/+source/flute/+changelog
[16:36] <pitti> that should be published
[16:36] <pitti> Sweetshark: but it's in universe
[16:37] <pitti> Sweetshark: it's new, so needs an MIR if you need it
[16:38] <Sweetshark> pitti: ah, right. then there is is source pentaho-reporting-flow-engine, which seems to be missing from precise
[16:38] <pitti> https://launchpad.net/ubuntu/+source/pentaho-reporting-flow-engine/+changelog
[16:38] <pitti> Deleted in karmic-release (Reason: renamed to pentaho-reporting-flow-engine-openoffice.org)
[16:39] <pitti> Sweetshark: unblacklisted, added to sync TODO list
[16:40] <Sweetshark> pitti: and source liblayout also missing from precise
[16:40] <ricotz> Sweetshark, i noticed this conversation, so havent actually looked at it yet
[16:41] <ronoc> mterry, hey
[16:41] <ronoc> just trying to apply a patch to lightdm at the mo
[16:41] <pitti> Sweetshark: added to TODO
[16:41] <mterry> ronoc, ok
[16:41] <ronoc> mterry, I get this http://paste.ubuntu.com/840568/
[16:41] <ronoc> from unity-greeter trunk
[16:41] <pitti> Sweetshark: similar ones are jcommon-serializer, flute-1.3-jfree, librepository, libformula, libfonts-java
[16:41] <ronoc> mterry, I have installed lightdm from the desktop ppa
[16:42] <Sweetshark> pitti: I need librepository, libformula and libfonts too.
[16:42] <ronoc> mterry, should i branch from your branch to apply the patch there
[16:42] <pitti> Sweetshark: libformula sounds like a similar case
[16:42] <ronoc> it's a small patch
[16:42] <ronoc> mterry, will not conflict with your work
[16:42] <mterry> ronoc, I fixed that already...
[16:42] <pitti> Sweetshark: libfonts-java is not in Debian again
[16:42] <ronoc> hence why i would like to do from trunk
[16:42] <mterry> should be in unity-greeter trunk
[16:42] <ronoc> hmmm
[16:42] <ronoc> ok ill pull again
[16:43] <mterry> ronoc, let me doulbe-confirm
[16:43] <pitti> Sweetshark: sorry, mis-looked
[16:43] <mterry> ronoc, sorry, I think I forgot to push :( Will do so now
[16:44] <ronoc> mterry, sound
[16:44] <Sweetshark> pitti: and MIR for libbase
[16:44] <pitti> Sweetshark: so, flute-1.3-jfree again?
[16:46] <mterry> ronoc, done
[16:46] <ronoc> mterry, thanks
[16:46] <mterry> (r285)
[16:47] <ronoc> mterry, compiling now fine, thanks
[16:48] <Sweetshark> pitti: http://packages.debian.org/source/sid/flute
[16:49] <pitti> Sweetshark: ack
[16:49] <pitti> good night everyone!
[16:52] <Sweetshark> pitti: hey there! dont leave, without a big "thank you!"
[16:52] <Sweetshark> pitti: ;)
[17:02] <Sweetshark> pitti: four more: pentaho-reporting-flow-engine needs a sync and sacjava, libserializer and libbase a MIR
=== fenris is now known as Guest17978
[17:25] <desrt> DBO: hey. how persistent are bamfwindow structs?
[17:25] <desrt> do they drop when i unref them or do they drop when the window stops existing?
[17:37] <lifeless> RAOF: pastebin the confi, let me see.
[18:07] <didrocks> have a good evening everyone
[18:24] <m4n1sh> seb128: there?
[18:24] <seb128> m4n1sh, yes
[18:25] <m4n1sh> I am getting this issue at console
[18:25] <m4n1sh> Gtk-WARNING **: Failed to set text from markup due to error parsing markup: Error on line 1: Entity did not end with a semicolon; most likely you used an ampersand character without intending to start an entity - escape ampersand as &amp;
[18:25] <m4n1sh> I see a lot of bugs with this warning
[18:25] <m4n1sh> is it some issue with Gtk or the code?
[18:25] <seb128> m4n1sh, the code
[18:25] <seb128> you don't properly escape strings you use in a markup function
[18:25] <m4n1sh> you mean set_markup ?
[18:25] <seb128> it's like if you had invalid html syntax in an html document
[18:26] <seb128> m4n1sh, well any label which is in markup format yes
[18:26] <m4n1sh> checking
[18:26] <seb128> m4n1sh, gtk_set_label_markup for example
[18:26] <seb128> mterry, wb!
