| Riddell | Sweetshark: https://bugs.launchpad.net/qt/+bug/914033 | 00:02 |
|---|---|---|
| ubot2` | Launchpad bug 914033 in qt4-x11 "threading crashes in qt with kde theme" [Undecided,New] | 00:02 |
| cyphermox | jasoncwarner_: cjk input in unity dash? | 08:08 |
| jasoncwarner_ | https://bugs.launchpad.net/ubuntu/oneiric/+source/ibus/+bug/880876 | 08:09 |
| ubot2` | Launchpad bug 880876 in unity "Unity causes ibus to not work correctly (spaces incorrectly placed)" [High,Confirmed] | 08:09 |
| seb128 | jasoncwarner_, cyphermox: so the ibus fix I uploaded before holidays is not enough? | 08:15 |
| cyphermox | maybe not, I'll try it here now | 08:15 |
| cyphermox | seb128: jasoncwarner_: indeed, I can reproduce the bug still | 08:18 |
| cyphermox | (in unity) | 08:18 |
| BigWhale | Good Morning | 08:19 |
| BigWhale | What is the preferred position for dialog buttons in Ubuntu? OK and CANCEL grouped together with OK being on the right side? | 08:22 |
| seb128 | cyphermox, jasoncwarner_: http://launchpadlibrarian.net/87652650/ibus_1.4.0-1ubuntu2_1.4.0-1ubuntu3.diff.gz is the fix I uploaded | 08:24 |
| cyphermox | yup, the patch is applied | 08:24 |
| cyphermox | I checked in case it was somehow dropped in the merge, but it seems not | 08:25 |
| seb128 | TheMuso`, do you think you could take on bug #880886 from rodrigo? | 08:40 |
| ubot2` | Launchpad bug 880886 in gnome-control-center "Doesn't associated with mnemonic_widget property or accessibility related properties with some preference tool dialogs in GNOME Control Center application" [Undecided,New] https://launchpad.net/bugs/880886 | 08:40 |
| seb128 | TheMuso`, seems a simple "check gtkbuilder files for a11y properties" | 08:40 |
| seb128 | mvo, hey | 08:47 |
| seb128 | mvo, could you roll an ubuntu-system-service to fix bug #877088 when you have some time? | 08:48 |
| ubot2` | Launchpad bug 877088 in ubuntu-system-service "Network proxy setting does not set apt.conf from the second time" [Low,In progress] https://launchpad.net/bugs/877088 | 08:48 |
| seb128 | mvo, i.e it's fixed in the vcs, just need an update to precise | 08:48 |
| seb128 | rodrigo_, hey | 09:08 |
| rodrigo_ | hi seb128 | 09:14 |
| seb128 | rodrigo_, how are you? | 09:14 |
| seb128 | rodrigo_, congrats for your new job ;-) | 09:14 |
| rodrigo_ | seb128, I'm fine, and you? | 09:14 |
| rodrigo_ | seb128, thanks! | 09:14 |
| seb128 | I'm great thanks | 09:14 |
| Riddell | Sweetshark: around today? | 09:29 |
| Sweetshark | yes, | 09:30 |
| Sweetshark | saw the bug, whatssup with it? | 09:30 |
| pitti | ubuntuone--syncdaemon --help | 09:41 |
| pitti | --im_ok_with_being_root_pretty_please_let_me_be_root | 09:41 |
| pitti | WTH :) | 09:41 |
| pitti | dobey: ^ that's you, isn't it :) | 09:41 |
| Sweetshark | add a --funroll-loops ... that always helps | 09:42 |
| pitti | dobey: input on bug 906462 greatly appreciated; I'm happy to work on a patch/branch, but would like to confirm that we can disable it by default | 09:51 |
| ubot2` | Launchpad bug 906462 in ubuntuone-client "ubuntuone sync daemon is writing IDLE messages to the log every 2 minutes" [High,Triaged] https://launchpad.net/bugs/906462 | 09:51 |
| tkamppeter | pitti, hi | 09:53 |
| ricotz | jbicha, hi, i am not sure if this a good idea yet, but since the cogl break it is probably fine now and should be quite 1.9.4 already | 09:53 |
| pitti | hello tkamppeter | 09:53 |
| seb128 | robert_ancell, https://docs.google.com/a/canonical.com/document/d/1ILTJDiDCd25Npt2AmgzF8aOnZZECxTfM0hvsbWT2BxA/edit?ndplr=1 section 2.6 | 09:55 |
| tkamppeter | pitti, it is about the USB backend of CUPS. | 09:56 |
| tkamppeter | pitti, as you perhaps know, Mike Sweet did not accept our libusb/usblp hybrid patch and suggested to deprecate usblp which we did in Oneiric by blacklisting. | 09:57 |