[18:27] <mterry> seb128, heh, went for lunch and forgot to re-launch IRC
[18:27] <mterry> seb128, did I miss anything?
[18:27] <m4n1sh> seb128: yes. Looking at any such
[18:27] <seb128> mterry, you missed me looking for you ;-)
[18:27] <mterry> seb128, what's up?
[18:27] <seb128> mterry, so...
[18:28] <seb128> mterry, when do you plan to update those ppa versions of yours? at least accountsservice and g-s-d
[18:28] <mterry> seb128, you mean upload?
[18:28] <mterry> seb128, I just uploaded accountsservice and lightdm
[18:28] <seb128> mterry, I've a g-s-d 3.3.5 test package I want to push to the ppa which needs to go with an unity-greeter upload
[18:28] <seb128> mterry, but I'm trying to avoid too much different testing versions around so I would like yours to be mainlined first
[18:28] <mterry> seb128, fixed the switch-user thing btw
[18:29] <seb128> great
[18:29] <mterry> seb128, right. I'm mainlining them now. I can push g-s-d and will roll a release of unity-greeter 0.2.1 and push it in
[18:29] <seb128> great, thanks
[18:29] <mterry> seb128, will ping you when done
[18:29] <seb128> mterry, I'm pondering trying to get you to rush a small fix in 0.2.1 ;-)
[18:30] <mterry> seb128, OK, shoot
[18:30] <seb128> mterry, so, small issue which becomes a problem thanks to desrt
[18:31] <seb128> g-s-d 3.3 renamed org.gnome.settings-daemon.plugins.wacom to org.gnome.settings-daemon.plugins.gsdwacom
[18:31] <mterry> seb128, ah, easy fix. sure
[18:31] <seb128> which of course leads unity-greeter to abort
[18:31] <seb128> mterry, well, easy "fix", I plan to do an upload with the rename in the ppa
[18:31] <mterry> seb128, man, why would they make that change...
[18:32] <seb128> mterry, the real fix would be make settings-daemon.vala check if schemas exist before trying to use them
[18:32] <seb128> to avoid any exit on broken schemas
[18:32] <seb128> mterry, yeah, that's a stupid thing to do to start with, thanks bastien...
[18:33] <seb128> mterry, don't bother for that version, I will use a breaks and do an unity-greeter matching upload
[18:33] <mterry> seb128, but if I'm going to mainline 0.2.1... why bother fixing it in PPA?
[18:33] <seb128> mterry, because g-s-d 3.3 is a to test, we stayed on 3.2 until now
[18:33] <seb128> and you will hit the opposite problem if you upload the rename with g-s-d 3.2
[18:33] <seb128> the name will mismatch the other way around
[18:33] <seb128> the real fix would be to skip invalid schemas
[18:33] <mterry> seb128, right. So what's the fix you want in 0.2.1? the skip invalid?
[18:34] <seb128> if possible
[18:35] <mterry> seb128, so I can certainly do that, but there will still need to be some coordination for the release, else gsdwacom will suddenly be enabled in unity-greeter. But regardless of that, I suppose it makes sense to avoid an abort
[18:35] <seb128> mterry, well you could list both if we don't abort on missing schemas ;-)
[18:36] <mterry> seb128, oh yeah. :)
[18:36] <mterry> That's TWO changes seb128
[18:36] <seb128> lol
[18:36] <seb128> indeed!
[18:36] <seb128> mterry, you know how to do the "handle missing schemas"?
[18:36] <mterry> I believe so
[18:36] <seb128> basically it should be check if the schemas is in g_settings_list_schemas ();
[18:36] <seb128> if not skip it
[18:37] <seb128> mterry, ok, since that's 2 changes you win 2 beers if you get that in your upload ;-)
[18:37] <mterry> I have to list all? Is this some plot of desrt's to discourage people from doing this crap?
[18:38] <seb128> mterry, yeah, I believe so :-(
[18:38] <cyphermox> seb128: re webkit, did you test the linker flags that are supposed to help? fwiw I tried --no-keep-memory and bluez builds locally, with a ulimit -Hv 3700000
[18:38] <mterry> seb128, he should have made it harder to rename schemas then
[18:38] <cyphermox> seb128: seems to peek at around 2.7 G memory or so when linking libwebkitgtk
[18:39] <seb128> cyphermox, bluez?
[18:39] <cyphermox> seb128: ignore me, I just noticed I did something incredibly stupid in testing this
[18:39] <cyphermox> seb128: what about bluez?
[18:39] <seb128> cyphermox, no I didn't yet, I delayed that to after ff
[18:39] <seb128> cyphermox, you wrote "--no-keep-memory and bluez builds locally"
[18:39] <mterry> seb128, there's also g_settings_schema_source_lookup
[18:39] <cyphermox> seb128: thinko. :)
[18:40] <seb128> mterry, it's not on http://developer.gnome.org/gio/unstable/GSettings.html ?
[18:41] <seb128> cyphermox, I didn't try yet in any case, it picks a bit over 3g without it, but your testing is appreciated
[18:41] <mterry> seb128, http://developer.gnome.org/gio/unstable/gio-GSettingsSchema-GSettingsSchemaSource.html
[18:41] <seb128> cyphermox, you say the testing is not correct? let me know if you try again
[18:41] <cyphermox> seb128: I will in a minute
[18:41] <seb128> mterry, oh ok, thanks
[18:41] <cyphermox> seb128: let me know if I can spare cycles again to help with this
[18:42] <seb128> cyphermox, well if you want to help you are most than welcome
[18:42] <cyphermox> the issue is that I was building on amd64 instead of i386
[18:43] <seb128> ok
[18:43] <seb128> you should still see if the option helps there I guess
[18:44] <seb128> jbicha, ricotz: ok, I've a working g-s-d 3.3.5 with a fixed unity-greeter, I will upload to ppa, do a bug with details for review, email etc later
[18:44] <seb128> jbicha, ricotz: I'm waiting first for mterry to clear up the ppa and upload his g-s-d and unity-greeter to precise
[18:45] <mterry> seb128, you need the new unity-greeter specifically? I can update g-s-d now, but will have to squeeze in some changes for unity-greeter (am going to try to fix the drag-window-by-menubar too
[18:46] <sil2100> Hi
[18:46] <sil2100> I would like to test out https://bugs.launchpad.net/compiz-core/+bug/864478
[18:46] <ubot2`> Launchpad bug 864478 in unity "Window shading is broken" [High,Fix released]
[18:46] <seb128> mterry, not if you are fine with me upload unity-greeter in the ppa with wacom->gsdwacom stuff to get with my gsd
[18:46] <sil2100> The description there says to modify '/apps/compiz-1/gwd/titlebar_action', but I don't see that in gconf settings
[18:46] <seb128> mterry, just be careful to not sneak the rename in the archive later on
[18:46] <seb128> sil2100, try #ubuntu-unity
[18:47] <seb128> sil2100, didrocks called it a day and nobody else has been looking at those issues, you have a better chance with the unity,compiz guys on #ubuntu-unity
[18:47] <seb128> sil2100, #ubuntu-unity is the right channel for unity question anyway since that's where you will find the people working on the code
[18:47] <sil2100> Understood
[18:47] <mterry> seb128, that's fine. I don't need further testing for unity-greeter in the PPA
[18:47] <seb128> mterry, ok, thanks
[18:48] <mterry> seb128, so I'll update g-s-d now then in archive
[18:48] <seb128> mterry, I'm out for dinner, if you upload your g-s-d version to precise I will upload mine later
[18:48] <seb128> mterry, excellent, thanks!
[18:48] <mterry> cool
[18:48] <mterry> have a good dinner
[18:48] <seb128> thanks ;-)
[19:05] <ricotz> seb128, alright, looking forward to it
[19:05] <ricotz> mterry, ^ :)
[19:05] <mterry> :)
[19:27] <DBO> desrt, bamf owns the ref
[19:28] <desrt> to me that says "this lives forever"
[19:29] <ricotz> DBO, hey
[19:29] <DBO> hey ricotz
[19:29] <ricotz> DBO, are there plans to enable the introspection build for bamf again?
[19:29] <DBO> desrt, it lives until bamf decides it shouldn't live
[19:30] <DBO> ricotz, uhhhh I think so
[19:30] <desrt> DBO: so i should take a ref on it is what you're saying :)
[19:30] <ricotz> DBO, i was hoping to get some official vala bindings instead of handwriting them ;)
[19:32] <ricotz> desrt, hey, are you playing with bamf in vala?