| tkamppeter | pitti, now I get many bug reports about problems communicating with USB printers. | 09:58 |
| pitti | tkamppeter: i. e. the libusb backend doesn't work properly? | 09:58 |
| tkamppeter | pitti, now we need a solution to make the libusb-based CUPS backend working as well as the former usblp-based one. | 09:58 |
| tkamppeter | pitti, especially the libusb-based backend does not support bi-di at all, what the usblp-based one does. | 09:59 |
| ricotz | hello | 10:00 |
| tkamppeter | pitti, for precise we would need to get it at least working as well as the usblp-based backend, later on packet mode should be added to remove HP's necessity to supply their own "hp" backend. | 10:00 |
| ricotz | seb128, i pushed a new rhythmbox snapshot to my staging ppa which is merged with debian, and it is running fine, but no changelog :( | 10:01 |
| tkamppeter | pitti, Mike Sweet is not willing to work on that as it is not in the interest of Apple (libusb is Linux-only AFAIK). | 10:01 |
| seb128 | ricotz, ok, I can sponsor that for you | 10:01 |
| pitti | tkamppeter: what kind of USB backend do they use in MacOS? | 10:01 |
| seb128 | ricotz, I guess you are not interested to write the changelog? :-) | 10:01 |
| ricotz | seb128, sorry, currently not, just wanted to point you there | 10:02 |
| tkamppeter | pitti, what we need is some libusb expert, preferably at Canonical, to review and improve the code of the libusb-based USB CUPS backend. | 10:02 |
| seb128 | ricotz, no worry, thanks | 10:02 |
| ricotz | seb128, please notice the glib patch which is needed for 2.31 | 10:03 |
| pitti | tkamppeter: hm, I don't think we have any in C :( | 10:03 |
| seb128 | ricotz, ok | 10:03 |
| tkamppeter | pitti, there is a separate source file for the Mac-OS-X version of the USB CUPS backend in the backend/ directory but I did not look into it. | 10:03 |
| pitti | tkamppeter: why was the usblp one removed, because the kernel module is deprecated, right? | 10:03 |
| tkamppeter | pitti, mostly because our hybrid backend was not accepted upstream and Mike answered my suggestion with deprecating usblp. | 10:04 |
| tkamppeter | pitti, in addition libusb gives more flexibility, like the possibility to use packet mode. | 10:04 |
| tkamppeter | pitti, and for me as distro maintainer, I could simplify auto-setup code as I need to only take into account libusb to detect and identify USB printers. | 10:05 |
| ricotz | seb128, please use the version string with "git" so the debian breaks/conflicts are working | 10:06 |
| tkamppeter | pitti, note that udev-config-printer (the Plug'n'Print) has awkward code to search and identify printers always with both methods. | 10:06 |
| pitti | tkamppeter: right, I thought that's why we blacklisted usblp, so that we could drop the hybrid one and only use libusb | 10:07 |
| tkamppeter | pitti, and now we need a libusb expert to make the libusb backend doing everything the usblp backend is capable of. | 10:08 |
| tkamppeter | pitti, looking into changelogs of libusb seems that Canonical's libusb competence is having a package sync bot ... | 10:10 |
| pitti | tkamppeter: that's pretty much correct | 10:10 |
| tkamppeter | pitti, is there someone at Canonical who develops on software which uses libusb to give Ubuntu certain additional USB functionality? | 10:11 |
| pitti | bryce: when sponsoring patches like in bug 645974, can you please make sure that they get reported/forwarded upstream? thanks | 10:11 |
| ubot2` | Launchpad bug 645974 in pm-utils "xfs handling on battery not working correctly" [Low,Fix released] https://launchpad.net/bugs/645974 | 10:11 |
| pitti | tkamppeter: note that cups uses a very ancient libusb (0.1), there is not even an upstream for this any more | 10:13 |
| pitti | tkamppeter: libusb-1.0-0-dev is much more modern, and perhaps even supports bidi | 10:13 |