[19:32] <DBO> desrt, yes
[19:32] <desrt> ricotz: i wish :)
[19:32] <DBO> ricotz, I'll ask thomi about it today
[19:33] <ricotz> hmm, i am avoiding taking refs on views :\
[19:33] <broder> who should i bother about looking at bug #682338? it's been sitting in the sponsor queue for over a month
[19:33] <ubot2`> Launchpad bug 682338 in libcairo "GTK programs in Ubuntu 10.10 are sluggish over NX" [Medium,Confirmed] https://launchpad.net/bugs/682338
[19:33] <ricotz> desrt, DBO, http://bazaar.launchpad.net/~docky-core/plank/trunk/view/head:/vapi/bamf.vapi
[19:33] <broder> my original thought was to figure out how to detect NX in particular, but i can't find any way to distinguish it from the X server it was forked from
[19:33] <DBO> ricotz, you dont *have* to take a ref
[19:33] <DBO> but if you dont it will die on its own when you get a closed signal
[19:34] <ricotz> DBO, yeah, i guess i am hitting this problem
[19:37] <seb128> broder, you can upload if you feel like taking responsability for the change ;-)
[19:38] <broder> seb128: well, i think the implications of the change are fairly well documented: if someone were to happen to use the patched cairo with a *really* old X server, it might crash
[19:38] <seb128> broder, the issue is, what happens if you i.e ssh -X from a debian stable system?
[19:39] <broder> debian oldstable isn't old enough to trigger this bug, IIRC
[19:39] <mterry> stgraber, heyo. Is edubuntu ready for a gsettings unity-greeter?
[19:39] <seb128> broder, ok, I can upload that
[19:39] <seb128> broder, I will do it today
[19:39] <stgraber> mterry: no, but upload anyway, I'll take care of it this afternoon so it's fixed for our next daily
[19:40] <broder> seb128: instead of the patch that's there, i might change src/cairo-xlib-display.c instead to not set ->buggy_repeat = TRUE in the first place
[19:40] <mterry> stgraber, cool
[19:40] <seb128> broder, if you want to add a new patch to the bug please do
[19:40] <broder> seb128: i can make that adjustment and do the upload; i just wasn't comfortable with making the judgement call on whether the patch was conceptually acceptable
[19:40] <seb128> broder, or feel free to upload, I think the fix is fine
[19:40] <broder> ok. i'll double-check that debian oldstable isn't old enough to be problematic
[19:41] <seb128> broder, I think we should be fine and we are early enough in the cycle to change that if there are issues
[19:41] <broder> seb128: awesome, thanks. i'll take care of it later this evening then
[19:41] <seb128> broder, debian oldstable has xorg 7.3 it seems and the fdo bug mention 6.9
[19:41] <seb128> broder, thanks
[19:51] <desrt> DBO: please fix your bamf docs...
[20:30] <desrt> RAOF: more input issues :/
[20:31] <desrt> RAOF: the mouse cursor is often 'getting stuck'
[20:31] <desrt> usually (but not always) near the boundaries of displays
[20:31] <desrt> happens maybe 50% of the time
[20:32] <desrt> pulling the mouse back and 'trying again' is often enough to unstick it
[20:39] <mterry> seb128, ronoc, stgraber: released and uploaded unity-greeter 0.2.1
[20:40] <stgraber> mterry: thanks, will update edubuntu-artwork and upload
[20:43] <ronoc> mterry, good stuff
[21:07] <Beret> qq so I don't file bugs for no reason - is the current build of unity in precise known to be broken?
[21:12] <seb128> Beret, how broken?
[21:29] <Beret> seb128, keyboard shortcuts for changing desktops ignored
[21:29] <Beret> I file #931698
[21:30] <seb128> ok
[21:38] <RAOF> desrt: Can you be more specific? Unity's use of barriers on the screen edges could do something like that.
[21:39] <desrt> RAOF: since you mention it, i did switch to unity at the same time i did dist-upgrade
[21:39] <desrt> but it's not consistent
[21:39] <desrt> sometimes it happens, sometimes not
[21:39] <desrt> and it happens going both ways... not just when there's a panel there
[21:39] <RAOF> Dependent on how fast you're moving your pointer? :)
[21:40] <RAOF> Yeah, both ways is a bug.
[21:40] <RAOF> Allow me to propose a merge to fix that.
[21:40] <RAOF> lifeless: http://paste.ubuntu.com/840954/ is the config; throwing away /var/cache/squid-deb-proxy allows it to start annd work. Presumably that's a result of a corrupt cache going weird? I've got the original (broken) cache available.
[21:40] <desrt> RAOF: it seems to semi-depend on speed?
[21:40] <desrt> it happens a bit less often if i go really fast
[21:41] <desrt> is this seriously a feature?
[21:41] <RAOF> It was more of a feature when the launcher wasn't always visible.
[21:42] <desrt> maybe they should drop it now?
[21:42] <RAOF> Or at least turn off the barriers when it's not autohiding.
[21:42] <desrt> indeed
[21:42] <desrt> gnome-shell fixed that issue back in the late days of 3.0 by turning off autohide for multiple monitors
[21:42] <RAOF> It's useful when the launcher *is* being autohidden.
[21:44] <desrt> ya. i can imagine
[21:45] <desrt> hitting the 1 pixel required to show it would be pretty tricky otherwise
[21:45] <dobey> is there any way to turn off the launcher/panel on my second monitor, completely?
[21:47] <dobey> desrt: does gnome-shell put the dock on all screens too?
[21:47] <desrt> no
[21:48] <desrt> it doesn't even put one on the main screen except when in overview
[21:48] <desrt> it takes the minimalist approach to chrome
[21:50] <dobey> so the mouse thing was only an issue in gnome-shell if you had a screen to the left of the main screen i guess?
[21:50] <lifeless> RAOF: yes, thats weird - file a squid bug upstream :)
[21:58] <Laney> kenvandine: yo, just finishing (imho) libproxy
[21:58] <Laney> will push after testing if you want to give it a whirl, and then i'll propose it to pkg-gnome
[22:04] <desrt> dobey: right, actually
[22:04] <TheMuso> /c/c
[22:04] <desrt> dobey: the dash never auto-hid
[22:04] <desrt> dobey: but the workspaces switcher did
[22:05] <desrt> and it's actually clever enough to only disable autohide if there is a monitor to the right of the main screen
[22:05] <desrt> (ie: only disabled if having it enabled would cause a problem)
[22:05] <dobey> ah
[22:07] <dobey> i really just want to get rid of all the shell bits on the non-primary screens. though i suppose with the way unity works, it's a bit hard for it to do that. however, having the launcher right in the middle is really annoying
[22:09] <desrt> ya.... particularly with those barriers
[22:10] <dobey> indeed. though the barrier tends to only be annoying for me, while dragging a window across screens
[22:14] <dobey> and the unlock screen popping up on whichever screen has the mouse on it when i move it, seems a bit weird as well. but eh
[22:20] <kenvandine> Laney, awesome
[22:21] <kenvandine> i can try to test it tomorrow
[22:22] <Laney> rock
[22:26] * RAOF returns from some impromptu fglrx wrangling.
[22:27] <RAOF> desrt, dobey: We really want the barriers when dragging windows - again, snap-to-maximise(/half-maximise) is much easier with such a target :)
[22:27] <RAOF> gnome-shell probably wants it as well :)
[22:29] <RAOF> (Also for the lower-right message tray, for vertical multi-head)
[22:30] <desrt> RAOF: we actually have the barrier in the message tray and top panel
[22:30] <desrt> so you aim for those targets if you want a half-maximised window
[22:30] <RAOF> Oh, ow.
[22:30] <RAOF> You've got an impenetrable barrier in the lower left?
[22:30] <RAOF> I mean, lower right.
[22:31] <desrt> not impenetrable
[22:31] <desrt> but pretty hard
[22:31] <RAOF> How does it get dropped?
[22:31] <RAOF> What's the behaviour for pushing through?
[22:32] <RAOF> Also, wouldn't it be nicer to just hit the side of the screen rather than opposite corners?
[22:33] <dobey> RAOF: but that is something i *never* want. and afaik, there's no way to turn it off
[22:33] <dobey> RAOF: all i want to do is turn it off. :)
[22:33] <RAOF> dobey: There's no way to turn it off, no.
[22:34] <RAOF> I'm working out if we can get better threshold information in the server, so it's easier to both deliberately go through and hit.
[22:35] <dobey> i am more interested in just not having all the visual noise on my second screen though. the barrier i can deal with (at least i can as it is running right now, i guess i need to update and reboot again)
=== dduffey is now known as dduffey_afk
[22:40] <Laney> kenvandine: ok, pushed: git://git.debian.org/git/users/laney/libproxy.git - you'll need to git-dch --auto
[22:41] <Laney> I attempted to multiarch it too, perhaps you could review that