| pitti | tkamppeter: but I'd say that any effort on improving libusb 0.1 is just wasted | 10:13 |
| pitti | tkamppeter: I checked the rdepends, there is no canonical project using libusb :/ | 10:14 |
| tkamppeter | pitti, I was also rather expecting that the USB backend code needs to be improved, not the code of libusb, probably all libusb versions are bi-di capable, as most devices (like scanners) used bi-di all the time. | 10:16 |
| tkamppeter | pitti, but advancing the backend to the new libusb would be a good idea anyway, only that this will not automatically introduce bi-di to it. | 10:17 |
| mvo | seb128: sure, done | 10:17 |
| pitti | tkamppeter: ah, ok | 10:17 |
| seb128 | mvo, danke | 10:20 |
| seb128 | ricotz, so basically getting your ppa version, tweaking the changelog and upload should work? | 10:20 |
| seb128 | pitti, recycled https://blueprints.launchpad.net/ubuntu/+spec/desktop-p-control-center-cleanup for the g-c-c design changes we discussed yesterday | 10:21 |
| pitti | ah, good | 10:21 |
| seb128 | robert_ancell, didrocks, mterry, agateau: ^ you get work items there | 10:23 |
| ricotz | seb128, yes | 10:23 |
| mterry | yay | 10:23 |
| seb128 | ricotz, thanks | 10:23 |
| pitti | bryce: sent upstream now, and committed to debian git | 10:24 |
| jbicha | we're still undecided about whether precise with get g-c-c 3.4 or 3.2? | 10:25 |
| seb128 | jbicha, 3.2 | 10:26 |
| Riddell | Sweetshark: does libreoffice still need the patch as a SRU? | 10:26 |
| seb128 | jbicha, 3.4 would move keybindings to gsettings which would mean requiring to port compiz to gsettings because they share that | 10:26 |
| Riddell | Sweetshark: or only qt? | 10:27 |
| jbicha | Compiz isn't going to gsettings this cycle either? | 10:27 |
| seb128 | jbicha, no, we have enough stability issues, we want to focus on fixing bugs | 10:27 |
| seb128 | not spend the cycle landing a new backend to then spend the end of the cycle debugging it | 10:28 |
| seb128 | you know how it goes with compiz... | 10:28 |
| seb128 | well maybe it wouldn't be that bad, but still we don't want to focus on bug fixing only | 10:28 |
| ricotz | seb128, this might be interesting to you https://bugzilla.gnome.org/show_bug.cgi?id=639875 | 10:30 |
| ubot2` | Gnome bug 639875 in GtkNotebook "crashes when unparenting a tab from a window (drag-n-drop it outside of it)" [Critical,Resolved: fixed] | 10:30 |
| jbicha | then I guess gnome-shell 3.4 will have to stay in the PPA then | 10:31 |
| seb128 | ricotz, thanks | 10:31 |
| seb128 | jbicha, yeah, it's a bit unfortunate since it means we will maybe have again a situation where the ppa will break unity | 10:31 |
| ricotz | jbicha, note to properly work g-s/mutter will need g-c-c/g-s-d too :\ | 10:32 |
| ricotz | jbicha, you can give them a try and see what it breaks (staging ppa) | 10:33 |
| jbicha | the breakage should be minimal, perhaps only the keyboard shortcuts panel of g-c-c | 10:33 |
| ricotz | yes | 10:34 |
| seb128 | jbicha, well, still not cool | 10:38 |
| seb128 | it means unity keybindings wouldn't be possible to update from g-c-c | 10:38 |
| jbicha | well compiz had a gsettings port months ago but I guess it didn't work well enough | 10:40 |
| jbicha | anyway, bye for now :) | 10:41 |
| seb128 | jbicha, it's supposed to work, still it's quite some change, and we were reluctant to update g-c-c and g-s-d to new versions for the lts for GNOME stability reasons | 10:43 |
| seb128 | jbicha, bye ;-) | 10:43 |
| BigWhale | This is how they name a function "Gtk.RadioButton.new_with_label_from_widget" and then they expect people to have lines within 80 characters ... | 10:45 |
| tkamppeter | pitti, any idea about the libusb-based backend problem? | 10:45 |
| BigWhale | I asked this before, but everyeone were still sleeping: What is the preferred position for dialog buttons in Ubuntu? OK and CANCEL grouped together with OK being on the right side? | 10:46 |
| pitti | tkamppeter: I'm afraid not :( this is again a case of orphaned piece of software :( | 10:46 |
| pitti | tkamppeter: so I guess we can just keep what we have right now, and possibly get rid of the hybrid backend to make things easier | 10:46 |
| tkamppeter | pitti, ignoring all the bug reports? | 10:48 |
| pitti | tkamppeter: well, if you have a better idea.. | 10:48 |
| tkamppeter | pitti, we have already dropped the hybrid backend. We are working with the upstream libusb-based backend. | 10:48 |
| tkamppeter | pitti, the move of deprecating usblp and dropping the hybrid backend introduced the problems. | 10:49 |
| Riddell | Sweetshark: nudge | 10:58 |
| BigWhale | What is the equivalent to GdkPixbuf in Gtk3? :/ | 12:23 |
| BigWhale | ok, just ignore me | 12:24 |
| dobey | pitti: not really me, no. ubuntuone-foundations team, but the idea sounds plausible | 14:00 |
| Riddell | Sweetshark: nudge | 14:21 |
| Sweetshark | Riddell: pong | 14:39 |
| Riddell | Sweetshark: yo, does libreoffice need the patch backported? | 14:41 |
| Sweetshark | Riddell: rechecking | 14:46 |
| Sweetshark | Riddell: http://cgit.freedesktop.org/libreoffice/libs-gui/commit/vcl/unx/kde4/main.cxx?h=libreoffice-3-4-4&id=4c2b93d96689f62c24ebdb2d4e87ac08d51ed53a <- is already in 3.4.4 (which we have in oneiric) | 14:47 |
| Sweetshark | Riddell: and we will SRU 3.4.5 rsn | 14:47 |
| Riddell | Sweetshark: ok so just qt that needs the SRU | 14:48 |
| === james_w` is now known as james_w | ||
| === m_conley_away is now known as m_conley | ||
| dobey | mterry: hey. can you poke at https://bugs.launchpad.net/ubuntu/+source/dirspec/+bug/913908 ? should be pretty easy to deal with i think :) | 15:41 |
| ubot2` | Launchpad bug 913908 in dirspec "[MIR] dirspec" [Undecided,New] | 15:41 |
| mterry | dobey, you need more than GLib's access functions for those directories? | 15:44 |
| dobey | mterry: we need to not use glib to get those directories, and we need it to be cross-platform. also, there are a *lot* of packages using python-xdg, so would be useful to move them off it, either to dirspec or to glib, as appropriate | 15:48 |
| mterry | dobey, and python-xdg is dead? | 15:52 |
| dobey | mterry: it is unmaintained, still sat in fd.o cvs, and hasn't been touched in years; so yeah, it's dead :) | 15:53 |
| dobey | dirspec isn't a replacement for all of pyxdg though. only the directory specification bits | 15:53 |
| mterry | dobey, can't we just patch python-xdg instead of coming up with a new upstream we maintain? :) | 15:53 |
| dobey | mterry: by patch you mean, fork and maintain pyxdg? | 15:55 |
| dobey | :) | 15:55 |
| mterry | dobey, well, patch it in ubuntu. I guess you need to worry about win32 too | 15:55 |
| dobey | i don't want to maintain all the other bits of it. and the directory spec bits are fairly trivial. | 15:56 |
| dobey | yeah, we have to worry about win32 | 15:56 |
| dobey | also, maintaining patches to a dead project is a waste of resources | 15:56 |
| mterry | dobey, less than maintaining a new package :) it's not like patches to a dead project need much maintenance | 15:57 |
| mterry | dobey, whatever, I'm not trying to be stop energy, just whining | 15:58 |
| dobey | mterry: and the dead project is either still dead, or you take it over and maintain it. and neither of those are amicable i think. so maintaining only the bits we need is more useful it seems :) | 15:58 |
| dobey | plus pyxdg has no test suite or anything nice really | 15:59 |
| dobey | and then there's the whole multi-platform issue. so yeah, i have thought about all this quite a bit, before actually making dirspec. and it was the best solution :) | 16:01 |
| === nekohayo_ is now known as nekohayo | ||
| === m_conley is now known as m_conley_away | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!